INVESTIGATING THE NEUROPLASTICITY OF EMOTIONAL MEMORIES · INVESTIGATING THE NEUROPLASTICITY OF...

INVESTIGATING THE NEUROPLASTICITY

OF EMOTIONAL MEMORIES.

Angela Jacques BBiomedSc

Submitted in fulfilment of the requirements for the degree of

Doctor of Philosophy

School of Clinical Sciences

Faculty of Health

Queensland University of Technology

2019

1

Keywords

Activity-regulated cytoskeleton-associated protein (Arc / Arg3.1), α-amino-3-

hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPAR), amygdala (AMG),

anxiety, basal amygdala (B), basolateral amygdala complex (BLC), brain derived

neurotrophic factor (BDNF), c-Fos, conditioned response (CR), conditioned stimulus

(CS), consolidation, extinction, extinction training, contextual fear conditioning

(CFC), contextual fear memory, dentate gyrus (DG), fear, hippocampus, fear

memory, fear memory trace, glutamate, hypothalamus–pituitary–adrenal (HPA) axis,

immediate-early genes (IEGs), immunohistochemistry, lateral amygdala (LA), long

term potentiation (LTP), N-methyl-D-aspartate receptor (NDMAR), memory

consolidation, memory maintenance, microglia, neuroanatomy, neurobiology,

neuroplasticity, Pavlovian fear conditioning, phosphorylated mitogen-activated

protein kinase (pMAPK), post-traumatic stress disorder, prefrontal cortex (PFC),

serotonin, synaptic plasticity, threat, unconditioned stimulus (US), vesicular

glutamate transporter 3.

2

Abstract

A central task in contemporary neuroscience is to identify the cellular and molecular

mechanisms underlying cognitive brain functions, and how alterations of these

mechanisms can lead to neuropsychiatric disease states. There are about 100 billion

neurons in the human brain, and approximately the same number of glial or

supporting cells and both types of cells play a unique role in neuroplasticity. The

complex web of connections they form is constantly being remodelled through

genetics, individual behaviour and our environment. This remodelling may involve

neurogenesis, increased dendritic branching and synaptic connections. However,

neoplasticity is biphasic and may take on a maladaptive nature resulting in atrophy,

reduced branching with smaller dendritic spines and decreased synaptic connections,

all referred to as negative plasticity. Negative neuroplasticity has been linked to

depression, stress, anxiety disorders, schizophrenia, phobias and post-traumatic stress

disorder. Although technological advancements abound we are yet to establish many

successful therapeutic interventions for physical and mental neurological disorders.

Only through continued research will we develop a greater understanding of

neuroplasticity, improved methods of quantification and identification of neural

correlates essential to these changes.

3

Table of Contents

Keywords .................................................................................................................................. 1

Abstract ..................................................................................................................................... 2

Table of Contents ...................................................................................................................... 3

List of Figures ........................................................................................................................... 6

List of Tables ............................................................................................................................ 9

List of Abbreviations .............................................................................................................. 10

Statement of Original Authorship ........................................................................................... 12

List of Publications and Presentations .................................................................................... 13

Acknowledgements ................................................................................................................. 18

Introduction .......................................................................................... 19

1.1 The Research Problem: Understanding the role of neuroplastic adaptations in emotional pathophysiology. ..................................................................................................................... 20

1.2 Context of Research: The mechanisms and biomarkers underlying neuroplasticity associated with emotional behaviours ..................................................................................... 28

1.3 Purposes ............................................................................................................................ 36

1.4 Significance and Scope: .................................................................................................... 40

1.5 Thesis Outline ................................................................................................................... 41

Functional Neuronal Topography: A Statistical Approach to Micro Mapping Neuronal Location ................................................................... 44

2.1 Abstract ............................................................................................................................. 46

2.2 Introduction ....................................................................................................................... 46

2.3 Step-By-Step Methods ...................................................................................................... 51

2.4 Statistical Analysis of topographic neuron density data ................................................... 68

2.5 Discussion ......................................................................................................................... 76

2.6 Conclusion ........................................................................................................................ 80

Localization of Contextual and Context Removed Auditory Fear Memory within the Basolateral Amygdala Complex ............................................ 83

3.1 Abstract ............................................................................................................................. 85

3.2 Introduction ....................................................................................................................... 86

3.3 Experimental Procedures .................................................................................................. 90

3.4 Results ............................................................................................................................. 102

3.5 Discussion ....................................................................................................................... 116

Micro-Topography of Fear Memory Consolidation and Extinction Retrieval within Prefrontal Cortex and Amygdala ............................................ 128

4.1 Abstract ........................................................................................................................... 130

4.2 Introduction ..................................................................................................................... 131

4.3 Material and Methods ..................................................................................................... 133

4

4.4 Results ............................................................................................................................. 142

4.5 Discussion ....................................................................................................................... 157

Microglial phenotype alters with varied fear memory recall in the prefrontal cortex..................................................................................................... 161

5.1 Abstract ........................................................................................................................... 163

5.2 Introduction ..................................................................................................................... 163

5.3 Methods .......................................................................................................................... 165

5.4 Results ............................................................................................................................. 174

5.5 Discussion ....................................................................................................................... 182

Contextual Fear Conditioning Alter Microglia Number and Morphology in the Rat Dorsal Hippocampus ...................................................... 186

6.1 Abstract ........................................................................................................................... 188

6.2 Introduction ..................................................................................................................... 189

6.3 Method ............................................................................................................................ 193

6.4 Results ............................................................................................................................. 204

6.5 Discussion ....................................................................................................................... 211

Axonal nonsegregation of the Vesicular Glutamate Transporter VGLUT3 within serotonergic projections in the mouse forebrain. ................... 220

7.1 Abstract ........................................................................................................................... 222

7.2 Introduction ..................................................................................................................... 223

7.3 Materials and Methods .................................................................................................... 226

7.4 Results ............................................................................................................................. 228

7.5 Discussion ....................................................................................................................... 240

7.6 References ....................................................................................................................... 243

Fear extinction recall mediated by 5-HT/VGLUT3 colocalisation 248

8.1 Abstract ........................................................................................................................... 250

8.2 Introduction ..................................................................................................................... 251

8.3 Methods .......................................................................................................................... 254

8.4 Results ............................................................................................................................. 262

8.5 Discussion ....................................................................................................................... 268

General Discussion ............................................................................. 274

9.1 Summary of findings ....................................................................................................... 275

9.2 Significance ..................................................................................................................... 277

9.3 Advanced considerations ................................................................................................ 278

9.4 Future directions ............................................................................................................. 280

9.5 Concluding remarks ........................................................................................................ 281

Appendix A: .......................................................................................................................... 283

Supplementary Material for chapter 7. ................................................................................. 283

5

Appendix B: .......................................................................................................................... 292

The Impact of Sugar Consumption on Stress Driven, Emotional and Addictive Behaviours. .................................................................................... 292

10.1 Abstract ......................................................................................................................... 293

10.2 Introduction ................................................................................................................... 294

10.3 Common neuronal pathways for sucrose consumption, addiction, emotions and obesity296

10.4 Physiological and neural substrates of sugar consumption ........................................... 302

10.5 Common neurochemistry underlie consumptive behaviours and emotions .................. 320

10.6 Common anatomical structures and neural substrates of stress driven, emotional behaviour .............................................................................................................................. 331

10.7 Sucrose Consumption Investigated ............................................................................... 341

10.8 Therapeutics for obesity, derived from studies of sucrose consumption ...................... 345

10.9 Discussion ..................................................................................................................... 350

Appendix C: .......................................................................................................................... 354

Glucocorticoid Receptor (GR) ......................................................... 354

Appendix D: .......................................................................................................................... 365

Mineralocorticoid Receptor ............................................................. 365

Bibliography ......................................................................................................................... 375

6

List of Figures

Figure 1-1 Auditory Pavlovian fear conditioning and extinction. ....................................... 23

Figure 1-2 Acquisition, consolidation and retrieval of a memory. ...................................... 24

Figure 1-3 MAPK pathway to neuroplasticity. .................................................................... 32

Figure 2-1: Steps for tissue sampling and measurement from behavioural data. ................ 53

Figure 2-2: Steps for producing raw coordinate data from identified neurons. ................... 60

Figure 2-3: Steps for producing and analysing topographical density maps. ...................... 66

Figure 3-1: Experimental design for behavioural training. .................................................. 91

Figure 3-2: Schematic illustration of aligned sections. ........................................................ 98

Figure 3-3: Arc, c-Fos and co-localised labelling of amygdala neurons. ............................ 99

Figure 3-4: Freezing to context and tone. .......................................................................... 104

Figure 3-5: Immediate early gene expression in Bregma coordinate -3.36mm. ................ 107




Figure 3-9: Schematic representation of IEG expression following conditioning. ............ 115

Figure 3-10: Schematic representation of total IEG expression following conditioning. ................................................................................................ 116

Figure 4-1: Experimental design. ....................................................................................... 136

Figure 4-2: Recall of auditory fear consolidation and extended extinction training result in differing levels of freezing. ........................................................... 144

Figure 4-3: Recall of a conditioned fear memory and extinction memory result in spatially different patterns of pMAPK expression in subregions of the amygdala. .............................................................................................. 146

Figure 4-4: Spatial analysis of pMAPK expression in the LA reveals a stable population of neurons specific to the recall of an extinction memory. ....... 149

Figure 4-5: Recall of auditory fear and extinction memory both activate pMAPK expression in the medial prefrontal cortex. ................................................. 150

Figure 4-6: Remote recall of an extinction memory and recent recall of an auditory fear memory both activate pMAPK expression in the infralimbic cortex. .......................................................................................................... 152

Figure 4-7: Spatial analysis of pMAPK expression in the PL reveals a different neuronal distribution between recent and remote recall of an auditory fear memory. ............................................................................................... 154

Figure 4-8: pMAPK expression in the IL cortex following recall of an extinction memory. ....................................................................................................... 156

Figure 5-1 Recall of recent and remote fear and extinction memories result in different levels of freezing. ......................................................................... 169

Figure 5-2 Schematic drawing showing the location of the acquired micrographs. .......... 172

7

Figure 5-3 c-Fos and Arc expression in the PFC are greater in the recall of a recent fear memory. ............................................................................................... 178

Figure 5-4 Microglia alter morphologically in response to fear memory recall. ............... 180

Figure 5-5 Morphological analysis of microglia. ............................................................... 182

Figure 6-1. Experimental Design for Behavioural Training. ............................................. 195

Figure 6-2. Illustration of LA and DH subregions and labelling in these Regions. ........... 202

Figure 6-3. Freezing to Context and Tone Data. ................................................................ 205

Figure 6-4. pCREB Expression in LA and DH. ................................................................. 206

Figure 6-5. BDNF Expression in LA and DH. .................................................................. 207

Figure 6-6. IBA-1 Expression in LA and DH. ................................................................... 208

Figure 6-7. IBA-1 Morphology in DG. .............................................................................. 211

Figure 7-1 Distribution of VGLUT3+ boutons within 5-HT+ and SERT+ axons in the prelimbic cortex. .................................................................................... 230

Figure 7-2 Distribution of VGLUT3+ boutons within 5-HT+ and SERT+ axons in the nucleus accumbens. ............................................................................... 231

Figure 7-3 Distribution of VGLUT3+ boutons within 5-HT+ and SERT+ axons in the posterior shell of the nucleus accumbens. ............................................. 233

Figure 7-4 Distribution of VGLUT3+ boutons within 5-HT+ and SERT+ axons in the striatum and lateral septum. ................................................................... 234

Figure 7-5 Distribution of VGLUT3+ boutons within 5-HT+ and SERT+ axons in the bed nucleus of the stria terminalis. ........................................................ 235

Figure 7-6 Distribution of VGLUT3+ boutons within 5-HT+ and SERT+ axons in the basolateral amygdala and central nucleus of the amygdala. .................. 236

Figure 7-7 Distribution of VGLUT3+ boutons within 5-HT+ and SERT+ axons in the CA1, CA2, CA3, dentate gyrus of the hippocampus. ............................ 238

Figure 7-8 Quantification of 5-HT neurons expressing vesicular glutamate transporter (VGLUT3) in various regions of the mouse forebrain. ............. 239

Figure 7-9 Visual representation of the proportion of 5-HT varicosities expressing the vesicular glutamate transporter VGLUT3 in various regions of the mouse forebrain. .................................................................................... 240

Figure 8-1 The recall of recent and remote fear results in different levels of freezing compared to extinction memory recall. ......................................... 257

Figure 8-2 Schematic drawing showing the location of the acquired micrographs. .......... 260

Figure 8-3 pMAPK labelling in the prefrontal cortex. ....................................................... 265

Figure 8-4 Colocalisation of VGLUT3, 5-HT and pMAPK. ............................................. 267

Figure 10-1 Reward pathway encompassing the mesocorticolimbic distribution of dopaminergic neurons. ................................................................................ 298

Figure 10-2 Regulation of feeding behaviour and food intake by central and peripheral appetite-regulating hormones and peptides. ............................... 308

Figure 10-3 Hypothalamic-pituitary-adrenal axis. Stress causes the release of corticotrophin-releasing hormone and vasopressin from the hypothalamus. ............................................................................................. 316

Figure 11-1 Glucocorticoid Receptor (GR) ....................................................................... 357

8

Figure 11-2 Glucocorticoid Receptor (GR) ....................................................................... 359

Figure 12-1 Mineralocorticoid Receptor ............................................................................ 371

9

List of Tables

Table 1 Options for statistical analysis ................................................................................ 76

Table 2 One-way ANOVA analysis of the volumetric density of 5-HT/VGLUT3 in the mouse forebrain. (Significant changes are highlighted in light grey). ........................................................................................................... 283

Table 3 One-way ANOVA analysis of the volumetric density of 5-HT/VGLUT3 in the mouse forebrain (Significant changes are highlighted in light grey). ........................................................................................................... 286

Table 4 One-way ANOVA analysis of the relative density of 5-HT/VGLUT3 in the mouse forebrain (Significant changes are highlighted in light grey). ......... 288

Table 5 The effects of sugar consumption on the reward pathway. ................................... 305

Table 6 The effects of sugar consumption on the orexigenic pathway. ............................. 312

Table 7 Published reports on the effect of sucrose and sweetener consumption on cognition, emotion and stress. ..................................................................... 341

Table 8 Therapeutics used in sugar consumption trials. .................................................... 346

10

List of Abbreviations

1dR Recent fear memory recall 5-HT Serotonin 7dR Remote fear memory recall AAFC Adjusted auditory fear conditioning ACTH Adrenocorticotrophic hormone AFC Auditory fear conditioning AMG Amygdala AMPAR α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid

receptor ANOVA One-way analysis of variance Arc / Arg3.1 Activity-regulated cytoskeleton-associated protein BC Box control BDNF Brain derived neurotrophic factor BLA Basolateral amygdala BLC Basolateral amygdala complex BLP Posterior portion of the basolateral amygdala CatFISH Compartmental analysis of temporal gene transcription by

fluorescent in situ hybridization CeA Central nucleus of amygdala CFC Contextual fear conditioning CNS Central nervous system CO Context only CORT Corticosterone CR Conditioned response CREB Cyclic-AMP response element-binding CS Conditioned stimulus CV Coefficient of variance DA Dopamine Den Dorsal endopiriform nucleus DG Dentate gyrus DH Dorsal Hippocampus ER Extinction recall ERK/MAPK Extracellular signal-regulated kinases/mitogen-activated

protein kinases FDR False discovery rate FMT Fear memory test HPA axis Hypothalamus–pituitary–adrenal axis IBA-1 Ionised calcium binding adaptor molecule 1 IEG Immediate early gene IL Infralimbic cortex ISD Immediate shock deficit LA Lateral amygdala LaDL Dorsolateral portion of the lateral amygdala LaVL Ventrolateral portion of the lateral amygdala LaVM Ventromedial portion of the lateral amygdala

11

LI Latent inhibition LTP Long term potentiation LV Lateral ventricle MAPK Mitogen-activated protein kinase MDA Multiple discriminant analysis MMTT Multiple memory trace theory mPFC Medial prefrontal cortex MROI Micro regions of interest MTL Medial temporal lobe Nac Nucleus accumbens NDMAR N-methyl-D-aspartate receptor PCA Principal component analysis pCREB Phosphorylated cylic-AMP response element-binding PL Prelimbic cortex pMAPK Phosphorylated mitogen-activated protein kinase PTSD Post-traumatic stress disorder RE Extinction recall ROI Region of interest SCT Standard consolidation theory SEM Standard error of the mean µm Micrometers UFC Unpaired fear conditioning US Unconditioned stimulus Vglut3 Vesicular glutamate transporter VTA Ventral tegmental area

12

Statement of Original Authorship

Date: 22-8-2019

QUT Verified Signature

13

List of Publications and Presentations

REFEREED JOURNAL ARTICLES (included in thesis) Chaaya, N., Jacques, A., Belmer, A., Beecher, K., Ali, S.A., Chehrehasa,

F., Battle, A.R., Johnson, L.R., Bartlett, S.E. Contextual fear conditioning

alter microglia number and morphology in the rat dorsal hippocampus.

Frontiers in Cellular Neuroscience.

Published: 14 May, 2019.

https://doi.org/10.3389/fncel.2019.00214

Belmer, A., Beecher, K., Jacques, A., Patkar, O. L., Sicherre, F., Bartlett,

S. E. Axonal nonsegregation of the Vesicular Glutamate Transporter

VGLUT3 within serotonergic projections in the mouse forebrain.

Frontiers in Cellular Neuroscience.

Published: 10th May, 2019.

https://doi.org/10.3389/fncel.2019.00193

Chaaya, N., Jacques, A., Belmer, A., Richard, D., Bartlett, S., Battle, A.,

& Johnson, L. (2018). Localization of Contextual and Context Removed

Auditory Fear Memory within the Basolateral Amygdala Complex.

Neuroscience.

Published: 1 February, 2019.

https://doi.org/10.1016/j.neuroscience.2018.12.004

Jacques, A., Chaaya, N., Hettiarachchi, C., Carmody, M.-L., Beecher, K.,

Belmer, A., Chehrehasa, F., Bartlett, S.E., Battle, A. R. Johnson, L. R.

14

(2019). Microtopography of fear memory consolidation and extinction

retrieval within prefrontal cortex and amygdala. Psychopharmacology.

Published: January, 2019.

https://link.springer.com/article/10.1007%2Fs00213-018-5068-4

Jacques, A., Wright, A., Chaaya, N., Overell, A., Bergstrom, H. C.,

McDonald, C., Battle, A.R., Johnson, L. R. (2018). Functional Neuronal

Topography: A Statistical Approach to Micro Mapping Neuronal

Location. Frontiers in Neural Circuits. Frontiers in Neural Circuits.

Published: 16 October, 2018.

https://doi.org/10.3389/fncir.2018.00084

Jacques, A., Chaaya, N., Beecher, K., Ali, S. O., Patkar, O. L., Battle, A.

R., Johnson, L. R., Belmer, A., Chehrehasa, F., Bartlett, S. E. Microglial

phenotype alters with varied fear memory recall in the prefrontal cortex.

Brain, Behavior and Immunity.

Submitted 27th December, 2018.

In revision.

Jacques, A., Chaaya, N., Belmer, A., Beecher, K., Ali, S. A., Battle, A. R.,

Johnson, L. R., Chehrehasa, F., Bartlett, S. E. (2019) 5-HT and glutamate

co-transmission in PFC during fear extinction recall. Frontiers in

Behavioural Neuroscience.

Submitted: 20 March, 2019.

In revision.

15

REFEREED JOURNAL ARTICLES (not related to this thesis)

Patkar, O. L., Belmer, A., Beecher, K., Jacques, A., Bartlett, S. E., The

therapeutic effects of pindolol’s unique pharmacology on the maladaptive

emotional and neurogenic consequences of alcohol. Frontiers in

Behavioral Neuroscience. Submitted 28th November, 2018.

In revision.

Chehrehasa, F., Jacques, A., St John, J. A., & Ekberg, J. A. (2018). The

Grueneberg olfactory organ neuroepithelium recovers after injury.

Brain research.

Published: 1 June, 2018.

https://doi.org/10.1016/j.brainres.2018.03.020

REVIEW ARTICLE (Completed during candidature – included in thesis appendix)

Jacques, A., Chaaya, N., Beecher, K., Ali, S.O., Belmer, A., Bartlett, SE.

The Impact of Sugar Consumption on Stress Driven, Emotional and

Addictive Behaviours.

Neuroscience and Biobehavioural Reviews.

Published: 21st May 2019

https://doi.org/10.1016/j.neubiorev.2019.05.021

MANUSCRIPTS IN PREPARATION (Completed during candidature – not included in thesis)

Chaaya, N., Jacques, A., Belmer, A., Beecher, K., Ali, S. A., Chaaya, M.,

Chehrehasa, F., Battle, A. R., Johnson, L. R., Bartlett, S. E. (2019)

16

Contextual fear memory maintenance in the pre-limbic cortex: Evaluation of

pMAPK, BDNF and Microglia.

Brain Structure and Function. Submitted 14th May, 2019

Under review

Jacques, A.*, Chaaya, N.*, Beecher, K., Ali, S. A., Chehrehasa, F., Belmer,

A., Bartlett, S. E. (2019). The Role of 5-HT/VGLUT3 Colocalisation in the

Mouse Prefrontal Cortex after Sugar Consumption. * Co-first authors.

Manuscript in preparation.

BOOK CHAPTERS (Completed during candidature – included in thesis appendix)

Jacques, A., Battle, A.R., Johnson, L.R., (2017). The Glucocorticoid

Receptor (GR). In Choi, Sangdun (Ed.) Encyclopedia of Signaling

Molecules [2nd edition]. Springer Reference, New York.

Published: 3 January, 2107.

https://link.springer.com/referenceworkentry/10.1007/978-3-319-67199-

4_101536

Jacques, A., Johnson, L. R., Battle, A. R., (2017). The Mineralocorticoid

Receptor (MR). In Choi, Sangdun (Ed.) Encyclopedia of Signaling

Molecules [2nd edition]. Springer Reference, New York.



4_101537

17

POSTER PRESENTATIONS Jacques, A. Chaaya, N., Battle, A. R., & Johnson, L. R. (2017).

Microanatomy of fear memory consolidation and extinction within sub-

regions of the prefrontal cortex and amygdala revealed by Arc and

pERK/MAPK activity. Society for Neuroscience, Washington DC, USA.

Jacques, A. Chaaya, N., Hettiarachchi, C., Carmody, M-L., Battle, A. R.,

& Johnson, L. R. (2017). Microanatomy of fear memory within subregions

of the prefrontal cortex and amygdala. Winter Conference on Brain

Research, Queenstown, New Zealand.

Jacques, A. Chaaya, N., Hettiarachchi, C., Carmody, M-L., Battle, A. R.,

& Johnson, L. R. (2016). Microanatomy of fear extinction within

subregions of the prefrontal cortex and amygdala revealed by arc and P

erk/mapk activity. Australasian Neuroscience Society 36th Annual

Scientific Meeting 2016, Hobart, Australia.

AWARDS AND ACHIEVEMENTS Australian-German Joint Research Cooperation Scheme, Australian

Universities. 2018 ($6250)

Grant-In-Aid (GIA) QUT. 2017 ($2,000)

Research Training Program Stipend (Domestic) [RTPSD], QUT.

2017- 2018 ($26,682)

Australian Neuroscience Society Student Travel Award 2016 ($215)

18

Acknowledgements

I gratefully acknowledge the unwavering support and guidance of my Principal

Supervisor Professor Selena Bartlett, Associate Supervisors Dr Fatemeh Chehrehasa

Dr Andrew Battle, and Dr Arnauld Belmer and my external supervisor Associate

Professor Luke Johnson. I consider it an honour and pleasure to have worked with

each of these experts in their fields. I would particularly like to thank Professor

Selena Bartlett and Associate Professor Luke Johnson for allowing me the

opportunity to be part of their teams, and to participate in the study of neuroscience.

Special thanks go to Dr Chehrehasa for her unwavering support which began during

my undergraduate degree and has been a constant throughout my PhD.

I am eternally indebted to the members of both the Bartlett and Johnson labs.

Without their tireless assistance, advice, humour and kindness this document would

not have been possible. I would like to express my deepest appreciation to my life

partner Raymond Penny, my sister Antoinette Turner and brother-in-law Scott

Turner and my lab partner Nicholas Chaaya. I extend many thanks to all my family

and friends for their past and continued support and encouragement throughout this

journey. Finally, I wish to acknowledge the sacrifice my rats made in order to assist

this investigation and bring us closer to understanding the neurobiology of

neuroplasticity. I would also like to gratefully acknowledge the financial support I

received from the Australian Government Research Training Program Stipend.

19

Introduction

This chapter provides a brief introduction detailing the research problem of

understanding the role of neuroplasticity in emotional pathophysiology, the context

of the research and the aims followed to advance knowledge in the field of

behavioural neuroscience.

20

1.1 The Research Problem: Understanding the role of neuroplastic adaptations in emotional pathophysiology.

Mental illness and neurological disorders are widespread in Australia, as in

most developed nations and generate a large personal, social and economic burden.

In 2003, 43% of years lost due to disability were reported to be the result of mental

and neurological disorders (Begg et al., 2007). Children suffering anxiety disorders

are twice as likely to attempt suicide, (Weissman et al., 1999) and with 264 million

people reported to be suffering from anxiety disorders in 2016/17 (Organization,

2017a), it is imperative we uncover the links between our emotions and simultaneous

neuroadaptations occurring in the brain.

Since the 1960s brain research has suggested we have a dynamic and ever

changing brain which is constantly being remodelled through genetics, individual

behaviour and our environment (Chambers, Liu, & McCouch, 1973; Kays, Hurley, &

Taber, 2012; Leuner & Gould, 2010; Pang, Short, Bredy, & Hannan, 2017; Wojtalik,

Eack, Smith, & Keshavan, 2018). This remodelling is inclusive of neurogenesis (the

formation of new neurons), greater dendritic branching and increases in synaptic

connections (synaptogenesis). This remodelling is collectively known as

neuroplasticity. Neuroplastic changes may also take on a maladaptive nature

resulting in decreased levels of neurotransmitters, atrophy, reduced branching with

smaller spines and decreased synaptic connections, all referred to as negative

plasticity.

Negative neuroplastic changes have been linked to depression (Fuchs, Czéh,

Kole, Michaelis, & Lucassen, 2004), stress (Pittenger & Duman, 2008), anxiety /

mood disorders and schizophrenia (Krystal et al., 2009), phobias (Reznikov, Fadel,

& Reagan, 2011) and posttraumatic stress disorder (PTSD) (Deppermann, Storchak,

21

Fallgatter, & Ehlis, 2014). Anxiety disorders are often characterised by the avoidance

behaviour attributed to the normal reaction to a conditioned fear cue (Grillon, 2002),

suggesting an abnormally high response to fear is common to anxiety disorders.

Increased amygdala and prefrontal cortex activity have been noted in patients with

PTSD (Shin et al., 2004), obsessive compulsive disorder (Straube, Mentzel, &

Miltner, 2005) and phobic disorders (Straube et al., 2005).

The neuroplasticity of post-traumatic stress disorder

Traumatic events such as war, natural disasters, serious motor vehicle

accidents and physical or sexual abuse may lead to the development of PTSD in

some individuals (Kilpatrick et al., 2013). Witnessing such events and repeated

exposure to trauma, as experienced by many emergency services personnel, may also

lead to the development of pathological fear memories (Luke R Johnson, Jennifer

McGuire, Rachel Lazarus, & Abraham A Palmer, 2012; Weiss, Marmar, Metzler, &

Ronfeldt, 1995). PTSD is a memory-based disorder, clinically defined as the

presentation of behavioural symptoms three months post trauma (Gray & Liotta,

2012; L. R. Johnson, McGuire, Lazarus, & Palmer, 2012). In order to create abiding

treatments for trauma and stressor related disorders a thorough knowledge of the

neural circuits involved in the formation of fear memories must be attained.

Learning and memory pathology have long been associated with PTSD,

where failure to extinguish fearful memories intensifies survival mechanisms to

debilitating levels (Elzinga & Bremner, 2002). The social and economic burden from

comorbid substance abuse, depression, suicide and the high rate of relapse after

therapy substantiate the requirement for development of improved therapeutics for

PTSD (Brown, Recupero, & Stout, 1995; Campbell et al., 2007; Hendin & Haas,

1991; Possemato, Wade, Andersen, & Ouimette, 2010).

22

Originally thought to be part of the natural progression through trauma

processing, PTSD is now known to involve abnormalities in the neurobiology of the

hypothalamic-pituitary-adrenal (HPA) axis (Yehuda, 2001). The HPA axis, as shown

in figure 2.1 is an inhibitory loop involved in the production and release of

glucocorticoids. Glucocorticoids such as cortisol are released into the blood stream

as a normal homeostatic regulation throughout the day. However, during the body’s

response to stress, cortisol release from the adrenal glands is increased. It binds to

mineralocorticoid and glucocorticoid receptors in the brain where it has been shown

to enhance the consolidation of pathological fear memories (Buchanan & Lovallo,

2001; McFarlane, Atchison, & Yehuda, 1997). Disruption of the glucocorticoid

receptors in the amygdala may play a role in the development of pathological

memories (Keller et al., 2016). Genomic studies investigating the mechanisms

involved are attempting to detect an endophenotype for people at higher risk of

developing PTSD and related disorders (Keller et al., 2016; Mehta & Binder, 2012).

Fear memory consolidation

The consolidation of fear memory involves the process of stabilization from

short term to long term memories (Luke R Johnson et al., 2012). During the1920s

Ivan Pavlov developed an animal model of conditioning that entailed learning that

certain cues within our environment can be associated with other memories (Pavlov,

1927). The most acclaimed of these involved dogs salivating when Pavlov rang a bell

but extended to associations of danger which could be used as predictors to avoid

aversive situations (Sevil Duvarci & Denis Pare, 2014; Maren, 2001).

Pavlovian auditory fear conditioning in rodent models is generated by pairing a

conditioned stimulus (CS), such as an auditory tone, with a noxious unconditioned

stimulus (US), such as a low grade electric foot shock to form a consolidated

23

associative fear memory (Luke R Johnson et al., 2012; Pavlov, 1927). This method

as shown in figure 1.1 (a) is often utilised in PTSD studies as the mechanisms

involved are similar to those of the fear memory formation that may progress to

pathological fear (Milad, Rauch, Pitman, & Quirk, 2006). The animals learn to

associate the neutral tone with the aversive shock and when later presented with the

CS (tone) alone, the conditioned response (CR) of freezing is elicited (M. Fendt &

M. S. Fanselow, 1999; Michael T Rogan, Stäubli, & LeDoux, 1997). Freezing is

denoted by a stillness of movement, other than the motion produced by respiration

and is often used as a rudimentary quantification as opposed to measurements of

sympathetic stress responses (Gisquet-Verrier, Dutrieux, Richer, & Doyère, 1999).

(a) (b)

Figure 1-1 Auditory Pavlovian fear conditioning and extinction.

(a) Conditioning a rodent involves the rodent being placed into a chamber where it receives a mild foot shock at the same time it is presented with an auditory tone. (b) Extinction training is performed by repeated presentation of the tone alone. Extinction is said to be complete once the conditioned response of freezing is no longer apparent. Figure adapted from (Johansen, Cain, Ostroff, & LeDoux, 2011).

Watson and Rayner (Watson & Rayner, 1920) demonstrated the process to be

applicable to humans by conditioning a baby ‘Albert B’ to fear a white rat by pairing

the rat with the loud strike of a hammer on metal (Maren & Fanselow, 1996).

Pavlovian fear conditioning has been demonstrated in several other mammalian

species including monkeys, rodents and cats (Hadley C Bergstrom, McDonald, &

24

Johnson, 2011; Brady, Schreiner, Geller, & Kling, 1954; Joseph E LeDoux,

Cicchetti, Xagoraris, & Romanski, 1990).

The stabilisation of the associative memory occurs after the initial acquisition

of the fear memory and is known as memory consolidation (Alberini, 2005;

McGaugh, 2000). Once a fear memory has had time to consolidate it becomes

extremely resistant to change (Flavell, Lambert, Winters, & Bredy, 2013; H. J. Lee,

R. P. Haberman, R. F. Roquet, & M.-H. Monfils, 2016). The assimilation of a short

term memory to a more resilient long term memory (see figure 1.2) occurs through a

process known as long term potentiation (LTP) which is dependent on N-methyl-D-

aspartate receptors (NMDAR) (McGaugh, 2000; Pinel & Barnes, 2017). Studies have

shown this process of stabilisation also requires new protein synthesis (Santini, Ge,

Ren, de Ortiz, & Quirk, 2004; Schafe & LeDoux, 2000). During retrieval of a

consolidated fear memory, both reconsolidation and extinction of the memory may

occur (Monfils, Cowansage, Klann, & LeDoux, 2009; Nader, Schafe, & LeDoux,

2000). When a memory is recalled, it becomes labile and requires a new process of

consolidation or reconsolidation (Wiltgen, Brown, Talton, & Silva, 2004).

Figure 1-2 Acquisition, consolidation and retrieval of a memory.

Once a memory has been acquired it undergoes the process of consolidation. This alters the memory from short term, an easily erased memory to long term where it becomes more stable. The process of memory retrieval may include both deconsolidation and reconsolidation of the memory (Maren, 2011).

25

Fear memory extinction

In order to decrease the heightened reaction to fear, that is characteristic of

PTSD, fear extinction paradigms, based on exposure therapy, are a leading treatment

for PTSD. Exposure therapy involves repeated exposure to the object of fear,

provided within a safe environment (Foa et al., 1999; Marek, Strobel, Bredy, & Sah,

2013) and has proven effective in the reduction of avoidance behaviours

symptomatic of PTSD but less effective in treating other symptoms such as the

recurring memories, anhedonia, dissociation and hyperarousal (Levin, 2008; Tarrier,

2001). Research by Tarrier and colleagues revealed approximately one third of

patients treated with exposure experienced an increase on the Clinician Administered

PTSD Scale, suggesting a worsening of symptoms occurred post treatment (Tarrier,

2001).

Pavlovlovian fear extinction training consisting of fear conditioning followed

by presentation of the CS (tone) alone (figure 1.1 (b)), which is widely used today in

behavioural models to mimic exposure therapy (for a comprehensive review see

(Maren, 2011). To reduce the fear associated with the CS, extinction training is

introduced at least 24 hours after fear conditioning. This initially results in freezing,

regardless of the absence of the shock, demonstrating the animal has learned the

correlation between the tone and the shock (Tronson & Taylor, 2007). After repeated

exposure there is a cessation of the conditioned response.

Pioneering neuroscientist Jerzy Konorsky (Konorski, 1967), who studied the

physiological properties of behaviour, suggested that extinction involved new

learning and the formation of a new association in competition with the original

memory. Bouton led the way through the 80’s and 90’s qualifying this proposal by

showing fear memories could recover spontaneously, renew in context and become

26

reinstated if exposed to the original threat (Bouton & Bolles, 1979; Bouton & Peck,

1989; Brooks & Bouton, 1993).

Extinction training reduces the expression of fearful behaviour, but the

extinction memory is not permanent and the pre-existing fear memory is not

extinguished, as demonstrated by a return of the fear response 24 hours after training

(Milad et al., 2008; Monfils et al., 2009). The new extinction memory may suppress

the original fear memory for a time, but the introduction of certain stimuli may result

in reappearance of the original fear memory (S. Li, Kim, & Richardson, 2012;

Maren, 2011; Onoue, Nakayama, Ikegaya, Matsuki, & Nomura, 2014). Recent

research by Li and colleagues suggests DNA modifications in the brain accompany

the formation of fear extinction memories (X. Li et al., 2018). Therefore, it would

appear that divergent molecular mechanisms and the neural circuits involved in the

extinction process relevant to cognitive extinction based therapies are a long way

from being understood.

Neural pathways

Although investigation of distinct neural pathways is not within the scope of

the research discussed herein, it is relevant to understand the functional brain nuclei

complicit in fear memory formation. The auditory thalamus and auditory association

cortex process the CS, while the somatosensory cortex and thalamus assimilate the

US (Medina, Repa, Mauk, & LeDoux, 2002). These sensory pathways, along with

contextual information processed by the hippocampus, synapse with principle

neurons in the lateral amygdala (LA) (Maren, 2001). Damage to the LA has been

shown to prevent auditory fear conditioning in rodent models (Joseph E LeDoux et

al., 1990). The LA and basolateral amygdala (BLA) consist of approximately 80%

glutamatergic pyramidal neurons with pyramidal shaped somas, long axons and

27

multipolar dendritic trees, often forming excitatory synapses with other principle

cells (H.-C. Pape & D. Pare, 2010).

The reconsolidation process involves protein synthesis to allow the retrieved

memory to continue to exist and to permit information update to existing memories

(Monfils et al., 2009; Nader et al., 2000). It is argued that consolidation and

reconsolidation involve common mechanisms and signalling pathways though

investigations of cellular correlates and specific brain subregions involved in the

process remain incomplete.

Fear extinction paradigms in rodent models show fear conditioning leads to

consolidation in the LA, while extinction learning involves the medial prefrontal

cortex (mPFC), the intercalated cells and the BLA (Hongjoo J Lee et al., 2016;

Onoue et al., 2014). Conflicting results are emerging within the literature regarding

these pathways and the involvement of the intercalated cells (Strobel, Marek, Gooch,

Sullivan, & Sah, 2015). Neural pathways used in extinction have been suggested, but

there is little consistency in the paradigms used to elicit the extinction. It is possible

that temporal differences in the paradigms may contribute to the array of pathways

suggested.

Functional neuroimaging studies in humans have shown results implicating

the amygdala, prefrontal cortex, anterior cingulate and the hippocampus in the

anatomy of patients with PTSD (Ursano et al., 2009). These studies demonstrate how

acquisition and extinction of fear memories is analogous between human and rodent

models (Herry et al., 2010; LaBar, Gatenby, Gore, LeDoux, & Phelps, 1998) and

therefore animal models provide a useful translational model to highlight the

microanatomy of the neurobiological mechanisms of fear memory extinction

(Gunduz-Cinar et al., 2013; Milad & Quirk, 2012).

28

Summary

In conclusion, neuroplasticity, is a fundamental process in the formation of

memories and alteration of behaviours (Berlucchi & Buchtel, 2009). Furthermore,

utilizing a behavioural model of fear memory formation and extinction provides a

robust platform upon which to build, as investigation into this model has detailed the

neuronal circuits, neurochemicals and some of the molecular mechanisms involved.

Quantifying and visualising the densities of neurons involved in neuronal

plasticity is of clinical relevance as the relapse of pathological fear is a leading

clinical obstacle in the treatment of disorders such as anxiety, phobia and PTSD.

Although there have been many technological advancements to increase our

understanding of neuroplastic change we are yet to establish many successful

therapeutic interventions for physical or mental neurological disorders. Electrical

brain stimulation, cognitive and motor training and some neuropharmacological

therapeutics have been identified with limited success. Only by continued research

will we develop a greater understanding of neuroplasticity, improved methods of

quantifying plastic changes in the brain and identification of biomarkers central to

these changes.

1.2 Context of Research: The mechanisms and biomarkers underlying neuroplasticity associated with emotional behaviours

Fight, flight and freeze are mechanisms that evolved to assist survival by

associating an appropriate behaviour to a potential threat. Many of these mechanisms

are conserved between species and now form the basis of the stress response first

described in the 1920’s by development of threat or fear memory pathologies such as

those involved in PTSD, phobias and anxiety disorders (Bouton, Mineka, & Barlow,

29

2001; Cannon, 1935; Hinkle Jr, 1973). To adequately investigate the neuroplasticity

of emotional memories, these mechanisms must be examined.

Acute stress elicits the fight-or-flight response through the activation of

numerous neuropeptide-secreting systems. This mechanism is designed to be

engaged briefly to provide the necessary energy and resources required to respond to

life threatening situations. A hypothesis supported by many studies demonstrating

neural adaptations that vary between acute and chronic stress and cognitive function

(for review see (De Kloet, Joëls, & Holsboer, 2005)), suggests that sustained

activation of this system leads to pathophysiological changes in the brain.

Limbic brain structures such as the hippocampus, prefrontal cortex and

amygdala process incoming sensory information regarding potential threats and if

necessary, trigger the release of corticotropin-releasing hormone (CRH). This

activates the hypothalamic–pituitary–adrenal (HPA) axis to stimulate corticosteroid

secretion from the adrenal cortex. Corticosteroids act through glucocorticoid (A.

Jacques, Battle, & Johnson, 2016) and mineralocorticoid receptors (A. Jacques,

Johnson, & Battle, 2016). The mineralocorticoid receptor moderates initiation of the

stress response while the glucocorticoid receptor terminates it and promotes memory

storage in preparation for future adverse encounters (for further information see

Appendix B and C). Interestingly both receptors are involved in the metabolism of

energy storage, metabolism and appetite, environmentally influenced factors that

affect neuroplasticity. Long-term exposure to adrenal glucocorticoids may lead to

atrophy of neurons, or negative neuroplasticity in the PFC and hippocampus, similar

to the effects of chronic stress (Hu et al., 2016). This effect may be exacerbated

through epigenetics with elevated paternal glucocorticoid exposure having been

shown to modify anxiety traits in offspring (Short et al., 2016). Impairment of the

30

HPA axis negative feedback system and stimulation of the HPA axis have been

observed in over 50% of depressed patients, which highlights the influence this

endocrine system exerts mood disorders (Varghese & Brown, 2001).

Originally thought to be part of the natural progression through trauma

processing, PTSD is now known to involve abnormalities in the neurobiology of the

HPA axis (Yehuda, 2001). The HPA axis is an inhibitory loop involved in the

production and release of glucocorticoids. Glucocorticoids such as cortisol are

released into the blood stream as a normal homeostatic regulation throughout the

day. However, during the body’s response to stress, cortisol release from the adrenal

glands is increased. It binds to mineralocorticoid and glucocorticoid receptors in the

brain where it has been shown to enhance the consolidation of pathological fear

memories (Buchanan & Lovallo, 2001; McFarlane, Atchison, & Yehuda, 1997).

Disruption of the glucocorticoid receptors in the amygdala may play a role in the

development of pathological memories (Keller et al., 2017). Genomic studies

investigating the mechanisms involved are attempting to detect an endophenotype for

people at higher risk of developing PTSD and related disorders (D. Mehta & Binder,

2012).

Technological advancements in whole brain imaging (Gao et al., 2019),

calcium imaging (Mishne & Charles, 2019) and electrical stimulation (Tucker,

Anderson, & Luu, 2019) have significantly improved our understanding of

neuroplastic changes and our ability to visualise generalised plasticity in the brain,

particularly after physical trauma. However, visualisation of precise subregions of

neuronal reorganisation and correlation with behaviour requires precise methods

capable of identifying and quantifying this functional circuitry. Functional magnetic

resonance imaging and positron emission tomography investigations into PTSD,

31

revealed hypo-activation in the ventromedial prefrontal cortex, a region modulating

emotional memories (for review see (Etkin & Wager, 2007).

In addition to a need for improved methods of neuroplastic quantification there

is little known about the underlying interactions between signal transduction

pathways, the effect of negative environmental influences on neuroplasticity and the

subsequent behavioural adaptations. As an inability to suppress a fear reaction and

HPA axis dysregulation have been associated with amygdala over activation in

subjects with PTSD (Jovanovic et al., 2010), factors capable of influencing stress

induced plasticity to circuits modulating the HPA axis require further elucidation and

may provide insight into novel therapeutic targets.

Neurochemical influence: pMAPK, Arc, serotonin and glutamate.

Factors influencing neuroplasticity include genetics, environmental stimuli,

and individual behaviour. From the food we ingest, to learning from our social

interactions, the environment appears to have a great capacity to keep our brains in a

constant state of flux, but what mediates this state? Experience, or how we interact

with our environment, has been shown to generate the greatest plastic neural change

(Kerr, Cheng, & Jones, 2011).

Both the formation and storage of fear and extinction memories are reliant on

neural plasticity i.e. changes within the neuron to allow a greater or lesser number of

connections to form with surrounding neurons (Izquierdo & Medina, 1997;

Rosenzweig & Bennett, 1996). The extracellular signal-regulated kinases / mitogen-

activated protein kinase (ERK/MAPK) pathway (see figure 1.3) has been

investigated and identified as essential during fear memory formation and extinction

learning (Hadley C Bergstrom et al., 2011; Herry, Trifilieff, Micheau, Lüthi, &

Mons, 2006). Phosphorylated mitogen-activated protein kinase (pMAPK) has been

32

well established as playing a role in cell growth and differentiation (G. L. Johnson &

Lapadat, 2002; Lai et al., 2001) and defined as fundamental in the plasticity required

for learning and memory.

Reports also suggest activity-regulated cytoskeleton-associated protein

(Arc/Arg3.1), an immediate-early gene (IEG) and downstream marker within this

pathway, along with serotonin, an inhibitory neurotransmitter, are required for α-

amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor trafficking,

which regulates synaptic plasticity and LTP (Chowdhury et al., 2006; Lesch &

Waider, 2012; Plath et al., 2006). The excitatory neurotransmitter glutamate, released

from the presynaptic neuron binds to NMDA receptors on the postsynaptic

membrane causing the cascade of events that lead to this gene expression and protein

phosphorylation. (Bauer, Schafe, & LeDoux, 2002; Maren, 2011).

Figure 1-3 MAPK pathway to neuroplasticity.

33

Glutamate NMDA receptor activation permits the influx of calcium which activates Ras protein (not shown). This causes consequent activation of phosphorylated MAPK, which conducts the extracellular signal to the nucleus. In the nucleus, transcription factors such as cyclic AMP response element binding (CREB) protein trigger the rapid transcription of Arc mRNA. The synthesized Arc protein is conveyed to dendrites where it moderates α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPAR) regulation and F-actin expansion. Originally adapted from (Rodrigues, Schafe, & LeDoux, 2004).

Serotonin (5-HT), is a monoamine inhibitory neurotransmitter which plays a

role in mood, appetite, learning, memory and cellular metabolism (Hodges,

Cummings, & Pilowsky, 2018; McAllister-Williams, Ferrier, & Young, 1998).

Changes in the levels of 5-HT receptors present within the nervous system have been

noted under various environmental conditions. There are currently sixteen catalogued

5-HT receptor subtypes involved in mediation of either excitatory or inhibitory

neurotransmission. (Hodges et al., 2018). Chemical manipulation of the

serotonergic system has shown that 5-HT2A agonists increase fear learning while 5-

HT1A agonists impair the learning (for review see (Bauer, 2015)), but how they

impact the neuroplasticity of memory is not well understood.

Glutamate signalling also plays a role in processing stressful situations as

demonstrated by use of ketamine (N-methyl-D-aspartic acid receptor antagonist) to

distort emotional processing similar to that seen in patients suffering PTSD (Cortese

& Phan, 2005). As glutamate and 5-HT can be released from the same neuron a

possible relationship may exist between the release of serotonin and glutamate during

the recall of stressful memories.

Neurobiological influence

Recently research across diverse mental pathologies has revolved around

inflammatory conditions (Fourrier, Singhal, & Baune, 2019; N. Mehta, Li,

Woolwine, Haroon, & Felger, 2019; Niraula, Sheridan, & Godbout, 2019). In the

brain, microglia are the immune cells that respond to all pathophysiological events.

They also play a role in neuroplasticity within the healthy brain. Microglia serve as

34

surveillance cells which actively search the brain for unrequired or damaged

synapses, establish contact with that synapse and prune it in preparation for

phagocytic removal (Hong, Dissing-Olesen, & Stevens, 2016). In this manor they

contribute to homeostatic regulation of the brain and the HPA axis (Silverman,

Pearce, Biron, & Miller, 2005). It is widely acknowledged that microglia respond to

stressful environmental challenges, however, their role in the plasticity of fear

memory is as yet undiscovered.

With respect to memory formation, microglia facilitate learning-induced

plasticity in glutamatergic synapses (Parkhurst et al., 2013) through the secretion of

brain-derived neurotrophic factor (BDNF) and constantly undergo morphological

change whilst monitoring their microenvironment(Wake, Moorhouse, Jinno,

Kohsaka, & Nabekura, 2009). Microglia sense environmental change and may use

this information to modulate hippocampal wiring through the formation, elimination

and relocation of synapses, thereby impacting memory function. Microglial-derived

IL-1β has been directly implicated in normal hippocampus-dependent learning

(Williamson, Sholar, Mistry, Smith, & Bilbo, 2011).

Summary

It is well known that fear memories are encoded by neurons in the amygdala,

with the BLA indicated in the acquisition of fear memories (Hadley C Bergstrom,

Craig G McDonald, Smita Dey, Haying Tang, et al., 2013; Herry et al., 2010). The

BLA encompasses the lateral and basal nuclei of the amygdala, with direct and

indirect neuronal connections to the central nuclear group (Maren, 2011). The spatial

allocation, temporal divergence and stability of these neuronal populations within

these nuclei are as yet far from complete.

35

Most of the research implicating the mPFC in extinction learning has come

through lesions which attenuate extinction (Falls, Miserendino, & Davis, 1992).

Several theories have emerged suggesting the mPFC may inhibit the amygdala after

extinction training (Gregory J Quirk, Likhtik, Pelletier, & Paré, 2003; Rosenkranz,

Moore, & Grace, 2003) but while some studies show that extinction learning recruits

glutamate and NMDA receptors (Myers & Davis, 2007; Sotres-Bayon, Cain, &

LeDoux, 2006) others suggest the neuroplasticity may be moderated by γ-

aminobutyric acid transmission (Sotres-Bayon et al., 2006).

One of the most successful ways to study the neuroplasticity of emotion is to

target an emotion with a thoroughly investigated and known circuitry. Fear is such an

emotion as the circuitry involved in the formation, extinction and recall of fear

memories have been extensively researched and defined. The pathways involved

have been silenced through lesion studies (McGaugh et al., 1995), drug infusions

(Miserendino, Sananes, Melia, & Davis, 1990) optogenetics (Do-Monte, Quinones-

Laracuente, & Quirk, 2015) and most recently, chemogenetics (Marek et al., 2018a).

Fear conditioning itself is a well-defined method of creating associative memories

reliant on neuroplasticity for their formation and has been fundamental in the

development of cognitive behavioural therapies for disorders such as phobia, anxiety

and post-traumatic stress (Maren & Holmes, 2016).

Treatment regimens encompassing neuroplastic change have long been

suggested for patients suffering mental illness but are now also being incorporated

into treatment for neurological disorders such as dementia, malnutrition associated

disorders, chronic pain, stroke and traumatic brain injuries (Altman, Richards,

Goldberg, Frucht, & McCabe, 2013; DeFina et al., 2009; Dimyan & Cohen, 2011;

Kass, Kolko, & Wilfley, 2013; Sibille, Bartsch, Reddy, Fillingim, & Keil, 2016).

36

Therefore, investigation of neural substrates and neuroplastic change may facilitate

the development of increasingly effective pharmacotherapeutics.

1.3 Purposes

PTSD represents a form of emotional memory reliant on neuroplastic changes

within the brain and may occur following exposure to a life threatening, traumatic

event. It is defined by symptoms such as intrusive memories, anhedonia (inability to

feel happiness) and avoidance behaviours, which may manifest as permanent

personality alterations (Andreasen, 2011). PTSD is a memory-based disorder,

clinically defined as the presentation of behavioural symptoms three months post

trauma (Gray & Liotta, 2012). In order to create abiding treatments for anxiety

related disorders a thorough knowledge of neuroplasticity must be attained.

Hypothesis

As these neuroplastic adaptations require modulation through neurochemical

and neurobiological factors, it was hypothesised that a new method to map neuronal

density would answer fundamental questions about fear related learning and memory

and its recall. Further to the examination of functional neuronal populations, the role

of the brains immune cells (microglia), known to alter phenotype in stressful

situations, was investigated during memory recall. It was hypothesised that

phenotype changes would occur during the recall of emotional memories, suggesting

reactive microglia play an active cellular level role that may prove to be a therapeutic

target for improved learning techniques. Investigation into the role of specific

neurotransmitters (5-HT and glutamate) was undertaken to establish a molecular

level target for improved treatment of anxiety related disorders.

37

Aim 1

To partially address the research problem surrounding the neuroplasticity of

emotional memories, a new method to map and quantify neuroplastic changes was

developed. Density topography (or heatmaps) generated after several behavioural

paradigms could assist in predicting the amount of cellular change or neuroplasticity

that occurs as a result of a specific environmental influence. Utilizing the expression

of neuroplasticity markers pMAPK and Arc, topographic density maps of the

amygdala and prefrontal cortex were generated.

The initial project was designed to

a) develop an economical approach to statistically map molecular markers

of neuroplasticity in specific neuronal networks for the identification of

new functional micro regions within established nuclei subdivisions

b) be adaptable to all fields of neuroscience and suitable to study any brain

region that could be identified through an anatomical anchor

c) This topographic visualisation was used to distinguish differences in

location derived through varying the type of memory recalled and the

temporal recall of auditory fear memories in the amygdala and

prefrontal cortex

d) illustrate the density of pMAPK expressing neurons and the

microanatomy involved in the encoding process of extinction versus

non-extinction of fear memories in the amygdala and mPFC

e) visualise differential neuronal activation within specific prefrontal

cortex cell layers in auditory fear memory and extinction memory recall

38

f) evaluate amygdala sub region involvement in the neuroplastic creation

and storage of emotional memories such as pure, context-removed and

auditory fear memories

Aim 2

In an attempt to identify novel cellular mechanisms in the prefrontal

cortex, amygdala and hippocampus that may influence the neuroplasticity of fear

memory consolidation and extinction, investigations were conducted into possible

morphological changes of microglia. It is well established that microglia are involved

in the neuroinflammatory response triggered by exposure to psychological stress but

their role in fear is as yet undefined. Arc plays a role in the activity-dependent

neuroplasticity of dendrites and c-Fos is a commonly used marker of neuronal

activation. pCREB and BDNF expression are present in fear memory consolidation.

Brain derived neurotrophic factor (BDNF) is a neurotrophin that modulates neuronal

survival and differentiation and may be released by microglia. Phosphorylated

cyclic-AMP response element binding (pCREB) a marker of plasticity can be

induced by BDNF. The studies undertaken aimed to

a) establish levels of neuronal activity through the expression of the immediate

early genes c-Fos and Arc

b) determine the extent to which memory recall influences the reactivity of

microglia within the prefrontal cortex

c) define microglial phenotypic changes as a consequence of recent and remote

fear memory recall, extinction memory recall

d) investigate the possibility of microglial morphological changes occurring as a

result of fear memories progressing from short term to long term

39

e) observe any alterations to amygdala and hippocampus BDNF and pCREB as

a consequence of different contextual fear memories

f) evaluate the maintenance of two different contextual fear memories in the

prelimbic cortex as demonstrated by pMAPK and BDNF expression and

through the microglia number present and phenotype displayed

Aim 3

To further define possible therapeutic targets capable of altering

negative neuroplasticity, current literature was examined to determine the molecular

mechanisms involved. It is well documented that manipulations of the 5-HT system

are widely used to treat anxiety and phobias; the pMAPK pathway to neuroplasticity

relies on the release of glutamate from the presynaptic neuron; 5-HT neurons express

VGLUT3, a transporter that concentrates glutamate into synaptic vesicles in

preparation for exocytosis; imbalanced levels of astrocytes and microglia has led to a

reduction of serotonin and consequent overabundance of glutamate and deletion of

VGLUT3 has been found to increase anxiety-related behaviours in mice. As

serotonergic neurons and glutamate release are known to be pivotal in anxiety

disorders, involved in fear memory consolidation, induce AMPAR mediated synaptic

plasticity and play a role in the activation of microglia, it seemed pertinent to observe

their role in the neuroplasticity of fear memory recall. Investigations were designed

to

a) map brain region specific contributions of colocalized 5-HT and

glutamatergic inputs specific to serotonergic neurons

a) determine if neuroplastic changes, denoted by pMAPK activation and due to

temporally different varieties of fear recall, were modulated through 5-HT

expression in the presence of the vesicular glutamate transporter type 3

40

1.4 Significance and Scope:

As many as 1.4 million Australians suffer PTSD annually and the estimated

harm, personal, social and economic burden facilitate a dire need for effective

clinical treatments (Maren, 2001; Statistics, 2008). Behavioural therapies targeting

enhancement of positive neuroplasticity have long been used to treat mental illness

however, very few pharmacological interventions have been developed in the last 40

years, in part due to the lack of novel mechanisms identified during this period (Insel

et al., 2013). The extinction of pathological fear is central to the treatment of

phobias, PTSD and anxiety disorders and involves a learning process, which forms

the basis of exposure therapy.

Exposure therapy utilizes fear memory extinction procedures, but the high rate

of relapse is a leading dilemma (Bouton & Peck, 1989; R. Bryant et al., 2008). This

thesis builds on previous research, which has defined a microanatomy of fear

memory by identifying a micro-topography of fear memory encoding in the

amygdala (Hadley C Bergstrom, Craig G McDonald, Smita Dey, Gina M Fernandez,

& Luke R Johnson, 2013). Identifying the precise biological mechanism of fear

memory acquisition and extinction is of great clinical relevance as the relapse of

pathological fear is a leading clinical obstacle in the treatment of PTSD and anxiety

disorders. Advances in imaging technology are permitting a greater understanding of

brain region adaptability. Developing a method to quantify neuroplastic change in

microanatomical regions and defining factors, both neurochemical and

neurobiological, that contribute to negative neuroplastic change will enhance our

ability to locate therapeutic targets.

41

1.5 Thesis Outline

This thesis consists of 10 chapters, 5 appendices and a bibliography.

Chapter 1 consists of an introduction, discussing the research problem, the

background behind the research undertaken, the purpose and scope of the research

and the aims of the study.

Chapter 2 is a published methods paper which provides a detailed

documentation of the method developed to locate and quantify specific subregions

involved in neuroplasticity.

Chapter 3 is a published data manuscript utilising the developed method to

anatomically define subregions of the basolateral amygdala undergoing localised

neuroplastic change due to contextual and auditory fear.

Chapter 4 is a published data manuscript expanding our understanding of

the microanatomical subregions of the amygdala and prefrontal cortex involved in

recent and remote fear memory and the formation of a new extinction memory. In

this study the micro regions are mapped and statistically quantified, to provide

insight into the individual cell populations and layers involved in the neuroplasticity

of fear memory formation.

Chapter 5 is a data manuscript in revision. This chapter investigates the

effect of fear memory recall on the number and phenotype of microglia within the

prefrontal cortex. It expands on our knowledge of neurobiological factors that

influence neuroplasticity.

Chapter 6 is a data manuscript investigating morphological change in

microglia in the rat dorsal hippocampus as a result of contextual fear conditioning.

This manuscript is under review.

42

Chapter 7 is a data manuscript mapping the colocalisation of serotonin,

serotonin transporters and vesicular glutamate type 3 transporters. These

neurotransmitters are well known for playing a role in anxiety disorders. This

manuscript is currently in revision.

Chapter 8 is a data manuscript recently submitted which explores the

activation of serotonin and glutamate with regards to the neuroplasticity of fear

memory and extinction memory recall.

Chapter 9 draws conclusions from the preceding chapters to create a cohesive

discussion to improve knowledge surrounding the neuroplasticity of emotional

memories. It increases our understanding of the micro regions involved in the

consolidation and recall of fear memories. Furthermore it discusses the role of

neurobiological and neurochemical factors affecting negative neuroplastic

adaptations.

Appendicies:

Appendix A displays the tables of statistical quantification outlined in chapter

7: Axonal nonsegregation of the Vesicular Glutamate Transporter VGLUT3 within

serotonergic projections in the mouse forebrain.

Appendix B holds chapter 10 a review manuscript discussing the most recent

clinical and pre-clinical work involving the impact of sugar, a common

environmental factor and its complicated relationship with neural adaptation. It

discusses the neuroplastic changes within the reward system, the neurobiology and

the neurochemistry behind consumptive behaviours and how our emotions are

intricately linked to neural adaptations. This review has been published.

43

Appendix C contains chapter 11, a published book chapter detailing the

function of the glucocorticoid receptor and its role in regulation of the HPA axis.

Appendix D is chapter 12, a published book chapter discussing the

mineralocorticoid receptor, its function within the brain and its role in fear, learning

and memory.

Bibliography.

44

Functional Neuronal Topography: A Statistical

Approach to Micro Mapping Neuronal Location

This chapter comprises the following published article: Jacques, A., Wright, A., Chaaya, N., Overell, A., Bergstrom, H. C., McDonald, C., Battle, A.R., Johnson, L. R. Functional Neuronal Topography: A Statistical Approach to Micro Mapping Neuronal Location. Frontiers in Neural Circuits, Published: 16 October, 2018 https://doi.org/10.3389/fncir.2018.00084

The project documented in this chapter encompasses part of aim 1, to develop a

method of mapping neuroplasticity that would be economical, adaptable to all fields

of neuroscience, capable of quantifying neurons, fibres and puncta, suitable to any

brain region and widely accessible to the general scientific public.

45

Statement of Contribution of Co-Authors for Thesis by Published Paper

The authors listed below have certified that:

1. they meet the criteria for authorship in that they have participated in the conception, execution, orinterpretation, of at least that part of the publication in their field of expertise;

2. they take public responsibility for their part of the publication, except for the responsible author who accepts overall responsibility for the publication;

3. there are no other authors of the publication according to these criteria;

4. potential conflicts of interest have been disclosed to (a) granting bodies, (b) the editor or publisher of journals or other publications, and (c) the head of the responsible academic unit, and

5. they agree to the use of the publication in the student’s thesis and its publication on the QUT’s ePrints site consistent with any limitations set by publisher requirements.

In the case of chapter 2:

Chapter 2: Functional Neuronal Topography: A Statistical Approach to Micro Mapping Neuronal Location Publication status: Published

Contributor Statement of contribution*Angela Jacques Contributed to methods development, conducted behavioural and laboratory

experiments, analysed data and wrote and edited the manuscript, created the figures.

Alison Wright Assisted with writing and editing the manuscript, involved in the development of the technique.

Nicholas Chaaya Assisted with writing and editing the manuscript, assisted with data analysis, assisted with behavioural experiments.

Ann Overell Assisted with writing and editing the manuscript.

Hadley Bergstrom Involved in the conception and design of the project, development of the technique, assisted with writing and editing the manuscript.

Craig McDonald Involved in the conception and design of the project, development of the technique.

Andrew Battle Assisted with creation of figures and editing the manuscript.

Luke Johnson Development of the technique, conceptual design of the methods paper, assisted with behavioural experiments, assisted with writing and editing the manuscript.

QUT Verified

Signature

46

2.1 Abstract

In order to understand the relationship between neuronal organization and

behaviour, precise methods that identify and quantify functional cellular ensembles

are required. This is especially true in the quest to understand the mechanisms of

memory. Brain structures involved in memory formation and storage, as well as the

molecular determinates of memory are well-known, however, the microanatomy of

functional neuronal networks remain largely unidentified. We developed a novel

approach to statistically map molecular markers in neuronal networks through

quantitative topographic measurement. Brain nuclei and their subdivisions are well-

defined – our approach allows for the identification of new functional micro regions

within established subdivisions. A set of analytic methods relevant for measurement

of discrete neuronal data across a diverse range of brain subdivisions are presented.

We provide a methodology for the measurement and quantitative comparison of

functional micro- neural network activity based on immunohistochemical markers

matched across individual brains using micro-binning and heat mapping within brain

sub-nuclei. These techniques were applied to the measurement of different memory

traces, allowing for greater understanding of the functional encoding within sub-

nuclei and its behaviour mediated change. These approaches can be used to

understand other functional and behavioural questions, including sub-circuit

organization, normal memory function and the complexities of pathology. Precise

micro-mapping of functional neuronal topography provides essential data to decode

network activity underlying behaviour.

2.2 Introduction

Following Cajal’s identification of the neuron as the fundamental functional

unit of the nervous system (López-Muñoz, Boya, & Alamo, 2006), neuroscience has

47

endeavoured to understand how neurons operates in local groups (ensembles) and

distributed networks to bring about behaviour. In 1894, Cajal (Cajal, 1894) proposed

a theory that memory storage requires the formation of new connections between

neurons in the brain. How neurons and their thousands of synaptic connections act

together to encode a memory was first conceptualized by Donald Hebb (Hebb, 1949)

as neuronal ensembles that both spatially and temporally act together to encode a

component of the memory. Since these foundational anatomical and theoretical

works, newer studies involving fluorescent imaging and electron microscopy have

since provided growing evidence for the modification of neuronal synapses as a

result of information storage, now known as synaptic plasticity (Kandel, 2001; Korb

& Finkbeiner, 2011). Thus, at the sub-cellular level knowledge of mechanisms of

memory encoding is more established, in contrast at the neuronal ensemble level

memory encoding mechanisms are not yet understood. Some functional evidence for

Hebbian reverberatory networks connecting ensembles of neurons (Hebb, 1949) has

been identified in memory circuits (L. R. Johnson et al., 2008; L. R. Johnson,

Ledoux, & Doyere, 2009; Josselyn, Kohler, & Frankland, 2017). However, key

challenges in neuroscience remain around how neurons collectively undergo

plasticity in ensembles to encode memories and behaviours. Aspects of neural

ensemble activity has been demonstrated in Hippocampus (N. H. Nakamura et al.,

2010) and Caudate (Barbera et al., 2016) and in Amygdala (Davis & Reijmers, 2017;

L. R. Johnson et al., 2008; L. R. Johnson et al., 2009; Josselyn & Frankland, 2018;

Josselyn et al., 2017; Rogerson et al., 2014). A key challenge in the neuroscience of

memory is in identifying which neurons have been allocated to the memory trace and

which have not, while some progress has been made (H. C. Bergstrom, 2016; Hadley

C Bergstrom & Luke R Johnson, 2014; Hadley C Bergstrom, Craig G McDonald,

48

Smita Dey, Gina M Fernandez, et al., 2013; Hadley C Bergstrom, Craig G

McDonald, Smita Dey, Haying Tang, et al., 2013; H. C. Bergstrom, McDonald,

French, & Smith, 2008; Hadley C Bergstrom et al., 2011; P. W. Frankland &

Josselyn, 2015; Josselyn & Frankland, 2018; Mayford, 2014; Rogerson et al., 2014),

more details are needed. This aim can be aided by the development of methods and

approaches to help reliably identify and quantify systematic topographies of neurons

allocated to specific memory traces.

Here we developed a method for topographical analysis and measurement of

neurons allocated to memory traces. We have applied this method to study aspects of

the neurobiological encoding of fear memory. We termed this method “neuronal

topographic density mapping” and have devised it to identify and map the degree of

stability within a micro-topography of neurons encoding Pavlovian fear memory

across different animals undergoing fear memory acquisition or extinction. The

methods described in detail below were developed over multiple studies,

investigating the location and distribution of neurons activated in fear memory in

amygdala (Hadley C Bergstrom, Craig G McDonald, Smita Dey, Gina M Fernandez,

et al., 2013; Hadley C Bergstrom, Craig G McDonald, Smita Dey, Haying Tang, et

al., 2013; Hadley C Bergstrom et al., 2011; Haranhalli, Massie, Little, LeDoux, &

Johnson, 2007; Luke R Johnson et al., 2012; Joseph E LeDoux, Haranhalli, Massie,

Little, & Johnson, 2006). For illustrative purposes and to expand on the scope of

these techniques, here we employed a small data set drawn from the study of

activity-regulated cytoskeleton-associated protein (Arc) expression in prefrontal

cortex.

In our studies to date, we have investigated the micro-topography of memory

using Pavlovian fear conditioning. In Pavlovian or classical fear conditioning a mild

49

foot shock (unconditioned stimulus, US) is temporally paired with an auditory tone

or comparable visual stimuli (conditioned stimulus, CS) (Hadley C Bergstrom &

Luke R Johnson, 2014; Hadley C Bergstrom, Craig G McDonald, Smita Dey, Gina

M Fernandez, et al., 2013; Luke R Johnson et al., 2012). The animal learns to

associate the US with the CS and exhibits typical behaviours including freezing,

typical of fear / threat behaviour (described extensively by other authors; see Johnson

and colleagues (2012) (Luke R Johnson et al., 2012) for details). We measured

neurons expressing plasticity associated proteins identified by

immunocytochemistry. Other functional protein and RNA expression in neurons and

glia can also be used with this approach. We tested for differences in the localization

of neurons among the conditioned memory groups. We have provided a

methodological approach to produce topographic neuron data from brain within

precisely aligned anatomical regions. This approach enables investigation of the

topographic patterns of neurons expressing plasticity associated proteins in the

associative fear memory formation and its extinction. We propose that this method

can also be used in the reproduction of neuronal density maps with regard to many

forms of neuroscience data for example, drug treatments, stress and addiction or

neurodegenerative disorders.

Our methodological approach to neuron topography, described here, provides

useful advantages for localizing function across behavioural conditions. Other

analysis methods to measure topography also provide useful topographic data. For

example, Nakamura and colleagues (N. H. Nakamura et al., 2010) identified that

memory activated neurons formed small anatomical clusters in hippocampus during

place preference formation, which was identified using a cluster analysis approach.

Recent studies by Barbera et al., (2016) (Barbera et al., 2016) used measures of

50

neuronal clustering of medium spiny neurons to predict locomotive states of

behaviour in mice. They reported that behavioural decoding accuracy improved

using spatially distinct neural clusters over single neurons (Barbera et al., 2016).

The brain is a complex interconnection of billions of neurons and decoding

how information is processed and stored by this circuitry requires the ability to

capture specific neuronal populations undergoing plasticity as a result of specific

behaviours. In vivo techniques such as calcium imaging are valuable tools to observe

functional imaging of neuronal populations in awake behaving subjects (Castanares,

Stuart, & Daria, 2019; Ohki & Reid, 2014; Romano et al., 2017). Of particular note

is a method developed by Romano and associates, to analyse neuronal population

dynamics (Romano et al., 2017). However, this method is invasive and not

conducive to the study of multiple brain regions or spatially segregated regions and

sub regions.

Yongsoo Kim and associates used rodent behaviour to develop a spatial IEG-

based mapping technique as a method to view whole-brain activity (Y. Kim et al.,

2015; Y. Kim et al., 2017). Furthermore, whole brain mapping methods have been

developed by Vousden and colleagues and Renier and colleagues (Renier et al.,

2016; Vousden et al., 2015). Each of these in vitro methods provides the advantage

of visualizing patterns of neural activity across brain regions to map distributed brain

networks and could be utilized to generate structure-function hypotheses prior to the

mapping of sub regions and micro circuits involved. The creation of neuronal

topographic density maps, as described here, can be used for a variety of studies to

pinpoint functional microcircuits in the brain.

Using our approach to mapping and measuring topography we have

characterized the microanatomy and topography of neurons involved in different

51

phases of memory, consolidation, reconsolidation and extinction (Hadley C

Bergstrom & Luke R Johnson, 2014; Hadley C Bergstrom, Craig G McDonald,

Smita Dey, Gina M Fernandez, et al., 2013; Hadley C Bergstrom, Craig G

McDonald, Smita Dey, Haying Tang, et al., 2013; H. C. Bergstrom et al., 2008;

Hadley C Bergstrom et al., 2011; N Chaaya, Jacques, Battle, & Johnson, 2017;

Haranhalli et al., 2007; A. Jacques, Chaaya, Battle, & Johnson, 2017; Joseph E

LeDoux et al., 2006). These data have the potential to pinpoint neuronal topography

patterns underlying memory encoding in the mammalian brain in normal and

pathological situations (Luke R Johnson et al., 2012) and thereby facilitate current

treatments for pathological memory disorders (Luke R Johnson et al., 2012). The

generation of neuronal topographic density maps can be used to define and measure

memory allocation within the brain.

Throughout this methodological report we provide details of the rationale,

procedures and equipment needed to produce and analyse topographic neuronal data.

In addition, within each methodological section we provide ‘examples’ from our own

data in order to illustrate how the methods can be applied and used. The

methodological approaches we describe here have wide applications for

understanding and measuring neuronal topography. Applications, include measuring

the topography of neurons encoding different types of memory, different sensory

stimuli and motor behaviours.

2.3 Step-By-Step Methods

2.3.1 Data collection: Behavioural, tissue and neuron analysis in preparation for topographic investigation

52

Run behavioural models

Pavlovian fear conditioning forms associative memories. Synaptic plasticity,

dependent upon phosphorylation of extracellular signal-regulated kinase (pMAPK)

has been established as critical in the formation of these memories in the lateral

amygdala (LA) and medial prefrontal cortex (mPFC) (J. E. LeDoux, 2000).

Example: The sample data set consisted of fear conditioned adult male

Sprague-Dawley rats (RRID:RGD_5508397) (n=40) that underwent behavioural

procedures in standard Pavlovian fear conditioning chambers (Coulbourn

Instruments, Allentown, PA, USA) (see fig. 1A). The unconditioned stimulus, a 0.6

mA foot shock with duration of 500 ms, was paired with the conditioned stimulus, a

tone of 5 kHz and 75 dB (Digitech Professional Sound Level Meter,

https://www.jaycar.com.au/pro-sound-level-meter-with-calibrator/p/QM1592), 20 s

in duration, to produce an associative memory. Three pairings were presented with

an average 180 s inter-trial interval with total time in box of 10 min. Standard

conditioning and behavioural testing procedures were followed (Hadley C Bergstrom

& Luke R Johnson, 2014; Hadley C Bergstrom, Craig G McDonald, Smita Dey, Gina

M Fernandez, et al., 2013; Hadley C Bergstrom, Craig G McDonald, Smita Dey,

Haying Tang, et al., 2013; H. C. Bergstrom et al., 2008; Hadley C Bergstrom et al.,

2011; N Chaaya et al., 2017; Haranhalli et al., 2007; A. Jacques et al., 2017; Joseph

E LeDoux et al., 2006). The experimenter was blind to the experimental conditions

when scoring freezing behaviour, which was defined as a lack of movement except

that required for respiration (J. E. LeDoux, J. Iwata, P. Cicchetti, & D. Reis, 1988).

Next, brains were prepared for histological analysis and measurement.

53

Figure 2-1: Steps for tissue sampling and measurement from behavioural data.

(A) Run behavioural models. Any expression of a chosen behaviour can be used as a model. In our example we have used Auditory Pavlovian fear conditioning. Behavioural testing was conducted with adult male Sprague-Dawley rats in acoustic classical fear conditioning chambers. A 0.6 mA foot shock with duration of 500 ms was paired with a tone of 5 kHz and 75 dB, 20 s in duration to produce an associative fear memory. (B) Perform immunocytochemistry. Avidin–biotin peroxidase complex method is demonstrated here. Sections from the lateral amygdala (LA) were labelled for Arc, scanned using a slide scanner and cropped at 2x magnification. Enlarged inset square shows Arc+ neurons in the dorsolateral portion of the LA at 20x magnification. Inverted gray scale images of fluorescent immunocytochemistry would also be suitable. (C) Choose suitable anatomical marker to be used as an anchor. The caudate putamen and lateral ventricle are two examples of anatomical landmarks that we have used previously, and can be differentiated in serial sections for section alignment by Feret length within the ventricle or between anatomical landmarks. Photomicrographs show three consecutive 60

54

µm sections across the rostrocaudal axis of the rat brain, depicting 2.76, 2.70, and 2.64 mm anterior to Bregma in the medial prefrontal cortex (mPFC). Feret diameter is shown – red arrow. Brain sections at Bregma coordinates –3.32, –3.36, and –3.40 mm posterior from Bregma were used to align the LA (Source: see Bergstrom et al., 2011). The maximum Feret length of the caudate putamen in the prefrontal cortex was shown to be statistically different across Bregma coordinates, animals and conditions. (D) Establish section alignment. The Rat Brain Atlas (Paxinos and Watson, 2007) is an important tool to assist alignment of sections. Schematic diagrams are shown depicting the regions of interest. The dorsolateral portion of the lateral amygdala (LAd), the ventromedial portion of the lateral amygdala (LAvm) and the ventrolateral portion of the amygdala (LAvl) are shown in three serial sections caudal from bregma –3.36 mm. The prelimbic (PL) and infralimbic (IL) cortex are represented by three serial sections caudal from bregma 2.52 mm. Brain Atlas diagrams are adapted from Paxinos and Watson (2007).

Perform immunohistochemistry

Rats were transcardially perfused and brains were post-fixed in 4 % PFA

overnight then stored in 0.1 M phosphate buffered saline. Free-floating serial coronal

sections (40 μm) of the medial prefrontal cortex and amygdala were prepared using a

vibratome (M11000; Pelco easiSlicer, Ted Pella Inc, Redding, CA, USA). Sections

from the lateral amygdala and prefrontal cortex were labelled for pMAPK and Arc

activation using the avidin-biotin peroxidase method. Detailed immunocytochemical

methods can be obtained from our previous reports (see (Hadley C Bergstrom, Craig

G McDonald, Smita Dey, Haying Tang, et al., 2013; Hadley C Bergstrom et al.,

2011)). Slides were scanned with an Olympus VS120 slide scanner and cropped at

2x magnification. (see fig. 1B)

Choose anatomical marker

Establishing anatomical alignment between regions of interest (ROI) is

necessary for visual comparison of neuron density in neural images, for sectioning

the ROI into micro regions for analysis, and for both quantitative and visual analysis

of the data. Therefore, choosing an appropriate anatomical anchor is a key step. The

anchor point should: 1) be a readily visible anatomical feature that is close in

proximity to the ROI, 2) be stable across subjects and conditions, and 3) change

55

shape rapidly and distinctly as the viewing plane changes, so that different planes of

view can be discriminated clearly. These characteristics are identifiable

microscopically and importantly can also be quantified (see fig. 1C) .

Example: The amygdala and medial prefrontal cortex (mPFC) have been

implicated in Pavlovian fear conditioning (Michael S. Fanselow & Gale, 2003; Luke

R Johnson et al., 2012; H. J. Lee, Haberman, Roquet, & Monfils, 2015). In a series of

studies, we have focused on the amygdala and have used the opening of the Lateral

Vertical (LV) as an anatomical anchor (Hadley C Bergstrom & Luke R Johnson,

2014; Hadley C Bergstrom, Craig G McDonald, Smita Dey, Gina M Fernandez, et

al., 2013; Hadley C Bergstrom, Craig G McDonald, Smita Dey, Haying Tang, et al.,

2013; H. C. Bergstrom et al., 2008; Hadley C Bergstrom et al., 2011; N Chaaya et

al., 2017; Haranhalli et al., 2007; A. Jacques et al., 2017; Joseph E LeDoux et al.,

2006). The LV has proved a useful structure for the purpose because it meets the

criteria outlined above: 1) the LV is close in proximity to the amygdala, 2) the LV

changes rapidly in size along the longitudinal plane, 3) the LV is a stable anatomical

feature and 4) LV changes can be seen clearly, and measured, through the sequence

of planes on which the brains were sectioned, enabling quantitative analysis of the

changes section by section. In order to further demonstrate and measure the

properties of the LV for landmark suitability, in addition to histological

measurements, we made measurements of the LV with MRI. Here, the

morphological properties of the LV, including its increase in diameter along the

rostral-caudal axis, were confirmed in vivo, using 3-dimensional T2-weighted MRI

to quantify its area (Hadley C Bergstrom, Craig G McDonald, Smita Dey, Haying

Tang, et al., 2013). This rapid change from rostral to caudal allows for precise

quantitative section alignment from plane to plane. In our histological studies the

56

morphology of the LV was reconstructed from five consecutive planes (Bregma -

3.36 to -3.48). The coronal plane with the least variance between conditions was

found at Bregma -3.36 in the rat (Paxinos & Watson, 2007), the entrance of the LV,

so this was chosen as the most suitable anatomical anchor, in addition, it could be

readily visualized and measured. At -3.36 mm Bregma, in addition to the LV it is

also possible to identify the major anatomical structures of the ROI (the subnuclei of

the lateral amygdala (LA). The choice of the LV as an anatomical anchor was

therefore suitable because it is amygdala-centric, changes shape rapidly and clearly,

and is stable across subjects (Hadley C Bergstrom & Luke R Johnson, 2014; Hadley

C Bergstrom, Craig G McDonald, Smita Dey, Gina M Fernandez, et al., 2013;

Hadley C Bergstrom, Craig G McDonald, Smita Dey, Haying Tang, et al., 2013; H.

C. Bergstrom et al., 2008; Hadley C Bergstrom et al., 2011; N Chaaya et al., 2017;

Haranhalli et al., 2007; A. Jacques et al., 2017; Joseph E LeDoux et al., 2006).

We used the caudate putamen as an anatomical landmark to align sections in

the prefrontal cortex (described below). Aspects of the caudate putamen met the

criteria we previously set for landmark identification (see fig. 1D). Histological

images were captured as virtual slide images (OlyVia; format.vsi) using a slide

scanner (Olympus VS120). Capturing images with a slide scanner (used in this

example) is an alternative approach to live capturing of neuron data with a

microscope connected directly to Neurolucida as used in our previous published data

(Hadley C Bergstrom & Luke R Johnson, 2014; Hadley C Bergstrom, Craig G

McDonald, Smita Dey, Gina M Fernandez, et al., 2013; Hadley C Bergstrom, Craig

G McDonald, Smita Dey, Haying Tang, et al., 2013; H. C. Bergstrom et al., 2008;

Hadley C Bergstrom et al., 2011; N Chaaya et al., 2017; Haranhalli et al., 2007; A.

Jacques et al., 2017; Joseph E LeDoux et al., 2006). In this example, we used

57

OlyVIA XV Image Viewer (Olympus Australia Pty Ltd, Vic, RRID: SCR_014342)

to ascertain and measure images within a Bregma range that showed an alteration in

the size of the caudate putamen. The caudate putamen becomes visible 2.7 mm

anterior to Bregma, distinctly widens and lengthens in serial coronal sections across

the rostrocaudal axis. Three consecutive sections (Bregma 2.7 mm – 2.58 mm) were

aligned and verified across subjects and conditions by statistical comparison

(ANOVA) of the feret length (the maximum feret length or distance between two

perpendicular tangents) was measured with Neurolucida 360 software (Neurolucida,

MBF BioScience, VT, RRID:SCR_001775) and analysed with SPSS, (IBM SPSS

Statistics 23, WA, SCR_002865). A similar comparison of sections was calculated

using z-scores from each maximum feret measurement of the caudate putamen. No

outliers were detected using +/- 3.0 standard deviation (SD). This principle includes

99.9 % of values coming from the same normal distribution). Additionally, outliers

can also be checked using online software tools, e.g. GraphPad Prism. Next, in order

to test each Bregma point assignment was dissimilar and no difference existed

between experimental conditions, paired t-tests were performed on the feret

measures. Each distance was found to be statistically different (example 2.76 mm

Bregma; p = 0.000304). This data was used to help exclude misaligned sections due

to natural or histological induced variations. This quantitative analysis approach can

thus be used to assign sections to distinct groups maximizing alignment accuracy for

subsequent neuronal topography measures.

Section alignment

Quantitative topographical data was produced beginning with neuron

identification and section alignment. While LV and caudate putamen changes can be

observed through a sequence of many planes, the region of interest may be rostral or

58

caudal to this point. For this reason, the chosen landmark is used only as a point of

reference. Sections are aligned manually using the landmark and working rostrally or

caudally through the Bregmas using the measurement of width of each section as a

guide. For example, Bregma 2.76mm is 0.48mm away from Bregma 3.24mm;

therefore there will be 8 x 60µm sections or 12 x 40µm sections between the two

Bregma coordinates. This highlights the need for precision when slicing and marking

serial sections. Having mounted sections in the correct order on slides prior to

labelling decreases time taken during this stage.

2.3.2 Generate topography in preparation for analysis

Create contour

In order to ensure consistency and precision in neuron counting across all

subjects, a contour or tracing of the anatomical structure being investigated can be

prepared in Neurolucida (NL) 360 (Neurolucida, MBF Bioscience, VT). Prior to

importing an image into NL for tracing, it is necessary to calibrate the image to

approximate the dimensions of a single brain section bitmap image (cellSens

software, Olympus, Vic, RRID: SCR_014551). Within Neurolucida select >File,

>Image open to allow the image to appear and select x and y calibration pixel size.

These measurements are located in the image properties section in the cellSens

program. Choose >Trace, >Contour Mapping in NL to begin the trace (see fig. 2A).

The image lines may be enlarged using the zoom tool, to increase accuracy of the

trace. Use the curser to trace around the selected area and >Close Contour when

finished each area. This allows delineation of each section of the contour with a

separate colour using ‘User Line’.

59

Scale contour

At this point it is essential to align the contour. The size of the tracing can be

adjusted to fit the image using >Tools, >Adjust Scaling. Contour alignment must be

consistent across all groups, prior to neuron counting. It is advisable to open several

images to scale the contour, due to minor variation in dimensions across subjects.

Calibrate contour

Very importantly, the contour is then calibrated to a constant point (0, 0 on the

x, y axis) to preserve consistency of neuron marker coordinates. The reference point

is displayed by selecting >Options, >Display Preferences, >View. In this window,

the radius of the point can be set to a desired diameter. Apply the display grid setting

and enlarge with the magnification tools as required. The contour is moved (using

move tools) such that the 0,0 coordinates are placed in the superior left corner of the

contour. Once in position the contour must not move or be resized for the duration of

neuron counting across all groups to ensure the integrity of the quantitative data.

Save contour as a data file.

60

Figure 2-2: Steps for producing raw coordinate data from identified neurons.

(A) Create, scale, and calibrate contour. Neurolucida 360 (or an equivalent program) can be used to produce a nucleus or brain region contour from a rat brain atlas diagram. Using the contour mapping tool in Neurolucida 360 contours (in different colours) can be traced over a figure from an atlas. Lateral amygdala tracing shown was generated from Bregma –3.36 of Rat Atlas (Paxinos and Watson, 2007). (B) Align section to contour and mark immuno positive neurons. Prefrontal cortex section with contour overlaid. Immuno-positive neurons were marked within the contour. Saved data files can be opened in Neurolucida Explorer to gain data file information such as contour areas, Feret length measures, and neuron counts. Prelimbic contour and neurons were marked in aqua, infralimbic contour, and neurons marked in yellow. Once neurons are marked, Neurolucida Explorer (or equivalent) can import the data file to generate a contour and marker analysis, LA example shown. (C) Export marker coordinates. The x, y coordinates produced for each marked neuron are exported to an ASCII file which can be opened in graphing software such as Origin Pro (or equivalent). (D) Produce bin matrix. A data matrix is generated based on the area and density of marked neurons within the contour. Bin size is calculated using twice the area of the contour divided by the total number of neurons (De Smith et al., 2009). Once the x, y coordinates are highlighted in an Origin Pro (or equivalent) workbook, the 2D binning option under descriptive statistics is chosen. The bin ends and size can be manually entered into the dialog box once determined using the standard geospatial formula (De Smith et al., 2009).

Align sections to contour

Once the tracing has been saved >CTRL+S, a scanned and cropped image of a

single neural section may be opened (>File, >Image Open, >calibrate pixel size) and

the tracing can be overlaid using the move tools to move only the image. There may

be some minor variation in the size and properties of each subject, driven by natural

61

variation or variations introduced during tissue processing - therefore the contour

must be aligned to each section. To align the section and the contour, select >Image,

>Image Processing and >Orientation (see fig. 2B). Options are provided for a mirror

image, flip, 90° or 180° rotation of the image. Choose Arbitrary Rotation and use the

arrows to alter the Rotation in Degrees.

Mark immuno positive neurons

Once the section is aligned to the contour (or tracing), begin to mark neurons

by choosing a marker from the marker toolbar located down the length of the left

side of the screen. Right click the mouse button on the selected marker to rename,

recolour or resize the marker. Elect to use a different colour for markers in separate

areas of the contour for ease of analysis at later stages of the process (see fig. 2B).

Markers may be erased at any time during counting by >CTRL Z, or >Edit, >Undo,

to remove the last placed marker. If mapping to determine the organization of

synaptic connection strength the same procedure should be followed for marking

puncta. To map the density of dendrites and axons in the neuropile the coordinate

markers must be placed along their length.

Note: If mapping neurons using NeuroLucida directly connected to a

microscope for live imaging, then, following contour tracing and neuron mapping, a

final alignment of all data to be compared must be made before analysis of neuron

spatial distribution. Contours with mapped neurons are rotated for matched

alignment using the Neurolucida Contour Alignment function.

Example: A digital image of the ROI, the mPFC, was sourced from the rat

brain atlas, 6th edition, 2007 (Paxinos and Watson, 2007, RRID: SCR_006369).

Three locations, 3.3 mm, 3.24 mm and 3.18 mm anterior to Bregma (Paxinos &

62

Watson, 2007) were used for cell counting. This level was chosen as both the

prelimbic and infralimbic cortices were represented at this point. Specific markers

were recoloured and renamed for each subregion to be mapped (fig. 2B).

Export Neurolucida ASCII File into OriginPro (or alternative)

Once all the neurons in the ROI are counted with the aligned contours, the

marker coordinates (x,y,z), which Neurolucida has recorded relative to the nominated

reference point, can be exported as an ASCII (plain text) file (see fig 1C). To

accomplish this, select >File, >Export Marker Coordinates and save the file. At this

point it is also prudent to save the data file you have placed your makers on, by

choosing >File, >Save Data File As. The Data file can be opened in Neurolucida

Explorer >File, >open data file, >contour, >analysis, >markers and region analysis.

This program provides a full synopsis of the contour areas, required for later

mapping, perimeters, feret measures and neuron counts for each designated region.

Once this information has been saved the neuron markers can be cleared in NL 360

using >Edit, >Select Objects. A window will open to the right of the screen where

you can select Any Object, Only Markers, Select All, then press the Delete key.

Choose >File, >Image Open to import a new section and begin the entire sequence

again. Once two or more images are open, select >Image, >Image Organizer, to

choose which images you will Show, Hide or Delete. Files can also be closed by

selecting >File, >Close All Images. To analyse the data obtained the ASCII files can

be opened in Microsoft Excel where the x and y coordinates are quickly accessed and

can be cut and pasted into Origin Pro (see fig. 1C)

(http://www.scientificcomputing.com/product-release/2014/10/origin-and-originpro-

2015-data-analysis-and-graphing-software). Alternatively Origin Pro has the facility

63

to open all files at once by choosing >File, >Import, >Multiple ASCII, and following

the prompts to choose the files you wish to include in one density map. It is

recommended to import only files from one behavioural condition at a time to reduce

human error. Once coordinates are listed, select >Descriptive Statistics, >2D

Frequency Binning, which will require input of bin sizes. (Alternatives to Origin Pro

can also be used – see Discussion below).

Select binned data parameters within Origin Pro (or alternative)

Data binning, also known as discretization, involves grouping data into bins in

order to ascertain a quantitative understanding of neuronal distribution (Kerber,

1992). Developing an appropriate data matrix relies on the optimization of the

dimensions of micro regions of data (bins). This part of the analysis should be well

considered and standardized in order to closely match the bin number and

dimensions with the central experimental question being investigated and also to

ensure the repeatability across subjects and experiments. The number of bins can be

determined based on experimenter determined parameters or alternatively a formula

can be applied to standardize the selection on bin numbers and to reduce any bias in

bin number selection. An established formula for this type of spatial analysis is based

on twice the expected frequency of items identified in a random field (2*sampling

area/n, where n = mean number of items to be counted, e.g. activated neurons)

(Michael J De Smith, Goodchild, & Longley, 2009). This method can be used to

ensure an unbiased estimate of the optimal dimension of bins for sectioning the ROI

into a matrix for data analysis. The neuron counts, and contour area measurements

are obtained from the Neurolucida Explorer data. Once bin number has been

calculated, the minimum bin beginning and maximum bin end for the x axis and y

64

axis are adjusted to encompass the smallest and largest coordinates contained within

the ASCII files. In Origin Pro, all Auto windows must be unchecked to allow manual

input of data. The bin size is measured in micrometres squared (m2). Once these

measurements have been entered and the number of bins is calculated by the

program, select >OK (see fig. 2D). This converts the data into an appropriate matrix,

based on the area and density of the marked objects.

Produce bin matrix

The next step is to use the data from the calculated matrix of bins and their

corresponding neuron counts for graphing and statistical analysis. The table of bins

and neurons counts derived from Origin Pro (see fig. 2D) can now be copied into an

Excel spreadsheet (or equivalent program). Repeat this process for each ASCII file

obtained from one section, in one condition across all animals – this will be based on

the section alignment for a specific ‘Bregma” coordinate – as described above. For

validation purposes individual density maps can be produced at this point, for later

comparison to the mean map. For an example see a range of 26 maps produced from

raw values for each subject across 4 experimental conditions in comparison to mean

maps in Figure 2 of (Hadley C Bergstrom, Craig G McDonald, Smita Dey, Gina M

Fernandez, et al., 2013).

2.3.3 Topographic neuronal density maps (heat maps) and analysis

Create density maps

Using Excel, an average across all sheets can then be calculated – this is used

to plot a graph of the mean for an experimental condition (see fig. 3A). In addition,

65

from these combined and averaged data a coefficient of variance (CV) and other

measures can be calculated. The mean and CV data can be used to create separate

neuron topographic density ‘heat’ maps using graphing software SigmaPlot or

OriginPro (SigmaPlot v 12.5, Systat Software, San Jose, CA, RRID:SCR_003210)

(or alternatives). For producing a variety of graphs from the now binned data we

have used SigmaPlot, however other programs can be used. The data matrix, using

individual subject data or averaged data from Excel, is transferred beginning in the

third column of SigmaPlot. The x and y coordinates from Origin Pro are copied into

columns one and two of Sigma Plot. In order to produce a coloured neuron

topographic density ‘heat’ map, select >Create Graph, >Contour Tool (see Fig.3A).

The scale can be adjusted using the graph properties tool. The production of a

neuronal topographic density ‘heat’ map is also possible using Origin Pro.

Example: We have used bin matrix data from neurons identified and marked

in the prelimbic and infralimbic cortices and transferred this data to SigmaPlot. This

data was used to produce both prelimbic (PL) and infralimbic (IL) mean neuron

topographic density graph (heat maps). As described above, during the creation and

alignment of the contour the 0, 0 coordinate was aligned to the superior left corner of

the contour. The creation of an overlay was performed by aligning this same superior

left landmark of the contour with the 0,0 coordinates as displayed on the SigmaPlot

contour graph export. This process allowed aligned or registered heat maps from

different animals to be combined into signed maps of mean data for initial qualitative

analysis of the data sets. In our example we identified neurons activated during the

recall of an extinguished fear memory – initial qualitative analysis of this data

reveals increased neuron density within the deep layers of the PL and IL.

66

Figure 2-3: Steps for producing and analysing topographical density maps.

(A) Create topographic density map. A neuronal topographic density mean map is produced by transferring binned data from Excel to Sigma Plot (or equivalent software) (X data = x coordinates, Y data = y coordinates). Density maps can be created for each sub region. A coefficient of variance map can be prepared by dividing the standard deviation by the mean across all samples in one condition. Difference maps can also be created between conditions. The data matrix from Origin Pro (or equivalent) is transferred to a spreadsheet. This procedure is followed for each animal from a single condition/group. An average across all sheets produces the data for a mean density map. The standard deviation is calculated and divided by the mean, producing the data required for the coefficient of variance (CV) map. Example – topographic density (mean and CV) maps shown for Bregma –3.36, pMAPKC neurons in the ventrolateral portion of the LA of rats that underwent extinction training (n = 7). (B) Align density map with contour and brain sections. To enhance visualization of specific neuronal subsets, density maps can be inserted into the contours or superimposed over brain sections. Density maps may be edited to change the styles, colours, font sizes, labels etc., providing alternatives conducive to individual requirements. Information regarding cell layers can be determined from visualizing the distribution of activated neurons as shown in the pMAPK labelling of the mPFC of rats that have undergone auditory fear conditioning (n = 7): mean map generated in Sigma Plot (or equivalent), map placed into contour, map overlaid on rat brain section. (C) Quantitative analysis of variance between conditions. A variety of statistical analysis can be performed to compare binned data such as Bonferroni correction, principal component analysis (PCA), false discovery rate (FDR), multiple discriminant analysis and mixed model ANOVA. Example of mean maps for the expression of pMAPK in the LA provides visual comparison between auditory fear conditioned (n = 6) and naïve (n = 7) rats. pMAPK ranks comparing extinction (n = 7) and no extinction (n = 5) groups within the ventrolateral portion of the LA p = 0.0022 (t-test, Mann–Whitney rank and SEM)

Align maps with contours and sections

We recommend two methods to enhance visualization of specific neuronal

subsets and gain visual information regarding distribution of activated neurons with

regards to cell layer. The density maps can be inserted into the contours generated

67

from an atlas, or superimposed over the original brain sections (see fig. 3B). To

ensure ease of fit it is prudent to place a marker in the corner of each contour which

can be removed prior to statistical analysis. Density maps may be edited in Sigma

Plot to change the styles, colours, font sizes, labels etc., providing alternatives

conducive to individual requirements.

Example: Information regarding cell layers can be determined from

visualising the distribution of activated neurons as shown in the pMAPK labelling of

the medial prefrontal cortex (see fig. 3B) of rats that have undergone auditory fear

conditioning (n=7): mean map generated in Sigma Plot (Systat Software). The map

was placed into the prefrontal contour and overlaid onto a rat brain section.

Analysis of binned data

Graphing topographic neuron density data is an important step to provide

visual evidence for changes in topography associated with behavioural and other

experimental manipulations, as described above. However, when further evidence is

needed to support conclusions of changes to neuronal topographic patterns then

statistical analysis of the topographic data is required. Quantitative analysis can be

performed with a variety of methods (discussed below) to compare the coefficient of

variance and topographical differences between conditions. GraphPad Prism 7

(GraphPad Software Co., CA, USA) can be utilized for each of the discussed

methods as well as linear regression and Pearson’s r coefficient which can also be

collected for correlation between groups.

Example: To evaluate the bins in each data matrix, two-way ANOVA with a

false discovery rate correction for multiple comparisons was conducted. The

discovered bins were termed micro regions of interest (MORIs) and assigned a

68

colour to represent the density of neuronal cell bodies located in that position (see

fig. 3C). Post hoc analysis of MROIs was conducted using t-tests.

2.4 Statistical Analysis of topographic neuron density data

In the next section, we describe statistical methods than can be applied to

binned data sets of topographic data combined with behavioural manipulations to

groups of experimental and control subjects. We also provide examples of

application of statistical analysis from our own behavioural and neuronal topography

data sets. The major challenge with the statistical analysis of multiple topographical

binned data sets, combined with several experimental groups, is statistical error due

to multiple comparisons. In order to best handle the analysis of topographical data

we have investigated and utilized a variety of statistical approaches for large multiple

comparison data sets – these include ANOVA and its variants; principle components

analysis (PCA); and false discovery rate (FDR) correction (see table 1). A very

important step in performing statistical analysis of topographic data is to perform the

statistical analysis in very close consultation with the Data produced from the

topographic maps as described above. Through careful observation and consultation

of the heat maps, derived from both individual animals and importantly behavioural

group mean heat maps together with their measures of variance (CV maps), the most

meaningful analyses can be performed and interpreted.

2.4.1 ANOVA followed by Bonferroni corrected t-tests

A question addressed in topographic data analysis is whether there is a

significant difference in the data (e.g. number of activated neurons in the ROI) across

all experimental conditions and in all ROI. One way to assess the overall difference

69

in experimental manipulation is with analysis of variance (ANOVA), followed by a

post hoc t-test with a correction for multiple comparisons (e.g. Bonferroni), among

specific ROI and experimental groups to determine where the significance arises.

Where multiple comparisons are necessary, a Bonferroni-type correction may be

employed (see use in (Hadley C Bergstrom et al., 2011), however it has the risk of

being too strict and likely to sacrifice power in the attempt to exert stringent control

over error. The potential for false negatives (type II errors) can be controlled

effectively, while still retaining sufficient power, with false discovery rate (FDR)

correction (Benjamini & Hochberg, 1995).

Example: We have analysed topographic neuron density data from Pavlovian

fear conditioning experiments in order to determine whether there was a significant

differences in topographic neuron density data across conditions by comparison of

activated neuron density in each of the micro ROIs (46 bins) across all conditions via

multiple comparisons (Hadley C Bergstrom, Craig G McDonald, Smita Dey, Gina M

Fernandez, et al., 2013). The mean numbers of activated neurons identified in the

ROI from topographic data were used to conduct ANOVA across all conditions.

Where a significant difference was found, planned contrasts between experimental

and control groups were performed to assess where the differences lay (Hadley C

Bergstrom, Craig G McDonald, Smita Dey, Gina M Fernandez, et al., 2013).

Multiple comparison tests involved three contrasts using one-way ANOVA. The first

compared the fear conditioned and conditioned stimulus reactivated groups to the

control groups: In this example, we compared box alone and conditioned stimulus

(memory not reactivated groups). The second contrast was between the fear

conditioned and conditioned stimulus reactivated groups and the third compared the

box alone to the conditioned stimulus group. Having established a significant

70

difference across conditions and located the main effect between experimental and

control conditions, the next step was to locate the region of greatest variance in the

ROI, requiring assessment of the differences in micro ROIs between groups (Hadley

C Bergstrom, Craig G McDonald, Smita Dey, Gina M Fernandez, et al., 2013). In

addition in this example, we also ran correlations with behavioural data as additional

analysis (Hadley C Bergstrom, Craig G McDonald, Smita Dey, Gina M Fernandez,

et al., 2013).

2.4.2 False Discovery Rate (FDR)

Where the area under investigation has been sectioned into topographical units,

each having its own data set, multiple ANOVAs on all topographical units may

determine more precisely any variance between experimental conditions. FDR

controls the expected rate of false rejection of the null hypothesis, by setting a

parameter, the quotient q, as the “tolerable” FDR (Genovese, Lazar, & Nichols,

2002). The q value is used as an alternative to p value when reporting significance,

and while it may be set at a conventional level (.05), a higher level may be

reasonable (Genovese et al., 2002). FDR has been used effectively in neuroscientific

studies (Hadley C Bergstrom & Luke R Johnson, 2014; Hadley C Bergstrom, Craig

G McDonald, Smita Dey, Gina M Fernandez, et al., 2013; Genovese et al., 2002;

Groppe, Urbach, & Kutas, 2011). Once the region of greatest variance across all

conditions is identified, follow up tests focus the investigation on the variance

between experimental conditions, in those locations.

Example: We have previously successfully applied FDR for type II error

minimization and identification of significance in specific topographic ROI in

behavioural experiments (see (Hadley C Bergstrom & Luke R Johnson, 2014;

Hadley C Bergstrom, Craig G McDonald, Smita Dey, Gina M Fernandez, et al.,

71

2013)). In these studies, we conducted mass univariate ANOVAs to assess

differences in neuron activation across all conditions in each of 46 bins. FDR

correction was used, with the tolerable limit set at q=.1. Significant differences

across conditions were found in certain micro ROIs (nine of 46 bins), so comparisons

were performed on those particular data to locate (1) the effect of the experimental

versus control groups and (2) the difference between two experimental groups

(Hadley C Bergstrom & Luke R Johnson, 2014; Hadley C Bergstrom, Craig G

McDonald, Smita Dey, Gina M Fernandez, et al., 2013). The q values were mapped

onto the topographical matrix (bins) to reveal the highly localized topography of

neuronal activation. The spatial distribution of these points of significance was

confirmed on visual analysis of the neuronal topographic density maps compiled

from topographic data, and also reflected earlier findings (Hadley C Bergstrom et al.,

2011). Subsequent correlational analysis was used to confirm the relationship

between the density of marked neurons and behaviour.

2.4.3 Principle Components Analysis (PCA)

Another approach to topographical data with multiple ROI and group

comparisons is principal components analysis (PCA). PCA seeks to identify and rank

combinations of variables that account for variance within the data set. PCA enables

the relationships between these patterns of variables to be identified, tested and

confirmed (Jolliffe, 2002). PCA has been applied by ourselves and others to address

a variety of anatomical questions, for example in morphological studies of microglial

cells (Soltys et al., 2005); and vagus nerves (Horn & Friedman, 2003); localization of

sensory cells in the thalamus in facial recognition (Chapin & Nicolelis, 1999); the

segregation of pyramidal neurons into morphological defined cell populations (H. C.

72

Bergstrom et al., 2008); eye-tracking data (J. C. R. Bergstrom, Olmsted-Hawala, &

Bergstrom, 2016) and extensivley in MRI data (F. Lin et al., 2006).

Example: We have successfully applied PCA for the analysis of topographic

neuronal density data activated in studies of Pavlovian fear conditioning. Activated

neurons were mapped and the area sectioned into micro ROIs (bins) as described

above, to produce a matrix of memory data (Hadley C Bergstrom, Craig G

McDonald, Smita Dey, Haying Tang, et al., 2013; Hadley C Bergstrom et al., 2011) .

Ten components (of spatial data) were revealed, with one of these (SC1) being

associated with the pattern of greatest difference (principal component score) in the

spatial distribution of activated neurons between experimental conditions. SC1

displayed a unique pattern of activated neurons in a particular subnucleus of the

amygdala (the LAd) across all brain samples in the experimental group. This was

confirmed by t-test comparisons (Bonferroni corrected) of the bins with the most

prominent loading values, and these also correlated with the area of highest density

in the topographic analysis outlined above. That is, as described above, the statistical

pattern could be confirmed by visual patterns seen in the neuronal topographic

density maps generated by color-coding neuron densities. PCA has proved a useful

statistical tool to extract meaningful patterns of variance related to the experimental

manipulation, which could be confirmed by both comparison with visual

representations of the data and Bonferroni corrected t-tests (Hadley C Bergstrom,

Craig G McDonald, Smita Dey, Haying Tang, et al., 2013; Hadley C Bergstrom et

al., 2011).

2.4.4 Multiple discriminative analyses (MDA)

73

Multiple Discriminant Analysis (MDA) is a method of visualizing patterns

within complex data sets (L. Lin, Osan, & Tsien, 2006). With complex data, such a

topographic data with many anatomical sub-regions and bins combined with multiple

experimental conditions, where both location and distribution across area, are under

investigation it can be important to identify patterns within this data set, in order to

help understand and interpret the data. MDA can be used to determine how a set of

continuous variables can discriminate groups (Hadley C Bergstrom, Craig G

McDonald, Smita Dey, Haying Tang, et al., 2013), for example, how the pattern of

neuron density in certain subnuclei (the independent or predictor variable) can

predict the experimental condition the subject brain best fits into (the grouping or

independent variable). MDA gives loading values (canonical variate correlation

coefficients) that represent the relative contribution of each variable in a set of

variables (a dimension) that discriminates groups from each other (see (L. Lin et al.,

2006); and (Hadley C Bergstrom, Craig G McDonald, Smita Dey, Haying Tang, et

al., 2013)).

Example: In one topography of Pavlovian fear memory study, we were

interested in the relative contribution of lateral and basal amygdala (LA) subnuclei to

the overall density of activated (pERK/MAPK expressing) neurons among each

experimental condition (Hadley C Bergstrom, Craig G McDonald, Smita Dey,

Haying Tang, et al., 2013). First, MANOVA was performed to examine the

relationship among the subnuclei. Where a significant relationship was found, one-

way ANOVA on each subnucleus tested for significant differences between

conditions. Next, MDA was used to test the relative contribution of each subnucleus

to the overall difference in density of activated neurons between conditions. The

MDA revealed a single underlying pattern in density of activated neurons across

74

lateral and basal amygdala subnuclei that discriminated the experimental and control

groups. It also showed the subnucleus (the LAd) that contributed most to the overall

difference between conditions. Having used MDA to help identify the region with

the most significant contribution to the overall pattern of variance between

conditions, it was possible to go further and explore more fine-grained details within

the data. To confirm the pattern identified with MDA, post hoc comparisons with

Bonferroni correction were performed, verifying the findings on the location and

experimental condition of the greatest activation, and reinforcing ours and others

previous findings about the predominance of LAd neural plasticity in fear memory

(Hadley C Bergstrom et al., 2011; Rodrigues et al., 2004).

2.4.5 Mixed Model ANOVA

The Mixed Model ANOVA also known as a Mixed Design ANOVA or a Split-

Plot ANOVA, allows for testing for differences between independent groups (in

functional topography experiment these will be the impendent behavioural groups,

i.e. experiment and control groups) while using repeated measures (bins in

topography experiments). Thus, the Mixed Model ANOVA can be employed for

microanatomy data comprising neuron counts within bins contrasted across several

independent groups. For our studies of functional neuronal topography, we typically

derive 20-80 bins per animal comprising the within-group dependent variable. For

the independent variable, several independent groups of animals are used including

experiment and control groups. Mixed Models allow for the analysis of data from all

locations and all animals in one analysis. Thus, Mixed Models a have strong

potential for analysis of topographic data combined with experimental manipulations

– such as behavioural or pharmacological manipulations. Using a Mixed Model

analysis data between anatomic locations can be compared and no adjustment for

75

multiple comparisons is required. Mixed Models can be thought as an advancement

of ANOVA and regression models. One, very important but often overlooked,

assumption of ANOVA/Regression, is that the data are independent of each other.

Thus, the analysis cannot have the same individual represented twice in the same

dataset. For example, measurements on lateral amygdala have to be analysed

separately from infralimbic cortex.

Mixed models ANOVA offers a toolbox to account for the dependence of

measurements taken on the same individual, by accounting for, so called, random

effects. Random effects are variables for which we are not interested in the actual

levels that we have sampled but on what they represent as a sample from a

population. The most usual random effect would be the individual animal (for further

definitions of random effects readers are directed to (Fitzmaurice, Laird, & Ware; A.

Zuur, Ieno, & Smith, 2007; A. F. Zuur, Ieno, Walker, Saveliev, & Smith, 2009).

Methods related to Mixed Model ANOVA that could also be applied to topographic

data sets with is the Generalized Estimated Equations (GEE) and the Generalized

Additive Mixed Models (GAMM) which can accommodate non-linear relationships

(for further information see, Zuur and Ieno, 2016 for GAMM and Fitzmaurice et al.

2004 on Mixed Model ANOVA and GEEs and their differences) (Fitzmaurice et al.;

A. F. Zuur & Ieno, 2016).

76

Table 1 Options for statistical analysis

Method

Purpose Advantage

ANOVA followed by Bonferroni corrected t-tests

To define where there is a significant difference in the data across conditions

Stringent control over type II errors

False discovery rate

To locate specific topographic regions of greatest variance across all conditions

- Controls the expected rate of false rejection of the null hypothesis) - Greater power - Can be useful prior to correlational analysis

Principle components analysis

Identifies and ranks combinations of variables that account for variance within the data set

Extract meaningful patterns of variance related to the experimental manipulation

Multiple discriminative analysis

To visualize patterns within complex data sets

Determines how a set of continuous variables can discriminate groups

Mixed model ANOVA

Tests for differences between independent groups while using repeated measures to analyse topographic data combined with experimental manipulations

- No adjustment for multiple comparisons is required - Accounts for random effects - *GEE and **GAMM can be applied after it, to accommodate non-linear relationships

* GEE: generalized estimated equations, **GAMM: generalized additive mixed models

2.5 Discussion

Understanding neural network organization and predicting memory and

behaviour from neural network functionality is a critical goal of neuroscience. While

various imaging techniques are capable of large-scale analysis of functional brain

regions, it is not suitable for imaging the spatial distribution, connectivity and

stability of neurons at the micro-network level. The ability to accurately map,

measure and compare neural network spatial properties, as described here,

contributes to our fundamental awareness of the organization and structure of

functional neural circuits. Classic cellular and molecular analysis of neuronal tissue

77

assists in the identification of molecular machinery underlying behaviour but does

not answer questions relative to the fundamental organizational properties and their

functional changes associated with behaviour. We have developed a combined

topographic and statistical approach for producing and analysing micro-topographic

data. This method provides clear visualization of the spatial organization and degree

of consistent neuronal patterns across brains from individual subjects and in different

experimental conditions.

Neuronal material used for topographic mapping can include both exogenously

labelled, such as immunocytochemistry and in situ hybridization, as well as

endogenous genetic labelling with green fluorescent protein (GFP) and other

fluorescent probes. Consistency in labelling is important with regard to whichever

neuron marking system is selected for topographic mapping. The statistical methods

recommended and applied here allow for natural variation in measured populations.

Nonetheless, reduction of variability will improve outcome consistency and

statistical verifications. Marking neurons requires consistent labelling and consistent

identification of neurons. To verify consistency, ideally experimenters blind to the

experimental conditions are employed throughout or for verification checks of large

data sets. The general principles outlined here for micro-topographic mapping can

be applied to sectioned brain material as well as whole brain analysis approaches

using CLARITY, CUBIC or iDISCO. Three-dimensional analysis also requires focus

and comparative measurements on specific anatomic regions of interest. Both 2D and

3D analysis ultimately requires localization and correlation of cellular activity with

behavioural function using the approaches described here.

78

2.5.1Topographic Mapping

The first step in the approach to visual and quantitative analysis of functional

neuronal topography between animals is to establish section alignment. Careful

choice of an appropriate and stable landmark or anchor point associated with the

region of interest is essential (Hadley C Bergstrom & Luke R Johnson, 2014; Hadley


Hadley C Bergstrom, Craig G McDonald, Smita Dey, Haying Tang, et al., 2013;

Hadley C Bergstrom et al., 2011; Haranhalli et al., 2007; Luke R Johnson et al.,

2012; Joseph E LeDoux et al., 2006). Identification of an anchor point which has

rapid and distinct conformational change through sectional view planes will ensure

success at this level. The second stage involves fitting a contour to the region of

interest, which ensures precision of the region in which the neurons will be counted,

as well as consistency in the area across subjects. A limitation at this stage is small

variation between sections from each subject, which can come from animal

variations and also from histological processing, therefore care is needed to minimize

variation. The contour must be fitted to each section with a degree of individual

judgement. Specific brain regions, such as the hippocampus, may also significantly

change in shape along the longitudinal axis and therefore a single contour is not

feasible. An alternate approach entails producing a unique mean contour section for a

specific data set. The rat brain atlas, developed by Paxinos and Watson in the 1980’s

(Paxinos & Watson, 2007), is one of the most established and detailed sources of

anatomical coordinates available at this time. Other brain atlases are available and

can also be used. In the Paxinos and Watson atlases, the depicted brain sections can

appear up to 480 micrometres apart necessitating several brain sections to be mapped

79

to individual atlas plates. Our method is therefore limited in part by the standardized

atlas information currently available (Paxinos & Watson, 2007).

Prior to creating a contour an atlas image generally requires resizing, which

can represent an amount of time spent making adjustments with various software

packages. Due to the number of software packages used to produce the images, it is

essential to note both the accepted file types (as listed in methods above) for

compatibility as images are moved between programs. Furthermore, it is very

important to note the numerical functions involved in any resizing, so that

consistency is maintained. Computer processing speed and memory requirements

must also be considered when using the large data files produced by slide scanning.

Free, open source programs are available for some procedures, making our

described method economically viable to all. For example, Image J and FIJI

(National Institutes of Health) can be substituted for some elements of the

topographic mapping, as it is able to perform cell counts and export x,y coordinate

data. Image J has many plugins available and runs in Java which is editable. Prior to

this the contours must be calibrated to a zero point to facilitate precise individual

comparisons. Once the coordinates have been exported a data matrix may be

developed. Data bins are created using a geospatial analysis formula to establish

unbiased bin dimensions. Open source programs are also available for this step

requiring some degree of coding for specific features. QtiPlot (Free Software

Foundation) is a free replacement for Origin and SigmaPlot. It will enable binning of

x, y coordinates into a two-dimensional matrix and has contour generating

capabilities for producing neuronal topographic density maps. Free online software

for FDR analysis, as described above, is also available (sdmproject.com). While we

have outlined and described our methodical approach using a series of standalone

80

commercial software packages for each of the steps descried, free software is also

available making the methodical approaches described here freely available for all

worldwide.

2.5.2 Analysis of Topographic Data

Although we have presented several arguments for the use of binned data for

micro-topographic analysis, there remains the opinion that discretization has

limitations (Langseth, 2008; MacCallum, Zhang, Preacher, & Rucker, 2002). We

have used both PCA as well as Mass Univariate ANOVA with FDR correction as a

useful way to locate areas of most variance in complex data, and to confirm the

qualitative data from our mean heat maps. This method assists in decreasing the

reduction in power generated with Bonferroni procedures (Verhoeven, Simonsen, &

McIntyre, 2005). While we provide general guidance for analysis of binned micro-

anatomical data sets, we advise the reader to liaise with statisticians to evaluate the

methodical approaches described here with the chosen data analysis techniques for

the analysis of unique data sets and research questions.

2.6 Conclusion

Neuronal micro-topographic density maps can assist in defining specific brain

regions involved in behaviour. Statistically verified microanatomical mapping has

the ability to advance our knowledge of the multi layered, complex organization of

the brain and its cognitive systems. Our approach for the measurement and

contrasting of neuronal topographic data in behavioural experiments has been

successfully applied to the study of the microanatomy of memory formation. It has

enabled us to visualize the spatial allocation of neurons activated during the

acquisition of fear memories (Hadley C Bergstrom & Luke R Johnson, 2014; Hadley

81


Hadley C Bergstrom, Craig G McDonald, Smita Dey, Haying Tang, et al., 2013;

Hadley C Bergstrom et al., 2011; Haranhalli et al., 2007; Luke R Johnson et al.,

2012; Joseph E LeDoux et al., 2006). We propose this method will prove

advantageous to other forms of neuroscience, including the cellular basis of

addiction; pathological memory models; pharmacological manipulations and other

forms of functional microanatomy (Holmes & Singewald, 2013; Luke R Johnson et

al., 2012). Existing nuclei catalogued in brain atlases have been defined

histologically, our approach allows for the identification of new functional micro

regions within established brain nuclei. By providing this walk-through tutorial we

encourage further development of these goals.

83

Localization of Contextual and Context Removed

Auditory Fear Memory within the Basolateral Amygdala Complex

This chapter comprises the following published article: Chaaya, N., Jacques, A., Belmer, A., Richard, D. J., Bartlett, S. E., Battle, A. R., & Johnson, L.R. (2019). Localization of Contextual and Context Removed Auditory Fear Memory within the Basolateral Amygdala Complex. Neuroscience. Published 13th December, 2018 https://doi.org/10.1016/j.neuroscience.2018.12.004

Chapters 3 and 4 encompass the remaining objectives listed in aim 1, inclusive

of identifying visual and quantitative differences in functional neuronal population

involved in memory consolidation and recall.

84







5. they agree to the use of the publication in the student’s thesis and its publication on the QUT’s ePrints site

consistent with any limitations set by publisher requirements.

In the case of chapter 3: Localization of Contextual and Context Removed Auditory Fear Memory within the Basolateral Amygdala Complex Publication status: Published

Contributor Statement of contribution*

Nicholas Chaaya Involved in the conception and design of the project, behavioural and laboratory experiments, analysed the data and wrote and edited the manuscript and produced the figures.

Angela Jacques Assisted with behavioural and laboratory experiments

Arnauld Belmer Assisted with laboratory experiments.

Derek Richard Assisted with laboratory experiments.

Selena Bartlett Assisted in editing the manuscript.

Andrew Battle Assisted in editing the manuscript.

Luke Johnson Assisted with behavioural experiments, assisted in reviewing and editing the manuscript.

QUT Verified

Signature

85

3.1 Abstract

Debilitating and persistent fear memories can rapidly form in humans

following exposure to traumatic events. Fear memories can also be generated and

studied in animals via Pavlovian fear conditioning. The current study was designed

to evaluate basolateral amygdala complex involvement following the formation of

different fear memories (two contextual fear memories and one adjusted auditory

fear memory). Fear memories were created in the same context with five 1.0 mA

(0.50 s) foot-shocks and, where necessary, five auditory tones (5 kHz, 75 dB, 20 s).

The adjusted auditory fear conditioning protocol was employed to remove

background contextual fear and produce isolated auditory fear memories.

Immunofluorescent labelling was utilised to identify neurons expressing immediate

early genes. We found the two contextual fear conditioning procedures to produce

similar levels of fear-related freezing to context. Contextual fear memories produced

increases in basolateral amygdala complex immediate early gene expression with

distinct and separate patterns of expression. These data suggest contextual fear

memories created in slightly altered contexts, can produce unique patterns of

amygdala activation. The adjusted auditory fear conditioning procedure produced

memories to tone, but not to context. This group, where no contextual fear was

present, had a significant reduction in basolateral amygdala complex immediate early

gene expression. These data suggest background contextual fear memories, created

in standard auditory fear conditioning protocols, contribute significantly to increases

in amygdala activation.

86

Keywords: Topography, Memory Allocation, Threat Conditioning, Lateral

Amygdala; Fear Conditioning, Immediate Early Genes.

3.2 Introduction

Pathological and persistent fear memories can rapidly form in humans

following exposure to trauma (Boschen, Neumann, & Waters, 2009; Michael S.

Fanselow, 2010; Maren, 2011). In animals, such as rodents, similar fear memories

can be experimentally generated by Pavlovian fear conditioning protocols (Foa,

Zinbarg, & Rothbaum, 1992; L. R. Johnson, J. McGuire, R. Lazarus, & A. A.

Palmer, 2012; Joseph E. LeDoux, 2014; Rothbaum & Davis, 2003). The neuronal

mechanisms of both contextual fear conditioning (CFC) and auditory fear

conditioning (AFC) have been, and continue to be deeply explored (for example, see

(Hadley C. Bergstrom, 2016; Michael S. Fanselow, 2010; Izquierdo, Furini, &

Myskiw, 2016; L. R. Johnson et al., 2012; Joseph LeDoux, 2012)). CFC and AFC are

both naturally occurring phenomenon which can also be replicated experimentally.

In the laboratory, noxious unconditioned stimuli (US), typically mild foot-shocks in

rodent research, are paired with specific contexts (for CFC), or contexts and

conditioned stimuli (CS) such as tones for AFC (Nicholas Chaaya, Battle, &

Johnson, 2018; L. R. Johnson et al., 2012; Joseph LeDoux, 2012). Numerous

investigations have shown abolishment, damage or inhibition of amygdala, either

permanently or temporarily, to result in attenuated or eradicated fear memories

(Michael S. Fanselow, 2010; J. LeDoux, 2000; Joseph LeDoux, 2003; Rudy, Huff, &

Matus-Amat, 2004). Our lab has previously identified specific sub-regions of the

amygdala responsible for auditory fear memory formation (H. C. Bergstrom & L. R.

Johnson, 2014; H. C. Bergstrom, C. G. McDonald, S. Dey, G. M. Fernandez, & L. R.

Johnson, 2013; H. C. Bergstrom et al., 2012; Hadley C Bergstrom et al., 2011). We

87

showed CFC and AFC (H. C. Bergstrom et al., 2012) to result in different numbers

of amygdala neurons expressing phosphorylated mitogen-activated protein kinase

(pMAPK), a well-documented marker of neuronal plasticity (Giese & Mizuno, 2013;

S. Peng, Zhang, Zhang, Wang, & Ren, 2010). Data suggested lateral amygdala (LA)

to be essential for auditory fear memory formation, but not for contextual fear

memory formation (H. C. Bergstrom et al., 2012). More recently, we identified

unique and separate patterns of auditory and visual fear memory consolidation in the

amygdala (H. C. Bergstrom & L. R. Johnson, 2014). These studies provided

evidence into the organisation of differential fear memories in the rodent brain. The

current study is designed to further investigate the complex organisation of fear,

focusing on fears of context and auditory tone. Specifically, using novel modified

behavioural approaches, the differences in the neural organisation of contextual fear

memories (created by two separate conditioning protocols) and auditory fear

memories are evaluated here.

Contextual and auditory fear conditioning are often referred to as unsignaled

and signalled fear conditioning, respectively (Russel G Phillips & LeDoux, 1994).

Both fear conditioning protocols produce contextual fear memories (Lehmann,

Lacanilao, & Sutherland, 2007; Russel G Phillips & LeDoux, 1994). “Foreground”

contextual fear memories develop following unsignaled fear conditioning, whereas

“background” contextual fear memories develop following signalled fear

conditioning as the signal (tone) is now in the foreground (Russel G Phillips &

LeDoux, 1994). Evaluations of auditory (signalled) fear memories are, in fact,

evaluations of both (1) auditory fear and (2) background contextual fear (Nicholas

Chaaya et al., 2018). The current study is designed to evaluate amygdala

involvement following the creation of pure, context-removed, auditory fear

88

memories. Rats in the AFC group underwent high exposure to the context prior to

conditioning. As per latent inhibition (LI), we hypothesised that prior exposure to the

fear conditioning context would lead to the development of an attenuated, or even

abolished, background contextual fear memory (R. Lubow, Rifkin, & Alek, 1976; R.

E. Lubow, 1965; R. E. Lubow & Moore, 1959). The ability for prior context

exposure to abolish foreground contextual fear memories has been demonstrated

previously (Barot, Chung, Kim, & Bernstein, 2009). By exposing rodents in the AFC

group to the conditioning chamber for 30 minutes a day for 10 days prior to

conditioning, the current study aimed to create pure, context-removed, auditory fear

memories (Barot et al., 2009).

Context and spatial memory research has delineated the influence of single

discrete stimuli (e.g. sound or smell) on the brain (M. I. Anderson & Jeffery, 2003;

D. M. Smith & Mizumori, 2006a, 2006b; Song, Kim, Kim, & Jung, 2005). Inclusion

of single new elements into a context can directly alter hippocampal place cell

activity (M. I. Anderson & Jeffery, 2003). Similarly, presentation of a single new

element (such as an acoustic stimulus) can alter amygdala activity (Romanski,

Clugnet, Bordi, & LeDoux, 1993). Unpaired fear conditioning (UFC) protocols are

identical to standard CFC protocols except for the addition of a single new discrete

element (in many cases, a tone). Following UFC, rodents develop fear memories to

the conditioning context, but not the auditory tone (H. C. Bergstrom et al., 2012).

Therefore, UFC protocols offer a method to evaluate how alterations to the fear

conditioning context alter contextual fear related behaviour and anatomy.

Traditionally, UFC protocols have been utilised as control conditions to paired or

signalled fear conditioning (H. C. Bergstrom et al., 2012; Hadley C Bergstrom et al.,

2011; Majak & Pitkänen, 2003; McKernan & Shinnick-Gallagher, 1997; Radley et

89

al., 2006; Michael T. Rogan, Staubli, & LeDoux, 1997). However, the ability for

UFC to produce associative fear memories to context, as demonstrated previously

(H. C. Bergstrom et al., 2012; Hadley C Bergstrom et al., 2011), suggests activation

of amygdala likely occurs. This was demonstrated by Trifilieff and colleagues, who

showed UFC to produce greater amygdala-related plasticity as compared to AFC

(Trifilieff, Calandreau, Herry, Mons, & Micheau, 2007). The current study,

therefore, evaluates if, and how, UFC differs from standard CFC and context-

removed AFC.

The objectives of the current study were to explore the complex neural

organisation of contextual fear memories (created via a standard CFC or an UFC

protocol) and context-removed auditory fear memories in the rodent basolateral

amygdala complex (BLC). While these fear memories were created with a consistent

noxious unconditioned stimulus (US; foot-shocks), the neutral conditioned stimulus

(CS) differs. Specifically, the CS is the context (CFC), the context and tone (UFC) or

just the tone with context removed (adjusted AFC; AAFC). The current study

explored the number of neurons expressing the immediate early genes (IEGs) activity

regulated cytoskeleton-associated protein (Arc) and c-Fos. Co-localisation of these

two IEGs was also explored to determine if a separate distribution of neurons express

both markers following fear memory formation. In order to obtain a large

representation of the BLC, comparisons between IEG expression were conducted at

four accurately aligned Bregma coordinates (see (H. C. Bergstrom et al., 2012;

Hadley C Bergstrom et al., 2011; A Jacques et al., 2018)). We focus on consistent

differences between conditioned groups as compared to two control groups: a context

only (CO) control and a tone alone (TA) control. We found our adjusted AFC

protocol to produce a reduced pattern of Arc and c-Fos expressing neurons as

90

compared to the pattern evident in rats that developed contextual fear memories

either via a CFC or UFC protocol. Furthermore, contextual fear memories created

with CFC lead to a different pattern of IEG expression in amygdala when compared

to those created by UFC.

3.3 Experimental Procedures

3.3.1 Animals

Experimentally naïve adult male Sprague Dawley Rats (Animal Resources

Centre, WA, Australia) weighing between 176-200 g at the time of arrival were

housed (two per cage) on a 12-hour light/dark cycle by the University of Queensland

Biological Resources (UQBR) facility. Food and water were provided ad libitum. All

procedures were approved by the University of Queensland (Ethics approval no.

023/17) and Queensland University of Technology (QUT approval number:

1700000295) animal ethics committees and complied with policies and regulations

regarding animal experimentation and other ethical matters, in accordance with the

Queensland Government Animal Research Act 2001, associated Animal Care and

Protection Regulation (2002 and 2008), as well as the Australian Code for the Care

of Animals for Scientific Purposes, 8th Edition (National Health and Medical

Research Council, 2013). Rats were acclimatised to the UQBR Facility for 8 days,

and then underwent 10 days of handling or habituation prior to fear conditioning.

Rats (weighing 324.8 ± 2 g at the time of training) were divided into five separate

groups, three experimental (Contextual Fear Conditioned; CFC n = 18, Adjusted

Auditory Fear Conditioned; AAFC n = 18 and Unpaired Fear Conditioned; UFC n =

18) and two control (Tone Alone; TA n = 18 and Context Only; CO n = 18) groups.

91

Following fear conditioning, rats were further divided into an anatomical (n = 12 per

group) and a behavioural (n = 6 per group) group.

3.3.2 Apparatus

Rats were exposed to two Plexiglas conditioning chambers (Coulbourn

Instruments, Lehigh Valley, Pennsylvania, USA). Both chambers (context A and B)

were dimly illuminated with a single house light (2 – 3 lux), sound insulated

(background dB = 55), equipped with a speaker and contained an infrared camera.

Context A was fitted with a metal grid floor which was connected to a shock

generator. Context A was undecorated and cleaned with ethanol (EtOH) 80%

following presentation of each rat. Context B had a flat floor, coloured decoration,

alterations to the roof to change its physical dimensions, and was lightly covered

with bedding. Following the presentation of each rat, context B was cleaned with

orange scented hand soap and the bedding replaced.

3.3.3 Procedure and Design

Behavioural procedures are outlined in Figure 1 and explained in detail below.

Figure 3-1: Experimental design for behavioural training.

Following acclimatisation, rats in the AAFC group underwent nine days of habituation to context A, while rats in all other groups were handled. Following, all rats were habituated to the same context for one day. On day 11, rats underwent their respective conditioning procedures, and then were either perfused 90 minutes later (anatomical rats) or provided with a fear memory test (behavioural rats) to context 24 hours later, and then to tone three days later. CFC: contextual fear conditioning; UFC: unpaired fear conditioning; TA: tone alone; CO: context only; AAFC: adjusted auditory fear conditioning; FMT: fear memory test.

92

3.3.4 Acclimatisation and Habituation

All rats had eight days of acclimatisation to the vivarium prior to behavioural

procedures. Rats in all groups, except the AAFC group, were handled for nine days,

and then habituated to context A (30 minutes per rat) for one day. Alternatively, rats

in the AAFC group were habituated to context A for 10 days (30 minutes each). This

adjustment was designed to induce latent inhibition to the fear conditioning context.

Previous investigations into latent inhibition suggest that 10 days is sufficient (Barot

et al., 2009).

3.3.5 Fear Conditioning

One day following habituation, rats in the CFC group were placed into context

A and presented with five non-overlapping and random shocks to the foot (1.0 mA,

0.50 s). Rats in the AAFC group were placed into context A and presented with five

pairings of an auditory tone (5 kHz, 75 dB, 20 s) that co-terminated with the foot-

shocks (1.0 mA, 0.50 s). Rats in the UFC group were placed into context A and

presented with five non-overlapping presentations of the same auditory tones (5 kHz,

75 dB, 20 s) and foot-shocks (1.0 mA, 0.50 s). Rats in the TA control group were

placed into context A presented with five non-overlapping and random auditory

tones (5 kHz, 75 dB, 20 s). Rats in all groups were permitted to explore the context

for 180 seconds prior to the presentation of any stimuli. Rats were allowed 60

seconds following the presentation of the final stimulus before being removed and

returned to their home-cage. Rats in the CO control group were left to explore

context A without any added stimuli. Fear conditioning procedures were 660 seconds

93

long for all groups except the UFC group, which lasted 880 seconds to account for

the non-overlapping foot-shocks and tones.

3.3.6 After Fear Conditioning

Behavioural group

Rats in the behavioural group (n = 30) had freezing behaviour manually scored

during training (fear conditioning) and testing (fear memory test; FMT). To avoid

inaccuracies that may occur from manual counting, and to remain consistent with

previous investigations, scoring occurred in 20 second blocks (H. C. Bergstrom & L.

R. Johnson, 2014; H. C. Bergstrom et al., 2012; R. G. Phillips & LeDoux, 1992;

Russel G Phillips & LeDoux, 1994; Gregory J. Quirk, Armony, & LeDoux, 1997;

Radley et al., 2006). During training, freezing was scored prior to (base line), during

(at cue 1, cue 2, cue 3, cue 4 and cue 5), and following fear conditioning (final). This

provided a progressive measure of fear. Immediately following fear conditioning, all

rats in the behavioural group were returned to their home-cage for 24 hours. Rats

were then re-exposed to context A for a 10 minute FMT to context. Neither foot-

shock nor auditory tones were presented during this time. Freezing was scored for

the final 20 seconds of every minute that rats were in context A. Rats were

immediately returned to home-cages for an additional three days and then re-exposed

to context B for a 10 minute FMT to tone. During this time, 10 of the same auditory

tones (5 kHz, 75 dB, 20 s) used for fear conditioning were presented in the final 20

seconds of every minute. Freezing behaviour was scored during this time.

Scoring of freezing

94

Freezing behaviour was defined as the inhibition, absence or suppression of

movement, except that necessary for autonomic nervous system functioning, such as

heart rate and respiration (Michael S. Fanselow, 1980). Head scanning and sleeping

were not included as freezing, however, heavy breathing and minimal movement

was. During training, freezing behaviour was scored in the final 20 seconds of the

first minute, in the final 20 seconds of the last minute, and the 20 seconds prior to

each foot-shock (or equivalent time-point for control groups). For rodents in the

AAFC and TA groups, this 20 second period is during tone presentation.

Alternatively, for rodents in the CFC and CO, this 20 second period occurred at the

same time-point, but without the presentation of a tone. Similarly, rodents in the

UFC had freezing scored during the 20 second time point prior to receiving foot-

shocks, as opposed to the 20 second time-point at which auditory tones are presented,

as the development of contextual fear memories is of interest here. During testing,

freezing behaviour was scored in the final 20 seconds of every minute. This was

done in order to account for extinction learning that naturally occurs following

exposure to fear conditioned stimuli (Maren, Phan, & Liberzon, 2013). For rodents

that received CFC and UFC (thus developing contextual fear memories), extinction

learning begins immediately upon re-exposure to context A, and continues until

rodents are removed from said context. This results in the gradual reduction of fear

(Bouton, 1988; Bouton, 2004). Alternatively, during the FMT to tone (in context B),

extinction learning will not occur for rodents that have auditory fear memories

(AAFC group) until they are presented with the tone. Therefore, in an attempt to

equalize extinction during the FMT to context in rodents with contextual fear

memories and during the FMT to tone in rodents with auditory fear memories, the

auditory tone was presented ten times. To obtain an accurate and full measure of

95

fear, freezing was scored during each of these presentations, or at identical time-

points during the FMT to context (H. C. Bergstrom et al., 2012; Hadley C Bergstrom

et al., 2011).

Anatomical group

Following fear conditioning, all rats in the anatomical group (n = 60) were

removed from the fear conditioning chamber and placed in their home cages. Ninety-

minutes following, rats were anesthetised and sacrificed via perfusion for double

fluorescent labelling.

3.3.7 Arc and c-Fos Fluorescent Immunohistochemistry

Tissue preparation

Arc and c-Fos protein expression have been shown to peak between 60 and 180

minutes following learning (Ivashkina, Toropova, Ivanov, Chekhov, & Anokhin,

2016; Lonergan, Gafford, Jarome, & Helmstetter, 2010; Morgan & Curran, 1991;

Ramírez-Amaya et al., 2005). We, therefore, aimed for a 90 minute timeframe.

Intraperitoneal (i.p.) injections of Ketamine/Xylazine (100 mg/kg, 10 mg/kg) were

administered. Once anaesthetised, rats were transcardially perfused via the ascending

aorta with ice-cold saline (200 mL per rat), followed by 4% paraformaldehyde/0.1 M

phosphate buffer (PB; pH of 7.4; 400 mL per rat). Brains were subsequently removed

and stored in the paraformaldehyde fixative for 24 hours (4 oC), and then stored in

phosphate buffered saline (PBS)/0.02% Azide for a minimum of three days. Free-

floating sequential coronal brain sections containing the amygdala were sliced on a

vibratome (M11000; Pelco easiSlicer, Ted Pella Inc, CA, USA) at 40 μm per section.

96

Left hemisphere sections were stored in PBS/0.02% Azide (4 oC) until

immunohistochemistry commenced.

Immunohistochemistry

Brain sections were removed from PBS/0.02% Azide and washed thoroughly

in PBS at room temperature. All further washes and incubations were conducted at

room temperature. To begin, sections were permeabilised with 1% Triton/0.1%

Tween 20 in PBS for one hour. Sections were washed with PBS, and then blocked

with 0.3% Triton/0.05% Tween 20/3% NGS/1% BSA in PBS for one hour. Blocking

solution was removed, and sections were immediately incubated in a c-Fos (9F6)

rabbit monoclonal antibody (1:300; Cell Signalling Technology, MA, USA) and an

Arc (C-7) mouse monoclonal antibody (1:300; Santa Cruz Biotechnology, TX, USA)

in blocking solution for 24 hours. Brain sections were washed in blocking solution

and immediately incubated in a pre-adsorbed goat anit-rabbit IgG H&L (Alexa Fluor

594) secondary antibody (1:500; Abcam, VIC, Aus) and a pre-adsorbed goat anit-

mouse IgG H&L (Alexa Fluor 488) secondary antibody (1:500; Abcam, VIC, Aus)

in blocking solution for 30 minutes. Brain sections were then washed in blocking

solution, then PBS, and mounted in numerical order on silane coated slides. Mounted

sections were immediately cover-slipped, left to dry and stored at 4 oC.

Brain Scans

Following immunohistochemistry, cover-slipped brain sections were scanned

using the InCell Analyser 2200 (GE Healthcare Life Sciences, NSW, Aus), provided

by Translational Cell Imaging Queensland (TCIQ). The InCell Analyser 2200 is an

imaging system capable of producing confocal-like images of up to two colour

97

channels. The InCell Analyser 2200 was set up to obtain 10x magnified consecutive

fields of the amygdala (15% overlap, horizontal acquisition pattern, 10% laser

power) in FITC (Arc protein) and Cy3 (c-Fos protein) channels. The InCell Analyser

2200 does not stitch or merge channels. Individual images are provided per scan

area, with an associated .xdce file. Channels were manually merged in Fiji

(Schindelin et al., 2012) by opening the images via the .xdce file in the Bio-Formats

Import Option. Individual merged images were subsequently stitched (Preibisch,

Saalfeld, & Tomancak, 2009) using Fiji.

Amygdala Brain Section Alignment and Neuron Quantification

Neuron counting was conducted blind to the experimental conditions.

Aligned Bregma coordinates -3.36 mm, -3.24 mm, -3.12 mm and -3.00 mm were

counted. At these levels, the BLC is well represented (Paxinos & Watson, 2006).

Merged and stitched files were imported into Neurolucida 360 (Neurolucida 360,

MBF Bioscience, VT). To separate the various subregions of the BLC, a contour of

the BLC was created using Neurolucida trace option. Subregions (see Figure 2) were

as follows: dorsolateral portion of the lateral amygdala (LaDL); ventromedial portion

of the lateral amygdala (LaVM); ventrolateral portion of the lateral amygdala

(LaVL); anterior portion of the basal amygdala (BLA); and posterior portion of the

basal amygdala (BLP). The resulting contour was scaled and superimposed on

imported amygdala images. Precise stereotaxic alignment is required to accurately

analyse neurons expressing Arc and c-Fos protein. The lateral ventricle (LV) is a

major anatomical landmark which changes rapidly at different rostral to caudal

locations. The LV becomes noticeable at Bregma coordinate -3.32, and consistently

increases in size towards more caudal locations. Bregma coordinate -3.36 mm in the

Rat Brain Atlas depicts the LV as a noticeable tear-drop size, allowing for easy

98

identification and matching in all rats (Paxinos & Watson, 2006). Bregma coordinate

-3.36 mm was identified in all rats by this anatomical landmark. Preceding Bregma

coordinates were identified by measuring back from Bregma coordinate -3.36 mm.

Anatomical landmarks were utilised to accurately and consistently align a contour

across all rats. These anatomical landmarks were the LV (at Bregma coordinate -3.36

mm only), rhinal fissure (RF), dorsal endopiriform nucleus (DEn), central amygdala

(CeA), stria terminalis (st) and the optic tract (opt; see Figure 2).

Figure 3-2: Schematic illustration of aligned sections.

Bregma coordinate -3.36 mm was identified following delineation of the LV. Following, Bregma coordinates -3.24 mm to -3.00 mm were identified by measuring back from Bregma coordinate -3.36 mm. To accurately align stencils onto brain images, the following anatomical landmarks were utilised (shaded grey in Bregma coordinate -3.36 mm): opt, st, central amygdala, DEn and RF and the LV in Bregma coordinate -3.36 mm. Neurons expressing the IEGs Arc and c-Fos were quantified in sub-regions (outlined in all Bregma coordinates above) LaDL, LaVM, LaVL, BLA and BLP (Paxinos & Watson, 2006). LaDL: dorsolateral portion of the lateral amygdala; LaVM: ventromedial portion of the lateral amygdala; LaVL: ventrolateral portion of the lateral amygdala; BLA: anterior portion of the basal amygdala; BLP: posterior portion of the basal amygdala; RF: rhinal fissure; Den: dorsal endopiriform nucleus; CeA: central amygdala; st: stria terminalis; opt: optic tract.

Neuron Counting and Identification

Neurons were manually tagged and counted using Neurolucida 360. Neurons

within each subregion were tagged with a different colour. Arc protein expression is

present in neuronal cell bodies and dendrite (Lyford et al., 1995), while c-Fos protein

is only expressed in the nucleus (Curran, Miller, Zokas, & Verma, 1984; Morgan,

99

Cohen, Hempstead, & Curran, 1987). Therefore, Arc expressing neurons were

counted if they met the following criteria: triangular or oval shape with a clear

neuronal form, significantly more perceptible compared to the background and

containing at least one dendritic protrusion per cell body (see Figure 3a). c-Fos

expressing neurons were counted if they met the following criteria: clear circular

shape without a neuronal form, significantly more perceptible compared to

background, with no dendritic protrusions (see Figure 3b). Arc expressing and c-Fos

expressing neurons were counted separately. Co-localisation (see Figure 3c) was

obtained using the co-localisation option in Neurolucida 360. Correcting for double

counting was not necessary as sections were non-consecutive.

Figure 3-3: Arc, c-Fos and co-localised labelling of amygdala neurons.

A Arc expressing neurons are represented in green. B c-Fos expressing neurons are represented in red. C Co-localised expressing neurons are represented in yellow. Green arrows indicate single labelled Arc expressing neurons; red arrows indicate single labelled c-Fos expressing neurons; and yellow arrows represent co-localised expression of both Arc and c-Fos.

Raw Neuron Topographic Density Maps (Heat Maps)

Neuron topographic density maps, known as heat maps, allow for the

visualization of neuronal densities in particular brain regions, with neuron densities

depicted in increasingly ‘hot’ (blue to red) colors. In order to build heat maps, data

regarding XY coordinates of individual neurons were provided by Neurolucida. XY

coordinates of neurons were binned by Origin (2018, OriginLab, Northampton, MA).

100

Each bin contains data regarding the number of Arc, c-Fos and co-expressing

neurons. Spatial dimension of each bin was 100m x 100m. Consequently, each bin

represents the same portion of amygdala. Values within each bin (average number of

neurons activated within each bin) were represented by different colors (SigmaPlot v

12.5, Systat Software, San Jose, CA), allowing for the creation of the heat maps (for

further details on specific methodology used see (H. C. Bergstrom et al., 2012;

Hadley C Bergstrom et al., 2011; A Jacques et al., 2018))

Data Analysis

Anatomical data from the current experiment explores the difference between

behavioural groups at different (1) BLC sub-regions, (2) Bregma coordinates and (3)

IEG marker. Focus of analyses was dedicated to the differences between behavioural

groups, ANOVAs were utilised to compare differences between behavioural groups

at individual IEG marker, individual sub-region and individual Bregma coordinate

(e.g. difference in Arc expressing neurons between behavioural groups in the LaDL

of Bregma coordinate -3.36 mm). This method of analysis captures differences

following the various forms of conditioning, which is the focus of this study. Prior to

analysis, normality and homogeneity of variance were tested for. Normality was

found to be present in almost all cases, with some breaches noted in subregions BLA

and BLP. As these subregions are not the focus of this paper, analyses continued

with the normality assumption confirmed. Homogeneity of variance was found to be

breached on several occasions – suggesting that inflation of type I errors is possible.

Non-parametric tests are designed to account for these breaches by reducing the

possibility of producing type I errors. Similar to non-parametric tests the Bonferroni

adjustment accounts for type I errors that may arise for other reasons, namely,

multiple comparisons. The data reported here includes both unequal variances and

101

multiple comparisons. Nevertheless, we used the Bonferroni correction to account

for both unequal variance and multiple comparisons. This is because non-parametric

tests are strongly recommended for breaches of normality, but are less effective

against breaches of homogeneity (Zimmerman & Zumbo, 1993). Similar to

parametric tests, non-parametric tests can be sensitive to unequal variance, unequal

sample size and outliers (Parra-Frutos, 2013; Zimmerman & Zumbo, 1993).

Alternatively, the Bonferroni correction is robust and effective – and is often cited to

be too severe, thus producing type II errors (Perneger, 1998). For these reasons, we

believe that the level of severity employed by the Bonferroni adjustment accounted

for all possible inflations in type I errors that may have arose (unequal variance and

multiple comparisons). Therefore, all behavioural and anatomical analyses were

conducted using one-way ANOVAs, followed by Bonferroni-corrected post-hoc

tests. All values embedded in the text are expressed as the mean +/- standard error of

the mean. P values at or below 0.05 were considered statistically significant. All

major statistical analysis was conducted using the Statistical Package for the Social

Sciences (SPSS) v22 software (IBM Corporation, NY, USA). GraphPad Prism v7

software (GraphPad, CA, USA) was utilised to create graphs and identify outliers.

Throughout the results section asterisks denote levels of statistical significance (* p ≤

0.05; ** p ≤ 0.01; *** p ≤ 0.001; **** p ≤ 0.0001).

Excluded Cases

Anatomical cases were excluded if they served as statistical outliers, or had

significant damage to the brain tissue, which occurred either during the perfusion,

labelling or cover slipping process. Statistical outliers were identified in GraphPad

Prism using the ROUT method, with the maximum false discovery rate set at 1%.

102

The ROUT method was used as SPSS does not provide an automatic test for outlier

identification. The ROUT method – as compared to the alternate option in GraphPad:

Grubb’s test – is effective at identifying multiple outliers in large datasets. Outlier

analysis was conducted on individual groups. Identified outliers were excluded on a

pairwise basis. Ten cases were excluded on a list wise basis due to significant

damage to brain tissue. Behavioural cases were excluded if they served as a

statistical outlier (identified using the ROUT method). Further, one behavioural rat

was interrupted during the FMT to tone by the fire alarm. Fear to tone data from this

rat was excluded.

3.4 Results

3.4.1 Behavioural Results

The amount of time rats displayed freezing behaviour was quantified during

training and testing. During training, the mean amount of time (as a percentage) rats

displayed freezing behaviour was quantified as a function of condition and time-

point (baseline, cue 1, cue 2, cue 3, cue 4, cue 5 and final). A two-way mixed design

ANOVA was conducted with condition as the between-subjects factor and time-point

as the within-subjects factor. Mauchly’s test of sphericity was breached, therefore

corrected results from the Greenhouse-Geisser test is reported. The Greenhouse-

Geisser test revealed a significant interaction of freezing behaviour as a function of

condition and time-point F(10.103, 45.466) = 8.921, p < 0.0001. The main effects for

condition (F[4, 18] = 24.424, p < 0.0001) and time-point (F[2.526, 45.466] = 46.573,

p < 0.0001) were significant. Bonferroni correct post-hoc tests (see Figure 4a)

revealed no differences between conditions at baseline, cue 1 or cue 2. At cue 3,

differences between conditions become apparent and significant between rats that

103

received CFC (freezing response was 73%) and UFC (69%) as compared to the TA

(0%) group. At cue 4 rats that received CFC (99%), UFC (71%) and AAFC (85%)

exhibited significantly more freezing as compared to both the TA (0%) and CO (7%)

controls. This significant difference between fear conditioned and control groups

remained present at cue 5 and at the final time-point, indicating progressive

development of fear-learning. as a function of condition.

One-way ANOVA of freezing data obtained from the FMT to context revealed

significant differences between groups as seen, F(4, 25) = 20.899, p > 0.0001.

Bonferroni corrected post-hoc tests revealed rats that underwent CFC (freezing

response was 71%) and UFC (56%) exhibited significantly more freezing as

compared to those that underwent AAFC (21%) during the FMT to context. Further,

those that underwent CFC and UFC exhibited significantly more freezing than the

TA (6%) or CO (3%) controls. No further differences between groups were observed

(see Figure 4a). One-way ANOVA of freezing data obtained from the FMT to tone

revealed significant differences between groups as seen, F(4, 24) = 9.113, p > 0.001.

Bonferroni corrected post-hoc tests showed rats that underwent AAFC (49%)

exhibited significantly more freezing as compared to those in all other conditions

(CFC = 11%; UFC = 10%; TA = 1%; CO = 0.3%). No further differences between

groups were observed (see Figure 4b). These data confirm the behavioural

procedures employed in this experiment successfully create fear memories. Further,

these data suggest AAFC, where rodents underwent extreme prior exposure to the

context, significantly attenuated background contextual fear memories, while leaving

auditory fear memories intact.

104

Figure 3-4: Freezing to context and tone.

A Freezing behaviour was scored during training to obtain a progressive measure of fear learning. For rodents in the AAFC and TA group, this 20 second period occurred during auditory tone presentation. For rodents in the CFC and CO group, this 20 second period occurred at the same time-point, but without tone presentation. Alternatively, rodents in the UFC had freezing scored during the 20 second time point prior to receiving foot-shocks, as opposed to the 20 second time-point at which auditory tones were presented (as this group develops contextual fear memories, not auditory fear memories). Analysis of mean freezing during training showed that rats in the CFC, UFC and AAFC groups exhibited equivalent freezing to rats in the CO and TA control groups at baseline, cue 1 and cue 2. At cue 3, differences between conditions become apparent and significant between rats that received CFC and UFC as compared to the TA (CFC *, p < 0.05; UFC *, p <0.05) group. At cue 4 rats that received CFC, UFC and AAFC exhibited significantly more freezing as compared to both the TA (CFC ****, p < 0.0001; UFC **, p < 0.01; AAFC ***, p < 0.001) and CO (CFC ****, p < 0.0001; UFC **, p < 0.01; AAFC ***, p < 0.001) controls. This significant difference between fear conditioned and control groups remained present at cue 5 (all differences between the conditioned and control groups were ****, p <0.0001) and at the final time-point (all differences between the conditioned and control groups were ****, p <0.0001). Symbols in figure denote level of statistical significance between average (of all time points) freezing behaviour between all fear conditioned groups compared to TA (* p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001; **** p ≤ 0.0001) and CO (‡ p ≤ 0.05; ‡‡ p ≤ 0.01; ‡‡‡ p ≤ 0.001; ‡‡‡‡ p ≤ 0.0001) control groups B One day post fear conditioning, all rats were re-exposed to context A. Rats that underwent CFC or UFC had significantly more freezing as compared to controls, indicating that these fear conditioning protocols reliably created contextual fear memories. Rats that underwent AAFC did not exhibit any statistical differences in freezing as compared to controls, suggesting a loss of background contextual fear memories. Asterisks denote levels of statistical significance between groups (* p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001; **** p ≤ 0.0001). C Three days following, all rats were exposed to a new context and provided with auditory tones. Rodents that underwent AAFC had significantly more freezing to tone as compared to all other groups. This suggests that only these rodents developed auditory fear memories. Asterisks denote levels of statistical significance between groups (* p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001; **** p ≤ 0.0001). CFC: contextual fear conditioning; UFC: unpaired fear conditioning; AAFC: adjusted auditory fear conditioning; TA: tone alone; CO: context only.

105

3.4.2 Anatomical Results

The following section is divided by Bregma coordinate. We focus on

statistical significance between the experimental groups (CFC, AAFC or UFC) to the

control groups (TA or CO). Importance is attributed to significant differences when

compared to both controls, as opposed to just one. Patterns of consistent statistical

significance between behavioural groups are of interest. Differences between said

behavioural groups identified between multiple Bregma coordinates and identified

with multiple IEG markers are most important as they represent clear and stable

involvement of a particular subregion to a behavioural condition. Investigations

revealed little consistent differences between rats that were conditioned and controls

in subregions BLA and BLP. Therefore, while some consistent differences are noted

in text and heat maps include these sub-regions, they are excluded in graphs.

IEG Expression at Bregma Coordinate -3.36 mm

To determine involvement of BLC in CFC, AAFC and UFC, one-way

ANOVAs were conducted. ANOVAs revealed group differences to exist in Arc

expressing neurons in LaDL (F[4, 41] = 5.453, p < 0.01) , LaVM (F[4, 37] = 6.575,

p < 0.001) and LaVL (F[4, 37] = 2.879, p < 0.05), but not BLA (F[4, 42] = 1.853, p

= 0.137) or BLP (F[4, 39] = 1.111, p = 0.365). Group differences were consistently

observed in all BLC subregions in c-Fos expressing neurons (LaDL F[4, 41] = 7.888,

p < 0.0001; LaVM F[4, 43] = 7.048, p < 0.001; LaVL F[4, 41] = 7.845, p < 0.0001;

BLA F[4, 43] = 4.7335, p < 0.01; BLP F[4, 41] = 6.956, p < 0.001) as well as co-

expressing neurons (LaDL F[4, 43] = 7.088, p < 0.001; LaVM F[4, 42] = 7.080, p <

0.001; LaVL F[4, 41] = 6.485, p < 0.00; BLA F[4, 40] = 7.749, p < 0.001; BLP F[4,

37] = 3.225, p < 0.05). Bonferroni corrected post-hoc tests (see Figure 5) revealed

those that underwent CFC expressed significantly more Arc, c-Fos and co-expressing

106

neurons in LaDL as compared to both controls. Comparatively, those that underwent

UFC had consistent significant increases in Arc, c-Fos and co-expressing neurons in

LaDL and LaVM. LaVL c-Fos and co-localised, but not Arc, expression was

significantly higher in those that underwent UFC as compared to controls. Further

differences were present in BLA for c-Fos and co-expressing neurons (data visually

present in heat maps). No significant differences were observed between those that

underwent AAFC as compared to both controls.

107

Figure 3-5: Immediate early gene expression in Bregma coordinate -3.36mm.

A Arc expression in LaDL was significantly higher in rats that underwent CFC as compared to both controls. Rats that underwent UFC exhibited similar differences as compared to both controls in LaDL. Further differences were noted in those that underwent UFC as compared to controls in subregion LaVM, but not LaVL. No differences were identified in those that underwent AAFC as compared to controls in any lateral amygdala subregion. B Examination of c-Fos expression revealed significant increases in rats that underwent CFC and UFC as compared to both controls in subregion LaDL. Further, those that underwent CFC exhibited a significant increase as compared to the CO group in subregion LaVM, whereas those that underwent UFC exhibited further differences as compared to controls in subregions LaVM and LaVL. No differences were observed in those that underwent AAFC. C Co-localised neuron expression revealed similar differences in rats that underwent CFC and UFC as compared to both controls in subregion LaDL. Those that underwent UFC exhibited further differences as compared to controls in subregions LaVM and LaVL. Once

108

again, no differences were observed in those that underwent AAFC. D Frequency of significant differences as compared to control groups in those that underwent CFC, UFC or AAFC. Six significant differences per subregion can exist: three IEG markers and two controls. As seen, those that underwent UFC had the highest number of significant differences as compared to controls in all three LA subregions. Rats that underwent CFC also had many significant differences, mostly contained to LaDL, whereas rats that underwent AAFC had no significant differences. Density heat maps visually depict mean Arc (E), c-Fos (F) and co-expression (G) in all five behavioural groups. Asterisks denote levels of statistical significance (* p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001; **** p ≤ 0.0001). CFC: contextual fear conditioning; UFC: unpaired fear conditioning; AAFC: adjusted auditory fear conditioning; TA: tone alone; CO: context only; LaDL: dorsolateral portion of the lateral amygdala; LaVM: ventromedial portion of the lateral amygdala; LaVL: ventrolateral portion of the lateral amygdala.


Immediate early gene expression in Bregma coordinate -3.24 mm was

compared across the five behavioural groups. Similar to above, ANOVAs were

utilised to reveal group differences in Arc expressing neurons in subregion LaDL

(F[4, 42] = 4.526, p < 0.01), LaVM (F[4, 41] = 3.762, p < 0.01) and LaVL (F[4, 40]

= 3.36 mm, p < 0.05), but not BLA (F[4, 42] = 1.021, p = 0.408) or BLP (F[4, 42] =

0.686, p = 0.606). Alternatively, significant group differences were observed in all

BLC subregions in c-Fos (LaDL F[4, 42] = 3.061, p < 0.05; LaVM F[4, 41] = 4.734,

p < 0.01; LaVL F[4, 41] = 6.587, p < 0.001; BLA F[4, 40] = 4.873, p < 0.01; BLP

F[4, 41] = 2.642, p < 0.05) and co-expressing neurons (LaDL F[4, 40] = 5.901, p <

0.001; LaVM F[4, 41] = 6.017, p < 0.001; LaVL F[4, 40] = 7.374, p < 0.001; BLA

F[4, 40] = 4.843, p < 0.01; BLP F[4, 40] = 3.451, p < 0.05). Bonferroni correct post-

hoc tests (see Figure 6) revealed group differences to primarily exist between UFC

rodents and the two controls. Some differences existed between CFC rodents and

controls in subregion LaVM, and between AAFC and controls in LaVM and LaVL.

Minor differences in subregion BLA between UFC and controls should be noted,

along with smaller differences in subregion BLP (data visually present in heat maps).

109


A Arc expression in LaDL revealed consistent significant differences between rats that underwent UFC as compared to both controls. Further differences were observed in this group in subregions LaVM and LaVL when compared to the TA control. No differences were identified in those that underwent CFC or AAFC as compared to controls in any lateral amygdala subregion. B Examination of c-Fos expression in LaDL revealed no differences in any groups. Alternatively, LaVM expression was significantly higher in all three conditioning groups as compared to the TA control. LaVL specific c-Fos expression was higher in the UFC as compared to both controls, and in the AAFC as compared to the TA control. C Co-localised neuron expression revealed consistent differences in rats that underwent UFC as compared to both controls in subregion LaDL, LaVM and LaVL. Those that underwent CFC exhibited an increase as compared to the TA control in subregions LaVM D As seen, the frequency of significant differences in the LA in those that underwent UFC was highest. Rats that

110

underwent CFC and AAFC had some differences in LaVM (both groups) and LaVL (for AAFC group only) as compared to controls. Density heat maps visually depict mean Arc (E), c-Fos (F) and co-expression (G) in all five behavioural groups. Asterisks denote levels of statistical significance (* p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001; **** p ≤ 0.0001). CFC: contextual fear conditioning; UFC: unpaired fear conditioning; AAFC: adjusted auditory fear conditioning; TA: tone alone; CO: context only; LaDL: dorsolateral portion of the lateral amygdala; LaVM: ventromedial portion of the lateral amygdala; LaVL: ventrolateral portion of the lateral amygdala.


Analysis of Bregma coordinate -3.12 mm revealed significant group

differences in Arc expressing neurons in BLC subregions LaDL (F[4, 40] = 6.505, p

< 0.001), LaVM (F[4, 38] = 7.354, p < 0.001), LaVL (F[4, 39] = 3.457, p < 0.05)

and BLA (F[4, 40] = 2.606, p = 0.05), but not BLP (F[4, 40] = 1.965, p = 0.118). In

c-Fos expressing neurons, significant group differences were observed in subregions

LaDL (F[4, 40] = 5.291, p < 0.01), LaVM (F[4, 41] = 4.843, p < 0.01), LaVL( F[4,

41] = 6.481, p < 0.001) and BLP (F[4, 41] = 2.522, p = 0.056), but not BLA (F[4,

41] = 3.542, p < 0.05 ). Alternatively, in co-expressing neurons, significant group

differences existed in all BLC subregions (LaDL F[4, 39] = 6.913, p < 0.001; LaVM

F[4, 39] = 10.411, p < 0.00001; LaVL F[4, 40] = 6.953, p < 0.001; BLA F[4, 39] =

7.997, p < 0.0001; BLP F[4, 38] = 3.941, p < 0.01). Bonferroni corrected post-hocs

(see Figure 7) revealed these differences to, once again, exist primarily between

rodents that underwent UFC as compared to controls. These differences were also

noted (albeit less consistently) in subregion BLA and BLP (data presented visually in

heat maps). Some minor differences were observed between those that underwent

CFC as compared to controls in subregion LaVM. Further, some differences were

present in rodents that underwent AAFC in LaDL (as compared to one control in c-

Fos expressing neurons) and LaVM (as compared to both controls in c-Fos

expressing neurons). Consistent differences were noted in those that underwent

111

AAFC in subregion LaVL when compared to both controls in c-Fos and co-

expressing neurons.


A Arc expression in LaDL and LaVM was significantly higher in rats that underwent UFC as compared to both controls. Furthermore, Arc expression was significantly higher in this group in subregion LaVL when compared to the TA control. B c-Fos expression in LaDL was significantly higher in the UFC and AAFC as compared to the CO control. Alternatively, LaVM expression was

112

significantly higher in the AAFC conditioning group as compared to both controls, while LaVL specific c-Fos expression was higher in the UFC and AAFC as compared to both controls. C Co-localised neuron expression was higher in rats that underwent UFC as compared to both controls in subregions LaDL, LaVM and LaVL. In addition, those that underwent CFC exhibited an increase as compared to the both control in subregions LaVM, whereas those that underwent AAFC had a significant increase in LaVL as compared to both controls D The frequency of significant differences in the LA in those that underwent UFC was, once again, highest. Rats that underwent CFC had minimal differences in LaVM, with rats that underwent AAFC exhibiting some differences throughout the LA. Density heat maps visually depict mean Arc (E), c-Fos (F) and co-expression (G) in all five behavioural groups. Asterisks denote levels of statistical significance (* p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001; **** p ≤ 0.0001). CFC: contextual fear conditioning; UFC: unpaired fear conditioning; AAFC: adjusted auditory fear conditioning; TA: tone alone; CO: context only; LaDL: dorsolateral portion of the lateral amygdala; LaVM: ventromedial portion of the lateral amygdala; LaVL: ventrolateral portion of the lateral amygdala.


Investigations of Bregma coordinate -3.00 mm revealed the most number of

group differences. Group differences were present in Arc expression in all

subregions except BLP (LaDL F[4, 41] = 5.867, p < 0.001; LaVM F[4, 37] = 6.643,

p < 0.001; LaVL F[4, 38] = 3.842, p < 0.05; BLA F[4, 39] = 3.362, p < 0.05; BLP

F[4, 40] = 1.275, p = 0.296). Group differences in c-Fos expressing neurons were

present in all subregions except BLA (LaDL F[4, 41] = 6.528, p < 0.001; LaVM F[4,

41] = 4.267, p < 0.01; LaVL F[4, 41] = 3.938, p < 0.01; BLA F[4, 41] = 2.261, p =

0.079; BLP F[4, 41] = 3.139, p < 0.05). Similarly, group differences in co-expressing

neurons were present in all subregions except BLA (LaDL F[4, 41] = 7.620, p <

0.001; LaVM F[4, 41] = 7.552, p < 0.001; LaVL F[4, 40] = 7.546, p < 0.001; BLA

F[4, 41] = 2.144, p = 0.093; BLP F[4, 39] = 2.716, p < 0.05). Bonferroni correct

post-hoc tests (see Figure 8) identified group differences to consistently exist

between rats that were CFC as compared to both controls in LaDL. Some further

differences existed between those that underwent CFC as compared to controls in

LaVM and LaVL. Rodents that underwent UFC had consistent significant increases

as compared to controls in subregions LaDL and LaVM. Further differences were

noted in LaVM. Once again, rodents that underwent AAFC had little amygdala-

113

related differences as compared to controls. Those that were present existed in LaVM

(as compared to one control in co-expressing neurons) and LaVL (as compared to

both controls in co-expressing neurons).

114


A Arc expression in LaDL and revealed significant increases between rats that underwent CFC and UFC as compared to both controls. Further differences were observed in the rats that underwent CFC as compared to TA rats in subregion LaVM, and rats that underwent UFC as compared to both controls in LaVM, and as compared to TA controls in LaVL. B Similar to Arc expression, c-Fos expression in LaDL was significantly higher in the CFC and UFC as compared to both controls. LaVM expression was significantly higher in the UFC conditioning group as compared to both controls, while LaVL expression was higher in the UFC as compared to the CO control. C Once again, co-localised neuron expression was, once again, significantly higher in rats that underwent CFC and UFC as compared to both controls in subregion LaDL. Further differences in LaVM were observed in those that received CFC and UFC as compared to both controls, and between those that received AAFC and the TA control. In subregion LaVL significant differences were observed between those that received UFC and AAFC as compared to both controls, and between these that received CFC and the TA controls. D The frequency of significant differences in the LA in those that underwent UFC was, once again, highest. Rats that underwent CFC also exhibited large differences in LaDL, with some minor differences in LaVM and LaVL. AAFC produces, once again, smallest differences. Density heat maps visually depict mean Arc (E), c-Fos (F) and co-expression (G) in all five behavioural groups. Asterisks denote levels of statistical significance (* p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001; **** p ≤ 0.0001). CFC: contextual fear conditioning; UFC: unpaired fear conditioning; AAFC: adjusted auditory fear conditioning; TA: tone alone; CO: context only; LaDL: dorsolateral portion of the lateral amygdala; LaVM: ventromedial portion of the lateral amygdala; LaVL: ventrolateral portion of the lateral amygdala.

3.4.3 Summary of Results

In the current study, we evaluated involvement of the BLC following three

different forms of fear conditioning. The current study extensively evaluates

amygdala IEG expression, focusing on Arc, c-Fos and their co-localisation in

Bregma coordinates -3.36 mm, -3.24 mm, -3.12 mm and -3.00 mm. In this section

results are visually summarised (see Figure 9 and 10). As seen, a role for LaDL and

LaVM exists following CFC. Alternatively, widespread BLC IEG expression is

noted following UFC, with limited role for BLC noted following AAFC, besides a

minor difference in IEG expression at subregion LaVL.

115

Figure 3-9: Schematic representation of IEG expression following conditioning.

Within each sub-region, statistically significant differences between conditioned rats as compared to both controls are represented by dots. Stronger significance results in more representative dots (p ≤ 0.05 = five dots; p ≤ 0.01 = 10 dots; p ≤ 0.001= 15 dots; p ≤ 0.0001 = 20 dots). Green dots represent significant difference in Arc expression, red dots represent c-Fos expression and yellow dots represent co-localised expression. As seen, following CFC a specific role for LaDL is present at Bregma coordinate -3.36 mm and -3.00 mm, whereas a more consistent, albeit weaker role for LaVM is present in all four Bregma coordinates. Comparatively, following UFC, a largescale and consistent role for the entire BLC is noted, except at Bregma coordinate -3.00 whereby IEG expression seems mostly localised to the LA. Following AAFC, little consistent differences are noted in BLC, besides at subregion LaVL, whereby a c-Fos and co-localised neuron expression seemed to increase at Bregma coordinates -3.12 mm and -3.00 mm. BLC: basolateral amygdala complex; LA: lateral amygdala; LaDL: dorsolateral portion of the lateral amygdala; LaVM: ventromedial portion of the lateral amygdala; LaVL: ventrolateral portion of the lateral amygdala; IEG: immediate early genes CFC: contextual fear conditioning; UFC: unpaired fear conditioning; AAFC: adjusted auditory fear conditioning.

116

Figure 3-10: Schematic representation of total IEG expression following conditioning.

Merged IEG expression across Bregma coordinates -3.36 mm, -3.24 mm, -3.12 mm and -3.00 mm reveal a differing pattern of activation following the three different fear conditioning protocols. IEG expression was seen in LaDL and LaVM following CFC, while widespread IEG was present within the BLC following UFC. IEG expression was found to be much lower following the AAFC protocol. CFC: contextual fear conditioning; UFC: unpaired fear conditioning; AAFC: adjusted auditory fear conditioning; BLC: basolateral amygdala complex; LaDL: dorsolateral portion of the lateral amygdala; LaVM: ventromedial portion of the lateral amygdala.

3.5 Discussion

In this study we examined the behavioural and anatomical changes following

fear conditioning protocols designed to isolate and investigate the component of

contextual memory within the basolateral amygdala. We studied contextual fear

memories by way of two different protocols: a standard CFC protocol and an UFC

protocol. Behavioural and anatomical differences in rodents that underwent CFC and

UFC were compared to a group that underwent an AAFC protocol. Behaviourally,

we show both CFC and UFC to produce fear memories to context but not tone. The

AAFC protocol employed here was successful in creating fear memories to tone with

accompanying background fear memories to context significantly attenuated. This

suggests creation of a context-removed or context-reduced auditory fear memory.

Data reported here suggest differential contribution of BLC subregions

following different fear conditioning protocols. In this investigation of the BLC, we

evaluate differences in the expression of Arc, c-Fos and their co-localisation in five

117

amygdala subregions across four rostral to caudal planes identified by consecutive

atlas Bregma coordinates. We demonstrate a role for the LA in the creation of fear

memories to context. Interestingly, LA involvement differed depending on the fear

conditioning protocol. Contextual fear conditioning produced specific changes to

LaDL and LaVM, compared to UFC which produced wide-scale BLC changes. Our

data suggests that the creation of context-removed/context-reduced auditory fear

memories become less dependent upon BLC, with only minor differences being

consistently present in subregion LaVL. Cumulatively, data suggest contextual fear

memories to rely heavily upon BLC. Alternatively, auditory fear memories, with

accompanying background contextual fear memories removed, seem to be less

reliant upon the BLC, and perhaps more reliant upon alternative brain areas.

3.5.1 Contextual Fear Conditioning

Contextual fear memories produced with a standard CFC protocol rely upon

the LaDL and LaVM subregion of the BLC (see Figure 9 and 10). In these

subregions, we found rats that underwent CFC to have significantly more neurons

expressing Arc and c-Fos as compared to the TA and CO control groups. Further,

examinations of co-localising neurons suggest that a unique population of neurons

express both Arc and c-Fos in response to CFC in these regions. Cumulatively, data

highlight contextual fear memory formation to be reliant upon LaDL and LaVM.

Many previous investigations have identified increases in IEG expression following

contextual fear memory formation (Barot et al., 2009; Hall, Thomas, & Everitt,

2001a; Malkani & Rosen, 2000; Perez-Villalba, Mackintosh, & Canales, 2008;

Trogrlic, Wilson, Newman, & Murphy, 2011; Y. M. Wilson & Murphy, 2009;

Zelikowsky, Hersman, Chawla, Barnes, & Fanselow, 2014). However, these

investigations examined the BLC as a whole (Barot et al., 2009; Hall et al., 2001a),

118

or only explored or found differences in the LaVL (Trogrlic et al., 2011; Y. M.

Wilson & Murphy, 2009), the BLA (Perez-Villalba et al., 2008), or the basal portion

of the BLC (Zelikowsky et al., 2014). Only one previous study, to our knowledge,

identified a specific role for LaDL following CFC (Malkani & Rosen, 2000). When

these studies, along with new data provided here, are viewed together, little

advancement is made in the ability to identify a specific BLC subregion whereby

contextual fear memories may be located. However, the in-depth and voluminous

analysis of BLC provided here can explain why particular subregion activation

following contextual fear memory formation cannot be consistently identified.

Data presented here provides some insight into the organisation of contextual

fear memories. Reported differences in LaDL between Arc, c-Fos and co-expressing

neurons were consistently present in Bregma coordinates -3.36 mm and -3.00 mm,

but not -3.12 mm or -3.24 mm. Alternatively, LaVM involvement was present in all

Bregma coordinates examined, but the amount of increase in IEG expression differed

depending upon the Bregma coordinate at which analyses occurred. The various

investigations above, demonstrate differential involvement of BLC subregions

following contextual fear memory formation. In these studies, an array of divergent

rostral-caudal locations are examined. Our data may explain the contradictory results

from previous studies. Our data provides evidence for a potential dynamic and

changing role for LaDL and LaVM involvement, depending upon rostral-caudal

location, following contextual fear memory formation.

3.5.2 Unpaired Fear Conditioning

Unpaired fear conditioning resulted in a broad and consistent (across all

Bregma coordinates studied) increase in Arc expressing, c-Fos expressing and co-

expressing neurons at subregions LaDL, LaVM and LaVL. The LA subregion

119

changes were accompanied by a less pronounced level of IEG activation in the BLA

and BLP (see Figure 9 and 10). Data suggests UFC leads to consistent and high BLC

activation. Immediate early gene expression was highest in these rats as compared to

rodents that underwent CFC or AFC. Interestingly, behavioural data did not show

UFC to result in stronger fear memories. More importantly, these data are in contrast

to many previously reported studies showing amygdala activation/plasticity to be

comparatively low following UFC, as opposed to AFC (H. C. Bergstrom et al., 2012;

Hadley C Bergstrom et al., 2011; Majak & Pitkänen, 2003; McKernan & Shinnick-

Gallagher, 1997; Radley et al., 2006; Michael T. Rogan et al., 1997). Unpaired fear

conditioning protocols are often utilised as controls for AFC. Leading hypotheses

suggest that, while tone alone or foot-shock alone may activate amygdala neurons,

the pairing of tone and foot-shock leads to plasticity in a unique set of BLC neurons

(Joseph LeDoux, 2003; Romanski et al., 1993). The UFC protocol acts as a

behavioural and anatomical control, as associative fear memories are not produced,

and therefore BLC involvement is reduced (H. C. Bergstrom et al., 2012; Hadley C

Bergstrom et al., 2011; Majak & Pitkänen, 2003; McKernan & Shinnick-Gallagher,

1997; Radley et al., 2006; Michael T. Rogan et al., 1997). Data from this study,

however, contrast these findings showing high amygdala activation following UFC.

Associative learning can occur without explicit presentation of two stimuli.

Foreground CFC is a prime example of this (Michael S. Fanselow, 1980).

Foreground contextual fear memories are formed by association of a foot-shock to a

context (Calandreau, Desmedt, Decorte, & Jaffard, 2005; Calandreau et al., 2006;

Desmedt, Garcia, & Jaffard, 1998; Russel G Phillips & LeDoux, 1994; Trifilieff et

al., 2007; Trifilieff et al., 2006). This context (for example, the chamber) is

constantly present as the environment (or backdrop) under which psychological

120

processes occur. Unpaired fear conditioning is identical to CFC, except the ‘context’

now includes further stimuli (in this case, auditory tones). These stimuli are not

paired with the foot-shock, and therefore are a component of the context.

Consequently, amygdala involvement following UFC is not surprising. In fact,

previous research has demonstrated significantly higher levels of pMAPK expression

following UFC, but not AFC, in the BLC (Trifilieff et al., 2007). Despite this,

Trifilieff et al. (2007) is the only paper, to our knowledge, showing significant

amygdala involvement following UFC, but only limited involvement following AFC,

as compared to controls. Previously we showed a non-significant trend level increase

of pMAPK expressing neurons following UFC as compared to controls in LaDL,

LaVM and LaVL (H. C. Bergstrom et al., 2012; Hadley C Bergstrom et al., 2011).

Similar results have been produced by others (Hall, Thomas, & Everitt, 2001b;

Schafe et al., 2000). However, no further research has demonstrated consistently high

levels of amygdala involvement, as compared to controls, following UFC.

Consequently, based on previous evidence, it was unexpected that UFC produced

high BLC IEG expression, while AFC did not. Data presented here, however, can be

explained when the adjustment made to the AFC protocol is considered. Our specific

AAFC protocol, whereby latent inhibition reduced background contextual fear, may

have reduced amygdala involvement. Without such latent inhibition, amygdala

involvement may have been above and beyond that of rats that underwent UFC as

extensively demonstrated (H. C. Bergstrom et al., 2012; Hadley C Bergstrom et al.,

2011; Majak & Pitkänen, 2003; McKernan & Shinnick-Gallagher, 1997; Radley et

al., 2006; Michael T. Rogan et al., 1997).

The UFC protocol here activated nearly the entire BLC, whereas the CFC

protocol did not. This is despite both protocols producing contextual fear memories.

121

In a 1993 investigation of LA activity during fear conditioning, subregion LaDL

responded both to somatosensory (foot-shock) and acoustic (clicks) stimuli

(Romanski et al., 1993). In the current study, presentation of both foot-shock and

acoustic tones, although not exclusively paired, may have led to large scale BLC

activation. Alternate explanations for large scale BLC activation include a non-

significant trend level development of auditory fear memories as a result of UFC, as

seen in the current and some previous investigations (H. C. Bergstrom et al., 2012;

Hadley C Bergstrom et al., 2011; Hall et al., 2001b). Nevertheless, UFC seems to

produce complex associative fear memories which may be responsible for the greater

BLC activation.

3.5.3 Auditory Fear Conditioning

In contrast to our contextual fear memory data, we found limited consistent

pattern of BLC IEG activity in rodents that underwent AAFC (see Figures 10 and

11). Countless investigations have consistently identified numerous plasticity and

activity related molecules to be highly expressed in the BLC following auditory fear

memory formation (see reviews by (Hadley C. Bergstrom, 2016; L. R. Johnson et al.,

2012; J. J. Kim & Jung, 2006; J. LeDoux, 2000; Joseph E. LeDoux, 2014; Maren,

2011; Sah, Westbrook, & Lüthi, 2008). Data here, therefore suggests that the

removal of context, by way of latent inhibition, significantly reduces BLC

involvement following auditory fear memory formation to a level whereby some,

albeit limited, LaVL IEG expression is present. Two possible explanations for this

reduction in IEG expression are considered. First, the reduction of background

contextual fear memories may have directly altered the level of IEG expression in the

BLC. This indicates that background contextual fear memories (that are present

following all forms of discrete fear conditioning, such as AFC) may be responsible

122

for a major degree of BLC activation/plasticity. Some evidence for this is presented

in our previous research, where we demonstrated visual and auditory fear memories

(both discrete fear memories with the contextual fear memory in the background) to

activate a similar density of plasticity-related pMAPK-expressing neurons in the LA

(H. C. Bergstrom & L. R. Johnson, 2014). Interestingly, the anatomical distribution

of pMAPK expressing neurons differed between these groups (H. C. Bergstrom & L.

R. Johnson, 2014). This suggests that alterations to discrete elements that are paired

with foot-shocks (auditory versus visual in this case) in fear conditioning have less

impact on the overall number of neurons undergoing plasticity in the amygdala when

background context is kept stable. Alternatively, context-related modifications (CFC

group versus UFC in this study) produce clear alterations in the number of neurons

expressing IEGs in the amygdala. This highlights the large and quantitative

importance of context in the amygdala during fear memory consolidation.

An alternative explanation for the reduction in LA IEG expression relates to

the clear reduction in freezing to tone in the AAFC group, as compared to previous

investigations. In our previous reports, when the same number and intensity of foot-

shocks were utilised, rodents froze to tone 96% of the time (Bergstrom et al., 2012).

Only with fewer presentations (two tone-shock pairings) and less severe (0.6mA

foot-shocks) fear conditioning does freezing to tone drop to 55% (Bergstrom et al.,

2014), which is still higher than the 49% reported here. This suggests the removal of

context, by way of latent inhibition, directly affects the ability to effectively create

auditory fear memories. Previous investigations have provided evidence for this

(Balaz, Capra, Hartl, & Miller, 1981; Balaz, Capra, Kasprow, & Miller, 1982;

Urcelay & Miller, 2010). These investigations, pioneered by Miller, demonstrated

some minor reductions to auditory fear memory (testing occurred in training context)

123

following attenuation or abolishment of contextual fear via latent inhibition or

extinction. Importantly, these reductions were shown to become severe when fear

memory testing occurred in a new context (Balaz et al., 1981; Balaz et al., 1982).

This suggests that context memory (sever reduction of fear when contextual fear is

attenuated and testing is in a new chamber), but not context fear memory (minor

reductions in fear when contextual fear is attenuated but testing is in the same

chamber) influences the association of discrete stimuli with foot-shocks. Two

primary explanations have attempted to explain how context influences discrete

stimuli and foot-shock associations (Balaz et al., 1981; Balaz et al., 1982; Grahame,

Hallam, Geier, & Miller, 1990; Urcelay & Miller, 2010). First, the conditioning

context is hypothesised to be responsible for retrieving associations between the

discrete stimuli and foot-shock – without context recall via exposure, or other

mechanisms, the discrete stimuli and foot-shock association is not effectively

retrieved (Balaz et al., 1982). Alternatively, attenuation in fear, especially following

latent inhibition, arises due to discrimination. Latent inhibition allows rodents to

more effectively identify and discriminate one context from another. Therefore, upon

placement to a new context, rodents quickly determine that it is not the same as the

training context – resulting in attenuation of fear (Balaz et al., 1982). While these

studies examine the possibility of deficits at recall, anatomical data reported here

(reduction of BLC involvement immediately following AAFC) suggest that latent

inhibition may cause encoding deficits.

Following extreme prior exposure to a discrete or contextual CS, conditioned

fear is attenuated (Barot et al., 2009; R. E. Lubow, 1965; R. E. Lubow & Moore,

1959). This occurs due to an encoding deficit, arising from competition for

associative strength. For contextual fear memory, context pre-exposure leads to the

124

development of a neutral associative contextual memory which, following CFC,

competes with the fearful associative memory (Bouton, 1993; Nicholas Chaaya et al.,

2018; Traverso, Ruiz, & De la Casa, 2012). Background contextual fear was

attenuated in the AAFC group via this mechanism. Importantly, we propose that the

(foreground) auditory fear may have also been attenuated by related mechanisms. It

is hypothesised that the auditory tone used to condition rodents became included as

part of the ‘neutral’ context during training. Evidence for this is provided in studies

of the hippocampus. In these studies, the addition of a discrete stimulus (such as an

auditory tone) to a known context largely alters place cell (neurons that fire to form

spatial representations of ‘contexts’) activity (M. I. Anderson & Jeffery, 2003; D. M.

Smith & Mizumori, 2006a). Further investigations have demonstrated a complete

remapping of hippocampal place cell activity to occur following similar

environmental changes to a context that rats have been heavily exposed to (Bostock,

Muller, & Kubie, 1991; Jeffery, Gilbert, Burton, & Strudwick, 2003). In the current

study, it is hypothesised that an encoding deficit occurred due to competition for

associative strength between the ‘neutral’ context – which, following a single

presentation, now includes the auditory tone – and the aversive foot-shock. While the

role of context is still being outlined during discrete fear conditioning (Urcelay &

Miller, 2010), this hypothesis suggests that latent inhibition of background

contextual fear memories, directly alters the efficacy of auditory fear conditioning.

This serves as a limitation of the AAFC protocol. Future research is required to

investigate uninhibited auditory fear memories, with the contextual component

removed.

For the above reason, the data here is reported cautiously. Additionally,

methodological limitations have produced some caveats. Firstly, to correct for

125

multiple comparisons, a Bonferroni adjustment was applied to all statistical analyses.

The Bonferroni adjustment has been cited to be too severe and capable of producing

type II errors (Perneger, 1998). Indeed, when evaluated without the Bonferroni

adjustment (data not reported), a pattern of IEG expression arises in this AAFC

group, mostly in c-Fos expressing and co-expressing, but not Arc expressing

neurons. These statistical differences, however, are much weaker than those seen in

the two contextual fear memory groups (CFC and UFC), and lose statistical

significance following minor and less severe adjustments to alpha. Furthermore, LA

involvement was not compared in this group to a non-adjusted AFC group. However,

the involvement of LA following non-adjusted AFC has been previous delineated on

numerous occasions (see reviews by (Hadley C. Bergstrom, 2016; L. R. Johnson et

al., 2012; J. LeDoux, 2000; Joseph E. LeDoux, 2014)). Therefore, we did not

determine a need for this group.

3.5.4 Technical Considerations

Data reported in the current study highlight a dependence on LA activation

following contextual fear memory formation, but a reduced role following auditory

fear memory formation. These conclusions are made with the IEGs Arc and c-Fos,

demonstrating their reliability as indicators of neuronal activity representing

differences in behaviour. Due to the large-scale investigations conducted here,

behavioural differences were found to produce a number of statistically significant

anatomical alterations. In this voluminous exploration of amygdala, we focus on

consistent (across multiple Bregma coordinates and also identified in Arc, c-Fos and

co-expressing neurons) statistical differences between conditioned rats and controls.

Many previous investigations focus primarily on differences in one Bregma

coordinate, utilising a single plasticity or activity marker, and with one behavioural

126

control. Here we sought to develop and use a multifaceted and more powerful

approach, as it reduces the chances of obtaining false-positives. For example, a

difference between groups may exist in one Bregma coordinate when measured with

one marker but may not exist in other Bregma coordinates or when measured with

many other related markers. A further advantage to the current approach is the ability

to identify patterns. For example, in the present study, CFC led to an increase in

LaDL IEG expression in two distinct Bregma coordinates (-3.00 mm and -3.36 mm),

suggesting CFC may activate particular areas of LaDL. Evaluation of a single

Bregma coordinate may not have accurately identified the complex involvement of

LaDL following CFC. The ability to highlight the complex involvement of various

brain regions following fear conditioning of various forms is essential. Treatment of

post-traumatic stress disorder (PTSD) relies upon a clear and thorough understanding

of the neural mechanisms involved following fear memory formation (L. R. Johnson

et al., 2012; Maren et al., 2013). Therefore, while we do not discount individual and

less consistent differences between groups, emphasis is on those differences which

were reliably and consistently identified. These highlight a clear pattern of amygdala

involvement which is less likely to have arisen due to sampling and statistical error.

3.5.5 General Discussion

In this study we found modification to fear memory conditioning protocols

leads to differing IEG activation patterns in BLC. While foot-shock presentation, and

corresponding freezing to context, did not differ between these two groups (CFC and

UFC), the ‘context’ did (i.e. the presence or absence of tone in the in the UFC box

versus CFC box, respectively). The hippocampus has high functional significance in

context/spatial memory formation and contextual fear memory formation (see

(Nicholas Chaaya et al., 2018) for review, along with recent chemogenetic and

127

optogenetic data (Huff, Emmons, Narayanan, & LaLumiere, 2016; Park et al., 2016;

Sparta et al., 2014; C. Xu et al., 2016)). We, therefore, hypothesise the alterations in

BLC following the two different fear conditioning protocols may be regulated by the

hippocampus. This highlights the interaction between amygdala and hippocampus as

essential brain regions to contextual fear memory formation (see (Nicholas Chaaya et

al., 2018)).

In this study, we found that when auditory fear memories are formed with

background contextual fear memories removed, BLC activation is significantly

reduced. Reduction in BLC activation suggests that background contextual fear

memories, formed by standard and non-adjusted AFC protocols, may be responsible

for a significant portion of amygdala activation found in previous investigations (see

(Hadley C. Bergstrom, 2016; L. R. Johnson et al., 2012; J. LeDoux, 2000; Joseph E.

LeDoux, 2014) for reviews of these investigations). This suggests auditory fear

memories may be modulated by context/spatial memory specific brain regions;

namely the dorsal portion of the hippocampus (Nicholas Chaaya et al., 2018).

Indeed, this was previously demonstrated by Fanselow and Quinn (Pierson, Pullins,

& Quinn, 2015; Jennifer J. Quinn, Loya, Ma, & Fanselow, 2005; Jennifer J Quinn,

Wied, Ma, Tinsley, & Fanselow, 2008). Similar to contextual fear memory

formation, we conclude that the interaction between amygdala and hippocampus may

be important to auditory fear memory formation. This warrants further investigation.

128

Micro-Topography of Fear Memory Consolidation and

Extinction Retrieval within Prefrontal Cortex and Amygdala

This chapter comprises the following published article: Jacques, A., Chaaya, N., Hettiarachchi, C., Carmody, M., Beecher, K., Belmer, A., Chehrehasa, F., Bartlett, S., Battle, A.R., Johnson, L. R. Micro-Topography of Fear Memory Consolidation and Extinction Retrieval within Prefrontal Cortex and Amygdala. Psychopharmacology. Published 4th January, 2019 https://link.springer.com/article/10.1007%2Fs00213-018-5068-4

129








In the case of chapter 4: Micro-Topography of Fear Memory Consolidation and Extinction Retrieval within Prefrontal Cortex and Amygdala

Publication status: Published

Contributor Statement of contribution*Angela Jacques Involved in the conception and design of the project, behavioural and

laboratory experiments, analysed data, created the figures and wrote and edited the manuscript.

Nicholas Chaaya Assisted with editing the manuscript, behavioural and laboratory experiments.

Cheimi Hettiarachchi Assisted with laboratory experiments.

Marie Carmody Assisted with behavioural experiments.

Kate Beecher Assisted with creating figures.

Arnauld Belmer Assisted with editing the manuscript.

Fatemeh Chehrehasa Assisted with reviewing and editing the manuscript.

Selena Bartlett Assisted with editing the manuscript.

Andrew Battle Assisted with editing the manuscript.

Luke Johnson Involved in the conception and design of the project, assisted with behavioural experiments and reviewing and editing the manuscript.

QUT Verified

Signature

130

4.1 Abstract

The precise neural circuitry that encodes fear memory and its extinction within

the brain are not yet fully understood. Fearful memories can be persistent, resistant to

extinction and associated with psychiatric disorders, especially post-traumatic stress

disorder (PTSD). Here, we investigated the micro-topography of neurons activated

during the recall of an extinguished fear memory, and the influence of time, on this

micro-topography. We used the plasticity-related phosphorylated mitogen activated

protein kinase (pMAPK) to identify neurons activated in the recall of consolidated

and extinguished auditory Pavlovian fear memories in rats. We used quantitatively

matched brain regions to investigate activity in the amygdala and prefrontal cortex.

Recall of a consolidated, non-extinguished auditory fear memory resulted in a

significantly greater number of activated neurons located in the dorsolateral sub-

division of the lateral amygdala (LADL) when recalled 24 hours after consolidation

but not when recalled seven days later. We found the recall of an extinction memory

was associated with pMAPK activation in the ventrolateral sub-division of the lateral

amygdala (LAVL). Next, we showed the pattern of pMAPK expression in the

prelimbic cortex differed spatially following temporal variation in the recall of that

memory. The deep and superficial layers of the pre-limbic cortex were engaged in

recent recall of a fear memory, but only the superficial layers were recruited if the

recall occurred seven days later. Collectively, our findings demonstrate a functional

micro-topography of auditory fear memory during consolidation and extinction at the

micro-anatomical level within the lateral amygdala and medial prefrontal cortex.

131

Keywords:

pMAPK; Topography; Neuroplasticity; Amygdala; Medial prefrontal cortex; Extinction; Fear conditioning.

4.2 Introduction

Individuals with post-traumatic stress disorder (PTSD) have difficulty learning

that cues previously associated with a threat no longer pose danger once removed

from the threatening situation, suggesting deficits in extinction learning (Luke R

Johnson et al., 2012; Yehuda, 2002) Pavlovian, or classical fear conditioning, has

long been used to study an animal’s response to threatening situations (Maren, 2001;

Rescorla, 1988; Schafe, Nader, Blair, & LeDoux, 2001), which involves a form of

associative learning. An animal will exhibit a defensive action such as freezing, in

response to a conditioned stimulus (CS) paired with an unpleasant, unconditioned

stimulus (US) (á. Fendt & M. Fanselow, 1999). The new associative memories

undergo consolidation, a process of stabilization from short-term to long-term

memories (Hebb, 1949; Luke R Johnson et al., 2012; Nader et al., 2000). In auditory

Pavlovian fear conditioning a learned association forms between a mild foot shock

(US) paired with an auditory tone (CS) (S. Duvarci & D. Pare, 2014; Michael S.

Fanselow & Gale, 2003; Luke R Johnson et al., 2012; J. LeDoux & Daw, 2018;

Maren, 2001, 2011; Maren & Quirk, 2004; H. C. Pape & D. Pare, 2010)

It is generally accepted that only a small subset of neurons within a given

population is required to encode a fear memory (Hadley C Bergstrom & Luke R

Johnson, 2014; X. Liu et al., 2012; Rumpel, LeDoux, Zador, & Malinow, 2005). This

has help direct investigations into localizations and mechanisms of memory

consolidation and extinction within amygdala and prefrontal cortex (Baker, Bisby, &

132

Richardson, 2016; Hadley C Bergstrom & Luke R Johnson, 2014; Bukalo et al.,

2015; Maren, 2011). Studies on the molecular mechanisms of fear memories have

identified activation of the extracellular signal-regulated kinase-mitogen-activated

protein kinase (ERK/MAPK) pathway as necessary for memory consolidation and

extinction learning (Adams & Sweatt, 2002; Hadley C Bergstrom et al., 2011;

Cestari, Rossi-Arnaud, Saraulli, & Costanzi, 2014; Herry et al., 2006; Schafe et al.,

2000). Molecular and cellular measures to date have provided evidence that the

medial prefrontal cortex (mPFC) and the amygdala engage varied neuronal networks

during fear memory consolidation, extinction and retrieval (Bukalo et al., 2015; H. J.

Lee, R. P. Haberman, R. Roquet, & M. H. Monfils, 2016; Milad & Quirk, 2002;

Shin, Rauch, & Pitman, 2006). This is consistent with studies in human subjects with

PTSD have been shown to have morphological changes and functional abnormalities

in the PFC suggesting compromised extinction circuits associated with the disorder

(Milad & Quirk, 2012).

The micro-topographical organization of neurons encoding fear memories

and their extinction memories remains elusive. Moreover, the degree to which

neuronal architecture that encodes memories is stable over time is not fully known.

Topographic stability, or reproducibility, is a function of repeated reappearance of

activation within a mapped space. We hypothesized that the spatial distribution of

pMAPK within the ERK/MAPK cascade in response to remote fear memory recall

and remote extinction memory recall could be visualised using density heat maps and

would occur in different sub regions of the amygdala. Furthermore pMAPK positive

neurons would be present in both the infralimbic and prelimbic regions of the mPFC,

however, their spatial distribution would be predominantly prelimbic during

conditioning recall and extinction recall. In addition, we predicted the magnitude of

133

plasticity would vary between remote and recent fear memory recall. Specifically, we

suggested the neurons expressing pMAPK would be distributed in distinct micro-

anatomical regions dependent on their memory group and be visible within distinct

cortical layers through the generation of density maps. Our objectives included

establishing the microanatomical properties of a marker of neuroplasticity - pMAPK

in subregions of the amygdala and the mPFC during fear memory and remote

extinction memory recall. We sought to define the specific sub regions undergoing

neuroplastic changes due to the recall of a remote extinction memory and define

differences if any, with the spatial distribution of this neuroplasticity after recent and

remote fear memory retrieval. We utilizedclassicalPavlovianconditioningandan

extinction paradigm spanning 3 days, combined with cytochemical

quantifications to determine themicroanatomy of fearmemory consolidation

andremoterecallofanextinctionmemory.Our method of quantitative functional

microanatomy and spatial analysis of micro-anatomical sub-regions involved in fear

and extinction memory recall in rodents was used to measure stable neuronal patterns

and differences among animals.

4.3 Material and Methods

4.3.1 Subjects

Experimentally naive adult male Sprague-Dawley rats (supplied by Animal

Resource Center, ARC, Western Australia), were housed in pairs in temperature (≈

24 °C) and humidity (35 %) controlled clear Plexiglas cages, maintained on a 12-

hour light/dark cycle. Behavioral procedures were conducted during the light cycle

as fear conditioning is documented to be more effective during the nocturnal phase

134

for rodents i.e. when the lights are on (Albrecht & Stork, 2017). Food and water were

provided ad libitum. Rats were acclimatized to the vivarium for 7 days prior to

training. At time of memory recall testing rats weighed 326 ± 3.6 g (276-391 g). All

procedures were conducted in compliance with the Animal Welfare Unit, The

University of Queensland Research and Innovation Ethics Committee and the

Research Ethics Committee of the Queensland University of Technology, Australia.

Procedures complied with policies, regulations and ethical standards for animal

experimentation, in accordance with the Queensland Government Animal Research

Act 2001, associated Animal Care and Protection Regulations (2002 and 2008), and

the Australian Code for the Care and Use of Animals for Scientific Purposes, 8th

Edition (National Health and Medical Research Council, 2013).

4.3.2 Apparatus

Two clear Plexiglas conditioning chambers were utilized for behavioral

procedures: Context A and Context B. Each was encased in an acoustic isolation box

(Coulbourn Instruments, Allentown, PA, USA). Background noise level was

measured at 55 dB, using a sound level meter (Digitech Professional Sound Level

Meter QM1592). Each chamber was equipped with a speaker, a low-level house light

(2-3 lux), an infrared light and an infrared camera. Modification to the chambers

occurred to create unique testing environments for fear conditioning or extinction

and to restrict any reaction to context. Context A contained a stainless-steel grid floor

connected to a shock generator and computer (Freeze frame software, Coulbourn

instruments). The novel odorant and cleaning agent, ethanol (70 %) was used in

Context A. To differentiate between contexts, Context B contained a plastic floor

covered with fresh bedding and internal colored decoration on the walls and ceiling.

135

Orange scented antibacterial soap was used to produce a smell unique to Context B.

The chambers were cleaned between each animal.

4.3.3 Behavioral procedures

A total of 52 rats were randomized into four groups (Figure 1). Extinction

recall (ER, n = 14), animals underwent auditory fear conditioning followed by

extinction training. Remote fear memory recall (7dR, n = 12), was a no extinction

control group that underwent auditory fear conditioning followed by no extinction

training. A recent fear memory recall (1dR, n = 16) group acted as a control for fear

memory consolidation. This group underwent auditory fear conditioning 24 h prior to

a memory recall test on the final day of behavioral training. Box control (BC, n =

10), a group of naive animals that underwent context training for the same duration

as the extinguished and non-extinguished groups (Figure 1). In addition to providing

a control to the extinction group, the no extinction animals represented the remote (7

d) retrieval of a fear memory in comparison to the fear conditioned group, where

recent memory retrieval occurred 24 h after memory formation. Therefore, the

groups were named accordingly.

136

Figure 4-1: Experimental design.

Experimental design of extended extinction paradigm with controls (no extinction, fear conditioned alone and naive). An extinction recall (ER, n = 14) group underwent auditory fear conditioning followed by three days extinction training followed by three days in their home cage and a fear memory extinction retrieval test. A no extinction group underwent auditory fear conditioning, context exposure for three days (no tone) and home cage prior to a remote fear memory recall test performed seven days after conditioning (7dR, n = 12). A box control group followed this protocol but received no stimuli (BC, n = 10). A group of rats remained in their home cages until the seventh day of experimentation when they received auditory fear conditioning. One day later they underwent a recent fear memory recall test (1dR, n = 16).

For two days prior to training, rats were habituated to each context for 30

minutes. On day one the ER and 7dR groups underwent three minutes of acclimation

to context A, followed by a 10-minute conditioning protocol involving three pairings

of an auditory conditioned stimulus (CS, tone - 5kHz, 75 dB, 20 s) that co-terminated

with an unconditioned stimulus (US, foot shock - 0.6 mA, 500 ms). The 1dR fear

conditioning control group underwent the same conditioning on day seven. Stimuli

were controlled through Freeze Frame software (Coulbourn Instruments) and were

separated by a mean inter-trial interval of 180 s. Rats were returned to home cages 60

s after the final stimulus presentation. 24 h, 48 h and 72 h after fear conditioning the

137

ER group underwent 30 minutes of extinction training (20 x CS alone, 5 kHz, 75 dB,

20 s) in Context B after which they remained in their home cages for three days to

allow for memory consolidation and a delay before memory was recalled. The 7dR

and BC groups followed this protocol, but received no stimuli in context B. The

intertrial interval throughout training and testing averaged 180s.

On the eighth day the ER, 7dR and 1dR groups underwent a 10-minute fear

memory test consisting of three 20 s CS presentations to test recall of the

consolidated auditory cued fear memory or extinction memory. BC animals were

exposed to context B for the same duration but did not undergo memory testing.

Freezing, an accepted behavioral index to quantitate the formation of a CS-US

association (Blanchard & Blanchard, 1969; Michael S Fanselow, 1984), was scored

during the 20 s CS intervals by an experimenter blind to the conditions. These

intervals were indicated in the recordings by an infrared light. Behaviour of the box

control (BC) and no extinction (7dR) groups were scored for the same 20 s, at the

same time points as each CS interval. Behavioral results were expressed as

percentage time freezing (dependent variable). Due to the low level of foot shock

given, not all animals formed an associative fear memory, therefore only data from

naïve animals and animals shown to have undergone successful fear conditioning

through the display of freezing behaviour when tested for the recall of an associative

fear memory (N = 40) were included in the behavioural analyses (n = 10 in each

group). These animals were identified using the ROUT method of statistical outlier

identification in GraphPad Prism 7 (GraphPad Software Co., CA, USA).

4.3.4 Tissue preparation

Rats were anesthetized with an injection of ketamine-xylazine (200 µl/100 g,

i.p.) and transcardially perfused through the ascending aorta with ice-cold saline (0.9

138

% NaCl) followed by 4 % paraformaldehyde in 0.1 M phosphate buffer 60 minutes

post memory recall. Brains were extracted and post-fixed in 4 % PFA overnight then

stored in 0.1 M phosphate buffered saline (PBS). Free-floating serial coronal sections

(40 μm) of the medial prefrontal cortex and amygdala were prepared using a

vibratome (M11000; Pelco easiSlicer, Ted Pella Inc, Redding, CA, USA).

4.3.5 Immunohistochemistry

To visualize pMAPK activation in neurons as a neuroplastic marker of fear

memory consolidation (Hadley C Bergstrom, Craig G McDonald, Smita Dey, Gina

M Fernandez, et al., 2013; Brambilla et al., 1997; Schafe et al., 2000) the sections

were immunolabelled with an antibody against pMAPK. Briefly, the sections were

washed three times in PBS (pH 7.4) and were blocked in PBS containing 1 % bovine

serum albumin (BSA, Sigma – Aldrich Pty. Ltd, NSW, Aus), 0.02 % Triton X-100

and 3 % Normal Goat Serum (NGS, Abcam, Vic, Aus) for 1 h. After blocking

sections were incubated in a rabbit monoclonal antibody to phospo-p44/42 MAPK

(Erk 1/2) (Thr 202/Tyr 204) (1:250; #4370, Cell Signaling Technology, MA, USA)

for 24 h at room temperature. After wash, the sections were incubated with

biotinylated goat anti-rabbit IgG (1:100 dilution, Vector Laboratories, Burlingame,

CA, USA) in PBS, 0.02 % Triton X-100 for 3 h. Avidin–biotin HRP complex (ABC

Elite, Vector Laboratories, Burlingame, CA, USA) was applied for 1.5 h. Activated

neurons were developed in SG chromagen (Vector Laboratories, Burlingame, CA,

USA) for 10 minutes. Serial sections were mounted on silane-coated slides, air-dried

then dehydrated in a graded series of alcohol, xylene and cover slipped using DPX

mountant (Sigma-Aldrich). Slides were scanned at 20 x magnification with an

Olympus VS120 bright field slide scanner.

139

4.3.6 Section alignment

Topographic spatial analysis relies on precision alignment of brain sections.

This ensures comparison of equivalent neuronal populations across subjects (Hadley

C Bergstrom & Luke R Johnson, 2014; Hadley C Bergstrom, Craig G McDonald,

Smita Dey, Gina M Fernandez, et al., 2013; Hadley C Bergstrom et al., 2011).

Precise alignment was achieved by reference to Paxinos and Watson’s stereotaxic

coordinates (Paxinos & Watson, 2007), and anatomical features easily visualized in

coronal sections. Methods for this process were described in detail for amygdala

studies using the entrance of the lateral ventricle as a land mark (Hadley C

Bergstrom, Craig G McDonald, Smita Dey, Gina M Fernandez, et al., 2013).

Adopting the same principles, the caudate putamen was used as a landmark for the

mPFC sections. In each animal, three locations, bregma +3.24 mm, +3.00 mm and

+2.76 mm were aligned and verified across subjects and conditions by statistical

comparison of the feret length (distance between two perpendicular tangents).

Sections were identically matched across each animal at specific Bregma locations.

Animals were excluded from a specific location if the section required was missing,

visibly torn, damaged or misshapen. This same animal however, may have supplied a

section for a map generated at a more caudal Bregma location.

4.3.7 Neuron density quantifications

Neuron cell quantifications were performed by an experimenter blind to the

experimental conditions. Sections from the left brain hemisphere were counted at

Bregma -3.36 mm and +3.00 mm, providing suitable representations of the LA and

mPFC (Paxinos & Watson, 2007). Subdivisions selected for quantitative analysis

140

included the lateral amygdaloid nucleus dorsolateral (LAd), ventromedial (LAvm)

and ventrolateral (LAvl) parts and the prelimbic and infralimbic regions of the

mPFC. Regional subdivisions were assigned by tracing contours from digital

depictions of the rat brain atlas (Paxinos & Watson, 2007) calibrated, scaled, aligned

and superimposed on the immunolabeled brain sections (Neurolucida 360, MBF

Bioscience, VT). To ensure position of the contour remained stable across

experimental groups, anatomical features used for alignment included the lateral

ventricle, central amygdala, external capsule and rhinal fissure for the LA. Forceps

minor of the corpus callosum, claustrum, olfactory part of the lateral ventricle, the

nucleus accumbens (shell) and the caudate putamen were employed to align the

mPFC. The XY coordinates of pMAPK+ neurons were marked and exported as

ASCII files (Neurolucida 360, MBF Bioscience, VT).

4.3.8 Topographic density maps

An important graphing technique for the visualization of neuronal topography

is the generation of neuron density maps (Hadley C Bergstrom & Luke R Johnson,

2014; Hadley C Bergstrom, Craig G McDonald, Smita Dey, Gina M Fernandez, et

al., 2013; Hadley C Bergstrom, Craig G McDonald, Smita Dey, Haying Tang, et al.,

2013; Hadley C Bergstrom et al., 2011; Luke R Johnson et al., 2012). A series of

plotting and analytic measures were employed to generate mean density maps. The

ASCII files containing the XY coordinates were imported into graphing software

where the data was binned (OriginPro v 9, Origin Lab, Northampton, MA). Bin

spatial dimensions (100 µm2) were determined by averaging twice the expected

frequency of points in a random distribution (D) across all subnuclei, calculated by D

= 2 (sampling area/n), where (n) is the mean number of pMAPK+ neurons for all

141

subjects (Michael John De Smith, Goodchild, & Longley, 2007). Therefore, each bin

portrayed the total number of pMAPK expressing neurons within an equivalent sized

portion of each sub region. To produce a contoured density map, the value within

each bin was assigned a colour (SigmaPlot v 12, Systat Software, San Jose CA).

Difference maps were generated to localize the neurons specific to fear or

extinction memory formation. Subtraction of the mean neuron density of the control

condition from the experimental condition was used to exclude non-specific neuronal

activation from the associative fear or extinction learning. Coefficient of variance

maps were produced to profile the variation of activated neurons within one

experimental condition, both within groups and across the sampling area. The

coefficient of variance (CV) for each bin was calculated by dividing the standard

deviation by the mean (CV = SD/mean). As the CV is normalized to the mean of the

neurons located within each bin, there is equal comparison of the relative variability

across the bin matrix. Low variation (CV < 1.0) was considered as representative of a

stable neuronal population.

4.3.9 Statistical analysis

Analysis of behavioural conditions and neuron counts were conducted using

one-way and two-way ANOVAs. Post hoc Bonferroni correction was used to reduce

type 1 errors synonymous with multiple comparisons. Outliers were removed from

neuron counts and binned data analysis using the ROUT method with the maximum

false discovery rate (q) set at 1 %. To evaluate all bins in each data matrix, two-way

ANOVA with a false discovery rate (FDR) correction for multiple comparisons was

conducted with the total number of activated neurons in each bin considered the

dependent variable. FDR cut-off was set to q ≤ 0.10. FDR correction has been

142

applied to similar data sets previously (Benjamini & Hochberg, 1995; Hadley C

Bergstrom & Luke R Johnson, 2014; Hadley C Bergstrom, Craig G McDonald,

Smita Dey, Gina M Fernandez, et al., 2013). The discovered bins represent the

region of specific interest when defining a population of functional neurons and as

such were termed micro regions of interest (MROIs) and subsequent analysis of

MROIs was conducted using t-tests. A p value ≤ 0.05 was stated as significant, *: p ≤

0.05, **: p ≤ 0.01, ***: p ≤ 0.0001, ****: p ≤ 0.0001. All statistical analysis was

generated with GraphPad Prism 7 and values are expressed as the mean ± standard

error of the mean (SEM).

4.4 Results

4.4.1 Extended extinction training results in decreased freezing in comparison to recent and remote auditory fear memories.

We assessed whether auditory fear conditioning conducted seven days apart

(ER, 7dR and 1dR) would result in comparative levels of freezing to CS/US pairings.

All animals presented similar freezing levels to each pairing with a two-way

ANOVA showing no significant interaction (F4,81 = 0.41,ns, p = 0.8018). Percentage

freezing increased significantly (F2,81 = 42.36, ****, p < 0.0001 ) after each

additional pairing (Figure 2a). This data indicate freezing levels to auditory fear

conditioning were consistent across all groups prior to extinction training. The

association formed between the CS and US was of equivalent strength across groups,

as revealed by the small difference in magnitude of freezing response at each pairing

(F2,81 = 0.55, ns, p = 0.5817). Twenty-four hours following fear conditioning, we

tested the effect of extinction training on freezing levels. The training consisted of 20

presentations of the CS alone. For within-session analyses, five presentations were

143

averaged to represent one trial or block of extinction (Figure 2b-d). Subjects showed

higher % freezing during the first block (CS x 5) of extinction trials compared to the

last block (ER: 1st block 73.7 ± 4.8%, 4th block 16.2 ± 4.0%) on the first day of

training, indicating successful recall of auditory conditioned fear at the beginning of

the extinction session. A significant interaction of time was seen between the first

and last block of presentations (beginning vs end of training, first day: t 588 = 11.58,

****, p < 0.0001; second day: t 588 = 8.92, ****, p < 0.0001; third day: t 588 = 1.61,

***, p = 0.0006 ) indicating the effectiveness of extinction training. Together this

data suggests successful extinction learning occurred, but highlights the spontaneous

renewal that can occur after the passage of time (24 h) (Bouton & King, 1983;

Gregory J Quirk, 2002; Rescorla & Heth, 1975) by showing the increase in freezing

at the beginning of each days extinction training.

144

Figure 4-2: Recall of auditory fear consolidation and extended extinction training result in differing levels of freezing.

(a) Freezing due to auditory fear conditioning conducted on different days resulted in similar levels of freezing for each pairing of US and CS. F4,81 = 0.41 , ns, p = 0.81, with a significant increase in freezing between the first and last tones: F2,81 = 42.36 ****, p < 0.0001, % freezing on final pairing: ER 54%, 7dR 61%, 1dR 54%. (b-d) Auditory fear conditioning was successfully recalled and extinguished as revealed by blocks of extinction training. A significant difference in freezing levels was seen in the beginning versus end of training each day; Day one: block one = 73.7 ± 4.8 % in comparison to the last block of training, block four = 16.2 ± 4.0 %; t 588 = 11.58, ****, p < 0.0001. Day two: t 588 = 8.92, ****, p < 0.0001 and day three: t 588 = 1.61, ***, p = 0.0006. Extinction training resulted in lower freezing levels at the beginning of each session due to repeated exposure to tone on the previous day: day one = 57.5 ± 7.3 %; day 2 = 35.8 ± 5.9 %; day three = 14.3 ± 4.7 %) and negligible levels of freezing by the third day: ER 2.2 %, 7dR 2.7 %, BC 0.6 %. ER subjects exhibited significantly higher levels of freezing than both the 7dR and BC groups in all four blocks on the first day of training (block 1: ER vs 7dR, ****, p < 0.0001; ER vs BC, ****, p < 0.0001; block 2: ER vs 7dR, ****, p < 0.0001; ER vs BC, ****, p < 0.0001; block 3: ER vs 7dR, ****, p < 0.0001; ER vs BC, ****, p < 0.0001; block 4: ER vs 7dR, **, p = 0.0040; ER vs BC, *, p = 0.0243. In the first two blocks on the second day of training: block 1: ER vs 7dR, ****, p < 0.0001; ER vs BC, ****, p < 0.0001; 7dR vs BC, **, p = 0.0038; block 2: ER vs 7dR, ***, p = 0.0002; ER vs BC, **, p = 0.0030.

145

On the third day of extinction training: block 1: ER vs 7dR, ***, p = 0.0008; ER vs BC, ***, p < 0.0005. On each day of training the interaction (indicated by bar) remained significant but decreased on the final day of training (first day: ****, p < 0.0001; second day: ****, p < 0.0001; third day: *, p = 0.0431). (e) Lower freezing levels were observed in the extinction group (ER) compared to animals that did not undergo extinction training (7dR, 1dR); ER = 8.0 ± 2.9 %, 7dR = 53.6 ± 3.1 %, BC = 4.8 ± 4.1 %, 1dR = 38.2 ± 4.6 %. Significant differences were noted between the extinction and fear conditioned groups at the second tone and third tones respectively, (second tone: ER vs7dR, **, p = 0.0012, ER vs 1dR, *, p = 0.0265; third tone: ER vs7dR, **, p = 0.0015). Recent and remote retrieval of a fear memory did not reveal significantly different rates of freezing when compared to each other. Variability among groups as shown by significance bar: F (3,108) = 18.68, ****, p < 0.0001. Data are presented as mean % freezing ± S.E.M; n=10 rats/group, **: p ≤ 0.01, ****: p ≤ 0.0001, two-way ANOVA followed by Bonferroni post hoc analysis.

To evaluate the recall of a consolidated fear or extinction memory, the animals

underwent a 10 minute memory test consisting of 3 x CS only presentations or 10

minutes in context for the box control group (Figure 2e). Freezing levels were

significantly lower in extinction animals compared with those that did not undergo

extinction training (ER vs 7dR, F3,108 = 18.68, ****, p < 0.0001). Bonferroni’s post

hoc comparisons revealed % freezing to be relatively equivalent between the

extinction and box control groups (ER vs BC recall 1: t108 = 1.01, ns, p > 0.9999;

recall 2: t108 = 0.41, ns, p > 0.9999; recall 3: t108 = 0.71, ns, p > 0.9999 ), but higher

in the recent and remote retrieval animals compared to box control (BC vs 7dR recall

1: t108 = 3.57, **, p = 0.0032; recall 2: t108 = 4.26, ***, p = 0.0003; recall 3: t108 =

3.08, ***, p = 0.0158; BC vs 1dR recall 1: t108 = 2.87, *, p = 0.0296; recall 2: t108 =

3.32, **, p = 0.0074; recall 3: t108 = 1.27, ns, p > 0.9999). This suggests the extended

extinction protocol to be highly effective in consolidating an extinction memory.

4.4.2 Spatial patterns of pMAPK expression differ in subregions of the amygdala due to the recall of consolidated or extinguished fear memories.

As pMAPK is implicated in the neuroplasticity required for fear memory

consolidation and extinction learning, we evaluated expression levels within sub

regions of the amygdala and prefrontal cortex. At Bregma -3.36 mm (Figure 3a) the

lateral amygdala is well represented and easily identified by the opening of the lateral

146

ventricle. Across 24 subjects (ER n = 7, 7dR n = 5, BC n = 7, 1dR n = 5) we

observed a significant difference in neuron counts between the box control and all

other experimental conditions, confirming the activation of pMAPK during fear and

extinction memory recall (F3,68 = 4.76, **, p = 0.0045). This was further established

with post hoc analyses ER: t68 = 3.07, *, p = 0.0187; 7dR: t68 = 2.77, *, p = 0.0431

and 1dR: t68 = 3.12, *, p = 0.0161, suggesting pMAPK activation increased in the LA

due to the recall of both fear conditioned and extinction memories.

Figure 4-3: Recall of a conditioned fear memory and extinction memory result in spatially different patterns of pMAPK expression in subregions of the amygdala.

(a) Depiction of the lateral amygdala at Bregma -3.36 mm (Paxinos & Watson, 2007). (b) The LAd was revealed to have a significantly higher number of activated neurons (F3,20 = 14.46, ****, p < 0.0001) due to the recall of both recent and remote auditory fear conditioned memories compared to

147

box control, 7dR: t20 = 4.37, **, p = 0.0018; 1dR: t20 = 6.13, ****, p < 0.0001. The extinction group was also observed to have significantly less activation than the recent auditory fear conditioned group, ER vs 1dR: t20 = 4.08, **, p = 0.0035. Neuron counts in the LAvm revealed a significantly greater number of pMAPK+ neurons in the remote fear memory group (7dR) than the box control, F3,20 = 4.30, *, p = 0.0170; 7dR vs. BC: t20 = 3.37, *, p = 0.0184. Investigation of the LAvl revealed a significantly higher activation after the recall of an extinction memory in comparison to all other groups, F3,20 = 10.99, ***, p = 0.0002; ER vs 7dR: t20 = 4.15, **, p = 0.0030; ER vs BC: t20 = 4.65, ***, p = 0.0009; ER vs 1dR: t20 = 4.78, ***, p = 0.0007, indicating a pattern of activation specific to extinction learning occurred in the LAvl. (c) Topographic density maps were generated using a matrix created from the XY coordinates of pMAPK+ neurons within 100 µm2 bins, to visualize the spatial distribution of the activation. Below each map is its coefficient of variance (CV) map, generated by dividing the standard deviation by the mean. Regions showing < 1 depict stable neuronal populations. Data are presented as mean ± S.E.M; ER n = 7, 7dR n = 5, BC n = 7, 1dR n = 5, *: p ≤ 0.05, **: p ≤ 0.01, ***: p ≤ 0.001, ****: p ≤ 0.0001, one-way ANOVA followed by Bonferroni post hoc analysis.

The LA was divided into three sub regions for further analysis (Figure 3b). The

LAd showed a significantly higher number of activated neurons (F3,20 = 14.46, ****,

p < 0.0001) due to the recall of both recent and remote auditory fear conditioned

memories compared to box control (7dR vs BC **, p = 0.0018, 1dR vs BC ****, p <

0.0001). Significantly less pMAPK activation was observed in the LAd of the

extinction memory recall group compared to the recent fear memory recall group

(ER vs 1dR **, p = 0.0035). In 2011, Bergstrom et al. showed similar findings

regarding the localization of auditory conditioned fear memories within the LAd

(Hadley C Bergstrom et al., 2011). A difference was noted in the LAvm between the

7dR and box control groups (7dR vs. BC *, p = 0.0184), with no differences seen in

the distribution of pMAPK activated neurons in either the 1dR or ER groups when

compared with the naive animals. This may suggest the recruitment of a second sub

region for the remote recall of an auditory fear memory. Analysis of the LAvl

revealed a significantly higher number of pMAPK+ neurons after the recall of an

extinction memory in comparison to all other groups (ER vs 7dR: **, p = 0.0030; ER

vs BC: ***, p = 0.0009; ER vs 1dR: ***, p = 0.0007), indicating a pattern of

activation specific to extinction learning. This can be visualized in the topographic

density maps and CV maps for stability depicted in Figure 3c.

148

4.4.3 Spatial analysis of pMAPK expression in the LAvl reveals a stable population of neurons specific to the recall of an extinction memory.

Difference and CV maps were generated to visualize any patterns of pMAPK

expression unique to the extinction condition (Figure 4a). To do this, binned data

from the non-extinguished control animals (7dR) was subtracted from the extinction

matrix (ER). The CV map was generated to show the stability of any neuronal

population identified. CV values < 1 that corresponded to the regions of greatest

difference (ER – 7dR) indicated stable populations. As the LAvl appeared to be the

focal point for pMAPK expression after the recall of an extinction memory, spatial

analyses were conducted on the LAvl, across all subjects, on 263 neurons in 25 bins

at Bregma -3.36 mm. The LAvl data matrix was compared to reveal a significant

difference between conditions (ER vs 7dR, ***, p = 0.0006) (Figure 4c). The

topography of pMAPK+ neurons was further analysed with FDR corrected mass

multiple comparisons, which resulted in a discovery of 2/25 (8 %) significantly

different bins between the ER and 7dR groups (Figure 4b). Post hoc analysis

revealed significantly more activation in the extinguished animals relative to controls

(mean of two sig. bins ***, p = 0.0004, 4:1 ratio) (Figure 4d). Reported CV values of

the MROIs where < 1 suggesting a statistically stable pattern of neurons encoding

extinction memory recall within the LAvl, across individual brains at Bregma -3.36

mm. Together these results indicate a stable neuronal population within the LAvl,

undergo neuroplastic changes to recall extinction memories.

149

Figure 4-4: Spatial analysis of pMAPK expression in the LA reveals a stable population of neurons specific to the recall of an extinction memory.

(a) A subpopulation of neurons specific to the recall of an extinction memory was identified within the LAvl. A difference map was generated by subtracting non-extinguished group (7dR) binned data from the extinguished group (ER). The CV map allowed visualization of the stability of the neuronal population within the LAvl. Regions showing < 1 depict stable neuronal populations. (b) A q value matrix revealed a discovery of 2/25 (8 %) bins between the ER and 7dR groups, q < 0.10. (c) The LAvl data matrix was compared to reveal a significant difference between conditions, ER 0.79 ± 0.1 vs 7dR 0.32 ± 0.1, n = 25, R2 = 0.22, t = 3.66, ***, p = 0.0006. (d) Post hoc analysis of discovered bins showed the mean of the two significant bins was 0.4 ± 0.2, ***, p = 0.0004, t10 = 5.18, reporting a 4:1 ratio of neurons between conditions, with a difference between means of 1.29 ± 0.7, R2 = 0.73. CV values of the MROIs where < 0 suggesting a statistically stable neuronal population. Data are presented as mean ± S.E.M; ER n = 7, 7dR n = 5, BC n = 7, 1dR n = 5, ***: p ≤ 0.001, t-tests and two-way ANOVA with FDR correction for multiple comparisons.

150

4.4.4 pMAPK expression was observed in PL cortex following recent and remote recall of an auditory fear memory.

Analysis of pMAPK+ neurons at Bregma +3.00 mm was conducted on the

medial prefrontal cortex of 31 rats: ER n = 10, 7dR n = 6, BC n = 6, 1dR n = 9

(Figure 5a). A between groups one way ANOVA of the mPFC (PL + IL) showed an

effect of memory recall on the number of pMAPK neurons activated (F3,58 = 3.91, R2

= 0.17, *, p = 0.0130) revealing increased densities of pMAPK+ neurons in the

recent recall of an auditory fear memory (BC vs 1dR: t58 = 3.37, **, p = 0.0080).

When analysed across 3 Bregmas of the PFC (Bregma +3.24 mm, +3.00 mm and

+2.76 mm) the difference increased, (F3,164 = 8.36, R2 = 0.13, ****, p < 0.0001)

(Figure 5b). Post hoc tests revealed significantly higher pMAPK+ neuron densities in

two groups (ER and 1dR) suggesting the activation of the ERK/MAP kinase pathway

is necessary for recall of both memory types. This data supports the importance of

mPFC modulation in recent fear memory recall.

Figure 4-5: Recall of auditory fear and extinction memory both activate pMAPK expression in the medial prefrontal cortex.

(a) Quantification of activated pMAPK neurons in the medial prefrontal cortex at Bregma +3.00mm showed significant difference between groups t58 = 3.91, *, p = 0.0130 and increased numbers of pMAPK + neurons were seen due to the recall of recent auditory fear conditioned memories in comparison to box control, 1dR vs. BC: t58 = 3.37, **, p = 0.0080. (b) Across 3 Bregma locations in the mPFC ( +3.24mm, +3.00mm, +2.76mm), the difference between groups increased t164 = 8.36, ****, p < 0.0001. Post hoc tests revealed higher pMAPK+ neuron densities in two groups (ER vs BC: t164 = 3.16, *, p = 0.0113 and 1dR vs BC: t164 = 4.92, ****, p < 0.0001 ) as compared to box control

151

suggesting the activation of the ERK/MAP kinase pathway is activated for recall of both memory types. Data are presented as mean ± S.E.M; ER n = 10, 7dR n = 6, BC n = 6, 1dR n = 9, *: p ≤ 0.05, ***: p ≤ 0.001, one-way ANOVA followed by Bonferroni post hoc analysis.

The mPFC subdivisions used for analysis consisted of the PL cortex and IL

cortex (Figure 6a). The PL is well known for its role in fear memory recall (Corcoran

& Quirk, 2007). One-way ANOVA and post hoc analysis with Bonferroni correction

was conducted on neuron counts within the PL cortex. Our findings supported the

previously noted discovery with greater pMAPK expression in the recent recall

group (1dR vs. ER: *, p = 0.0276; 1dR vs. BC: ***, 0.0008), with no significant

difference between the number of pMAPK+ neurons activated in the PL due to the

recall of a recent or remote fear memory.

2

0

PL

IL

0

50

100

150

200

250

PL + 3.00

No.

of p

MA

PK

+ N

euro

ns

****

IL + 3.00

**

ER

7dR

BC

1dR

Condition

152

Figure 4-6: Remote recall of an extinction memory and recent recall of an auditory fear memory both activate pMAPK expression in the infralimbic cortex.

(a) Depiction of the medial prefrontal cortex at Bregma +3.00mm (Paxinos & Watson, 2007). (b) The PL cortex showed increased numbers of pMAPK + neurons due to the recall of recent auditory fear conditioned memories and extinction memories compared to box control, 1dR vs. ER: t27 = 3.09, *, p = 0.0276; 1dR vs. BC: t27 = 4.44, ***, p = 0.0008. There was a nonsignificant trend level difference between remote recall and box control (ns, p = 0.0684). Significant changes in neuronal numbers were seen in the IL cortex between the extinguished (ER) and recent recall (1dR) groups when compared to the naive animals (BC): ER vs BC: t27 = 3.17, *, p = 0.0224; 1dR vs BC: t27 = 3.33, *, p = 0.0152. (c) Topographic density maps were generated to visualize the spatial distribution of the activation. Below each map is its coefficient of variance (CV) map, regions showing < 1 depict stable neuronal populations. Data are presented as mean ± S.E.M; ER n = 10, 7dR n = 6, BC n = 6, 1dR n = 9, *: p ≤ 0.05, ***: p ≤ 0.001, one-way ANOVA followed by Bonferroni post hoc analysis.

There is an established view that the IL is involved in the recall of extinction

memories (Milad & Quirk, 2002) and more recently, argument for involvement only

in extinction learning and not recall (Do-Monte, Manzano-Nieves, Quiñones-

Laracuente, Ramos-Medina, & Quirk, 2015). Results from our statistical evaluation

of neuron counts in the IL, after the remote retrieval of an extinction memory

indicated no significant difference between the memory recall of extinguished and

non-extinguished animals (F2,22 = 1.71, ns, p = 0.2046). Post hoc tests comparing all

conditions revealed differences between the box control animals and some conditions

(ER vs BC *, p = 0.0276; 1dR vs BC ***, p = 0.0008, see Figure 6b). To test the

veracity of this result, analysis was performed on the number of activated pMAPK

neurons within the IL across 3 Bregmas (+3.24 mm, +3.00 mm and +2.76 mm). The

outcome confirmed the original finding with significant difference only to the box

control animals (ER vs BC: ***, p = 0.0002; 1dR vs BC: ***, p = 0.0004). These

results are in accordance with the later findings on the suspected role the IL cortex

153

plays in fear memory extinction, whereby it only exhorts influence during learning.

However, visualization of the spatial distribution of pMAPK+ neurons in the IL

cortex shows a slightly greater density of expression during extinction recall, in

support of the original findings on IL activation for recall. Figure 6c depicts density

and CV maps for both the PL and IL cortex, where the neuron distribution can be

visualized within specific cortical layers.

4.4.5 Patterns of pMAPK expression in the PL cortex reveal activation of cortical layers differ between the recall of recent and remote auditory fear memories.

A visual difference was noted in the density of neurons in the superficial and

deep layers of the recent (1dR) and remote (7dR) auditory fear memory groups (see

Figure 6c). This prompted further investigation into pMAPK activation in the PL

cortex. Difference and CV maps for recent recall less remote recall are depicted in

Figure 7a. The topographic density matrix of pMAPK+ neurons in the PL was greater

due to recent recall of an auditory fear memory when compared to the remote recall

of a fear memory (1dR vs 7dR: ****, p < 0.0001) (Figure 7c). FDR corrected mass

multiple comparison revealed a discovery of 33/261 (~ 8 %) significantly different

MROIs (q < 0.10) between the 1dR and 7dR groups (Figure 7b). Post hoc analysis of

the 33 MROIs showed significant difference between conditions (1dR vs 7dR: ****,

p < 0.0001) (Figure 7d). A CV was calculated for each MROI and 26 bins reported

densities < 1, defining a stable distribution of neurons encoding recent fear memory

recall within the PL across individual brains at Bregma +3.00 mm.

154

Figure 4-7: Spatial analysis of pMAPK expression in the PL reveals a different neuronal distribution between recent and remote recall of an auditory fear memory.

(a) A greater density of pMAPK+ neurons was observed in the deep layers of the PL cortex when remote recall (7dR) was subtracted from recent (1dR) of auditory fear memories (difference map). The CV map showed this region to be highly stable. (b) A q value matrix was developed from the PL cortex binned data. FDR corrected mass multiple comparison revealed a discovery of 33/261 significantly different MROIs (q < 0.10) between the 1dR and 7dR groups. (c) Post hoc analysis of the binned matrix revealed a significant difference between groups, t520 = 5.37, ****, p < 0.0001, difference between means 0.28 ± 0.05 and variance F260 = 1.43. (d) Post hoc analysis of the 33 MROIs showed significant difference, t64 = 15.47, ****, p < 0.0001, with a difference between means of 1.39 ± 0.09, R2 = 0.79 and variance of F32 = 2.37. A CV was calculated for each 26 out of 33 MROIs

7dR 1dR0.0

0.5

1.0

1.5

2.0

PL bin matrix

Condition

No

. of

pM

AP

K+

Neu

ron

s

****

7dR 1dR0.0

0.5

1.0

1.5

2.0

Mean of 33 MROIs PL

Condition

No

. of

pM

AP

K+

Neu

ron

s

****

155

showed values < 1, defining a stable distribution of neurons encoding recent fear memory recall within the deep layers of the PL cortex. Data are presented as mean ± S.E.M; ER n = 10, 7dR n = 6, BC n = 6, 1dR n = 9, ****: p ≤ 0.0001, t-tests and two-way ANOVA with FDR correction for multiple comparisons.

4.4.6 Distribution of pMAPK+ neurons in the IL cortex supports findings of IL activation during extinction memory recall.

Spatial analysis was performed on the mPFC (PL + IL) across subjects, on

3954 neurons in 348 bins. A difference was noted in the IL cortex between the

distributions of pMAPK+ neurons after the recall of memories from extinguished

(ER) and non-extinguished animals (7dR), supported by the CV map displaying

regions of stability (Figure 8a). This may suggest a function for the infralimbic

cortex in the recall of a remote extinction memory. Post hoc analysis of the binned

matrix for the IL cortex resulted in significant difference between conditions (ER vs

7dR: **, p = 0.003) (Figure 8c) and a q value matrix (q < 0.10) revealed a discovery

of three bins (Figure 8b). Analysis of the three MROIs resulted in a significant

difference (ER vs 7dR: **, p = 0.0030), with two of the three bins considered

representative of stable neuronal populations, with CV values < 1 (Figure 8d).

156

Figure 4-8: pMAPK expression in the IL cortex following recall of an extinction memory.

(a) A subpopulation of neurons specific to the recall of an extinction memory was identified within the IL cortex. A difference map was generated by subtracting the non-extinguished group (7dR) binned data from the extinguished group (ER). The CV map allowed visualization of the stability of the neuronal population within the IL cortex. Regions showing < 1 depict stable neuronal populations. (b) A q value matrix revealed a discovery of 3/87 (3 %) bins between the ER and 7dR groups, q < 0.10. (c) Post hoc analysis of the binned matrix for the IL cortex resulted in a significant difference ER vs 7dR: t 172 = 3.00, **, p = 0.003, 0.14 ± 0.05, R2 = 0.05. (d) Post hoc analysis of discovered bins showed significant difference between conditions t4 = 6.45, **, p = 0.0030 with the difference between means of three significant bins 1.056 ± 0.16, R2 = 0.91. CV values of 2 of the MROIs where < 1 suggesting a stable neuronal population. Data are presented as mean ± S.E.M; ER n = 10, 7dR n = 6, BC n = 6, 1dR n = 9, **: p ≤ 0.01, t-tests and two-way ANOVA with FDR correction for multiple comparisons, q = 0.10.

ER 7dR0.0

0.5

1.0

1.5

2.0

IL bin matrix

Condition

No

. o

f p

MA

PK

+ N

eu

ron

s

**

ER 7dR0.0

0.5

1.0

1.5

2.0

Mean of 3 MROIs in IL

Condition

No

. of

pM

AP

K+

Neu

ron

s

**

157

4.5 Discussion

This study sought to identify the micro topography of neurons associated with

extinction memory encoding with prefrontal cortex and amygdala. We aimed to

identify the stable topography of neurons undergoing plasticity following extinction

memory recall in amygdala and prefrontal cortex. We identified a population of

pMAPK+ neurons consistently activated in the LAvl and the IL cortex following

extinction memory recall, elucidating the contributions of these sub regions in

extinction memory encoding.

Our results also show fear memory encoding alters between recent and remote

memory recall. We identified a specific role for the dorsolateral portion of the lateral

amygdala in the recent recall of a fear memory, consistent with our previous findings

(Hadley C Bergstrom, Craig G McDonald, Smita Dey, Gina M Fernandez, et al.,

2013; Hadley C Bergstrom, Craig G McDonald, Smita Dey, Haying Tang, et al.,

2013; Hadley C Bergstrom et al., 2011; Luke R Johnson et al., 2012). In contrast,

reduced activation in the LAd was observed following seven-day remote recall of

extinction memory, which coincided with an increased pattern of activation within

the ventromedial portion of the lateral amygdala. Time dependent differential

activation was also identified in the PL involving cell layer diversity as opposed to

sub regional changes. Superficial cortical layers showed consistent activation in both

recent and remote memory recall, whereas deeper layers were activated only during

recent recall of fear memories suggesting changed IL involvement in the remote

recall of fear memories.

Using an innovative and comprehensive methodology we were able to

identify a difference in the spatial allocation of pMAPK+ neurons in the PL cortex

due to the recent and remote recall of fear memories. Our topographic maps revealed

158

a direct quantitative comparison of pMAPK activation within specific cell layers of

the mPFC. We showed the recall of a fear memory 24h after conditioning

predominantly required layers V and VI of the PL cortex in conjunction with layers

II and III. In 2004, Vertes used anterograde anatomical tracers to show labelled

neurons in layers V and VI of the PL cortex projected to many limbic regions

(Vertes, 2004). Layers II and III primarily process information from the midline

thalamus and ventral hippocampus (Little & Carter, 2012). When the recall of the

conditioned memory was remote by seven days we found pMAPK activated neurons

restricted to superficial layers (II and III) only, suggesting the possibility of a greater

role for the broader limbic system in recent recall. Together, our data demonstrate a

consistent requirement for both amygdala and PL cortex in recent and remote

auditory fear memory retrieval. Interestingly, we show an internal re-organisation of

both amygdala (from LAd to LAvl) and within PL when recent auditory fear

memories are made remote. This highlights the complexities of systems

consolidation within these micro anatomical regions.

Numerous animal studies revealed the IL, in particular, plays a role in fear

memory extinction and for many years the inference that the IL was required for all

aspects of extinction memory formation remained uncontested (Delamater &

Westbrook, 2014; Milad & Quirk, 2002). With recent advances in technology, Do-

Monte and associates reported optogenetic inactivation of glutamatergic neurons

within the IL did not affect the recall of an extinction memory (Do-Monte, Manzano-

Nieves, et al., 2015). However, in support of our work, earlier studies by Laurent and

Westbrook inactivated the IL resulting in impaired extinction retrieval, suggesting

other glutamatergic neurons may have been affected (Laurent & Westbrook, 2009).

159

Our study suggests a possible role for the IL in the recall of remote extinction

memories. Our spatial analysis affirmed this trend by displaying a small pattern of

consistently activated neurons within the IL cortex, specific to the extinguished

group. This is in accord with recent chemogenetic studies showing inhibition of the

IL to BLA pathway at time of extinction learning can reduce the recall of an

extinction memory (Bloodgood, Sugam, Holmes, & Kash, 2018). The relatively

small sample size of the non-extinguished group in the present study may account for

the lack of significance. This was evident by the small observed power size (power =

0.367) between the non-extinguished and extinguished groups. Further work is

therefore needed to confirm this result.

Emotional memories form part of the mechanism required to determine the

appropriate behavioural response to a threatening situation (Faliagkas, Rao-Ruiz, &

Kindt, 2018; Van der Kolk, 1994). PTSD sufferers not only have an impairment to

this mechanism but have an extinction retention deficit after training to extinguish

the fearful memory, for review see (Adams & Sweatt, 2002). Our study shows that a

portion of extinction memory recall is encoded in a stable specific subpopulation of

neurons localized to the LAvl, based on the relative distribution of neurons within

this region after the remote recall of extinction memories. Examination of a protein

synthesis marker such as arc (Arg 3.1) (Mamiya et al., 2009a) could also enhance the

knowledge gained from studying pMAPK, a single molecular marker of neuronal

plasticity. Further to this, a larger number of bregma sites and time points could

expand the current recorded data and identify other regions involved in extinction

memory recall. To further investigate this fundamental molecular finding,

DREADDs (K. S. Smith, Bucci, Luikart, & Mahler, 2016), optogenetics (Do-Monte,

Manzano-Nieves, et al., 2015) or local micro-infusions of pMAPK inhibitors (Herry

160

et al., 2006) could further expand our knowledge of the significant role this sub-

region plays in the extinction of fear memories.

Our findings within the mPFC provide new insight into the organization of

laminar activation due to fear conditioning and extinction recall, which enhances our

understanding of the neurobiological substrates of fearful memories. Future studies

involving retrograde tracers may be useful to assist in determining the specific

circuits mediating this behaviour (Bloodgood et al., 2018; Marek et al., 2018b).

Analysis revealed several regions with a four-fold increase in MAPK activated

neurons following extinction memory recall, proposing potential micro-target regions

for future therapeutic interventions. As the neurobiological mechanisms of extinction

are thought to be compromised in PTSD, a thorough understanding of these

mechanisms is necessary for providing potential new therapeutic sites and targets.

161

Microglial phenotype alters with varied fear memory

recall in the prefrontal cortex.

This chapter comprises the following article, currently undergoing revision: Jacques, A., Chaaya, N., Beecher, K., Ali, S. A., Patkar, S. O., Battle, A. R., Johnson, L. R. Belmer, A., Chehrehasa, F., Bartlett, S. E. Microglia phenotype alters in the prefrontal cortex with varied fear memory recall. Brain, Behavior and Immunity. Submitted 27th December, 2018. In revision.

Chapters 5 and 6 investigate the role of microglia in fear memory consolidation

and recall, as outlined in aim 2. Briefly this entails establishing levels of c-Fos, Arc,

pMAPK and BDNF, and defining changes to microglia number and phenotype as a

result of differing fear memory types.

162








In the case of chapter 5

Microglial phenotype alters with varied fear memory recall in the prefrontal cortex.

Publication status: In revision

Contributor Statement of contribution*Angela Jacques Involved in the conception and design of the project, behavioural and

laboratory experiments, imaging, analysed data, created the figures and wrote and edited the manuscript.

Nicholas Chaaya Assisted with editing the manuscript, behavioural and laboratory experiments and imaging.

Kate Beecher Assisted with editing the manuscript.

Syed Aoun Ali Assisted with editing the manuscript.

Omkar Patkar Assisted with laboratory experiments.


Luke Johnson Involved in the conception and design of the project and assisted with editing the manuscript.

Arnauld Belmer Assisted with imaging.

Fatemeh Chehrehasa

Assisted with interpretation of data and editing of manuscript

Selena Bartlett Assisted with reviewing and editing the manuscript.

QUT Verified

Signature

163

5.1 Abstract

In the central nervous system, microglia act as immune cells and respond to all forms

of pathophysiological events from infection to traumatic brain injury. In the healthy

brain, resting (or surveillance) microglia actively survey functional brain tissue to

maintain homeostasis. It is accepted that microglia respond to environmental

challenges such as stress but it is unknown if their morphology is altered as a result

of fear. First to establish levels of neuronal activity we investigated the influence of

fear memory recall on the expression of the immediate early genes c-Fos and Arc.

Following, we note any microglial phenotypic changes as a consequence of three

types of fear memory recall: recent fear, remote fear and extinction memory.

Histological analysis of ionized calcium binding adaptor molecule 1 (Iba1), was

employed to identify microglia while cell tracing was utilised to determine

morphological adaptations. Our results demonstrate that microglia alter their

phenotype after the recall of temporally distinct fearful events.

5.2 Introduction

Microglia facilitate learning-induced plasticity of glutamatergic synapses (Parkhurst

et al., 2013) through the secretion of brain-derived neurotrophic factor (BDNF) and

constantly undergo morphological change whilst monitoring their microenvironment

(Wake et al., 2009). They are myeloid cells which continuously monitor the brain

for invading micro-organisms, damaged neurons and to prune synaptic terminals

(Paolicelli et al., 2011) by extension or retraction of their processes. These non-

neuronal cells may respond to pathological brain changes in a similar manner to

164

macrophages (Hanisch & Kettenmann, 2007) and are capable of evoking an innate

immune response (Kettenmann, Hanisch, Noda, & Verkhratsky, 2011).

Microglia are capable of morphological and functional adaptation from a

heavily branched or ramified shape to an amoeboid shape. To date, reports show

unramified microglia tend to have a large or occasionally elongated cell body with

short, distended extensions, are said to be “phagocytic” and provide a pro-

inflammatory role (Karperien, Ahammer, & Jelinek, 2013; Stence, Waite, & Dailey,

2001). Microglia with a highly ramified phenotype are called surveillance or resting

microglia (Swanton et al., 2018) and provide neurotrophic factors (Nakajima &

Kohsaka, 2002), mediate pain (Watkins, Milligan, & Maier, 2001), monitor and

prune synapses (Pow, Perry, Morris, & Gordon, 1989). Although many studies

involve the microglial response to inflammatory conditions, structural remodelling of

surveillance microglia can occur as a result of natural environmental challenges

without any concomitant inflammation (Kongsui, Beynon, Johnson, & Walker, 2014)

(Tremblay, Lowery, & Majewska, 2010). Essential to neuroplasticity, surveillance

microglia monitor the synaptic connections within their micro environment at a rate

of once per hour (Wake et al., 2009). Reduced neuronal activity results in a reduced

frequency of this monitoring (Wake et al., 2009). The duration of a microglia-

synapse connection may range from 5 minutes to 1 hour, dependant on the functional

status of the synapse and whether synaptic pruning is required (Wake et al., 2009).

Stress has been shown to facilitate increased proliferation and adaptation to

the activated phenotype of microglia (Tynan et al., 2010; Varvel et al., 2012; Wohleb

et al., 2012). Therefore, we hypothesised that recalling a fear memory would also

alter the phenotype and function of these cells. We examined the prefrontal cortex

(PFC) as studies suggest remote memories are stored here (Knowlton & Fanselow,

165

1998; Squire & Alvarez, 1995) with the medial PFC indicated to integrate emotional

memories such as fear (Paul W Frankland, Bontempi, Talton, Kaczmarek, & Silva,

2004). To investigate the changes in neuronal activity we examined immediate early

genes (IEGs) c-Fos and Arc which can provide a measure of neural activity and are

implicated in memory recall (Minatohara, Akiyoshi, & Okuno, 2016; Plath et al.,

2006). c-Fos expression has been shown to be greater in the prelimbic cortex

following fear conditioning than during extinction learning (Morrow, Elsworth,

Inglis, & Roth, 1999), with Arc being considered a contributing factor in the synaptic

plasticity required for memory storage (Plath et al., 2006). Our findings show recent

fear memory recall results in high neuronal activation, as evidenced by increased

expression of c-Fos and Arc. Further to this, we observed microglial phenotype

adaptations present in each case of fear recall. In this study prelimbic cortex tissue is

labelled with markers of immediate early genes (c-Fos and Arc) and microglia

(Iba1). Quantification and tracing of cells demonstrates a distinct morphological

change of microglia in response to temporally diverse fear memory recall.

5.3 Methods

5.3.1 Subjects

Experimentally naive adult male Sprague-Dawley rats (N = 21) were supplied

by Animal Resource Center, ARC, Western Australia and housed 3 per cage, in

temperature (≈ 24 °C) and humidity (35 %) controlled Plexiglas cages. The cages

were maintained on a 12-hour light/dark cycle with the behavioral procedures

conducted during the light cycle, as fear conditioning is reported to be more

effective during the nocturnal phase for rodents i.e. during the light cycle (Albrecht

& Stork, 2017). Rats were acclimatized to the vivarium for 7 days prior to training,

with food and water provided ad libitum. At time of memory recall testing, rats

166

weighed 326.4 ± 6.7 g. All procedures were conducted in compliance with the

Animal Welfare Unit, The University of Queensland Research and Innovation Ethics

Committee and the Research Ethics Committee of the Queensland University of

Technology, Australia. Procedures complied with policies, regulations and ethical

standards for animal experimentation, in accordance with the Queensland

Government Animal Research Act 2001, associated Animal Care and Protection

Regulations (2002 and 2008), and the Australian Code for the Care and Use of

Animals for Scientific Purposes, 8th Edition (National Health and Medical Research

Council, 2013).

5.3.2 Apparatus

Two conditioning chambers were utilized for behavioral procedures, referred

to as Context A and Context B. The chambers were modified to create unique testing

environments for fear conditioning or extinction and to restrict any reaction to

context. Context A included a stainless-steel grid floor connected to a shock

generator and computer (Freeze frame software, Coulbourn instruments). Ethanol (80

%) was used as the cleaning agent and odorant in Context A and the chambers were

cleaned between each animal. To create contextual difference, Context B contained a

solid plastic floor with fresh bedding and internal colored decoration on the walls and

ceiling. Orange scented antibacterial soap was used as the cleaning agent and to

provide a unique smell to Context B. The chambers were contained in an acoustic

isolation box (Coulbourn Instruments, Allentown, PA, USA) with the background

noise level measured at 55 dB, using a sound level meter (Digitech Professional

Sound Level Meter QM1592). Each chamber contained a speaker, low-level house

light (2-3 lux), infrared light and infrared camera.

167

5.3.3 Behavioral procedures

A total of 21 rats were randomized into three groups. Recent and remote

auditory fear conditioning was run using an extended extinction group as a control.

Both fear to context and extinction of fear are types of learning, therefore any

differences in IEG expression and changes in microglia morphology can be assumed

to be specific to the type of learning undertaken. Changes in microglia morphology

specific to exposure to the context alone in fear conditioning was recently published

by our lab (Nicholas Chaaya et al., 2019), therefore the unique control group of

extinction was chosen to ensure each group of animals had undergone the same fear

conditioning protocol and to highlight the changes in microglia through the recall of

different types of fear memories specifically. For two days prior to training, rats were

habituated to each context for 30 minutes. On the first day of training the RE and 7dF

groups underwent three minutes of acclimation to context A, followed by a 10-

minute conditioning protocol involving three pairings of an auditory conditioned

stimulus (CS, tone - 5kHz, 75 dB, 20 s) that co-terminated with an unconditioned

stimulus (US, foot shock - 0.6 mA, 500 ms). Stimuli were controlled through Freeze

Frame software (Coulbourn Instruments) and training was separated by a mean inter-

trial interval of 180 s. Rats were returned to home cages after final stimulus

presentation. The remote extinction recall (RE, n = 7) group then underwent three

days of extinction training. Each day the animals had 30 minutes of extinction

training (20 x CS alone, 5 kHz, 75 dB, 20 s) in Context B. The average inter-trial

interval for extinction training was 76 s and began after 5 minutes of habituation to

the context. To facilitate memory consolidation they were returned to their home

cage for three days then provided with a fear memory test (FMT) 1 hr prior to

168

sacrifice on day 8. The FMT was a 10-minute memory recall test consisting of three

20 s CS presentations to observe the recall of a consolidated auditory cued fear

memory or extinction memory. The remote fear memory group (7dF, n = 7),

followed the same protocol, without exposure to tone during the extinction training

days. The recent fear memory group (1dF, n = 7), remained in their home cages until

the seventh day when they received auditory fear conditioning and 24 hr later

underwent a FMT (see Figure 1A).

Freezing, a behavioral index to quantitate a CS-US association

(Blanchard & Blanchard, 1969; Michael S Fanselow, 1984), was scored during the

20 s CS intervals by an experimenter blind to the conditions. These intervals were

indicated in the video recordings by an infrared light allowing the behaviour of the

7dF group to be scored for the same length of time and at the same time points as

each CS interval. Behavioral results (see Figure 1B) were expressed as percentage

time freezing (dependent variable).

169

Figure 5-1 Recall of recent and remote fear and extinction memories result in different levels of freezing.

A. Experimental design. Recent and remote auditory fear conditioning was run using an extended extinction group as a control. Remote extinction recall (RE, n = 7) group underwent auditory fear conditioning followed by three days of extinction training and three days in the home cage, with a retrieval test 1 hr prior to sacrifice. A remote fear memory group (7dF, n = 7), underwent the same conditioning and followed the same protocol as the remote extinction group, however, were not exposed to the tone. The recent fear memory group (1dF, n = 7), remained in their home cages until the seventh day of experimentation when they received auditory fear conditioning and underwent a memory recall test 24 hr later. B. Behavioural results. Habituation: Habituation to context provided a base line freezing measure (0.2 ± 0.1%) with no difference between contexts (0.04 ± 0.2 %, t 34 = 0.23, ns, p = 0.8218). Fear conditioning: Auditory fear conditioning was conducted seven days apart with groups RE and 7dF on day 1 and 1dF on day 7. Percentage freezing increased at each pairing of CS and US (F2,48 = 23.22, ****, p < 0.0001 ). 1st, 2nd and 3rd extinction: The effect of extinction or no extinction training (20 presentations of CS alone or context alone, in context B) is shown as four CS presentations averaged to one block. Higher % freezing in the first block of extinction trials compared to the last block (RE: 1st block 83.5 ± 5.5%, 4th block 10.3 ± 3.3%) on the first day of training. There was a significant temporal interaction between the beginning vs end of training, first day: t 54 = 11.46, ****, p < 0.0001; second day: t 54 = 7.08, ****, p < 0.0001; third day: t 54 = 2.28, *, p = 0.0265). FMT: Recall of fear memory or extinction memory was tested with a 10-minute recall test consisting

170

of 3 x CS only presentations. % freezing levels were lower due to remote extinction recall as opposed to fear memory recall (RE vs 7dF and 1dF, **, p = 0.0026). Dataarepresentedasmean%freezing±S.E.M,**:p≤0.01,****:p≤0.0001,two‐wayANOVAfollowedbyBonferroniposthocanalysis.


Rats were anesthetized with an injection of lethabarb (400 mg/kg, i.p.) and

transcardially perfused through the ascending aorta with ice-cold 1% (wt/vol)

paraformaldehyde (PFA) with 0.125% (vol/vol) glutaraldehyde followed by 4 %

(wt/vol) PFA with 0.125% (vol/vol) glutaraldehyde in 0.1 M phosphate buffer, 60

minutes post memory recall. Brains were extracted and post-fixed in 4 % PFA with

0.125% (vol/vol) glutaraldehyde overnight then stored in 0.1 M phosphate buffered

saline (PBS). Free-floating serial coronal sections (40 μm) of the medial prefrontal

cortex were prepared using a vibratome (M11000; Pelco easiSlicer, Ted Pella Inc,

Redding, CA, USA). Analysis was performed solely on the prelimbic section of the

medial prefrontal cortex. Tissue from the animals used for this research was also

used in other studies. These studies required the neuronal cell walls to hold a specific

resistance, thus the perfusion was performed without saline solution and with

glutaraldehyde added to the PFA.


Three prefrontal sections were taken from each subject to visualize Iba1

expression in microglia and cFos / Arc activation in neurons. The sections were

immunolabelled with antibodies against each. It must be noted that Iba-1 staining

may label microglia, infiltrating monocytes, macrophages and blood vessel-

associated macrophages. Briefly, the sections were washed three times in PBS

(containing 0.02 % sodium azide) and blocked in PBS containing, 0.3 % Triton X-

100, 0.05 % Tween 20 and 2 % Normal Horse Serum (NHS, Abcam, Vic, Aus) for 1

171

h. After blocking sections were incubated with a goat monoclonal antibody (mAb) to

Iba1 (1:500; ab5076, Abcam, Vic, Australia), rabbit mAb to cFos (9F6) (1:300;

#2250, Cell Signaling Technology, MA, USA) and mouse mAb to Arc (c-7) (1:300;

sc-17839, Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA), for 24 h at room

temperature. After wash, the sections were incubated with Alexa Fluor donkey anti-

sheep IgG 594, donkey anti-rabbit IgG 488 and donkey anti-mouse IgG 647 (1:500;

Invitrogen, Life Technologies, CA, USA) in blocking solution (see above) for 3 h.

Sections were mounted on silane-coated slides and cover slipped using ProLong

Gold antifade reagent (Invitrogen, DR, USA).

5.3.5 Imaging

3 sections per animal, n = 5 animals per group, totalling 15 sections per

condition, were imaged on an Olympus FV3000 confocal laser scanning microscope

(Olympus Australia Pty Ltd., VIC, Australia) using a 30X oil-immersion objective

(NA 1.05) with a 1.5 x zoom and a Z-axis step of 0.8 μm, using sequential scanning.

The sections were taken from the prelimbic cortex around Bregma +2.76mm, and

mosaics of the regions of interest (as depicted in Figure 2) were acquired in OIR file

format. Each mosaic consisted of a data volume of 1452000 µm3.

172

Figure 5-2 Schematic drawing showing the location of the acquired micrographs.

A. Images were acquired around bregma +2.76mm (depicted by the blue / purple vertical line) in the medial prefrontal cortex. B. Representation of brain slice showing the prelimbic (PL) and infralimbic (IL) cortex. Mosaic images were captured from the PL, layer II & III (blue / purple square). Drawings depicted from (Paxinos & Watson, 2007).

5.3.6 Cell quantifications

For the purpose of cell quantification, volumes equaling x = 212 µm, y = 614

µm and z = 12 µm were scanned from the medial prelimbic cortex of each subject.

Cell quantifications were performed by an experimenter blind to the conditions,

using IMARIS software (IMARIS 9.1.2, Bitplane, ZH, CHE). Using the measuring

feature in IMARIS an average diameter was defined for each cell type. The cells

were reconstructed in 3D using the ‘spot detection’ function (cFos 4.3 µm, Arc 6 µm

and Iba1 2.7 µm) and automatically counted. Filter intensity was set by the

experimenter and images were batch processed using the same thresholding

parameters across experimental groups. The cell numbers including colocalisation

were obtained from the statistics function in Imaris (see Figure 3A-C, 4A). These

steps have been previously described in (Belmer, Klenowski, Patkar, & Bartlett,

2017; Nicholas Chaaya et al., 2019; Tarren, Lester, Belmer, & Bartlett, 2017).

173

Neurolucida 360 software (Neurolucida 360, MBF Bioscience, VT) was used

to trace three microglia from each brain slice (RE n = 42, 7dF = 42, 1dF = 36). These

microglia were chosen by meeting the criteria of being centrally located in both the

middle of the scan and centered within the scan to allow for all branching to be

complete within the scans parameters. The cell body of each microglia was

automatically selected using soma detector sensitivity of 50, an interactive search

region of 20µm and a size constraint of 2 µm. The extensions were traced manually.

The traces were used to analyse the cell body volume and microglial processes,

inclusive of the number of extensions leaving the cell body (trees), the number of

branches off these extensions (branches) and the total length of all extensions (see

Figures 4 and 5). Neurolucida 360 uses the formula (sum of the terminal orders +

number of terminals) ∗ (total process length/number of primary branches) (Pillai et

al., 2012) to determine the ramification of microglia branching.

5.3.7 Statistical analysis

Analysis of behavioral conditions and cell counts were performed using t-

tests, one-way and two-way ANOVAs. Post hoc Bonferroni correction was used in

all cases to reduce type one errors synonymous with multiple comparisons. A p value

≤ 0.05 was stated as significant, *: p ≤ 0.05, **: p ≤ 0.01, ***: p ≤ 0.0001, ****: p ≤

0.0001. All statistical analyses were generated with GraphPad Prism 7 (GraphPad

Software Co., CA, USA) and values are expressed as the mean ± standard error of

the mean (SEM). As foot shock levels for this protocol were low, not all animals

formed an associative fear memory, therefore only data from animals shown to have

undergone successful fear conditioning through the display of freezing behaviour

when tested for the recall of an associative fear memory were included in the

behavioral analyses (removed: 7dF n = 1, 1dF n = 1). Four animals were removed

174

due to inadequate perfusion (RE n = 2, 7dF n=1, 1dF n=1) and significant tissue

damage resulting in an inability to analyze the region of interest resulted in the

removal of several brain slices (RE n = 1, 7dF n = 1, 1dF n = 3). Statistical outliers

were identified and excluded using the ROUT method of statistical outlier

identification in GraphPad Prism 7. The ROUT method used combined robust

regression and outlier removal. It is used to fit a curve that is not influenced by

outliers. The residuals are analysed with a test adapted from the False Discovery

Rate approach for multiple comparisons. Once the outliers are removed ordinary

least-squares regression is performed on the remaining data (Graphpad, 2016).

5.4 Results

5.4.1 Remote extinction memory recall results in decreased freezing compared to recent and remote auditory fear memory recall.

For all behavioral results see Figure 1B. Baseline freezing behavior (0.2 ±

0.1%) was obtained to both fear conditioning and extinction contexts during the

habituation phase. Freezing behaviour to context A and context B was averaged

across both days and then compared. There was no significant difference between

contexts (0.04 ± 0.2 %, t 34 = 0.23, ns, p = 0.8218). Auditory fear conditioning was

conducted seven days apart with groups RE (n = 7) and 7dF (n = 6) on day 1 and 1dF

(n=6) on day 7. A two-way ANOVA was conducted to show freezing behaviour to

be equivalent between groups, while progressively increasing as a function of CS/US

pairing. Results from the two-way ANOVA showed no interaction to exist (F4, 48 =

0.22, ns, p = 0.93). Follow-up main effect confirmed that no differences between

behavioural conditions existed (F2, 48 = 0.01, ns, p = 0.99), while statistical

significant difference between CS/US pairing existed (F2, 48 = 23.22, ****, p <

0.0001), with each pairing resulting in higher fear expression. These data indicate the

175

fear conditioning protocol employed here resulted in equivalent freezing to tone

across all groups.

Twenty-four hours following fear conditioning (RE and 7dF), the effect of

extinction training (20 presentations of CS alone in context B) on freezing levels was

tested. For analyses, four CS presentations were averaged to represent one trial or

block of extinction. Higher % freezing was observed during the first block (CS x 4)

of extinction trials compared to the last block (RE: 1st block 83.5 ± 5.5%, 4th block

10.3 ± 3.3%) on the first day of training, an indication of successful recall of auditory

conditioned fear at the beginning of the extinction session, and extinction of fear by

the end of the session. Statistical analyses across all extinction days revealed a

significant temporal interaction between the first and last block of presentations

(beginning vs end of training, first day: t 54 = 11.46, ****, p < 0.0001; second day: t

54 = 7.08, ****, p < 0.0001; third day: t 54 = 2.28, *, p = 0.0265) indicating the

effectiveness of extinction training over a period of three days. Together these data

suggest successful extinction learning occurred and highlights the spontaneous

renewal that occurred after each 24 h period (Bouton & King, 1983; Gregory J

Quirk, 2002; Rescorla & Heth, 1975) as seen by the increase in freezing at the

beginning of each day’s extinction training.

Following extinction training, remote fear conditioning (7dF), or recent fear

conditioning (1dF), recall of fear memory or extinction memory was tested with a

10-minute recall test consisting of 3 x CS only presentations. Two-way ANOVA

revealed no interaction as a function of condition versus CS/US presentation (F4, 48 =

0.80, ns, p = 0.53). Similarly, follow-up main effect of CS/US presentation was not

significant (F2, 48 = 0.97, ns, p = 0.39). Importantly, follow-up main effect of

condition was statistically significant (F2, 48 = 0.29.84, ****, p < 0.0001). Follow-up

176

of simple main effect via Bonferroni corrected post hoc comparisons revealed %

freezing levels were significantly lower in extinction animals compared with those

that did not undergo extinction training (RE vs 7dF and 1dF, **, p = 0.0026),

suggesting the extended extinction protocol to be highly effective in consolidating

extinction memories.

5.4.2 Recall of recent fear memory results in increased c-Fos and Arc activation.

Analysis of c-Fos+ and Arc+ neurons and Iba1+ microglia was conducted on

the medial prelimbic cortex of 15 rats, 5 per condition (Figure 3). Following our labs

previous work mapping the location of neuroplastic change produced by the recall of

recent and remote fear memories and remote extinction memories, we noted

activation of the ERK/MAP kinase pathway was evident in the prelimbic cortex (A.

Jacques et al., 2017). These findings provided purpose to investigate microglial

activity in the prelimbic rather than the infralimbic cortex. 3 sample sections, each

with a volume of 2 x 106 µm3, were taken for analysis, from each animal and

averaged across the group. A between groups one way ANOVA showed an effect of

memory recall on the number of cFos+ neurons activated (F2,37 = 3.41, R2 = 0.16, *, p

= 0.0437). Post hoc tests revealed an increased density of cFos+ neurons in the

recent recall of an auditory fear memory in comparison to a remote extinction

memory and no significant difference between recent and remote memory recall (1dF

vs RE: t37 = 2.60, ns, p = 0.0399; 1dF vs 7dF: t37 = 1.20, *, p = 0.7161; 7dF vs RE: t37

= 1.46, ns, p = 0.4578). Analysis of Arc activation correlated these results with the

between groups one-way ANOVA showing memory recall activated Arc expression

(F2,37 = 16.6, R2 = 0.47, ****, p < 0.0001). Post hoc tests revealed an increased

177

density of Arc+ neurons in the recent recall of a recent auditory fear memory in

comparison to both remote fear and remote extinction memories (1dF vs 7dF: t37 =

5.00, ****, p < 0.0001; 1dF vs RE: t37 = 5.10, ****, p < 0.0001; 7dF vs RE: t37 =

0.10, ns, p > 0.9999). Together these data are in concurrence with studies suggesting

greater activation of the prelimbic cortex occurs during the recall of recent fear

memories (for review see (Giustino & Maren, 2015)). There is an established view

that Arc mRNA activation colocalizes with c-Fos expressing cells (Fanous et al.,

2013). Our data revealed Arc immunoreactivity was present in c-Fos+ neurons and

this colocalisation was not affected by the type of memory recall tested (F2,37 = 0.9,

R2 = 0.04, ns, p = 0.4361; 1dF vs 7dF: t37 = 0.32, ns, p > 0.9999; 1dF vs RE: t37 =

1.24, ns, p = 0.6733; 7dF vs RE: t37 = 0.96, ns, p > 0.9999).

178

Figure 5-3 c-Fos and Arc expression in the PFC are greater in the recall of a recent fear memory.

A. Analysis of c-Fos+ and Arc+ neurons was conducted on 3 sections (volume of 2 x 106 µm3) from each animal and averaged across groups. An increased density of cFos+ neurons was apparent after recent fear memory recall compared to remote extinction memory recall, with no change between recent and remote memory recall (1dF vs RE: t37 = 2.60, ns, p = 0.0399; 1dF vs 7dF: t37 = 1.20, *, p = 0.7161; 7dF vs RE: t37 = 1.46, ns, p = 0.4578). B. Arc expression (F2,37 = 16.6, R2 = 0.47, ****, p < 0.0001) increased due to the recent recall of a recent auditory fear memory in comparison to both remote fear and remote extinction memories (1dF vs 7dF: t37 = 5.00, ****, p < 0.0001; 1dF vs RE: t37

= 5.10, ****, p < 0.0001; 7dF vs RE: t37 = 0.10, ns, p > 0.9999). C. Arc immunoreactivity was present in c-Fos+ neurons however, colocalisation was not affected by memory type (1dF vs 7dF: t37 = 0.32, ns, p > 0.9999; 1dF vs RE: t37 = 1.24, ns, p = 0.6733; 7dF vs RE: t37 = 0.96, ns, p > 0.9999). Dataarepresentedasmean±S.E.M;*:p≤0.05,****:p≤0.0001,one‐wayANOVAfollowedbyBonferroniposthocanalysis.D.Representativeimagesofc‐Fos,ArcandIba1expression.

5.4.3 Microglia alter morphologically in response to recent and remote fear recall.

Social defeat stress is known to increase the number of Iba-1+ microglia in

the hippocampus, prefrontal cortex and amygdala (Wohleb et al., 2012). In line with

this finding fear may also contribute to the number and morphology of microglia.

Quantification of microglia nuclei (Figure 4A) revealed a variation in microglia

density between our memory recall groups (F2,37 = 10.37, R2 = 0.36, ***, p = 0.0003)

with post hoc analysis demonstrating the greatest difference in cell number between

early fear recall and remote extinction recall (1dF vs 7dF: t37 = 2.34, ns, p = 0.0743;

1dF vs RE: t37 = 4.55, ***, p = 0.0002; 7dF vs RE: t37 = 2.30, ns, p = 0.0810). This

data suggests an increased number of microglia are present within the prefrontal

cortex after the recall of recent fear memories. Figure 4 depicts graphically and

pictorially, the number and morphology of microglia found in each condition. The

complexity of the microglia refers to the normalization and comparison of processes

among microglia. It is calculated using the dendritic complexity index, (sum of the

terminal orders + number of terminals) * (total process length / number of primary

branches)(Pillai et al., 2012). Statistical analyses (ANOVA) revealed the degree of

ramification (complexity of the extensions, Figure 4B) was different between groups

179

(F2,112 = 10.97, R2 = 0.16, ****, p < 0.0001), with post hoc analysis showing recent

fear memory recall to be significantly different from remote fear and extinction

memory recall (1dF vs 7dF: t112 = 3.02, **, p = 0.0093; 1dF vs RE: t112 = 4.63, ****,

p < 0.0001; 7dF vs RE: t112 = 1.58, ns, p = 0.3545). These data reveal well ramified

cells only in the remote extinction group, as clearly depicted in Figure 4C.

Alongside proliferation and ramification changes, experience dependent

modifications include morphological adaptations of cell body size and shape, and the

number and length of extensions protruding from the cell body (see Figure 5). A

between groups ANOVA of cell body volume was significant (F2,117 = 17.76, R2 =

0.23, ****, p < 0.0001), with Bonferroni corrected post hoc tests revealing

differences between the remote extinction versus both recent and remote fear

memory recall (RE vs 7dF: t117 = 5.18, ****, p < 0.0001; RE vs 1dF: t17 = 5.07, ****,

p < 0.0001; 7dF vs 1dF: t117 = 0.10, ns, p > 0.9999). This finding was repeated in

measurements of the cell body area (F2,117 = 18.94, R2 = 0.25, ****, p < 0.0001; RE

vs 7dF: t117 = 5.18, ****, p < 0.0001; RE vs 1dF: t117 = 5.41, ****, p < 0.0001; 7dF

vs 1dF: t117 = 0.10, ns, p > 0.9999). One-way ANOVAs revealed significant

differences between groups (number of extensions: F2,117 = 17.27, R2 = 0.23, ****, p

< 0.0001; number of branches: F2,117 = 21.38, R2 = 0.27, ****, p < 0.0001; length of

extensions: F2,580 = 11.44, R2 = 0.04, ****, p < 0.0001). Post hoc evaluation revealed

the number of extensions (7dF vs 1dF: t117 = 4.69, ****, p < 0.0001; 1dF vs RE: t117

= 5.51, ****, p < 0.0001; 7dF vs RE: t117 = 0.86, ns, p < 0.9999), their branches (7dF

vs 1dF: t117 = 5.06, ****, p < 0.0001; 1dF vs RE: t117 = 6.21, ****, p < 0.0001; 7dF

vs RE: t117 = 1.20, ****, p < 0.0001) and their overall length (7dF vs 1dF: t117 = 4.21,

****, p < 0.0001; 1dF vs RE: t117 = 4.43, ****, p < 0.0001; 7dF vs RE: t117 = 0.21,

ns, p > 0.9999) were greater in both the extinction and remote fear memory recall

180

groups versus recent fear recall. Interestingly, together this data suggests the

morphology of the microglia shows increased ramification reminiscent of resting

state microglia in the remote extinction recall group and a trend towards this

ramification in the remote fear recall subjects.

Figure 5-4 Microglia alter morphologically in response to fear memory recall.

A. Quantification of microglia nuclei revealed density variations between memory recall groups with the greatest difference in cell number between early fear recall and remote extinction recall (1dF vs 7dF: t37 = 2.34, ns, p = 0.0743; 1dF vs RE: t37 = 4.55, ***, p = 0.0002; 7dF vs RE: t37 = 2.30, ns, p = 0.0810). B. The measure of complexity (sum of the terminal orders + number of terminals) * (total process length / number of primary branches) varied across groups with recent fear memory recall significantly different from remote fear and extinction memory recall (1dF vs 7dF: t112 = 3.02, **, p = 0.0093; 1dF vs RE: t112 = 4.63, ****, p < 0.0001; 7dF vs RE: t112 = 1.58, ns, p = 0.3545). Dataarepresented asmean ± S.E.M; **: p ≤ 0.01, ***: p ≤ 0.0001, ****: p ≤ 0.0001, one‐way ANOVAfollowedbyBonferroniposthocanalysis.C. Top panel shows Iba1+ microglia with tracing overlaid and bottom panel shows the tracing alone. Scale bar 50µm.

Alongside proliferation and ramification changes, experience dependent

modifications include morphological adaptations of cell body size and shape, and the

181

number and length of extensions protruding from the cell body (see figure 4). A

between groups ANOVA of cell body volume was significant (F2,117 = 17.76, R2 =

0.23, ****, p > 0.0001), with Bonferroni corrected post hoc tests revealing

differences between the remote extinction versus both recent and remote fear

memory recall (RE vs 7dF: t117 = 5.18, ****, p < 0.0001; RE vs 1dF: t17 = 5.07, ****,

p < 0.0001; 7dF vs 1dF: t117 = 0.10, ns, p > 0.9999). This finding was repeated in

measurements of the cell body area (F2,117 = 18.94, R2 = 0.25, ****, p > 0.0001; RE

vs 7dF: t117 = 5.18, ****, p < 0.0001; RE vs 1dF: t117 = 5.41, ****, p < 0.0001; 7dF

vs 1dF: t117 = 0.10, ns, p > 0.9999). One-way ANOVAs for each revealed significant

differences between groups (number of extensions: F2,117 = 17.27, R2 = 0.23, ****, p

> 0.0001; number of branches: F2,117 = 21.38, R2 = 0.27, ****, p > 0.0001; length of

extensions: F2,580 = 11.44, R2 = 0.04, ****, p > 0.0001). Post hoc evaluation revealed

the number of extensions (7dF vs 1dF: t117 = 4.69, ****, p < 0.0001; 1dF vs RE: t117

= 5.51, ****, p < 0.0001; 7dF vs RE: t117 = 0.86, ns, p < 0.9999), their branches (7dF

vs 1dF: t117 = 5.06, ****, p < 0.0001; 1dF vs RE: t117 = 6.21, ****, p < 0.0001; 7dF

vs RE: t117 = 1.20, ****, p < 0.0001) and their overall length (7dF vs 1dF: t117 = 4.21,

****, p < 0.0001; 1dF vs RE: t117 = 4.43, ****, p < 0.0001; 7dF vs RE: t117 = 0.21,

ns, p > 0.9999) were greater in both the extinction and remote fear memory recall

groups versus recent fear recall. Interestingly, together this data suggests the

morphology of the microglia shows increased ramification reminiscent of resting

state microglia in healthy brain tissue in the remote extinction recall group and a

trend towards this ramification in the remote fear recall subjects.

182

Figure 5-5 Morphological analysis of microglia.

A. Depiction of cell tracing showing the body, tree (extension from the body) and a branch (extension from a tree). B. Analysis of cell body volume revealed differences between the remote extinction versus both recent and remote fear memory recall (RE vs 7dF: t117 = 5.18, ****, p < 0.0001; RE vs 1dF: t17 = 5.07, ****, p < 0.0001; 7dF vs 1dF: t117 = 0.10, ns, p > 0.9999). C. Significant differences were noted between the number of extensions (7dF vs 1dF: t117 = 4.69, ****, p < 0.0001; 1dF vs RE: t117 = 5.51, ****, p < 0.0001; 7dF vs RE: t117 = 0.86, ns, p < 0.9999), (D) their branches (7dF vs 1dF: t117 = 5.06, ****, p < 0.0001; 1dF vs RE: t117 = 6.21, ****, p < 0.0001; 7dF vs RE: t117 = 1.20, ****, p < 0.0001) and (E) their overall length (7dF vs 1dF: t117 = 4.21, ****, p < 0.0001; 1dF vs RE: t117 = 4.43, ****, p < 0.0001; 7dF vs RE: t117 = 0.21, ns, p > 0.9999). Data are presented as mean ± S.E.M; ****: p ≤ 0.0001, one-way ANOVA followed by Bonferroni post hoc analysis. Scale bar 10 µm.

5.5 Discussion

The purpose of this study was to determine the extent to which fear memory recall

influences neuronal activity, and importantly, the morphology of microglia within

the prefrontal cortex. Our data indicate that recall of recent fear resulted in

significant morphological change in microglia. This finding is consistent with past

substantiation that microglia change due to exposure to psychological stress (Aji Nair

& Robert H Bonneau, 2006; Tynan et al., 2010). We extend on previous evidence

183

through demonstration that these morphological adaptations are time and memory

type dependent. As diversity in phenotype equates to diversity of function, this

outcome may suggest that microglial activation could play an important role in the

control of and adaptation to the fear response. If further examination supports this

theory microglia may constitute a novel target for attenuating the emotional and

physiological consequences of pathological fear.

Our behavioural findings provide evidence of the effective training and recall

of each memory category. The extended extinction protocol appeared to be

extremely successful at facilitating the development of a new extinction memory

more robust than the original fear memory. The levels of freezing to tone were

relatively similar in the case of a recent and remote fear memory, suggesting the

level of fear demonstrated by these two groups was equivocal at time of memory

recall.

Reduced neuronal activity has been reported as attenuating monitoring of

synapses by microglia (Wake et al., 2009). Our study reported lower c-Fos and Arc

expression in the extinction memory recall group, alongside smaller densities of

Iba1+ microglia. It appears possible that the lower number of surveillance cells

present may contribute to a lower frequency of monitoring. The recall of a recent

fear resulted in both increased neuronal activity as quantified by increased numbers

of c-Fos and Arc, and correspondent increases in Iba1+ microglia present in the

sample. However, the phenotype of the cells present was not conducive to synaptic

monitoring as they had short, thickened extensions more akin to an active microglia.

In contrast to microglia involved in previously documented studies on stress response

(Hinwood et al., 2012), these cells did not possess the larger cell bodies typical of

active microglia. This may be explained by the fact that the recent fear memory is an

184

example of an acute stressor as opposed to chronic. Therefore, the adaptation process

from surveillance to activated microglia may take longer than the 60 minutes we

permitted or that an immune response does not occur as a result of fear memory

recall. The morphology of microglia across all groups are not typical of pro or anti-

inflammatory microglia as the cell body increase observed was consistent with an

increased number of process as opposed to a reduction. Observations of microglia in

the rodent and human prefrontal cortex have identified four divisions of microglia:

ramified, primed, reactive and amoeboid microglia (Kongsui et al., 2014; Torres-

Platas et al., 2014). Unlike ramified microglia that have small cell bodies with large

extensions, the primed microglia have wider cell bodies but retain the standard

ramified extensions. Similar to the remote extinction recall group, the remote

memory group displayed ramified extensions. However, the cell bodies in this group

were smaller (not typical of an active microglia), indicating a morphology that may

have been transitioning towards a more active state, or reverting to a more

surveillance / resting state phenotype. In recent research from our lab examining

microglial morphology after contextual fear conditioning, cell body volumes were

not found to be different between contextually fear conditioned animals versus box

control in the rat dentate gyrus (Nicholas Chaaya et al., 2019), suggesting that

context alone may not generate these changes.

Although the experiments documented in this report reveal a phenotype

change and this change is generally associated with functional alteration

(Kettenmann et al., 2011), it should be noted that a marker of neuroinflammation

(such as MHC II) could provide additional evidence of functional changes between

memory classes. A box control group could define similarities between microglial

morphology and density in a group that had no fear memory to begin with in

185

comparison to a fear memory that has been updated with a new extinction memory.

An increase of cell number was found in the recent fear memory recall group.

Labelling with EDU (5-ethynyl-2’-deoxyuridine), a marker of cell proliferation

would assist in determining the nature of the increase as it may occur through

migration as opposed to proliferation. Transmembrane protein 119 (Tmem119)

staining may also provide insight into the proliferation versus infiltration issue as it

labels microglia exclusively. To further investigate the functional role microglia play

in fear memory, molecular analyses of mediators of inflammation could be

conducted at the RNA level, similar to those conducted by (Butovsky et al., 2014).

Colony stimulating factor 1 receptor (CSF1R), could be used after fear conditioning

to decrease microglia proliferation, a function previously shown by (Gerber et al.,

2018).

In conclusion this study is the first to characterise microglial phenotype after

fear memory recall. Our findings are consistent with the observation that microglial

activation could play an important role in facilitating fear responses and further

investigation into the functional changes of microglia and their effect on fear

memory itself may provide new therapeutic targets for phobias and pathological fear

as is present in post-traumatic stress disorder.

186

Contextual Fear Conditioning Alter Microglia Number

and Morphology in the Rat Dorsal Hippocampus

This chapter comprises the following published article:

Chaaya, N., Jacques, A., Belmer, A., Beecher, K., Ali, S. A., Chehrehasa, F., Battle, A. R., Johnson, L. R., Bartlett, S. E. Contextual fear conditioning alter microglia number and morphology in dorsal hippocampus. Frontiers in Cellular Neuroscience. Published 14th May, 2019. https://doi.org/10.3389/fncel.2019.00214

187








In the case of chapter 6: Contextual Fear Conditioning Alter Microglia Number and Morphology in the Rat Dorsal Hippocampus Publication status: Published

Contributor Statement of contribution*

Nicholas Chaaya Design of the project, behavioural and laboratory experiments, analysed data, created the figures and wrote and edited the manuscript.

Angela Jacques Assisted with behavioural experiment, laboratory experiments and imaging.

Arnauld Belmer Assisted with laboratory experiments, imaging and editing the manuscript.

Kate Beecher Assisted with laboratory experiments, assisted with editing the manuscript.

Syed Aoun Ali Assisted with laboratory experiments, assisted with editing the manuscript.

Fatemeh Chehrehasa Assisted with interpretation of data and editing of manuscript.


Luke Johnson Assisted with design of study and behavioural protocols

Selena Bartlett Involved in the conception and design of the study and interpretation of results.

QUT Verified

Signature

188

6.1 Abstract

Contextual fear conditioning is a Pavlovian conditioning paradigm capable of

rapidly creating fear memories to contexts, such as rooms or chambers. Contextual

fear conditioning protocols have long been utilised to study how fear memories are

consolidated, maintained, expressed, recalled and extinguished. This study has

identified the lateral portion of the amygdala and the dorsal portion of the

hippocampus as essential for contextual fear memory consolidation. The current

study was designed to evaluate how two different contextual fear memories alter

amygdala and hippocampus microglia, brain derived neurotrophic factor (BDNF;

which can be released by microglia), and phosphorylated cyclic-AMP response

element binding (pCREB; phosphorylation can be induced by BDNF). We find rats

provided with standard contextual fear conditioning to have more microglia and

express more BDNF in the dentate gyrus as compared to a context only control

group. Interestingly, the unpaired fear conditioning procedure, despite producing

equivalent levels of fear as the standard procedure, did not alter microglia, BDNF or

pCREB number in any dorsal hippocampus or lateral amygdala brain regions. In the

dentate gyrus, standard contextual fear conditioning alters microglia morphology to

become amoeboid in shape, which is a common response to central nervous system

insult, such as traumatic brain injury, infection, ischemia and more. Additionally, the

unpaired fear conditioning protocol produced some alterations in microglia

morphology, also becoming more amoeboid in shape. These data suggest that

contextual fear conditioning is capable of producing large alterations to dentate gyrus

function, whereas unpaired fear conditioning produces minor changes. Cumulatively,

these data suggest that Pavlovian fear conditioning protocols induce similar

189

responses as trauma, infection or other direct causes within the central nervous

system.

Key words: Contextual Fear Conditioning, Microglia, BDNF (brain-derived

neurotrophic factor), Hippocampus & Amygdala, Dentate Gyrus

6.2 Introduction

Contextual fear conditioning (CFC) is a Pavlovian conditioning protocol

whereby an animal, typically a rodent, is placed into a context (conditioned stimulus;

CS) and provided with noxious stimuli (unconditioned stimulus; US) (Nicholas

Chaaya et al., 2018; Michael S. Fanselow, 2010; Foa et al., 1992; Joseph E. LeDoux,

2014; Rothbaum & Davis, 2003). CFC, along with similar fear conditioning

protocols are utilised to replicate the behavioural events that lead to the development

of fear-based disorders, namely, post-traumatic stress disorder (PTSD) (Nicholas

Chaaya et al., 2018; Michael S. Fanselow, 2010; Foa et al., 1992; Joseph E. LeDoux,

2014; Maren, 2011; Rothbaum & Davis, 2003). Utilising these conditioning

protocols, researchers have identified various essential brain regions, circuits and

molecules involved in the consolidation, maintenance, expression, recall and

extinction of fear (Michael S. Fanselow, 2010; Joseph E. LeDoux, 2014; Maren,

2011). During CFC, the amygdala and hippocampus, or more specifically, the lateral

amygdala (LA) and dorsal hippocampus (DH), have been identified as two brain

regions critical for its consolidation (Nicholas Chaaya et al., 2018). Numerous

investigations, largely directed by Fanselow, have demonstrated an inability for

contextual fear memories to be consolidated when these regions are inhibited or

190

abolished (Anagnostaras, Maren, & Fanselow, 1999; Michael S. Fanselow, 2010;

Maren, Aharonov, & Fanselow, 1997; Maren & Fanselow, 1997). Additional

investigations exploring cellular and molecular alterations following learning in these

regions have also identified a critical role for various plasticity and activity related

proteins and immediate early genes (IEGs) in the LA and DH following CFC (Barot

et al., 2009; Besnard, Laroche, & Caboche, 2014; N. Chaaya et al., 2019; Choi et al.,

2016; Impey et al., 1998; Perez-Villalba et al., 2008; Pignataro, Middei, Borreca, &

Ammassari-Teule, 2013; Sananbenesi, Fischer, Schrick, Spiess, & Radulovic, 2002;

Y. M. Wilson & Murphy, 2009; Zelikowsky et al., 2014; Zheng, Luo, Guo, Cheng, &

Li, 2015). The current study expands on these and our previous work (N. Chaaya et

al., 2019) which demonstrated how minor changes to the context during CFC alter

LA activity. This study explores a variety of cellular and molecular alterations in LA

as well as the DH.

The unpaired fear conditioning (UFC) protocol is an alternate Pavlovian

conditioning procedure capable of robustly producing contextual fear memories (H.

C. Bergstrom et al., 2012; Hadley C Bergstrom et al., 2011; Trifilieff et al., 2007).

Despite this, UFC protocols have traditionally been utilised as controls for cued fear

learning (e.g. auditory fear conditioning; AFC) procedures (H. C. Bergstrom et al.,

2012; Hadley C Bergstrom et al., 2011; Majak & Pitkänen, 2003; McKernan &

Shinnick-Gallagher, 1997; Radley et al., 2006; Michael T. Rogan et al., 1997). While

the CS (e.g. tone) and US (e.g. foot-shock) are explicitly paired in cued fear

conditioning, they are presented at random non-overlapping times during UFC. This

led to the hypothesis that the amygdala would not be activated, as no explicit CS-US

associative memory was formed (H. C. Bergstrom et al., 2012; Hadley C Bergstrom

et al., 2011; Majak & Pitkänen, 2003; McKernan & Shinnick-Gallagher, 1997;

191

Radley et al., 2006; Michael T. Rogan et al., 1997). However, as noted above,

contextual fear memories activate both LA and DH. Therefore, the ability for fear

memories to context to be formed following UFC suggest the LA, DH, or other

related brain region are similarly recruited. Indeed, this has been demonstrated

previously in LA by Trifilieff and colleagues (quantified phosphorylated mitogen-

activated protein kinase; pMAPK (Trifilieff et al., 2007)), and more recently by our

lab (quantified IEGs (N. Chaaya et al., 2019)). The current study aims to expand on

this research by exploring additional molecular and cellular alternations in both DH

and LA.

Microglia are functionally and anatomically distinct central nervous system

(CNS) cells that possess macrophage-like function (Walker et al., 2014). They are

crucially involved in responding to infection, trauma, ischemia and other insults of

the brain, and participate in maintaining neuronal integrity (Kettenmann et al., 2011;

Pósfai, Cserép, Orsolits, & Dénes, 2018). Briefly, microglia respond to insult in two

main ways: they increase in number in the affected area and their morphology (cell

body size and extension number/size) alter (Calcia et al., 2016; Kettenmann et al.,

2011). Resting (ramified) microglia in a healthy CNS system have a small cell body

and long, thin extensions, with many processes (Calcia et al., 2016; Dwyer & Ross,

2016; Kettenmann et al., 2011). In this ramified stage microglia search for signals of

insult (Dwyer & Ross, 2016; Kettenmann et al., 2011; Walker et al., 2014). Upon

detection of harmful stimuli, microglia number increase in the affected area, and

morphology alter to become amoeboid, with most notable changes being an increase

in cell body size, and reduced branching, and number of extensions (Dwyer & Ross,

2016; Kettenmann et al., 2011; Walker et al., 2014). Furthermore, in this state,

microglia release a number of factors and compounds, one such being brain derive

192

neurotrophic factor (BDNF), an important neurotrophin involved in neuronal

survival and differentiation (Ferrini & De Koninck, 2013; Pósfai et al., 2018). Recent

research has begun to identify how microglia respond to psychosocial stressors, with

various studies (extensively reviewed by (Calcia et al., 2016)) showing alterations

corresponding to that of an injured or insulted brain. To our knowledge, one study

has identified such alterations following chemically induced fear (Vollmer et al.,

2016). However, no research has directly examined how Pavlovian fear conditioning

protocols alter microglia number and morphology. The current study, is therefore

designed to examine how CFC and UFC result in alterations in microglia number and

morphology. Furthermore, we investigate how these protocols alter BDNF

expression (which can be released by microglia), as well as phosphorylated cyclic-

AMP response element binding (pCREB) expression (plasticity marker that can be

activated by BDNF) (Ferrini & De Koninck, 2013; Minichiello, 2009). Both pCREB

and BDNF expression have been demonstrated to be involved in fear memory

formation, making them good candidate proteins for the investigation of differential

fear memory consolidation (Bekinschtein et al., 2007; Hall, Thomas, & Everitt,

2000; Impey et al., 1998; I. Y. C. Liu, Lyons, Mamounas, & Thompson, 2004;

Mamiya et al., 2009b; Mizuno, Dempster, Mill, & Giese, 2012).

The objectives of the current study were to explore how two different

Pavlovian fear conditioning protocols, capable of creating contextual fear memories,

alter LA and DH microglia (identified by labelling for ionized calcium binding

adaptor molecule 1; IBA-1) number and morphology, BDNF number and pCREB

number. The current study provides the first insights into how these Pavlovian fear

conditioning protocols alter microglia number and morphology. We find that

microglia number and BDNF expression increase in the dentate gyrus (DG)

193

subregions of the DH, following CFC, as compared to a context only (CO) control

group. Further investigations show that CFC alter microglia morphology to become

amoeboid (responding to insult). Interestingly, we show that UFC does not lead to a

change in BDNF or microglia number. Despite this, investigations of microglia

morphology in UFC suggest they also appear to have some characteristics of

amoeboid microglia; signifying they also responded to insult.

6.3 Method

6.3.1 Animals

Animals were experimentally naïve adult male Sprague Dawley Rats

(Animal Resources Centre, WA, Australia). Data reported here was gathered from

rats that make part of a larger dataset. Rats weighed 176-200 g at arrival and were

housed, two per cage, by the University of Queensland Biological Resources

(UQBR) facility on a 12-hour light/dark cycle. Food and water was provided ad

libitum. All behavioural procedures were approved by the University of Queensland

(Ethics approval no. 023/17) and Queensland University of Technology (QUT

approval number: 1700000295) animal ethics unit. All procedures complied with the

Queensland Government Animal Research Act 2001, associated Animal Care and

Protection Regulation (2002 and 2008), as well as the Australian Code for the Care

of Animals for Scientific Purposes, 8th Edition (National Health and Medical

Research Council, 2013) policies and regulations of animal experimentation and

other ethical matters. Upon arrival, rats were acclimatised to the UQBR Facility for

eight days, handled by the experimenter for nine days, habituated to the fear

conditioning context for one day, and then, 24 hours later, fear conditioned (now

weighing 326.56g +/- 2.8g on the fear conditioning day) as explained previously (N.

Chaaya et al., 2019). There were two experimental (Contextual Fear Conditioned;

194

CFC n = 18, Unpaired Fear Conditioned; UFC n = 18) groups and one control

(Context Only; CO n = 18) group. Rats were divided into anatomical (n = 12 per

group) and behavioural (n = 6 per group) subgroups following experimental

procedures.

6.3.2 Apparatus

All procedures occurred in one of two Plexiglas conditioning chambers

(Coulbourn Instruments, Lehigh Valley, Pennsylvania, USA). A single house light (2

– 3 lux) dimly illuminated both chambers (context A and B). Chambers contained an

infrared camera, were equipped with a speaker and sound insulated (background dB

= 55). Context A contained a metal grid floor which connected to an electric shock

generator. This context contained no decorations, and was cleaned with ethanol

(EtOH) 80% following the presentation of each rat. Alternatively, context B was

fitted with a flat floor that was lightly covered with bedding. The walls were

coloured, and alterations were made to the roof which altered its physical

dimensions. Following the presentation of each rat, orange scented hand soap was

used to clean context B. The bedding was also replaced.

6.3.3 Procedures and Design

Figure 1 briefly outlines behavioural procedures. These procedures

are explained in detail below and have been outlined previously (MS1; (N. Chaaya et

al., 2019)).

195

Acclimatisation, Habituation and Fear Conditioning

Prior to behavioural procedures, rats in all conditions were acclimatised to the

vivarium for eight days. Rats were handled for nine days by the experimenter, and

then each placed in context A for 30 minutes on the tenth day. After twenty-four

hours, rats in the CFC and UFC group were placed into context A for fear

conditioning. Rats were permitted 180 s to explore the context before receiving any

Figure 6-1. Experimental Design for Behavioural Training. Following an acclimatisation, handling and habituation period, rats were divided into three distinct behavioural groups. Rats in the CFC group were placed into a chamber and provided with five non-overlapping and random electric shocks to the foot. Rats in the UFC were placed in the same chamber and provided with the same foot-shocks. However, five non-overlapping and random auditory tones were also presented during the fear conditioning procedure. Rats in the CO control group were placed in the same chamber and provided with no further stimuli. Following conditioning, rats in all three groups were separated into an anatomy (perfused 90 minutes post-conditioning) and behavioural group (provided with two fear memory test 24 hours after conditioning, and four days after conditioning). CFC: contextual fear conditioning; UFC: unpaired fear conditioning; CO: context only.

196

stimuli. Rats in the CFC group were then presented with five non-overlapping and

random electric shocks to the foot (1.0 mA, 0.50 s). Whilst in receiving these foot-

shocks in context A, rats in the UFC group also received five presentations of

auditory tones (5 kHz, 75 dB, 20 s). These auditory tones did not overlap with each

other, or with the foot-shocks. Following presentation of the final stimulus, rats

remained in the context for 60 s. The experimenter then removed the rats, and

returned them to their home-cage. In total, fear conditioning procedures were 660 s

long for rats in the CFC group, and 880 s long for rats in the UFC (extra time was

required to account for the addition of auditory tones). Rats in the CO control group

were placed into context A without any added stimuli, and left to explore for 660 s.

Following Fear Conditioning

Behavioural subgroup.

The behavioural subgroup of rats (n = 6 per group) had freezing behaviour

manually scored during training (fear conditioning) and testing (fear memory test:

FMT). As per previous investigations (H. C. Bergstrom & L. R. Johnson, 2014; H. C.

Bergstrom et al., 2012; R. G. Phillips & LeDoux, 1992; Russel G Phillips & LeDoux,

1994; Gregory J. Quirk et al., 1997; Radley et al., 2006), all scoring occurred in 20

second blocks. During training a progressive measure of fear was obtained by

scoring freezing behaviour before (baseline), during (cue 1 - 5), and after fear

conditioning (final). Following training, rats were returned to their home-cages, and

kept there for 24 hours. Rats were then placed back into context A, and freezing

behaviour was scored for a 10 minute FMT to context (no foot-shocks or auditory

tones provided). Freezing behaviour was scored during the final 20 seconds of every

minute that rats were undergoing their FMT to context. Rats were returned to their

home-cages for 72 hours, until which a FMT to tone was conducted. During the FMT

197

to tone, rats were placed in context B for 10 minutes and presented with auditory

tones (5 kHz, 75 dB, 20 s) at the end of each minute. Freezing behaviour was scored

during the 20 s tone presentation periods.

Scoring of freezing.

Freezing behaviour was defined as the inhibition, absence or suppression of

movement, besides from that required for autonomic nervous system functioning

(Michael S. Fanselow, 1980). Head scanning and sleeping were not included as

freezing. Heavy breathing, minimal movement and other movements required for

normal respiration and autonomic function were considered as freezing behaviour.

During training, freezing behaviour was scored in the CFC and UFC groups during

the final 20 s of the first minute, the final 20 s of the last minute, and the 20 seconds

prior to each foot-shock. Freezing behaviour was not scored prior to auditory tone

presentation in the UFC, as the development of contextual fear memories are of

interest here. For rats in the CO control group, freezing behaviour was scored at

identical time points as that of the CFC group (as their trials had identical durations).

During testing, freezing behaviour was scored in the final 20 seconds of every

minute (H. C. Bergstrom et al., 2012; Hadley C Bergstrom et al., 2011).

Anatomical group.

Following completion of fear conditioning, rats in the anatomical group (n =

12 per group) were removed from the conditioning context and immediately returned

to their home cages. Ninety minutes following, rats were anesthetised and sacrificed

via perfusion for fluorescent labelling of pCREB, BDNF and IBA-1.

6.3.4 pCREB, BDNF and IBA-1 Immunohistochemistry

Tissue preparation

To anaesthetise rats, intraperitoneal (i.p.) injections of Ketamine/Xylazine

(100mg/kg, 10mg/kg) were administered. Following anaesthetisation, rats were

198

transcardially perfused with ice-cold saline (200 mL per rat) followed by 4%

paraformaldehyde/0.1 M phosphate buffer (PB; pH of 7.4; 400 mL per rat) via the

ascending aorta. Subsequently, brains were removed and stored at 4 oC in the 4%

paraformaldehyde fixative for 24 hours. Following, brains were stored in phosphate

buffered saline (PBS)/0.02% Azide for a minimum of three days, at which point free-

floating sequential coronal brain sections were obtained. These sections, sliced on a

vibratome (M11000; Pelco easiSlicer, Ted Pella Inc, CA, USA) at 40 μm per section,

contained the lateral amygdala and the hippocampus. Sections were stored at 4 oC in

PBS/0.02% azide until immunohistochemistry commenced. Immunohistochemistry

was conducted on right hemisphere sections.

Immunohistochemistry

Brain sections were removed from PBS/0.02% azide and washed thoroughly

with PBS. Optimisation of each antibody required altered protocols. These

alterations are outlined below. First, all sections were post-fixed for an additional

five minutes with the 4% paraformaldehyde fixative used for perfusion. Sections

were then thoroughly washed with PBS. Sections labelled for BDNF and IBA-1 (but

not pCREB) were incubated in 3% H2O2/10% Methanol in PBS for five minutes.

Once washed in PBS, all sections were permeabilised with 1% Triton/0.1% Tween

20 in PBS for one hour, and then washed in PBS again. Labelling for BDNF required

antigen retrieval; whereby sections were incubated in Citrate Buffer (10mM Sodium

Citrate, 0.05% Tween 20, pH 6.0) for five minutes (80 oC). Once sections returned to

room temperature, they were washed in PBS. All sections were then blocked with

0.3% Triton/0.05% Tween 20/2% normal horse serum (NHS) in PBS for one hour.

Blocking solution was removed, and sections were incubated in their respective

primary (either pCREB, BDNF or IBA-1) antibody diluted in the blocking solution

199

for 24 hours. Sections labelled for pCREB were incubated in anti-phospho-CREB

(Ser133) rabbit polycloncal antibody (1:500; Merck Milipore, HE, DEU). Sections

labelled for BDNF were incubated in anti-BDNF [EPR1292] (ab108319) rabbit

monoclonal antibody (1:500; Abcam, VIC, Aus). Sections labelled for IBA-1 were

incubated in anti-IBA1 (ab5076) goat polyclonal antibody (1:500; Abcam, VIC,

Aus). Following incubation in primary antibody, all sections were washed with

blocking solution. Sections labelled for pCREB and BDNF were immediately

incubated in a pre-absorbed goat anti-rabbit IgG H&L (Alexa Fluor 594) secondary

antibody (1:500; Abcam, VIC, Aus) in blocking solution, whereas sections labelled

for IBA-1 were immediately incubated in a cross-absorbed donkey anti-sheep IgG

H&L (Alexa Fluor 594) secondary antibody (1:500; ThermoFisher Scientific, VIC,

Aus). Brain sections were then washed in blocking solution and then PBS.

Following, sections were incubated in 4′,6-diamidino-2-phenylindole (DAPI) diluted

in PBS for five minutes (1:1000; D1306 ThermoFisher Scientific, VIC, Aus), washed

a final time, and then mounted on silane coated slides. Mounted sections were

immediately cover-slipped using ProLong Gold antifade reagent (Invitrogen, DR,

USA), left to dry and stored at 4 oC.

Image Acquisition

Cover-slipped brain sections were scanned using a Nikon/Spectral Spinning

Disc Confocal Microscope (Nikon Instruments Inc, NY, USA) to take 20x magnified

tile-scan mosaics of the amygdala and hippocampus. Scan settings are as follows: x =

14, y = 10 consecutive fields (horizontal acquisition pattern) with 10% overlap and 7

z-stacks with 4 μm step-size. Laser channels were 405nm for DAPI (20ms exposure

time) and 561nm (high) for pCREB, BDNF and IBA-1 (300ms exposure). Individual

scans (each z-stack and wavelength/channel as a separate image) were saved as

200

separate .tiff files, and manually merged using imageJ (Schindelin et al., 2012).

Merged z-stacks and channels were then stitched (Preibisch et al., 2009) in imageJ.

The Olympus FV3000 Confocal Laser Scanning Microscope (Olympus Australia Pty

Ltd, VIC, Aus) was used to take 40x magnified (1.5x zoom) scans (x = 212 μm, y =

212 μm) with 30 z-stacks of 0.50 μm thickness (z = 15 μm). These scans were only

conducted on IBA-1 sections to allow for the tracing of these cells. Only brain

subregions that were identified to have significant differences in pCREB, BDNF and

IBA-1 number (as identified with initial Spinning Disc Confocal scans) were scanned

with the FV3000.

Amygdala and Hippocampus Subregion Identification

All brain sections contained the LA and its three subregions: dorsolateral

portion of the lateral amygdala (LaDL), ventromedial portion of the lateral amygdala

(LaVM) and ventrolateral portion of the lateral amygdala (LaVL), as well as the

dorsal hippocampus (DH) and its three subregions: dentate gyrus (DG), CA1 and

CA3. The CA2 was excluded from analysis as it is significantly smaller than the

other hippocampal subregions, and is difficult to accurately outline (Caruana,

Alexander, & Dudek, 2012). Stereotaxic alignment was utilised to accurately analyse

pCREB, BDNF and IBA-1 expression in the same rostral-caudal location of each rat

brain. Briefly, the lateral ventricle (LV) – a rapidly changing anatomical landmark –

becomes present at Bregma coordinate -3.32 mm, and consistently grows towards

more caudal locations (Paxinos & Watson, 2006). The LV becomes easily

identifiable at Bregma coordinate -3.36 mm (depicted as a tear-drop size; Paxinos &

Watson, 2006), allowing for accurate identification and alignment. Following

identification of Bregma coordinate -3.36 mm, preceding sections can be identified

by counting back. Identification of Bregma coordinate -3.36 mm allowed for all rat

201

brains to be aligned at the same rostral-caudal location. To provide a larger

representation of the LA and DH, two sections per rat (equal distances apart) were

labelled with each antibody. Anatomical landmarks, such as the external capsule

(ec), rhinal fissure (RF), central amygdala (CeA), dorsal endopiriform nucleus

(DEn), optic tract (opt) and stria terminalis (st) were used to identify the three LA

subregions (see Figure 2). The hippocampal subregions were more easily identified

by the clear structural alterations of each subregion (see Figure 2).

6.3.5 Neuron/Microglia Quantification

Neurons and Microglia were automatically counted and tagged using the spot

detection option in IMARIS (IMARIS 9.1.2, Bitplane, ZH, CHE). An average

diameter was obtained for cells labelled with pCREB, BDNF and IBA-1. Filter

intensity was determined by the experimenter, and kept constant across experimental

groups. Cell diameter and filter intensity were altered, if necessary, depending on

amygdala or hippocampus subregion, but remained constant across experimental

groups for each subregion. Representative areas of LA subregions (x = 266.66 μm, y

= 266.66 μm) were selected and the number of cells and microglia quantified (see

Figure 2). Due to the relatively large size of the DH and its subregions, three areas (x

= 333.33 μm, y = 333.33 μm) were counted from each DH subregion (see Figure 2).

Tracing of microglia was dependent upon pCREB, BDNF and IBA-1 quantification.

In identified brain subregions, a maximum of three microglia per section were traced

using Neurolucida 360 (Neurolucida 360, MBF Bioscience, VT, USA). The average

length of microglia extensions, number of trees or ends for these extensions, the cell

body volume and complexity of the microglia branching (measure of ramification

state) were quantified from these traces. The following formula was utilised to

determine the complexity of microglia branching (sum of the terminal orders +

202

number of terminals) * (total process length / number of primary branches) (Pillai et

al., 2012).

6.3.6 Data Analysis

The results section is separated into three parts: pCREB expression, BDNF

expression and IBA-1 expression. Within each part, differences as a function of

Figure 6-2. Illustration of LA and DH subregions and labelling in these Regions. Three subregions of interest within the LA, and three subregions of interest within the DH, were identified in the rat brains. (A) To accurately identify LA subregions LaDL, LaVM and LaVL, the CeA, LV (when present), st, opt, Den and RF were used (shaded in grey) as anatomical landmarks. Following identification of subregions, pCREB (B), BDNF (C) and IBA-1 (D) were quantified. (E) The hippocampus has easily distinguishable subregions as a result of its unique anatomical architecture. Therefore, identification of CA1, CA3 and DG relied upon the assistance of the (Paxinos & Watson, 2006) rat brain atlas. Due to the limited research and relative size of the CA2 subregion, this region was utilised as a border between the CA1 and CA3. Following identification of subregions, pCREB (F), BDNF (G) and IBA-1 (H) were quantified. LA: lateral amygdala; DH: dorsal hippocampus; LaDL: dorsolateral portion of the lateral amygdala; LaVM: ventromedial portion of the lateral amygdala; LaVL: ventrolateral portion of the lateral amygdala; RF: rhinal fissure; DEn: dorsal endopiriform nucleus; CeA: central amygdala; st: stria terminalis; opt: optic tract; pCREB: phosphorylated cyclic-AMP response element binding; BDNF: brain derived neurotrophic factor; IBA-1: ionized calcium binding adaptor molecule 1; CA1: hippocampal subregion CA1; CA2: hippocampal subregion CA2; CA3: hippocampal subregion CA; DG: dentate gyrus. Scale bar: 100 µm.

203

conditioning (CFC, UFC and CO control) across all LA and DH subregions are

explored. For this reason, analyses of variances (ANOVAs) were utilised to compare

differences between behavioural groups within each subregion (e.g. differences in

pCREB expressing neurons between behavioural groups in the LaDL). Prior to

analysis, normality and homogeneity of variance were tested for. Assumptions of

normality and homogeneity of variance were confirmed in the majority of cases. To

control for possible type I errors arising from these breaches, a Bonferroni

adjustment (Perneger, 1998) was utilised in all follow-up post-hoc tests. The

Bonferroni adjustment also controlled for multiple comparisons that were conducted

in these analyses. Therefore, all analyses to anatomical data were conducted with

one-way ANOVAs, followed by Bonferroni corrected post-hoc tests. All values in

the text and graphs are expressed as the mean +/- standard error of the mean. P

values at or below 0.05 are considered statistically significant. All major statistical

analyses, outlier analyses and graph generation were conducted using GraphPad

Prism v7 software (GraphPad, CA, USA). Asterisks are used to denote levels of

statistical significance within all graphs (* p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001; ****

p ≤ 0.0001). Behavioural data is previously reported in MS1 (N. Chaaya et al., 2019),

and therefore summarised briefly at the beginning of the results section.

Excluded Cases

Statistical outliers or significantly damaged brain tissue (occurring from

perfusion, labelling or cover-slipping process) were excluded from analyses.

Statistical outliers were identified via the ROUT method (GraphPad Prism), with the

maximum false discovery rate set to 1%. The ROUT method is capable of effectively

identifying multiple outliers in large datasets. Outlier analysis was conducted on

individual groups, and identified outliers were excluded on a pairwise basis.

204

6.4 Results

6.4.1 Behavioural Results

In order to determine confirm that the behavioural procedures produced

differences in fear to context, freezing behavioural during training and testing was

quantified. The percentage of time rats displayed freezing behaviour as a function of

condition (CFC, UFC or CO) and time-point (baseline minute, cue 1 – 5, and final

minute) were quantified as explained previously (N. Chaaya et al., 2019). A two-way

mixed design ANOVA revealed a significant interaction of freezing behaviour as a

function of condition and time-point (p < 0.0001), a significant main effect of

condition (p < 0.0001) and a significant main effect of time-point (p < 0.0001).

Bonferroni corrected post-hoc tests (see Figure 3A) demonstrated a progressive

acquisition of fear to context, becoming significantly different between both

conditioned groups as compared to the CO control group starting from cue 4. One-

way ANOVA of freezing during the FMT to context revealed significant differences

between groups (p < 0.0001). Bonferroni corrected post-hoc tests revealed rats that

underwent CFC and UFC to exhibit significantly more freezing to context as

compared to rats in the CO control group (Figure 3B). Alternatively, one-way

ANOVA of freezing during the FMT to tone revealed no statistical differences to

exist between groups (see Figure 3C).

205

6.4.2 Anatomical Results

pCREB Expression Does Not Changed by both CFC and UFC

One-way ANOVA of pCREB (a marker for neuronal plasticity) expression in

LA and DH revealed no group differences to exist (see Figure 4). No differences in

Figure 6-3. Freezing to Context and Tone Data. Previously reported freezing to context and tone (N. Chaaya et al., 2019) reveal both training protocols to robustly create fear memories to context. (A) Analysis of fear-related freezing during conditioning reveal rats that underwent both CFC and UFC to progressively develop fear memories. By the final 20 second period, rats in the CFC and UFC both expressed significantly more fear-related freezing to context as compared to the CO control. (B) During the FMT to context provided 24 hours following conditioning, rats in both the CFC and UFC expressed significantly more fear-related freezing behaviour as compared to the CO control. There was no difference in fear-related freezing between the CFC and UFC group. (C) During the FMT to tone provided three days after the FMT to context, fear-related freezing was equivalent in all groups. CFC: contextual fear conditioning; UFC: unpaired fear conditioning; CO: context only; FMT: fear memory test. Asterisks denote level of statistical significance between groups * p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001; **** p ≤ 0.0001.

206

pCREB number were observed as a function of conditioning in LA subregions LaDL

(F[2, 55] = 0.0431, p = 0.9578), LaVM (F[2, 55] = 1.282, p = 0.2856), or LaVL

(F[2, 55] = 0.1709, p = 0.8433). Similarly, no group differences were observed in

DH subregions CA1 (F[2, 61] = 0.5423, p = 0.5842), CA3 (F[2, 60] = 0.7473, p =

0.4780) and DG (F[2, 60] = 0.5430, p = 0.5838).

BDNF Expression in DG is Increased by CFC

One-way ANOVA of BDNF expression in LA and DH subregions CA1 and

CA3 provided mostly similar results as that of pCREB expression. No differences in

BDNF expression were noted in LA (see Figure 5A) subregion LaDL (F[2, 57] =

0.3216, p = 0.7263), LaVM (F[2, 56] = 0.1123, p = 0.8940) or LaVL (F[2, 57] =

1.0571, p = 0.3542). Furthermore, no differences in BDNF expression were noted in

DH subregion (see Figure 5B) CA1(F[2, 56] = 0.6008, p = 0.5519) or CA3 (F[2, 56]

= 0.8187, p = 0.4462). Contrary to pCREB data above, one-way ANOVA revealed a

significant difference in BDNF expression as a function of condition in DH

Figure 6-4. pCREB Expression in LA and DH. Evaluation of pCREB expression in LA (A) and its subregions and DH (B) and its subregions revealed no statistically significant differences as a function of conditioning. LA: lateral amygdala; DH: dorsal hippocampus; pCREB: phosphorylated cyclic-AMP response element binding.

207

subregion DG (F[2, 55] = 8.1509, p < 0.001). Bonferroni correct post-hoc tests (see

Figure 5B) found rats that underwent CFC to have significantly more BDNF

expression as compared to the control group. Interestingly, rats that underwent CFC

also exhibited significantly more BDNF expression in DG as compared to those that

underwent UFC.

IBA-1 Number in CA1 and DG is Increased by CFC

Similar to pCREB and BDNF expression, one-way ANOVA of IBA-1

number revealed no differences in the LA (see Figure 6A) as a function of

conditioning. The number of IBA-1 remained the same in subregions LaDL (F[2, 57]

= 0.9217, p = 0.4037), LaVM (F[2, 58] = 0.3137, p = 0.7320) and LaVL (F[2, 58] =

0.9302, p = 0.4003) as a function of conditioning. Alternatively, in DH (see Figure

Figure 6-5. BDNF Expression in LA and DH. (A) Evaluation of BDNF expression in LA and its subregions revealed no statistically significant differences as a function of conditioning. (B) Evaluation of BDNF expression in DH revealed statistically significant differences in subregion DG as a function of condition. Rats that underwent CFC had significantly more BDNF expression as compared to both rats that underwent UFC and the CO control rats. LA: lateral amygdala; DH: dorsal hippocampus; BDNF: brain derived neurotrophic factor; DG: dentate gyrus; CFC: contextual fear conditioning; UFC: unpaired fear conditioning; CO: context only. Asterisks denote level of statistical significance between groups * p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001; **** p ≤ 0.0001.

208

6B), one-way ANOVA revealed differences in the number of IBA-1 in subregions

CA1 (F[2, 55] = 3.4574, p < 0.05) and DG (F[2, 54] = 5.928, p < 0.01), but not CA3

(F[2, 53] = 0.8517, p = 0.4324). Bonferroni corrected post-hoc tests (see Figure 6B)

showed rats that underwent CFC to have significantly more IBA-1 as compared to

the CO control in subregions CA1 and DG. Similar to BDNF data, IBA-1 number in

DG was also significantly higher in the CFC group as compared to the UFC group in

DG.

IBA-1 Morphology in DG is Altered by CFC

Due to consistent differences in BDNF expression and IBA-1 number

between groups (CFC versus CO control) following conditioning, IBA-1 was traced

in DG (see Figure 7A). Four measures were obtained from traced IBA-1: average

Figure 6-6. IBA-1 Expression in LA and DH. (A) Evaluation of IBA-1 number in LA and its subregions revealed no statistically significant differences as a function of conditioning. (B) Evaluation of IBA-1 number in DH revealed statistically significant differences in subregions CA1 and DG as a function of condition. Rats that underwent CFC had significantly more IBA-1 as compared to the CO control in subregions CA1 and DG. Furthermore, rats that underwent CFC had significantly more IBA-1 than rats that underwent UFC in the DG. LA: lateral amygdala; DH: dorsal hippocampus; IBA-1: ionized calcium binding adaptor molecule 1; CA1: hippocampal subregion CA1; DG: dentate gyrus; CFC: contextual fear conditioning; UFC: unpaired fear conditioning; CO: context only. Asterisks denote level of statistical significance between groups * p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001; **** p ≤ 0.0001.

209

length of extensions, number of ends, cell body volume and complexity (as a

measure of IBA-1 ramification or activation state). One-way ANOVA revealed

significant differences in IBA-1 morphology in average length of extensions (F[2,

76] = 18.3192, p < 0.0001), number of ends (F[2, 80] = 8.1239, p < 0.001),

complexity (F[2, 70] = 17.2069, p < 0.0001), but not cell body volume (F[2, 80] =

1.493, p = 0.2308) (see Figure 7B – E). Bonferroni corrected post-hoc tests (see

Figure 7B – E) revealed rats in the CFC group to have significantly larger extensions

than both the UFC group and CO control group. Interestingly, despite no difference

in IBA-1 number between UFC and CO controls, rodents in the UFC had

significantly larger extensions than the CO control. Bonferroni corrected post-hoc

tests revealed rats in the CFC group to have significantly more ends than those in the

UFC and CO control group. No differences were observed between the UFC group

and CO control group. Finally, statistical analyses on complexity data revealed rats

in the CFC and UFC group to have less complex (or less ramified), IBA-1 as

compared to the CO control group.

211

6.5 Discussion

This study investigated the involvement of amygdala and hippocampus

following the formation of contextual fear memories created via separate protocols.

The primary difference between these two protocols was the amount of time spent in

the fear conditioning context (660 s for rats in the CFC group versus 880 s for rats in

the UFC group), and importantly, the inclusion of five non-reinforced auditory tones

for rats in the UFC group. Expression of pCREB and BDNF, along with the number

of IBA-1 were quantified and compared in these two groups to a CO control group

that explored the same context, but with no stimuli provided. Behavioural results

(previously reported in (N. Chaaya et al., 2019)) demonstrate successful acquisition

of fear to context, but not to tone, in both conditioned groups. Despite these

behavioural changes, investigations of LA pCREB, BDNF and IBA-1 did not reveal

Figure 6-7. IBA-1 Morphology in DG. (A) The 3D projection of scanned microglia were manually traced in each behavioural group. These traces provided data regarding the average length of extensions, the number of trees/endings, the cell body volume and the complexity of the extensions. (B) Rats that underwent CFC were found to have the shortest average length of extensions. Their extensions were significantly shorter than those in the CO control group. Rats that underwent UFC had significantly longer extensions than those in the CFC group, but significantly shorter extensions than those in the CO control group. (C) Rats that underwent CFC were found to have significantly less endings as compared to both the UFC group and the CO control group. (D) No difference microglia cell body volume was identified as a function of fear conditioning. (E) Analysis of the complexity of extensions (providing a measure of microglia ramification state) revealed the CFC and UFC group to both have significantly less complex extensions than the CO control group. This indicates that only the CO group had ramified or resting microglia. (E) Max projection of IBA-1 microglia with tracing overlaid (top) and the tracing alone (bottom). DG; dentate gyrus; CFC: contextual fear conditioning; UFC: unpaired fear conditioning; CO: context only; IBA-1: ionized calcium binding adaptor molecule 1. Asterisks denote level of statistical significance between groups * p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001; **** p ≤ 0.0001. Scale bar = 10 µm.

212

any changes following either CFC or UFC. These data suggest the markers examined

here may not be capable of identifying alterations in the LA that exist following

these forms of conditioning protocols. Contrary to this, differences in BDNF and

IBA-1 number were present in hippocampus following CFC, but not UFC.

Specifically, an increase in the expression of BDNF in hippocampal subregion DG,

and IBA-1 in hippocampal subregions CA1 and DG was present. Interestingly, DG-

specific BDNF and IBA-1 were significantly higher following CFC as compared to

both the CO control group, and the UFC group. These data suggest that contextual

fear memories created with standard CFC protocols rely upon the DG. However,

when contextual fear memories are altered with non-reinforced auditory tones,

dependency for the DG appears to be lost.

6.5.1 Lateral Amygdala

Contextual fear memories created with either the standard CFC protocol or

the altered UFC protocol does not appear to rely upon the LA. Evaluation of pCREB,

BDNF and IBA-1 number in the LaDL, LaVM and LaVL did not reveal any changes

as a function of fear conditioning protocol. These data appear to contradict previous

reports (Barot et al., 2009; Nicholas Chaaya et al., 2018; Michael S. Fanselow, 2010;

Hall et al., 2000, 2001a; Impey et al., 1998; Malkani & Rosen, 2000; Perez-Villalba

et al., 2008; Pignataro et al., 2013; Trogrlic et al., 2011; Y. M. Wilson & Murphy,

2009; Zelikowsky et al., 2014). Furthermore, these results are in direct contrast with

our previous investigations, demonstrating wide scale BLC activation following

UFC, and specific LaDL activation following CFC (N. Chaaya et al., 2019). The

particular proteins and cells explored here may explain this discrepancy. For

example, early research could not identify any changes in LA BDNF expression

following CFC (Hall et al., 2000). Further research found amygdala BDNF to

213

increase as a function of cue fear conditioning, but not as a function of contextual

fear conditioning (Rattiner, Davis, French, & Ressler, 2004). Similar to this,

investigations into microglia following fear conditioning appears to be limited, with

only recent research beginning to examine its functional role in psychiatric disorders

(Dwyer & Ross, 2016). For example, a recent study demonstrated the requirement

for microglia activation in chemically induced fear (Vollmer et al., 2016), with

numerous other investigations showing microglia to respond to stressful stimuli

(Calcia et al., 2016; Frank, Baratta, Sprunger, Watkins, & Maier, 2007; Aji Nair &

Robert H. Bonneau, 2006; Tynan et al., 2010). However, no research to our

knowledge has specifically examined the alterations in microglia number and

morphology that occur following fear conditioning protocols. The data here,

therefore, provides evidence that microglia number in LA remains stable following

contextual fear conditioning. Additionally, despite many documented cases of altered

activity and protein expression in LA following CFC (see recent review (Nicholas

Chaaya et al., 2018)), data reported here confirm that BDNF expression in LA is

unchanged.

Contrary to BDNF and IBA-1 data, pCREB expression in amygdala

following CFC has been examined. Early research on CRE-lac Z transgenic mice

investigated CREB (expressed as β-galactosidase in these mice) expression following

contextual, unpaired and cued fear conditioning (Impey et al., 1998). They noted an

increase in amygdala CREB as a function of both CFC and UFC (Impey et al., 1998).

More recent research conducted in two phases demonstrated that (1) oral

administration of DMP696, a corticotropin-releasing factor 1 antagonist significantly

reduced BLC pCREB expression and subsequent contextual fear expression, and (2)

bilateral microinjections of DMP696 in BLC significantly reduce fear-related

214

freezing to context (Hubbard, Nakashima, Lee, & Takahashi, 2007). Interestingly,

DMP696 had no effect on CeA pCREB expression, and microinjections of the

antagonist into the CeA did not affect fear-related freezing to context (Hubbard et al.,

2007). This suggests the deficits in fear-related freezing to context occurred due to

disturbances in contextual fear learning, as opposed to fear-expression. Numerous

other investigations have also confirmed a role for amygdala pCREB expression in

CFC (Hagewoud, Bultsma, Barf, Koolhaas, & Meerlo, 2011; Mamiya et al., 2009b;

Stanciu, Radulovic, & Spiess, 2001). Considering these studies, the results reported

here are unexpected, and may be attributed to methodological confounds. These are

explained in detail below.

6.5.2 Dorsal Hippocampus

The data reported here suggest that CFC, but not UFC, rely upon the DG

subregion of the hippocampus. Following CFC, BDNF expression was significantly

increased in DG, while IBA-1 number was significantly increased in DG and CA1 as

compared to the CO control. However, similar to above, no difference in pCREB

expression was present as a function of conditioning. Once again, this pCREB data is

contradictory to previous reports. On numerous occasions, hippocampal pCREB has

been found to be essential to CFC (Hagewoud et al., 2011; Impey et al., 1998; Kudo,

Qiao, Kanba, & Arita, 2004; Mamiya et al., 2009b; Stanciu et al., 2001; Trifilieff et

al., 2006). Nevertheless, the increase in BDNF expression is expected. Several

studies have found a functional role for hippocampal BDNF following CFC (R. M.

Barrientos et al., 2004; Bekinschtein et al., 2007; Hall et al., 2000; I. Y. C. Liu et al.,

2004; Lubin, Roth, & Sweatt, 2008; Mizuno et al., 2012; Takei et al., 2011).

However, the majority of this research found CA1 BDNF to be essential for CFC

(Bekinschtein et al., 2007; Hall et al., 2000; J. L. C. Lee, Everitt, & Thomas, 2004;

215

Lubin et al., 2008). Our recent review also concluded that the DH subregions most

involved in CFC appear to be the CA1 and CA3, but not the DG (Nicholas Chaaya et

al., 2018). Nevertheless, some evidence has found increases in BDNF expression in

the DG (R. M. Barrientos et al., 2004), with many other studies showing the

requirement for BDNF in the DH as a whole (I. Y. C. Liu et al., 2004; Mizuno et al.,

2012; Takei et al., 2011). The data here, therefore, suggest that DG BDNF is required

for contextual fear memory consolidation.

This work showed a significant increase in the number of IBA-1 in

hippocampal subregions CA1 and DG. Examination of microglia morphological

states within the DG revealed altered branching and processes. As compared to CO

controls, rats that underwent CFC had significantly shorter extensions, with fewer

ends and less complex extensions, suggesting a clear morphological response to this

fear conditioning protocol. An increase in the number of microglia (via the blood

system or proliferation) typically occurs due to central nervous system response to

harmful stimuli (Calcia et al., 2016; Kettenmann et al., 2011). Additionally,

microglia alter morphology in response to harmful stimuli (Calcia et al., 2016;

Dwyer & Ross, 2016; Kettenmann et al., 2011). During resting phase, without

harmful stimuli, microglia are ramified; when responding to harmful stimuli,

microglia become amoeboid (Calcia et al., 2016). When ramified, or “resting”,

microglia have long, thin extensions with many processes that search for signals of

insult (Dwyer & Ross, 2016; Kettenmann et al., 2011; Walker et al., 2014). When

amoeboid, microglia extensions retract, cell bodies enlarge, and they respond to

insult by release of proinflammatory and immunoregulatory factors and compounds

(Dwyer & Ross, 2016; Kettenmann et al., 2011; Walker et al., 2014). One such factor

that can be released by microglia in response to injury is BDNF (Ferrini & De

216

Koninck, 2013; Pósfai et al., 2018). Microglia in the DG following CFC in the

current study were found to represent the typical morphological state of those

responding to insult. Not only did they represent the morphological structure of an

amoeboid microglia, they may have been responsible for the increase in BDNF.

While further research is required to confirm the direct relationship of microglia and

BDNF in fear, this is the first study to examine and report alterations in the number

and morphological state of microglia following CFC.

Both BDNF and IBA-1 number in DG were significantly higher in the CFC

group as compared to the UFC group. The UFC group has typically been used as a

control to cued fear conditioning groups (H. C. Bergstrom et al., 2012; Hadley C

Bergstrom et al., 2011; Majak & Pitkänen, 2003; McKernan & Shinnick-Gallagher,

1997; Radley et al., 2006; Michael T. Rogan et al., 1997). This is because the UFC

do not have overlapping cue and foot-shock, and therefore no associative memory is

formed (Joseph LeDoux, 2003; Romanski et al., 1993). Nevertheless, during CFC,

the ‘context’ becomes paired with the foot-shock (Calandreau et al., 2005;

Calandreau et al., 2006; Desmedt et al., 1998; Russel G Phillips & LeDoux, 1994;

Trifilieff et al., 2007; Trifilieff et al., 2006). While this context is altered during the

UFC protocol, it still gets paired with the foot-shock. We previously showed the

UFC protocol to result in wide-spread BLC activation, whereas the CFC protocol

only resulted in specific LaDL activation (N. Chaaya et al., 2019). Contrastingly,

despite equivalent levels of fear to context, the UFC protocol here resulted in

significantly less BDNF and IBA-1 number in DG as compared to the CFC protocol.

We hypothesise that this occurred as a result of the auditory tones altering the

contextual fear memories. Trace fear conditioning is a behavioural paradigm similar

to the UFC protocol; with the only difference being that the auditory tones are

217

presented before a foot-shock (the period in between the tone and foot-shock

represent the trace period) in a consistent and ordered manner (e.g. 10 s before each

foot-shock) (Rogers, Hunsaker, & Kesner, 2006). While the DH is well-documented

to be essential to CFC (reviewed in (Nicholas Chaaya et al., 2018)), some research

has shown that it may be less involved in trace fear conditioning (Rogers et al.,

2006). Nevertheless, the majority of research has reported the DH to be essential to

trace fear conditioning (Pierson et al., 2015; Jennifer J. Quinn et al., 2005; Reichelt,

Maniam, Westbrook, & Morris, 2015). This suggests the difference between the UFC

protocol and trace fear conditioning protocol may drastically change the need for

DH. Contrary to BDNF and IBA-1 number data, traced microglia in DG did appear

to have some morphological alterations (significant decrease in length of extensions

and complexity of extensions as compared to CO controls) following UFC. Given

that both protocols successfully produce contextual fear memories, further research is

required to fully delineate DH involvement following CFC and UFC.

6.5.3 Technical Considerations

Evaluation of BDNF and IBA-1, but not pCREB revealed a functional role

for DG following CFC. The stability of pCREB expression in hippocampus, and also

amygdala, was in direct contrast to many previous investigations (Hagewoud et al.,

2011; Impey et al., 1998; Kudo et al., 2004; Mamiya et al., 2009b; Stanciu et al.,

2001; Trifilieff et al., 2006). Methodological considerations may explain these

results. First, no previous research identifying a role for pCREB found a difference

90 minutes post-learning. Various proteins and IEG’s have a peak expression time

following learning (Ivashkina et al., 2016; Lonergan et al., 2010; Morgan & Curran,

1991; Ramírez-Amaya et al., 2005; Schafe et al., 2000). Behavioural differences

cannot be identified if animals are sacrificed before or after this time-point (Schafe et

218

al., 2000). An alternate explanation for the stability of pCREB expression relates to

the quality of labelling. While labelling in previous investigations allow for clear

differentiation between neurons (see (Kudo et al., 2004; Mamiya et al., 2009b;

Stanciu et al., 2001; Trifilieff et al., 2006)), there appear to be more total pCREB

expressing neurons in the current study, which overlap. This made it difficult to

accurately quantify the number of pCREB positive neurons. For this reason, the

pCREB-specific results are reported cautiously.

The further caveat of the current project is the small differences in actual

number of BDNF (mean difference of 16.17 between CFC and CO control) and IBA-

1 (mean difference of 7.167) number. The small standard errors make these

differences statistically significant. Nevertheless, the change in number may limit the

clinical significance of this study. Targeted pharmacological inhibitors can produce

large-scale alterations in protein expression (see pCREB study (Hubbard et al.,

2007)). While large-scale alterations may be beneficial to the treatment of fear-

related disorders, it may have numerous confounding behavioural consequences.

While this particular caveat exists in all neuroscientific research, it appears to be

exaggerated by the atypically small number in differences reported here.

Nevertheless, identification of these differences may be essential to understanding

how the healthy brain can become pathological. For example, the small increase in

BDNF and IBA-1 number lead to more in-depth analyses into microglia morphology,

which were found to be have large and clear alterations.

6.5.4 Conclusion

The current study investigated how two differing contextual fear memories

are represented in the rat brain. While fear to context was relatively constant, we

report differences in BDNF and IBA-1 number between these two conditioned

219

groups. Standard CFC leads to an increase in BDNF expression, IBA-1 number,

while UFC did not. Furthermore, we provide some of the first data showing

microglia morphology to become altered as a function of CFC and UFC.

Interestingly, while BDNF expression and IBA-1 number increased only in following

CFC, morphological differences were identified in both the CFC and UFC as

compared to the CO control. While the change in BDNF and IBA-1 number may

have limited clinical significance, the alteration in microglia morphology may be

clinically significant. When active, or amoeboid, microglia can alter neuronal activity

via the BDNF-TrkB pathway (Ferrini & De Koninck, 2013; Pósfai et al., 2018), and

therefore targeted treatment of microglia may affect only the necessary neurons

responding to fear-inducing stimuli. This suggests that targeted drug treatments

aimed at inhibiting microglia activity (Cheng et al., 2015; Greter & Merad, 2013)

may provide a new therapeutic tool for sufferers of fear-based disorders. Further

research is required to investigate how reducing microglia activity influences fear

memory consolidation and maintenance.

220

Axonal nonsegregation of the Vesicular Glutamate

Transporter VGLUT3 within serotonergic projections in the mouse

forebrain.

This chapter comprises the following published article: Belmer, A., Beecher, K., Jacques, A., Patkar, O. L., Sicherre, F., Bartlett, S. E. Axonal nonsegregation of the Vesicular Glutamate Transporter VGLUT3 within serotonergic projections in the mouse forebrain. Frontiers in Cellular Neuroscience. Published 10th May, 2019. https://doi.org/10.3389/fncel.2019.00193

Chapters 7 and 8 address the objectives listed in aim 3. The co-release of

glutamate and serotonin are central to the encoding of reward and anxiety type

behaviours, therefore the mapping of colocalized 5-HT and Vglut3 in relation to

brain region and to pMAPK expressing neurons may reveal potential therapeutic

targets.

221








In the case of chapter 7: Axonal nonsegregation of the Vesicular Glutamate Transporter VGLUT3 within serotonergic projections in the mouse forebrain. Publication status: Published

Contributor Statement of contribution*Arnauld Belmer Performed IHC experiments, imaging, design of experiments, data analysis,

drafted and edited the manuscript

Kate Beecher Assisted with imaging, design of experiments, data analysis, drafted and edited the manuscript

Angela Jacques Assisted with data collection, figure formatting and editing the manuscript.

Omkar Patkar Assisted with data collection, figure formatting and editing the manuscript.

Florian Sicherre Assisted with data collection, figure formatting and editing the manuscript.

Selena Bartlett Assisted with the design of experiments and editing the manuscript

QUT Verified

Signature

222

7.1 Abstract

A subpopulation of raphe 5-HT neurons expresses the vesicular glutamate

transporter VGLUT3 with the co-release of glutamate and serotonin proposed to play

a pivotal role in encoding reward- and anxiety-related behaviors. Serotonin axons are

identifiable by immunolabelling of either serotonin (5-HT) or the plasma membrane

5-HT transporter (SERT), with SERT labeling demonstrated to be only partially

overlapping with 5-HT staining. Studies investigating the colocalization or

segregation of VGLUT3 within SERT or 5-HT immunolabeled boutons have led to

inconsistent results. Therefore, we combined immunohistochemistry, high resolution

confocal imaging and 3D-reconstruction techniques to map and quantify the

distribution of VGLUT3 immunoreactive boutons within 5-HT vs SERT- positive

axons in various regions of the mouse forebrain, including the prefrontal cortex,

nucleus accumbens core and shell, bed nucleus of the stria terminalis, dorsal

striatum, lateral septum, basolateral and central amygdala and hippocampus. Our

results demonstrate that about 90% of 5-HT boutons are colocalized with SERT in

almost all the brain regions studied, which therefore reveals that VGLUT3 and SERT

do not segregate. However, in the posterior part of the NAC shell, we confirmed the

presence of a subtype of 5-HT immunoreactive axons that lack the SERT.

Interestingly, about 90% of the 5-HT/VGLUT3 boutons were labelled for the SERT

in this region, suggesting that VGLUT3 is preferentially located in SERT

immunoreactive 5-HT boutons. This work demonstrate that VGLUT3 and SERT

cannot be used as specific markers to classify the different subtypes of 5-HT axons.

223

7.2 Introduction

Intensive efforts have long been made to understand the complexity of the

serotonin (5-Hydroxytryptamine, 5-HT) system and to identify specific markers for

serotonin neuron diversity. Although it is becoming evident that raphe serotonin

neurons are morphologically, functionally and molecularly heterogeneous (Calizo et

al., 2011; Fernandez et al., 2016; Gaspar and Lillesaar, 2012; Kiyasova et al., 2011,

2013), the diversity of serotonergic axonal projections to the forebrain is not

completely understood.

Pioneer electron or light microscopy and anterograde tracing studies have

revealed the existence of 5-HT axon terminals with different sizes, shapes, contents

of their small vesicles, and the presence or absence of dense-core vesicles (for review

see(Descarries et al., 2010)). In rats, two types of axons were reported, with axons

originating from the dorsal raphe showing fine beaded or fusiform varicosities

separated by smooth axon segments of variable length (type D), while axons

originating from the median raphe displayed large spherical varicosities with fine and

smooth inter-varicosity segments (type M) (Kosofsky and Molliver, 1987). In

primates, two types of axons were also described, with sparse, small, ovoid or large,

spheroidal varicosities (Hornung et al., 1990). However, it is likely that this axonal

morphology classification cannot longer be considered as valid criteria for

distinguishing the cellular origin or the chemical identity of the 5-HT neurons.

Indeed a chemically defined 5-HT neuron can send several types of axonal

projections with different morphologies to different brain regions (Gagnon and

Parent, 2014). Hence, research has been rather devoted to studying the molecular or

physiological diversity of 5-HT neurons, identifying various 5-HT neuronal subtypes

that differentially express the 5-HT1A autoreceptor (Bonnavion et al., 2010;

224

Fernandez et al., 2016; Kirby et al., 2003; Kiyasova et al., 2013; Sotelo et al., 1990),

substance P/neurokinin receptor 1 (NK1)(Lacoste et al., 2006), galanin and its

receptor (Larm et al., 2003; Xu and Hökfelt, 1997), neuronal nitric oxide synthase

(nNOS) (Xu and Hökfelt, 1997), gamma-aminobutyric acid (GABA)-synthesizing

enzyme glutamic acid decarboxylase (GAD) (Fu et al., 2010), alpha7 nicotinic

receptor (Aznar et al., 2005), MET receptor tyrosine kinase (Kast et al., 2017) or

display different pharmacological and electrophysiological properties (Calizo et al.,

2011; Hajós et al., 2007; Kirby et al., 2003). This heterogeneity appears to be target-

specific (Fernandez et al., 2016; Prouty et al., 2017) and could therefore be used to

establish a specific anatomy/function cartography of raphe serotonin sub-systems

(Ren et al., 2018).

In addition, a subpopulation of dorsal and median raphe 5-HT neurons was

found to co-express transcripts of the vesicular glutamate transporter type 3

(VGLUT3) (Gras et al., 2002; Hioki et al., 2010), suggesting that 5-HT and

glutamate could be stored in the same vesicles and co-released. VGLUT3 protein

was also reported to be located in the some 5-HT immunoreactive axonal varicosities

in the forebrain (Mintz and Scott, 2006; Schäfer et al., 2002), including the granular

cell layer of the olfactory bulb, cerebral cortex, central amygdaloid nuclei,

hippocampal CA3 field, dorsolateral septum and supra-ependymal plexus of the third

ventricle (Shutoh et al., 2008). A classification of two serotonergic axons subtypes

depending on the presence or absence of VGLUT3 was therefore proposed (Shutoh

et al., 2008). It is likely that the co-expression of VGLUT3 and the vesicular

monoamine transporter 2 (Vmat2) in serotonin terminals (Schäfer et al., 2002)

synergizes the filling of 5-HT and glutamate in the same synaptic vesicles (Amilhon

et al., 2010), with 5-HT/glutamate cotransmission proposed to play a pivotal role in

225

the control of reward- and emotion-related neural circuitry (Liu et al., 2014;

Sengupta et al., 2017) and their plasticity/adaptability during development or

pathological processes (Gagnon and Parent, 2014).

However, some discrepancies have emerged from the aforementioned studies,

regarding the total or sparse colocalization of the serotonin transporter SERT in 5-

HT axon varicosities. While Gagnon and Parent (2014) observed that all 5-HT axon

varicosities contain the SERT in rats, two studies have reported a very sparse

colocalization of SERT and 5-HT immunolabeling in mice, with VGLUT3 and

SERT mostly segregrated within 5-HT varicosities (Amilhon et al., 2010; Voisin et

al., 2016), especially in the prefrontal cortex, hippocampus, dorsal striatum and

lateral septum. This data suggests that different subtypes of 5-HT axonal varicosities

(SERT-/VGLUT3+ or SERT+/VGLUT3-) may coexist in the mouse forebrain.

In the present study, we therefore investigated the immunohistological distribution of

SERT and VGLUT3 within 5-HT axon varicosities in various regions of the mouse

forebrain, including the prefrontal cortex, nucleus accumbens core and shell, bed

nucleus of the stria terminalis, dorsal striatum, lateral septum, basolateral and central

amygdala and hippocampus. For this, we imaged ±1.55x108 um3 of tissue and 3D-

reconstructed ±106 5-HT varicosities to determine the volumetric density and the

proportion of varicosities co-labelled with SERT and/or VGLUT3, in each brain

region. We found that the great majority (≈ 90%) of 5-HT varicosities express the

SERT in every brain regions analyzed except the posterior shell of the nucleus

accumbens (50%). We herein report that VGLUT3 is preferentially located in SERT+

5-HT varicosities. Our results demonstrate that VGLUT3 and SERT do not

particularly segregate in 5-HT axonal varicosities of the mouse forebrain.

226

7.3 Materials and Methods

7.3.1 Animals

Six 8-10 week-old C57Bl6 mice (3 males, 3 females) were housed in standard

ventilated cages in climate-controlled rooms. Food, water, and environmental

enrichment were available ad libitum. This study was carried out in accordance with

the recommendations of National Health and Medical Research Council (NHMRC)

guidelines to promote the well-being of animals used for scientific purposes and the

Australian code for the care and use of animals for scientific purposes. The protocol

was approved by the Queensland University of Technology Animal Ethics

Committee and the University of Queensland Animal Ethics Committee.

7.3.2 Histology

Mice were transcardially perfused with 4% paraformaldehyde (PFA) prior to

decapitation. Brains were harvested and post-fixed overnight at 4°C. Forty-micron

thick coronal vibratome sections were collected and incubated overnight in blocking

solution (2% normal goat serum, 0.3% Triton and 0.05% Tween 20 in 0.1M

phosphate-buffer saline (PBS)).


Sections containing the pre-limbic cortex (Bregma +2.46 ± 0.3 mm), the

NAC (Bregma +1.42 ± 0.2 mm), the posterior NAC, the dorsal striatum and the

lateral septum (Bregma +1.00 ± 0.2 mm), the BNST (Bregma -0.22 ± 0.3 mm), the

hippocampus (Bregma -1-70 ± 0.3 mm) or the amygdala (Bregma -1.40 ± 0.2 mm)

were incubated with primary antibodies diluted in the blocking solution: rat anti-5-

HT (Millipore #MAB352, 1:100) for 48 hours at room temperature followed by

rabbit anti-SERT (Millipore #PC177L, 1:1000) and guinea-pig anti-VGLUT3

227

(Synaptic System #135204, 1:500) overnight at 4 degrees. After three washes in the

blocking solution, the slices were incubated for 4 hours at room temperature with

secondary antibodies diluted in the blocking solution: goat anti-rabbit-Alexa 488,

goat anti-guinea pig-Alexa 647 (Thermofisher Scientific, #A11034 and #A21450,

1:500) and goat anti-rat biotinylated (Jackson Laboratory # 112-065-003, 1:200).

After three washes in the blocking solution, slices were incubated for 30 min in

Streptavidin-Cy3 (Thermofisher Scientific #438315, 1:1000), washed 3 times in

PBS, and mounted in Prolong Gold antifade mountant (Thermofisher Scientific,

#P36934).

7.3.4 Imaging and Analysis

Sections (3 sections per animal, n = 6 animals, 18 sections/brain region) were

imaged on an Olympus FV3000 using a 60X oil-immersion objective (NA 1.35) with

a 2.5 x zoom and a Z-axis step of 0.3 µm, using sequential scanning. Mosaics of the

regions of interest were acquired as depicted in A of Figures 1 to 7, in OIR file

format. The 5-HT immunoreactive boutons were reconstructed in 3D using the

surface rendering function with Imaris 9.2.1 (Bitplane), as previously described

(Belmer et al., 2016; Tarren et al., 2017). All the images were processed in batch

using the same surface thresholding parameters. Mean fluorescence intensities of

SERT or VGLUT3 labeling within 5-HT boutons and image volumes were obtained

from the surface statistics in Imaris. Since the level of background of confocal

images can reach as much as 30% of maximum image intensity (Landmann and

Marbet, 2004), we use this threshold as a criteria to define the 5-HT boutons

colocalized and non-colocalized with SERT and/or VGLUT3 (i.e mean intensity

>30% or <30% of the maximum intensity, respectively). For each brain region, the

frequency distribution of the colocalized and non-colocalized boutons within each

228

brain region was analyzed using Excel 365, averaged for each animal, and plotted in

Graphpad Prism 7.0 (Graph Pad Software Co., San Diego, CA, United States) as

replicates (n = 6). 3D representation of the proportion of VGLUT3 immunoreactive

5-HT boutons within each brain region was generated using the online Scalable

Brain Atlas (Bakker et al., 2015) with a custom color-coded scale (Figure 9).

7.3.5 Statistics

Statistical analyses were carried out using GraphPad Prism 7.0. The

proportion of 5-HT or 5-HT/VGLUT3 varicosities immunopositive for SERT were

compared to SERT immunonegative boutons within each brain region using a two-

tailed unpaired t-test and across all studied regions using a one-way ANOVA. The

densities and relative densities of 5-HT/VGLUT3 boutons within all the analyzed

brain regions were compared using one-way ANOVA with Sidak correction for

multiple comparison. A p-value < 0.05 was considered significant, with all values

expressed as the mean ± SEM.

7.4 Results

Previous studies have suggested that VGLUT3 always segregates with SERT

within the varicosities of the 5-HT axons projecting to the prelimbic region of the

prefrontal cortex (Amilhon et al., 2010). Therefore, we first examined the

distribution of VGLUT3 and SERT immunoreactive 5-HT varicosities within the

different layers of the prelimbic cortex (PrL), at bregma +2.46 mm (Fig. 1A). Co-

labeling of the SERT and 5-HT revealed that most of the varicosities reconstructed

were co-labeled for the two markers (Fig. 1B: a-i). Labeling of VGLUT3 showed the

typical scattered punctate fluorescence (Fig. 1B: j) and by combining the 3D-

reconstruction and the masking functions of Imaris, we isolated the VGLUT3

229

labeling that was only contained in 3D-reconstructed 5-HT varicosities (Fig. 1B: k).

We observed that almost all the 5-HT/VGLUT3 boutons were co-labeled with SERT

(arrow heads, Fig. 1B: c, f, i, l). Indeed, the absolute quantifications of the proportion

of 5-HT varicosities that were co-labeled with SERT confirmed that 85 % and 91 %

of 5-HT varicosities were co-labeled with SERT in the layers 4-5 and 1-3,

respectively (Fig 1C, a, b; ****: p<0.0001). Similar proportions of 5-HT/VGLUT3

boutons (82 % and 91 %) were also co-labeled with SERT (Fig 1C, c, d). These

results suggest that only a small proportion of 5-HT boutons (9-18 %) do not express

detectable levels of SERT in the PrL and, that VGLUT3 and SERT do not

predominantly segregate within the 5-HT axonal varicosities in this brain region.

230

Figure 7-1 Distribution of VGLUT3+ boutons within 5-HT+ and SERT+ axons in the prelimbic cortex.

A. Schematic drawing showing the location of the acquired micrographs. Prelimbic (PrL) cortex mosaic images of layer I to V were acquired at bregma +2.46 ± 0.2 mm (red vertical line) in the dorsal half of the medial cortex (yellow/red square). B. Micrograph showing the distribution of the SERT (green, a-c), 5-HT (red, d-f), SERT (green) + 5-HT (red) (g-i), and VGLUT3 (magenta, j-l). Left panel shows lower magnification. Scale bar: 50 µm. Middle and right panels show higher magnification of the white dashed box in the left panel. Scale bar: 2 µm. The right panel shows the VGLUT3 boutons co-labeled with SERT (c), 5-HT (f) and SERT+5-HT (i). VGLUT3 puncta located within 5-HT varicosities were isolated using Imaris (l). Arrow heads show 5-HT varicosities collocated with SERT and VGLUT3 (c, f, i and l). C. (a, b) Quantification of the proportion of 5-HT varicosities not co-labeled (SERT-) and co-labeled (SERT+) with SERT in the upper (a) or deeper (b) layers of the PrL, showing a great majority of 5-HT varicosities co-labeled with SERT. (c, d) Quantification of the proportion of 5-HT+/VGLUT3+ not co-labeled (SERT-) and co-labeled (SERT+) with SERT in the upper (a) or deeper (b) layers of the PrL, showing a great majority of 5-HT/VGLUT3 varicosities co-labeled with SERT (t test, ****: p<0.0001).

The co-release of 5-HT and glutamate by 5-HT neurons from the dorsal raphe

has been proposed to play an important role in the regulation of the

neurotransmission in the ventral striatum and the modulation of reward-related

behaviors (Liu et al., 2014). Hence, we next investigated the distribution of SERT

and VGLUT3 immunolabeled varicosities within the NAc core and shell, at bregma

231

+1.42 (Fig. 2A). Both in the core (94 %) and the shell (95 %) regions of the NAc, the

great majority of 5-HT boutons were also immunoreactive for SERT (Fig. 2B and

2C: a-i; 2D: a-b, **: p<0.01; ****: p<0.0001). Hence, a great proportion of 5-

HT/VGLUT3 boutons were co-labeled with the SERT, both in the core (92 %) and

the shell (89 %) (arrow heads, Fig. 2B and 2C: c-l; Fig. 2D: c-d, ****: p<0.0001).

These results support a high degree of overlap between 5-HT and SERT

immunoreactive axons in the core and the shell of the rostral NAc (Brown and

Molliver, 2000), and further highlight the absence of any particular segregation

between SERT and VGLUT3 within 5-HT varicosities.

Figure 7-2 Distribution of VGLUT3+ boutons within 5-HT+ and SERT+ axons in the nucleus accumbens.

A. Schematic drawing showing the location of the acquired micrographs. Nucleus accumbens (NAC) mosaic images were acquired at bregma +1.42 ± 0.2 mm (red vertical line) in the dorsomedial shell and the core (ventrolateral to the anterior commissure) of the nucleus accumbens (green/red areas). B. Micrograph showing the distribution of the SERT (green, a-c), 5-HT (red, d-f), SERT (green) + 5-HT (red) (g-i), and VGLUT3 (magenta, j-i). Left panel shows lower magnification. Scale bar: 50 µm. Middle and right panels show higher magnification of the white dashed box in the left panel. Scale bar: 2 µm. The right panel shows the VGLUT3 boutons co-labeled with SERT (c), 5-HT (f) and

232

SERT+5-HT (i). VGLUT3 puncta located within 5-HT varicosities were isolated using Imaris (l). Arrow heads show 5-HT varicosities collocated with SERT and VGLUT3 (c, f, i and l). C. (a, b) Quantification of the proportion of 5-HT varicosities not co-labeled (SERT-) and co-labeled (SERT+) with SERT in the NAC, showing a great majority of 5-HT varicosities co-labeled with SERT. (c, d) Quantification of the proportion of 5-HT+/VGLUT3+ not co-labeled (SERT-) and co-labeled (SERT+) with SERT in the NAC, showing a great majority of 5-HT/VGLUT3 varicosities co-labeled with SERT (t test, **: p<0.01, ****: p<0.0001).

Brown and Molliver also observed that a subset of 5-HT axons lack SERT in

the caudal part of the rat NAc shell(Brown and Molliver, 2000). Therefore, we

investigated the distribution of SERT and VGLUT3 within the 5-HT varicosities in

the posterior NAc shell, at bregma +1.00 mm, at a similar level to the

aforementioned study in rat (i.e “septal pole” or “cone region”) (Fig. 3A).

Interestingly, we found the 5-HT varicosities either co-labeled (arrow heads) or not

labelled (or only weakly/partially labeled, arrows) for the SERT in the posterior NAc

shell (Fig. 3B: a-c, e-g) and both SERT+ and SERT- serotonergic varicosities were

co-labeled for VGLUT3 (Fig. 3B: i-l). The quantification revealed that half of the 5-

HT varicosities do not express the SERT in the posterior shell of the mouse NAc

(Fig. 3C: a) as previously described in rats. Although VGLUT3 was observed in

some SERT- varicosities, this subtype of serotonergic boutons only represents a

minority (11 %), as the great majority (89 %, Fig. 3C: b, ****: p<0.0001) of the 5-

HT+/VGLUT3+ boutons were also co-labeled for SERT. These results further

suggest that VGLUT3 are rather colocalized than segregated with the SERT in 5-HT

varicosities.

233

Figure 7-3 Distribution of VGLUT3+ boutons within 5-HT+ and SERT+ axons in the posterior shell of the nucleus accumbens.

A. Schematic drawing showing the location of the acquired micrographs. The posterior (caudal) shell of the nucleus accumbens (NAC) (green/red area) mosaic images were acquired at bregma +1 ± 0.2 mm (red vertical line). B. Micrograph showing the distribution of the SERT (green, a, e, i), 5-HT (red, b, f, j), SERT (green) + 5-HT (red) (c, g, k), and VGLUT3 (magenta, d, h, l). Top panel shows lower magnification. Scale bar: 50 µm. Middle and bottom panels show higher magnification of the white dashed box in the left panel. Scale bar: 2 µm. The bottom panel shows the VGLUT3 boutons co-labeled with SERT (i), 5-HT (j) and SERT+5-HT (k). VGLUT3 puncta located within 5-HT varicosities were isolated using Imaris (l). Arrow heads show 5-HT varicosities collocated with SERT and VGLUT3 (i-l). C. (a) Quantification of the proportion of 5-HT varicosities not co-labeled (SERT-) and co-labeled (SERT+) with SERT in the posterior shell of the NAC, showing a great majority of 5-HT varicosities co-labeled with SERT. (b) Quantification of the proportion of 5-HT+/VGLUT3+ not co-labeled (SERT-) and co-labeled (SERT+) with SERT in the posterior shell of the NAC, showing a great majority of 5-HT/VGLUT3 varicosities co-labeled with SERT (t test, ****: p<0.0001).

The segregation of SERT and VGLUT3 was also reported in the dorsal

striatum and the lateral septum(Voisin et al., 2016). Hence, we investigated their

distribution at bregma +1.00 mm (Fig. 4A). Again, the great majority of the 5-HT

varicosities express the SERT in both the dorsal striatum (92 %, Fig. 4B: a-i; Fig.

4D: a) and lateral septum (75 %, Fig. 4C: a-i; Fig. 4D: b; ***: p<0.001; ****:

p<0.0001). Therefore, no axonal segregation could be observed, indeed more than 80

% of the VGLUT3 immunoreactive 5-HT boutons were also co-labeled for the

234

SERT, in both the striatum and lateral septum (Fig. 4B and 4C: a-i, arrow heads;

Fig. 4D: c-d; ***: p<0.001; ****: p<0.0001). These results further evidence that

SERT and VGLUT3 cannot be used as markers to classify different 5-HTergic

axonal subtypes.

Figure 7-4 Distribution of VGLUT3+ boutons within 5-HT+ and SERT+ axons in the striatum and lateral septum.

A. Schematic drawing showing the location of the acquired micrographs. Mediolateral dorsal striatum and lateral septum (LS) mosaic images were acquired at bregma +1 ± 0.2 mm (red vertical line) (green/red circles). B. Micrograph showing the distribution of the SERT (green, a-c), 5-HT (red, d-f), SERT (green) + 5-HT (red) (g-i), and VGLUT3 (magenta, j-i) in the striatum. Left panel shows lower magnification. Scale bar: 50 µm. Middle and right panels show higher magnification of the white dashed box in the left panel. Scale bar: 2 µm. The right panel shows the VGLUT3 boutons co-labeled with SERT (c), 5-HT (f) and SERT+5-HT (i). VGLUT3 puncta located within 5-HT varicosities were isolated using Imaris (l). Arrow heads show 5-HT varicosities collocated with SERT and VGLUT3 (c, f, i and l). C. Micrograph showing the distribution of the SERT (green, a-c), 5-HT (red, d-f), SERT (green) + 5-HT (red) (g-i), and VGLUT3 (magenta, j-l) in the LS. Left panel shows lower magnification. Scale bar: 50µm. Middle and right panels show higher magnification of the white dashed box in the left panel. Scale bar: 2µm. The right panel shows the VGLUT3 boutons co-labeled with SERT (c), 5-HT (f) and SERT+5-HT (i). VGLUT3 puncta located within 5-HT varicosities (l). Arrow heads show 5-HT varicosities collocated with SERT and VGLUT3 (c, f, i and l). D. (a, b) Quantification of the proportion of 5-HT varicosities not co-labeled (SERT-) and co-labeled (SERT+) with SERT in the striatum (a) and LS (b), showing a great majority of 5-HT varicosities co-labeled with SERT. (c, d) Quantification of the proportion of 5-HT+/VGLUT3+ not co-labeled (SERT-) and co-labeled (SERT+) with SERT in the striatum (c) and LS (d), showing a great majority of 5-HT/VGLUT3 varicosities co-labeled with SERT (t test, ****: p<0.0001; ***:p<0.001).

235

DR 5-HT neurons that send VGLUT3 immunoreactive axons to the NAc also

send collaterals to different brain regions, including the bed nucleus of the stria

terminalis (BNST) and central (CeA) and basolateral (BLA) amygdala. We therefore

investigated the distribution of SERT and VGLUT3 within 5-HT varicosities in those

brain regions (Fig. 5 and 6). In the anterior BNST at bregma -0.22 mm (Fig. 5A), we

again observed a high degree of overlapping between SERT and 5-HT

immunoreactivity (82 %) (Fig. 5B: a-i and 5C: a; ****: p<0.0001), with a high

proportion (89 %) of 5-HT/VGLUT3 varicosities that were co-labeled for SERT

(Fig. 5B: a-i, arrow heads and 5C: b; ****: p<0.0001).

Figure 7-5 Distribution of VGLUT3+ boutons within 5-HT+ and SERT+ axons in the bed nucleus of the stria terminalis.

A. Schematic drawing showing the location of the acquired micrographs. The bed nucleus of the stria terminalis (BNST) mosaic images were acquired at bregma -0.22 ± 0.3 mm (red vertical line) in the medial posteromedial BNST (BSTMPM) and posterointermediate part of the BNST (BSTMPI) (green/red area). B. Micrograph showing the distribution of the SERT (green, a-c), 5-HT (red, d-f),

236

SERT (green) + 5-HT (red) (g-i), and VGLUT3 (magenta, j-l). Left panel shows lower magnification. Scale bar: 50 µm. Middle and right panels show higher magnification of the white dashed box in the left panel. Scale bar: 2 µm. The right panel shows the VGLUT3 boutons co-labeled with SERT (c), 5-HT (f) and SERT+5-HT (i). VGLUT3 puncta located within 5-HT varicosities were isolated using Imaris (l). Arrow heads show 5-HT varicosities collocated with SERT and VGLUT3 (c, f, i and l). C. (a) Quantification of the proportion of 5-HT varicosities not co-labeled (SERT-) and co-labeled (SERT+) with SERT in the BNST, showing a great majority of 5-HT varicosities co-labeled with SERT. (b) Quantification of the proportion of 5-HT+/VGLUT3+ not co-labeled (SERT-) and co-labeled (SERT+) with SERT in the BNST, showing a great majority of 5-HT/VGLUT3 varicosities co-labeled with SERT (t test, ****: p<0.0001).

Similarly, in the amygdala at bregma -1.70 (Fig. 6A), the great majority of 5-

HT boutons were co-labeled for SERT in both the BLA (87 %) and CeA (79 %) (Fig.

6B: a-i; Fig. 6C: a-b, ***: p<0.001; ****: p<0.0001). Consequently, a great

proportion of 5-HT/VGLUT3 boutons were co-labeled with the SERT within the

BLA (89 %) and CeA (73 %) (arrow heads, Fig. 6B: g-l; Fig. 6C: c-d; ****:

p<0.0001; ***: p<0.001).

Figure 7-6 Distribution of VGLUT3+ boutons within 5-HT+ and SERT+ axons in the basolateral amygdala and central nucleus of the amygdala.

237

A. Schematic drawing showing the location of the acquired micrographs. The basolateral amygdala (BLA) and central nucleus of the amygdala (CeA) mosaic images were acquired at bregma -1.70 ± 0.2 mm (red vertical line) in the anterior part of the basolateral amygdala and the lateral CeA (green/red areas). B. Micrograph showing the distribution of the SERT (green, a-c), 5-HT (red, d-f), SERT (green) + 5-HT (red) (g-i), and VGLUT3 (magenta, j-l) in the CeA. Left panel shows lower magnification. Scale bar: 150 µm. Middle and right panels show higher magnification of the white dashed areas (circle = CeA) in the left panel. Scale bar: 2 µm. The right panel shows the VGLUT3 boutons co-labeled with SERT (c), 5-HT (f) and SERT+5-HT (i) in the CeA. VGLUT3 puncta located within 5-HT varicosities were isolated using Imaris (l). Arrow heads show 5-HT varicosities collocated with SERT and VGLUT3 (c, f, i and l) in the CeA. C. (a-b) Quantification of the proportion of 5-HT varicosities not co-labeled (SERT-) and co-labeled (SERT+) with SERT in the BLA (a) and CeA (b), showing a great majority of 5-HT varicosities co-labeled with SERT. (c-d) Quantification of the proportion of 5-HT+/VGLUT3+ not co-labeled (SERT-) and co-labeled (SERT+) with SERT in the BLA (c) and CeA (d), showing a great majority of 5-HT/VG varicosities co-labeled with SERT (t test, ****: p<0.0001; ***: p<0.001).

In the hippocampus, VGLUT3 was shown to modulate 5-HTergic tone, and

to stimulate VMAT2-dependent accumulation of 5-HT in synaptic vesicles (Amilhon

et al., 2010), which further suggests that VGLUT3/5-HT synaptic cross-talk may

play an important role in hippocampal-mediated behaviors such as anxiety and

depression. Since these behaviors are also dependent upon SERT activity/blockade

by serotonergic antidepressant, whether VGLUT3 and SERT segregate or co-localize

within 5-HT varicosities in the hippocampus is question of great interest and could

further help in developing improved therapeutics for the treatment of anxiety- or

depression-related disorders. Therefore, we rigorously investigated the distribution of

SERT and VGLUT3 in 5-HT axonal varicosities within the different subregions of

the hippocampus, at bregma -1.70 ± 0.3 mm (Fig. 7A). Most of 5-HT boutons were

also immunoreactive for SERT in all four regions of the hippocampus, CA1 (90 %),

CA2 (85 %), CA3 (84 %) and DG (86 %) (Fig. 7B: g-i; 7C: a-b, ****: p<0.0001). A

similar proportion of 5-HT/VGLUT3 boutons were co-labeled with SERT in the

CA1 (87 %), CA2 (87 %), CA3 (82 %) and DG (78 %) (arrow heads, Fig 7B: c, f, i,

l; 7C: c-d; ****: p<0.0001; **: p<0.01).

238

Figure 7-7 Distribution of VGLUT3+ boutons within 5-HT+ and SERT+ axons in the CA1, CA2, CA3, dentate gyrus of the hippocampus.

A. Schematic drawing showing the location of the acquired micrographs. The hippocampus (HIP) mosaic images were acquired at bregma -1.70 ± 0.3 mm (red vertical line) in the dorsal hippocampus (green/red area). B. Micrograph showing the distribution of the SERT (green, a-c), 5-HT (red, d-f), SERT (green) + 5-HT (red) (g-i), and VGLUT3 (magenta, j-l) in the CA1. Left panel shows lower magnification. Scale bar: 150 µm. Middle and right panels show higher magnification of the CA1 in the white dashed box in the left panel. Scale bar: 2 µm. The right panel shows the VGLUT3 boutons co-labeled with SERT (c), 5-HT (f) and SERT+5-HT (i). VGLUT3 puncta located within 5-HT varicosities were isolated using Imaris in the CA1 (l). Arrow heads show 5-HT varicosities collocated with SERT and VGLUT3 in the CA1 (c, f, i and l). C. (a-d) Quantification of the proportion of 5-HT varicosities not co-labeled (SERT-) and co-labeled (SERT+) with SERT in the CA1, CA2, CA3 and DG of the HIP, showing a great majority of 5-HT varicosities co-labeled with SERT. (e-h) Quantification of the proportion of 5-HT+/VGLUT3+ not co-labeled (SERT-) and co-labeled (SERT+) with SERT in the CA1, CA2, CA3 and DG of the HIP, showing a great majority of 5-HT/VG varicosities co-labeled with SERT (t test, ****: p<0.0001; **: p<0.01).

We have reconstructed a total of 106 serotonergic varicosities from various

brain regions, including the prelimbic cortex (5x104), nucleus accumbens (10x104),

lateral septum (15x104), BNST (25x104), amygdala (27x104), hippocampus (15x104)

and dorsal striatum (5x104). Our quantification of the volumetric density of 5-

HT+/VGLUT3+ boutons revealed a high heterogeneity along the different brain

regions. The, CA1-3, lateral septum, amygdala (BLA and CeA) and NAc shell show

the highest density, and the NAc core, striatum, dentate gyrus and prelimbic cortex

239

show the lowest density of 5-HT+/VGLUT3+ varicosities (Fig. 8A) (see table 1 for

detailed statistics). The relative density of 5-HT+/VGLUT3+ boutons, calculated as a

percentage of total 5-HT varicosities within each brain region, shows the largest

proportion is located in CA1-3 (24 to 32 %), followed by the PrL (20-23 %) > DG

(19 %) > CeA (16 %) > LS (14 %) > BLA (13 %) > NAc shell (10 %) > BNST (7 %)

> striatum (6 %) > NAc core (4.2 %) (Fig. 8B and Fig. 9). Correlation analysis shows

that the density of 5-HT+/VGLUT3+ varicosities is totally independent of the density

of 5-HT boutons within each brain region (Pearson’s coefficients r = 0.32 and R2 =

0.10, p = 0.28). This data suggests that the heterogeneous distribution of 5-

HT+/VGLUT3+ varicosities in the forebrain represents a functionally relevant feature

of 5-HT neurons complex topology, rather than a biased detection of those

varicosities in our methodological approach.

Figure 7-8 Quantification of 5-HT neurons expressing vesicular glutamate transporter (VGLUT3) in various regions of the mouse forebrain.

Volumetric quantification of the density of VGLUT3 immunoreactive boutons within 5-HT-labeled fibers in each brain region. The results are expressed as density of boutons per 106 mm3 of 5-HT+ fiber (a), or percent of boutons in 5-HT+ fiber (b) and represented as the mean ± SEM.

240

Figure 7-9 Visual representation of the proportion of 5-HT varicosities expressing the vesicular glutamate transporter VGLUT3 in various regions of the mouse forebrain.

The percentage of total 5-HT varicosities that are co-labeled with VGLUT3 is represented with a color coding from 0 % (dark blue) to 32 % (dark red) within each analyzed brain region, from various viewing angle. The highest proportion of VGLUT3 was found in CA2, followed by CA3, CA1, PrL IV-V, PrL I-III, DG, CeA, BLA, NAC shell, BNST and NAC core.

7.5 Discussion

The main finding of the present study is that SERT and VGLUT3 rarely

segregate within 5-HT varicosities, but rather preferentially co-localize in most of the

mouse forebrain regions we analyzed. These results are in agreement with a previous

observation in rats (Gagnon and Parent, 2014), but surprisingly differ from two

previous studies in WT littermates (vglut3+/+) of a transgenic mouse line, obtained by

the breeding of vglut3+/- mice (Amilhon et al., 2010; Voisin et al., 2016). In these

mice, a clear segregation of SERT and VGLUT3 was reported in the varicosities of

the 5-HTergic axons projecting to the prelimbic cortex, the ventral and dorsal CA3

field of the hippocampus, the dorsal striatum and the lateral septum (Amilhon et al.,

2010; Voisin et al., 2016). In our study in wild-type C57Bl6 mice, we did not

observe this segregation between SERT and VGLUT3 in5-HT varicosities. This

suggests that the aforementioned discrepancies may originate from technical issues,

241

probably related to antibody specificity or sensitivity, rather than true interspecies

differences. Further studies using a different methodological approach are therefore

needed to confirm the co-localization or segregation of SERT and VGLUT3 within

5-HT axonal varicosities.

Axonal segregation between SERT and VGLUT3 within 5-HT varicosities

would hence imply that a significant subset of 5-HT varicosities do not express the

SERT. However, our results demonstrate that the great majority of 5-HT boutons

(about 90%) are immunoreactive for the SERT, in almost all the brain regions

studied. The only brain region with an equivalent proportion of SERT-positive and

SERT-negative 5-HT boutons was the posterior part of the nucleus accumbens shell.

This is in line with previous reports in rats showing that only subtle differences could

be detected between 5-HT and of SERT immunostaining in particular brain regions,

with a subset of 5-HT axonal projections that lack the SERT in the posterior part of

the nucleus accumbens (Brown and Molliver, 2000). Since 5-HT could be stored in

dopaminergic neurons of the substantia nigra or the ventral tegmental area (Zhou et

al., 2005), we assessed the dopaminergic nature of these SERT-/5-HT+ varicosities.

The absence of immunoreactivity for the tyrosine hydroxylase has confirmed the

non-catecholaminergic phenotype of these varicosities (data not shown). Retrograde

tracing studies are currently performed in our laboratory to determine the origin of

this particular subset of SERT-/5-HT+ axons, i.e. whether these projections originate

from the dorsal, median or caudal raphe and whether this particular subtype of 5-HT

neurons are restricted to specific raphe subnuclei.

The high density of 5-HT/glutamate co-release sites identified within brain

regions known to be involved in the control of emotion, reward and decision-making

such as the amygdala, NAc shell, hippocampus and prefrontal cortex, evoke new

242

potential mechanisms for the control of the neuroplasticity in these brain regions, and

its dysregulation in the development of addictive disorders (Kauer and Malenka,

2007). In line with this, the genetic ablation of vglut3 in mice was shown to

predispose to cocaine abuse (Sakae et al., 2015), however, the authors reported that

this mechanism is likely independent of 5-HT signaling but rather linked to an

increased dopamine and glutamate signaling in the nucleus accumbens, from VTA

dopaminergic and cortical glutamatergic inputs, respectively. Further studies are

hence needed to identify the contribution of serotonin/glutamate co-signaling in this

effect.

Indeed, recent evidence supports a significant role of 5-HT/glutamate co-

transmission in both reward and emotion (Amilhon et al., 2010; Liu et al., 2014).

Further, the activation of dopamine- and cAMP-regulated phosphoprotein of M(r)

32,000 (DARPP-32) by dopamine/glutamate co-transmission has been postulated to

act as a molecular switch that control the reward pathway plasticity that mediates the

behavioral sensitization to various drugs of abuse (Nairn et al., 2004; Valjent et al.,

2005). DARPP-32 also appears to be involved in some biochemical and behavioral

actions of 5-HT (Svenningsson et al., 2002), however, there is no evidence to date

that serotonin-induced DARPP-32 activation is mediated by 5-HT/glutamate co-

transmission. Further studies are thus required to determine the precise role(s) played

by 5-HT and/or glutamate and their different receptors in DARPP-32 activation, the

subsequent neuroplastic changes and associated cognitive/emotional deficits

produced by drugs of abuse. Furthermore, genetic ablation of VGLUT3 in mice

(vglut3-/-) produces increased anxiety (Amilhon et al., 2010; Sakae et al., 2018),

enhanced fear and altered stress axis regulation (Balázsfi et al., 2018), concomitant to

the desensitization of 5-HT1A autoreceptors (Amilhon et al., 2010), suggesting a

243

potential cross-regulation between 5-HT1A receptors and glutamate co-release. Since

the desensitization of 5-HT1A autoreceptors is essential for the antidepressant effect

of selective serotonin reuptake inhibitors (SSRIs)(Popa et al., 2010), this further

suggests that selective pharmacological ablation of VGLUT3 might represent a

potential adjunct for antidepressant therapy.

7.6 References

Amilhon, B., Lepicard, E., Renoir, T., Mongeau, R., Popa, D., Poirel, O., et al. (2010). VGLUT3 (vesicular glutamate transporter type 3) contribution to the regulation of serotonergic transmission and anxiety. J. Neurosci. Off. J. Soc. Neurosci. 30, 2198–2210. doi:10.1523/JNEUROSCI.5196-09.2010. Aznar, S., Kostova, V., Christiansen, S. H., and Knudsen, G. M. (2005). Alpha 7 nicotinic receptor subunit is present on serotonin neurons projecting to hippocampus and septum. Synap. N. Y. N 55, 196–200. doi:10.1002/syn.20108. Bakker, R., Tiesinga, P., and Kötter, R. (2015). The Scalable Brain Atlas: Instant Web-Based Access to Public Brain Atlases and Related Content. Neuroinformatics 13, 353–366. doi:10.1007/s12021-014-9258-x. Balázsfi, D., Fodor, A., Török, B., Ferenczi, S., Kovács, K. J., Haller, J., et al. (2018). Enhanced innate fear and altered stress axis regulation in VGluT3 knockout mice. Stress Amst. Neth. 21, 151–161. doi:10.1080/10253890.2017.1423053. Belmer, A., Klenowski, P. M., Patkar, O. L., and Bartlett, S. E. (2016). Mapping the connectivity of serotonin transporter immunoreactive axons to excitatory and inhibitory neurochemical synapses in the mouse limbic brain. Brain Struct. Funct. doi:10.1007/s00429-016-1278-x. Bonnavion, P., Bernard, J.-F., Hamon, M., Adrien, J., and Fabre, V. (2010). Heterogeneous distribution of the serotonin 5-HT(1A) receptor mRNA in chemically identified neurons of the mouse rostral brainstem: Implications for the role of serotonin in the regulation of wakefulness and REM sleep. J. Comp. Neurol. 518, 2744–2770. doi:10.1002/cne.22331. Brown, P., and Molliver, M. E. (2000). Dual serotonin (5-HT) projections to the nucleus accumbens core and shell: relation of the 5-HT transporter to amphetamine-induced neurotoxicity. J. Neurosci. Off. J. Soc. Neurosci. 20, 1952–1963. Calizo, L. H., Akanwa, A., Ma, X., Pan, Y., Lemos, J. C., Craige, C., et al. (2011). Raphe serotonin neurons are not homogenous: Electrophysiological, morphological and neurochemical evidence. Neuropharmacology 61, 524–543. doi:10.1016/j.neuropharm.2011.04.008.

244

Descarries, L., Riad, M., and Parent, M. (2010). “CHAPTER 1.4 - Ultrastructure of the Serotonin Innervation in the Mammalian Central Nervous System,” in Handbook of Behavioral Neuroscience Handbook of the Behavioral Neurobiology of Serotonin., ed. C. P. M. and B. L. Jacobs (Elsevier), 65–101. Available at: http://www.sciencedirect.com/science/article/pii/S1569733910700722 [Accessed November 4, 2015]. Fernandez, S. P., Cauli, B., Cabezas, C., Muzerelle, A., Poncer, J.-C., and Gaspar, P. (2016). Multiscale single-cell analysis reveals unique phenotypes of raphe 5-HT neurons projecting to the forebrain. Brain Struct. Funct. 221, 4007–4025. doi:10.1007/s00429-015-1142-4. Fu, W., Le Maître, E., Fabre, V., Bernard, J.-F., David Xu, Z.-Q., and Hökfelt, T. (2010). Chemical neuroanatomy of the dorsal raphe nucleus and adjacent structures of the mouse brain. J. Comp. Neurol. 518, 3464–3494. doi:10.1002/cne.22407. Gagnon, D., and Parent, M. (2014). Distribution of VGLUT3 in highly collateralized axons from the rat dorsal raphe nucleus as revealed by single-neuron reconstructions. PloS One 9, e87709. doi:10.1371/journal.pone.0087709. Gaspar, P., and Lillesaar, C. (2012). Probing the diversity of serotonin neurons. Philos. Trans. R. Soc. B Biol. Sci. 367, 2382–2394. doi:10.1098/rstb.2011.0378. Gras, C., Herzog, E., Bellenchi, G. C., Bernard, V., Ravassard, P., Pohl, M., et al. (2002). A third vesicular glutamate transporter expressed by cholinergic and serotoninergic neurons. J. Neurosci. Off. J. Soc. Neurosci. 22, 5442–5451. Hajós, M., Allers, K. A., Jennings, K., Sharp, T., Charette, G., Sík, A., et al. (2007). Neurochemical identification of stereotypic burst-firing neurons in the rat dorsal raphe nucleus using juxtacellular labelling methods. Eur. J. Neurosci. 25, 119–126. doi:10.1111/j.1460-9568.2006.05276.x. Hioki, H., Nakamura, H., Ma, Y.-F., Konno, M., Hayakawa, T., Nakamura, K. C., et al. (2010). Vesicular glutamate transporter 3-expressing nonserotonergic projection neurons constitute a subregion in the rat midbrain raphe nuclei. J. Comp. Neurol. 518, 668–686. doi:10.1002/cne.22237. Hornung, J. P., Fritschy, J. M., and Törk, I. (1990). Distribution of two morphologically distinct subsets of serotoninergic axons in the cerebral cortex of the marmoset. J. Comp. Neurol. 297, 165–181. doi:10.1002/cne.902970202. Kast, R. J., Wu, H.-H., Williams, P., Gaspar, P., and Levitt, P. (2017). Specific Connectivity and Unique Molecular Identity of MET Receptor Tyrosine Kinase Expressing Serotonergic Neurons in the Caudal Dorsal Raphe Nuclei. ACS Chem. Neurosci. 8, 1053–1064. doi:10.1021/acschemneuro.7b00020. Kauer, J. A., and Malenka, R. C. (2007). Synaptic plasticity and addiction. Nat. Rev. Neurosci. 8, 844–858. doi:10.1038/nrn2234.

245

Kirby, L. G., Pernar, L., Valentino, R. J., and Beck, S. G. (2003). Distinguishing characteristics of serotonin and non-serotonin-containing cells in the dorsal raphe nucleus: electrophysiological and immunohistochemical studies. Neuroscience 116, 669–683. Kiyasova, V., Bonnavion, P., Scotto-Lomassese, S., Fabre, V., Sahly, I., Tronche, F., et al. (2013). A subpopulation of serotonergic neurons that do not express the 5-HT1A autoreceptor. ACS Chem. Neurosci. 4, 89–95. doi:10.1021/cn300157s. Kiyasova, V., Fernandez, S. P., Laine, J., Stankovski, L., Muzerelle, A., Doly, S., et al. (2011). A genetically defined morphologically and functionally unique subset of 5-HT neurons in the mouse raphe nuclei. J. Neurosci. Off. J. Soc. Neurosci. 31, 2756–2768. doi:10.1523/JNEUROSCI.4080-10.2011. Kosofsky, B. E., and Molliver, M. E. (1987). The serotoninergic innervation of cerebral cortex: different classes of axon terminals arise from dorsal and median raphe nuclei. Synap. N. Y. N 1, 153–168. doi:10.1002/syn.890010204. Lacoste, B., Riad, M., and Descarries, L. (2006). Immunocytochemical evidence for the existence of substance P receptor (NK1) in serotonin neurons of rat and mouse dorsal raphe nucleus. Eur. J. Neurosci. 23, 2947–2958. doi:10.1111/j.1460-9568.2006.04833.x. Landmann, L., and Marbet, P. (2004). Colocalization analysis yields superior results after image restoration. Microsc. Res. Tech. 64, 103–112. doi:10.1002/jemt.20066. Larm, J. A., Shen, P.-J., and Gundlach, A. L. (2003). Differential galanin receptor-1 and galanin expression by 5-HT neurons in dorsal raphé nucleus of rat and mouse: evidence for species-dependent modulation of serotonin transmission. Eur. J. Neurosci. 17, 481–493. Liu, Z., Zhou, J., Li, Y., Hu, F., Lu, Y., Ma, M., et al. (2014). Dorsal raphe neurons signal reward through 5-HT and glutamate. Neuron 81, 1360–1374. doi:10.1016/j.neuron.2014.02.010. Mintz, E. M., and Scott, T. J. (2006). Colocalization of serotonin and vesicular glutamate transporter 3-like immunoreactivity in the midbrain raphe of Syrian hamsters (Mesocricetus auratus). Neurosci. Lett. 394, 97–100. doi:10.1016/j.neulet.2005.10.033. Nairn, A. C., Svenningsson, P., Nishi, A., Fisone, G., Girault, J.-A., and Greengard, P. (2004). The role of DARPP-32 in the actions of drugs of abuse. Neuropharmacology 47 Suppl 1, 14–23. doi:10.1016/j.neuropharm.2004.05.010. Popa, D., Cerdan, J., Repérant, C., Guiard, B. P., Guilloux, J.-P., David, D. J., et al. (2010). A longitudinal study of 5-HT outflow during chronic fluoxetine treatment using a new technique of chronic microdialysis in a highly emotional mouse strain. Eur. J. Pharmacol. 628, 83–90. doi:10.1016/j.ejphar.2009.11.037.

246

Prouty, E. W., Chandler, D. J., and Waterhouse, B. D. (2017). Neurochemical differences between target-specific populations of rat dorsal raphe projection neurons. Brain Res. 1675, 28–40. doi:10.1016/j.brainres.2017.08.031. Ren, J., Friedmann, D., Xiong, J., Liu, C. D., Ferguson, B. R., Weerakkody, T., et al. (2018). Anatomically Defined and Functionally Distinct Dorsal Raphe Serotonin Sub-systems. Cell 175, 472-487.e20. doi:10.1016/j.cell.2018.07.043. Sakae, D. Y., Marti, F., Lecca, S., Vorspan, F., Martín-García, E., Morel, L. J., et al. (2015). The absence of VGLUT3 predisposes to cocaine abuse by increasing dopamine and glutamate signaling in the nucleus accumbens. Mol. Psychiatry 20, 1448–1459. doi:10.1038/mp.2015.104. Sakae, D. Y., Ramet, L., Henrion, A., Poirel, O., Jamain, S., El Mestikawy, S., et al. (2018). Differential expression of VGLUT3 in laboratory mouse strains: Impact on drug-induced hyperlocomotion and anxiety-related behaviors. Genes Brain Behav., e12528. doi:10.1111/gbb.12528. Schäfer, M. K.-H., Varoqui, H., Defamie, N., Weihe, E., and Erickson, J. D. (2002). Molecular cloning and functional identification of mouse vesicular glutamate transporter 3 and its expression in subsets of novel excitatory neurons. J. Biol. Chem. 277, 50734–50748. doi:10.1074/jbc.M206738200. Sengupta, A., Bocchio, M., Bannerman, D. M., Sharp, T., and Capogna, M. (2017). Control of Amygdala Circuits by 5-HT Neurons via 5-HT and Glutamate Cotransmission. J. Neurosci. 37, 1785–1796. doi:10.1523/JNEUROSCI.2238-16.2016. Shutoh, F., Ina, A., Yoshida, S., Konno, J., and Hisano, S. (2008). Two distinct subtypes of serotonergic fibers classified by co-expression with vesicular glutamate transporter 3 in rat forebrain. Neurosci. Lett. 432, 132–136. doi:10.1016/j.neulet.2007.12.050. Sotelo, C., Cholley, B., El Mestikawy, S., Gozlan, H., and Hamon, M. (1990). Direct Immunohistochemical Evidence of the Existence of 5-HT1A Autoreceptors on Serotoninergic Neurons in the Midbrain Raphe Nuclei. Eur. J. Neurosci. 2, 1144–1154. Svenningsson, P., Tzavara, E. T., Liu, F., Fienberg, A. A., Nomikos, G. G., and Greengard, P. (2002). DARPP-32 mediates serotonergic neurotransmission in the forebrain. Proc. Natl. Acad. Sci. U. S. A. 99, 3188–3193. doi:10.1073/pnas.052712699. Tarren, J. R., Lester, H. A., Belmer, A., and Bartlett, S. E. (2017). Acute Ethanol Administration Upregulates Synaptic α4-Subunit of Neuronal Nicotinic Acetylcholine Receptors within the Nucleus Accumbens and Amygdala. Front. Mol. Neurosci. 10, 338. doi:10.3389/fnmol.2017.00338. Valjent, E., Pascoli, V., Svenningsson, P., Paul, S., Enslen, H., Corvol, J.-C., et al. (2005). Regulation of a protein phosphatase cascade allows convergent dopamine

247

and glutamate signals to activate ERK in the striatum. Proc. Natl. Acad. Sci. U. S. A. 102, 491–496. doi:10.1073/pnas.0408305102. Voisin, A. N., Mnie-Filali, O., Giguère, N., Fortin, G. M., Vigneault, E., El Mestikawy, S., et al. (2016). Axonal Segregation and Role of the Vesicular Glutamate Transporter VGLUT3 in Serotonin Neurons. Front. Neuroanat. 10, 39. doi:10.3389/fnana.2016.00039. Xu, Z. Q., and Hökfelt, T. (1997). Expression of galanin and nitric oxide synthase in subpopulations of serotonin neurons of the rat dorsal raphe nucleus. J. Chem. Neuroanat. 13, 169–187. Zhou, F.-M., Liang, Y., Salas, R., Zhang, L., De Biasi, M., and Dani, J. A. (2005). Corelease of dopamine and serotonin from striatal dopamine terminals. Neuron 46, 65–74. doi:10.1016/j.neuron.2005.02.010.

248

Fear extinction recall mediated by 5-HT/VGLUT3

colocalisation

This chapter comprises the following submitted article: Jacques, A., Chaaya, N., Belmer, A., Beecher, K., Ali, S. A., Battle, A. R., Johnson, L. R., Chehrehasa, F., Bartlett, S. E. (2019) Fear extinction recall mediated by 5-HT/VGLUT3 colocalisation. Frontiers in Behavioural Neuroscience Submitted: 20 March, 2019.

249








In the case of chapter 8: Fear extinction recall facilitated by 5-HT/VGLUT3 colocalisation Publication status: Submitted

Contributor Statement of contribution*Angela Jacques Designed the project, conducted behavioural and laboratory experiments,

analysed data, created the figures and wrote and edited the manuscript.

Nicholas Chaaya Assisted with behavioural and laboratory experiments.

Kate Beecher Assisted with editing the manuscript.

Syed Aoun Ali Assisted with editing the manuscript.


Luke Johnson Involved in the conception and design of the project.

Arnauld Belmer Involved in the conception and design of the project, and assisted with editing the manuscript and assisted with laboratory experiments, imaging and data analysis.

Fatemeh Chehrehasa Involved in the conception and design of the project, and assisted with editing the manuscript and assisted with editing the manuscript

Selena Bartlett Involved in the conception and design of the project, assisted with reviewing and editing the manuscript.

QUT Verified Signature

250

8.1 Abstract

The hallmark of social anxiety and panic disorder is excessive fear. The

neurobiological mechanisms that underpin the recall of fear and extinction memories

may assist in development of strategies to improve the treatment of these disorders.

Extinction memories are particularly labile and the introduction of certain stimuli

may result in a re-emergence of the original fear. As the divergent molecular

mechanisms and complex neural circuits involved in the extinction process relevant

to cognitive extinction based therapies are a long way from being understood, this

study examined two of the essential elements. Serotonin (5-Hydroxytryptamine, 5-

HT) modulates the acquisition and storage of conditioned emotional memories and

glutamate facilitates learning and memory. In addition, the loss of vesicular

glutamate transporter 3 (VGLUT3) expression may lead to anxiety-like behaviour.

Here we tested the hypothesis that the colocalisation of VGLUT3 puncta within 5-

HT varicosities (reported as 5-HT+/VGLUT3+ boutons) may correlate with the recall

of extinction memories. To test this hypothesis, male Sprague Dawley rats underwent

auditory Pavlovian fear conditioning and extinction using an extended protocol. We

quantified the distribution of VGLUT3 puncta within 5-HT immunoreactive

varicosities apposed to pMAPK+ neurons within the prefrontal cortex using

immunohistochemistry, high resolution confocal microscopy and 3D image

reconstruction techniques. Significant increases in the density of pMAPK+ neurons

was observed in the extinction recall group with a corresponding increase in the

percentage of 5-HT+/VGLUT3+ boutons in their vicinity. This data suggests these

VGLUT3+ serotonergic axons may play a role in plasticity and represent a potential

target in the extinction of pathological fear and anxiety.

251

8.2 Introduction

Social anxiety disorder, panic disorder, post-traumatic stress disorder and

phobias are characterised by failures of extinction and inhibitory learning

(Cannistraro & Rauch, 2003). Central nervous system serotonin (5-

Hydroxytryptamine, 5-HT) originates from the raphe nuclei, modulates mood,

emotion and behaviours (Jacobs & Azmitia, 1992), and enhances learning and

memory (Dai et al., 2008; Kandel, 2004; Zhang et al., 2013). Pathophysiological

changes in levels of 5-HT transmission occur in cases of memory dysfunction,

anxiety, depression and post-traumatic stress disorder (Owens & Nemeroff, 1994).

Well documented treatment of anxiety related disorders revolve around modulation

of the serotonergic system (Bezchlibnyk-Butler, Aleksic, & Kennedy, 2000; Guiard

& Di Giovanni, 2018) however, the role of the glutamate transporter within

serotonergic projections has not been observed during relevant therapies for these

disorders.

Extinction and inhibitory learning underpin exposure therapy, a cognitive

behavioural therapy often utilised in the treatment of anxiety and pathological fear.

Understanding the functional role of the serotonergic system in extinction and

inhibitory learning can be facilitated through research utilizing a Pavlovian fear

conditioning protocol. In auditory fear conditioning a tone (CS - conditioned

stimulus) is paired with a mild foot shock (US - unconditioned stimulus) resulting in

a behavioural reaction to the associated CS. In rodents, this behaviour is known as

freezing and is depicted by the cessation of movement other than for breathing (J. E.

LeDoux, J. Iwata, P. Cicchetti, & D. J. Reis, 1988). Repeated exposure to the CS

alone generates a new extinction memory whereby the CS no longer elicits a fear

response (Bouton, Westbrook, Corcoran, & Maren, 2006; Haubrich et al., 2017;

252

Gregory J Quirk & Mueller, 2008). This is a robust and well validated protocol

capable of modelling the learning and recall of fear and extinction memories and one

in which the neural circuitry has been clearly defined (Bezchlibnyk-Butler et al.,

2000; Wilensky, Schafe, Kristensen, & LeDoux, 2006).

The current study examines the medial prefrontal cortex (mPFC), which is

implicated in the expression of both fear and extinction memories (Bredy et al.,

2007; Angela Jacques et al., 2019) and is densely innervated by 5-HT neurons

(Cornea‐Hébert, Riad, Wu, Singh, & Descarries, 1999). Increased levels of

extracellular 5-HT in the rat PFC have been observed immediately post fear

conditioning (Yoshioka, Matsumoto, Togashi, & Saito, 1995) suggesting its

functional role in the consolidation of fearful memories and correlated behaviours.

As extracellular 5-HT levels increase, activation of the 5-HT2A receptor stimulates

intracellular signalling cascades that lead to the phosphorylation of mitogen-activated

protein kinase (pMAPK), a well-established marker of neuroplasticity (Musazzi et

al., 2010; Orsini & Maren, 2012). Indeed, extensive evidence has demonstrated the

pivotal role of 5-HT in fear conditioning (for review see (Haubrich et al., 2017)). As

neuroplastic change is imperative to the formation of new extinction memories the

goal of our investigation was to discern whether there was a correlation between

certain serotonergic inputs, (specifically those with the capacity to corelease

glutamate) and the recall of extinction memories.

Glutamate N-methyl-d-aspartate (NMDA) receptors in mPFC are activated

during the acquisition of an extinction memory but not the acquisition of a

conditioned fear memory (Burgos-Robles, Vidal-Gonzalez, Santini, & Quirk, 2007).

Glutamate elicits fast excitatory neurotransmission and mediates both memory

consolidation and retrieval (Hassel & Dingledine, 2012). Glutamate is loaded into

253

synaptic vesicles through proteins known as vesicular glutamate transporters

(VGLUTs). Of particular note is VGLUT3, a transporter expressed in discrete

neuronal populations including 5-HT neurons projecting to the PFC (Herzog et al.,

2004), which allows for serotonin and glutamate to be co-released from the same

terminals (Gras et al., 2002), a mechanism proposed to play an important role in

neuronal plasticity (Gagnon & Parent, 2014).

This study tests the hypothesis that the colocalisation of VGLUT3 puncta

within 5-HT varicosities (reported as 5-HT+/VGLUT3+ boutons) may correlate with

the plasticity generated through the recall of extinction memories. The search for

molecular markers to strengthen extinction memory formation is essential to the

evolution of anxiety and fear related therapeutics.

Here we used Pavlovian fear conditioning and an extended extinction protocol

whereby the extinction to tone training spanned 3 days, followed by three days with

no training. The protocol was designed to ensure the long-term potentiation of both a

fear memory and an extinction memory prior to reconsolidation. To evaluate whether

VGLUT3 immunoreactive 5-HT varicosities were closely apposed (< 1 µm) to

pMAPK+ neurons in the PFC of male Sprague Dawley rats we combined

immunohistochemistry, high resolution confocal imaging and 3D image

reconstruction (Belmer et al., 2019; Belmer et al., 2017). Our results showed that the

recall of an extinction memory both increased the density of pMAPK neurons and

upregulated the proportion of VGLUT3-containing 5-HT neuronal inputs.

254

8.3 Methods

8.3.1 Subjects

Experimentally naive adult male Sprague-Dawley rats (N = 21, 3 groups, n =

7) were supplied by Animal Resource Centre, ARC, Western Australia. They were

housed 3 per cage, in temperature (≈ 24 °C) and humidity (35 %) controlled

Plexiglas cages maintained on a 12-hour light/dark cycle with the behavioural

procedures conducted during the light cycle. Fear conditioning has been observed as

more effective during the nocturnal phase for rodents, equating to the light cycle

(Albrecht & Stork, 2017). Rats were acclimatized to the vivarium for 7 days prior to

training, with food and water provided ad libitum and by end of training protocols

weighed 326.4 ± 6.7 g on average. All procedures were reviewed and approved

(AEC Approval Number: QUT/280/17) by the Animal Welfare Unit, University of

Queensland Research and Innovation Ethics Committee and the Research Ethics

Committee of the Queensland University of Technology, Australia. All procedures

complied with the policies, regulations and ethical standards for animal

experimentation, in accordance with the Queensland Government Animal Research

Act 2001, associated Animal Care and Protection Regulations (2002 and 2008), and

the Australian Code for the Care and Use of Animals for Scientific Purposes, 8th

Edition (National Health and Medical Research Council, 2013).

8.3.2 Apparatus

Behavioural procedures were conducted in two conditioning chambers

referred to as Context A and Context B. The chambers were modified to create

unique testing environments for fear conditioning or extinction thus restricting any

reaction to context. Context A contained a stainless-steel grid floor connected to a

shock generator and computer (Freeze frame software, Coulbourn instruments).

255

Ethanol (80 %) was used in Context A to clean the chamber between each animal.

Context B contained a solid plastic floor with fresh bedding and internal coloured

decoration on the walls and ceiling to provide contextual difference. Orange scented

antibacterial soap was the cleaning agent that provided a unique smell to the

chamber. The chambers were housed within an acoustic isolation box (Coulbourn

Instruments, Allentown, PA, USA) with the background noise level measured at 55

dB, using a sound level meter (Digitech Professional Sound Level Meter QM1592).

Each chamber contained a speaker, low-level house light (2-3 lux), infrared light and

infrared camera.

8.3.3 Behavioural procedures

A total of 21 rats were randomized into three groups, a remote extinction

memory recall group (n = 7), remote fear memory recall group (n = 7), and a recent

fear memory recall group (n = 7). Experimental design and behavioural results are

depicted in Figure 1. Each of the behavioural training protocols have been previously

described in (Angela Jacques et al., 2019). For two days prior to training, rats were

habituated to context A and context B for 30 minutes in each. On the third day of

training the remote extinction and remote fear groups underwent three minutes of

acclimation to context A, followed by a 10-minute fear conditioning protocol that

consisted of three pairings of an auditory conditioned stimulus (CS, tone - 5kHz, 75

dB, 20 s) that co-terminated with an unconditioned stimulus (US, foot shock - 0.6

mA, 500 ms). Stimuli were controlled through Freeze Frame software (Coulbourn

Instruments) with training separated by a mean inter-trial interval of 180 s. The rats

were returned to their home cages after final stimulus presentation. On the fourth,

fifth and sixth days of the protocol the remote extinction group underwent 30

minutes of extinction training (20 x CS alone, 5 kHz, 75 dB, 20 s) in context B after

256

which they were returned to their home cages for three days (days 7, 8 & 9) to allow

for memory consolidation. The remote fear memory recall group followed the same

training regime, but received no stimuli in context B.

On the ninth day the recent fear memory recall group underwent the same

fear conditioning training the other two groups had received. On the tenth day all

groups underwent a 10-minute memory recall test consisting of three 20 s CS

presentations to test for recall of the consolidated auditory fear memory or extinction

memory. Freezing, a behavioural index used to quantify CS-US association

(Blanchard & Blanchard, 1969; Michael S Fanselow, 1984), was scored during the

20 s CS intervals by an experimenter blind to the conditions. These intervals were

indicated in the video recordings by an infrared light allowing the behaviour of the

remote fear memory group to be scored for the same length of time and at the same

time points as the remote extinction group. Behavioural results were expressed as

percentage time freezing (dependent variable).

257

Figure 8-1 The recall of recent and remote fear results in different levels of freezing compared to extinction memory recall. (a) Habituation: Habituation to context provided a base line freezing measure (0.2 ± 0.1%) with no difference noted between freezing levels to each context (0.04 ± 0.2 %, t 34 = 0.23, ns, p = 0.8218). (b) Fear conditioning: Auditory fear conditioning (3 x CS + US) was conducted seven days apart with the

258

remote extinction and remote fear groups on day 3 and the recent fear group on day 9. Percentage freezing increased at each pairing of CS and US (F2,4 = 106.3, ***, p = 0.0003 ) but did not differ between groups. (c-e) 1st, 2nd and 3rd extinction session occurred on days 4, 5 & 6: The effect of extinction or no extinction training (20 presentations of CS alone or context alone, in context B) is shown as four CS presentations averaged to one block. Higher % freezing was observed in the first block of extinction trials as compared with the last block (remote extinction: 1st block 83.5 ± 5.5%, 4th block 10.3 ± 3.3%) on the first day of training. There was a significant temporal interaction between the beginning vs end of training, first day: t 54 = 11.46, ****, p < 0.0001; second day: t 54 = 7.08, ****, p < 0.0001; third day: t 54 = 2.28, *, p = 0.0265). (f) FMT: Recall of fear memory or extinction memory was tested via a 10-minute recall test consisting of 3 x CS only presentations. % freezing levels were lower due to remote extinction recall as opposed to fear memory recall (remote extinction vs remote fear and recent fear, F2,4 = 37.11, **, p = 0.0026). Data are presented as mean % freezing ± S.E.M, *: p ≤ 0.05, **: p ≤ 0.01, ***: p ≤ 0.0001, ****: p ≤ 0.0001, two-way ANOVA followed by Bonferroni post hoc analysis.


Rats were anesthetized with lethabarb (400 mg/kg, i.p.) and transcardially

perfused through the ascending aorta with ice-cold 1% (wt/vol) PFA with 0.125%

(vol/vol) glutaraldehyde followed by 4 % (wt/vol) PFA with 0.125% (vol/vol)

glutaraldehyde in 0.1 M phosphate buffer, 60 minutes after the fear memory recall

test. Brains were extracted and post-fixed in 4 % PFA with 0.125% (vol/vol)

glutaraldehyde overnight then stored in 0.1 M phosphate buffered saline (PBS). Free-

floating serial coronal sections (40 μm) of the medial prefrontal cortex were prepared

using a vibratome (M11000; Pelco easiSlicer, Ted Pella Inc, Redding, CA, USA).


Immunohistochemistry was performed as previously described (Belmer et al.,

2019). Briefly, three prefrontal sections containing the pre-limbic cortex (every 2nd

section caudally from Bregma +3.24 mm), were taken from each subject and

immunolabelled with antibodies against VGLUT3, 5-HT and pMAPK. Briefly, the

sections were washed three times in PBS (containing 0.02 % sodium azide) and

blocked in PBS containing, 0.3 % Triton X-100, 0.05 % Tween 20 and 2 % Normal

goat serum (NGS, Abcam, Vic, Aus) for 1 h. After blocking sections were incubated

259

with primary antibody: rat anti-5-HT (Millipore #MAB352, 1:100) diluted in the

blocking solution for 48 hours at room temperature. Primary antibody guinea-pig

anti-VGLUT3 (Synaptic System #135204, 1:500) was then added directly to each

well and incubated for a further 24 hours at room temperature. Sections were

washed 3 times in blocking solution and incubated for 4 hours at room temperature

in secondary antibodies goat anti-guinea pig-Alexa 647 (Thermofisher Scientific,

#A21450, 1:500) and goat anti-rat 568 (Thermofisher Scientific, #A11077, 1:500).

Sections were washed 4 times in PBS and once in blocking solution as listed above

without the Triton X-100, as triton had previously been observed as affecting the

pMAPK labelling quality. Sections were incubated for 24 hours in primary antibody

phosphor-p44/42 MAPK (Erk 1/2) (Thr 202/Tyr 204) (1:250; #4370, Cell Signalling

Technology, MA, USA) diluted in the no-triton blocking solution. After three washes

in the blocking solution, the slices were incubated for 4 hours at room temperature

with secondary antibody goat anti rabbit-Alexa 488 (Thermofisher Scientific,

#A11034; 1:500) diluted in blocking solution. Sections were washed 3 times in PBS,

mounted on silane-coated slides and cover slipped using ProLong Gold antifade

reagent (Thermofisher Scientific, #P36934, Invitrogen, DR, USA).

8.3.6 Imaging

3 sections per animal, n = 5 animals per group, totalling 15 sections per

condition, were imaged on an Olympus FV3000 confocal microscope using a 60X

oil-immersion objective (NA 1.35) with a 1.5 x zoom and a Z-axis step of 0.5 μm,

using sequential scanning. The sections were taken from the prefrontal cortex around

Bregma +3.24mm, and mosaics of the regions of interest were acquired as depicted

in Figure (2 a, b), in OIR file format. Each mosaic consisted of a data volume of 1.95

260

x 106 µm3. The pMAPK immunoreactive neurons and 5-HT boutons were

reconstructed in 3D using the surface rendering function with Imaris 9.2.1 (Bitplane),

similarly to our previous work (Belmer et al., 2019; Belmer et al., 2017). The

removal (or masking) of the VGLUT3 fluorescence signal outside of the 5-HT

surface ensured that only intra-fibre VGLUT3 labelling remained, allowing for

quantification of 5-HT boutons containing VGLUT3 puncta. The ‘spot detection’

function was applied for each created mask. Images were batch processed using the

same surface thresholding parameters. Figure (2 c-f) depicts Imaris images. Mean

fluorescence intensities and image volumes were obtained from the statistics function

in Imaris. Each of these steps have been previously described in (Belmer et al., 2017;

Tarren et al., 2017).

Figure 8-2 Schematic drawing showing the location of the acquired micrographs.

(a-b) Prelimbic (PL), infralimbic (IL) cortex mosaic images of layer II & III were acquired around bregma +3.24mm (depicted by the blue / purple vertical line) in the medial prefrontal cortex (blue / purple square) Drawings depicted from (Paxinos & Watson, 2007). (c) Sample of section labelled for

261

pMAPK expressing neurons, 5-HT boutons and VGLUT3 puncta and processed using surface and spot detection functions in Imaris. (d) pMAPK surface. (e) 5-HT surface. (f) Masked VGLUT3 (colocalized with 5-HT boutons). Scale bar 50µm.

8.3.7 Analysis

Analyses of behavioural conditions were performed by two-way ANOVAs.

Post hoc Bonferroni correction was used in all cases to reduce type one errors

synonymous with multiple comparisons. A p value ≤ 0.05 was stated as significant,

*: p ≤ 0.05, **: p ≤ 0.01, ***: p ≤ 0.0001, ****: p ≤ 0.0001. All statistical analyses

were generated with GraphPad Prism 7 (GraphPad Software Co., CA, USA) and

values are expressed as the mean ± standard error of the mean (SEM). The

quantification of pMAPK labelling, 5-HT-/VGLUT3+, 5-HT+/VGLUT3- and 5-

HT+/VGLUT3+ boutons < 1µm from pMAPK+ neurons (Fogarty, Hammond,

Kanjhan, Bellingham, & Noakes, 2013) was analysed using one-way ANOVAs again

with Bonferroni correction. Excluded cases included subjects that failed to acquire a

fear memory during conditioning (as defined by the fear to tone test) (7dF n = 1, 1dF

n = 1), failed to adequately perfuse (as perfusion without saline may result in

thromboembolism) (RE n = 2, 7dF n=1, 1dF n=1), or sections that sustained

significant damage during processing resulting in an inability to analyse the region of

interest in the PFC (RE n = 3, 7dF n = 1, 1dF n = 3). Furthermore, statistical outliers

were identified and excluded using the ROUT method of statistical outlier

identification in GraphPad Prism 7. The ROUT method combines robust regression

and outlier removal, and is used to fit a curve not influenced by outliers. The

residuals are analysed using a test adapted from the False Discovery Rate approach

of testing for multiple comparisons. The outliers are then removed and ordinary

least-squares regression is performed on the remaining data (Graphpad, 2016).

262

8.4 Results

8.4.1 Decreased freezing due to remote extinction memory recall as compared to both recent and remote auditory fear memory recall.

For all behavioural results see Figure 2. Baseline freezing behaviour (0.2 ±

0.1%) was obtained to both context A and B during the first 2 days of the

behavioural protocol (habituation). No significant difference was observed between

contexts (0.04 ± 0.2 %, t 34 = 0.23, ns, p = 0.8218). Auditory Pavlovian fear

conditioning was conducted seven days apart with the remote extinction group and

remote fear recall group on day 3 and the recent fear recall group on day 9. A two-

way ANOVA was conducted to show freezing behaviour to be equivalent between

groups, while progressively increasing as a function of CS/US pairing. Results from

the two-way ANOVA showed no interaction to exist (F4, 48 = 0.22, ns, p = 0.93).

Follow-up main effect confirmed that no differences between behavioural conditions

existed (F2, 48 = 0.01, ns, p = 0.99), while a statistically significant difference between

CS/US pairing existed (F2, 48 = 23.22, ****, p < 0.0001). These data indicate the fear

conditioning protocol employed here produced equivalent freezing to tone across all

three groups regardless of the temporal difference.

Twenty-four hours following fear conditioning, the effect of extinction

training (day 4 - 20 presentations of CS alone in context B) on freezing levels was

tested. For analyses, 4 CS presentations were averaged and presented as one block of

extinction training. Higher % freezing was observed during the first block (CS x 4)

of extinction trials compared to the last block (remote extinction: 1st block 83.5 ±

5.5%, 4th block 10.3 ± 3.3%) on the first day of training, suggesting successful recall

of auditory conditioned fear at the beginning of the extinction session, and extinction

of fear by the end of the session. Statistical analyses across all three extinction days

revealed a significant temporal interaction between the first and last block of

263

presentations (beginning vs end of training, first day: t 54 = 11.46, ****, p < 0.0001;

second day: t 54 = 7.08, ****, p < 0.0001; third day: t 54 = 2.28, *, p = 0.0265)

indicating the effectiveness of extended extinction training over a period of three

days. Together these data suggest that extinction learning occurred and highlights the

spontaneous renewal that occurred after each 24 h period (Bouton & King, 1983;

Gregory J Quirk, 2002; Rescorla & Heth, 1975) as seen by the increase in freezing at

the beginning of each day i.e. in the first block of extinction training.

On the tenth day of the training protocol recall of fear memory or extinction

memory was tested with a 10-minute recall test consisting of 3 x CS only

presentations. Two-way ANOVA revealed, a significant difference between groups

(**, p = 0.0026). This was followed with Bonferroni’s post hoc comparisons

revealing % freezing levels were significantly lower in extinction animals as

compared with those that did not undergo extinction training upon presentation of

both the second and third tone (second tone: remote extinction vs remote fear, t 4 =

5.07, *, p = 0.0214, remote extinction vs recent fear, t 4 = 4.69, *, p = 0.0281; third

tone: remote extinction vs remote fear, t 4 = 4.68, *, p = 0.0284, remote extinction vs

recent fear, t 4 = 5.36, *, p = 0.0176) suggesting the extended extinction protocol to

be highly effective in consolidating extinction memories.

8.4.2 pMAPK activation differs for conditions when measured as neuron counts or volumetric density.

The number of neuronal cell bodies expressing pMAPK (fig. 3a) and the

density of pMAPK labelling per cubic centimetre of tissue in the sample (fig. 3b),

were compared as functions of condition. One-way ANOVA revealed no significant

264

difference between groups in the number of pMAPK expressing neuronal cell bodies

(ns, p= 0.3545). Paradoxically, one-way ANOVA revealed a significant group

difference (***, p = 0.0002) in the volumetric density of pMAPK immunoreactivity

within the tissue sample. Bonferroni-corrected post-hoc tests revealed significantly

higher density in the remote extinction group as compared to the remote fear (t34 =

2.92, *, p = 0.0187) and recent fear (t34 = 4.74, ***, p = 0.0001) groups. No

significant difference in density was noted in remote versus recent fear recall (t33 =

2.07, ns, p = 0.1393). These data suggest the measurement of volumetric density in

observations of neuroplastic change may provide a more comprehensive data set. In

Figures 3 (c – e) representative images of pMAPK expression are shown for each

condition.

265

Figure 8-3 pMAPK labelling in the prefrontal cortex. (a) The number of pMAPK expressing neuronal cell bodies was counted for each sample with no significant difference noted between conditions (F[2, 33] = 1.07, p= 0.3545). (b) A significant difference between groups (F[2, 34] = 11.43, p = 0.0002) was observed in pMAPK immunoreactivity within the tissue sample. Bonferroni-corrected post-hoc tests showed significantly higher density in the remote extinction recall group as compared to the remote fear recall group (t34 = 2.92, *, p = 0.0187) and the recent fear recall (t34 = 4.74, ***, p = 0.0001) group. No significant difference in density was observed between the remote versus recent fear recall groups (t33 = 2.07, ns, p = 0.1393). (c) pMAPK immunoreactivity due to remote extinction memory recall. (d) pMAPK immunoreactivity due to remote fear memory recall. (e) pMAPK immunoreactivity due to recent fear memory recall. Data are presented as mean ± S.E.M, *: p ≤ 0.05, **: p ≤ 0.01, ***: p ≤ 0.0001, ****: p ≤ 0.0001, one-way ANOVA followed by Bonferroni post hoc analysis. Scale bar 50µm.

8.4.3 Increased levels of colocalized VGLUT3 in 5-HT axon varicosities were located adjacent to pMAPK during extinction memory recall.

Labelling of VGLUT3 was demonstrated by scattered punctate fluorescence

(Fig. 4d-e) and by combining the 3D-reconstruction and masking functions of Imaris,

we isolated the VGLUT3 labelling that was only contained in 3D-reconstructed 5-HT

varicosities (Fig. 4a, h). We were interested in the levels of colocalized VGLUT3, 5-

HT and pMAPK relative to the volume of pMAPK in each condition. We observed a

266

significant difference between groups in the % of 5-HT/VGLUT3 boutons located

within 1µm of a pMAPK+ neuron (***, p= 0.0007). Post hoc analysis revealed a

significantly higher % in the remote extinction group as compared to the remote fear

(t34 = 3.04, *, p = 0.0137) and recent fear (t34 = 4.09, ***, p = 0.0007) groups. No

significant difference was observed in remote versus recent fear recall (t34 = 1.28,

ns, p = 0.6325). These data suggest that corelease of serotonin and glutamate may be

involved in the recall of extinction memories, in particular, neuroplasticity in the

medial prefrontal cortex.

To define the level of involvement of 5-HT+/VGLUT3+ fibres, in comparison

to other glutamatergic, cholinergic or serotonergic neurons, we quantified the levels

of 5-HT-/VGLUT3+ puncta and 5-HT+/VGLUT3- boutons separately, relative to the

density of pMAPK expression in each condition. The percentage of 5-HT-

/VGLUT3+ puncta located within 1µm of pMAPK+ neurons was significantly

different between groups (***, p= 0.0023). Once again post hoc analysis revealed a

significantly higher % in the remote extinction group as compared to the remote fear

(t35 = 3.37, **, p = 0.0055) and recent fear (t35 = 3.27, **, p = 0.0073) groups. No

difference was observed in remote versus recent fear recall (t35 = 0.02, ns, p

>0.9999).

The percentage of 5-HT+/VGLUT3- boutons located within 1µm of pMAPK+

neurons was also significantly different between groups (***, p= 0.0007).

Interestingly, post hoc analysis showed a higher % of 5-HT+/VGLUT3- boutons in

the recent fear group (t35 = 4.23, ***, p = 0.0005) when compared to the remote fear

recall group. There was no significant difference between recent fear and remote

extinction (t35 = 1.89, ns, p = 0.2020) nor between remote fear and remote extinction

groups (t35 = 2.27, ns, p >0.0891). This finding suggests greater serotonergic

267

innervation was present in the prefrontal cortex during the recall of a recent fear

memory however, this expression did not correlate with increased expression of the

neuroplasticity marker.

Together these data suggest a greater role for the corelease of serotonin and

glutamate in the mPFC due to the recall of extinction memories; however, it would

appear the glutamatergic input may have a larger functional role. Figure 4 (a-i)

shows representative images of pMAPK, VGLUT3 and 5-HT expression before and

after surface and spot functions in IMARIS were utilised.

Figure 8-4 Colocalisation of VGLUT3, 5-HT and pMAPK.

(a) Sample of section labelled for pMAPK expressing neurons, 5-HT boutons and masked VGLUT3 puncta. (b) Immunoreactive pMAPK neurons. (c) Immunoreactive 5-HT boutons. (d) Immunoreactive VGLUT3 puncta (e) IMARIS created surface of pMAPK and 5-HT with VGLUT3 spots (f) pMAPK surface. (g) 5-HT surface (h) masked VGLUT3 spots (colocalized with 5-HT boutons). (i) enlarged view of pMAPK surfaces, 5-HT surfaces and VGLUT3 spots. (j) Post hoc analysis of the percentage of colocalized VGLUT3 puncta in 5-HT boutons near pMAPK expressing neurons, relative to the density of pMAPK within the tissue revealed a significantly higher % in the remote extinction group

268

as compared to the remote fear (t34 = 3.04, *, p = 0.0137) and recent fear (t34 = 4.09, ***, p = 0.0007) groups. (k) The percentage of VGLUT3 puncta relative to the density of pMAPK expression in each condition revealed a significantly higher % in the remote extinction recall group as compared to the remote fear recall group (t35 = 3.37, **, p = 0.0055) and recent fear recall group (t35 = 3.27, **, p = 0.0073). (l) The percentage of 5-HT immunoreactive boutons located within 1µm of pMAPK+ neurons was higher in the recent fear recall group (t35 = 4.23, ***, p = 0.0005) when compared to the remote fear recall group. Data are presented as mean ± S.E.M, *: p ≤ 0.05, **: p ≤ 0.01, ***: p ≤ 0.0001, ****: p ≤ 0.0001, one-way ANOVA followed by Bonferroni post hoc analysis. Scale bars 50µm and 15µm.

8.5 Discussion

Anxiety related illness contributes to the greatest share of global mental health

disorders. Current therapeutics are associated with significant side effects, high rates

of relapse and low efficacy. Neuroplastic changes occur during both the formation

and treatment of anxiety disorders. To further explore the changes that occur during

the recall of noxious memories we investigated the correlation between 5-

HT+/VGLUT3+ boutons and the plasticity generated through the recall of extinction

memories. Research conducted by Amilhon and colleagues showed VGLUT3

increases 5-HT transmission and the loss of the transporter leads to anxiety like

behaviours (Amilhon et al., 2010). In this study, we used Pavlovian fear conditioning

to show that synaptic plasticity in the prefrontal cortex correlates with the presence

of 5-HT+/VGLUT3+ fibres during extinction memory recall, but not with recent or

remote fear recall. We further demonstrated that the density of non-serotonergic

VGLUT3+ varicosities was higher during extinction memory recall and established a

recruitment of 5-HT+/VGLUT3- immunoreactive boutons in the prefrontal cortex

during recent fear recall. These results suggest that 5-HT/glutamate release could

exert a particular type of neurotransmission underlying the recall of extinguished

memories. This study further highlights the potential therapeutic effect of 5-

HT/glutamate co-release to enhance exposure therapy in the treatment of fear-related

anxiety disorders.

Behaviour

269

The behavioural findings in this study provide evidence of effective training

and memory recall for each group. The extended extinction protocol (3 days of

exposure to tone) facilitated the development of a new extinction memory. The

percentage of freezing to tone was relatively equivalent between recent and remote

fear memory, suggesting the level of behavioural defence demonstrated by these two

groups was equivocal at time of memory recall, independent of the temporal

variation in conditioning (N. Chaaya et al., 2019; A. Jacques et al., 2017).

pMAPK expression

The behavioural studies were followed by immunohistological observations

where we compared the number of neuronal cell bodies expressing pMAPK and the

density of pMAPK labelling per cubic centimetre of tissue in the sample.

Interestingly, the number of pMAPK expressing neuronal cell bodies did not differ

between groups. In contrast to this finding, a significant group difference was found

in pMAPK immunoreactivity within the tissue sample with a greater density of

pMAPK labelling in the remote extinction group compared to the two fear recall

groups. This finding suggests the measurement of volumetric density may provide

greater scope when observing neuroplastic adaptations in the brain.

5HT+/VGLUT3+ fibres correlate with neuroplasticity-like changes during

extinction memory recall

Glutamatergic neurons transmit sensory information from the thalamus to the

frontal cortex (Romanski et al., 1993) and are therefore complicit in forming an

associative memory involving an unconditioned stimulus such as a tone (as in

auditory fear conditioning). Glutamate elicits depolarization (via sodium influx),

permitting magnesium efflux through NMDA receptors to trigger synaptic plasticity

(Nicoll, Kauer, & Malenka, 1988) and has long been implicated in extinction

270

memory retention (Cox & Westbrook, 1994; D. M. Johnson, Baker, & Azorlosa,

2000). Drug infusion studies on glutamatergic AMPA receptors in the amygdala

have been used to demonstrate the involvement of these receptors in extinction

training and retention (M. Kim, Campeau, Falls, & Davis, 1993; H. C. Lin, Mao, &

Gean, 2009). At present, therapeutic targets specific to long-term retention of

extinction memories remain elusive.

Glutamatergic neurons store the excitatory neurotransmitter glutamate in

synaptic vesicles via three isoforms of vesicular glutamate transporter, named

VGLUT1 to 3. VGLUT3 is structurally similar to VGLUT 1 and 2 but differs

through its limited distribution within the brain and its location in cholinergic

neurons in the caudate-putamen and 5-HT neurons in the brainstem raphe nuclei. Of

specific interest to this study was the role 5-HT+/VGLUT3+ axonal varicosities

played in memory recall in the prefrontal cortex.

Serotonergic function in the brain is regulated through its release and reuptake

(Wong-Lin, Wang, Moustafa, Cohen, & Nakamura, 2017). It is both complex and

contradictory in the sense that it is an inhibitory neurotransmitter that can facilitate

both increased and decreased excitability through different receptors, for review see

(Lesch & Waider, 2012). The serotonin 2A receptor has been evidenced facilitating

both consolidation and extinction of fear memories (Zhang et al., 2013) and is widely

spread throughout the rat cortex, in the dendritic spines of pyramidal

neurons (Willins, Deutch, & Roth, 1997; Yoshida et al., 2011). Selective serotonin

reuptake inhibitors (SSRI) increase 5-HT levels by inhibition of 5-HTT (the

transporter responsible for the reuptake of serotonin) and are implicated in extinction

memory formation (Homberg, 2012). It has been shown that fluoxetine treatment (an

SSRI antidepressant) dampens the reactivation of the original fear memory, thereby

271

enabling strengthening or stability of the extinction memory (Deschaux, Motanis,

Spennato, Moreau, & Garcia, 2011), suggesting the important role serotonin plays.

This accumulated data on serotonin and glutamate is in accordance with our

findings of increased neuroplasticity (indicated by pMAPK expression) in

conjunction with increased 5-HT+/VGLUT3+ axonal varicosities in the prefrontal

cortex after extinction memory recall. Our data is further supported by previous

research conducted by Karpova and colleagues (Karpova et al., 2011) demonstrating

increased extracellular 5-HT in the IL cortex (of the mPFC), leads to neuroplastic

changes via alterations to the peri neuronal nets, thereby contributing to the erasure

of fear memory (Karpova et al., 2011).

5HT+/VGLUT3- fibres present during recent fear recall

Kawahar and colleagues (Kawahara, 1995) and Suzuki and colleagues

(Suzuki, Ishigooka, Watanabe, & Miyaoka, 2002) undertook micro dialysis studies

that established a role for serotonin in the recall of conditioned fear in the amygdala.

Further to this, serotonin does not interact with NMDA receptors to mediate

plasticity, but has been shown to inhibit glutamate mediated sensory input to the

lateral amygdala (via GABA interneurons) demonstrating its role in defence against

fear, as would be required during recent fear recall (Stutzmann & LeDoux, 1999;

Yokoyama et al., 2005). Our studies expand on this accumulated information by

demonstrating increased serotonergic innervation during recent (but not remote) fear

memory recall in the prefrontal cortex, in association with low levels of the

neuroplasticity marker pMAPK. Previous studies from our lab have shown divergent

cell layer involvement in the prefrontal cortex during recent and remote fear memory

recall (Angela Jacques et al., 2019), therefore it is not surprising that the same

neurotransmitter is not involved at the same level in both temporal recall points.

272

5-HT1A and 5-HT2A receptor subtypes are abundant in the prefrontal cortex

and involved in the regulation of mood (Caliendo, Santagada, Perissutti, & Fiorino,

2005; Weisstaub et al., 2006). Highly expressed in the deep layers of the rodent

cerebral cortex, 5-HT2A receptors are particularly located in Layer V (Pazos &

Palacios, 1985), which we previously showed is activated during recent but not

remote fear conditioning (Angela Jacques et al., 2019). As our current findings of

increased serotonergic innervation after fear recall do not correlate with increased

neuroplasticity (as measured through pMAPK activation) we suggest 5-HT is

exerting an inhibitory influence in this instance. However, the current study observed

tissue from layer 2 of the prefrontal cortex only. It may prove interesting to examine

the deeper layers of the prefrontal cortex with regard to the 5-HT+/VGLUT3+ inputs.

Svenningsson and colleagues have reported dopamine- and cAMP regulated

phosphoprotein of M(r) 32,000 (DARPP-32) is involved in both the biochemical and

behavioural consequences of 5-HT activation (Yoshida et al., 2011). Further to this,

serotonin-induced DARPP-32 activation, mediated by 5-HT/glutamate co-

transmission may also warrant investigation.

Technical considerations

Future studies would benefit by examining a no recall group of animals to

establish a baseline for associative learning, a recent extinction group for comparison

to remote extinction and a group that has undergone extinction training without

recall, to establish changes rought through consolidation only. It is possible the

colocalisation correlating with reconsolidation observed here may also occur after

memory consolidation. A control marker for the colocalisation of 5-HT+/VGLUT3+

inputs to markers that do not indicate plasticity may provide further relevant

information.

273

Conclusion

In conclusion, it is apparent that 5HT+/glutamate co-release plays a role

during the recall of extinction memories in the prefrontal cortex of rats and further

studies are necessary to determine how neuromodulators drive the associative

learning that encodes extinction memory recall to enhance treatment of anxiety

disorders.

274

General Discussion

This dissertation details findings that advance our knowledge of the

neuroplasticity involved in the development of emotional memories. This chapter

presents a summary of results, followed by the limitations of the research conducted

within. Possible future directions of this research are detailed, ensued by concluding

remarks.

275

9.1 Summary of findings

Initially, a topographic and statistical method to accurately map, quantify and

compare the spatial properties of neural networks was developed. This technique

contributes to our knowledge of the organization and structure of functional neural

circuits by allowing visualization of the spatial organization of the neuroplastic

changes incurred through memory formation.

The initial aim of identifying precise biological mechanism of fear memory

acquisition and extinction was achieved in part by identifying specific patterns of

neuronal activation. In Chapter 3 the developed method was utilized to anatomically

define subregions of the basolateral amygdala undergoing localised neuroplastic

change due to contextual and context-removed auditory fear. The key finding was

that modification to fear memory conditioning protocols lead to varied spatial

patterns of IEG activation in the BLC. Interestingly this was not represented by

alterations in freezing behaviour, leading us to believe neuroplastic changes in the

BLC may be regulated by the hippocampus. Furthermore, auditory fear memories

that are formed without background contextual fear memories were found to reduce

BLC activation also suggesting hippocampal modulation.

Emotional memories such as the pathological fear suffered by those with PTSD

occur due to impairment in the mechanism required to determine the appropriate

behavioural response to threatening situations. Statistically verified neuronal micro-

mapping can advance our knowledge of the elaborate composition of the brain

during the formation of extinction memories that may ultimately override this

response. Chapter 4 investigated and analysed the spatial and temporal micro

topography of pMAPK+ neurons involved in encoding recent, remote and extinction

memory recall within prefrontal cortex and amygdala. A small consistent (stable)

276

population of neurons in the ventrolateral portion of the lateral amygdala (LAvl)

were found to be activated only following extinction memory recall and a reduction

of activation in the dorsolateral portion of the lateral amygdala after remote fear

recall was visualised, in comparison to recent recall. Importantly, time dependent

activation was noted in the PL as expression in diverse cell layers. Superficial layers

of the prelimbic cortex were consistently activated after both recent and remote

memory recall, as opposed to the deeper layers which were only activated through

recent recall of the auditory fear memory. This data suggests changed PL

involvement due to the remote recall of fear memories.

There are a multitude of neurobiological factors that may influence

neuroplasticity due to fear, however, to our knowledge, this is the first time

adaptations to microglia as a result of fear, have been assessed. Therefore chapters 5,

6 and 7 were dedicated to observations of phenotypical changes microglia underwent

when subjected to contextual fear, auditory fear and extinction memories.

Examination of microglial morphology was conducted for short and long-term

consolidation of memories as well as recent and remote recall. Findings included

significant morphological change and an increase in number of microglia present in

the PFC due to recent fear memory recall suggesting some form of activation occurs

to the brains equivalent of immune cells, during fear states. Additionally, reduced

neuronal activity as demonstrated by lower c-Fos and Arc expression was noted in

the extinction memory recall group, correlated with reduced densities of Iba1+

microglia.

Examination of the rat dorsal hippocampus and amygdala as a result of

contextual fear conditioning revealed that microglia released more BDNF in the

dentate gyrus as compared to controls. Surprisingly, unpaired fear conditioning

277

resulted in equivalent levels of fear behaviour though did not alter microglia, BDNF

or pCREB number in either the dorsal hippocampus or lateral amygdala. Contextual

fear conditioning resulted in microglia altering morphology to become more

amoeboid-like within the dentate gyrus. This is a common response to traumatic

brain injury and infection. Further to these findings, a spatial relationship between

pMAPK+ neurons and microglia was noted in the prelimbic cortex, in which

microglia located near pMAPK expressing neurons correlated with enhanced group

differences. These data indicate alterations in PL may occur for up to two weeks post

contextual fear conditioning.

The two overarching neurotransmitters reported as contributors to anxiety, fear

memory formation, microglial activation and synaptic plasticity are serotonin and

glutamate with recent evidence suggesting 5-HT/glutamate co-transmission plays a

role in both reward and emotion processing. In chapter 8 the colocalisation of

serotonin, serotonin transporters and vesicular glutamate type 3 transporters were

mapped in brain regions involved in emotional memories. Results revealed that 90%

of 5-HT boutons are immunoreactive for the SERT and furthermore, SERT and

VGLUT3 preferentially co-localize within 5-HT varicosities. Chapter 9 built on this

finding by exploring the impact of 5-HT and Vglut3 with regards to the

neuroplasticity of fear memory, as indicated by pMAPK activation.

9.2 Significance

The extinction of pathological fear is central to the treatment of several anxiety

disorders and involves a learning process, which forms the basis of exposure therapy.

At the present time pharmacological interventions target neurotransmitters in a

whole brain manner, resulting in adverse side effects and high rates of relapse. This

thesis builds on previous research, which has identified a micro-topography of fear

278

memory encoding in the amygdala by defining the precise brain subregion necessary

for the extinction of auditory fear. In addition to this anatomical finding corelease of

the neurotransmitters glutamate and serotonin have been correlated with the recall of

these extinction memories. As technology and pharmaceutical treatment advances

this basic science has defined both the precise target region and molecules involved

in the extinction of fear.

9.3 Advanced considerations

The technique developed to define functional neuronal topography has both

auspicious and inauspicious traits. While providing ease with which to identify

patterns of neuronal activation during set behaviours, once identified the clinical

relevance becomes limited as there are currently no pharmacotherapeutics designed

to target such small regions of functionality. Both the time required to produce

comprehensive quantitative results using this method and the current standardised

brain atlases available to guide the process, are somewhat prohibitive. As mentioned

in chapter 2 there is also a degree of brain tissue variability and individual judgment

required to align sections with contours. There is exigency for an improved

methodological approach to contouring brain nuclei and perhaps an atlas detailing

such contours would prove beneficial. Despite these limitations chapters 2, 3 and 4

provide essential and comprehensive conclusions with regards to the spatial and

temporal alterations in functional neuronal populations associated with varied fear

consolidation, recall and extinction.

Conjecture surrounding the translatability of animal models is a constant

technical consideration in behavioural neuroscience. Fear conditioning paradigms are

used to model traumatic events leading to PTSD in humans (J. LeDoux, 2000;

Maren, 2011), however, expression of neuroplasticity markers may differ

279

considerably between species and procedures fail to capture the emotional

complexity of PTSD (Beckers, Krypotos, Boddez, Effting, & Kindt, 2013; Newman,

Riggs, & Roth, 1997).

The analysis of microglial cell bodies described in chapters 4, 5 and 6 was

performed through volume and area quantification as opposed to the perimeter and

diameter measurements utilized in most studies to date. This difference may account

for the lack of adherence to the definitive morphological stereotypes previously

described in the literature. The observation that microglia may be phagocytic in

nature suggests the elongation of cell bodies may routinely occur and therefore skew

cell body measurement data. In chapter 4 comparisons were made between microglia

after the recall of fear and extinction memories. Although the variation in

ramification was enlightening, a box control group may have supplemented

informative comparison by establishing a baseline of unreactive microglial

morphology.

The research conducted to generate chapters 8 and 9, highlight the need for

lower cost high resolution imaging. The costs sustained to procure quality images

prohibited the size of the brain region studied. The finding of increased dendritic

presence (as noted by pMAPK expression) in the case of extinction memory recall,

as opposed to the number of actual cell bodies counted, indicates the necessity of

improved quantitative techniques in the study of neuroplasticity. As plasticity occurs

at a synaptic level it does not appear as imperative to know whether cell number

itself has increased. Even in studies measuring the size and number of dendritic

spines present, suggestions pertaining to large brain regions must be inferred from

small samples of tissue.

280

9.4 Future directions

Further to the topographic density maps described within this dissertation,

mapping of neurotransmitter release may prove to be more clinically relevant,

particularly in association with addiction and depression studies and research

involving pharmacological interventions. Examination of additional protein synthesis

markers, behavioural models, time points and brain nuclei may also result in

instrumental advances in knowledge surrounding the neuroplasticity of emotional

memories.

Conceivable future avenues of inquiry into the role of microglial activation in

emotional situations are virtually limitless due to the infancy of this field of study.

Markers capable of characterising the activation state of microglia at time of memory

recall should be foremost in the research agenda. Investigation of the inflammatory

status and molecular correlates of microglial activation as a result of various

emotional states would also enhance development of potential therapeutics.

At the opposite end of the spectrum is an almost endless supply of research into

the action of neurotransmitters serotonin and glutamate. However, this research is

often confounded by the complexity of neuronal networks. Serotonin has been touted

as the “mediator that spans evolution” and many scientists have dedicated their life’s

work to uncovering its secrets (Pilowsky, 2018). To date, 16 different serotonin

receptors of varying excitatory and inhibitory transmission have been uncovered and

characterised (Ohno, 2019), with perhaps many more to come. The frequency of 5-

HT/glutamate co-release sites within brain regions involved in emotional regulation

would suggest moderation of neuroplasticity and raise possibilities of potential

mechanisms for the control of that neuroplasticity (Kauer and Malenka, 2007).

281

Further research is required to discover the role of co-released serotonin and

glutamate on the neuroplasticity induced by emotional states.

9.5 Concluding remarks

In conclusion, understanding neural network organization may lead to the

ability to predict memory and behaviour from functional neural networks while

microglial activation may play an important role in the facilitation of this neuronal

organisation. Knowledge and understanding of anxiety related disorders and the

neuroplasticity underlying consequential behaviours, expands with each vital new

piece of research conducted.

283

Appendix A:

Supplementary Material for chapter 7.

Table 2 One-way ANOVA analysis of the volumetric density of 5-HT/VGLUT3 in the mouse forebrain. (Significant changes are highlighted in light grey).

Tukey'smultiplecomparisonstest

MeanDiff.

95.00%CIofdiff.

Significant?

Summary

AdjustedPValue

PrL1‐3vs.PrL4‐6 6.418 ‐6.531to No ns 0.9009

PrL1‐3vs.NACc ‐1.444 ‐14.39to No ns >0.9999

PrL1‐3vs.NACs ‐1.718 ‐14.67to No ns >0.9999

PrL1‐3vs.pNACs 41.99 29.04to Yes **** <0.0001

PrL1‐3vs.CPU ‐ ‐13.33to No ns >0.9999

PrL1‐3vs.BNST 10.8 ‐2.149to No ns 0.2037

PrL1‐3vs.LS 19.31 6.363to Yes *** 0.0002

PrL1‐3vs.BLA 6.46 ‐6.489to No ns 0.8965

PrL1‐3vs.CeA 16.48 3.535to Yes ** 0.0026

PrL1‐3vs.CA1 1.724 ‐11.22to No ns >0.9999

PrL1‐3vs.CA2 6.995 ‐5.955to No ns 0.8317

PrL1‐3vs.CA3 8.201 ‐4.748to No ns 0.6304

PrL1‐3vs.DG 6.255 ‐6.694to No ns 0.9165

PrL4‐6vs.NACc ‐7.861 ‐20.81to No ns 0.6925

PrL4‐6vs.NACs ‐8.135 ‐21.08to No ns 0.6426

PrL4‐6vs.pNACs 35.57 22.62to Yes **** <0.0001

PrL4‐6vs.CPU ‐6.8 ‐19.75to No ns 0.8574

PrL4‐6vs.BNST 4.383 ‐8.566to No ns 0.9956

PrL4‐6vs.LS 12.89 ‐0.05477to No ns 0.052

PrL4‐6vs.BLA 0.0428 ‐12.91to No ns >0.9999

PrL4‐6vs.CeA 10.07 ‐2.882to No ns 0.3005

PrL4‐6vs.CA1 ‐4.693 ‐17.64to No ns 0.9916

PrL4‐6vs.CA2 0.577 ‐12.37to No ns >0.9999

PrL4‐6vs.CA3 1.783 ‐11.17to No ns >0.9999

PrL4‐6vs.DG ‐ ‐13.11to No ns >0.9999

284

NACcvs.NACs ‐ ‐13.22to No ns >0.9999

NACcvs.pNACs 43.43 30.48to Yes **** <0.0001

NACcvs.CPU 1.061 ‐11.89to No ns >0.9999

NACcvs.BNST 12.24 ‐0.705to No ns 0.0825

NACcvs.LS 20.76 7.807to Yes **** <0.0001

NACcvs.BLA 7.904 ‐5.045to No ns 0.6848

NACcvs.CeA 17.93 4.979to Yes *** 0.0007

NACcvs.CA1 3.168 ‐9.781to No ns 0.9998

NACcvs.CA2 8.438 ‐4.511to No ns 0.5859

NACcvs.CA3 9.645 ‐3.305to No ns 0.367

NACcvs.DG 7.699 ‐5.25to No ns 0.7211

NACsvs.pNACs 43.7 30.75to Yes **** <0.0001

NACsvs.CPU 1.335 ‐11.61to No ns >0.9999

NACsvs.BNST 12.52 ‐0.431to No ns 0.0682

NACsvs.LS 21.03 8.081to Yes **** <0.0001

NACsvs.BLA 8.178 ‐4.771to No ns 0.6346

NACsvs.CeA 18.2 5.253to Yes *** 0.0005

NACsvs.CA1 3.442 ‐9.507to No ns 0.9996

NACsvs.CA2 8.712 ‐4.237to No ns 0.5342

NACsvs.CA3 9.919 ‐3.031to No ns 0.323

NACsvs.DG 7.973 ‐4.976to No ns 0.6724

pNACsvs.CPU ‐42.37 ‐55.32to‐ Yes **** <0.0001

pNACsvs.BNST ‐31.18 ‐44.13to‐ Yes **** <0.0001

pNACsvs.LS ‐22.67 ‐35.62to‐ Yes **** <0.0001

pNACsvs.BLA ‐35.52 ‐48.47to‐ Yes **** <0.0001

pNACsvs.CeA ‐25.5 ‐38.45to‐ Yes **** <0.0001

pNACsvs.CA1 ‐40.26 ‐53.21to‐ Yes **** <0.0001

pNACsvs.CA2 ‐34.99 ‐47.94to‐ Yes **** <0.0001

pNACsvs.CA3 ‐33.78 ‐46.73to‐ Yes **** <0.0001

pNACsvs.DG ‐35.73 ‐48.68to‐ Yes **** <0.0001

285

CPUvs.BNST 11.18 ‐1.766to No ns 0.163

CPUvs.LS 19.69 6.745to Yes *** 0.0001

CPUvs.BLA 6.843 ‐6.106to No ns 0.8519

CPUvs.CeA 16.87 3.918to Yes ** 0.0018

CPUvs.CA1 2.107 ‐10.84to No ns >0.9999

CPUvs.CA2 7.377 ‐5.572to No ns 0.7746

CPUvs.CA3 8.583 ‐4.366to No ns 0.5585

CPUvs.DG 6.638 ‐6.312to No ns 0.877

BNSTvs.LS 8.512 ‐4.438to No ns 0.572

BNSTvs.BLA ‐4.34 ‐17.29to No ns 0.996

BNSTvs.CeA 5.684 ‐7.265to No ns 0.9582

BNSTvs.CA1 ‐9.076 ‐22.03to No ns 0.4666

BNSTvs.CA2 ‐3.806 ‐16.75to No ns 0.9989

BNSTvs.CA3 ‐2.6 ‐15.55to No ns >0.9999

BNSTvs.DG ‐4.545 ‐17.49to No ns 0.9937

LSvs.BLA ‐12.85 ‐25.8to No ns 0.0537

LSvs.CeA ‐2.827 ‐15.78to No ns >0.9999

LSvs.CA1 ‐17.59 ‐30.54to‐ Yes *** 0.0009

LSvs.CA2 ‐12.32 ‐25.27to No ns 0.0784

LSvs.CA3 ‐11.11 ‐24.06to No ns 0.1702

LSvs.DG ‐13.06 ‐26.01to‐ Yes * 0.0462

BLAvs.CeA 10.02 ‐2.925to No ns 0.3069

BLAvs.CA1 ‐4.736 ‐17.69to No ns 0.9909

BLAvs.CA2 0.5341 ‐12.41to No ns >0.9999

BLAvs.CA3 1.74 ‐11.21to No ns >0.9999

BLAvs.DG ‐ ‐13.15to No ns >0.9999

CeAvs.CA1 ‐14.76 ‐27.71to‐ Yes * 0.0119

CeAvs.CA2 ‐9.49 ‐22.44to No ns 0.3931

CeAvs.CA3 ‐8.284 ‐21.23to No ns 0.6149

CeAvs.DG ‐10.23 ‐23.18to No ns 0.2769

286

CA1vs.CA2 5.27 ‐7.679to No ns 0.9771

CA1vs.CA3 6.477 ‐6.473to No ns 0.8948

CA1vs.DG 4.531 ‐8.419to No ns 0.9939

CA2vs.CA3 1.206 ‐11.74to No ns >0.9999

CA2vs.DG ‐ ‐13.69to No ns >0.9999

CA3vs.DG ‐1.946 ‐14.9to11 No ns >0.9999 Table 3 One-way ANOVA analysis of the volumetric density of 5-HT/VGLUT3 in the mouse forebrain (Significant changes are highlighted in light grey).

Sidak'smultiplecomparisonstest

MeanDiff.

95.00%CIofdiff.

Significant?

Summary

AdjustedPValue

BLAvs.CeA 0.0888 ‐0.2917to No ns 0.9999

BLAvs.BNST 0.6377 0.3401to Yes **** <0.0001

BLAvs.NACc 1.053 0.7212to Yes **** <0.0001

BLAvs.NACs 0.0501 ‐0.3138to No ns >0.9999

BLAvs.CPU 1.131 0.8124to Yes **** <0.0001

BLAvs.CA1 ‐ ‐0.4975to No ns >0.9999

BLAvs.CA2 ‐ ‐0.7175to No ns 0.5757

BLAvs.CA3 0.0161 ‐0.4178to No ns >0.9999

BLAvs.DG 1.022 0.588to Yes **** <0.0001

BLAvs.LS ‐ ‐0.5869to No ns 0.8206

BLAvs.PrL1‐3 0.9156 0.5752to Yes **** <0.0001

BLAvs.PrL4‐6 0.9743 0.634to Yes **** <0.0001

CeAvs.BNST 0.5489 0.206to Yes **** <0.0001

CeAvs.NACc 0.9642 0.5913to Yes **** <0.0001

CeAvs.NACs ‐ ‐0.4405to No ns >0.9999

CeAvs.CPU 1.042 0.681to Yes **** <0.0001

CeAvs.CA1 ‐ ‐0.6185to No ns 0.9959

CeAvs.CA2 ‐ ‐0.8385to No ns 0.2596

CeAvs.CA3 ‐ ‐0.5388to No ns >0.9999

CeAvs.DG 0.9331 0.467to Yes **** <0.0001

287

CeAvs.LS ‐ ‐0.7121to No ns 0.4462

CeAvs.PrL1‐3 0.8267 0.4462to Yes **** <0.0001

CeAvs.PrL4‐6 0.8855 0.5049to Yes **** <0.0001

BNSTvs.NACc 0.4153 0.1275to Yes *** 0.0003

BNSTvs.NACs ‐ ‐0.9119to‐ Yes **** <0.0001

BNSTvs.CPU 0.4931 0.2209to Yes **** <0.0001

BNSTvs.CA1 ‐ ‐1.103to‐ Yes **** <0.0001

BNSTvs.CA2 ‐ ‐1.323to‐ Yes **** <0.0001

BNSTvs.CA3 ‐ ‐1.023to‐ Yes **** <0.0001

BNSTvs.DG 0.3842 ‐0.01709to No ns 0.0742

BNSTvs.LS ‐ ‐1.187to‐ Yes **** <0.0001

BNSTvs.PrL1‐3 0.2779 ‐0.01981to No ns 0.0917

BNSTvs.PrL4‐6 0.3366 0.03895to Yes * 0.0131

NACcvs.NACs ‐1.003 ‐1.359to‐ Yes **** <0.0001

NACcvs.CPU 0.0778 ‐0.2313to No ns 0.9997

NACcvs.CA1 ‐1.117 ‐1.544to‐ Yes **** <0.0001

NACcvs.CA2 ‐1.337 ‐1.764to‐ Yes **** <0.0001

NACcvs.CA3 ‐1.037 ‐1.464to‐ Yes **** <0.0001

NACcvs.DG ‐ ‐0.4583to No ns >0.9999

NACcvs.LS ‐1.259 ‐1.632to‐ Yes **** <0.0001

NACcvs.PrL1‐3 ‐ ‐0.4692to No ns 0.9699

NACcvs.PrL4‐6 ‐ ‐0.4104to No ns 0.9998

NACsvs.CPU 1.081 0.7373to Yes **** <0.0001

NACsvs.CA1 ‐ ‐0.5663to No ns 0.9997

NACsvs.CA2 ‐ ‐0.7863to No ns 0.3808

NACsvs.CA3 ‐ ‐0.4865to No ns >0.9999

NACsvs.DG 0.9718 0.5192to Yes **** <0.0001

NACsvs.LS ‐ ‐0.6582to No ns 0.6123

NACsvs.PrL1‐3 0.8655 0.5016to Yes **** <0.0001

NACsvs.PrL4‐6 0.9242 0.5604to Yes **** <0.0001

288

CPUvs.CA1 ‐1.194 ‐1.611to‐ Yes **** <0.0001

CPUvs.CA2 ‐1.414 ‐1.831to‐ Yes **** <0.0001

CPUvs.CA3 ‐1.115 ‐1.532to‐ Yes **** <0.0001

CPUvs.DG ‐ ‐0.5258to No ns 0.9995

CPUvs.LS ‐1.337 ‐1.698to‐ Yes **** <0.0001

CPUvs.PrL1‐3 ‐ ‐0.5337to No ns 0.5214

CPUvs.PrL4‐6 ‐ ‐0.4749to No ns 0.8996

CA1vs.CA2 ‐0.22 ‐0.7306to No ns 0.9595

CA1vs.CA3 0.0797 ‐0.4308to No ns >0.9999

CA1vs.DG 1.086 0.5749to Yes **** <0.0001

CA1vs.LS ‐ ‐0.6089to No ns 0.9978

CA1vs.PrL1‐3 0.9792 0.5452to Yes **** <0.0001

CA1vs.PrL4‐6 1.038 0.604to Yes **** <0.0001

CA2vs.CA3 0.2998 ‐0.2108to No ns 0.7304

CA2vs.DG 1.306 0.7949to Yes **** <0.0001

CA2vs.LS 0.0772 ‐0.3889to No ns >0.9999

CA2vs.PrL1‐3 1.199 0.7652to Yes **** <0.0001

CA2vs.PrL4‐6 1.258 0.824to Yes **** <0.0001

CA3vs.DG 1.006 0.4952to Yes **** <0.0001

CA3vs.LS ‐ ‐0.6886to No ns 0.9169

CA3vs.PrL1‐3 0.8994 0.4655to Yes **** <0.0001

CA3vs.PrL4‐6 0.9582 0.5243to Yes **** <0.0001

DGvs.LS ‐1.228 ‐1.694to‐ Yes **** <0.0001

DGvs.PrL1‐3 ‐ ‐0.5403to No ns 0.9998

DGvs.PrL4‐6 ‐ ‐0.4815to No ns >0.9999

LSvs.PrL1‐3 1.122 0.7413to Yes **** <0.0001

LSvs.PrL4‐6 1.181 0.8001to Yes **** <0.0001

PrL1‐3vs.PrL4‐6 0.0587 ‐0.2816to No ns >0.9999

Table 4 One-way ANOVA analysis of the relative density of 5-HT/VGLUT3 in the mouse forebrain (Significant changes are highlighted in light grey).

289

Sidak'smultiplecomparisonstest

MeanDiff.

95.00%CIofdiff.

Significant?

Summary

AdjustedPValue

BLAvs.CeA ‐3.801 ‐16.86to No ns >0.9999

BLAvs.BNST 6.116 ‐3.858to No ns 0.9282

BLAvs.NACc 8.216 ‐2.967to No ns 0.5826

BLAvs.NACs 2.695 ‐8.207to No ns >0.9999

BLAvs.CPU 6.465 ‐4.198to No ns 0.9389

BLAvs.CA1 ‐12.17 ‐27.24to No ns 0.3484

BLAvs.CA2 ‐19.15 ‐34.23to‐ Yes ** 0.0017

BLAvs.CA3 ‐17.12 ‐32.2to‐ Yes ** 0.0098

BLAvs.DG ‐7.094 ‐22.17to No ns 0.9997

BLAvs.LS ‐1.215 ‐14.27to No ns >0.9999

BLAvs.PrL1‐3 ‐8.508 ‐20.02to No ns 0.5686

BLAvs.PrL4‐6 ‐11.08 ‐22.6to No ns 0.0762

CeAvs.BNST 9.918 ‐2.585to No ns 0.3881

CeAvs.NACc 12.02 ‐1.469to No ns 0.1614

CeAvs.NACs 6.497 ‐6.759to No ns 0.9992

CeAvs.CPU 10.27 ‐2.793to No ns 0.41

CeAvs.CA1 ‐8.364 ‐25.22to No ns 0.9989

CeAvs.CA2 ‐15.35 ‐32.21to No ns 0.1325

CeAvs.CA3 ‐13.32 ‐30.18to No ns 0.3983

CeAvs.DG ‐3.292 ‐20.15to No ns >0.9999

CeAvs.LS 2.586 ‐12.49to No ns >0.9999

CeAvs.PrL1‐3 ‐4.706 ‐18.47to No ns >0.9999

CeAvs.PrL4‐6 ‐7.283 ‐21.05to No ns 0.9949

BNSTvs.NACc 2.1 ‐8.428to No ns >0.9999

BNSTvs.NACs ‐3.421 ‐13.65to No ns >0.9999

BNSTvs.CPU 0.3483 ‐9.626to No ns >0.9999

BNSTvs.CA1 ‐18.28 ‐32.88to‐ Yes ** 0.0021

BNSTvs.CA2 ‐25.27 ‐39.87to‐ Yes **** <0.0001

290

BNSTvs.CA3 ‐23.23 ‐37.83to‐ Yes **** <0.0001

BNSTvs.DG ‐13.21 ‐27.81to No ns 0.1406

BNSTvs.LS ‐7.331 ‐19.83to No ns 0.9634

BNSTvs.PrL1‐3 ‐14.62 ‐25.51to‐ Yes *** 0.0006

BNSTvs.PrL4‐6 ‐17.2 ‐28.08to‐ Yes **** <0.0001

NACcvs.NACs ‐5.521 ‐16.93to No ns 0.9994

NACcvs.CPU ‐1.752 ‐12.93to No ns >0.9999

NACcvs.CA1 ‐20.38 ‐35.83to‐ Yes *** 0.0008

NACcvs.CA2 ‐27.37 ‐42.82to‐ Yes **** <0.0001

NACcvs.CA3 ‐25.33 ‐40.79to‐ Yes **** <0.0001

NACcvs.DG ‐15.31 ‐30.76to No ns 0.0555

NACcvs.LS ‐9.431 ‐22.92to No ns 0.705

NACcvs.PrL1‐3 ‐16.72 ‐28.72to‐ Yes *** 0.0003

NACcvs.PrL4‐6 ‐19.3 ‐31.3to‐ Yes **** <0.0001

NACsvs.CPU 3.77 ‐7.133to No ns >0.9999

NACsvs.CA1 ‐14.86 ‐30.11to No ns 0.0667

NACsvs.CA2 ‐21.85 ‐37.1to‐ Yes *** 0.0002

NACsvs.CA3 ‐19.81 ‐35.06to‐ Yes ** 0.0011

NACsvs.DG ‐9.789 ‐25.04to No ns 0.8716

NACsvs.LS ‐3.91 ‐17.17to No ns >0.9999

NACsvs.PrL1‐3 ‐11.2 ‐22.94to No ns 0.0833

NACsvs.PrL4‐6 ‐13.78 ‐25.52to‐ Yes ** 0.006

CPUvs.CA1 ‐18.63 ‐33.71to‐ Yes ** 0.0026

CPUvs.CA2 ‐25.62 ‐40.7to‐ Yes **** <0.0001

CPUvs.CA3 ‐23.58 ‐38.66to‐ Yes **** <0.0001

CPUvs.DG ‐13.56 ‐28.64to No ns 0.1489

CPUvs.LS ‐7.68 ‐20.74to No ns 0.9616

CPUvs.PrL1‐3 ‐14.97 ‐26.49to‐ Yes ** 0.0011

CPUvs.PrL4‐6 ‐17.55 ‐29.07to‐ Yes **** <0.0001

CA1vs.CA2 ‐6.988 ‐25.46to No ns >0.9999

291

CA1vs.CA3 ‐4.953 ‐23.42to No ns >0.9999

CA1vs.DG 5.071 ‐13.4to No ns >0.9999

CA1vs.LS 10.95 ‐5.909to No ns 0.8533

CA1vs.PrL1‐3 3.657 ‐12.04to No ns >0.9999

CA1vs.PrL4‐6 1.081 ‐14.61to No ns >0.9999

CA2vs.CA3 2.034 ‐16.43to No ns >0.9999

CA2vs.DG 12.06 ‐6.409to No ns 0.8445

CA2vs.LS 17.94 1.078to Yes * 0.0236

CA2vs.PrL1‐3 10.64 ‐5.05to No ns 0.7728

CA2vs.PrL4‐6 8.068 ‐7.627to No ns 0.9973

CA3vs.DG 10.02 ‐8.444to No ns 0.9911

CA3vs.LS 15.9 ‐0.9559to No ns 0.0939

CA3vs.PrL1‐3 8.61 ‐7.084to No ns 0.989

CA3vs.PrL4‐6 6.034 ‐9.661to No ns >0.9999

DGvs.LS 5.879 ‐10.98to No ns >0.9999

DGvs.PrL1‐3 ‐1.414 ‐17.11to No ns >0.9999

DGvs.PrL4‐6 ‐3.991 ‐19.69to No ns >0.9999

LSvs.PrL1‐3 ‐7.293 ‐21.06to No ns 0.9947

LSvs.PrL4‐6 ‐9.869 ‐23.63to No ns 0.6446

PrL1‐3vs.PrL4‐6 ‐2.577 ‐14.89to No ns >0.9999

292

Appendix B:

The Impact of Sugar Consumption on Stress Driven,

Emotional and Addictive Behaviours.

This appendix comprises the following published article:

Jacques, A., Chaaya, N., Beecher, K., Ali, S.O., Belmer, A., Bartlett, SE. The Impact of Sugar Consumption on Stress Driven, Emotional and Addictive Behaviours.

Neuroscience and Biobehavioral Reviews. Published August, 2019. https://doi.org/10.1016/j.neubiorev.2019.05.021

Owing to the substantial comorbidity of mental illness, neurodegeneration and

the current world obesity epidemic (Lopresti & Drummond, 2013; Spielman, Little,

& Klegeris, 2014) there is also a crucial need to investigate any co-localising

mechanisms of action between neuroplastic changes and high caloric substances. A

growing body of evidence has implicated a critical role for negative neuroplasticity

in the development of these disorders (Bansal, Hellerstein, & Peterson, 2018;

Matikainen‐Ankney & Kravitz, 2018). This review summarizes neural adaptations

that influence emotion following sugar consumption.

293

10.1 Abstract

In 2016 the World Health Organization reported 39% of the world’s adult

population (over 18 y) was overweight, with western countries such as Australia and

the United States of America at 64.5% and 67.9% respectively. Overconsumption of

high fat/sugar containing food and beverages contribute to the development of

obesity. Neural plasticity that occurs as a result of long term sugar consumption has

been shown to reduce impulse control and therefore lower the ability to resist the

high fat/sugar foods contributing to the obesity epidemic. There is significant overlap

between the neural pathways involved in emotions that guide behavioural responses

to survival situations with those regulating overconsumption of highly palatable

food. This suggests that having a clearer understanding of the role of stress and

emotions in the development of obesity will lead to the development of novel

therapeutic strategies. Sucrose consumption activates the mesocorticolimbic system

in a manner synonymous with substances of abuse. There is overwhelming evidence

to support the hypothesis that sucrose consumption results in pathophysiological

consequences such as morphological neuronal changes, altered emotional processing

and modified behaviour in rodent and human models. In this comprehensive review,

we examined >300 studies investigating the interaction between sugar consumption,

stress and emotions. Preclinical and clinical trials investigating highly palatable

foods and stress, anxiety, depression and fear are reviewed. Importantly, the synergy

between sugar consumption and neurobiology is addressed. This review summarizes

the neurochemical changes and neural adaptations – including changes in the

dopaminergic system – that influence emotion and behaviour following sugar

consumption.

294

Keywords: sucrose consumption; stress; anxiety; depression; fear; obesity; addiction;

emotion; behaviour

10.2 Introduction

A sedentary lifestyle combined with a high caloric diet plays a significant role in

obesity(Organization, 2017b). In 2016 the World Health Organization reported more

than 1.9 billion adults were overweight (Organization, 2016). World obesity has

essentially tripled since 1975(Organization, 2016). Excessive sugar consumption has

been shown to be one of the leading contributors to weight gain (Malik, Pan, Willett,

& Hu, 2013). Furthermore, a diet high in sugars has been linked to cognitive

impairments, negative neuroplasticity changes such as hippocampal dysfunction

(Kanoski & Davidson, 2011; Noble, Hsu, Liang, & Kanoski, 2017) and emotional

disorders such as anxiety and depression (S. Kim, Shou, Abera, & Ziff, 2018). High

sugar intake increases the risk of cancer, oxidative stress, inflammation, and obesity

(Makarem, Bandera, Nicholson, & Parekh, 2018), as well as impacting cognitive

function (Barnes & Joyner, 2012) and mental health (Peet, 2004). For example, a diet

higher in refined sugar has been shown to predict a worsening of schizophrenic

behaviour over a two year period (Peet, 2004). Despite the many psychological,

physical and neurological burdens of sugar overconsumption and consequent obesity,

there are no therapies directed at reducing sugar consumption, and few therapies

capable of successfully treating obesity (for review see (Kanoski & Davidson,

2011)).

Obesity, arising from overconsumption of rewarding foods such as those with a

high sugar content, may result in negative consequences, such as a loss of self-

control and subsequent poor decision-making (Beilharz, Maniam, & Morris, 2014).

This loss of control poses a significant challenge for overweight individuals

295

attempting to lose weight. The desire to eat, is regulated by brain regions known as

feeding centers, located in the arcuate nucleus of the hypothalamus. Importantly,

these regions are interconnected with the limbic system and cerebral cortex

(specifically the hippocampus and amygdala), which are responsible for the

modulation of emotions (Ahima & Antwi, 2008; C. Liu, Lee, & Elmquist, 2014;

Simon et al., 2006; Stunkard, Faith, & Allison, 2003; Sweeney & Yang, 2017;

Ulrich-Lai et al., 2010).

Sugar, artificial sweeteners and obesity

Sugar became embedded in the food chain in the late 1960s and replaced fats

to mask bitterness and make food more palatable (Bakke et al., 2018; Moss, 2013).

In the 1970’s a shift toward increased sugar-sweetened beverages became apparent

(Wolf, Bray, & Popkin, 2008). Our early ancestors obtained sugar from either fruit,

limited by seasons, or honey, protected by bees. In the last half century, sugar

consumption has tripled worldwide, partially due to the hidden use of added sugars

in processed food (Lustig, Schmidt, & Brindis, 2012). The first artificial sweetener

(saccharin) was introduced in 1879 with low production costs during wartime,

increasing its popularity (Weihrauch & Diehl, 2004). In the 1950s as sugar became

readily available, the use of sweeteners shifted to so-called ‘diet products’ with low

caloric content.

The development of obesity relies on both the hedonic, sweet taste of food in

conjunction with the negative emotional properties of food consumption (Berthoud,

Münzberg, & Morrison, 2017; Meye & Adan, 2014; Yu & Kim, 2012). Hedonic

reactions to a 10% sucrose solution (and as low as 3.4%) were tested and found to be

significantly higher in adolescent rats when compared to adult rats (Wilmouth &

Spear, 2009).

296

Sugar and emotions

Emotional eating has been show to stem from the desire to mitigate the

effects of stress (Van Strien, Frijters, Bergers, & Defares, 1986) and stress is partially

regulated by the hypothalamic-pituitary-adrenal (HPA) axis. Interestingly, activity of

the HPA axis has been shown to be reduced through the consumption of sugar

containing foods (Ulrich-Lai, Ostrander, & Herman, 2011). Following consumption,

hormones are released to reduce the feelings of stress, which also increase the desire

for comfort foods, thus perpetuating emotional eating habits (Ulrich-Lai et al., 2011;

Ursano et al., 2009).

The objective of this review is to summarize research that examines how the

consumption of sugar leads to changes in neurobiological brain function that alters

emotional states and subsequent behaviours. We will examine findings from studies

at the intersection between the consumption of sucrose and changes in the

performance of tasks with a stressful or emotional component and review

neurobiological and neurochemical mechanisms involved in addiction, stress, fear,

anxiety and depression to determine whether there are overlapping neural

mechanisms. Lastly, we will determine whether there are novel

pharmacotherapeutics and/or interventions that target these brain circuits or

neurochemical pathways to improve the current approaches for the treatment of

obesity.

10.3 Common neuronal pathways for sucrose consumption, addiction, emotions and obesity

Addiction is characterized by a difficulty to control habitual behaviour even in

the face of negative consequences (Lindgren et al., 2018). Early addiction research

focused on drugs of abuse such as alcohol, morphine and nicotine. This has since

297

been extended to include gambling, eating and more recently, sugar consumption

(Benowitz, 1988; Benton, 2010; Comings et al., 2001; Gearhardt, Corbin, &

Brownell, 2009; Jellinek, 1952; Klenowski et al., 2016; Weeks, 1962). Addiction to

substances of abuse relies on the drug binding to specific protein targets which elicits

certain physiological and behavioural responses unique to that drug(A. M. Lee &

Messing, 2008). Psychoactive drugs commonly result in rewarding sensations that

lead to repeated use and, depending on genetic susceptibility, environmental factors

and subsequent addiction(A. M. Lee & Messing, 2008). Stress has long been

associated with both the motivation to use rewarding substances and the result of not

attaining those substances (for review see (Sinha, 2008)). The negative symptoms

produced through withdrawal are common to all forms of addictive substances

including highly palatable food. These include prolonged sensitization to the

substance of choice and associative learning where environmental cues become

associated with the pleasure derived from the substance (Hebebrand et al., 2014).

These associative memories, combined with intense cravings, increase the incidence

of relapse even after sustained abstinence. These commonalities may be due to

substances of addiction utilizing the same circuitry within the brain’s

mesocorticolimbic system (Figure 1) (Baron, Blum, Chen, Gold, & Badgaiyan, 2018;

G. F. Koob & Le Moal, 2001; Volkow, Fowler, Wang, & Swanson, 2004).

Food consumption is necessary to regulate homeostatic energy balance.

However, humans also consume food for pleasure or comfort. The hedonistic desire

for palatable food is considered reward-related and may result in maladaptive or

negative neuroplasticity that can override homeostatic regulation and result in

overeating behaviours(Kenny, 2011). Reward is delivered through stimuli that

produce pleasurable or enjoyable experiences, contrasting with addiction, which

298

involves compulsive and sometimes painful behaviour, yet both share similar

neuroadaptive responses and overlapping neuronal pathways (Adinoff, 2004). The

reward or mesocorticolimbic pathway plays an influential role in what we choose to

eat as demonstrated by studies where rats appear willing to endure noxious stimuli

such as extreme cold, heat and foot-shock to procure highly palatable foods over

standard rat chow(Cabanac & Johnson, 1983; Foo & Mason, 2005; Oswald,

Murdaugh, King, & Boggiano, 2011).

Figure 10-1 Reward pathway encompassing the mesocorticolimbic distribution of dopaminergic neurons.

The neural regions of the reward pathway include the prefrontal cortex (PFC), amygdala (AMG), ventral-tegmental area (VTA) and nucleus accumbens (NAc). Each anatomical region modulates individual behaviours and contributes to general behaviours through cross-connectivity. Regions shown and behaviours listed are consistent between human and rodent brains(Belin & Dalley, 2012).

The reward pathway consists of the prefrontal cortex (PFC), amygdala (AMG),

ventral-tegmental area (VTA) and nucleus accumbens (NAc) and in accord with

drugs of abuse (Alsiö, Olszewski, Levine, & Schiöth, 2012; Sinha, 2018), is thought

299

to be stimulated by the overconsumption of sugar, thus contributing to the

development of obesity (Leigh & Morris, 2016; Murray, Tulloch, Criscitelli, &

Avena, 2016; Shariff et al., 2016). The reward pathway is highly associated with the

efflux of dopamine which regulates the motivational state of wanting or craving that

substance or behaviour (Adinoff, 2004). Once overstimulated, the pathway becomes

primed to require that particular stimuli when presented with contextual cues or

emotional stress (Adinoff, 2004). The VTA induces this sensitization while the NAc

modulates its expression through dopaminergic control (Huang, Wu, Lee, Huang, &

Chen, 2018; Xiong et al., 2018; C. Zhang et al., 2018).

Nucleus accumbens

The NAc consists of two sub regions (core and shell), each containing specific

neuronal populations fundamental in processes of motivation (Reker et al., 2018;

Yang et al., 2018), aversion(Rosas et al., 2018), fear related avoidance (Moscarello

& Maren, 2018), reinforcement learning (E. M. Anderson et al., 2018; Floresco,

Montes, Maric, & van Holstein, 2018), pleasure seeking, addiction (C. Barrientos et

al., 2018; S. Kim et al., 2018; Lam & Jadavji, 2018) and behavioural sensitization

(Nona & Nobrega, 2018). To examine motivation and the role played by α-amino-3-

hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors in the NAc, intra-

NAc infusions of AMPAR antagonists were given to obesity prone and obesity

resistant male rats (Derman & Ferrario, 2017). Compared with controls, findings

showed the obesity-prone animals exhibited cue-triggered food seeking and the

behaviour was mediated by increased surface expression of AMPA receptors in the

core of the NAc. This data suggests cue-triggered food seeking may elicit greater

motivational feedback in individuals prone to obesity, and thus play a role in driving

the compulsive over-consumption of food (Berridge, Ho, Richard, &

300

DiFeliceantonio, 2010; Derman & Ferrario, 2017). Similarly, consumption of a 10%

sucrose solution resulted in alterations in synaptic strength via AMPA receptor

trafficking, which modulates the compulsive tendencies characteristic of any drug-

seeking (X. X. Peng, Ziff, & Carr, 2011). .

Drug induced neuroplasticity has been observed in the NAc through alterations

in dendritic morphology and altered gene expression (Huang et al., 2018;

Legastelois, Botia, Coune, Jeanblanc, & Naassila, 2014; Y. Li, Acerbo, & Robinson,

2004). Similar to psychoactive substances, sugar binging has been shown to cause

repeated increases in dopamine release and altered expression of NAc Delta FosB as

demonstrated in rats with an increase in Delta Fos B expression after consumption of

a high sugar diet (Avena, Rada, Moise, & Hoebel, 2006; Wallace et al., 2008).

Dopamine release increases as a direct effect of chronic drug use and results in

postsynaptic changes of Delta FosB and CREB accumulation (Nestler & Aghajanian,

1997). The intracellular build-up of Delta FosB can alter the gene production of

receptors which may result in reinstatement during withdrawal (Nestler &

Aghajanian, 1997).This data suggest that neuroadaptations in the brain reward

pathway in obese subjects may contribute to the progression of compulsive eating.

The NAc is considered to be the main region to undergo neuroadaptation after sugar

consumption, but changes have been noted in many brain regions encompassing the

mesocorticolimbic system (De Jong, Vanderschuren, & Adan, 2016).

Amygdala

The AMG plays a key role in negative reinforcement, the progression towards

addiction and extensively in the learned associations that lead to relapse (See, Fuchs,

Ledford, & McLAUGHLIN, 2003). Neuroplastic changes occur in the AMG which

facilitate the level of dependence to a substance of abuse to move from impulsivity to

301

compulsion(G. F. Koob, 2009a). This was demonstrated by long term sucrose

consumption in rats which produced maladaptive alterations in the apical dendritic

morphology of AMG principle neurons(Shariff et al., 2017). Optogenetic stimulation

of central AMG neurons was paired with a sucrose reward to test reward incentive in

rats(M. J. Robinson, Warlow, & Berridge, 2014). Stimulation increased incentive to

choose the sucrose reward over a similar sweet alternative and amplified the degree

of effort the rats were willing to exert to obtain the reward(M. J. Robinson et al.,

2014). These findings suggest neuroplastic changes in the AMG may occur due to

sucrose consumption and increased addictive-like compulsive behaviour. The AMG

and PFC share a role in motivation, associative learning (including negative

reinforcement) (G. F. Koob, 2009a), compulsive behaviour and deficits in executive

functioning (Zehra et al., 2018). Relapsing into addictive behaviours is most likely

due to dysregulation of motivational processes. Negative reinforcement, or substance

abuse to alleviate an hedonic, anxious, irritable or dysphoric state, is characteristic of

addiction withdrawal (G. Koob, 2018).

Prefrontal cortex

The orbitofrontal cortex and anterior cingulate are associated with compulsive

cravings for drugs of abuse (Adinoff, 2004). Interestingly, neurons in the

orbitofrontal cortex of rats were shown to encode this same compulsion to seek a

sucrose reward solution (15%) (Moorman & Aston-Jones, 2014). The PFC is the

brain region responsible for executive functions such as planning and decision

making. Addiction and periods of withdrawal from substances of abuse are generally

accompanied by a loss of executive function, which is due to dysfunctional

neurocircuitry in the medial PFC (Molnar et al., 2018). Sucrose and fructose

consumption have also been linked to metabolic and electrophysiological changes in

302

the hippocampus (Hsu et al., 2015; Lemos et al., 2016), the thalamus (Jastreboff et

al., 2016) and the hypothalamus (Berthoud et al., 2017; Ulrich-Lai et al., 2011); brain

regions involved in the associative learning of cues that lead to receiving rewards (L.

C. Anderson & Petrovich, 2018; Atkins, Selcher, Petraitis, Trzaskos, & Sweatt,

1998).

10.4 Physiological and neural substrates of sugar consumption

Feeding centers linked to limbic regions relaying emotional information

influence our behaviour towards food(C. Liu et al., 2014). The current view of what

drives us to eat takes into account the drive to maintain an internal balance between

energy expenditure and consumption. However, the influence of external cues that

promise immediate reward may negate this balance(Lowe & Butryn, 2007; Stroebe,

Papies, & Aarts, 2008). Sucrose added to food provides this immediate reward, with

the human desire for sucrose or any sweet taste being comparable to the degree of

yearning and reward produced by drugs of addiction (Ahmed, Guillem, & Vandaele,

2013). Sugar or sucrose (table sugar) is a disaccharide composed of the

monosaccharides glucose and fructose (I. Smith, 1960; Southgate, 1995). Unlike

most drugs of addiction (excluding alcohol), sugar does not cross the blood brain

barrier to bind to molecular substrates/receptors on the cell surface and subsequently

alter neural plasticity. Recent research has observed astrocytes, a type of glial cell

that helps maintain the blood-brain barrier appear to sense and uptake sugar to

regulate neuronal signaling related to appetite(García-Cáceres, 2016). However, as

sugar is a food, ingestion begins in the mouth.

Sugar is initially sensed by heterodimeric G protein-coupled receptors on taste

cells located in the mouth and gut(Brown & Rother, 2012). Once in the small

intestine, sucrose is broken down into glucose and fructose; which are metabolized

303

by separate and disparate mechanisms(Southgate, 1995). Glucose increases the

absorption of fructose from the gut, whereas fructose acts as a catalyst for the uptake

and storage of glucose by the liver(Laughlin, 2014). Fructose is absorbed into the

bloodstream at a slower rate and persists for longer(Teff et al., 2004). As opposed to

glucose, fructose is not well absorbed by pancreatic beta cells and therefore

stimulates very little insulin secretion(Kyriazis, Soundarapandian, & Tyrberg, 2012).

Insulin increases satiety and subdues the reward value derived from food, suggesting

fructose may play a more complex role in the development of obesity than glucose

consumption(Grant, Christie, & Ashcroft, 1980). In addition to these observed

features fructose can increase the rate of carbohydrate oxidation after a meal, also

decreasing satiety levels(Laughlin, 2014).

Glucose and fructose have the ability to cross the blood-brain-barrier. However,

as the gut and liver rapidly break down fructose, the blood concentration levels are

generally low, resulting in only small amounts available to cross the

barrier(Gonzalez & Betts, 2018). Once across the barrier glucose signaling

mechanisms are activated. As demonstrated in animal studies(Asnaghi,

Gerhardinger, Hoehn, Adeboje, & Lorenzi, 2003; Kikkawa et al., 1987), emerging

arguments suggest fructose is produced from glucose in the brain via the polyol

pathway (glucose → aldose reductase → sorbitol → sorbitol dehydrogenase →

fructose), driven by hyperglycemia(J. J. Hwang et al., 2017). Glucose consumption

increases functional connectivity between the hypothalamus, thalamus and striatum

where fructose does not affect striatal connectivity(Cha, Wolfgang, Tokutake,

Chohnan, & Lane, 2008). Regional cerebral blood flow patterns also differ post

ingestion of these monosaccharides, with reduced flow in appetite and reward

regions of the brain after glucose consumption inclusive of the hypothalamus,

304

thalamus, insula, anterior cingulate, and striatum. Alternate to this, fructose reduces

flow in the thalamus, hippocampus, posterior cingulate cortex, fusiform, and visual

cortex(Page et al., 2013). These differences may assist in clarifying the role of

sucrose consumption in obesity and diabetes mellitus. Both human and rat studies

have shown that fructose and not glucose consumption results in conditions such as

metabolic syndrome and contributes to obesity(Nakagawa et al., 2006).

On the neuroanatomical level, regulation of energy homeostasis is dependent on

the hypothalamus(Benoit, Tracy, Davis, Choi, & Clegg, 2008), however, the NAc,

part of the reward-pathway, is pivotal in the pathophysiology of sugar

consumption(Geiger et al., 2008). Table 1 depicts the findings of several studies

examining the effect of sugar consumption on brain regions involved in the reward

pathway, defining the molecular mechanisms resultant of the neural adaptations and

the subsequent behavioural changes demonstrated by the animals.

To determine separation of neural networks engaged in choosing to eat based on

palatability or nutritional status, striatal dopamine levels were measured in sugar

consuming mice (Tellez et al., 2016). Only when the sweet solution contained energy

was the dorsal striatum, basal ganglia descending pathway recruited. Energy content

drives the release of dopamine in the NAc, however, sweetness modulated this

efflux, inhibiting dopamine release as palatability decreased(Tellez et al., 2016).

Alternatively, suppression of the sweeteners nutritional value inhibited dopamine

release in the dorsal striatum, demonstrating recruitment of different circuitry when

separating energy requirements from pleasurable taste(Tellez et al., 2016).

305

Table 5 The effects of sugar consumption on the reward pathway.

Sugar consumption causes neural adaptation to brain regions in the reward pathway

Sugar consumption alters activation of molecular substrates

Sugar consumption alters behaviour

Ventral tegmental area Nucleus accumbens Amygdala Prefrontal cortex

Dopamine (VTA-NAc)

Sensitisation (Sharpe, Clemens, Morris, & Westbrook, 2016; Wyvell & Berridge, 2001) Reward seeking (B. Gosnell, 1987; Spangler et al., 2004) (Avena & Hoebel, 2003)

Dopamine Kir2.1 CREB (NAc)(Leão, Cruz, Marin, & da Silva Planeta, 2012)

Anxiety (S. Kim et al., 2018) Depression (Avena, Rada, & Hoebel, 2008)

Serotonin

Motivation (Mathes, Gregson, & Spector, 2013)

α6β2 nAChRs α4β2 nAChRs

Anticipation of reward Motivation (Shariff et al., 2016)

AMPAR (NAc core)

Motivational feedback Cue-triggered food seeking (Derman & Ferrario, 2017)

IL-6 Leptin IL-6 TNF-α

Anxiety Fear Depression (Santos et al., 2018)

Protein oxidation (PFC)

Executive function Anxiety (Chepulis, Starkey, Waas, & Molan, 2009)

306

Delta FosB (NAc) Delta FosB (PFC)

Reward seeking (Wallace et al., 2008) Fear extinction memory retention (Baker & Reichelt, 2016)

10.4.1 The effect of sugar on the neurobiology of food consumption

The hedonic value of food, the reward associated with its consumption and the

learned external cues which trigger the desire to feed are modulated via bidirectional

circuitry between the reward pathway, hippocampus and orexigenic cells ( which

induce appetite and stimulate food intake) in the lateral hypothalamus (Kastin, 2013;

Petrovich & Gallagher, 2003; Petrovich, Hobin, & Reppucci, 2012) (see Figure 2)

and for a comprehensive review see (Murray et al., 2016). Studies activating the

lateral hypothalamus show glutamatergic inhibition of feeding (Jennings, Rizzi,

Stamatakis, Ung, & Stuber, 2013) and GABAergic stimulation of feeding (Jennings

et al., 2015). Furthermore, a subset of these GABAergic neurons project from the

VTA to the hypothalamus and express galanin (a neuropeptide) which enhances

motivation for sucrose consumption (Bocarsly, 2018; Qualls-Creekmore et al., 2017).

Lateral hypothalamic orexin neurons gain input regarding food intake from the

arcuate nucleus of the hypothalamus via endogenous melanocortin receptor

antagonists; neuropeptide Y (NPY) and agouti-related peptide-expressing neurons

(AgRP). Further, metabolic homeostasis is sensed by the lateral hypothalamus

through surrounding glucose, ghrelin and leptin levels, and this drives food seeking

behaviours (Yamanaka et al., 2000).

Hypothalamic orexinergic and anorexinergic pathways are regulated by

NPY/AgRP and POMC/CART peptides and are affected in different ways by the

307

consumption of sugar (Murray et al., 2016). The orexigenic pathway is influenced by

two specific neuronal populations within the arcuate nucleus, the first population

expresses NPY and AgRP and stimulates food intake (Benoit et al., 2008; Yamanaka

et al., 2000). Variations in both NPY and AgRP were associated with rats consuming

more chow after being provided with a 30% sucrose solution (Gaysinskaya,

Karatayev, Shuluk, & Leibowitz, 2011). Reduced expression of these peptides was

noticed following sucrose consumption, which then increased thirty to sixty minutes

later, prior to feeding (Gaysinskaya et al., 2011).

308

Figure 10-2 Regulation of feeding behaviour and food intake by central and peripheral appetite-regulating hormones and peptides.

Hypothalamic orexinergic and anorexinergic pathways and their regulation by NPY/AgRP and POMC/CART peptides, respectively, are depicted in the right-hand side pannel. Peripheral appettite-stimulating (green) and –inhibiting (red) hormones and peptides crossing the blood-brain barrier (BBB), and the organs their originate from are also represented. The effect of these hormones/peptides on the stimulation or ihibition of the orexigenic (dashed green arrows) and anorixegenic (dashed red arrows) is depicted as (+) or (-), respectively.

The link between p53 (a gene encoding a protein involved in regulating the cell

cycle) in AgRP neurons (which regulate ghrelin-induced appetite) and obesity, has

been demonstrated in mice, with its overexpression resulting in excessive weight loss

(Quiñones et al., 2018). Under normal circumstances, a lack of available nutrients

drives AgRP neurons to initiate feeding behaviour. In mice, it has been shown that

the desire to feed does necessarily involve AgRP neuronal activation when the food

has been enriched with sugar and fat (Denis et al., 2015). In this case dopamine

signaling initiates the desire to eat and feeding behaviour becomes driven by reward,

through the neural circuits involved in emotion, as opposed to the orexigenic

pathway induced by metabolic need. This appears to be the case when eating for

comfort (Denis et al., 2015). Further examinations directly link sucrose

consumption with the orexigenic pathway. In two separate studies sucrose intake was

found to be increased following NPY infusion into the lateral ventricle or AgRP

administration into the NAc shell (Badia‐Elder, Stewart, Powrozek, Murphy, & Li,

2003; Pandit et al., 2015). The effects of AgRP administration into the NAc shell is,

however, halted by pre-treatment of α-flupenthixol, a non-selective dopamine

receptor antagonist (Pandit et al., 2015). Furthermore, chemogenetic and optogenetic

manipulation of AgRP neurons activity modulates the emotional valence of feeding.

Specifically, arcuate nucleus AgRP neurons were shown to regulate the emotional

309

aspects of feeding involved with anxiety, fear-like behaviour and aggression (Burnett

et al., 2016; Joly-Amado et al., 2014; Padilla et al., 2016; Sweeney & Yang, 2017).

The second, or anorexigenic pathway contains proopiomelanocortin (POMC), a

pre-cursor to melanocortin receptor agonist α-MS, and cocaine- and amphetamine-

regulated transcript (CART). This pathway is used to inhibit the consumption of food

(Benoit et al., 2008; Lau & Herzog, 2014). Disruption of the genes encoding POMC,

CART or the melanocortin-4 receptor may lead to obesity (Graham, Shutter,

Sarmiento, & Stark, 1997). Mice fed a high fat diet and then provided with sucrose-

sweetened water showed a down-regulation of POMC mRNA expression in the

hypothalamus (Soto, 2015). Prolonged limited access to sucrose lead to decreased

activity of the anorexigenic oxytocin system (associated with satiety and termination

of feeding), in the hypothalamus of rats (Mitra, 2010). Together, these findings

suggest sugar plays a role in limiting the activation of this pathway, thus increasing

the desire to feed.

Neural pathways connecting feeding centres of the brain to the limbic system

have been identified through optogenetic activation of melanin-concentrating

hormone neurons during the intake of an artificial sweetener (sucralose) (Domingos

et al., 2013). Sucralose was found to increase striatal dopamine levels which

transposed the preference to sugar normally shown by mice into a preference for the

sweetener (Domingos et al., 2013). Further to this, Domingos and colleagues showed

that melanin-concentrating hormone neurons projecting to reward areas are necessary

for the rewarding effects of sucrose (Domingos et al., 2013).

Furthermore, the serotonergic system is indicated in the regulation of hedonic

feeding, with increases of serotonin causing decreased food intake and decreased

serotonin increasing motivation to feed (Halford, Boyland, Blundell, Kirkham, &

310

Harrold, 2010). The suggested mechanism of action for serotonin in feeding

behaviour is via AgRP and POMC neurons in the arcuate nucleus (Sohn, Elmquist, &

Williams, 2013). The serotonin transporter-linked polymorphic region (5-HTTLPR)

regulates vulnerability to stress, which is increased in cases of pathological fear, and

influences energy intake, suggesting its role on stress related overeating (Capello &

Markus, 2014). Neural adaptations resulting from sugar dependence in rodents

include alterations in dopamine and opioid receptor binding in the mesolimbic

cortex, changed expression of encephalin mRNA and modified NAc release of

dopamine and acetylcholine(Avena et al., 2008).

Leptin and Ghrelin

There are several alternate neuronal populations, pathways and brain nuclei

which drive overeating behaviour and sucrose consumption. Along with

hypothalamic orexin neurons, MCH producing neurons and leptin-receptor cells

projecting from the lateral hypothalamus to the VTA also influence hedonic feeding

and reward seeking behaviour (Bocarsly, 2018). High fructose intake results in lower

insulin levels, decreased levels of leptin and increased concentrations of ghrelin

when compared to meals high in glucose(Teff et al., 2004). Alteration to levels of

these satiety hormones (which result in the feeling of ‘fullness’ following food

consumption) may be a precursor to overeating. These results concur with those

found after intracerebroventricular injection of concentrated fructose or glucose into

the hypothalamus of rats exposed to 2-deoxy-Dglucose (2DG) (Miller, Martin,

Whitney, & Edwards, 2002). DG is an analogue of glucose that cannot be

metabolized and is known to cause increased food intake in rats by interfering with

the process of glycolysis(Fiorentini & Müller, 1975). Rats provided with fructose

injections showed enhanced food intake both in the presence and absence of 2DG,

311

where glucose suppressed 2DG induced food intake suggesting a role for brain

glucoreceptors in the control of food consumption(Miller et al., 2002).

Glucoreceptors monitor blood glucose levels and are located within the

hypothalamus. They have been shown to be stimulated by low blood sugar levels to

increase feeding behaviour (Ritter, Slusser, & Stone, 1981).

Dopaminergic neurons projecting from the VTA to the NAc are inhibited by

leptin and insulin and stimulated by ghrelin(Palmiter, 2007). Overconsumption of

sugar has been shown to increase dopamine D1 receptor binding in the NAc core and

shell, decrease dopamine D2 binding in the dorsal striatum and increase binding to

dopamine transporters in the midbrain(N. T. Bello, Lucas, & Hajnal, 2002; C

Colantuoni et al., 2001). Increases in dopamine delays the release of acetylcholine

during feeding, which postpones satiety and paves the way for overconsumption(B.

Hoebel, Avena, Bocarsly, & Rada). Alternatively, opioid receptor binding (mu-1)

was increased in the NAc shell, hippocampus, locus coeruleus and cingulate cortex

after rats were permitted intermittent binging on a 25% glucose solution(C

Colantuoni et al., 2001). Together these findings suggest the overconsumption of

sugar may sensitize dopamine D1 and opioid mu-1 receptors in a similar manner to

drugs of dependence. Table 2 depicts studies examining the effect of sugar

consumption on brain regions, molecular mechanisms and behavioural changes

involved in the orexigenic pathway.

312

Table 6 The effects of sugar consumption on the orexigenic pathway.

Sugar consumption causes neural adaptation to brain regions connected to the orexigenic pathway

Sugar consumption alters activation of molecular substrates

Sugar consumption alters behaviour

Hypothalamus Hippocampus

Dopamine Acetylcholine

Satiety (B. Hoebel et al.)

Corticosterone CRH mRNA

Stress (Kinzig, Hargrave, & Honors, 2008)

Orexin / hypocretin

Cue-induced over-consumption (Wise et al., 1995)

BDNF

Spatial learning(Molteni, Barnard, Ying, Roberts, & Gomez-Pinilla, 2002)

AgRP

Stimulation to feed(Badia‐Elder et al., 2003; Pandit et al., 2015)

5-HT2CRs

Incentive motivation(Valencia-Torres et al., 2017)

Mu-1 binding

Overconsumption(C

Colantuoni et al., 2001)

TNF-α

Neurogenesis(Van der Borght et al., 2011)

NPY

Desire to feed Learned cue

associations(Badia‐Elder et al., 2003)

313

10.4.2 Compulsive sucrose seeking

Since the 1980’s it has been known that neurons from the lateral hypothalamus

encode reward-associated cues (K. Nakamura, Ono, & Tamura, 1987) and both

feeding (Schwartzbaum, 1988) and drinking (Tabuchi et al., 2002) behaviour. In

2015 an attempt was made to define the neural circuits specific to compulsive

sucrose seeking(Nieh et al., 2015). Photoinhibition was employed to show a selective

pathway from the lateral hypothalamus to the VTA reduced compulsive sucrose-

seeking in mice without affecting normal feeding behaviour (Nieh et al., 2015). The

study found a bidirectional circuit of both inhibitory and excitatory projections from

the lateral hypothalamus to dopaminergic and GABAergic neurons in the VTA (Nieh

et al., 2015). It is possible that this circuit increases the focus and intensity of sucrose

motivation, highlighting it as a potential therapeutic target for the compulsive

overeating of sugar.

As obesity occurs due to excess caloric intake, the incentive value of the food

consumed must be considered in parallel to the satiation of hunger. 5-

hydroxytryptamine 2C receptors (5-HT2CRs) play a role in incentive motivation via

hypothalamic-VTA feeding circuits and have therefore been considered a target for

obesity treatments (Valencia-Torres et al., 2017). Lorcaserin, a 5-HT2CR agonist

was administered to mice and shown to reduce both standard food intake and the

desire for chocolate pellets with a corresponding increase in c-fos expression in VTA

5-HT2CR GABAergic neurons (not dopaminergic neurons) (Valencia-Torres et al.,

2017), validating their role in the inhibition of motivational behaviour. Similar

findings arose from observations of sucrose drinking where leptin administered to the

VTA was found to reduce food intake and knockdown of the leptin receptors

increased sucrose seeking (Clifton, 2017). Interestingly, mediation of the orexigenic

pathway between the VTA and NAc may affect behavioural changes towards reward

314

seeking for drugs of abuse, including withdrawal and relapse behaviour (Carr &

Kalivas, 2006; Kastin, 2013).

Decreased dopamine levels and simultaneous increased acetylcholine levels are

associated with behavioural signs of withdrawal from drugs of abuse (Rada &

Hoebel, 2005; Rada, Jensen, & Hoebel, 2001). Rats provided intermittent sugar

access show similar imbalances to initiate withdrawal after provision of the opioid

receptor antagonist naloxone(Carlo Colantuoni et al., 2002). An animal will crave a

substance of abuse it has been deprived, as shown by elevated operant responding

even when unrewarded and increased responding to associated drug cues

(Bienkowski et al., 2004; Lu, Grimm, Hope, & Shaham, 2004). Rats deprived of

sugar after glucose overconsumption (i.e. permitted 25% glucose for 30 min per day

for 28 days and glucose access in their home cages for an additional 11.5 h per

day) responded significantly more in operant chambers compared to controls

suggesting that sensitization of the dopaminergic system and associative learning

leads to increased motivation to seek sugar (Avena, Long, & Hoebel, 2005). The

neural adaptations required to cause the behavioural change appear long lasting as

sugar consumption ceased 2 weeks prior to testing. Similar results were shown in rat

studies of alcohol consumption (Heyser, Schulteis, & Koob, 1997).

10.4.3 Sucrose consumption and the hypothalamic-pituitary-adrenal axis

Long term stress, depending on its severity, appears to correlate with a

preference for high sugar foods, suggesting its contribution to the progression of

obesity. Chronic stress may develop through an accumulation of physical

(traumatic), chemical (dietary), physiologic (painful), psychologic (fear) or social

(lifestyle) stressors (Powers & Howley, 2007). Extended periods of stress result in

hyper activation of the hypothalamic-pituitary-adrenal (HPA) axis, a mammalian

315

stress response system involving the endocrine and central nervous systems (S. M.

Smith & Vale, 2006). Hyper activation of the HPA axis leads to increased release of

corticosteroids (steroid hormones produced in the adrenal cortex) see figure 3

(BjÖrntorp, 2000; S. M. Smith & Vale, 2006).

Activation of the HPA axis leads to increased adrenocorticotropic hormone

(ACTH) levels and increased glucocorticoids, which affect the utilization of energy

stores (Ulrich-Lai et al., 2010). This is of particular importance, as people regularly

report choosing to consume sweet tasting food (of higher caloric content) due to its

ability to enhance their mood and relieve negative emotional states (S. M. Smith &

Vale, 2006). In an observation of rats undergoing acute stress, the provision of a

sucrose solution significantly reduced levels of both ACTH and corticosterone

secretion (Ulrich-Lai et al., 2010). Furthermore, rats fed with a sweetener

(Saccharin), also showed lower HPA axis responses to acute stress suggesting the

hedonic nature of sugar may be responsible for the reduced stress response (Ulrich-

Lai et al., 2010).

Contrastingly, a study involving nineteen women found consumption of sucrose

but not the sweetener (Aspartame) resulted in reduced reduced stress-induced

cortisol (Tryon et al., 2015). Sucrose and not Aspartame also produced greater

activity in the left hippocampus (Tryon et al., 2015) suggesting the HPA negative

feedback loop may assist in generating a tendency towards sugar consumption in

people dealing with stress. The following section reviews the complex relationship

between the HPA axis, food intake, energy stores and chronic stress.

316

Figure 10-3 Hypothalamic-pituitary-adrenal axis. Stress causes the release of corticotrophin-releasing hormone and vasopressin from the hypothalamus.

These hormones are transported to the anterior pituitary. Adrenocorticotropic hormone (ACTH) is then released from the pituitary gland which stimulates the adrenal glands to release cortisol, catecholamines and aldosterone into the bloodstream. Cortisol released from the adrenal glands targets systemic organs, including the brain where it increases sucrose seeking and exerts a feedback inhibition on the release of CRH by the hypothalamus and ACTH by the pituitary gland, via corticoid receptors (mostly glucocorticoid receptors (GRs), and to a lesser extent mineralocorticoid receptors).

Any discerned acute stressor may cause engagement of the HPA axis and result

in an emotional response eliciting fear, anxiety or defensive behaviours designed to

maximize chances of survival (Ulrich-Lai et al., 2010). The relationship between

stress, cortisol levels and sugar consumption is far from understood, however, it

appears that high sugar diets play a significant role in HPA axis regulation and

binge-like eating associated with stress. When ingested, foods high in sugar release

neuropeptides, elevating mood and reinforcing desire or selective preference towards

greater amounts of high caloric food (Ulrich-Lai et al., 2011). Sucrose and the

317

sweetener saccharin have been shown to dampen the HPA axis response and this

effect is possibly why highly palatable foods are consumed to assuage unpleasant

emotions (Ulrich-Lai et al., 2011). Sugar consumption regulates stress-like behaviour

via the HPA axis however, the precise mechanisms of action are yet to be elucidated.

Contrarily, examples of chronic stress reveal elevated levels of

glucocorticoids with increased risk of developing stress related illnesses such as

depression. Long term stress appears to alter brain function through changes in

negative feedback loops from energy stores and glucocorticoid modulation of neural

circuits (for review see (Dallman et al., 2003)). As glucocorticoid levels are tightly

controlled by the HPA axis, one avenue of investigation has been to examine levels

of mRNA expression of 11βHSD-1(a mediator of glucocorticoid metabolism in the

liver) in cases of early life stress(Maniam, Antoniadis, & Morris, 2015). Interested in

the role of sugar intake on lipid homeostasis, researchers found a 53% increase in

transcriptional levels of 11βHSD-1 in animals permitted chronic sugar (25% sucrose

solution) consumption in contrast to controls revealing a correlation between sucrose

consumption, stress and increased glucocorticoid metabolism(Maniam et al., 2015).

Additionally, when humans are exposed to stress at a young age, they become more

likely to develop anxiety and depression associated with dysregulation of the HPA

axis(Heim & Nemeroff, 2001). Increased levels of cortisol (the resultant product of

the HPA axis in humans) have also been positively correlated with higher visceral fat

deposits and insulin resistance (Misra et al., 2008). Similarly high cortisol levels

(22% higher) were reported in overweight/obese adolescents after 2 or more sugar-

sweetened beverages per day, suggesting significant increases in stress hormone

were not only due to visceral fat deposits but also increased sugar

consumption(Shearrer et al., 2016). One suggested mechanism may include

318

morphological alteration of the adrenal glands as shown through consumption of

high sugar content beverages which increased the risk of metabolic syndromes such

as diabetes and lead to dysfunction of the adrenal glands(Díaz-Aguila et al., 2016).

Young adult rats fed a 30% sucrose solution for 12 weeks showed increased visceral

fat deposits and insulin resistance (Díaz-Aguila et al., 2016). Interestingly, the

adrenal glands showed histomorphological changes in the adrenal cortex and medulla

due to sucrose consumption, suggesting hyperplasia and indicators of metabolic

syndrome(Díaz-Aguila et al., 2016).

10.4.4 Developmental neuroadaptation

In rodents prenatal stress can decrease glucocorticoid and mineralocorticoid

receptor levels in the hippocampus thus decreasing receptor availability for the

feedback inhibition of corticosterone, which may explain why novel stressors cause

an increased and longer lasting corticosterone response (Henry, Kabbaj, Simon, Le

Moal, & Maccari, 1994). Pregnant rats underwent restraint stress during the third

week of gestation(Henry et al., 1994). The HPA axis and hippocampal corticosteroid

receptors in the male offspring were investigated. Plasma corticosterone was found

to be significantly higher in the prenatally-stressed rats compared to controls and the

receptor subtypes (hippocampal type I and type II corticosteroid receptors) were

decreased in the hippocampus suggesting long term changes in the HPA axis may

occur following prenatal stress(Henry et al., 1994). Numerous studies have examined

maternal stress and corticosteroids, for a comprehensive review see (Welberg &

Seckl, 2001). Fewer studies involve these neuroadaptations and diet, therefore more

direct investigation of high sugar consumption in mothers may be significant in the

study of childhood obesity.

319

Furthermore, evidence exists for the pre-programming of HPA axis hyper

activation during prenatal periods (Welberg & Seckl, 2001). In adolescents deemed

to be at higher risk of becoming obese (by evaluating their parent’s body mass

index), a higher functional magnetic resonance imaging (fMRI) signal was observed

in brain regions associated with reward learning, processing and motivation after the

consumption of high sugar milkshakes as compared to high fat milkshakes (Grace E.

Shearrer, Eric Stice, & Kyle S. Burger, 2018). Although fat provides greater energy

and therefore contributes to obesity by providing excess calories, sugar is more often

associated with modulation of habitual overeating and therefore the addiction-like

behaviour associated with obesity (Grace E. Shearrer et al., 2018).

Overconsumption during adolescence leads to long-lasting changes in the

dopaminergic reward system and may cause stimuli-induced sensitization that is

observed in adulthood (Naneix et al., 2018; T. E. Robinson & Berridge, 2008).

Stimuli-induced sensitization refers to an increased effect of the stimuli, following

repeated exposures eg. repeated exposure to a loud noise may create a sensitization

to noise, generating an enhanced response, the same being true for repeated

exposures to a drug of abuse. In drug addiction, sensitization causes changes in

dopamine transmission and delta FosB expression which contributes to increased

craving and relapse (T. E. Robinson & Berridge, 2001).

Adolescent rats permitted sucrose consumption (5% sucrose solution) for sixteen

days were later tested for motivation to seek either saccharin, maltodextrin or cocaine

(Vendruscolo, Gueye, Darnaudéry, Ahmed, & Cador, 2010). The adult rats showed a

reduction in motivation to procure saccharin and maltodextrin, however this was not

the case for cocaine (Vendruscolo et al., 2010). Further to these findings, rats

presented with a choice between a saccharin solution and intravenous cocaine

320

demonstrated the reward provided by the sweetness surpassed the desire for cocaine

(Lenoir, Serre, Cantin, & Ahmed, 2007); an intense debate remains as to whether

sugar is in fact addictive (Benton, 2010; Corsica & Pelchat, 2010; G. T. Wilson,

2010).

It has been hypothesized the addictive-like behaviours present after long-term

sucrose consumption arise from the highly palatable nature of sugar and not the

neurochemical effects of sugar itself (for reviews see (Ahmed et al., 2013; Avena,

Rada, & Hoebel, 2009; Criscitelli & Avena, 2017; B. Hoebel et al.; Kendig, 2014;

Westwater, Fletcher, & Ziauddeen, 2016)). The purpose of this review is not to

debate the argument, but to examine the role of sugar in emotional dysfunction.

10.5 Common neurochemistry underlie consumptive behaviours and emotions

Neuroimaging studies and animal models investigating the mechanisms

involved in the progression to obesity have revealed neurobiological correlates

participating in the neuroadaptations of obesity and sugar consumption (Lindgren et

al., 2018; Novelle & Diéguez, 2018). Preclinical rodent models of consumption are

useful for investigating neural regions and pathways underlying consumptive

behaviours (Bonin et al., 2018; Holgate, Shariff, Mu, & Bartlett, 2017) with

intermittent access to sucrose shown to have an effect on opioid, cholinergic and,

importantly, dopaminergic receptors (Avena et al., 2008; Carlo Colantuoni et al.,

2002; Pratt & Kelley, 2004; Shariff et al., 2016).

10.5.1 Opioids

Opioid receptors are expressed throughout the limbic system and play key roles

in the regulation of fear, happiness, anger, arousal, motivation and reward related

feeding(Levine & Billington, 2004; Nummenmaa & Tuominen, 2017). Opioid-

321

induced behaviour includes the modulation of pain, drug addiction and control of the

autonomic nervous system, which includes emotions, cognitive processes and stress

coping (Belzeaux, Lalanne, Kieffer, & Lutz, 2018; Benarroch, 2012). Opioids

influence the way we process rewards through interaction with the dopamine

system(Bray & Bouchard, 2014). The way an animal behaves towards food is also

modulated by endogenous opioid neurotransmitters, which are neural substrates that

create susceptibility to reward (Kavaliers & Hirst, 1987; Wise, 1996). Alterations in

hedonic processing are demonstrated to be highly dependent on endogenous opioid

receptors in the NAc suggesting these changes may result in the development of

reward-related disorders such as obesity (Naneix, Darlot, Coutureau, & Cador,

2016).

Opioid peptides (β-endorphin, enkephalins, and dynorphins), and activation

of their receptor types, μ (mu), δ (delta) and κ (kappa) in the NAc cause inhibition at

both pre and post synapses (Benarroch, 2012). Opioid agonists modulate pleasure,

reward and reinforcement through activation of mu and delta receptor ligands via the

mesolimbic dopamine system whereas the dysphoria associated with withdrawal

relies on kappa receptors (Benarroch, 2012; Herz, 1997; Nummenmaa & Tuominen,

2017). Mu receptor binding in the NAc is significantly increased after

cocaine(Bailey, Gianotti, Ho, & Kreek, 2005), morphine (Vigano et al., 2003) and

sucrose (C Colantuoni et al., 2001).

In the endeavor to define specific peptide and receptor involvement in the

motivation and reward seeking behaviour characteristic of hedonistic feeding, several

animal studies have combined opioid antagonists and agonists with sucrose

consumption (Beczkowska, Bowen, & Bodnar, 1992; Hayward, Schaich-Borg,

Pintar, & Low, 2006; Ruegg, Yu, & Bodnar, 1997). Opioid receptor antagonists

322

(naltrexone and naloxone), used to treat drug and alcohol dependence, were shown to

use kappa and mu2 binding sites to inhibit sucrose intake, thus demonstrating the

role of endogenous opioids in regulation of hedonic rewards in rats (Beczkowska et

al., 1992).

Substance P and the NK1 receptor system, interact with opioid receptor systems

to regulate reward related behaviours, suggesting its role in addictive

behaviour(Steensland et al., 2010). Substance P is also a neuropeptide that affects

the orexigenic pathway, via the NK1 receptor(Karagiannides et al., 2008). Its

presence in the brain assists the regulation of feeding, however, it is also located in

the stomach and small intestine suggesting it may be a potential therapeutic target for

obesity. Administration of a substance P antagonist (CJ 012,255) was shown to

prevent weight gain in obese mice following a 2w high fat diet while treatment with

CJ 012,255 in obese mice resulted in a loss of weight loss and reduction improved

insulin sensitivity, partially due to reduced food intake(Karagiannides et al., 2008).

Using a two-bottle choice drinking paradigm, knockout mouse models (lacking

either one or two opioid peptides) were used to identify opioid receptor ligands

modulating sucrose preference (Hayward et al., 2006). Enkephalin and dynorphin

were found to modulate preference to sucrose but were not considered necessary for

its consumption, supporting the role these peptides play in the motivation to consume

unnecessary calories(Hayward et al., 2006). A decrease in enkephalin mRNA

expression was observed in the NAc after rats were provided with intermittent access

to sugar (Spangler et al., 2004). The link between opioids, dopamine and motivation

for hedonic rewards is further reinforced with data showing rats preference for sweet

taste after the introduction of morphine to the NAc (Berridge, 1996; Pecina &

Berridge, 1995).

323

10.5.2 Acetylcholine

Neuronal nicotinic acetylcholine receptors (nAChRs) act as biological targets for

ethanol, nicotine (Bito‐Onon, Simms, Chatterjee, Holgate, & Bartlett, 2011) and

sucrose(Löf, Olausson, Stomberg, Taylor, & Söderpalm, 2010; Shariff et al., 2016)

and modulate the neurotransmission of GABA, glutamate, dopamine, serotonin,

acetylcholine and noradrenaline(Wu, 2009). Homeostatic dysregulation of

acetylcholine in the limbic system has been shown to modulate the motivation and

reward seeking behaviours characteristic of addiction and relapse (Aosaki, Miura,

Suzuki, Nishimura, & Masuda, 2010; B. G. Hoebel, Avena, & Rada, 2007; Rahman,

Engleman, & Bell, 2015), neurodegeneration (Alzheimer’s disease, mild cognitive

impairment) and mental illness (anxiety, depression and schizophrenia)(Posadas,

López-Hernández, & Ceña, 2013; Wu, 2009). Within the NAc, cholinergic

interneurons regulate the expression and release of enkephalin(Kelley, Baldo, &

Pratt, 2005) and many rat studies support the theory that they inhibit feeding

behaviou (Avena et al., 2006; B. G. Hoebel et al., 2007; Mark et al., 2006). As

increased serotonin reduces the motivation to feed, a study using rats to perform a

progressive ratio task and paroxetine (a selective serotonin reuptake inhibitor) found

that increased serotonergic activity decreased the appetitive-based responses to the

hedonic taste of sucrose and the aversive taste of quinine (Mathes et al., 2013).

Withdrawal from morphine increases acetylcholine levels in the NAc, while

dopamine simultaneously remains low, leading to the assertion that this mechanism

is involved in the unpleasant aspects associated with withdrawal (Avena et al., 2008;

Rada et al., 2001). Arecholine (a muscarinic agonist) inhibits feeding but can be

blocked by pirenzapine (a muscarinic acetylcholine 1 receptor antagonist),

suggesting a role for acetylcholine in food intake (Avena et al., 2006).

324

In 2010, Lof and associates utilized rats to demonstrate the role of nAChRs in

the conditioned reinforcement of lever pressing for a sucrose reward (Löf et al.,

2010). Using a selective antagonist at α7 nAChRs (methyllycaconitine) they showed

reduced lever pressing for the sucrose solution down to the level of the control,

suggesting mediation of the amount of influence the cue had on the desirability of the

sucrose (Löf et al., 2010). More recently our laboratory elucidated the effects of

varenicline (an FDA-approved drug to reduce nicotine cravings) on sucrose

consumption in rats (Shariff et al., 2016). Varenicline, a partial agonist at α4β2* and

antagonist α6β2* nAChRs subtypes, significantly reduced sucrose consumption in

both short and long term binge like (intermittent access) drinking. While α4β2*

nAChR binding sites were increased, α6β2* nAChRs were significantly decreased in

the NAc as a result of both short-term and long-term sucrose consumption (Shariff et

al., 2016). By modulating dopamine release in the NAc (Feduccia, Simms, Mill, Yi,

& Bartlett, 2014) the control of sucrose consumption by nicotinic receptors and their

subtypes may provide promising therapeutic strategies for obesity (Cocores & Gold,

2008).

10.5.3 Dopamine

Sucrose is considered a primary or natural reinforcer that does not require a

learning process to be considered desirable (Corsini, 1999). Primary reinforcers are

known to trigger dopaminergic activation in the VTA(Moore, Sabino, & Cottone,

2018; Olds & Olds, 1963; Wise, 1996), which act on the NAc via the medial

forebrain bundles (Yu & Kim, 2012). Activation of dopaminergic neurons in the

VTA modulates processes of memory, learning and motivation(Adcock, Thangavel,

Whitfield-Gabrieli, Knutson, & Gabrieli, 2006; Otmakhova, Duzel, Deutch, &

Lisman, 2013; Wise, 2004; Yang et al., 2018) as well as the intense emotions

325

associated with love, sexual desire, orgasm(Acevedo, Aron, Fisher, & Brown, 2012;

Holstege et al., 2003; Ulrich, Stauß, & Grön, 2018), fear, stress, anxiety (Bouchet et

al., 2018; Moghaddam, 2018; S. Zhang et al., 2018) and psychosis (De Nijs et al.,

2018; Knolle et al., 2018). Rat models of obesity display low levels of dopamine and

impairment in the release of dopamine (Geiger et al., 2008).

The mesocorticolimbic dopaminergic system has been implicated in modulation

of addictive reward seeking behaviour, (for review see (S. Liu, 2018)). Drugs of

abuse such as alcohol, nicotine, cocaine and amphetamines that cause increased

extracellular dopamine and subsequent heightened pleasure levels in humans (T. E.

Robinson & Berridge, 1993; Volkow, Fowler, & Wang, 2002) also cause increases in

dopamine release in the NAc in rats (Di Chiara & Imperato, 1988; Wise et al., 1995).

Commonly abused (opiates, ethanol, cocaine, amphetamine and nicotine) given to

rats resulted in increased extracellular dopamine in the NAc and the dorsal caudate

nucleus(Di Chiara & Imperato, 1988). Similar results were found in rats permitted

intermittent access to sugar with subsequent binge drinking resulting in increased

dopamine release in the NAc (Avena et al., 2006; Rada, Avena, & Hoebel, 2005).

Supporting both the hedonic potency of sucrose and its reliance on dopaminergic

regulation, orosensory stimulation in sucrose sham-fed rats resulted in increased

dopamine release within the NAc, thus increasing reward seeking behaviour (Hajnal,

Smith, & Norgren, 2004; Schneider, 1989). Compelling evidence shows drugs not

routinely abused, such as imipramine, atropine and diphenhydramine do not display

rewarding properties and do not alter synaptic dopamine levels(Di Chiara &

Imperato, 1988). Studies show an attenuated dopamine system in obesity prone rats

in the NAc, PFC and dorsal striatum and suggest the reduction in hedonic response

available to these animals is related to hyperphagia (increased appetite) and resultant

326

obesity (Geiger et al., 2008). These findings were similar to those of a human study

showing reduced levels of dopamine D2 receptors correlated with increased body

mass index in 10 obese individuals (Wang et al., 2001).

Dopamine sensitization

In 1993 Robinson and Berridge postulated the incentive-sensitization theory,

which suggested that repeated exposure to rewarding substances sensitizes the

dopaminergic system resulting in exaggerated “cue-triggered wanting” which can

transform into a compulsion to seek out an associated reward (T. E. Robinson &

Berridge, 1993; Wyvell & Berridge, 2001). Cue-reward learning depicts the learned

association between the rewarding effects of substances of abuse and environmental

cues (Bianchi et al., 2018; Tye, Stuber, De Ridder, Bonci, & Janak, 2008). This

learning contributes to obsessive overuse and the tendency to relapse after long

periods of abstinence (Bianchi et al., 2018; Ziauddeen, Farooqi, & Fletcher, 2012).

Repeated or prolonged substance abuse can cause modifications to neurotransmitter

release and alterations in synaptic strengths (Gerdeman, Partridge, Lupica, &

Lovinger, 2003). For example, rats provided with intermittent access to sucrose

display signs of dopamine sensitization through altered dopamine receptor

function(Sharpe et al., 2016).

Behavioural sensitization or hyper-locomotion is a motor response that increases

incrementally with repeated exposures to a drug. It is reflective of a hyposensitized

(or attenuated) reward pathway and is said to contribute to the craving or compulsive

seeking which characterizes addictive behaviour (Nona & Nobrega, 2018; T. E.

Robinson & Berridge, 2001). In both human and rodent studies, behavioural

sensitization has been observed as dose-dependent(Huber et al., 2018; Jing, Liu,

Zhang, & Liang, 2018) and shows considerable individual variation(T. E. Robinson,

327

1988). Brain-derived neurotrophic factor (BDNF) is believed to modulate this

behaviour through overexpression of the dopamine D3 receptor(Bordet et al., 1997)

and has been implicated in drug addiction, schizophrenia (Guillin et al., 2001;

Gurevich et al., 1997; Staley & Mash, 1996) and novelty seeking behaviour in

obese rats(Savage et al., 2014).

Animals sensitized to one drug are often cross-sensitized to other substances of

abuse as well as non-drug substances (Avena et al., 2008). This has been

demonstrated with rats exposed to amphetamines becoming sensitized to

cocaine(Pierce & Kalivas, 1995), cocaine cross-sensitizing to alcohol (Itzhak &

Martin, 1999) and cocaine cross-sensitizing with stress(Covington & Miczek, 2001).

Similar to these, short term sucrose consumption increases binding affinity to opiates

(Avena et al., 2008), resulting in cross tolerance to other opioids such as morphine,

while chronic sucrose consumption reduces the analgesic properties of morphine

(D'Anci, Kanarek, & Marks-Kaufman, 1996; Fidler, Kalman, Ziemer, & Green,

1993; Kanarek, White, Biegen, & Marks-Kaufman, 1991; Steensland et al., 2010). A

number of studies have also shown opioid agonists such as morphine to increase the

impulsivity to feed when injected systemically or directly into the NAc,

paraventricular nucleus, AMG, hypothalamus and tegmentum (B. Gosnell, 1987;

Stanley, Lanthier, & Leibowitz, 1988; Woods & Leibowitz, 1985).

Further investigations have demonstrated how sugar access can lead to cross-

sensitization of dopamine-altering drugs. Intermittent access to sugar has been found

to cross-sensitize with amphetamines up to 8 days post sugar ingestion (Avena &

Hoebel, 2003) Female rats provided with either a 10% sucrose solution or a rotation

of sucrose solution and withdrawal, displayed behavioural cross-sensitization to a

small dose of amphetamine (Avena & Hoebel, 2003; Avena et al., 2008). Similar

328

results for sucrose and cocaine (B. A. Gosnell, 2005) and sucrose and quinpirole (a

dopamine agonist) have also been found(Foley, Fudge, Kavaliers, & Ossenkopp,

2006). The rats maintained on the cyclic sucrose solution responded hyperactively to

amphetamine in comparison to controls, suggesting the binge-like sugar consumption

leads to increased amphetamine sensitivity, which may be due to neuroadaptations of

the dopamine system (Avena & Hoebel, 2003). Compelling evidence suggests

chronic stress leads to similar cross-sensitization (G. F. Koob, 2009b; Leão et al.,

2012). As globally the consumption of sugar is on the rise an interesting avenue of

investigation may be to examine the effect of sucrose consumption on stress derived

cross sensitization.

Dopamine and impulsivity

Dopamine is released in response to food cues, making it essential for the

motivational prompt to consume food (Flagel et al., 2011; Salamone & Correa,

2012). As dopamine neurons project from the VTA to the NAc, caudate, putamen,

PFC, hippocampus and AMG, they modulate a wide variety of behaviours and

emotions(de Jong, 2015) including impulsivity(Wade, de Wit, & Richards, 2000). In

relation to eating disorders such as obesity, there appears to be an impulsive

tendency to overconsume sweet foods in response to cravings for carbohydrates and

sugars. When a subject is withdrawn from an addictive substance, impulsivity

increases and can be quantified using a behavioural test called differential-

reinforcement-of-low-rate-schedule performance (DRL) (Mangabeira, Garcia-

Mijares, & Silva, 2015). The test requires the subject to withhold a response for a set

period of time before they are permitted to respond and thereby earn a

reward(Kirshenbaum, Brown, Hughes, & Doughty, 2008). Deprivation after long

term sugar consumption in rats resulted in impairment of DRL performance,

329

confirming its resemblance with drug addiction and suggesting an increase in

impulsive behaviour following sugar deprivation (Mangabeira et al., 2015). The

suggested mechanism of action for this behaviour is the dopaminergic system as

dopamine receptor antagonists (pimozide(Wise, 2006), raclopride (Maldonado,

Valverde, & Berrendero, 2006), SCH 23390 (Verty, McGregor, & Mallet, 2004)) can

attenuate the desirability of highly palatable foods in rats (Mangabeira et al., 2015;

Wise, 2006).

The progression towards obesity is linked with a decline in neural responses

to reward similarly induced by cocaine and heroin, which moderate the transition

from casual user to the impulsive actions of a compulsive drug-taker(P. M. Johnson

& Kenny, 2010). Impulsive actions such as over eating are modulated by the

excitation of D1 dopaminergic neurons and inhibition of D2 dopaminergic neurons in

the dorsal striatum and the NAc (P. M. Johnson & Kenny, 2010). Preclinical and

clinical studies have been conducted to investigate the role of dopamine in obesity

with findings showing obese subjects have an inverse relationship between the

abundance of dopamine D2Rs and body mass index, suggesting a dopaminergic role

in compulsive eating(P. M. Johnson & Kenny, 2010; Pijl, 2003; Volkow & Wise,

2005; Wang et al., 2001). Male rats self-administering amphetamine showed a

reduction in the ability of D2Rs to inhibit dopamine release in the NAc supporting

the theory that desensitization due to short term substance abuse modulates

addiction-related behaviours such as impulsivity (E. P. Bello et al., 2011; Calipari et

al., 2014; Cools, Barker, Sahakian, & Robbins, 2003). In a human study, twelve

patients with Parkinson’s disease, a disorder of the mesolimbic dopamine system,

underwent several cognitive tests designed to provide a measure of impulsivity.

330

Increased impulsive behaviour was found to correlate with greater dopamine

levels(Cools et al., 2003).

The neural mechanisms modulating motivation and drive are implicated in

both the loss of self-control exhibited by over eating and addiction to drugs of abuse

(Avena et al., 2008; Berridge, 1996). Neurofunctional imaging has documented

neuroadaptations in the PFC (Volkow & Fowler, 2000), hippocampus(Wang et al.,

2006), AMG and NAc (Volkow & Wise, 2005). A functional MRI study was

conducted showing female adolescents an image of either highly palatable food

(dessert) or a vegetable (Batterink, Yokum, & Stice, 2010). In comparison to leaner

adolescents, overweight adolescents had increased behavioural impulsivity and

decreased neural activation in frontal inhibitory regions of the brain (e.g. medial

prefrontal cortex, and orbitofrontal cortex) (Batterink et al., 2010). In addition,

activation of the reward region was positively correlated with body mass index when

shown the picture of dessert (Batterink et al., 2010). These findings suggest the

reduced functioning of inhibitory controls combined with increased responses in

food reward regions are relevant to weight gain (Batterink et al., 2010). The Stroop

test requires an individual to suppress their automatic response to certain stimuli.

This was used to assess the inhibitory control of overweight / obese 10 y old children

in comparison to normal weight children(Reyes, Peirano, Peigneux, Lozoff, &

Algarin, 2015). Results indicated changes to inhibitory control functions in

overweight children suggesting less emotional self-regulation, which in conjunction

with dysfunctional impulsivity control may contribute to over eating behaviour in the

overweight/ obese children(Reyes et al., 2015).

331

10.6 Common anatomical structures and neural substrates of stress driven, emotional behaviour

The insular cortex and cingulate cortex (part of the limbic system) are

responsible for the processing of emotions in conjunction with processes involving

higher cognition (Rajmohan & Mohandas, 2007). The hippocampus provides

negative feedback for the HPA axis with both neuroadaptations to its volume and

capacity for neurogenesis implicated in emotional disorders(Jacobson & Sapolsky,

1991). The AMG processes, stores and retrieves fear memories, and initiates

appropriate behavioural responses(Martin, Ressler, Binder, & Nemeroff, 2009). It is

responsible for the manner in which we express fear, aggression and defensive

behaviour(Martin et al., 2009). The PFC regulates executive functions and reward

processing while impulse control and mood are modulated by the orbitofrontal

cortex(Martin et al., 2009). Emotional disorders involve a complex interconnected

number of neuroendocrine, neuropeptide, neurotransmitter and neuroanatomical

adaptations (Gulpinar & Yegen, 2004). These alterations may occur due to genetic

predisposition or as a result of environmental influence(Martin et al., 2009). We

assert that abundant evidence exists regarding the role overconsumption of sugar

plays in altering these brain regions and contributing to emotional disorders such as

depression, anxiety and fear.

10.6.1 Anxiety

Generalized anxiety disorder is defined by uncontrolled, exaggerated concern

over numerous endeavours (Association, 2013). Treatment often consists of

anxiolytic drugs that act primarily on the monoamines serotonin, noradrenalin and

dopamine(Martin et al., 2009). In the central nervous system these neurotransmitters

are released in conjunction with neuropeptides that have strong links to anxiety, such

as neuropeptide Y and cholecystokinin expressed in the limbic cortex where they

332

influence emotions and stress levels(Martin et al., 2009). Other molecular substrates

involved in regulating the stress response include corticotropin-releasing factor and

adrenocorticotropic hormone in the HPA axis (Patriquin & Mathew, 2017). In cases

of anxiety, those carrying the short-allele of the serotonin transporter gene (5-

HTTLPR) demonstrate over activation of the AMGs in response to viewing fearful

and angry faces as compared to those carrying the long-allele(Hariri et al., 2005).

Two specific brain circuits are notable in anxiety disorders, the first involves the

dorsolateral PFC, anterior cingulate cortex, dorsal parietal cortex and precentral

gyrus, all associated with cognitive control, executive function, flexible cognition,

working memory and attention (Coplan, Webler, Gopinath, Abdallah, & Mathew,

2018). The second circuit is known as the negative affect circuit and as the name

suggests is activated by negative stimuli (Coplan et al., 2018). Anatomically the

hippocampus, AMG, medial PFC, and the dorsal, subgenual and pregenual regions of

the anterior cingulate cortex may undergo neural adaptation to skew negative bias

and threat by reducing attenuation through the cognitive control circuit (Coplan et

al., 2018).

Human studies show links between high caloric food consumption and anxiety.

Recent evidence from epidemiological studies found a suggestive link between

greater consumption of processed foods and widespread presence of anxiety

disorders (Jacka, Mykletun, Berk, Bjelland, & Tell, 2011; S. E. Quirk et al., 2013;

Westover & Marangell, 2002). There are also numerous studies linking adverse

childhood experiences (early life stress) with an increased risk of body weight gain

and obesity during adolescence and adulthood (D'Argenio et al., 2009; Gunstad et al.,

2006; Lumeng et al., 2013; Noll, Zeller, Trickett, & Putnam, 2007;

Ramasubramanian, Lane, & Rahman, 2013). More specific behavioural studies

333

observing acute and chronic withdrawal from sugar in rodents show anxiety is

induced when withdrawal follows extensive periods of sugar consumption (Avena et

al., 2008). Long exposures (1 month of 12 h daily access) to palatable foods did

result in rats showing increased anxiety-like behaviours on the elevated plus maze

(EPM) when tested 24 h after withdrawal (Avena et al., 2008). Rats exposed to 12 h

intermittent access to 10% sucrose solution for 28 days also displayed anxious

behaviour in the EPM after 36 h withdrawal (Avena et al., 2008). These rats showed

reduced conditioned suppression after 1 and 28 days of abstinence, measured by a

failure to significantly reduce the number of lever-presses for sucrose during the

presence of a tone stimulus paired with shock (Avena et al., 2008). Similar results

have been reported using longer cycling periods whereby rats given intermittent

access to a sucrose diet showed increased anxiety in the EPM and defensive

withdrawal tests after 8 and 9 weeks on the diet cycle followed by 48 h withdrawal

(Parylak, Cottone, Sabino, Rice, & Zorrilla, 2012). Alternate investigations have

shown rats fed a carbohydrate rich diet to have increased protein oxidation in the

frontal cortex which correlated with anxiogenic behaviour (Souza et al., 2007). These

studies suggest a high probability of long term sugar consumption may contribute to

symptoms of anxiety.

A long term study has shown an opposite effect in terms of anxiety and

depressive like behaviours(Cao, Lu, Lewis, & Li, 2007). Findings show that sugar

consumption had no significant effects on motor activity in an open field (test for

locomotor and anxious like behaviour), on exploration in a T-maze, or on anxiety in

an EMP(Cao et al., 2007). Another study looking at the anxiety effects of sugar after

a 1 year intervention of sucrose and honey found that anxiety decreased after 3

months of consumption, as tested with the EPM and open field test (Chepulis et al.,

334

2009). Importantly, this reduction was maintained until the end of the experiment at

12 months (Chepulis et al., 2009). Despite this, honey-fed rats showed significantly

less anxiety throughout the study as compared with those fed sucrose (Chepulis et al.,

2009). Together this data suggests that chronic (i.e., greater than 1 month)

intermittent access to high energy foods may increase anxiety-like behaviour

followed by a plateau observed after 3months (Chepulis et al., 2009).

10.6.2 Depression

Depression is characterized by feelings of sadness, hopelessness and reduced

pleasure in daily activities (Association, 2013). Functional, structural and

neurochemical factors associated with the pathophysiology of depression include

dysregulation of the HPA axis resulting in reductions in hippocampal, PFC and

striatal volumes (Koolschijn, van Haren, Lensvelt‐Mulders, Hulshoff Pol, & Kahn,

2009). In areas of emotional processing such as the AMG and PFC abnormal

metabolism of glucose and changes to cerebral blood flow have also been

demonstrated (Drevets, 1998). Although the association between obesity and

depression is well documented, less is known about the physiological and

corresponding psychological influence long-term sugar consumption has on

depression (Luppino et al., 2010). Recent research suggests a possible contributor to

the incidence rate of depression is sugar overconsumption (Sánchez-Villegas et al.,

2012; Westover & Marangell, 2002). We suggest the neuroadaptations that occur to

depression-related brain regions, namely the hippocampus, PFC and AMG,

following sugar intake (reviewed above) contributes to the incidence of depression

(Hsu et al., 2015; Moorman & Aston-Jones, 2014; M. J. Robinson et al., 2014).

The first study reporting a possible link between sugar and depression was

conducted in 2002 by Westover and Marangell (Westover & Marangell, 2002). Data

335

collected data from 6 countries showed a correlation between the consumption of

sugar (calories/capita/day) with the yearly rate of major depression disorder

(Westover & Marangell, 2002). Fast food further increases the potential of

developing depression, and commercial baked goods are also positively correlated

with depressive disorders (Sánchez-Villegas et al., 2012). Not only highly palatable

food, but sweetened beverages (either artificially or otherwise) also contribute to

depression (Sánchez-Villegas et al., 2012; Westover & Marangell, 2002). Regular

consumption of sweetened beverages indeed increases the occurrence of depression

and suicidal tendencies (Guo et al., 2014; Lien, Lien, Heyerdahl, Thoresen, &

Bjertness, 2006; Pan, Zhang, & Shi, 2011; Shi, Taylor, Wittert, Goldney, & Gill,

2010).

In rats given the same high sugar/chow diet cycle for 7 weeks, highly

palatable foods induce depressive-like behaviour as evidenced greater immobility in

the Forced Swim Test (test for depression) and decreased preference for 0.8%

sucrose (test for anhedonia) (Iemolo et al., 2012). Correlations with symptoms of

depression have been observed following long-term exposure to sucrose, with long

term overstimulation of the dopaminergic system during adolescence, which may

occur through sucrose binging (Rada et al., 2005) resulting in deficits in later life that

affect motivation, memory and happiness (Avena et al., 2008; Kendig, 2014; Naneix

et al., 2018). When a model designed to generate a depressed-like state (known as

social defeat-induced persistent stress) was used, rats showed increased motivation to

acquire a sucrose reward and reinstated sucrose-seeking induced by a cue (Riga,

Theijs, De Vries, Smit, & Spijker, 2015). Importantly, these studies showed the long-

term effects of stress exposure induced deficits in the ability to evaluate natural

rewards (Riga et al., 2015).

336

Congenitally helpless rats, a genetic model of predisposition to major

depression, were used to respond less to the reward of sucrose solution (Vollmayr et

al., 2004). This lack of motivation to partake in sweet rewards is conversely shown

after long term sucrose consumption by adolescent rats, which results in depressive

like behaviour in adulthood including symptoms of anhedonia (inability to feel

pleasure) and increased anxiety-like behaviour (Gueye et al., 2018). Adult rats

consuming sucrose showed a similar depressive-like behaviour, but to a lesser

degree, suggesting the critical period of brain development that occurs during

adolescence can be moderated by sucrose consumption and may increase the instance

of disorders related to rewards, such as depression (Gueye et al., 2018).

As depression correlates with altered glucose metabolism, it is not surprising

that comorbidity of diabetes mellitus (DM) and depression occurs quite frequently,

greatly increasing the mortality risk (van Dooren et al., 2013). A study highlighting

the link between glucose and depression in patients whose histories included DM and

cardiac disease found there was a 34% increased chance of depression in subjects

with elevated blood glucose levels (Azimova et al., 2015). A study including 70,000

postmenopausal women found a higher risk of developing depression in women

consuming a high-sugar diet than those consuming high naturally occurring sugars

(Gangwisch et al., 2015). However, research suggests sugar may contribute to

depression more in the male population as was demonstrated by the Whitehall Study

II which tracked dietary regimes and the corresponding medical health of 8,000

participants over a 22 year period (Knüppel, Shipley, Llewellyn, & Brunner, 2017).

Observing a five year duration they found a 23% increase in likelihood of men being

diagnosed with depression if they consumed ≥ 67 g of sugar per day compared to ≤

40 g (Knüppel et al., 2017). Cumulatively, these studies, while not able to confirm

337

that sugar causes depression, appear to produce enough evidence to show sugar

overconsumption contributes to an increased risk of developing depression.

10.6.3 Neurogenesis

Neurogenesis is the term given to new neurons generated from neural stem cells

(Ming & Song, 2011). The molecular substrates of stress, although not well

understood are implicated in the regulation of adult neurogenesis, through molecular

pathways modulated by glucocorticoids, inflammatory mediators and neurotrophic

factors (Egeland, Zunszain, & Pariante, 2015). Studies on antipsychotic

pharmacology support these findings by linking the blockade of dopamine D2

receptors with reductions in parkinsonian symptoms, anhedonia and increased

neurogenesis (Aringhieri et al., 2018; Chikama et al., 2017). The effect of diet on

hippocampal neurogenesis has been relatively unexplored.

Rats given 1 month access to fructose showed almost a 40% reduction in the

number of BrdU / NeuN-immunopositive cells (mature neurons) in the dentate gyrus

of the hippocampus(Van der Borght et al., 2011). Alternatively, rats consuming

glucose had a similar number of BrdU / NeuN-immunopositive cells as the water

controls(Van der Borght et al., 2011). Density of immature neurons labelled with

PSA-NCAM was decreased following sucrose, fructose and glucose(Van der Borght

et al., 2011). Sugar consumption increased the number of proliferating cells positive

for Ki67 (a marker of cell proliferation (Scholzen & Gerdes, 2000)) in all cases and

rats offered sucrose or fructose showed more cell death in the dentate gyrus of the

hippocampus (Van der Borght et al., 2011).

Rats given a high fructose corn syrup solution throughout adolescence were

tested for hippocampal-dependent contextual memory. Impairments in memory

function were found in later in life, suggesting that sugar consumption early in life

338

may have long-term negative effects on memory function (Noble et al., 2017). High

fructose consumption is also linked to insulin resistance (Stranahan et al., 2008) with

reduced hippocampal neurogenesis (Van der Borght et al., 2011). Hence it appears

convincing that a high-sugar diet negatively affects adult neurogenesis which may

contribute to the anxiety and depressive like behaviour demonstrated by animals after

withdrawal. Decreased neurogenesis in the hippocampus has further been implicated

in memory dysfunction and cognitive impairment disorders such as Alzheimer’s

disease(Price et al., 2001) and the learning and memory processes involved in

appetitive control(Davidson et al., 2009; Lathe, 2001). The hippocampus may hold

further insight into the cognitive processes by which food becomes entwined with

motivation and new rewards.

10.6.4 Fear

The functional neuroanatomy of fear encompasses the AMG, which stimulates

the HPA axis and the hippocampus, which suppresses activity of the axis (Martin et

al., 2009). Hyper activation of the AMG was recorded in response to viewing faces

construed as fearful (R. Bryant et al., 2008) and has been implicated in post-

traumatic stress disorder (PTSD) (Cortese & Phan, 2005). The PFC is required for

the extinction of fear memories (Giustino & Maren, 2015), and in line with these

findings changes in volume of the anterior cingulate cortex have been implicated in a

reduced ability to extinguish fearful memories (R. A. Bryant et al., 2008). When fear

memories become dysfunctional it is thought that the intrusive, recurring thoughts

result from an inability of the cognitive control circuit to repress the negative effect

circuit (Morey, Petty, Cooper, LaBar, & McCarthy, 2008). In patients with PTSD

their information processing may be overpowered by the hyper activation of the

AMG when exposed to threat related stimuli (Falconer et al., 2008).

339

At the molecular level corticotrophin-releasing factor from the paraventricular

nucleus of the hypothalamus initially activates the HPA axis in response to

threatening stimuli(Martin et al., 2009). Low levels of corticosterone, which acts

through binding to mineralocorticoid and glucocorticoid receptors, as well as

enhanced negative feedback of the HPA axis are associated with pathological fear

(Han, Ding, & Shi, 2014). Rats that underwent a single prolonged stress paradigm (a

reliable animal model of PTSD) have shown a down-regulation of both these

receptors in the hippocampus, enhanced fear and altered neuronal morphology of

AMG neurons(Han et al., 2014). Glucocorticoid receptors and dopamine receptors

have been implicated in processes within the PFC that drive extinction memory

learning. Infusions of corticosterone or a glucocorticoid receptor antagonist

(RU38486) into the infralimbic cortex and pretreatment with sulpiride (a dopamine

D2 receptor antagonist) attenuated fear expression suggesting enhanced fear

extinction (Dadkhah, Abdullahi, Rashidy-Pour, Sameni, & Vafaei, 2018). As

glucocorticoid receptors are also required for glucose homeostasis and play a role in

the development of hyperinsulinemia and obesity (Majer-Lobodzinska & Adamiec-

Mroczek, 2017) they are suspected as key players in neuroadaptations to the AMG,

hippocampus and PFC that occur after long-term sucrose consumption.

Juxtaposing the combined psychological and physical effects high sugar diets

have on pathological fear has, until recently, been an untouched topic of research.

Although no studies have been conducted on sugar consumption alone with regards

to pathological fear, some research has begun to explore the combined effect of high-

fat/sugar or high sugar and high carbohydrate diets (Baker & Reichelt, 2016;

Reichelt et al., 2015; Santos et al., 2018; Williams-Spooner, Richardson, & Baker,

2017). These studies are beginning to delineate the impact high caloric foods can

340

have on the intensity and duration of pathological fear, however research into the

effect of sugar alone is required to delineate the mechanisms involved. More recent

research has examined contextual fear following a standard contextual fear

conditioning protocol which was contradictory to this finding (Santos et al., 2018).

Specifically, high-sugar/carbohydrate diets significantly enhance fear-related

freezing to context (Santos et al., 2018). Whilst both examine contextual fear

memory formation, different fear conditioning protocols were used (trace fear

conditioning versus contextual fear conditioning). Indeed hippocampal function in

contextual fear memory has been noted as disparate, depending on the fear

conditioning protocol (see recent review by (Nicholas Chaaya, Battle, & Johnson,

2018, In Press)). Nevertheless, these studies highlight how diet can influence

hippocampal-dependent memory, as well as hippocampal function (Reichelt et al.,

2015; Santos et al., 2018).

Fear extinction is a protocol whereby, following excessive exposure to the

previously neutral fear conditioning CS (e.g. the context in contextual fear

conditioning, or a tone in auditory fear conditioning), a reduction of fear-related

freezing is seen (Maren, 2011). This reduction in fear is similar to basic theories

surrounding exposure therapy, whereby excessive exposure to fear-inducing stimuli

(via mental imagery, for example) results in a reduction in fear-related symptoms

(Maren, 2011; Milad & Quirk, 2012).

Investigation into the effect of PTSD symptoms on the consumption of highly

palatable ‘fast’ food and sweetened beverages was conducted to see if there was a

correlation with emotional eating behaviours and body mass index (Hirth, Rahman,

& Berenson, 2011). The answers to questions regarding frequency of consuming fast

food were collected from 3154 females and analyzed using regression analyses

341

(Hirth et al., 2011). To determine unhealthy eating habits, participants were asked if

they used diet pills, laxative, diuretics, skipped meals or vomited after eating.

Findings suggested that PTSD symptoms increased the frequency of consumption of

high caloric food and sodas, as well as contributing to unhealthy eating habits but did

not increase overall body mass index (Hirth et al., 2011). Cumulatively, data from

these various studies show a strong link between PTSD, fear memory and caloric

foods suggesting a possible correlation with high sugar consumption that requires

further investigation.

10.7 Sucrose Consumption Investigated

Negative states of emotion driven by substances of abuse are often shown by

examining anxiety-like behaviours, decreased pain tolerance, or an increase in the

point at which reward becomes sufficient to produce a stimulating effect and

memory deficits(G. Koob, 2018). The complexities of the systems involved, the

interconnectedness of the systems and the myriad of protocols used to study sucrose

dependence could potentially lead to a wide variety of outcomes and yet overall

investigations into high-sugar consumption in rats, mice and humans result in similar

outcomes with regards to memory, stress and emotion (see Table 3.)

Table 7 Published reports on the effect of sucrose and sweetener consumption on cognition, emotion and stress. Sucrose or sweetener consumption in rats

Authors and year

Subjects

Tasks Brain region involved

Findings

2h/day 10% sucrose for 28 days

Xu, T.J. and Reichelt, A.C., 2017 (T. J. Xu & Reichelt, 2018)

3 week-old male Sprague-Dawley rats (n = 8)

EPM, open-field, NPR and short- and long-term NOR1

hippocampus, basolateral amygdala

increased anxiety-like behaviours

Rat chow supplement

Reichelt, A., et al.,

6 week-old male

trace fear conditioning

hippocampus strengthened visual fears

342

ed with meat pies, cakes and biscuits for 10 weeks

2015 (Reichelt et al., 2015)

Sprague–Dawley rats (n = 16)

and attenuated contextual fears

Powdered diet: 7.9% sucrose for 52 weeks

Chepulis, L.M., et al., 2009 (Chepulis et al., 2009)

2-month old Sprague Dawley rats (n = 10-14)

1 EPM, NOR Y maze, C

NA reduced spatial memory and increased anxiety

35% sucrose solution for 9 weeks

Lemos, C., et al., 2016 (Lemos et al., 2016)

12-weeks old, male Wistar rats

open field, object displacement, NOR, forced swimming test

Hippocampus

decreased memory performance and increased helpless behaviour

32% sucrose solution for eight weeks

Jurdak, N. and Kanarek, R.B., 2008 (Jurdak & Kanarek, 2009)

6 week old male Long-Evans rats (n = 10)

1 NOR NA decreased cognitive performance on object recognition

10% fructose in drinking water for 7 months

Sangüesa, G., et al., 2017 (Sangüesa et al., 2018)

female Sprague- Dawley rats

Morris water maze, 1 NOR

frontal cortex and hippocampus

reduced performance in the NOR test

30% sucrose solution, 30m, twice daily, 2-4wk Or 0.1% saccharin

Ulrich-Lai, Y.M., et al., 2010 (Ulrich-Lai et al., 2010)

Adult male Long-Evans rats

Restraint stress, social interaction test, open field, EPM

hypothalamic paraventricular nucleus, BLA

- reduced ACTH by sucrose and saccharin - reduced corticosterone only after sucrose sucrose - reduced restraint-induced tachycardia and behavioural anxiety

10-34% sucrose

Wilmouth, C.E., and

Adolescent male

Taste reactivity

NA greater positive taste

343

solution Spear, L.P., 2009. (Wilmouth & Spear, 2009)

Sprague–Dawley rats

and voluntary consumption

reactivity and reduced negative responding

Free access to 5% sucrose or water for 16d

Vendruscolo, L.F., et al., (Vendruscolo et al., 2010)

Male Wistar rats

fixed- and progressive-ratio self administration of: - saccharin - maltodex-trin - cocaine

Sugar overconsumption during adolescence, reduced motivation for saccharin and maltodextrin

1% sugar solution (high fructose corn syrup) throughout the adolescent phase of development (post-natal day 26-56).

Noble, E.E., et al., 2019 (Noble et al., 2017)

Male rats novel object in context

- impairments in hippocampal-dependent memory function later in life - NOIC performance impaired at PN 175

Sucrose and sweetener consumption in mice

Authors and year

Subjects Tasks Brain region involved

Findings

High-carbohydrate diet (45% condensed milk, 10% sugar and 45% chow) for 8 weeks

Santos, C.J., et al., 2018 (Santos et al., 2018)

5 – 7 week old, male BALB/c mice

restraint stress, 1

EPM, contextual fear conditioning, tail suspension test

NA increased anxiety-like and depressive-like behaviour and aversive memory

10% sucrose, 3

I2BC for four weeks, withdrawal animals received only water for one week after

Kim, S., et al., 2017 (S. Kim et al., 2018)

C57BL/6 mice

tail suspension test,1 EPM, sucrose preference test

nucleus accumbens

withdrawal after sucrose overeating induces depression and anxiety-like behaviour.

344

the four weeks Acesulfame potassium and low carbohydrate diet for 4 weeks

Ibi, D., 2018 (Ibi, Suzuki, & Hiramatsu, 2018)

Male ddY strain mice (7–9 weeks old)

Y-maze and NOR, glucose levels

frontal cortex

decrease in short-term and object cognitive memories decreased glucose levels

Sucrose and sweetener consumption in humans

Authors and year

Subjects Tasks Brain region involved

Findings

Sugar, sweetened beverages, and fruit intake was assessed on a servings per day basis

Cohen, J.F.W., et al., 2018 (Cohen, Rifas-Shiman, Young, & Oken, 2018)

mother and child pairings, during pregnancy and childhood mean ages 3.3 yrs, 7.7 yrs (n = 1,234)

2 PPVT-III KBIT-II KBIT-II WRAVMA WRAML and HOME-SF

NA adverse impact on child memory and learning

High-sugar content milkshakes

Shearrer, G.E., et al., 2018 (Grace E Shearrer, Eric Stice, & Kyle S Burger, 2018)

133 adolescents

food picture exposure during fMRI

temporal gyrus, operculu, juxtapositional lobule, thalamus, caudate

increased signal in the reward learning, processing and motivation regions of the brain

Sucrose- or aspartame-sweetened beverage consumption three times per day for 2 weeks

Tyron, M.S., et al., 2015 (Tryon et al., 2015)

Nineteen women (age 18–40 y)

Salivary cortisol, Montreal Imaging Stress Task

hippocampus sucrose consumption resulted in: - higher activity in the left hippocampus - reduced stress-induced cortisol - lower reactivity to naltrexone,

345

lower nausea, and a trend toward lower cortisol

1 EPM: elevated plus maze; NPR: novel place recognition; NOR: novel object recognition. 2 PPVT-III: Peabody Picture Vocabulary Test, third edition for maternal testing; WRAVMA: Wide Range Assessment of Visual Motor Abilities, KBIT-II: Kaufman Brief Intelligence Test, second edition, WRAML: Wide Range Assessment of Memory and Learning for childhood testing; HOME-SF: Environment short form test to evaluate home environment for cognitive stimulation and emotional support. 3 I2BC: intermittent 2 bottle choice. 4 LTP: long-term potentiation.

10.8 Therapeutics for obesity, derived from studies of sucrose consumption

10.8.1 Pharmacological approaches

Sucrose is shown to be addictive in rodents(Avena et al., 2008; Carlo Colantuoni

et al., 2002; Rada et al., 2005) and the studies listed in Table 3 show correlations

between sucrose consumption and emotional disturbances. The NK1- (Neurokinin-1)

receptor system is involved in the reinforcement mechanism that motivates the desire

to have stimuli that no longer create pleasure, a characteristic of addiction(Sandweiss

& Vanderah, 2015), and is implicated in both anxiety and depression(Mantyh, 2002).

Using the intermittent access model with a 5% sucrose solution, our laboratory

showed the NK1-receptor antagonist ezlopitant (which possesses anxiolytic effects)

was able to attenuate and inhibit sucrose intake in Long Evans rats(Steensland et al.,

2010). This finding suggests the NK1- receptor system to be a potential target for

sugar-related obesity therapeutics(Steensland et al., 2010). Other potential options

include antidiabetic drugs such as sodium-glucose cotransporter 2 (SGLT2)

inhibitors, which impede renal glucose reabsorption and are generally considered

effective, though lose efficacy when taken long term(Gillies et al., 2007). Indeed, a 4

week oral administration of ipragliflozin, (a SGLT2 inhibitor) with antidiabetic

effects on type 2 diabetes, decreases the caloric balance and improves symptoms of

diabetes, including obesity in mice fed on 20% glucose or sucrose solution(Chao &

Henry, 2010; Tahara, Takasu, Yokono, Imamura, & Kurosaki, 2018)

346

A recent review detailing the current management for type 2 diabetes suggested

a step wise approach; including diet and exercise the patient would take metformin

(which lowers high blood glucose levels by decreasing the production of hepatic

glucose, increasing insulin sensitivity and lowering intestinal glucose

absorption(Rena, Hardie, & Pearson, 2017)), a glucagon-like peptide 1 receptor

agonist (to increase insulin output from the pancreas(Drucker & Nauck, 2006)) or a

SCLT2 inhibitor in addition to one of the approved weight-loss drugs (to control

appetite or the absorption of calories consumed(Padwal & Majumdar,

2007))(Burguera, Ali, & Brito, 2017). Table 4 lists doses and mechanisms of action

of therapeutics used in trials to alter sucrose consumption in an effort to discover

alternate treatments for diabetes and obesity.

Table 8 Therapeutics used in sugar consumption trials.

Author, year Drug / Dose Mechanism of action

Subjects Findings

Richard, D., et

al., 2000

(Richard,

Ferland,

Lalonde,

Samson, &

Deshaies,

2000)

Topiramate (30 mg / kg)

Blocks voltage gated sodium and calcium channels, glutamate, GABA (Anticonvulsant)

Rats Weight sucrose intake

Beczkowska,

I.W., 1992

(Beczkowska

et al., 1992)

Naltrexone Naloxone

Kappa and Mu2 (opioid) receptor antagonists (Treat drug and alcohol dependence

Rats Sucrose intake

Steensland, P.,

et al., 2010

(Steensland et

Ezlopitant (2,5,10 mg / kg)

NK1 receptor antagonist (Anxiolytic and

Rats obesity

347

al., 2010) antiemetic)

Shariff, M.,

2016 (Shariff

et al., 2016)

Varenicline (0.3,1,2 mg / kg)

nAChR partial agonist (Smoking cessation)

Rats sucrose intake

Tahara, A., et

al., 2018

(Tahara et al.,

2018)

Ipragliflozin (0.1 – 3 mg / kg)

SGLT2 inhibitor (Antidiabetic)

Mice sucrose intake

Muscat, R. and

Willner, P.,

1989 (Muscat

& Willner,

1989)

Sulpiride (20,40 mg / kg)

D2 and D3 receptor elective antagonist (Antipsychotic)

Rats desire for sucrose

Lof, E., et al.,

2010 (Löf et

al., 2010)

Methyllycaconitine(0.3,1 mg/kg)

α7 nAChRs selective antagonist

Rats sucrose

preference

Patkar, O., et

al., 2017

(Patkar et al.,

2017)

Buspirone (1,2.5,5 mg / kg)

5-HT1A/1B partial agonist (Anxiolytic)

Mice sucrose consumption

Lin, Z., et al.,

2013 (Z. Lin et

al., 2013)

Curcumin

(40 mg / kg)

Upregulation of

PPAR-γ activation

(Herbal

supplement)

Rats sucrose intake

Kurhe, Y., et al., 2014 (Kurhe, Radhakrishnan, & Gupta, 2014)

Ondansetron (1mg / kg)

Serotonin receptor (5-HT3) antagonist (Antiemetic)

Mice Sucrose

consumption

Badia‐Elder,

N.E., et al.,

2003 (Badia‐Elder et al.,

NPY (5µg / 5µl) (10µg / 10µl)

Inhibits GAD67 expression (Vasoconstrictor)

Rats sucrose intake

348

2003)

Pandit, R., et

al., 2015

(Pandit et al.,

2015)

AgRP (0.66 nmol) (1 nmol)

MC 3/4 receptor inverse agonist (Decreases metabolism)

Rats motivation for sucrose

10.8.2 Lifestyle Interventions

Diet and exercise are highly recommended approaches to treatment for obesity

(Bray & Bouchard, 2014). A 2007 systematic review and meta-analysis that

evaluated lifestyle interventions for patients with impaired glucose tolerance looked

at 21 randomized controlled trials designed to delay or prevent the onset of type 2

diabetes (Gillies et al., 2007). Data revealed evidence in support of interventions

such as diet and exercise being at least as effective as oral diabetes drugs, a Chinese

herbal remedy (jiangtang bushen) and the anti-obesity drug orlistat (Gillies et al.,

2007). The dilemma with lifestyle interventions is that regain of the weight lost is

common after a period of time and often results in weight gain greater than that

originally lost (Barte et al., 2010). Strategies that proved effective in maintaining

weight loss included a consistent regular meal pattern that included breakfast, a high

expectation to succeed, a good support system and behavioural self-monitoring

(Barte et al., 2010). Mindfulness meditation is another intervention studied as a way

to regulate emotional eating. A self-reported survey was conducted to examine

whether mindfulness practice could alter emotional eating (Levoy, Lazaridou,

Brewer, & Fulwiler, 2017). Surveys conducted before and after the intervention

revealed lower emotional eating scores after the meditation, suggesting a possible

role for mindfulness as a treatment for emotional eating (Levoy et al., 2017).

Mindfulness studies do not show a direct effect on weight loss itself, but appear to be

quite successful in lessening the addictive-like behaviours pursuant to relapse of

349

weight gain (Goldbacher, La Grotte, Komaroff, Vander Veur, & Foster, 2016;

Keesman, Aarts, Häfner, & Papies, 2017; Moor, Scott, & McIntosh, 2013).

10.8.3 Digital technology

Decreasing emotional eating has been proposed as a potential mechanism for the

long-term maintenance of weight loss (Barte et al., 2010). Digital technology may be

our best hope of achieving this goal. Interaction with online environments provide

the social support often lacking during maintenance of weight loss(K. O. Hwang et

al., 2010). Apps for exercise regimes, food tracking, meditation and positive thinking

all provide a support system to the user and promote networking with others in

similar situations. Support networks are useful tools as increases in body weight

decrease an individual’s ability to make informed decisions regarding highly

palatable food, resist temptation and regulate their emotions (Yeomans, 2017). These

challenges were supported through a study of 17 obese women and their

physiological reactions to verbal food cues which found that food and beverage

preferences affected physiologic responses as well as cognition and attention

(Fioravanti et al., 2004).

Wearable technology such as Fitbits (activity trackers), that interact directly with

the individual, provide information about the individual and promote long term

maintenance and support models of healthcare where the patient takes an active role

in their own wellness(Handel, 2011; Sama, Eapen, Weinfurt, Shah, & Schulman,

2014). Wearable technologies appear to be extremely effective in assisting weight

loss in patients with serious mental illness (Aschbrenner, Naslund, Shevenell,

Mueser, & Bartels, 2016). Thirteen obese individuals diagnosed with a serious

mental illness (e.g. schizophrenia and major depressive disorder) participated in 24

weeks of behavioural weight loss intervention encompassing group sessions for

350

weight management and exercise, mobile health technology and social media for

motivation, peer support and a self-monitoring tool (Aschbrenner et al., 2016). At the

end of 6 months of interventions, 45 % of participants had reduced their weight

(below their baseline) and increased fitness (as measured through walking distance)

(Aschbrenner et al., 2016).

10.9 Discussion

The data summarized in this article suggests overconsumption of sugar can

lead to brain adaptations involving many different neural systems, molecular

substrates and subsequent changes in behaviour. The ease of availability and cost

effectiveness of high caloric, sweetened food and beverages appear to be a

contributing factor in the world wide increase in obesity(He et al., 2018; Sinha,

2018). Paradigms designed to investigate emotional eating show increases in weight

gain due to higher caloric intake, nevertheless examination of the correlation

between environmental and social inputs, individual thought processes and behaviour

that maintains emotional eating is yet to be defined. It seems feasible to suggest the

pleasurable sensations brought about through the consumption of sugar may provide

a self-medicating method to deal with daily stresses (Brewerton, 2011; Fortuna,

2010). Common hedonistic mechanisms play a role in both obesity and addiction to

drugs of abuse (P. M. Johnson & Kenny, 2010). While the root cause of obesity

remains elusive, the estimated global annual healthcare cost of treating illnesses

related to obesity may reach US $1.2 trillion per year by 2025.

Every day new digital applications are being developed to assist our pursuit of

health and happiness. In the future it may be possible to collect personal information

about the consumption of high-sugar foods and beverages in conjunction with

emotions felt on a daily basis. The possibilities of personal wellness monitoring,

351

implantable in vivo monitoring and drug delivery devices will also require further

robust study before they prove to be effective treatments for obesity and food

addiction. It would be interesting to investigate a possible correlation between sugar

consumption, emotions and body mass index (BMI > 30 denotes obesity). It would

also be significant to know if such applications combining game rewards might assist

with the childhood obesity epidemic as current predictions state that globally, 2.7

billion adults will be overweight and/or obese by 2025 (Haby, Markwick, Peeters,

Shaw, & Vos, 2012).

Functional brain imaging studies substantiate the higher preference and

increased emotional activation that occurs in response to images and verbal cues

related to high sugar content foods and beverages, making it more difficult for

overweight individuals to resist eating unhealthy food (Batterink et al., 2010; Pursey

et al., 2014). Optogenetic and chemogenetic studies may assist in defining the

combination of neural pathways and substrates involved. Other tools designed to

unravel the complexities between neuronal organisation and behaviour include

methods to quantify cellular populations that are functional for particular behaviours

(Hadley C Bergstrom et al., 2011). Using analytic methods (including micro-binning

and density mapping) to accurately compare functional neural network activity it

may be possible to produce microanatomical topography of molecular activity

resultant from the influence of long term sugar consumption on a variety of stress

and anxiety related behaviours (A Jacques et al., 2018). This would assist in

demonstrating pathophysiological neuroadaptive changes and perhaps lead to the

development of enhanced pharmacotherapeutic and technological strategies to assist

the reduction of excess sugar consumption.

352

In conclusion, the ease of access to sugar rich diets today is an environmental

contributor to obesity, but it may be sugars ability to generate a superior neurological

reward signal which overrides self-control mechanisms and leads to obesity(Lenoir

et al., 2007; Yeomans, 2017). Obesity and long term sugar consumption both result

in low basal levels of dopamine, particularly in the NAc, which may be the

mechanism which induces the desire to overeat in the hope to restore homeostatic

dopamine levels and avoid mild depression. Opioids, which induce feeding through

an abundance of brain regions, may be responsible for cue-induced relapse into

overeating behaviours, and binging on sugar postpones the release of acetylcholine

required to signal satiety. Each of these neuroadaptations implicates sucrose

consumptions ability to alter the way we perceive and process our emotions and

consequential behaviour. Perhaps a greater understanding of the neural mechanisms

of impulsivity and overeating are required to assist in the development of improved

obesity treatments.

It has been estimated that by 2020, 1.5 million people will die each year by

suicide, with 15 to 30 million attempting it(Weissman et al., 1999). Children

suffering anxiety disorders are twice as likely to attempt suicide, while those

suffering major depressive illnesses show a 3 fold chance at attempt(Weissman et al.,

1999).In support of these statistics, over 300 million people were reported as

suffering depression in 2016/17, with 264 million people reported to be suffering

from anxiety disorders(Organization, 2017a). This review has examined how

negative emotion can exacerbate sugar overconsumption, and vice versa. If negative

emotions are so prevalent in our children, and sugar intake so common, its

consumption may be considered a threat to the emotional stability of our race (see

reviews on mental health in children and adolescents(Beesdo, Knappe, & Pine, 2009;

353

Pine, Cohen, Johnson, & Brook, 2002; Wehry, Beesdo-Baum, Hennelly, Connolly,

& Strawn, 2015)). More importantly, reduction of sugar overconsumption may be

capable of significantly reducing the prevalence of negative emotion in a vast

number of individuals around the world.

354

Appendix C:

Glucocorticoid Receptor (GR)

This appendix comprises the following published book chapter:

Jacques, A., Battle, A.R., Johnson, L.R., (2017). The Glucocorticoid Receptor

(GR). In Choi, Sangdun (Ed.) Encyclopedia of Signaling Molecules [2nd

edition]. Springer Reference, New York.



4_101536

355

Synonyms GCCR; GCR; GCRST; Glucocorticoid nuclear receptor variant 1; GR; GRL; Nuclear receptor subfamily 3 group C member 1 (glucocorticoid receptor)

Background

The gene NR3C1 encoding the glucocorticoid receptor (GR) is located on

chromosome 5q31.3 in humans, chromosome 18 in rats and mice, and chromosome

13 in chickens (Flaherty et al. 2012). It counts 2201 bp and contains 15 exons. The

five prime untranslated region consists of exon 1 while the protein-encoding region

involves exons 2–9. Homologs are conserved in chimpanzees, dogs, rats, zebra fish,

and frogs. The structural organization of the GR arises from exons 2 to 9 and

includes a DNA-binding domain and hinge region between the N and C termini.

Exon 2 codes the N-terminal domain. Exons 3 and 4 encode the DNA-binding

domain consisting of two zinc fingers. The GR consists of 777 amino acids and is

expressed throughout the body. Subcellular locations include the cytoplasm,

mitochondrion, cell nucleus, and plasma membrane. It is presently unknown if

different structural forms of the GR are associated with the cytoplasm and

membrane.

The name glucocorticoids derive from original descriptions of GR function in

the regulation of gluconeogenesis, however the diverse functions of the GR and now

beginning to be understood. GRs bind the glucocorticoid hormones cortisol (humans)

and corticosterone (rodents) and regulate genes facilitating processes such as energy

metabolism, immune responses, growth and development, and brain and body

responses to stress and challenge.

356

Glucocorticoids are released into the circulatory system throughout the day at

varying concentrations in a circadian-dependent manner. In addition, during times of

stress, additional volumes of glucocorticoids are released. Glucocorticoid release is

driven by corticotrophin-releasing hormone by the hypothalamus which is

transported to the anterior pituitary. Here adrenocorticotropic hormone (ACTH) is

released from the pituitary gland to the circulatory system which stimulates the

adrenal glands to release cortisol from the adrenal glands into the bloodstream.

Release is pulsatile which contributes to the fast “ultradian” rhythm of release.

Cortisol released from the adrenal gland targets organs throughout the body,

including the brain (Fig. 1).

Glucocorticoids bind to the GR to regulate gene transcription, this regulation

can result in either gene translation or transrepression (Prager and Johnson 2009). In

addition, GR may also act as fast-acting membrane-associated receptors regulating

cell structure and function (Prager and Johnson 2009; de Kloet 2014). Disorders

potentially associated with mutations in the gene or dysregulation of the GR function

include glucocorticoid resistance, Cushing’s syndrome lymphosarcoma, major

depressive disorder, posttraumatic stress disorder, and other diverse disorders

(Kadmiel and Cidlowski 2013).

357

Figure 11-1 Glucocorticoid Receptor (GR) The hypothalamic–pituitary–adrenal axis. During times of stress, corticotrophin-releasing hormone and vasopressin are released from the hypothalamus and transported to the anterior pituitary. Adrenocorticotropic hormone (ACTH) is released from the pituitary gland which stimulates the adrenal glands to release cortisol from the adrenal glands into the bloodstream. Cortisol released from the adrenal gland targets organs throughout the body, including the brain.

Structure of GR

The GR protein has domains arising from exons 2 to 9. Exon 2 codes for the N-

terminus, containing the main transcriptional domain. The central region of the

protein which consists of two zinc fingers involved in DNA binding and homo-

dimerization (see below under structural studies) are encoded by exons 3 and 4. The

DNA binding domain is coded with the composition “zinc subdomain-helix-8-

358

strand”. The two helixes are perpendicular to each other with the hydrophobic side

chains forming a protein core. The zinc sites are located at equidistance from the

outside of the core. The first subdomain is folded onto the core and connects with the

two helices and the C-terminal. The second protrudes out from the protein core,

forms a loop, an a-strand and a short a-helix (Haerd and Gustafsson 1993). The C

terminus includes the domains required for transcription and ligand binding.

The glucocorticoid receptor in humans has two splice variants labelled GRα

and GRβ. The two isoforms are structured identically from amino acid 1 to 727 and

then deviate. The GRα functions as a transcription factor, but GRβ does not bind

glucocorticoid and lacks transcriptional functionality. It has been implicated in

asthma-related glucocorticoid resistance due to its dominant- negative inhibition of

GRα.

Transcription Factor

Both GR and the mineralocorticoid receptor (MR) have similar mechanisms of

action and therefore similar functional architecture. The same receptor domains are

responsible for ligand binding interactions with a variety of heat shock proteins,

translocation to the nucleus, DNA binding, and other transcriptional regulatory

protein interactions (Haerd and Gustafsson 1993). The DNA-binding domain is a

highly conserved region of the GR, abundant in lysine, arginine, and cysteine. The 15

base pair glucocorticoid response element (GRE) core sequence

(GGTACANNNTGTTCT) contains two partially palindromic hexamer sequences

with three intervening nucleotides. This allows a recombinant DNA-binding domain

molecule to associate with one half-site of the GRE while a second one binds

cooperatively to the adjacent half-site. This action strongly depends on the three

nucleotides of the intervening sequence. A section of five amino acid residues has

359

been found to be essential for dimerization and the binding of GR to GRE (Haerd

and Gustafsson 1993).

The first reported crystallographic study of the rat GR was published in 1991

(Luisi et al. 1991), where it was found to dimerize when bound to DNA. Subsequent

studies have revealed that the mode of binding of GR to GREs plays a pivotal role in

regulating transcription through adopting different binding conformations. When in

positive mode (activating transcription) the GR binds to the GRE as a homodimer

(Meijsing et al. 2009; Hudson et al. 2016) (see Fig. 2), while repressing transcription

(nGRE), the GR binds at two different sides as monomers (Hudson et al. 2013) (Fig.

2).

Figure 11-2 Glucocorticoid Receptor (GR)

360

The glucocorticoid receptor (GR) binds as a dimer on +GRE sites on DNA (Meijsing et al. 2009; Hudson et al. 2016) (a) to activate transcription or binds as two monomers to repress transcription (Hudson et al. 2013). (b) Blue and red colors indicate individual GR molecules bound to white +GRE or nGRE DNA strands.

Function of GR

Influential early work from de Kloet and colleagues identified differences in

receptor binding affinity between MR and GR (for review see Prager and Johnson

(2009) and de Kloet (2014)). This finding leads to the important concept that at

resting levels of adrenal corticosterone release, corticosterone bounds predominately

to MR, while at periods to elevate corticosterone release including as a result of

stress, corticosterone also bound to the lower affinity GR. Within the brain, GRs are

found in dense concentrations in neurons of the hippocampus, amygdala, and the

prefrontal cortex. Its abundant expression throughout the limbic system suggests an

important role in stress and defense reactions (Wolf et al. 2016).

During times of stress, GR is recruited. Stress triggers the activation of the

hypothalamic-pituitary–adrenal (HPA) axis. Corticotrophin-releasing hormone is

secreted from the anterior hypothalamus stimulating the pituitary gland to release

adrenocorticotropic hormone (ACTH) into the blood stream. ACTH stimulates the

adrenal glands to release corticosteroids (cortisol in humans, corticosterone in

rodents; CORT) into the blood stream (Reul and Kloet 1985). Once CORT enters the

brain via the blood stream the elevated levels act on the HPA-axis to inhibit the

release of more corticosteroids, reducing the initial stress reaction (transrepression).

GR returns target cells back to baseline after an initial stress reaction and enhances

recovery by increasing energy metabolism. Stress-related pathology may result from

dysregulation of this CORT/HPA-axis interaction (for detailed review see Millan et

al. (2012)). Depending on the amount of CORT exposure, the slow genomic action

361

whereby CORT binding facilitates gene transcription may take up to 30 minutes

from receptor activation and can last for days (see Joels et al. 2012).

Pharmacology of GR

The endogenous ligands of GR include aldosterone, corticosterone, cortisol,

and deoxycortisone. Agonists include clobetasol, therapeutically used for skin

disorders such as psoriasis; fluticasone propionate – asthma medication;

desoximetasone – a metabolite; and methylprednisolone commonly used to treat

immunodeficiency syndromes. Mifepristone is a GR antagonist marketed as RU-486;

and a selective antagonist is onapristone, used to treat breast cancer and implicated as

a possible treatment for hormone-dependent tumors (Vilasco et al. 2011). Since

glucocorticoid therapy was instituted over 60 years ago it has become the pillar of

anti-inflammatory modulators (Kadmiel and Cidlowski 2013). Synthetic

glucocorticoids are now prescribed for conditions such as asthma and chronic

obstructive pulmonary disease. Dexamethasone is used in psychiatry to test for

functioning of the HPA axis and its feedback mechanisms (dexamethasone

suppression test). Dexamethasone, a synthetic CORT, acts centrally to suppress the

ongoing HPA activity and reduces endogenous CORT levels.

Membrane GR

Glucocorticoids can affect neuronal activity within seconds of exposure to

cells. When genomic regulation was found to be incompatible with rapid effects,

studies suggested these actions were mediated by the activation of membrane-

associated receptors (see Prager and Johnson (2009) and Wolf et al. (2016) for

review). Reports to date of these actions involve the limbic system and brainstem,

areas involving stress, learning, emotional memory, reproductive behavior, and

362

movement. Studies involving the effects of glucocorticoids on learning, memory, and

stress show both inhibitory and excitatory processes.

In a classic study, the reproductive behavior of male rough-skinned newts was

shown to be suppressed by rapid corticosterone action that inhibited neural circuits of

the brain stem (Orchinik et al. 1991). In 1996, Sandi et al. reported systemic

glucocorticoid increased locomotor activity of rats in a novel environment, finding

the effect was nitric oxide dependent (Sandi et al. 1996). Electrophysiological studies

by Joels et al. (2012) and Tasker and Herman(2011) have identified fast-acting GR

responses.

Direct anatomical localization of GR at neuron membranes including synapses

was shown by Johnson et al. (2005) who identified GR receptors localized in the

postsynaptic membranes of the lateral amygdala (Johnson et al. 2005; Prager et al.

2010). They established these receptors in the presynaptic terminals, the postsynaptic

density, dendrites, dendrites spines, and soma of neurons. Emerging data suggest that

these membrane GRs may rapidly regulate neuron dendrite spine structure (for

review see Russo et al. (2016)). Synaptic GR may play a role in the modulation of

synaptic plasticity related to memory (Prager and Johnson 2009; Wolf et al. 2016).

Evidence for mGRs in other cells have also been demonstrated. For example,

Gametchu and coworkers identified human mGR in leukemic cells membranes using

peptide antibody labelling (Gametchu et al. 1993).

Summary

Cortisol (or corticosterone) activates the glucocorticoid receptor which

functions both as a transcription factor itself, a regulator of other transcription

factors, and also as a fast-acting membrane receptor. The genomic GR is located in

363

the cytoplasm and it is transported to the cell nucleus when ligand bound, where it

plays a role in transcription processes. Its functions include regulation of cell

proliferation, tissue differentiation, inflammatory processes, and neuronal plasticity.

GR gene mutations are linked with Cushing’s disease and glucocorticoid resistance

and other disorders. As GRs are located in almost every tissue of the body and act in

both genomic and nongenomic capacities, a comprehensive understanding of their

mechanisms of action will ensure their role as therapeutic targets.

References

de Kloet ER. From receptor balance to rational glucocorticoid therapy. Endocrinology. 2014;155:2754–69.

Flaherty KT, Robert C, Hersey P, Nathan P, Garbe C, Milhem M, et al. Improved survival with MEK inhibition in BRAF-mutated melanoma. New Engl J Med. 2012;367:107–14. https://doi.org/10.1056/NEJMoa1203421.

Gametchu B, Watson C, Wu S. Use of receptor antibodies to demonstrate membrane glucocorticoid receptor in cells from human leukemic patients. FASEB J. 1993;7:1283–92.

Haerd T, Gustafsson JA. Structure and function of the DNA-binding domain of the glucocorticoid receptor and other members of the nuclear receptor supergene family. Acc Chem Res. 1993;26:644–50.

Hudson WH, Kossmann BR, de Vera IMS, Chuo SW, Weikum ER, Eick GN, et al. Distal substitutions drive divergent DNA specificity among paralogous transcription factors through subdivision of conformational space. Proc Natl Acad Sci USA. 2016;113:326–31. https://doi.org/10.1073/pnas.1518960113.

Hudson WH, Youn C, Ortlund EA. The structural basis of direct glucocorticoid-mediated transrepression. Nat Struct Mol Biol. 2013;20:53–U69. https://doi.org/10.1038/nsmb.2456.

Joels M, Sarabdjitsingh RA, Karst H. Unraveling the time domains of corticosteroid hormone influences on brain activity: rapid, slow, and chronic modes. Pharmacol Rev. 2012;64:901–38. https://doi.org/10.1124/pr.112.005892.

364

Johnson LR, Farb C, Morrison J, McEwen B, LeDoux J. Localization of glucocorticoid receptors at postsynaptic membranes in the lateral amygdala. Neuroscience. 2005;136:289–99.

Kadmiel M, Cidlowski JA. Glucocorticoid receptor signaling in health and disease. Trends Pharmacol Sci. 2013;34:518–30.

Luisi BF, Xu WX, Otwinowski Z, Freedman LP, Yamamoto KR, Sigler PB. Crystallographic analysis of the interaction of the glucocorticoid receptor with DNA. Nature. 1991;352:497–505. https://doi.org/10.1038/352497a0.

Meijsing SH, Pufall MA, So AY, Bates DL, Chen L, Yamamoto KR. DNA binding site sequence directs glucocorticoid receptor structure and activity. Science. 2009;324:407–10. https://doi.org/10.1126/science.1164265.

Millan MJ, Agid Y, Brüne M, Bullmore ET, Carter CS, Clayton NS, et al. Cognitive dysfunction in psychiatric disorders: characteristics, causes and the quest for improved therapy. Nat Rev Drug Discov. 2012;11:141–68.

Orchinik M, Murray TF, Moore FL. A corticosteroid receptor in neuronal membranes. Science. 1991;252:1848.

Prager EM, Brielmaier J, Bergstrom HC, McGuire J, Johnson LR. Localization of mineralocorticoid receptors at mammalian synapses. PLoS One. 2010;5. https://doi.org/10.1371/journal.pone.0014344.

Prager EM, Johnson LR. Stress at the synapse: signal transduction mechanisms of adrenal steroids at neuronal membranes. Sci Signal. 2009;2:re5.

Reul J, de Kloet ER. Two receptor systems for corticosterone in rat brain: microdistribution and differential occupation. Endocrinology. 1985;117:2505–11.

Russo MF, Loy SRA, Battle AR, Johnson LR. Membrane associated synaptic mineralocorticoid and glucocorticoid receptors are rapid regulators of dendritic spines. Front Cell Neurosci. 2016;10.

Sandi C, Venero C, Guaza C. Novelty-related rapid locomotor effects of corticosterone in rats. Eur J Neurosci. 1996;8:794–800.

Tasker JG, Herman JP. Mechanisms of rapid glucocorticoid feedback inhibition of the hypothalamic–pituitary–adrenal axis. Stress. 2011;14:398–406. https://doi.org/10.3109/10253890.2011.586446.

Vilasco M, Communal L, Mourra N, Courtin A, Forgez P, Gompel A. Glucocorticoid receptor and breast cancer. Breast Cancer Res Treat. 2011;130:1–10. https://doi.org/10.1007/s10549-011-1689-6.

Wolf OT, Atsak P, de Quervain DJ, Roozendaal B, Wingenfeld K. Stress and memory: a selective review on recent developments in the understanding of stress hormone effects on memory and their clinical relevance. J Neuroendocrinol. 2016;28. https://doi.org/10.1111/jne.12353.

365

Appendix D:

Mineralocorticoid Receptor

Jacques, A., Johnson, L. R., Battle, A. R., (2017). The Mineralocorticoid

Receptor (MR). In Choi, Sangdun (Ed.) Encyclopedia of Signaling Molecules

[2nd edition]. Springer Reference, New York.



4_101537

366

Synonyms MCR; MLR; MR; NR3C2; NR3C2VIT; Nuclear receptor subfamily 3 group C member 2

Historical Background

Located on chromosome 4q31.1 in humans, the gene NR3C2 encodes the

mineralocorticoid receptor (Fan et al. 1989; Morrison et al. 1990). It has 5201 bp, an

exon count of 12, and is located on chromosome 8 in mice, 19 in rats, 1 in zebra fish,

and 4 in chickens. Mineralocorticoids (MRs) belong to the nuclear receptor

subfamily 3 and are distributed throughout the epithelia of the kidneys, sweat glands,

and colon and the nonepithelial tissues of the heart and brain. The human MR was

first cloned in 1987, consists of 984 amino acids, and is similar in structure to the

glucocorticoid receptor, sharing 94% identity in the DNA-binding domain (Funder

1997).

It is unique among steroid receptors in that it plays signaling roles in both

mineralocorticoids (e.g., aldosterone and deoxycorticosterone) and the

glucocorticoids (cortisol in humans, corticosterone in rats). In particular, both

cortisol and corticosterone bind to MR with similar affinity as aldosterone, but

aldosterone will only bind to glucocorticoid (GR) at very high concentrations. This

was determined through structural studies that showed that despite both MR and GR

showing high sequence homology in their ligand-binding domains (Sturm et al.

2005), nonspecific amino acid interactions between sequences 804–844 were

identified as essential for aldosterone specificity (Rogerson et al. 1999). NR3C2

defects may result in autosomal dominant pseudohypoaldosteronism type I. This

disorder characteristically entails a high flow rate of very dilute urine. Other gene

mutations may result in early-onset hypertension severely exacerbated in pregnancy.

367

MR is expressed in many cells in the body, where it modulates ion and fluid

balance, response to injury, and early responses to stress. Given the ubiquitous nature

of MR, this encyclopedia chapter focuses on the role of MR as a signaling molecule

in the brain. For recent general information on MR, the reader is directed to a 2014

review by Gomez- Sanchez (Gomez-Sanchez and Gomez-Sanchez 2014).

MR Signaling in the Brain

The body’s ability to adapt to stress requires the facilitation of neuronal

plasticity mediated by cortisol (in humans) and corticosterone (in rodents)

(Sarabdjitsingh and Joels 2014). Cortisol binds to genomic receptors such as the

mineralocorticoid (MR) and glucocorticoid (GR) receptors which function as

transcription factors and permit stress-related information to be stored for subsequent

use (Prager and Johnson 2009). The receptors regulate a variety of gene

transcriptional processes including the synthesis of new proteins which facilitate

synaptic plasticity (Prager and Johnson 2009). These processes are modulated by

binding of steroids, including cortisol (corticosterone in rodents) to both MRs and

GRs. Cortisol binds with higher affinity to MR than to GR (Prager and Johnson

2009; Joels et al. 2012; de Kloet 2014).

Localization of MR in the Brain and Body

MRs are distributed throughout the tissues of the brain, heart, kidney, colon,

hippocampus, hypothalamus, and adrenal fasciculata. Subcellular locations

include the cytoplasm, nucleus, endoplasmic reticulum membrane, and plasma mem-

brane. Epithelial locations include parts of the nephron (distally), the colon (distally),

and sweat and salivary glands. Nonepithelial loci include the neurons of the central

368

nervous system, cardiac myocytes, and smooth muscle cells of large vessels such as

the aorta.

MRs are predominantly expressed in the learning and memory centers of the

brain such as the hippocampus and amygdala (Reul and Kloet 1985). Localized in

the membrane, they facilitate second messenger (e.g., G protein) cascades to directly

affect membrane proteins, including the regulation of membrane potential through

gated ion channels (Prager and Johnson 2009; Prager et al. 2010; Joels et al. 2012).

Electron microscopy studies have shown MRs expressed at nuclear locations within

the lateral amygdala on glutamatergic (excitatory) and GABAergic (inhibitory)

neurons and at the extranuclear loci of presynaptic terminals, dendrites, and their

spines (Prager et al. 2010). Electron microscopy has allowed identification of MRs

localized to Golgi apparatus and mitochondrial membranes (Johnson et al. 2005;

Prager et al. 2010).

Function of MR in the Brain and Body

Conditioned fear can coexist with or trigger the stress response, elevating

adrenal hormones in the brain. Amygdala-dependent activation of the hypothalamic-

pituitary-adrenal (HPA) axis facilitates expression of conditioned fear. The HPA axis

increases corticosterone blood concentrations which facilitates binding to MRs and

GRs within the limbic system. In the brain genomic MRs (gMRs) have high-affinity

binding with glucocorticoids (e.g., cortisol, corticosterone), mineralocorticoids (e.g.,

aldosterone), and progesterone (Krozowski and Funder 1983). The diverse

distribution of MRs implicates mineralocorticoid effects on neuronal function in

specific subregions of the brain. The lateral nucleus of the amygdala, which both

acquires and stores fear memories, is the primary site for the resultant synaptic

plasticity (Prager et al. 2010). Studies of chronic stress models have shown

369

hypertrophy of dendrites in amygdala principal neurons (Johnson et al. 2005). In

epithelial tissues aldosterone activates MRs, by converting cortisol to NAD and

NADH. This facilitates protein expression which regulates the epithelial sodium

channel, sodium potassium pump, and serum- and glucocorticoid-induced kinase.

Sodium and water are reabsorbed resulting in increased extracellular volume,

increased blood pressure, and reduced potassium levels due to potassium excretion

(to maintain homeostasis).

Cortisol Signaling to Intracellular Genomic MR (gMR)

During stress corticosterone enters the brain rapidly, binding to MRs in the

limbic brain regions such as the septum, hippocampus, and amygdala. In the

hippocampus, the receptors bind to a complex of heat shock proteins (e.g., HSP90,

70, 56, etc.) and, once activated, dissociate from these proteins to homodimerize with

other receptors (Rupprecht et al. 1993). The dimerized genomic MRs translocate to

the nucleus of the cell and bind to mineralocorticoid response element (MRE), which

modulates gene transcription into mRNA of the activated genes. MRE is a short

DNA dimer, denoted by a pair of inverted repeats that are partitioned by three

nucleotides. Located within the promoter of a gene, it specifically binds a steroid

hormone receptor complex in order to regulate transcription. MRs at postsynaptic

membrane densities of excitatory synapses have been shown to regulate synaptic

transmission (Prager et al. 2010).

Structure of MR

The MR adopts a quaternary structure and is com- prised of three domains: the

N-terminal domain, a DNA-binding domain, and a C-terminal ligand-binding domain

370

(Pawlak et al. 2012) In the absence of a ligand, MR will form a hetero- multimeric

cytoplasmic complex with heat shock proteins HSP90, HSP70, and FKBP4 (Bruner

et al. 1996) In the presence of a ligand, after binding, it translocates to the nucleus to

bind to DNA as a homodimer and as a heterodimer with NR3C1.

The crystal structure of the MR in complex with the GRE was published in

2014 by Ortlund and coworkers (Hudson et al. 2014), revealing that during positive

transcription processes, it forms a dimeric complex with GRE (Fig. 1). Interestingly,

despite both the MR and GR exhibiting high sequence homology in the DNA-

binding domain (DBD), only the GR is able to bind both GRE and nGREs, which is

described in the recent work by (Hudson et al. 2016). In this report, they have shown

ancestral DBD was able to bind both GRE and nGREs; however amino acid

substitution in duplicated daughter genes resulted in this ability being lost in MRs

(Hudson et al. 2016).

371

Figure 12-1 Mineralocorticoid Receptor The mineralocorticoid receptor (MR) binds as a dimer (Hudson et al. 2014) on GRE sites. Blue and red colors indicate MR molecules, white indicates GRE DNA strands

Cortisol Signaling to Membrane- Localized MR (mMR)

Like the closely related GR, MR has also been identified as a functional

membrane receptor (mMR). In neurons mMR are fast-acting receptors that have been

documented to regulate synaptic transmission (for review see Prager and Johnson

2009; Joels et al. 2012; Russo et al. 2016). Anatomical evidence for their existence

was recently documented using electron microscopy where possible mMR where

found in postsynaptic densities as well as presynaptic structures (Prager et al. 2010).

Functional evidence for a rapid signaling membrane-based MR has been documented

by extensive electrophysiological studies by Joels and colleagues; see Joels et al. for

review (Joels et al. 2012). In a classic study of fast-signaling actions, MR agonists

were found to facilitate presynaptic glutamate release (Karst et al. 2005).

Pharmacology of MR

The endogenous ligands of MR in order of potency are corticosterone, cortisol,

aldosterone, and progesterone. Nuclear MR binds with high affinity to corticosterone

and cortisol (Reul and Kloet 1985). Due to GR lower-binding affinity, its genomic

response occurs only after exposure to a stressful event (Joels 2008). The rate of

occupancy and activation of MR is comprehensive even when circulating hormone

levels are low, pointing to a role for MRs in variations of ultradian rhythm (Russo et

al. 2016). mMRs were found to regulate ion channels to moderate the speed of

neuronal depolarization and synaptic transmission (Prager and Johnson 2009).

Presynaptic calcium levels increased glutamate release in the hippocampus, while the

372

postsynaptic efflux of potassium was inhibited, reducing hyperpolarization (Prager

and Johnson 2009).

Apart from cortisol, agonists of MR include aldosterone, produced in the

adrenal glands, that acts on the renin-angiotensin system; prednisolone and

dexamethasone, both anti-inflammatories; and progesterone, a sex hormone and

fludrocortisone, used to treat cerebral salt- wasting syndrome. MR antagonists

include spironolactone, a diuretic to prevent salt absorption and potassium excretion,

eplerenone an antihypertensive, finerenone used in the treatment of chronic heart

failure, and onapristone for the treatment of breast cancer.

Summary

When presented with stressful situations, corticosteroids are released effecting

the brain and, as a consequence, behavior. Both MR and GR mediate these actions

through their expression on neurons within the limbic system, but MR has a much

higher binding affinity to the naturally occurring cortisol and therefore does not

require the introduction of stress to become active. MR plays a major role in the

regulation of ion and water transport via the renin-angiotensin system. Gene

mutations are linked with autosomal dominant pseudohypoaldosteronism type I and

pregnancy-exacerbated hypertension. Alternate splicing can result in multiple

transcript variants.

MR moderates ligand-dependent transcription and binds to MRE to

transactivate specific target genes. The high-affinity binding of cortisol to MRs lends

credence to the nongenomic effects of mMRs to induce fast responses in neuronal

second messenger systems to regulate synaptic transmission (Prager et al. 2010). The

mineralocorticoid receptor-binding properties, interactions with other genes, and

373

extensive tissue expression afford its great functional diversity and multifarious

physiological regulation.

References

Bruner KL, Derfoul A, Robertson NM, Guerriero G, Fernandes-Alnemri T, Alnemri ES, et al. The unliganded mineralocorticoid receptor is associated with heat shock proteins 70 and 90 and the immunophilin FKBP-52. Recept Sig Transduct. 1996;7:85–98.

de Kloet ER. From receptor balance to rational glucocorticoid therapy. Endocrinology. 2014;155:2754–69.

Fan YS, Eddy RL, Byers MG, Haley LL, Henry WM, Shows TB. Assignment of human mineralocorticoid receptor (Mlr) to human chromosome-4q31.2. Cytogenet Cell Genet. 1989;51:996.

Funder JW. Glucocorticoid and mineralocorticoid receptors: biology and clinical relevance. Annu Rev Med. 1997;48:231–40. doi:10.1146/annurev.med.48.1.231.

Gomez-Sanchez E, Gomez-Sanchez CE. The multifaceted mineralocorticoid receptor. Compr Physiol. 2014;4:965–94. doi:10.1002/cphy.c130044.

Hudson WH, Kossmann BR, de Vera IMS, Chuo SW, Weikum ER, Eick GN, et al. Distal substitutions drive divergent DNA specificity among paralogous transcription factors through subdivision of conformational space. P Natl Acad Sci USA. 2016;113:326–31. doi:10.1073/pnas.1518960113.

Hudson WH, Youn C, Ortlund EA. Crystal structure of the mineralocorticoid receptor DNA binding domain in complex with DNA. PLoS One. 2014;9. doi:ARTN e10700081371/journal.pone.0107000.

Joels M. Stress hormone actions in brain, in health disease – possibilities for new pharmacotherapies – preface. Eur J Pharmacol. 2008;583:173. doi:10.1016/j.ejphar.2008.01.002.

Joels M, Sarabdjitsingh RA, Karst H. Unraveling the time domains of corticosteroid hormone influences on brain activity: rapid, slow, and chronic modes. Pharmacol Rev. 2012;64:901–38. doi:10.1124/pr.112.005892.

Johnson LR, Farb C, Morrison J, McEwen B, LeDoux J. Localization of glucocorticoid receptors at postsynaptic membranes in the lateral amygdala. Neuroscience. 2005;136:289–99.

Karst H, Berger S, Turiault M, Tronche F, Schutz G, Joels M. Mineralocorticoid receptors are indispensable for nongenomic modulation of hippocampal glutamate

374

transmission by corticosterone. P Natl Acad Sci USA. 2005;102:19204–7. doi:10.1073/pnas.0507572102.

Krozowski ZS, Funder JW. Renal mineralocorticoid receptors and hippocampal corticosterone-binding species have identical intrinsic steroid specificity. P Natl Acad Sci-Biol. 1983;80:6056–60. doi:10.1073/pnas.80.19.6056.

Morrison N, Harrap SB, Arriza JL, Boyd E, Connor JM. Regional chromosomal assignment of the human mineralocorticoid receptor gene to 4q31.1. Hum Genet. 1990;85:130–2.

Pawlak M, Lefebvre P, Staels B. General molecular biology and architecture of nuclear receptors. Curr Top Med Chem. 2012;12:486–504.

Prager EM, Brielmaier J, Bergstrom HC, McGuire J, Johnson LR. Localization of mineralocorticoid receptors at mammalian synapses. PLoS One. 2010;5. doi:ARTN e14344171371/journal.pone.0014344.

Prager EM, Johnson LR. Stress at the synapse: signal transduction mechanisms of adrenal steroids at neuronal membranes. Sci Signal. 2009;2:re5-re.

Reul J, Kloet E. Two receptor systems for corticosterone in rat brain: microdistribution and differential occupation. Endocrinology. 1985;117:2505–11.

Rogerson FM, Dimopoulos N, Sluka P, Chu S, Curtis AJ, Fuller PJ. Structural determinants of aldosterone binding selectivity in the mineralocorticoid receptor. J Biol Chem. 1999;274:36305–11. doi:10.1074/jbc.274.51.36305.

Rupprecht R, Reul JM, van Steensel B, Spengler D, Söder M, Berning B, et al. Pharmacological and functional characterization of human mineralocorticoid and glucocorticoid receptor ligands. Eur J Pharmacol Mol Pharmacol. 1993;247:145–54.

Russo MF, Loy SRA, Battle AR, Johnson LR. Membrane associated synaptic mineralocorticoid and glucocorticoid receptors are rapid regulators of dendritic spines. Front Cell Neurosci. 2016;10.

Sarabdjitsingh RA, Joels M. Rapid corticosteroid actions on synaptic plasticity in the mouse basolateral amygdala: relevance of recent stress history and beta-adrenergic signaling. Neurobiol Learn Mem. 2014;112:168–75. doi:10.1016/j.nlm.2013.10.011.

Sturm A, Bury N, Dengreville L, Fagart J, Flouriot G, Rafestin-Oblin ME, et al. 11-deoxycorticosterone is a potent agonist of the rainbow trout (Oncorhynchus mykiss) mineralocorticoid receptor. Endocrinology. 2005;146:47–55. doi:10.1210/en.2004-0128.

375

Bibliography

Acevedo, B. P., Aron, A., Fisher, H. E., & Brown, L. L. (2012). Neural correlates of

long-term intense romantic love. Social cognitive and affective neuroscience, 7(2), 145-159.

Adams, J. P., & Sweatt, J. D. (2002). Molecular psychology: roles for the ERK MAP kinase cascade in memory. Annual review of pharmacology and toxicology, 42(1), 135-163.

Adcock, R. A., Thangavel, A., Whitfield-Gabrieli, S., Knutson, B., & Gabrieli, J. D. (2006). Reward-motivated learning: mesolimbic activation precedes memory formation. Neuron, 50(3), 507-517.

Adinoff, B. (2004). Neurobiologic processes in drug reward and addiction. Harvard review of psychiatry, 12(6), 305-320.

Ahima, R. S., & Antwi, D. A. (2008). Brain regulation of appetite and satiety. Endocrinology and metabolism clinics of North America, 37(4), 811-823.

Ahmed, S. H., Guillem, K., & Vandaele, Y. (2013). Sugar addiction: pushing the drug-sugar analogy to the limit. Current Opinion in Clinical Nutrition & Metabolic Care, 16(4), 434-439.

Alberini, C. M. (2005). Mechanisms of memory stabilization: are consolidation and reconsolidation similar or distinct processes? Trends in Neurosciences, 28(1), 51-56. doi:http://dx.doi.org/10.1016/j.tins.2004.11.001

Albrecht, A., & Stork, O. (2017). Circadian rhythms in fear conditioning: an overview of behavioral, brain system, and molecular interactions. Neural plasticity, 2017.

Alsiö, J., Olszewski, P. K., Levine, A. S., & Schiöth, H. B. (2012). Feed-forward mechanisms: addiction-like behavioral and molecular adaptations in overeating. Frontiers in neuroendocrinology, 33(2), 127-139.

Altman, K. W., Richards, A., Goldberg, L., Frucht, S., & McCabe, D. J. (2013). Dysphagia in stroke, neurodegenerative disease, and advanced dementia. Otolaryngologic Clinics of North America, 46(6), 1137-1149.

Amilhon, B., Lepicard, E., Renoir, T., Mongeau, R., Popa, D., Poirel, O., . . . Gallego, J. (2010). VGLUT3 (vesicular glutamate transporter type 3) contribution to the regulation of serotonergic transmission and anxiety. Journal of Neuroscience, 30(6), 2198-2210.

Anagnostaras, S. G., Maren, S., & Fanselow, M. S. (1999). Temporally graded retrograde amnesia of contextual fear after hippocampal damage in rats: within-subjects examination. The Journal of Neuroscience, 19(3), 1106-1114.

Anderson, E. M., Larson, E. B., Guzman, D., Wissman, A. M., Neve, R. L., Nestler, E. J., & Self, D. W. (2018). Overexpression of the histone dimethyltransferase G9a in nucleus accumbens shell increases cocaine self-administration, stress-induced reinstatement, and anxiety. Journal of Neuroscience, 38(4), 803-813.

376

Anderson, L. C., & Petrovich, G. D. (2018). Distinct recruitment of the hippocampal, thalamic, and amygdalar neurons projecting to the prelimbic cortex in male and female rats during context-mediated renewal of responding to food cues. Neurobiology of learning and memory, 150, 25-35.

Anderson, M. I., & Jeffery, K. J. (2003). Heterogeneous Modulation of Place Cell Firing by Changes in Context. The Journal of neuroscience, 23(26), 8827-8835.

Andreasen, N. C. (2011). What is post-traumatic stress disorder? Dialogues in clinical neuroscience, 13(3), 240.

Aosaki, T., Miura, M., Suzuki, T., Nishimura, K., & Masuda, M. (2010). Acetylcholine–dopamine balance hypothesis in the striatum: An update. Geriatrics & gerontology international, 10(s1).

Aringhieri, S., Carli, M., Kolachalam, S., Verdesca, V., Cini, E., Rossi, M., . . . Scarselli, M. (2018). Molecular targets of atypical antipsychotics: From mechanism of action to clinical differences. Pharmacol Ther. doi:10.1016/j.pharmthera.2018.06.012

Aschbrenner, K. A., Naslund, J. A., Shevenell, M., Mueser, K. T., & Bartels, S. J. (2016). Feasibility of behavioral weight loss treatment enhanced with peer support and mobile health technology for individuals with serious mental illness. Psychiatric Quarterly, 87(3), 401-415.

Asnaghi, V., Gerhardinger, C., Hoehn, T., Adeboje, A., & Lorenzi, M. (2003). A role for the polyol pathway in the early neuroretinal apoptosis and glial changes induced by diabetes in the rat. Diabetes, 52(2), 506-511.

Association, A. P. (2013). Diagnostic and statistical manual of mental disorders (DSM-5®): American Psychiatric Pub.

Atkins, C. M., Selcher, J. C., Petraitis, J. J., Trzaskos, J. M., & Sweatt, J. D. (1998). The MAPK cascade is required for mammalian associative learning. Nature Neuroscience, 1(7), 602-609.

Avena, N. M., & Hoebel, B. (2003). A diet promoting sugar dependency causes behavioral cross-sensitization to a low dose of amphetamine. Neuroscience, 122(1), 17-20.

Avena, N. M., Long, K. A., & Hoebel, B. G. (2005). Sugar-dependent rats show enhanced responding for sugar after abstinence: evidence of a sugar deprivation effect. Physiology & behavior, 84(3), 359-362.

Avena, N. M., Rada, P., & Hoebel, B. G. (2008). Evidence for sugar addiction: behavioral and neurochemical effects of intermittent, excessive sugar intake. Neuroscience & Biobehavioral Reviews, 32(1), 20-39.

Avena, N. M., Rada, P., & Hoebel, B. G. (2009). Sugar and fat bingeing have notable differences in addictive-like behavior. The Journal of Nutrition, 139(3), 623-628.

Avena, N. M., Rada, P., Moise, N., & Hoebel, B. (2006). Sucrose sham feeding on a binge schedule releases accumbens dopamine repeatedly and eliminates the acetylcholine satiety response. Neuroscience, 139(3), 813-820.

Azimova, K., Rude, J., Mallawaarachchi, I., Dwivedi, A., Sarosiek, J., & Mukherjee, D. (2015). Glucose levels and depression in Hispanic patients admitted to the cardiovascular intensive care unit: A cross-sectional study. Angiology, 66(1), 57-64.

Badia‐Elder, N. E., Stewart, R., Powrozek, T. A., Murphy, J., & Li, T. K. (2003). Effects of neuropeptide Y on sucrose and ethanol intake and on anxiety‐like

377

behavior in high alcohol drinking (HAD) and low alcohol drinking (LAD) rats. Alcoholism: Clinical and Experimental Research, 27(6), 894-899.

Bailey, A., Gianotti, R., Ho, A., & Kreek, M. J. (2005). Persistent upregulation of μ‐opioid, but not adenosine, receptors in brains of long‐term withdrawn escalating dose “binge” cocaine‐treated rats. Synapse, 57(3), 160-166.

Baker, K. D., Bisby, M. A., & Richardson, R. (2016). Impaired fear extinction in adolescent rodents: behavioural and neural analyses. Neuroscience & Biobehavioral Reviews, 70, 59-73.

Baker, K. D., & Reichelt, A. C. (2016). Impaired fear extinction retention and increased anxiety-like behaviours induced by limited daily access to a high-fat/high-sugar diet in male rats: Implications for diet-induced prefrontal cortex dysregulation. Neurobiology of learning and memory, 136, 127-138.

Bakke, A. J., Stubbs, C. A., McDowell, E. H., Moding, K. J., Johnson, S. L., & Hayes, J. E. (2018). Mary Poppins was right: Adding small amounts of sugar or salt reduces the bitterness of vegetables. Appetite, 126, 90-101.

Balaz, M. A., Capra, S., Hartl, P., & Miller, R. R. (1981). Contextual potentiation of acquired behavior after devaluing direct context-US associations. Learning and Motivation, 12(4), 383-397.

Balaz, M. A., Capra, S., Kasprow, W. J., & Miller, R. R. (1982). Latent inhibition of the conditioning context: Further evidence of contextual potentiation of retrieval in the absence of appreciable context-US associations. Animal Learning & Behavior, 10(2), 242-248.

Bansal, R., Hellerstein, D. J., & Peterson, B. S. (2018). Evidence for neuroplastic compensation in the cerebral cortex of persons with depressive illness. Molecular psychiatry, 23(2), 375.

Barbera, G., Liang, B., Zhang, L., Gerfen, C. R., Culurciello, E., Chen, R., . . . Lin, D.-T. (2016). Spatially compact neural clusters in the dorsal striatum encode locomotion relevant information. Neuron, 92(1), 202-213.

Barnes, J. N., & Joyner, M. J. (2012). Sugar highs and lows: the impact of diet on cognitive function. The Journal of physiology, 590(12), 2831-2831.

Baron, D., Blum, K., Chen, A., Gold, M., & Badgaiyan, R. D. (2018). Conceptualizing Addiction From an Osteopathic Perspective: Dopamine Homeostasis. The Journal of the American Osteopathic Association, 118(2), 115-118.

Barot, S. K., Chung, A., Kim, J. J., & Bernstein, I. L. (2009). Functional Imaging of Stimulus Convergence in Amygdalar Neurons during Pavlovian Fear Conditioning. PLoS One, 4(7), e6156. doi:10.1371/journal.pone.0006156

http://dx.doi.org/10.1371/journal.pone.0006156 Barrientos, C., Knowland, D., Wu, M. M., Lilascharoen, V., Huang, K. W., Malenka,

R. C., & Lim, B. K. (2018). Cocaine induced structural plasticity in input regions to distinct cell types in nucleus accumbens. Biological psychiatry.

Barrientos, R. M., Sprunger, D. B., Campeau, S., Watkins, L. R., Rudy, J. W., & Maier, S. F. (2004). BDNF mRNA expression in rat hippocampus following contextual learning is blocked by intrahippocampal IL-1β administration. Journal of neuroimmunology, 155(1), 119-126. doi:https://doi.org/10.1016/j.jneuroim.2004.06.009

Barte, J., Ter Bogt, N., Bogers, R., Teixeira, P., Blissmer, B., Mori, T., & Bemelmans, W. (2010). Maintenance of weight loss after lifestyle interventions for overweight and obesity, a systematic review. obesity reviews, 11(12), 899-906.

378

Batterink, L., Yokum, S., & Stice, E. (2010). Body mass correlates inversely with inhibitory control in response to food among adolescent girls: an fMRI study. Neuroimage, 52(4), 1696-1703.

Bauer, E. P. (2015). Serotonin in fear conditioning processes. Behavioural Brain Research, 277, 68-77.

Bauer, E. P., Schafe, G. E., & LeDoux, J. E. (2002). NMDA receptors and L-type voltage-gated calcium channels contribute to long-term potentiation and different components of fear memory formation in the lateral amygdala. Journal of Neuroscience, 22(12), 5239-5249.

Beckers, T., Krypotos, A.-M., Boddez, Y., Effting, M., & Kindt, M. (2013). What's wrong with fear conditioning? Biological Psychology, 92(1), 90-96. doi:http://dx.doi.org/10.1016/j.biopsycho.2011.12.015

Beczkowska, I. W., Bowen, W. D., & Bodnar, R. J. (1992). Central opioid receptor subtype antagonists differentially alter sucrose and deprivation-induced water intake in rats. Brain research, 589(2), 291-301.

Beesdo, K., Knappe, S., & Pine, D. S. (2009). Anxiety and anxiety disorders in children and adolescents: developmental issues and implications for DSM-V. Psychiatric Clinics, 32(3), 483-524.

Begg, S., Vos, T., Barker, B., Stevenson, C., Stanley, L., & Lopez, A. D. (2007). The burden of disease and injury in Australia 2003.

Beilharz, J. E., Maniam, J., & Morris, M. J. (2014). Short exposure to a diet rich in both fat and sugar or sugar alone impairs place, but not object recognition memory in rats. Brain, behavior, and immunity, 37, 134-141.

Bekinschtein, P., Cammarota, M., Igaz, L. M., Bevilaqua, L. R. M., Izquierdo, I., & Medina, J. H. (2007). Persistence of Long-Term Memory Storage Requires a Late Protein Synthesis- and BDNF- Dependent Phase in the Hippocampus. Neuron, 53(2), 261-277. doi:http://dx.doi.org/10.1016/j.neuron.2006.11.025

Belin, D., & Dalley, J. W. (2012). Animal models in addiction research Drug Abuse and Addiction in Medical Illness (pp. 73-93): Springer.

Bello, E. P., Mateo, Y., Gelman, D. M., Noaín, D., Shin, J. H., Low, M. J., . . . Rubinstein, M. (2011). Cocaine supersensitivity and enhanced motivation for reward in mice lacking dopamine D 2 autoreceptors. Nature neuroscience, 14(8), 1033.

Bello, N. T., Lucas, L. R., & Hajnal, A. (2002). Repeated sucrose access influences dopamine D2 receptor density in the striatum. Neuroreport, 13(12), 1575.

Belmer, A., Beecher, K., Jacques, A., Patkar, O. L., Sicherre, F., & Bartlett, S. E. (2019). Axonal nonsegregation of the Vesicular Glutamate Transporter VGLUT3 within serotonergic projections in the mouse forebrain. Frontiers in cellular neuroscience, 13, 193.

Belmer, A., Klenowski, P. M., Patkar, O. L., & Bartlett, S. E. (2017). Mapping the connectivity of serotonin transporter immunoreactive axons to excitatory and inhibitory neurochemical synapses in the mouse limbic brain. Brain Structure and Function, 222(3), 1297-1314.

Belzeaux, R., Lalanne, L., Kieffer, B. L., & Lutz, P.-E. (2018). Focusing on the Opioid System for Addiction Biomarker Discovery. Trends in molecular medicine, 24(2), 206-220.

Benarroch, E. E. (2012). Endogenous opioid systems Current concepts and clinical correlations. Neurology, 79(8), 807-814.

379

Benjamini, Y., & Hochberg, Y. (1995). Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the royal statistical society. Series B (Methodological), 289-300.

Benoit, S. C., Tracy, A. L., Davis, J. F., Choi, D., & Clegg, D. J. (2008). Novel functions of orexigenic hypothalamic peptides: from genes to behavior. Nutrition, 24(9), 843-847.

Benowitz, N. L. (1988). Pharmacologic aspects of cigarette smoking and nicotine addiction. New England journal of medicine, 319(20), 1318-1330.

Benton, D. (2010). The plausibility of sugar addiction and its role in obesity and eating disorders. Clinical Nutrition, 29(3), 288-303.

Bergstrom, H. C. (2016). The neurocircuitry of remote cued fear memory. Neurosci Biobehav Rev, 71, 409-417. doi:10.1016/j.neubiorev.2016.09.028

Bergstrom, H. C. (2016). The neurocircuitry of remote cued fear memory. Neuroscience & Biobehavioral Reviews, 71, 409-417. doi:http://dx.doi.org/10.1016/j.neubiorev.2016.09.028

Bergstrom, H. C., & Johnson, L. R. (2014). An organization of visual and auditory fear conditioning in the lateral amygdala. Neurobiology of learning and memory, 116, 1-13.

Bergstrom, H. C., & Johnson, L. R. (2014). An organization of visual and auditory fear conditioning in the lateral amygdala. Neurobiol Learn Mem, 116, 1-13. doi:10.1016/j.nlm.2014.07.008

Bergstrom, H. C., McDonald, C. G., Dey, S., Fernandez, G. M., & Johnson, L. R. (2013). Neurons Activated During Fear Memory Consolidation and Reconsolidation are Mapped to a Common and New Topography in the Lateral Amygdala. Brain topography, 26(3), 468-478.

Bergstrom, H. C., McDonald, C. G., Dey, S., Fernandez, G. M., & Johnson, L. R. (2013). Neurons activated during fear memory consolidation and reconsolidation are mapped to a common and new topography in the lateral amygdala. Brain Topogr, 26(3), 468-478. doi:10.1007/s10548-012-0266-6

Bergstrom, H. C., McDonald, C. G., Dey, S., Tang, H., Selwyn, R. G., & Johnson, L. R. (2012). The structure of Pavlovian fear conditioning in the amygdala. Brain Struct Funct, 218(6), 1569-1589. doi:10.1007/s00429-012-0478-2

Bergstrom, H. C., McDonald, C. G., Dey, S., Tang, H., Selwyn, R. G., & Johnson, L. R. (2013). The structure of Pavlovian fear conditioning in the amygdala. Brain Structure and Function, 218(6), 1569-1589.

Bergstrom, H. C., McDonald, C. G., French, H. T., & Smith, R. F. (2008). Continuous nicotine administration produces selective, age-dependent structural alteration of pyramidal neurons from prelimbic cortex. Synapse, 62(1), 31-39. doi:10.1002/syn.20467

Bergstrom, H. C., McDonald, C. G., & Johnson, L. R. (2011). Pavlovian fear conditioning activates a common pattern of neurons in the lateral amygdala of individual brains. PloS one, 6(1), e15698.

Bergstrom, J. C. R., Olmsted-Hawala, E. L., & Bergstrom, H. C. (2016). Older adults fail to see the periphery in a Web site task. Universal Access in the Information Society, 15(2), 261-270. doi:10.1007/s10209-014-0382-z

Berlucchi, G., & Buchtel, H. A. (2009). Neuronal plasticity: historical roots and evolution of meaning. Experimental brain research, 192(3), 307-319.

Berridge, K. C. (1996). Food reward: brain substrates of wanting and liking. Neuroscience & Biobehavioral Reviews, 20(1), 1-25.

380

Berridge, K. C., Ho, C.-Y., Richard, J. M., & DiFeliceantonio, A. G. (2010). The tempted brain eats: pleasure and desire circuits in obesity and eating disorders. Brain research, 1350, 43-64.

Berthoud, H.-R., Münzberg, H., & Morrison, C. D. (2017). Blaming the brain for obesity: integration of hedonic and homeostatic mechanisms. Gastroenterology, 152(7), 1728-1738.

Besnard, A., Laroche, S., & Caboche, J. (2014). Comparative dynamics of MAPK/ERK signalling components and immediate early genes in the hippocampus and amygdala following contextual fear conditioning and retrieval. Brain Structure and Function, 219(1), 415-430. doi:http://dx.doi.org/10.1007/s00429-013-0505-y

Bezchlibnyk-Butler, K., Aleksic, I., & Kennedy, S. H. (2000). Citalopram--a review of pharmacological and clinical effects. Journal of Psychiatry and Neuroscience, 25(3), 241.

Bianchi, P. C., de Oliveira, P. E. C., Palombo, P., Leão, R. M., Cogo-Moreira, H., da Silva Planeta, C., & Cruz, F. C. (2018). Functional inactivation of the orbitofrontal cortex disrupts context-induced reinstatement of alcohol seeking in rats. Drug and alcohol dependence, 186, 102-112.

Bienkowski, P., Rogowski, A., Korkosz, A., Mierzejewski, P., Radwanska, K., Kaczmarek, L., . . . Kostowski, W. (2004). Time-dependent changes in alcohol-seeking behaviour during abstinence. European Neuropsychopharmacology, 14(5), 355-360.

Bito‐Onon, J. J., Simms, J. A., Chatterjee, S., Holgate, J., & Bartlett, S. E. (2011). Varenicline, a partial agonist at neuronal nicotinic acetylcholine receptors, reduces nicotine‐induced increases in 20% ethanol operant self‐administration in Sprague‐Dawley rats. Addiction biology, 16(3), 440-449.

BjÖrntorp, P. (2000). Hypertension and the metabolic syndrome: closely related central origin? Blood pressure, 9(2-3), 71-82.

Blanchard, R. J., & Blanchard, D. C. (1969). Passive and active reactions to fear-eliciting stimuli. Journal of comparative and physiological psychology, 68(1p1), 129.

Bloodgood, D. W., Sugam, J. A., Holmes, A., & Kash, T. L. (2018). Fear extinction requires infralimbic cortex projections to the basolateral amygdala. Translational psychiatry, 8(1), 60.

Bocarsly, M. E. (2018). Pharmacological Interventions for Obesity: Current and Future Targets. Current Addiction Reports, 5(2), 202-211.

Bonin, R. P., Dubins, D., Mogil, J., Lecker, I., Mercik, A., & D'souza, A. (2018). Systems, methods and apparatus for rodent behavioural monitoring: Google Patents.

Bordet, R., Ridray, S., Carboni, S., Diaz, J., Sokoloff, P., & Schwartz, J.-C. (1997). Induction of dopamine D3 receptor expression as a mechanism of behavioral sensitization to levodopa. Proceedings of the National Academy of Sciences, 94(7), 3363-3367.

Boschen, M. J., Neumann, D. L., & Waters, A. M. (2009). Relapse of successfully treated anxiety and fear: theoretical issues and recommendations for clinical practice. Australian & New Zealand Journal of Psychiatry, 43(2), 89-100. doi:10.1080/00048670802607154

Bostock, E., Muller, R. U., & Kubie, J. L. (1991). Experience‐dependent modifications of hippocampal place cell firing. Hippocampus, 1(2), 193-205.

381

Bouchet, C. A., Miner, M. A., Loetz, E. C., Rosberg, A. J., Hake, H. S., Farmer, C. E., . . . Greenwood, B. N. (2018). Activation of nigrostriatal dopamine neurons during fear extinction prevents the renewal of fear. Neuropsychopharmacology, 43(3), 665.

Bouton, M. E. (1988). Context and ambiguity in the extinction of emotional learning: Implications for exposure therapy. Behaviour research and therapy, 26(2), 137-149. doi:http://dx.doi.org/10.1016/0005-7967(88)90113-1

Bouton, M. E. (1993). Context, time, and memory retrieval in the interference paradigms of Pavlovian learning. Psychological Bulletin, 114(1), 80-99. doi:10.1037/0033-2909.114.1.80

Bouton, M. E. (2004). Context and behavioral processes in extinction. Learning & Memory, 11(5), 485-494.

Bouton, M. E., & Bolles, R. C. (1979). Role of conditioned contextual stimuli in reinstatement of extinguished fear. Journal of Experimental Psychology: Animal Behavior Processes, 5(4), 368.

Bouton, M. E., & King, D. A. (1983). Contextual control of the extinction of conditioned fear: tests for the associative value of the context. Journal of Experimental Psychology: Animal Behavior Processes, 9(3), 248.

Bouton, M. E., Mineka, S., & Barlow, D. H. (2001). A modern learning theory perspective on the etiology of panic disorder. Psychological review, 108(1), 4.

Bouton, M. E., & Peck, C. A. (1989). Context effects on conditioning, extinction, and reinstatement in an appetitive conditioning preparation. Animal Learning & Behavior, 17(2), 188-198.

Bouton, M. E., Westbrook, R. F., Corcoran, K. A., & Maren, S. (2006). Contextual and temporal modulation of extinction: behavioral and biological mechanisms. Biological psychiatry, 60(4), 352-360.

Brady, J. V., Schreiner, L., Geller, I., & Kling, A. (1954). Subcortical mechanisms in emotional behavior: The effect of rhinencephalic injury upon the acquisition and retention of a conditioned avoidance response in cats. Journal of comparative and physiological psychology, 47(3), 179.

Brambilla, R., Gnesutta, N., Minichiello, L., White, G., Roylance, A. J., Herron, C. E., . . . Rossi-Arnaud, C. (1997). A role for the Ras signalling pathway in synaptic transmission and long-term memory. Nature, 390(6657), 281.

Bray, G. A., & Bouchard, C. (2014). Handbook of Obesity–Volume 2: Clinical Applications (Vol. 2): CRC Press.

Bredy, T. W., Wu, H., Crego, C., Zellhoefer, J., Sun, Y. E., & Barad, M. (2007). Histone modifications around individual BDNF gene promoters in prefrontal cortex are associated with extinction of conditioned fear. Learning & Memory, 14(4), 268-276.

Brewerton, T. D. (2011). Posttraumatic stress disorder and disordered eating: food addiction as self-medication: Mary Ann Liebert, Inc. 140 Huguenot Street, 3rd Floor New Rochelle, NY 10801 USA.

Brooks, D. C., & Bouton, M. E. (1993). A retrieval cue for extinction attenuates spontaneous recovery. Journal of Experimental Psychology: Animal Behavior Processes, 19(1), 77.

Brown, R. J., & Rother, K. I. (2012). Non-nutritive sweeteners and their role in the gastrointestinal tract. The Journal of Clinical Endocrinology & Metabolism, 97(8), 2597-2605.

382

Bryant, R., Felmingham, K., Kemp, A., Das, P., Hughes, G., Peduto, A., & Williams, L. (2008). Amygdala and ventral anterior cingulate activation predicts treatment response to cognitive behaviour therapy for post-traumatic stress disorder. Psychological medicine, 38(4), 555-561.

Bryant, R. A., Felmingham, K., Whitford, T. J., Kemp, A., Hughes, G., Peduto, A., & Williams, L. M. (2008). Rostral anterior cingulate volume predicts treatment response to cognitive-behavioural therapy for posttraumatic stress disorder. Journal of psychiatry & neuroscience: JPN, 33(2), 142.

Buchanan, T. W., & Lovallo, W. R. (2001). Enhanced memory for emotional material following stress-level cortisol treatment in humans. Psychoneuroendocrinology, 26(3), 307-317.

Bukalo, O., Pinard, C. R., Silverstein, S., Brehm, C., Hartley, N. D., Whittle, N., . . . Singewald, N. (2015). Prefrontal inputs to the amygdala instruct fear extinction memory formation. Science advances, 1(6), e1500251.

Burgos-Robles, A., Vidal-Gonzalez, I., Santini, E., & Quirk, G. J. (2007). Consolidation of fear extinction requires NMDA receptor-dependent bursting in the ventromedial prefrontal cortex. Neuron, 53(6), 871-880.

Burguera, B., Ali, K., & Brito, J. (2017). Antiobesity drugs in the management of type 2 diabetes: A shift in thinking? Cleveland Clinic journal of medicine, 84(7 Suppl 1), S39-S46.

Burnett, C. J., Li, C., Webber, E., Tsaousidou, E., Xue, S. Y., Brüning, J. C., & Krashes, M. J. (2016). Hunger-driven motivational state competition. Neuron, 92(1), 187-201.

Butovsky, O., Jedrychowski, M. P., Moore, C. S., Cialic, R., Lanser, A. J., Gabriely, G., . . . Doykan, C. E. (2014). Identification of a unique TGF-β–dependent molecular and functional signature in microglia. Nature neuroscience, 17(1), 131.

Cabanac, M., & Johnson, K. (1983). Analysis of a conflict between palatability and cold exposure in rats. Physiology & behavior, 31(2), 249-253.

Cajal, S. R. Y. (1894). The Croonian Lecture: La fine structure des centres nerveux. Proceedings of the Royal Society of London, 55(331-335), 444-468.

Calandreau, L., Desmedt, A., Decorte, L., & Jaffard, R. (2005). A different recruitment of the lateral and basolateral amygdala promotes contextual or elemental conditioned association in Pavlovian fear conditioning. Learning & Memory, 12(4), 383-388.

Calandreau, L., Trifilieff, P., Mons, N., Costes, L., Marien, M., Marighetto, A., . . . Desmedt, A. (2006). Extracellular Hippocampal Acetylcholine Level Controls Amygdala Function and Promotes Adaptive Conditioned Emotional Response. The Journal of neuroscience, 26(52), 13556-13566. doi:10.1523/jneurosci.3713-06.2006

Calcia, M. A., Bonsall, D. R., Bloomfield, P. S., Selvaraj, S., Barichello, T., & Howes, O. D. (2016). Stress and neuroinflammation: a systematic review of the effects of stress on microglia and the implications for mental illness. Psychopharmacology, 233(9), 1637-1650. doi:10.1007/s00213-016-4218-9

Caliendo, G., Santagada, V., Perissutti, E., & Fiorino, F. (2005). Derivatives as 5HT1A receptor ligands--past and present. Curr Med Chem, 12(15), 1721-1753.

Calipari, E. S., Sun, H., Eldeeb, K., Luessen, D. J., Feng, X., Howlett, A. C., . . . Chen, R. (2014). Amphetamine self-administration attenuates dopamine D2 autoreceptor function. Neuropsychopharmacology, 39(8), 1833.

383

Cannistraro, P. A., & Rauch, S. L. (2003). Neural circuitry of anxiety: evidence from structural and functional neuroimaging studies. Psychopharmacology bulletin, 37(4), 8-25.

Cannon, W. B. (1935). Stress and strains of homeostasis. American Journal of Medical Science, 189, 1-14.

Cao, D., Lu, H., Lewis, T. L., & Li, L. (2007). Intake of sucrose-sweetened water induces insulin resistance and exacerbates memory deficits and amyloidosis in a transgenic mouse model of Alzheimer disease. Journal of Biological Chemistry, 282(50), 36275-36282.

Capello, A. E., & Markus, C. R. (2014). Differential influence of the 5-HTTLPR genotype, neuroticism and real-life acute stress exposure on appetite and energy intake. Appetite, 77, 83-93. doi:10.1016/j.appet.2014.03.002

Carr, D., & Kalivas, P. W. (2006). Orexin: a gatekeeper of addiction. Nature medicine, 12(3), 274.

Caruana, D. A., Alexander, G. M., & Dudek, S. M. (2012). New insights into the regulation of synaptic plasticity from an unexpected place: hippocampal area CA2. Learning & Memory, 19(9), 391-400.

Castanares, M., Stuart, G. J., & Daria, V. (2019). Holographic Functional Calcium Imaging of Neuronal Circuit Activity Advanced Optical Methods for Brain Imaging (pp. 143-165): Springer.

Cestari, V., Rossi-Arnaud, C., Saraulli, D., & Costanzi, M. (2014). The MAP (K) of fear: from memory consolidation to memory extinction. Brain research bulletin, 105, 8-16.

Cha, S. H., Wolfgang, M., Tokutake, Y., Chohnan, S., & Lane, M. D. (2008). Differential effects of central fructose and glucose on hypothalamic malonyl–CoA and food intake. Proceedings of the National Academy of Sciences.

Chaaya, N., Battle, A. R., & Johnson, L. R. (2018). An update on contextual fear memory mechanisms: Transition between Amygdala and Hippocampus. Neuroscience & Biobehavioral Reviews, 92, 43-54. doi:https://doi.org/10.1016/j.neubiorev.2018.05.013

Chaaya, N., Battle, A. R., & Johnson, L. R. (2018, In Press). An Update on Contextual Fear Memory Mechanisms: Transition between Amygdala and Hippocampus. Neuroscience & Biobehavioral Reviews. doi:https://doi.org/10.1016/j.neubiorev.2018.05.013

Chaaya, N., Jacques, A., Battle, A. R., & Johnson, L. R. (2017). Consolidation of contextual and auditory fear memories within amygdala, hippocampus and prefrontal cortex. Abstract 749.22. Society for Neuroscience Abstracts.

Chaaya, N., Jacques, A., Belmer, A., Beecher, K., Ali, S. A., Chehrehasa, F., . . . Bartlett, S. E. (2019). Contextual Fear Conditioning Alter Microglia Number and Morphology in the Rat Dorsal Hippocampus. Frontiers in cellular neuroscience, 13, 214.

Chaaya, N., Jacques, A., Belmer, A., Richard, D. J., Bartlett, S. E., Battle, A. R., & Johnson, L. R. (2019). Localization of Contextual and Context Removed Auditory Fear Memory within the Basolateral Amygdala Complex. Neuroscience, 398, 231-251. doi:https://doi.org/10.1016/j.neuroscience.2018.12.004

Chambers, W., Liu, C., & McCouch, G. (1973). Anatomical and physiological correlates of plasticity in the central nervous system. Brain, behavior and evolution, 8(1-2), 5-26.

384

Chao, E. C., & Henry, R. R. (2010). SGLT2 inhibition—a novel strategy for diabetes treatment. Nature reviews Drug discovery, 9(7), 551.

Chapin, J. K., & Nicolelis, M. A. (1999). Principal component analysis of neuronal ensemble activity reveals multidimensional somatosensory representations. Journal of neuroscience methods, 94(1), 121-140.

Cheng, W., Li, Y., Hou, X., Bai, B., Li, F., Ding, F., . . . Wang, Y. (2015). Determining the neuroprotective effects of dextromethorphan in

lipopolysaccharide‑stimulated BV2 microglia. Molecular medicine reports, 11(2), 1132-1138.

Chepulis, L. M., Starkey, N. J., Waas, J. R., & Molan, P. C. (2009). The effects of long-term honey, sucrose or sugar-free diets on memory and anxiety in rats. Physiology & behavior, 97(3-4), 359-368.

Chikama, K., Yamada, H., Tsukamoto, T., Kajitani, K., Nakabeppu, Y., & Uchimura, N. (2017). Chronic atypical antipsychotics, but not haloperidol, increase neurogenesis in the hippocampus of adult mouse. Brain Res, 1676, 77-82. doi:10.1016/j.brainres.2017.09.006

Choi, M., Ahn, S., Yang, E.-J., Kim, H., Chong, Y. H., & Kim, H.-S. (2016). Hippocampus-based contextual memory alters the morphological characteristics of astrocytes in the dentate gyrus. Molecular brain, 9, 72. doi:10.1186/s13041-016-0253-z

Chowdhury, S., Shepherd, J. D., Okuno, H., Lyford, G., Petralia, R. S., Plath, N., . . . Worley, P. F. (2006). Arc/Arg3. 1 interacts with the endocytic machinery to regulate AMPA receptor trafficking. Neuron, 52(3), 445-459.

Clifton, P. G. (2017). Neural circuits of eating behaviour: Opportunities for therapeutic development. Journal of Psychopharmacology, 31(11), 1388-1402.

Cocores, J. A., & Gold, M. S. (2008). Varenicline, appetite, and weight reduction. The Journal of neuropsychiatry and clinical neurosciences, 20(4), 497-498.

Cohen, J. F., Rifas-Shiman, S. L., Young, J., & Oken, E. (2018). Associations of Prenatal and Child Sugar Intake With Child Cognition. American journal of preventive medicine, 54(6), 727-735.

Colantuoni, C., Rada, P., McCarthy, J., Patten, C., Avena, N. M., Chadeayne, A., & Hoebel, B. G. (2002). Evidence that intermittent, excessive sugar intake causes endogenous opioid dependence. Obesity, 10(6), 478-488.

Colantuoni, C., Schwenker, J., McCarthy, J., Rada, P., Ladenheim, B., Cadet, J.-L., . . . Hoebel, B. (2001). Excessive sugar intake alters binding to dopamine and mu-opioid receptors in the brain. Neuroreport, 12(16), 3549-3552.

Comings, D., Gade‐Andavolu, R., Gonzalez, N., Wu, S., Muhleman, D., Chen, C. e., . . . Dietz, G. (2001). The additive effect of neurotransmitter genes in pathological gambling. Clinical genetics, 60(2), 107-116.

Cools, R., Barker, R. A., Sahakian, B. J., & Robbins, T. W. (2003). L-Dopa medication remediates cognitive inflexibility, but increases impulsivity in patients with Parkinson’s disease. Neuropsychologia, 41(11), 1431-1441.

Coplan, J. D., Webler, R., Gopinath, S., Abdallah, C. G., & Mathew, S. J. (2018). Neurobiology of the dorsolateral prefrontal cortex in GAD: Aberrant neurometabolic correlation to hippocampus and relationship to anxiety sensitivity and IQ. Journal of affective disorders, 229, 1-13.

Corcoran, K. A., & Quirk, G. J. (2007). Activity in prelimbic cortex is necessary for the expression of learned, but not innate, fears. Journal of Neuroscience, 27(4), 840-844.

385

Cornea‐Hébert, V., Riad, M., Wu, C., Singh, S. K., & Descarries, L. (1999). Cellular and subcellular distribution of the serotonin 5‐HT2A receptor in the central nervous system of adult rat. Journal of Comparative Neurology, 409(2), 187-209.

Corsica, J. A., & Pelchat, M. L. (2010). Food addiction: true or false? Current opinion in gastroenterology, 26(2), 165-169.

Corsini, R. J. (1999). The dictionary of psychology: Psychology Press. Cortese, B. M., & Phan, K. L. (2005). The role of glutamate in anxiety and related

disorders. CNS spectrums, 10(10), 820-830. Covington, H. E., & Miczek, K. A. (2001). Repeated social-defeat stress, cocaine or

morphine. Psychopharmacology, 158(4), 388-398. Cox, J., & Westbrook, R. F. (1994). The NMDA receptor antagonist MK-801 blocks

acquisition and extinction of conditioned hypoalgesic responses in the rat. Q J Exp Psychol B, 47(2), 187-210.

Criscitelli, K., & Avena, N. M. (2017). 5 Sugar and Fat Addiction. Processed Food Addiction: Foundations, Assessment, and Recovery.

Curran, A. T., Miller, D., Zokas, L., & Verma, I. M. (1984). Viral and cellular fos proteins: a comparative analysis. Cell, 36(2), 259-268.

D'Anci, K. E., Kanarek, R. B., & Marks-Kaufman, R. (1996). Duration of sucrose availability differentially alters morphine-induced analgesia in rats. Pharmacology Biochemistry and Behavior, 54(4), 693-697.

D'Argenio, A., Mazzi, C., Pecchioli, L., Di Lorenzo, G., Siracusano, A., & Troisi, A. (2009). Early trauma and adult obesity: is psychological dysfunction the mediating mechanism? Physiology & behavior, 98(5), 543-546.

Dadkhah, M., Abdullahi, P. R., Rashidy-Pour, A., Sameni, H. R., & Vafaei, A. A. (2018). Infralimbic dopamine D2 receptors mediate glucocorticoid-induced facilitation of auditory fear memory extinction in rats. Brain Res, 1682, 84-92. doi:10.1016/j.brainres.2018.01.006

Dai, J.-X., Han, H.-L., Tian, M., Cao, J., Xiu, J.-B., Song, N.-N., . . . Xu, L. (2008). Enhanced contextual fear memory in central serotonin-deficient mice. Proceedings of the National Academy of Sciences.

Dallman, M. F., Akana, S. F., Laugero, K. D., Gomez, F., Manalo, S., Bell, M., & Bhatnagar, S. (2003). A spoonful of sugar: feedback signals of energy stores and corticosterone regulate responses to chronic stress. Physiology & behavior, 79(1), 3-12.

Davidson, T. L., Chan, K., Jarrard, L. E., Kanoski, S. E., Clegg, D. J., & Benoit, S. C. (2009). Contributions of the hippocampus and medial prefrontal cortex to energy and body weight regulation. Hippocampus, 19(3), 235-252.

Davis, P., & Reijmers, L. G. (2017). The dynamic nature of fear engrams in the basolateral amygdala. Brain Res Bull. doi:10.1016/j.brainresbull.2017.12.004

de Jong, J. W. (2015). Eating addiction? The nerves and fibers that control food intake. Utrecht University.

De Jong, J. W., Vanderschuren, L. J., & Adan, R. A. (2016). The mesolimbic system and eating addiction: What sugar does and does not do. Current Opinion in Behavioral Sciences, 9, 118-125.

De Kloet, E. R., Joëls, M., & Holsboer, F. (2005). Stress and the brain: from adaptation to disease. Nature reviews neuroscience, 6(6), 463.

De Nijs, J., van Opstal, D. P., Janssen, R. J., Cahn, W., Schnack, H., & Investigators, G. (2018). O9. 7. INDIVIDUALIZED LONG-TERM OUTCOME PREDICTION OF PSYCHOSIS IN AN OBSERVATIONAL STUDY: A

386

MACHINE LEARNING APPROACH. Schizophrenia Bulletin, 44(suppl_1), S101-S102.

De Smith, M. J., Goodchild, M. F., & Longley, P. (2007). Geospatial analysis: a comprehensive guide to principles, techniques and software tools: Troubador Publishing Ltd.

De Smith, M. J., Goodchild, M. F., & Longley, P. A. (2009). Geospatial analysis. Matador.

DeFina, P., Fellus, J., Polito, M. Z., Thompson, J. W., Moser, R. S., & DeLuca, J. (2009). The new neuroscience frontier: promoting neuroplasticity and brain repair in traumatic brain injury. The Clinical Neuropsychologist, 23(8), 1391-1399.

Delamater, A. R., & Westbrook, R. F. (2014). Psychological and neural mechanisms of experimental extinction: a selective review. Neurobiology of learning and memory, 108, 38-51.

Denis, R. G., Joly-Amado, A., Webber, E., Langlet, F., Schaeffer, M., Padilla, S. L., . . . Delbès, A.-S. (2015). Palatability can drive feeding independent of AgRP neurons. Cell metabolism, 22(4), 646-657.

Deppermann, S., Storchak, H., Fallgatter, A., & Ehlis, A.-C. (2014). Stress-induced neuroplasticity:(mal) adaptation to adverse life events in patients with PTSD–a critical overview. Neuroscience, 283, 166-177.

Derman, R. C., & Ferrario, C. R. (2017). Enhanced incentive motivation in obesity-prone rats is mediated by NAc core CP-AMPARs. Neuropharmacology.

Deschaux, O., Motanis, H., Spennato, G., Moreau, J.-L., & Garcia, R. (2011). Re-emergence of extinguished auditory-cued conditioned fear following a sub-conditioning procedure: effects of hippocampal and prefrontal tetanic stimulations. Neurobiology of learning and memory, 95(4), 510-518.

Desmedt, A., Garcia, R., & Jaffard, R. (1998). Differential Modulation of Changes in Hippocampal–Septal Synaptic Excitability by the Amygdala as a Function of Either Elemental or Contextual Fear Conditioning in Mice. The Journal of neuroscience, 18(1), 480-487.

Di Chiara, G., & Imperato, A. (1988). Drugs abused by humans preferentially increase synaptic dopamine concentrations in the mesolimbic system of freely moving rats. Proceedings of the National Academy of Sciences, 85(14), 5274-5278.

Díaz-Aguila, Y., Castelán, F., Cuevas, E., Zambrano, E., Martínez-Gómez, M., Muñoz, A., . . . Nicolás-Toledo, L. (2016). Consumption of sucrose from infancy increases the visceral fat accumulation, concentration of triglycerides, insulin and leptin, and generates abnormalities in the adrenal gland. Anatomical science international, 91(2), 151-162.

Dimyan, M. A., & Cohen, L. G. (2011). Neuroplasticity in the context of motor rehabilitation after stroke. Nature Reviews Neurology, 7(2), 76.

Do-Monte, F. H., Manzano-Nieves, G., Quiñones-Laracuente, K., Ramos-Medina, L., & Quirk, G. J. (2015). Revisiting the role of infralimbic cortex in fear extinction with optogenetics. Journal of Neuroscience, 35(8), 3607-3615.

Do-Monte, F. H., Quinones-Laracuente, K., & Quirk, G. J. (2015). A temporal shift in the circuits mediating retrieval of fear memory. Nature, 519(7544), 460.

Domingos, A. I., Sordillo, A., Dietrich, M. O., Liu, Z.-W., Tellez, L. A., Vaynshteyn, J., . . . de Araujo, I. E. (2013). Hypothalamic melanin concentrating hormone neurons communicate the nutrient value of sugar. Elife, 2, e01462.

387

Drevets, W. C. (1998). Functional neuroimaging studies of depression: the anatomy of melancholia. Annual review of medicine, 49(1), 341-361.

Drucker, D. J., & Nauck, M. A. (2006). The incretin system: glucagon-like peptide-1 receptor agonists and dipeptidyl peptidase-4 inhibitors in type 2 diabetes. The Lancet, 368(9548), 1696-1705.

Duvarci, S., & Pare, D. (2014). Amygdala microcircuits controlling learned fear. Neuron, 82(5), 966-980. doi:10.1016/j.neuron.2014.04.042

Duvarci, S., & Pare, D. (2014). Amygdala microcircuits controlling learned fear. Neuron, 82(5), 966-980.

Dwyer, J. B., & Ross, D. A. (2016). Modern Microglia: Novel Targets in Psychiatric Neuroscience. Biological psychiatry, 80(7), e47-e49. doi:https://doi.org/10.1016/j.biopsych.2016.08.006

Egeland, M., Zunszain, P. A., & Pariante, C. M. (2015). Molecular mechanisms in the regulation of adult neurogenesis during stress. Nat Rev Neurosci, 16(4), 189-200. doi:10.1038/nrn3855

Elzinga, B. M., & Bremner, J. D. (2002). Are the neural substrates of memory the final common pathway in posttraumatic stress disorder (PTSD)? Journal of affective disorders, 70(1), 1-17.

Etkin, A., & Wager, T. D. (2007). Functional neuroimaging of anxiety: a meta-analysis of emotional processing in PTSD, social anxiety disorder, and specific phobia. American Journal of Psychiatry, 164(10), 1476-1488.

Falconer, E., Bryant, R., Felmingham, K. L., Kemp, A. H., Gordon, E., Peduto, A., . . . Williams, L. M. (2008). The neural networks of inhibitory control in posttraumatic stress disorder. Journal of psychiatry & neuroscience: JPN, 33(5), 413.

Faliagkas, L., Rao-Ruiz, P., & Kindt, M. (2018). Emotional memory expression is misleading: delineating transitions between memory processes. Current Opinion in Behavioral Sciences, 19, 116-122.

Falls, W. A., Miserendino, M., & Davis, M. (1992). Extinction of fear-potentiated startle: blockade by infusion of an NMDA antagonist into the amygdala. Journal of Neuroscience, 12(3), 854-863.

Fanous, S., Guez‐Barber, D. H., Goldart, E. M., Schrama, R., Theberge, F. R., Shaham, Y., & Hope, B. T. (2013). Unique gene alterations are induced in FACS‐purified Fos‐positive neurons activated during cue‐induced relapse to heroin seeking. Journal of neurochemistry, 124(1), 100-108.

Fanselow, M. S. (1980). Conditional and unconditional components of post-shock freezing. The Pavlovian Journal of Biological Science: Official Journal of the Pavlovian, 15(4), 177-182.

Fanselow, M. S. (1984). What is conditioned fear? Trends in neurosciences, 7(12), 460-462.

Fanselow, M. S. (2010). From Contextual Fear to a Dynamic View of Memory Systems. Trends in cognitive sciences, 14(1), 7. doi:10.1016/j.tics.2009.10.008

Fanselow, M. S., & Gale, G. D. (2003). The Amygdala, Fear, and Memory. Annals of the New York Academy of Sciences, 985(1), 125-134. doi:10.1111/j.1749-6632.2003.tb07077.x

Feduccia, A., Simms, J., Mill, D., Yi, H., & Bartlett, S. (2014). Varenicline decreases ethanol intake and increases dopamine release via neuronal nicotinic acetylcholine receptors in the nucleus accumbens. British journal of pharmacology, 171(14), 3420-3431.

388

Fendt, á., & Fanselow, M. (1999). The neuroanatomical and neurochemical basis of conditioned fear. Neuroscience & Biobehavioral Reviews, 23(5), 743-760.

Fendt, M., & Fanselow, M. S. (1999). The neuroanatomical and neurochemical basis of conditioned fear. Neuroscience & Biobehavioral Reviews, 23(5), 743-760.

Ferrini, F., & De Koninck, Y. (2013). Microglia control neuronal network excitability via BDNF signalling. Neural plasticity, 2013.

Fidler, P., Kalman, B. A., Ziemer, H. E., & Green, K. F. (1993). Early onset of reduced morphine analgesia by ingestion of sweet solutions. Physiology & behavior, 53(1), 167-171.

Fioravanti, M., Polzonetti, C., Nocca, D., Spera, G., Falcone, S., Lazzari, R., & Colosimo, A. (2004). Emotional activation of obese and normal women due to imagery and food content of verbal stimuli in a memory task. Eating behaviors, 5(1), 47-54.

Fiorentini, A., & Müller, E. (1975). Sensitivity of central chemoreceptors controlling blood glucose and body temperature during glucose deprivation. The Journal of physiology, 248(2), 247-271.

Fitzmaurice, G., Laird, N., & Ware, J. Applied longitudinal analysis. 2004. Hoboken Wiley-Interscience.

Flagel, S. B., Clark, J. J., Robinson, T. E., Mayo, L., Czuj, A., Willuhn, I., . . . Akil, H. (2011). A selective role for dopamine in stimulus–reward learning. Nature, 469(7328), 53.

Flavell, C. R., Lambert, E., Winters, B. D., & Bredy, T. W. (2013). Mechanisms governing the reactivation-dependent destabilization of memories and their role in extinction. Frontiers in Behavioral Neuroscience, 7, 214.

Floresco, S. B., Montes, D. R., Maric, M., & van Holstein, M. (2018). Differential contributions of nucleus accumbens subregions to cue-guided risk/reward decision making and implementation of conditional rules. Journal of Neuroscience, 38(8), 1901-1914.

Foa, E. B., Dancu, C. V., Hembree, E. A., Jaycox, L. H., Meadows, E. A., & Street, G. P. (1999). A comparison of exposure therapy, stress inoculation training, and their combination for reducing posttraumatic stress disorder in female assault victims. Journal of consulting and clinical psychology, 67(2), 194.

Foa, E. B., Zinbarg, R., & Rothbaum, B. O. (1992). Uncontrollability and unpredictability in post-traumatic stress disorder: an animal model. Psychological bulletin, 112(2), 218.

Fogarty, M. J., Hammond, L. A., Kanjhan, R., Bellingham, M. C., & Noakes, P. G. (2013). A method for the three-dimensional reconstruction of Neurobiotin-filled neurons and the location of their synaptic inputs. Front Neural Circuits, 7, 153. doi:10.3389/fncir.2013.00153

Foley, K. A., Fudge, M. A., Kavaliers, M., & Ossenkopp, K.-P. (2006). Quinpirole-induced behavioral sensitization is enhanced by prior scheduled exposure to sucrose: A multi-variable examination of locomotor activity. Behavioural Brain Research, 167(1), 49-56.

Foo, H., & Mason, P. (2005). Sensory suppression during feeding. Proceedings of the National Academy of Sciences, 102(46), 16865-16869.

Fortuna, J. L. (2010). Sweet preference, sugar addiction and the familial history of alcohol dependence: shared neural pathways and genes. Journal of psychoactive drugs, 42(2), 147-151.

Fourrier, C., Singhal, G., & Baune, B. T. (2019). Neuroinflammation and cognition across psychiatric conditions. CNS spectrums, 1-12.

389

Frank, M. G., Baratta, M. V., Sprunger, D. B., Watkins, L. R., & Maier, S. F. (2007). Microglia serve as a neuroimmune substrate for stress-induced potentiation of CNS pro-inflammatory cytokine responses. Brain, behavior, and immunity, 21(1), 47-59. doi:https://doi.org/10.1016/j.bbi.2006.03.005

Frankland, P. W., Bontempi, B., Talton, L. E., Kaczmarek, L., & Silva, A. J. (2004). The involvement of the anterior cingulate cortex in remote contextual fear memory. science, 304(5672), 881-883.

Frankland, P. W., & Josselyn, S. A. (2015). Memory allocation. Neuropsychopharmacology, 40(1), 243. doi:10.1038/npp.2014.234

Fuchs, E., Czéh, B., Kole, M. H., Michaelis, T., & Lucassen, P. J. (2004). Alterations of neuroplasticity in depression: the hippocampus and beyond. European Neuropsychopharmacology, 14, S481-S490.

Gagnon, D., & Parent, M. (2014). Distribution of VGLUT3 in highly collateralized axons from the rat dorsal raphe nucleus as revealed by single-neuron reconstructions. PloS one, 9(2), e87709. doi:10.1371/journal.pone.0087709

Gangwisch, J. E., Hale, L., Garcia, L., Malaspina, D., Opler, M. G., Payne, M. E., . . . Lane, D. (2015). High glycemic index diet as a risk factor for depression: analyses from the Women’s Health Initiative. The American journal of clinical nutrition, 102(2), 454-463.

Gao, R., Asano, S. M., Upadhyayula, S., Pisarev, I., Milkie, D. E., Liu, T.-L., . . . Zhao, Y. (2019). Cortical column and whole-brain imaging with molecular contrast and nanoscale resolution. science, 363(6424), eaau8302.

García-Cáceres, C., Quarta, C., Varela, L., Gao, Y., Gruber, T., Legutko, B., ... & Le Thuc, O. . (2016). Astrocytic insulin signaling couples brain glucose uptake with nutrient availability. . Cell, 166(4), 867-880.

Gaysinskaya, V., Karatayev, O., Shuluk, J., & Leibowitz, S. (2011). Hyperphagia induced by sucrose: relation to circulating and CSF glucose and corticosterone and orexigenic peptides in the arcuate nucleus. Pharmacology Biochemistry and Behavior, 97(3), 521-530.

Gearhardt, A. N., Corbin, W. R., & Brownell, K. D. (2009). Preliminary validation of the Yale food addiction scale. Appetite, 52(2), 430-436.

Geiger, B. M., Behr, G. G., Frank, L. E., Caldera-Siu, A. D., Beinfeld, M. C., Kokkotou, E. G., & Pothos, E. N. (2008). Evidence for defective mesolimbic dopamine exocytosis in obesity-prone rats. The FASEB Journal, 22(8), 2740-2746.

Genovese, C. R., Lazar, N. A., & Nichols, T. (2002). Thresholding of statistical maps in functional neuroimaging using the false discovery rate. Neuroimage, 15(4), 870-878.

Gerber, Y. N., Saint-Martin, G. P., Bringuier, C. M., Bartolami, S., Goze-Bac, C., Noristani, H. N., & Perrin, F. E. (2018). CSF1R Inhibition Reduces Microglia Proliferation, Promotes Tissue Preservation and Improves Motor Recovery After Spinal Cord Injury. Frontiers in cellular neuroscience, 12.

Gerdeman, G. L., Partridge, J. G., Lupica, C. R., & Lovinger, D. M. (2003). It could be habit forming: drugs of abuse and striatal synaptic plasticity. Trends in neurosciences, 26(4), 184-192.

Giese, K. P., & Mizuno, K. (2013). The roles of protein kinases in learning and memory. Learning & Memory, 20(10), 540-552.

Gillies, C. L., Abrams, K. R., Lambert, P. C., Cooper, N. J., Sutton, A. J., Hsu, R. T., & Khunti, K. (2007). Pharmacological and lifestyle interventions to prevent

390

or delay type 2 diabetes in people with impaired glucose tolerance: systematic review and meta-analysis. Bmj, 334(7588), 299.

Gisquet-Verrier, P., Dutrieux, G., Richer, P., & Doyère, V. (1999). Effects of lesions to the hippocampus on contextual fear: evidence for a disruption of freezing and avoidance behavior but not context conditioning. Behavioral neuroscience, 113(3), 507.

Giustino, T. F., & Maren, S. (2015). The role of the medial prefrontal cortex in the conditioning and extinction of fear. Frontiers in Behavioral Neuroscience, 9, 298.

Goldbacher, E., La Grotte, C., Komaroff, E., Vander Veur, S., & Foster, G. D. (2016). An initial evaluation of a weight loss intervention for individuals who engage in emotional eating. Journal of behavioral medicine, 39(1), 139-150.

Gonzalez, J. T., & Betts, J. A. (2018). Dietary fructose metabolism by splanchnic organs: size matters. Cell metabolism, 27(3), 483-485.

Gosnell, B. (1987). Central structures involved in opioid-induced feeding. Paper presented at the Federation proceedings.

Gosnell, B. A. (2005). Sucrose intake enhances behavioral sensitization produced by cocaine. Brain research, 1031(2), 194-201.

Graham, M., Shutter, J. R., Sarmiento, U., & Stark, K. L. (1997). Overexpression of Agrt leads to obesity in transgenic mice. Nature genetics, 17(3), 273.

Grahame, N. J., Hallam, S. C., Geier, L., & Miller, R. R. (1990). Context as an occasion setter following either CS acquisition and extinction or CS acquisition alone. Learning and Motivation, 21(3), 237-265. doi:https://doi.org/10.1016/0023-9690(90)90008-C

Grant, A., Christie, M., & Ashcroft, S. (1980). Insulin release from human pancreatic islets in vitro. Diabetologia, 19(2), 114-117.

Graphpad. (2016). The ROUT method of identifying outliers. . Retrieved from https://www.graphpad.com/guides/prism/7/curve-fitting/reg_rout_method_outliers.htm?toc=0&printWindow

Gras, C., Herzog, E., Bellenchi, G. C., Bernard, V., Ravassard, P., Pohl, M., . . . El Mestikawy, S. (2002). A third vesicular glutamate transporter expressed by cholinergic and serotoninergic neurons. J Neurosci, 22(13), 5442-5451.

Gray, R. M., & Liotta, R. F. (2012). PTSD: Extinction, reconsolidation, and the visual-kinesthetic dissociation protocol. Traumatology, 18(2), 3-16.

Greter, M., & Merad, M. (2013). Regulation of microglia development and homeostasis. Glia, 61(1), 121-127. doi:10.1002/glia.22408

Grillon, C. (2002). Startle reactivity and anxiety disorders: aversive conditioning, context, and neurobiology. Biological psychiatry, 52(10), 958-975.

Groppe, D. M., Urbach, T. P., & Kutas, M. (2011). Mass univariate analysis of event‐related brain potentials/fields I: A critical tutorial review. Psychophysiology, 48(12), 1711-1725.

Gueye, A. B., Vendruscolo, L. F., Ávila, C., Moine, C., Darnaudéry, M., & Cador, M. (2018). Unlimited sucrose consumption during adolescence generates a depressive-like phenotype in adulthood. Neuropsychopharmacology, 1.

Guiard, B. P., & Di Giovanni, G. (2018). 5-HT2A Receptors in the Central Nervous System: Springer.

Guillin, O., Diaz, J., Carroll, P., Griffon, N., Schwartz, J.-C., & Sokoloff, P. (2001). BDNF controls dopamine D 3 receptor expression and triggers behavioural sensitization. Nature, 411(6833), 86.

391

Gulpinar, M., & Yegen, B. (2004). The physiology of learning and memory: role of peptides and stress. Current Protein and Peptide Science, 5(6), 457-473.

Gunduz-Cinar, O., MacPherson, K. P., Cinar, R., Gamble-George, J., Sugden, K., Williams, B., . . . Alapafuja, S. (2013). Convergent translational evidence of a role for anandamide in amygdala-mediated fear extinction, threat processing and stress-reactivity. Molecular psychiatry, 18(7), 813.

Gunstad, J., Paul, R. H., Spitznagel, M. B., Cohen, R. A., Williams, L. M., Kohn, M., & Gordon, E. (2006). Exposure to early life trauma is associated with adult obesity. Psychiatry research, 142(1), 31-37.

Guo, X., Park, Y., Freedman, N. D., Sinha, R., Hollenbeck, A. R., Blair, A., & Chen, H. (2014). Sweetened beverages, coffee, and tea and depression risk among older US adults. PloS one, 9(4), e94715.

Gurevich, E. V., Bordelon, Y., Shapiro, R. M., Arnold, S. E., Gur, R. E., & Joyce, J. N. (1997). Mesolimbic dopamine D3 receptors and use of antipsychotics in patients with schizophrenia: a postmortem study. Archives of general psychiatry, 54(3), 225-232.

Haby, M. M., Markwick, A., Peeters, A., Shaw, J., & Vos, T. (2012). Future predictions of body mass index and overweight prevalence in Australia, 2005–2025. Health Promotion International, 27(2), 250-260. doi:10.1093/heapro/dar036

Hagewoud, R., Bultsma, L. J., Barf, R. P., Koolhaas, J. M., & Meerlo, P. (2011). Sleep deprivation impairs contextual fear conditioning and attenuates subsequent behavioural, endocrine and neuronal responses. Journal of Sleep Research, 20(2), 259-266. doi:10.1111/j.1365-2869.2010.00895.x

Hajnal, A., Smith, G. P., & Norgren, R. (2004). Oral sucrose stimulation increases accumbens dopamine in the rat. American Journal of Physiology-Regulatory, Integrative and Comparative Physiology, 286(1), R31-R37.

Halford, J. C., Boyland, E. J., Blundell, J. E., Kirkham, T. C., & Harrold, J. A. (2010). Pharmacological management of appetite expression in obesity. Nature Reviews Endocrinology, 6(5), 255.

Hall, J., Thomas, K. L., & Everitt, B. J. (2000). Rapid and selective induction of BDNF expression in the hippocampus during contextual learning. Nature neuroscience, 3, 533. doi:10.1038/75698

Hall, J., Thomas, K. L., & Everitt, B. J. (2001a). Cellular Imaging of zif268 Expression in the Hippocampus and Amygdala during Contextual and Cued Fear Memory Retrieval: Selective Activation of Hippocampal CA1 Neurons during the Recall of Contextual Memories. The Journal of neuroscience, 21(6), 2186.

Hall, J., Thomas, K. L., & Everitt, B. J. (2001b). Fear memory retrieval induces CREB phosphorylation and Fos expression within the amygdala. European Journal of Neuroscience, 13(7), 1453-1458.

Han, F., Ding, J., & Shi, Y. (2014). Expression of amygdala mineralocorticoid receptor and glucocorticoid receptor in the single-prolonged stress rats. BMC Neurosci, 15, 77. doi:10.1186/1471-2202-15-77

Handel, M. J. (2011). mHealth (mobile health)—Using apps for health and wellness. EXPLORE: The Journal of Science and Healing, 7(4), 256-261.

Hanisch, U.-K., & Kettenmann, H. (2007). Microglia: active sensor and versatile effector cells in the normal and pathologic brain. Nature neuroscience, 10(11), 1387.

392

Haranhalli, N., Massie, L., Little, J., LeDoux, J., & Johnson, L. R. (2007). Relationship between memory 'strength' and the number of phosphorylated MAPK neurons in the lateral amygdala, Abstract number 307.9. Society for Neuroscience Abstracts.

Hariri, A. R., Drabant, E. M., Munoz, K. E., Kolachana, B. S., Mattay, V. S., Egan, M. F., & Weinberger, D. R. (2005). A susceptibility gene for affective disorders and the response of the human amygdala. Archives of general psychiatry, 62(2), 146-152.

Hassel, B., & Dingledine, R. (2012). Glutamate and glutamate receptors Basic Neurochemistry (pp. 342-366): Elsevier.

Haubrich, J., Machado, A., Boos, F. Z., Crestani, A. P., Sierra, R. O., de Oliveira Alvares, L., & Quillfeldt, J. A. (2017). Enhancement of extinction memory by pharmacological and behavioral interventions targeted to its reactivation. Scientific reports, 7(1), 10960.

Hayward, M. D., Schaich-Borg, A., Pintar, J. E., & Low, M. J. (2006). Differential involvement of endogenous opioids in sucrose consumption and food reinforcement. Pharmacology Biochemistry and Behavior, 85(3), 601-611.

He, B., Long, W., Li, X., Yang, W., Chen, Y., & Zhu, Y. (2018). Sugar-sweetened beverages consumption positively associated with the risks of obesity and hypertriglyceridemia among Children aged 7–18 years in South China. Journal of atherosclerosis and thrombosis, 25(1), 81-89.

Hebb, D. O. (1949). The organization of behavior: A neuropsychological theory: Psychology Press.

Hebebrand, J., Albayrak, Ö., Adan, R., Antel, J., Dieguez, C., de Jong, J., . . . Murphy, M. (2014). “Eating addiction”, rather than “food addiction”, better captures addictive-like eating behavior. Neuroscience & Biobehavioral Reviews, 47, 295-306.

Heim, C., & Nemeroff, C. B. (2001). The role of childhood trauma in the neurobiology of mood and anxiety disorders: preclinical and clinical studies. Biological psychiatry, 49(12), 1023-1039.

Henry, C., Kabbaj, M., Simon, H., Le Moal, M., & Maccari, S. (1994). Prenatal stress increases the hypothalamo‐pituitary‐adrenal axis response in young and adult rats. Journal of neuroendocrinology, 6(3), 341-345.

Herry, C., Ferraguti, F., Singewald, N., Letzkus, J. J., Ehrlich, I., & Lüthi, A. (2010). Neuronal circuits of fear extinction. European Journal of Neuroscience, 31(4), 599-612.

Herry, C., Trifilieff, P., Micheau, J., Lüthi, A., & Mons, N. (2006). Extinction of auditory fear conditioning requires MAPK/ERK activation in the basolateral amygdala. European Journal of Neuroscience, 24(1), 261-269.

Herz, A. (1997). Endogenous opioid systems and alcohol addiction. Psychopharmacology, 129(2), 99-111.

Herzog, E., Gilchrist, J., Gras, C., Muzerelle, A., Ravassard, P., Giros, B., . . . El Mestikawy, S. (2004). Localization of VGLUT3, the vesicular glutamate transporter type 3, in the rat brain. Neuroscience, 123(4), 983-1002.

Heyser, C. J., Schulteis, G., & Koob, G. F. (1997). Increased ethanol self‐administration after a period of imposed ethanol deprivation in rats trained in a limited access paradigm. Alcoholism: Clinical and Experimental Research, 21(5), 784-791.

Hinkle Jr, L. E. (1973). The concept of" stress" in the biological and social sciences. Science, Medicine and Man, 1(1), 31.

393

Hinwood, M., Tynan, R. J., Charnley, J. L., Beynon, S. B., Day, T. A., & Walker, F. R. (2012). Chronic stress induced remodeling of the prefrontal cortex: structural re-organization of microglia and the inhibitory effect of minocycline. Cerebral cortex, 23(8), 1784-1797.

Hirth, J. M., Rahman, M., & Berenson, A. B. (2011). The association of posttraumatic stress disorder with fast food and soda consumption and unhealthy weight loss behaviors among young women. Journal of Women's Health, 20(8), 1141-1149.

Hodges, M. R., Cummings, K. J., & Pilowsky, P. M. (2018). Serotonin: Physiological Homeostasis from Birth to Adulthood: Elsevier Science & Technology.

Hoebel, B., Avena, N., Bocarsly, M., & Rada, P. Natural addiction: a behavioral and circuit model based on sugar addiction in rats. J Addict Med 2009; 3: 33–41.

Hoebel, B. G., Avena, N. M., & Rada, P. (2007). Accumbens dopamine-acetylcholine balance in approach and avoidance. Current opinion in pharmacology, 7(6), 617-627.

Holgate, J. Y., Shariff, M., Mu, E. W., & Bartlett, S. (2017). A rat drinking in the dark model for studying ethanol and sucrose consumption. Frontiers in Behavioral Neuroscience, 11, 29.

Holmes, A., & Singewald, N. (2013). Individual differences in recovery from traumatic fear. Trends Neurosci, 36(1), 23-31. doi:10.1016/j.tins.2012.11.003

Holstege, G., Georgiadis, J. R., Paans, A. M., Meiners, L. C., van der Graaf, F. H., & Reinders, A. S. (2003). Brain activation during human male ejaculation. Journal of Neuroscience, 23(27), 9185-9193.

Homberg, J. R. (2012). Serotonergic modulation of conditioned fear. Scientifica, 2012.

Hong, S., Dissing-Olesen, L., & Stevens, B. (2016). New insights on the role of microglia in synaptic pruning in health and disease. Current opinion in neurobiology, 36, 128-134.

Horn, C. C., & Friedman, M. I. (2003). Detection of single unit activity from the rat vagus using cluster analysis of principal components. Journal of neuroscience methods, 122(2), 141-147.

Hsu, T. M., Konanur, V. R., Taing, L., Usui, R., Kayser, B. D., Goran, M. I., & Kanoski, S. E. (2015). Effects of sucrose and high fructose corn syrup consumption on spatial memory function and hippocampal neuroinflammation in adolescent rats. Hippocampus, 25(2), 227-239.

Hu, W., Zhang, Y., Wu, W., Yin, Y., Huang, D., Wang, Y., . . . Li, W. (2016). Chronic glucocorticoids exposure enhances neurodegeneration in the frontal cortex and hippocampus via NLRP-1 inflammasome activation in male mice. Brain, behavior, and immunity, 52, 58-70.

Huang, S.-H., Wu, W.-R., Lee, L.-M., Huang, P.-R., & Chen, J.-C. (2018). mTOR signaling in the nucleus accumbens mediates behavioral sensitization to methamphetamine. Progress in Neuro-Psychopharmacology and Biological Psychiatry.

Hubbard, D. T., Nakashima, B. R., Lee, I., & Takahashi, L. K. (2007). ACTIVATION OF BASOLATERAL AMYGDALA CRF(1) RECEPTORS MODULATES THE CONSOLIDATION OF CONTEXTUAL FEAR. Neuroscience, 150(4), 818-828. doi:10.1016/j.neuroscience.2007.10.001

Huber, R., Imeh-Nathaniel, A., Nathaniel, T. I., Gore, S., Datta, U., Bhimani, R., . . . van Staaden, M. J. (2018). Drug-sensitive Reward in Crayfish: Exploring the

394

Neural Basis of Addiction with Automated Learning Paradigms. Behavioural processes.

Huff, M. L., Emmons, E. B., Narayanan, N. S., & LaLumiere, R. T. (2016). Basolateral amygdala projections to ventral hippocampus modulate the consolidation of footshock, but not contextual, learning in rats. Learning & Memory, 23(2), 51-60. doi:10.1101/lm.039909.115

Hwang, J. J., Jiang, L., Hamza, M., Dai, F., Belfort-DeAguiar, R., Cline, G., . . . Sherwin, R. S. (2017). The human brain produces fructose from glucose. JCI insight, 2(4).

Hwang, K. O., Ottenbacher, A. J., Green, A. P., Cannon-Diehl, M. R., Richardson, O., Bernstam, E. V., & Thomas, E. J. (2010). Social support in an Internet weight loss community. International journal of medical informatics, 79(1), 5-13.

Ibi, D., Suzuki, F., & Hiramatsu, M. (2018). Effect of AceK (acesulfame potassium) on brain function under dietary restriction in mice. Physiology & behavior, 188, 291-297.

Iemolo, A., Valenza, M., Tozier, L., Knapp, C. M., Kornetsky, C., Steardo, L., . . . Cottone, P. (2012). Withdrawal from chronic, intermittent access to a highly palatable food induces depressive-like behavior in compulsive eating rats. Behavioural pharmacology, 23, 593.

Impey, S., Smith, D. M., Obrietan, K., Donahue, R., Wade, C., & Storm, D. R. (1998). Stimulation of cAMP response element (CRE)-mediated transcription during contextual learning. Nature Neuroscience, 1(7), 595-601.

Insel, T., Voon, V., Nye, J., Brown, V. J., Altevogt, B., Bullmore, E., . . . Malloch, G. (2013). Innovative solutions to novel drug development in mental health. Neuroscience & Biobehavioral Reviews, 37(10), 2438-2444.

Itzhak, Y., & Martin, J. L. (1999). Effects of cocaine, nicotine, dizocipline and alcohol on mice locomotor activity: cocaine–alcohol cross-sensitization involves upregulation of striatal dopamine transporter binding sites. Brain research, 818(2), 204-211.

Ivashkina, O. I., Toropova, K. A., Ivanov, A. A., Chekhov, S. A., & Anokhin, K. V. (2016). Waves of c-Fos and Arc Proteins Expression in Neuronal Populations of the Hippocampus in Response to a Single Episode of New Experience. Bulletin of Experimental Biology and Medicine, 160(6), 729-732. doi:10.1007/s10517-016-3296-3

Izquierdo, I., Furini, C. R. G., & Myskiw, J. C. (2016). Fear Memory. Physiological reviews, 96(2), 695-750.

Izquierdo, I., & Medina, J. H. (1997). Memory formation: the sequence of biochemical events in the hippocampus and its connection to activity in other brain structures. Neurobiology of learning and memory, 68(3), 285-316.

Jacka, F. N., Mykletun, A., Berk, M., Bjelland, I., & Tell, G. S. (2011). The association between habitual diet quality and the common mental disorders in community-dwelling adults: the Hordaland Health study. Psychosomatic medicine, 73(6), 483-490.

Jacobs, B. L., & Azmitia, E. C. (1992). Structure and function of the brain serotonin system. Physiological reviews, 72(1), 165-229.

Jacobson, L., & Sapolsky, R. (1991). The role of the hippocampus in feedback regulation of the hypothalamic-pituitary-adrenocortical axis. Endocrine reviews, 12(2), 118-134.

395

Jacques, A., Battle, A. R., & Johnson, L. R. (2016). Glucocorticoid Receptor (GR). In S. Choi (Ed.), Encyclopedia of Signaling Molecules (pp. 1-5). New York, NY: Springer New York.

Jacques, A., Chaaya, N., Battle, A. R., & Johnson, L. R. (2017). Microanatomy of fear memory consolidation and extinction within sub-regions of the prefrontal cortex and amygdala revealed by Arc and pERK/MAPK activity, Abstract 424.10. Society for Neuroscience Abstracts.

Jacques, A., Chaaya, N., Hettiarachchi, C., Carmody, M.-L., Beecher, K., Belmer, A., . . . Johnson, L. R. (2019). Microtopography of fear memory consolidation and extinction retrieval within prefrontal cortex and amygdala. Psychopharmacology, 1-15.

Jacques, A., Johnson, L. R., & Battle, A. R. (2016). Mineralocorticoid Receptor. In S. Choi (Ed.), Encyclopedia of Signaling Molecules (pp. 1-5). New York, NY: Springer New York.

Jacques, A., Wright, A., Chaaya, N., Overall, A., Bergstrom, H. C., McDonald, C., . . . Johnson, L. R. (2018). Techniques for Mapping and Measuring Functional Neuronal Topography for the Decoding and Prediction of Behavior. Frontiers in Neural Circuits, 12, 84.

Jastreboff, A. M., Sinha, R., Arora, J., Giannini, C., Kubat, J., Malik, S., . . . Pierpont, B. (2016). Altered brain response to drinking glucose and fructose in obese adolescents. Diabetes, db151216.

Jeffery, K. J., Gilbert, A., Burton, S., & Strudwick, A. (2003). Preserved performance in a hippocampal‐dependent spatial task despite complete place cell remapping. Hippocampus, 13(2), 175-189.

Jellinek, E. M. (1952). Phases of alcohol addiction. Quarterly journal of studies on alcohol, 13(4), 673-684.

Jennings, J. H., Rizzi, G., Stamatakis, A. M., Ung, R. L., & Stuber, G. D. (2013). The inhibitory circuit architecture of the lateral hypothalamus orchestrates feeding. science, 341(6153), 1517-1521. doi:10.1126/science.1241812

Jennings, J. H., Ung, R. L., Resendez, S. L., Stamatakis, A. M., Taylor, J. G., Huang, J., . . . Stuber, G. D. (2015). Visualizing hypothalamic network dynamics for appetitive and consummatory behaviors. Cell, 160(3), 516-527. doi:10.1016/j.cell.2014.12.026

Jing, L., Liu, B., Zhang, M., & Liang, J.-H. (2018). Involvement of dopamine D2 receptor in a single methamphetamine-induced behavioral sensitization in C57BL/6J mice. Neuroscience letters.

Johansen, J. P., Cain, C. K., Ostroff, L. E., & LeDoux, J. E. (2011). Molecular mechanisms of fear learning and memory. Cell, 147(3), 509-524.

Johnson, D. M., Baker, J. D., & Azorlosa, J. L. (2000). Acquisition, extinction, and reinstatement of Pavlovian fear conditioning: the roles of the NMDA receptor and nitric oxide. Brain Res, 857(1-2), 66-70.

Johnson, G. L., & Lapadat, R. (2002). Mitogen-activated protein kinase pathways mediated by ERK, JNK, and p38 protein kinases. science, 298(5600), 1911-1912.

Johnson, L. R., Hou, M., Ponce-Alvarez, A., Gribelyuk, L. M., Alphs, H. H., Albert, L., . . . Doyere, V. (2008). A recurrent network in the lateral amygdala: a mechanism for coincidence detection. Front Neural Circuits, 2, 3. doi:10.3389/neuro.04.003.2008

396

Johnson, L. R., Ledoux, J. E., & Doyere, V. (2009). Hebbian reverberations in emotional memory micro circuits. Front Neurosci, 3(2), 198-205. doi:10.3389/neuro.01.027.2009

Johnson, L. R., McGuire, J., Lazarus, R., & Palmer, A. A. (2012). Pavlovian fear memory circuits and phenotype models of PTSD. Neuropharmacology, 62(2), 638-646. doi:10.1016/j.neuropharm.2011.07.004

Johnson, L. R., McGuire, J., Lazarus, R., & Palmer, A. A. (2012). Pavlovian fear memory circuits and phenotype models of PTSD. Neuropharmacology, 62(2), 638-646.

Johnson, P. M., & Kenny, P. J. (2010). Dopamine D2 receptors in addiction-like reward dysfunction and compulsive eating in obese rats. Nature neuroscience, 13(5), 635.

Jolliffe, I. (2002). Principal component analysis: Wiley Online Library. Joly-Amado, A., Cansell, C., Denis, R. G., Delbes, A.-S., Castel, J., Martinez, S., &

Luquet, S. (2014). The hypothalamic arcuate nucleus and the control of peripheral substrates. Best practice & research Clinical endocrinology & metabolism, 28(5), 725-737.

Josselyn, S. A., & Frankland, P. W. (2018). Memory Allocation: Mechanisms and Function. Annu Rev Neurosci. doi:10.1146/annurev-neuro-080317-061956

Josselyn, S. A., Kohler, S., & Frankland, P. W. (2017). Heroes of the Engram. J Neurosci, 37(18), 4647-4657. doi:10.1523/JNEUROSCI.0056-17.2017

Jovanovic, T., Norrholm, S. D., Blanding, N. Q., Phifer, J. E., Weiss, T., Davis, M., . . . Ressler, K. (2010). Fear potentiation is associated with hypothalamic–pituitary–adrenal axis function in PTSD. Psychoneuroendocrinology, 35(6), 846-857.

Jurdak, N., & Kanarek, R. B. (2009). Sucrose-induced obesity impairs novel object recognition learning in young rats. Physiology & behavior, 96(1), 1-5.

Kanarek, R. B., White, E. S., Biegen, M. T., & Marks-Kaufman, R. (1991). Dietary influences on morphine-induced analgesia in rats. Pharmacology Biochemistry and Behavior, 38(3), 681-684.

Kandel, E. R. (2001). The molecular biology of memory storage: a dialogue between genes and synapses. science, 294(5544), 1030-1038.

Kandel, E. R. (2004). The molecular biology of memory storage: a dialog between genes and synapses. Bioscience reports, 24(4-5), 475-522.

Kanoski, S. E., & Davidson, T. L. (2011). Western diet consumption and cognitive impairment: links to hippocampal dysfunction and obesity. Physiology & behavior, 103(1), 59-68.

Karagiannides, I., Torres, D., Tseng, Y. H., Bowe, C., Carvalho, E., Espinoza, D., . . . Kokkotou, E. (2008). Substance P as a novel anti-obesity target. Gastroenterology, 134(3), 747-755. e741.

Karperien, A., Ahammer, H., & Jelinek, H. (2013). Quantitating the subtleties of microglial morphology with fractal analysis. Frontiers in cellular neuroscience, 7, 3.

Karpova, N. N., Pickenhagen, A., Lindholm, J., Tiraboschi, E., Kulesskaya, N., Ágústsdóttir, A., . . . Sullivan, R. (2011). Fear erasure in mice requires synergy between antidepressant drugs and extinction training. science, 334(6063), 1731-1734.

Kass, A. E., Kolko, R. P., & Wilfley, D. E. (2013). Psychological treatments for eating disorders. Current opinion in psychiatry, 26(6), 549.

Kastin, A. (2013). Handbook of biologically active peptides: Academic press.

397

Kavaliers, M., & Hirst, M. (1987). Slugs and snails and opiate tales: opioids and feeding behavior in invertebrates. Paper presented at the Federation proceedings.

Kawahara, H. (1995). Serotonin release in the rat amygdala during exposure to emotional stress. Serotonin in the central nervous system and periphery, 89-96.

Kays, J. L., Hurley, R. A., & Taber, K. H. (2012). The dynamic brain: neuroplasticity and mental health. The Journal of neuropsychiatry and clinical neurosciences, 24(2), 118-124.

Keesman, M., Aarts, H., Häfner, M., & Papies, E. K. (2017). Mindfulness Reduces Reactivity to Food Cues: Underlying Mechanisms and Applications in Daily Life. Current addiction reports, 4(2), 151-157.

Keller, J., Gomez, R., Williams, G., Lembke, A., Lazzeroni, L., Murphy Jr, G. M., & Schatzberg, A. F. (2017). HPA axis in major depression: cortisol, clinical symptomatology and genetic variation predict cognition. Molecular psychiatry, 22(4), 527.

Kelley, A. E., Baldo, B. A., & Pratt, W. E. (2005). A proposed hypothalamic–thalamic–striatal axis for the integration of energy balance, arousal, and food reward. Journal of Comparative Neurology, 493(1), 72-85.

Kendig, M. D. (2014). Cognitive and behavioural effects of sugar consumption in rodents. A review. Appetite, 80, 41-54.

Kenny, P. J. (2011). Reward mechanisms in obesity: new insights and future directions. Neuron, 69(4), 664-679.

Kerber, R. (1992, 1992). Chimerge: Discretization of numeric attributes. Paper presented at the Proceedings of the tenth national conference on Artificial intelligence.

Kerr, A. L., Cheng, S.-Y., & Jones, T. A. (2011). Experience-dependent neural plasticity in the adult damaged brain. Journal of communication disorders, 44(5), 538-548.

Kettenmann, H., Hanisch, U.-K., Noda, M., & Verkhratsky, A. (2011). Physiology of microglia. Physiological reviews, 91(2), 461-553.

Kikkawa, R., Umemura, K., Haneda, M., Arimura, T., Ebata, K., & Shigeta, Y. (1987). Evidence for existence of polyol pathway in cultured rat mesangial cells. Diabetes, 36(2), 240-243.

Kilpatrick, D. G., Resnick, H. S., Milanak, M. E., Miller, M. W., Keyes, K. M., & Friedman, M. J. (2013). National estimates of exposure to traumatic events and PTSD prevalence using DSM‐IV and DSM‐5 criteria. Journal of traumatic stress, 26(5), 537-547.

Kim, J. J., & Jung, M. W. (2006). Neural circuits and mechanisms involved in Pavlovian fear conditioning: A critical review. Neuroscience & Biobehavioral Reviews, 30(2), 188-202. doi:http://dx.doi.org/10.1016/j.neubiorev.2005.06.005

Kim, M., Campeau, S., Falls, W. A., & Davis, M. (1993). Infusion of the non-NMDA receptor antagonist CNQX into the amygdala blocks the expression of fear-potentiated startle. Behav Neural Biol, 59(1), 5-8.

Kim, S., Shou, J., Abera, S., & Ziff, E. B. (2018). Sucrose withdrawal induces depression and anxiety-like behavior by Kir2. 1 upregulation in the nucleus accumbens. Neuropharmacology, 130, 10-17.

398

Kim, Y., Venkataraju, K. U., Pradhan, K., Mende, C., Taranda, J., Turaga, S. C., . . . Rockland, K. S. (2015). Mapping social behavior-induced brain activation at cellular resolution in the mouse. Cell reports, 10(2), 292-305.

Kim, Y., Yang, G. R., Pradhan, K., Venkataraju, K. U., Bota, M., Del Molino, L. C. G., . . . Levine, J. M. (2017). Brain-wide maps reveal stereotyped cell-type-based cortical architecture and subcortical sexual dimorphism. Cell, 171(2), 456-469. e422.

Kinzig, K. P., Hargrave, S. L., & Honors, M. A. (2008). Binge-type eating attenuates corticosterone and hypophagic responses to restraint stress. Physiology & behavior, 95(1-2), 108-113.

Kirshenbaum, A. P., Brown, S. J., Hughes, D. M., & Doughty, A. H. (2008). Differential-reinforcement-of-low-rate-schedule performance and nicotine administration: a systematic investigation of dose, dose-regimen, and schedule requirement. Behavioural pharmacology, 19(7), 683.

Klenowski, P. M., Shariff, M. R., Belmer, A., Fogarty, M. J., Mu, E. W., Bellingham, M. C., & Bartlett, S. E. (2016). Prolonged consumption of sucrose in a binge-like manner, alters the morphology of medium spiny neurons in the nucleus accumbens shell. Frontiers in Behavioral Neuroscience, 10, 54.

Knolle, F., Ermakova, A. O., Justicia, A., Fletcher, P. C., Bunzeck, N., Duzel, E., & Murray, G. (2018). Altered cingulate and substantia nigra/ventral tegmental activation to novelty and emotional salience in antipsychotic naive first episode psychosis patients. bioRxiv, 263020.

Knowlton, B. J., & Fanselow, M. S. (1998). The hippocampus, consolidation and on-line memory.

Knüppel, A., Shipley, M. J., Llewellyn, C. H., & Brunner, E. J. (2017). Sugar intake from sweet food and beverages, common mental disorder and depression: prospective findings from the Whitehall II study. Scientific reports, 7(1), 6287.

Kongsui, R., Beynon, S. B., Johnson, S. J., & Walker, F. R. (2014). Quantitative assessment of microglial morphology and density reveals remarkable consistency in the distribution and morphology of cells within the healthy prefrontal cortex of the rat. Journal of neuroinflammation, 11(1), 182.

Konorski, J. (1967). Integrative activity of the brain; an interdisciplinary approach. Koob, G. (2018). 181. Drug Addiction: The Gain in the Brain is in the Pain.

Biological psychiatry, 83(9), S72-S73. Koob, G. F. (2009a). Brain stress systems in the amygdala and addiction. Brain

research, 1293, 61-75. Koob, G. F. (2009b). Neurobiological substrates for the dark side of compulsivity in

addiction. Neuropharmacology, 56, 18-31. Koob, G. F., & Le Moal, M. (2001). Drug addiction, dysregulation of reward, and

allostasis. Neuropsychopharmacology, 24(2), 97. Koolschijn, P. C. M., van Haren, N. E., Lensvelt‐Mulders, G. J., Hulshoff Pol, H. E.,

& Kahn, R. S. (2009). Brain volume abnormalities in major depressive disorder: A meta‐analysis of magnetic resonance imaging studies. Human brain mapping, 30(11), 3719-3735.

Korb, E., & Finkbeiner, S. (2011). Arc in synaptic plasticity: from gene to behavior. Trends in neurosciences, 34(11), 591-598.

Krystal, J. H., Tolin, D. F., Sanacora, G., Castner, S. A., Williams, G. V., Aikins, D. E., . . . D'Souza, D. C. (2009). Neuroplasticity as a target for the

399

pharmacotherapy of anxiety disorders, mood disorders, and schizophrenia. Drug discovery today, 14(13-14), 690-697.

Kudo, K., Qiao, C.-X., Kanba, S., & Arita, J. (2004). A selective increase in phosphorylation of cyclic AMP response element-binding protein in hippocampal CA1 region of male, but not female, rats following contextual fear and passive avoidance conditioning. Brain research, 1024(1), 233-243. doi:https://doi.org/10.1016/j.brainres.2004.08.007

Kurhe, Y., Radhakrishnan, M., & Gupta, D. (2014). Ondansetron attenuates depression co-morbid with obesity in obese mice subjected to chronic unpredictable mild stress; an approach using behavioral battery tests. Metabolic brain disease, 29(3), 701-710.

Kyriazis, G. A., Soundarapandian, M. M., & Tyrberg, B. (2012). Sweet taste receptor signaling in beta cells mediates fructose-induced potentiation of glucose-stimulated insulin secretion. Proceedings of the National Academy of Sciences, 109(8), E524-E532.

LaBar, K. S., Gatenby, J. C., Gore, J. C., LeDoux, J. E., & Phelps, E. A. (1998). Human amygdala activation during conditioned fear acquisition and extinction: a mixed-trial fMRI study. Neuron, 20(5), 937-945.

Lai, C.-F., Chaudhary, L., Fausto, A., Halstead, L. R., Ory, D. S., Avioli, L. V., & Cheng, S.-L. (2001). Erk is essential for growth, differentiation, integrin expression, and cell function in human osteoblastic cells. Journal of Biological Chemistry, 276(17), 14443-14450.

Lam, K. C., & Jadavji, N. M. (2018). Seeking Happiness: Understanding the Mechanisms of Mixing Music and Drugs.

Langseth, H. (2008). Bayesian networks in reliability: The good, the bad, and the ugly. Advances in Mathematical Modeling for Reliability, 1.

Lathe, R. (2001). Hormones and the hippocampus. Journal of Endocrinology, 169(2), 205-231.

Lau, J., & Herzog, H. (2014). CART in the regulation of appetite and energy homeostasis. Frontiers in neuroscience, 8, 313.

Laughlin, M. R. (2014). Normal roles for dietary fructose in carbohydrate metabolism. Nutrients, 6(8), 3117-3129.

Laurent, V., & Westbrook, R. F. (2009). Inactivation of the infralimbic but not the prelimbic cortex impairs consolidation and retrieval of fear extinction. Learning & Memory, 16(9), 520-529.

Leão, R. M., Cruz, F. C., Marin, M. T., & da Silva Planeta, C. (2012). Stress induces behavioral sensitization, increases nicotine-seeking behavior and leads to a decrease of CREB in the nucleus accumbens. Pharmacology Biochemistry and Behavior, 101(3), 434-442.

LeDoux, J. (2000). Emotion circuits in the brain. Annual Review of Neuroscience, 23, 155-184.

LeDoux, J. (2003). The emotional brain, fear, and the amygdala. Cellular and molecular neurobiology, 23(4-5), 727-738.

LeDoux, J. (2012). Rethinking the emotional brain. Neuron, 73(4), 653-676. doi:10.1016/j.neuron.2012.02.004

LeDoux, J., & Daw, N. D. (2018). Surviving threats: neural circuit and computational implications of a new taxonomy of defensive behaviour. Nat Rev Neurosci. doi:10.1038/nrn.2018.22

LeDoux, J. E. (2000). Emotion circuits in the brain. Annual review of neuroscience, 23(1), 155-184.

400

LeDoux, J. E. (2014). Coming to terms with fear. Proceedings of the National Academy of Sciences, 111(8), 2871.

LeDoux, J. E., Cicchetti, P., Xagoraris, A., & Romanski, L. M. (1990). The lateral amygdaloid nucleus: sensory interface of the amygdala in fear conditioning. Journal of Neuroscience, 10(4), 1062-1069.

LeDoux, J. E., Haranhalli, N., Massie, L. W., Little, J. P., & Johnson, L. R. (2006). Systematic constellations of neurons are activated in lateral amygdala following Pavlovian fear conditioning, Abstract number 370.8. Society for Neuroscience Abstracts.

LeDoux, J. E., Iwata, J., Cicchetti, P., & Reis, D. (1988). Different projections of the central amygdaloid nucleus mediate autonomic and behavioral correlates of conditioned fear. The Journal of neuroscience, 8(7), 2517-2529.

LeDoux, J. E., Iwata, J., Cicchetti, P., & Reis, D. J. (1988). Different projections of the central amygdaloid nucleus mediate autonomic and behavioral correlates of conditioned fear. Journal of Neuroscience, 8(7), 2517-2529.

Lee, A. M., & Messing, R. O. (2008). Protein kinases and addiction. Annals of the New York Academy of Sciences, 1141(1), 22-57.

Lee, H. J., Haberman, R. P., Roquet, R., & Monfils, M. H. (2015). Extinction and retrieval+ extinction of conditioned fear differentially activate medial prefrontal cortex and amygdala in rats. Frontiers in Behavioral Neuroscience, 9, 369.

Lee, H. J., Haberman, R. P., Roquet, R., & Monfils, M. H. (2016). Extinction and retrieval+extinction of conditioned fear differentially activate medial prefrontal cortex and amygdala in rats. Frontiers in Behavioral Neuroscience, 9. doi:10.3389/fnbeh.2015.00369

Lee, H. J., Haberman, R. P., Roquet, R. F., & Monfils, M.-H. (2016). Extinction and retrieval+ extinction of conditioned fear differentially activate medial prefrontal cortex and amygdala in rats. Frontiers in Behavioral Neuroscience, 9, 369.

Lee, J. L. C., Everitt, B. J., & Thomas, K. L. (2004). Independent Cellular Processes for Hippocampal Memory Consolidation and Reconsolidation. science, 304(5672), 839. doi:10.1126/science.1095760

Legastelois, R., Botia, B., Coune, F., Jeanblanc, J., & Naassila, M. (2014). Deciphering the relationship between vulnerability to ethanol‐induced behavioral sensitization and ethanol consumption in outbred mice. Addiction biology, 19(2), 210-224.

Lehmann, H., Lacanilao, S., & Sutherland, R. J. (2007). Complete or partial hippocampal damage produces equivalent retrograde amnesia for remote contextual fear memories. European Journal of Neuroscience, 25(5), 1278-1286. doi:10.1111/j.1460-9568.2007.05374.x

Leigh, S.-J., & Morris, M. J. (2016). The role of reward circuitry and food addiction in the obesity epidemic: An update. Biological psychology.

Lemos, C., Rial, D., Goncalves, F., Pires, J., Silva, H., Matheus, F., . . . Jarak, I. (2016). High sucrose consumption induces memory impairment in rats associated with electrophysiological modifications but not with metabolic changes in the hippocampus. Neuroscience, 315, 196-205.

Lenoir, M., Serre, F., Cantin, L., & Ahmed, S. H. (2007). Intense sweetness surpasses cocaine reward. PloS one, 2(8), e698.

401

Lesch, K.-P., & Waider, J. (2012). Serotonin in the modulation of neural plasticity and networks: implications for neurodevelopmental disorders. Neuron, 76(1), 175-191.

Leuner, B., & Gould, E. (2010). Structural plasticity and hippocampal function. Annual review of psychology, 61, 111-140.

Levin, A. (2008). Prolonged Exposure Therapy Helps Ward off PTSD. Psychiatric News, 43(13), 13.

Levine, A. S., & Billington, C. J. (2004). Opioids as agents of reward-related feeding: a consideration of the evidence. Physiology & behavior, 82(1), 57-61.

Levoy, E., Lazaridou, A., Brewer, J., & Fulwiler, C. (2017). An exploratory study of Mindfulness Based Stress Reduction for emotional eating. Appetite, 109, 124-130.

Li, S., Kim, J. H., & Richardson, R. (2012). Differential involvement of the medial prefrontal cortex in the expression of learned fear across development. Behavioral neuroscience, 126(2), 217.

Li, X., Zhao, Q., Wei, W., Lin, Q., Magnan, C., Emami, M., . . . Madugalle, S. U. (2018). The accumulation of N6-methyl-2-deoxyadenosine in DNA drives activity-induced gene expression and is required for the extinction of conditioned fear. bioRxiv, 059972.

Li, Y., Acerbo, M. J., & Robinson, T. E. (2004). The induction of behavioural sensitization is associated with cocaine‐induced structural plasticity in the core (but not shell) of the nucleus accumbens. European Journal of Neuroscience, 20(6), 1647-1654.

Lien, L., Lien, N., Heyerdahl, S., Thoresen, M., & Bjertness, E. (2006). Consumption of soft drinks and hyperactivity, mental distress, and conduct problems among adolescents in Oslo, Norway. American journal of public health, 96(10), 1815-1820.

Lin, F., Yu, C., Jiang, T., Li, K., Li, X., Qin, W., . . . Chan, P. (2006). Quantitative analysis along the pyramidal tract by length-normalized parameterization based on diffusion tensor tractography: application to patients with relapsing neuromyelitis optica. Neuroimage, 33(1), 154-160.

Lin, H. C., Mao, S. C., & Gean, P. W. (2009). Block of gamma-aminobutyric acid-A receptor insertion in the amygdala impairs extinction of conditioned fear. Biol Psychiatry, 66(7), 665-673. doi:10.1016/j.biopsych.2009.04.003

Lin, L., Osan, R., & Tsien, J. Z. (2006). Organizing principles of real-time memory encoding: neural clique assemblies and universal neural codes. Trends in neurosciences, 29(1), 48-57.

Lin, Z., Shi, L., Lu, J., Li, J., Hu, H., Zuo, C., . . . Xu, L. (2013). Effects of curcumin on glucose metabolism in the brains of rats subjected to chronic unpredictable stress: a 18 F-FDG micro-PET study. BMC complementary and alternative medicine, 13(1), 202.

Lindgren, E., Gray, K., Miller, G., Tyler, R., Wiers, C., Volkow, N., & Wang, G. (2018). Food addiction: A common neurobiological mechanism with drug abuse. Frontiers in bioscience (Landmark edition), 23, 811-836.

Little, J. P., & Carter, A. G. (2012). Subcellular synaptic connectivity of layer 2 pyramidal neurons in the medial prefrontal cortex. Journal of Neuroscience, 32(37), 12808-12819.

402

Liu, C., Lee, S., & Elmquist, J. K. (2014). Circuits controlling energy balance and mood: inherently intertwined or just complicated intersections? Cell metabolism, 19(6), 902-909.

Liu, I. Y. C., Lyons, W. E., Mamounas, L. A., & Thompson, R. F. (2004). Brain-Derived Neurotrophic Factor Plays a Critical Role in Contextual Fear Conditioning. The Journal of neuroscience, 24(36), 7958. doi:10.1523/JNEUROSCI.1948-04.2004

Liu, S. (2018). What Drives Us? Chasing Reward in a Dopaminergic Society. Columbia Undergraduate Science Journal, 11(2017).

Liu, X., Ramirez, S., Pang, P. T., Puryear, C. B., Govindarajan, A., Deisseroth, K., & Tonegawa, S. (2012). Optogenetic stimulation of a hippocampal engram activates fear memory recall. Nature, 484(7394), 381.

Löf, E., Olausson, P., Stomberg, R., Taylor, J. R., & Söderpalm, B. (2010). Nicotinic acetylcholine receptors are required for the conditioned reinforcing properties of sucrose-associated cues. Psychopharmacology, 212(3), 321-328.

Lonergan, M. E., Gafford, G. M., Jarome, T. J., & Helmstetter, F. J. (2010). Time-dependent expression of Arc and zif268 after acquisition of fear conditioning. Neural plasticity, 2010.

López-Muñoz, F., Boya, J., & Alamo, C. (2006). Neuron theory, the cornerstone of neuroscience, on the centenary of the Nobel Prize award to Santiago Ramón y Cajal. Brain research bulletin, 70(4), 391-405.

Lopresti, A. L., & Drummond, P. D. (2013). Obesity and psychiatric disorders: commonalities in dysregulated biological pathways and their implications for treatment. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 45, 92-99.

Lowe, M. R., & Butryn, M. L. (2007). Hedonic hunger: a new dimension of appetite? Physiology & behavior, 91(4), 432-439.

Lu, L., Grimm, J. W., Hope, B. T., & Shaham, Y. (2004). Incubation of cocaine craving after withdrawal: a review of preclinical data. Neuropharmacology, 47, 214-226.

Lubin, F. D., Roth, T. L., & Sweatt, J. D. (2008). Epigenetic Regulation of <em>bdnf</em> Gene Transcription in the Consolidation of Fear Memory. The Journal of neuroscience, 28(42), 10576. doi:10.1523/JNEUROSCI.1786-08.2008

Lubow, R., Rifkin, B., & Alek, M. (1976). The context effect: The relationship between stimulus preexposure and environmental preexposure determines subsequent learning. Journal of Experimental Psychology: Animal Behavior Processes, 2(1), 38.

Lubow, R. E. (1965). Latent inhibition: effects of frequency of nonreinforced preexposure of the CS. Journal of comparative and physiological psychology, 60(3), 454.

Lubow, R. E., & Moore, A. U. (1959). Latent inhibition: The effect of nonreinforced pre-exposure to the conditional stimulus. Journal of Comparative and Physiological Psychology, 52(4), 415-419. doi:10.1037/h0046700

Lumeng, J. C., Wendorf, K., Pesch, M. H., Appugliese, D. P., Kaciroti, N., Corwyn, R. F., & Bradley, R. H. (2013). Overweight adolescents and life events in childhood. Pediatrics, 132(6), e1506-e1512.

Luppino, F. S., de Wit, L. M., Bouvy, P. F., Stijnen, T., Cuijpers, P., Penninx, B. W., & Zitman, F. G. (2010). Overweight, obesity, and depression: a systematic

403

review and meta-analysis of longitudinal studies. Archives of general psychiatry, 67(3), 220-229.

Lustig, R. H., Schmidt, L. A., & Brindis, C. D. (2012). Public health: The toxic truth about sugar. Nature, 482(7383), 27.

Lyford, G. L., Yamagata, K., Kaufmann, W. E., Barnes, C. A., Sanders, L. K., Copeland, N. G., . . . Worley, P. F. (1995). Arc, a growth factor and activity-regulated gene, encodes a novel cytoskeleton-associated protein that is enriched in neuronal dendrites. Neuron, 14(2), 433-445. doi:https://doi.org/10.1016/0896-6273(95)90299-6

MacCallum, R. C., Zhang, S., Preacher, K. J., & Rucker, D. D. (2002). On the practice of dichotomization of quantitative variables. Psychological methods, 7(1), 19.

Majak, K., & Pitkänen, A. (2003). Activation of the amygdalo‐entorhinal pathway in fear‐conditioning in rat. European Journal of Neuroscience, 18(6), 1652-1659.

Majer-Lobodzinska, A., & Adamiec-Mroczek, J. (2017). Glucocorticoid receptor polymorphism in obesity and glucose homeostasis. Adv Clin Exp Med, 26(1), 143-148. doi:10.17219/acem/41231

Makarem, N., Bandera, E. V., Nicholson, J. M., & Parekh, N. (2018). Consumption of Sugars, Sugary Foods, and Sugary Beverages in Relation to Cancer Risk: A Systematic Review of Longitudinal Studies. Annual review of nutrition(0).

Maldonado, R., Valverde, O., & Berrendero, F. (2006). Involvement of the endocannabinoid system in drug addiction. Trends in neurosciences, 29(4), 225-232.

Malik, V. S., Pan, A., Willett, W. C., & Hu, F. B. (2013). Sugar-sweetened beverages and weight gain in children and adults: a systematic review and meta-analysis–. The American journal of clinical nutrition, 98(4), 1084-1102.

Malkani, S., & Rosen, J. B. (2000). Specific induction of early growth response gene 1 in the lateral nucleus of the amygdala following contextual fear conditioning in rats. Neuroscience, 97(4), 693-702. doi:https://doi.org/10.1016/S0306-4522(00)00058-0

Mamiya, N., Fukushima, H., Suzuki, A., Matsuyama, Z., Homma, S., Frankland, P. W., & Kida, S. (2009a). Brain region-specific gene expression activation required for reconsolidation and extinction of contextual fear memory. Journal of Neuroscience, 29(2), 402-413.

Mamiya, N., Fukushima, H., Suzuki, A., Matsuyama, Z., Homma, S., Frankland, P. W., & Kida, S. (2009b). Brain Region-Specific Gene Expression Activation Required for Reconsolidation and Extinction of Contextual Fear Memory. The Journal of Neuroscience, 29(2), 402. doi:10.1523/JNEUROSCI.4639-08.2009

Mangabeira, V., Garcia-Mijares, M., & Silva, M. T. A. (2015). Sugar withdrawal and differential reinforcement of low rate (DRL) performance in rats. Physiology & behavior, 139, 468-473.

Maniam, J., Antoniadis, C. P., & Morris, M. J. (2015). The effect of early-life stress and chronic high-sucrose diet on metabolic outcomes in female rats. Stress, 18(5), 524-537.

Mantyh, P. W. (2002). Neurobiology of substance P and the NK1 receptor. The Journal of clinical psychiatry, 63, 6-10.

404

Marek, R., Jin, J., Goode, T. D., Giustino, T. F., Wang, Q., Acca, G. M., . . . Lynagh, T. (2018a). Hippocampus-driven feed-forward inhibition of the prefrontal cortex mediates relapse of extinguished fear. Nature neuroscience, 1.

Marek, R., Jin, J., Goode, T. D., Giustino, T. F., Wang, Q., Acca, G. M., . . . Lynagh, T. (2018b). Hippocampus-driven feed-forward inhibition of the prefrontal cortex mediates relapse of extinguished fear. Nature neuroscience, 21(3), 384.

Marek, R., Strobel, C., Bredy, T. W., & Sah, P. (2013). The amygdala and medial prefrontal cortex: partners in the fear circuit. The Journal of physiology, 591(10), 2381-2391.

Maren, S. (2001). Neurobiology of Pavlovian fear conditioning. Annual review of neuroscience, 24(1), 897-931.

Maren, S. (2011). Seeking a Spotless Mind: Extinction, Deconsolidation, and Erasure of Fear Memory. Neuron, 70(5), 830-845. doi:10.1016/j.neuron.2011.04.023

Maren, S., Aharonov, G., & Fanselow, M. S. (1997). Neurotoxic lesions of the dorsal hippocampus and Pavlovian fear conditioning in rats. Behav Brain Res, 88(2), 261-274. doi:http://dx.doi.org/10.1016/S0166-4328(97)00088-0

Maren, S., & Fanselow, M. S. (1996). The amygdala and fear conditioning: has the nut been cracked? Neuron, 16(2), 237-240.

Maren, S., & Fanselow, M. S. (1997). Electrolytic lesions of the fimbria/fornix, dorsal hippocampus, or entorhinal cortex produce anterograde deficits in contextual fear conditioning in rats. Neurobiol Learn Mem, 67(2), 142-149.

Maren, S., & Holmes, A. (2016). Stress and fear extinction. Neuropsychopharmacology, 41(1), 58.

Maren, S., Phan, K. L., & Liberzon, I. (2013). The contextual brain: implications for fear conditioning, extinction and psychopathology. Nat Rev Neurosci, 14(6), 417-428. doi:10.1038/nrn3492

Maren, S., & Quirk, G. J. (2004). Neuronal signalling of fear memory. Nature reviews neuroscience, 5(11), 844.

Mark, G. P., Kinney, A. E., Grubb, M. C., Zhu, X., Finn, D. A., Mader, S. L., . . . Bechtholt, A. J. (2006). Injection of oxotremorine in nucleus accumbens shell reduces cocaine but not food self-administration in rats. Brain research, 1123(1), 51-59.

Martin, E. I., Ressler, K. J., Binder, E., & Nemeroff, C. B. (2009). The neurobiology of anxiety disorders: brain imaging, genetics, and psychoneuroendocrinology. Psychiatric Clinics, 32(3), 549-575.

Mathes, C. M., Gregson, J. R., & Spector, A. C. (2013). The selective serotonin reuptake inhibitor paroxetine decreases breakpoint of rats engaging in a progressive ratio licking task for sucrose and quinine solutions. Chem Senses, 38(3), 211-220. doi:10.1093/chemse/bjs096

Matikainen‐Ankney, B. A., & Kravitz, A. V. (2018). Persistent effects of obesity: a neuroplasticity hypothesis. Annals of the New York Academy of Sciences.

Mayford, M. (2014). The search for a hippocampal engram. Philos Trans R Soc Lond B Biol Sci, 369(1633), 20130161. doi:10.1098/rstb.2013.0161

McAllister-Williams, R., Ferrier, I., & Young, A. (1998). Mood and neuropsychological function in depression: the role of corticosteroids and serotonin. Psychological medicine, 28(3), 573-584.

McFarlane, A. C., Atchison, M., & Yehuda, R. (1997). The acute stress response following motor vehicle accidents and its relation to PTSD. Annals of the New York Academy of Sciences, 821(1), 437-441.

405

McGaugh, J. L. (2000). Memory--a century of consolidation. science, 287(5451), 248-251.

McGaugh, J. L., Cahill, L., Parent, M. B., Mesches, M. H., Coleman-Mesches, K., & Salinas, J. A. (1995). Involvement of the amygdala in the regulation of memory storage.

McKernan, M. G., & Shinnick-Gallagher, P. (1997). Fear conditioning induces a lasting potentiation of synaptic currents in vitro. Nature, 390(6660), 607-611. doi:http://dx.doi.org/10.1038/37605

Medina, J. F., Repa, J. C., Mauk, M. D., & LeDoux, J. E. (2002). Parallels between cerebellum-and amygdala-dependent conditioning. Nature reviews neuroscience, 3(2), 122.

Mehta, D., & Binder, E. B. (2012). Gene× environment vulnerability factors for PTSD: the HPA-axis. Neuropharmacology, 62(2), 654-662.

Mehta, N., Li, Z., Woolwine, B., Haroon, E., & Felger, J. (2019). Abstract# 3091 Inflammation is Associated with Decreased Amygdala to Ventromedial Prefrontal Functional Connectivity in Association with Symptoms of Anxiety in Patients with Depression. Brain, behavior, and immunity, 76, e15.

Meye, F. J., & Adan, R. A. (2014). Feelings about food: the ventral tegmental area in food reward and emotional eating. Trends in pharmacological sciences, 35(1), 31-40.

Milad, M. R., Orr, S. P., Lasko, N. B., Chang, Y., Rauch, S. L., & Pitman, R. K. (2008). Presence and acquired origin of reduced recall for fear extinction in PTSD: results of a twin study. Journal of psychiatric research, 42(7), 515-520.

Milad, M. R., & Quirk, G. J. (2002). Neurons in medial prefrontal cortex signal memory for fear extinction. Nature, 420(6911), 70-74.

Milad, M. R., & Quirk, G. J. (2012). Fear extinction as a model for translational neuroscience: ten years of progress. Annual review of psychology, 63, 129-151.

Milad, M. R., Rauch, S. L., Pitman, R. K., & Quirk, G. J. (2006). Fear extinction in rats: implications for human brain imaging and anxiety disorders. Biological psychology, 73(1), 61-71.

Miller, C. C., Martin, R. J., Whitney, M. L., & Edwards, G. L. (2002). Intracerebroventricular injection of fructose stimulates feeding in rats. Nutritional neuroscience, 5(5), 359-362.

Minatohara, K., Akiyoshi, M., & Okuno, H. (2016). Role of immediate-early genes in synaptic plasticity and neuronal ensembles underlying the memory trace. Frontiers in molecular neuroscience, 8, 78.

Ming, G.-l., & Song, H. (2011). Adult neurogenesis in the mammalian brain: significant answers and significant questions. Neuron, 70(4), 687-702.

Minichiello, L. (2009). TrkB signalling pathways in LTP and learning. Nature reviews neuroscience, 10, 850. doi:10.1038/nrn2738

https://www.nature.com/articles/nrn2738#supplementary-information Miserendino, M. J., Sananes, C. B., Melia, K. R., & Davis, M. (1990). Blocking of

acquisition but not expression of conditioned fear-potentiated startle by NMDA antagonists in the amygdala. Nature, 345(6277), 716.

Mishne, G., & Charles, A. S. (2019). Learning spatially-correlated temporal dictionaries for calcium imaging. arXiv preprint arXiv:1902.03132.

Misra, M., Bredella, M. A., Tsai, P., Mendes, N., Miller, K. K., & Klibanski, A. (2008). Lower growth hormone and higher cortisol are associated with

406

greater visceral adiposity, intramyocellular lipids, and insulin resistance in overweight girls. American Journal of Physiology-Endocrinology and Metabolism, 295(2), E385-E392.

Mitra, A., Gosnell, B. A., Schiöth, H. B., Grace, M. K., Klockars, A., Olszewski, P. K., & Levine, A. S. . (2010). Chronic sugar intake dampens feeding-related activity of neurons synthesizing a satiety mediator, oxytocin. Peptides, 31(7), 1346-1352.

Mizuno, K., Dempster, E., Mill, J., & Giese, K. P. (2012). Long-lasting regulation of hippocampal Bdnf gene transcription after contextual fear conditioning. Genes, Brain and Behavior, 11(6), 651-659. doi:10.1111/j.1601-183X.2012.00805.x

Moghaddam, B. (2018). 27.3 Discrete And Coordinated Encoding Of Rewarded Actions By Prefrontal Cortex And Dopamine Neurons. Schizophrenia Bulletin, 44(Suppl 1), S45.

Molnar, S. M., Beaton, L. E., Happer, J. P., Holcomb, L. A., Huang, S., Arienzo, D., & Marinkovic, K. (2018). Behavioral and Brain Activity Indices of Cognitive Control Deficits in Binge Drinkers. Brain sciences, 8(1), 9.

Molteni, R., Barnard, R., Ying, Z., Roberts, C., & Gomez-Pinilla, F. (2002). A high-fat, refined sugar diet reduces hippocampal brain-derived neurotrophic factor, neuronal plasticity, and learning. Neuroscience, 112(4), 803-814.

Monfils, M.-H., Cowansage, K. K., Klann, E., & LeDoux, J. E. (2009). Extinction-reconsolidation boundaries: key to persistent attenuation of fear memories. science, 324(5929), 951-955.

Moor, K. R., Scott, A. J., & McIntosh, W. D. (2013). Mindful eating and its relationship to body mass index and physical activity among university students. Mindfulness, 4(3), 269-274.

Moore, C. F., Sabino, V., & Cottone, P. (2018). Trace Amine Associated Receptor 1 (TAAR1) Modulation of Food Reward. Frontiers in pharmacology, 9, 129.

Moorman, D. E., & Aston-Jones, G. (2014). Orbitofrontal cortical neurons encode expectation-driven initiation of reward-seeking. Journal of Neuroscience, 34(31), 10234-10246.

Morey, R. A., Petty, C. M., Cooper, D. A., LaBar, K. S., & McCarthy, G. (2008). Neural systems for executive and emotional processing are modulated by symptoms of posttraumatic stress disorder in Iraq War veterans. Psychiatry Research: Neuroimaging, 162(1), 59-72.

Morgan, J. I., Cohen, D. R., Hempstead, J. L., & Curran, T. (1987). Mapping patterns of c-fos expression in the central nervous system after seizure. science, 237(4811), 192-197.

Morgan, J. I., & Curran, T. (1991). Stimulus-transcription coupling in the nervous system: involvement of the inducible proto-oncogenes fos and jun. Annual review of neuroscience, 14(1), 421-451.

Morrow, B. A., Elsworth, J. D., Inglis, F. M., & Roth, R. H. (1999). An antisense oligonucleotide reverses the footshock-induced expression of fos in the rat medial prefrontal cortex and the subsequent expression of conditioned fear-induced immobility. Journal of Neuroscience, 19(13), 5666-5673.

Moscarello, J. M., & Maren, S. (2018). Flexibility in the face of fear: hippocampal–prefrontal regulation of fear and avoidance. Current Opinion in Behavioral Sciences, 19, 44-49.

Moss, M. (2013). Salt, sugar, fat: How the food giants hooked us: Random House.

407

Murray, S., Tulloch, A., Criscitelli, K., & Avena, N. M. (2016). Recent studies of the effects of sugars on brain systems involved in energy balance and reward: Relevance to low calorie sweeteners. Physiology & behavior, 164, 504-508.

Musazzi, L., Mallei, A., Tardito, D., Gruber, S. H., El Khoury, A., Racagni, G., . . . Popoli, M. (2010). Early-life stress and antidepressant treatment involve synaptic signaling and Erk kinases in a gene-environment model of depression. Journal of psychiatric research, 44(8), 511-520.

Muscat, R., & Willner, P. (1989). Effects of dopamine receptor antagonists on sucrose consumption and preference. Psychopharmacology, 99(1), 98-102.

Myers, K. M., & Davis, M. (2007). Mechanisms of fear extinction. Molecular psychiatry, 12(2), 120.

Nader, K., Schafe, G. E., & LeDoux, J. E. (2000). Reply—Reconsolidation: The labile nature of consolidation theory. Nature reviews neuroscience, 1(3), 216-219.

Nair, A., & Bonneau, R. H. (2006). Stress-induced elevation of glucocorticoids increases microglia proliferation through NMDA receptor activation. Journal of neuroimmunology, 171(1-2), 72-85.

Nair, A., & Bonneau, R. H. (2006). Stress-induced elevation of glucocorticoids increases microglia proliferation through NMDA receptor activation. Journal of Neuroimmunology, 171(1), 72-85. doi:https://doi.org/10.1016/j.jneuroim.2005.09.012

Nakagawa, T., Hu, H., Zharikov, S., Tuttle, K. R., Short, R. A., Glushakova, O., . . . Herrera-Acosta, J. (2006). A causal role for uric acid in fructose-induced metabolic syndrome. American Journal of Physiology-Renal Physiology, 290(3), F625-F631.

Nakajima, K., & Kohsaka, S. (2002). Neuroprotective roles of microglia in the central nervous system Microglia in the Regenerating and Degenerating Central Nervous System (pp. 188-208): Springer.

Nakamura, K., Ono, T., & Tamura, R. (1987). Central sites involved in lateral hypothalamus conditioned neural responses to acoustic cues in the rat. Journal of Neurophysiology, 58(5), 1123-1148.

Nakamura, N. H., Fukunaga, M., Akama, K. T., Soga, T., Ogawa, S., & Pavlides, C. (2010). Hippocampal Cells Encode Places by Forming Small Anatomical Clusters. Neuroscience, 166(3), 994-1007. doi:10.1016/j.neuroscience.2009.12.069

Naneix, F., Darlot, F., Coutureau, E., & Cador, M. (2016). Long‐lasting deficits in hedonic and nucleus accumbens reactivity to sweet rewards by sugar overconsumption during adolescence. European Journal of Neuroscience, 43(5), 671-680.

Naneix, F., Darlot, F., De Smedt-Peyrusse, V., Pape, J.-R., Coutureau, E., & Cador, M. (2018). Protracted motivational dopamine-related deficits following adolescence sugar overconsumption. Neuropharmacology, 129, 16-25.

Nestler, E. J., & Aghajanian, G. K. (1997). Molecular and cellular basis of addiction. science, 278(5335), 58-63.

Newman, E., Riggs, D. S., & Roth, S. (1997). Thematic resolution, PTSD, and complex PTSD: The relationship between meaning and trauma-related diagnoses. Journal of traumatic stress, 10(2), 197-213.

Nicoll, R. A., Kauer, J., & Malenka, R. (1988). The current excitement in long term potentiation. Neuron, 1(2), 97-103.

408

Nieh, E. H., Matthews, G. A., Allsop, S. A., Presbrey, K. N., Leppla, C. A., Wichmann, R., . . . Tye, K. M. (2015). Decoding neural circuits that control compulsive sucrose seeking. Cell, 160(3), 528-541.

Niraula, A., Sheridan, J., & Godbout, J. (2019). Abstract# 3200 IL-6 production during repeated social defeat stress primes monocytes to induce anxiety-like behavior. Brain, behavior, and immunity, 76, e42-e43.

Noble, E. E., Hsu, T. M., Liang, J., & Kanoski, S. E. (2017). Early-life sugar consumption has long-term negative effects on memory function in male rats. Nutritional neuroscience, 1-11.

Noll, J. G., Zeller, M. H., Trickett, P. K., & Putnam, F. W. (2007). Obesity risk for female victims of childhood sexual abuse: a prospective study. Pediatrics, 120(1), e61-e67.

Nona, C. N., & Nobrega, J. N. (2018). A role for nucleus accumbens glutamate in the expression but not the induction of behavioural sensitization to ethanol. Behavioural Brain Research, 336, 269-281.

Novelle, M. G., & Diéguez, C. (2018). Food Addiction and Binge Eating: Lessons Learned from Animal Models. Nutrients, 10(1), 71.

Nummenmaa, L., & Tuominen, L. (2017). Opioid system and human emotions. British journal of pharmacology.

Ohki, K., & Reid, R. C. (2014). In vivo two-photon calcium imaging in the visual system. Cold Spring Harbor Protocols, 2014(4), pdb. prot081455.

Ohno, Y. (2019). Serotonin Receptors as the Therapeutic Target for Central Nervous System Disorders Serotonin (pp. 369-390): Elsevier.

Olds, M. E., & Olds, J. (1963). Approach‐avoidance analysis of rat diencephalon. Journal of Comparative Neurology, 120(2), 259-295.

Onoue, K., Nakayama, D., Ikegaya, Y., Matsuki, N., & Nomura, H. (2014). Fear extinction requires Arc/Arg3. 1 expression in the basolateral amygdala. Molecular brain, 7(1), 30.

Organization, W. H. (2016). World Health Organization obesity and overweight fact sheet.

Organization, W. H. (2017a). Depression and other common mental disorders: global health estimates.

Organization, W. H. (2017b). Health literacy. The solid facts. Health. Orsini, C. A., & Maren, S. (2012). Neural and cellular mechanisms of fear and

extinction memory formation. Neuroscience & Biobehavioral Reviews, 36(7), 1773-1802.

Oswald, K. D., Murdaugh, D. L., King, V. L., & Boggiano, M. M. (2011). Motivation for palatable food despite consequences in an animal model of binge eating. International journal of eating disorders, 44(3), 203-211.

Otmakhova, N., Duzel, E., Deutch, A. Y., & Lisman, J. (2013). The hippocampal-VTA loop: the role of novelty and motivation in controlling the entry of information into long-term memory Intrinsically motivated learning in natural and artificial systems (pp. 235-254): Springer.

Owens, M. J., & Nemeroff, C. B. (1994). Role of serotonin in the pathophysiology of depression: focus on the serotonin transporter. Clinical chemistry, 40(2), 288-295.

Padilla, S. L., Qiu, J., Soden, M. E., Sanz, E., Nestor, C. C., Barker, F. D., . . . Kelly, M. J. (2016). Agouti-related peptide neural circuits mediate adaptive behaviors in the starved state. Nature neuroscience, 19(5), 734.

409

Padwal, R. S., & Majumdar, S. R. (2007). Drug treatments for obesity: orlistat, sibutramine, and rimonabant. The Lancet, 369(9555), 71-77.

Page, K. A., Chan, O., Arora, J., Belfort-DeAguiar, R., Dzuira, J., Roehmholdt, B., . . . Constable, R. T. (2013). Effects of fructose vs glucose on regional cerebral blood flow in brain regions involved with appetite and reward pathways. Jama, 309(1), 63-70.

Palmiter, R. D. (2007). Is dopamine a physiologically relevant mediator of feeding behavior? Trends in neurosciences, 30(8), 375-381.

Pan, X., Zhang, C., & Shi, Z. (2011). Soft drink and sweet food consumption and suicidal behaviours among Chinese adolescents. Acta paediatrica, 100(11), e215-e222.

Pandit, R., Van Der Zwaal, E. M., Luijendijk, M. C., Brans, M. A., Van Rozen, A. J., Ophuis, R. J. O., . . . la Fleur, S. E. (2015). Central melanocortins regulate the motivation for sucrose reward. PloS one, 10(3), e0121768.

Pang, T. Y., Short, A. K., Bredy, T. W., & Hannan, A. J. (2017). Transgenerational paternal transmission of acquired traits: stress-induced modification of the sperm regulatory transcriptome and offspring phenotypes. Current Opinion in Behavioral Sciences, 14, 140-147.

Paolicelli, R. C., Bolasco, G., Pagani, F., Maggi, L., Scianni, M., Panzanelli, P., . . . Dumas, L. (2011). Synaptic pruning by microglia is necessary for normal brain development. science, 1202529.

Pape, H.-C., & Pare, D. (2010). Plastic synaptic networks of the amygdala for the acquisition, expression, and extinction of conditioned fear. Physiological reviews, 90(2), 419-463.

Pape, H. C., & Pare, D. (2010). Plastic synaptic networks of the amygdala for the acquisition, expression, and extinction of conditioned fear. Physiol Rev, 90(2), 419-463. doi:10.1152/physrev.00037.2009

Park, S., Kramer, E. E., Mercaldo, V., Rashid, A. J., Insel, N., Frankland, P. W., & Josselyn, S. A. (2016). Neuronal Allocation to a Hippocampal Engram. Neuropsychopharmacology: official publication of the American College of Neuropsychopharmacology.

Parkhurst, C. N., Yang, G., Ninan, I., Savas, J. N., Yates, J. R., Lafaille, J. J., . . . Gan, W.-B. (2013). Microglia promote learning-dependent synapse formation through brain-derived neurotrophic factor. Cell, 155(7), 1596-1609.

Parra-Frutos, I. (2013). Testing homogeneity of variances with unequal sample sizes. Computational Statistics, 28(3), 1269-1297.

Parylak, S. L., Cottone, P., Sabino, V., Rice, K. C., & Zorrilla, E. P. (2012). Effects of CB1 and CRF1 receptor antagonists on binge-like eating in rats with limited access to a sweet fat diet: lack of withdrawal-like responses. Physiology & behavior, 107(2), 231-242.

Patkar, O. L., Belmer, A., Holgate, J. Y., Tarren, J. R., Shariff, M. R., Morgan, M., . . . Klenowski, P. M. (2017). The antihypertensive drug pindolol attenuates long‐term but not short‐term binge‐like ethanol consumption in mice. Addiction biology, 22(3), 679-691.

Patriquin, M. A., & Mathew, S. J. (2017). The Neurobiological Mechanisms of Generalized Anxiety Disorder and Chronic Stress. Chronic Stress, 1, 2470547017703993.

Pavlov, I. P. (1927). Conditional reflexes: an investigation of the physiological activity of the cerebral cortex.

410

Paxinos, G., & Watson, C. (2006). The Rat Brain in Stereotaxic Coordinates: Academic Press.

Paxinos, G., & Watson, C. (2007). The rat brain in stereotaxic coordinates. Qingchuan Zhuge translate. People’s Medical Publishing House, Beijing, 32.

Pazos, A., & Palacios, J. (1985). Quantitative autoradiographic mapping of serotonin receptors in the rat brain. I. Serotonin-1 receptors. Brain research, 346(2), 205-230.

Pecina, S., & Berridge, K. (1995). Central enhancement of taste pleasure by intraventricular morphine. Neurobiology (Budapest, Hungary), 3(3-4), 269-280.

Peet, M. (2004). International variations in the outcome of schizophrenia and the prevalence of depression in relation to national dietary practices: an ecological analysis. The British Journal of Psychiatry, 184(5), 404-408.

Peng, S., Zhang, Y., Zhang, J., Wang, H., & Ren, B. (2010). ERK in Learning and Memory: A Review of Recent Research. International Journal of Molecular Sciences, 11(1), 222-232. doi:10.3390/ijms11010222

Peng, X. X., Ziff, E. B., & Carr, K. D. (2011). Effects of food restriction and sucrose intake on synaptic delivery of AMPA receptors in nucleus accumbens. Synapse, 65(10), 1024-1031.

Perez-Villalba, A., Mackintosh, N. J., & Canales, J. J. (2008). Influence of massed and distributed context preexposure on contextual fear and Egr-1 expression in the basolateral amygdala. Physiology & behavior, 93(1–2), 206-214. doi:http://dx.doi.org/10.1016/j.physbeh.2007.08.017

Perneger, T. V. (1998). What’s wrong with Bonferroni adjustments. BMJ : British Medical Journal, 316(7139), 1236-1238.

Petrovich, G. D., & Gallagher, M. (2003). Amygdala subsystems and control of feeding behavior by learned cues. Annals of the New York Academy of Sciences, 985(1), 251-262.

Petrovich, G. D., Hobin, M., & Reppucci, C. (2012). Selective Fos induction in hypothalamic orexin/hypocretin, but not melanin-concentrating hormone neurons, by a learned food-cue that stimulates feeding in sated rats. Neuroscience, 224, 70-80.

Phillips, R. G., & LeDoux, J. E. (1992). Differential contribution of amygdala and hippocampus to cued and contextual fear conditioning. Behavioral neuroscience, 106(2), 274-285. doi:10.1037/0735-7044.106.2.274

Phillips, R. G., & LeDoux, J. E. (1994). Lesions of the dorsal hippocampal formation interfere with background but not foreground contextual fear conditioning. Learning & Memory, 1(1), 34-44.

Pierce, R. C., & Kalivas, P. W. (1995). Amphetamine produces sensitized increases in locomotion and extracellular dopamine preferentially in the nucleus accumbens shell of rats administered repeated cocaine. Journal of Pharmacology and Experimental Therapeutics, 275(2), 1019-1029.

Pierson, J. L., Pullins, S. E., & Quinn, J. J. (2015). Dorsal hippocampus infusions of CNQX into the dentate gyrus disrupt expression of trace fear conditioning. Hippocampus, 25(7), 779-785. doi:10.1002/hipo.22413

Pignataro, A., Middei, S., Borreca, A., & Ammassari-Teule, M. (2013). Indistinguishable pattern of amygdala and hippocampus rewiring following tone or contextual fear conditioning in C57BL/6 mice. Frontiers in Behavioral Neuroscience, 7, 156. doi:10.3389/fnbeh.2013.00156

411

Pijl, H. (2003). Reduced dopaminergic tone in hypothalamic neural circuits: expression of a “thrifty” genotype underlying the metabolic syndrome? European journal of pharmacology, 480(1-3), 125-131.

Pillai, A. G., de Jong, D., Kanatsou, S., Krugers, H., Knapman, A., Heinzmann, J.-M., . . . Touma, C. (2012). Dendritic morphology of hippocampal and amygdalar neurons in adolescent mice is resilient to genetic differences in stress reactivity. PloS one, 7(6), e38971.

Pilowsky, P. M. (2018). Serotonin: The Mediator that Spans Evolution: Academic Press.

Pine, D. S., Cohen, P., Johnson, J. G., & Brook, J. S. (2002). Adolescent life events as predictors of adult depression. Journal of affective disorders, 68(1), 49-57.

Pinel, J. P., & Barnes, S. (2017). Biopsychology: Pearson Higher Ed. Pittenger, C., & Duman, R. S. (2008). Stress, depression, and neuroplasticity: a

convergence of mechanisms. Neuropsychopharmacology, 33(1), 88. Plath, N., Ohana, O., Dammermann, B., Errington, M. L., Schmitz, D., Gross, C., . . .

Welzl, H. (2006). Arc/Arg3. 1 is essential for the consolidation of synaptic plasticity and memories. Neuron, 52(3), 437-444.

Posadas, I., López-Hernández, B., & Ceña, V. (2013). Nicotinic receptors in neurodegeneration. Current neuropharmacology, 11(3), 298-314.

Pósfai, B., Cserép, C., Orsolits, B., & Dénes, Á. (2018). New Insights into Microglia–Neuron Interactions: A Neuron’s Perspective. Neuroscience. doi:https://doi.org/10.1016/j.neuroscience.2018.04.046

Pow, D., Perry, V., Morris, J., & Gordon, S. (1989). Microglia in the neurohypophysis associate with and endocytose terminal portions of neurosecretory neurons. Neuroscience, 33(3), 567-578.

Powers, S. K., & Howley, E. T. (2007). Exercise physiology: Theory and application to fitness and performance.

Pratt, W. E., & Kelley, A. E. (2004). Nucleus accumbens acetylcholine regulates appetitive learning and motivation for food via activation of muscarinic receptors. Behavioral neuroscience, 118(4), 730.

Preibisch, S., Saalfeld, S., & Tomancak, P. (2009). Globally optimal stitching of tiled 3D microscopic image acquisitions. Bioinformatics, 25(11), 1463-1465.

Price, J. L., Ko, A. I., Wade, M. J., Tsou, S. K., McKeel, D. W., & Morris, J. C. (2001). Neuron number in the entorhinal cortex and CA1 in preclinical Alzheimer disease. Archives of neurology, 58(9), 1395-1402.

Pursey, K. M., Stanwell, P., Callister, R. J., Brain, K., Collins, C. E., & Burrows, T. L. (2014). Neural Responses to Visual Food Cues According to Weight Status: A Systematic Review of Functional Magnetic Resonance Imaging Studies. Frontiers in Nutrition, 1(7). doi:10.3389/fnut.2014.00007

Qualls-Creekmore, E., Yu, S., Francois, M., Hoang, J., Huesing, C., Bruce-Keller, A., . . . Munzberg, H. (2017). Galanin-Expressing GABA Neurons in the Lateral Hypothalamus Modulate Food Reward and Noncompulsive Locomotion. J Neurosci, 37(25), 6053-6065. doi:10.1523/jneurosci.0155-17.2017

Quinn, J. J., Loya, F., Ma, Q. D., & Fanselow, M. S. (2005). Dorsal hippocampus NMDA receptors differentially mediate trace and contextual fear conditioning. Hippocampus, 15(5), 665-674. doi:10.1002/hipo.20088

Quinn, J. J., Wied, H. M., Ma, Q. D., Tinsley, M. R., & Fanselow, M. S. (2008). Dorsal hippocampus involvement in delay fear conditioning depends upon the strength of the tone‐footshock association. Hippocampus, 18(7), 640-654.

412

Quiñones, M., Al-Massadi, O., Folgueira, C., Bremser, S., Gallego, R., Torres-Leal, L., . . . Lam, B. Y. (2018). p53 in AgRP neurons is required for protection against diet-induced obesity via JNK1. Nature communications, 9(1), 3432.

Quirk, G. J. (2002). Memory for extinction of conditioned fear is long-lasting and persists following spontaneous recovery. Learning & Memory, 9(6), 402-407.

Quirk, G. J., Armony, J. L., & LeDoux, J. E. (1997). Fear conditioning enhances different temporal components of tone-evoked spike trains in auditory cortex and lateral amygdala. Neuron, 19(3), 613-624.

Quirk, G. J., Likhtik, E., Pelletier, J. G., & Paré, D. (2003). Stimulation of medial prefrontal cortex decreases the responsiveness of central amygdala output neurons. Journal of Neuroscience, 23(25), 8800-8807.

Quirk, G. J., & Mueller, D. (2008). Neural mechanisms of extinction learning and retrieval. Neuropsychopharmacology, 33(1), 56.

Quirk, S. E., Williams, L. J., O’Neil, A., Pasco, J. A., Jacka, F. N., Housden, S., . . . Brennan, S. L. (2013). The association between diet quality, dietary patterns and depression in adults: a systematic review. BMC psychiatry, 13(1), 175.

Rada, P., Avena, N., & Hoebel, B. (2005). Daily bingeing on sugar repeatedly releases dopamine in the accumbens shell. Neuroscience, 134(3), 737-744.

Rada, P., & Hoebel, B. G. (2005). Acetylcholine in the accumbens is decreased by diazepam and increased by benzodiazepine withdrawal: a possible mechanism for dependency. European journal of pharmacology, 508(1-3), 131-138.

Rada, P., Jensen, K., & Hoebel, B. G. (2001). Effects of nicotine and mecamylamine-induced withdrawal on extracellular dopamine and acetylcholine in the rat nucleus accumbens. Psychopharmacology, 157(1), 105-110.

Radley, J. J., Johnson, L. R., Janssen, W. G., Martino, J., Lamprecht, R., Hof, P. R., . . . Morrison, J. H. (2006). Associative Pavlovian conditioning leads to an increase in spinophilin-immunoreactive dendritic spines in the lateral amygdala. Eur J Neurosci, 24(3), 876-884. doi:10.1111/j.1460-9568.2006.04962.x

Rahman, S., Engleman, E. A., & Bell, R. L. (2015). Nicotinic receptor modulation to treat alcohol and drug dependence. Frontiers in neuroscience, 8, 426.

Rajmohan, V., & Mohandas, E. (2007). The limbic system. Indian journal of psychiatry, 49(2), 132.

Ramasubramanian, L., Lane, S., & Rahman, A. (2013). The association between maternal serious psychological distress and child obesity at 3 years: a cross‐sectional analysis of the UK Millennium Cohort Data. Child: care, health and development, 39(1), 134-140.

Ramírez-Amaya, V., Vazdarjanova, A., Mikhael, D., Rosi, S., Worley, P. F., & Barnes, C. A. (2005). Spatial exploration-induced Arc mRNA and protein expression: evidence for selective, network-specific reactivation. Journal of Neuroscience, 25(7), 1761-1768.

Rattiner, L. M., Davis, M., French, C. T., & Ressler, K. J. (2004). Brain-Derived Neurotrophic Factor and Tyrosine Kinase Receptor B Involvement in Amygdala-Dependent Fear Conditioning. The Journal of neuroscience, 24(20), 4796. doi:10.1523/JNEUROSCI.5654-03.2004

Reichelt, A. C., Maniam, J., Westbrook, R. F., & Morris, M. J. (2015). Dietary-induced obesity disrupts trace fear conditioning and decreases hippocampal reelin expression. Brain, behavior, and immunity, 43, 68-75. doi:https://doi.org/10.1016/j.bbi.2014.07.005

413

Reker, A. N., Oliveros, A., Sullivan, J. M., Nahar, L., Hinton, D. J., Kim, T., . . . Nam, H. W. (2018). Neurogranin in the nucleus accumbens regulates NMDA receptor tolerance and motivation for ethanol seeking. Neuropharmacology, 131, 58-67.

Rena, G., Hardie, D. G., & Pearson, E. R. (2017). The mechanisms of action of metformin. Diabetologia, 60(9), 1577-1585.

Renier, N., Adams, E. L., Kirst, C., Wu, Z., Azevedo, R., Kohl, J., . . . Zhou, Y. (2016). Mapping of brain activity by automated volume analysis of immediate early genes. Cell, 165(7), 1789-1802.

Rescorla, R. A. (1988). Behavioral studies of Pavlovian conditioning. Annual review of neuroscience, 11(1), 329-352.

Rescorla, R. A., & Heth, C. D. (1975). Reinstatement of fear to an extinguished conditioned stimulus. Journal of Experimental Psychology: Animal Behavior Processes, 1(1), 88.

Reyes, S., Peirano, P., Peigneux, P., Lozoff, B., & Algarin, C. (2015). Inhibitory control in otherwise healthy overweight 10-year-old children. International Journal of Obesity, 39(8), 1230.

Reznikov, L. R., Fadel, J. R., & Reagan, L. P. (2011). Glutamate-mediated neuroplasticity deficits in mood disorders Neuroplasticity (pp. 13-26): Springer.

Richard, D., Ferland, J., Lalonde, J., Samson, P., & Deshaies, Y. (2000). Influence of topiramate in the regulation of energy balance. Nutrition, 16(10), 961-966.

Riga, D., Theijs, J. T., De Vries, T. J., Smit, A. B., & Spijker, S. (2015). Social defeat-induced anhedonia: effects on operant sucrose-seeking behavior. Frontiers in Behavioral Neuroscience, 9, 195.

Ritter, R. C., Slusser, P. G., & Stone, S. (1981). Glucoreceptors controlling feeding and blood glucose: location in the hindbrain. science, 213(4506), 451-452.

Robinson, M. J., Warlow, S. M., & Berridge, K. C. (2014). Optogenetic excitation of central amygdala amplifies and narrows incentive motivation to pursue one reward above another. Journal of Neuroscience, 34(50), 16567-16580.

Robinson, T. E. (1988). Stimulant drugs and stress: factors influencing individual differences in the susceptibility to sensitization. Sensitization of the nervous system, 145-173.

Robinson, T. E., & Berridge, K. C. (1993). The neural basis of drug craving: an incentive-sensitization theory of addiction. Brain Research Reviews, 18(3), 247-291.

Robinson, T. E., & Berridge, K. C. (2001). Incentive‐sensitization and addiction. Addiction, 96(1), 103-114.

Robinson, T. E., & Berridge, K. C. (2008). The incentive sensitization theory of addiction: some current issues. Philosophical Transactions of the Royal Society B: Biological Sciences, 363(1507), 3137-3146.

Rodrigues, S. M., Schafe, G. E., & LeDoux, J. E. (2004). Molecular mechanisms underlying emotional learning and memory in the lateral amygdala. Neuron, 44(1), 75-91.

Rogan, M. T., Staubli, U. V., & LeDoux, J. E. (1997). Fear conditioning induces associative long-term potentiation in the amygdala. Nature, 390(6660), 604-607. doi:http://dx.doi.org/10.1038/37601

Rogan, M. T., Stäubli, U. V., & LeDoux, J. E. (1997). Fear conditioning induces associative long-term potentiation in the amygdala. Nature, 390(6660), 604.

414

Rogers, J. L., Hunsaker, M. R., & Kesner, R. P. (2006). Effects of ventral and dorsal CA1 subregional lesions on trace fear conditioning. Neurobiol Learn Mem, 86(1), 72-81. doi:http://dx.doi.org/10.1016/j.nlm.2006.01.002

Rogerson, T., Cai, D. J., Frank, A., Sano, Y., Shobe, J., Lopez-Aranda, M. F., & Silva, A. J. (2014). Synaptic tagging during memory allocation. Nat Rev Neurosci, 15(3), 157-169. doi:10.1038/nrn3667

Romano, S. A., Pérez-Schuster, V., Jouary, A., Boulanger-Weill, J., Candeo, A., Pietri, T., & Sumbre, G. (2017). An integrated calcium imaging processing toolbox for the analysis of neuronal population dynamics. PLoS computational biology, 13(6), e1005526.

Romanski, L. M., Clugnet, M.-C., Bordi, F., & LeDoux, J. E. (1993). Somatosensory and auditory convergence in the lateral nucleus of the amygdala. Behavioral neuroscience, 107(3), 444.

Rosas, M., Porru, S., Sabariego, M., Piludu, M. A., Giorgi, O., Corda, M. G., & Acquas, E. (2018). Effects of morphine on place conditioning and ERK 1/2 phosphorylation in the nucleus accumbens of psychogenetically selected Roman low-and high-avoidance rats. Psychopharmacology, 235(1), 59-69.

Rosenkranz, J. A., Moore, H., & Grace, A. A. (2003). The prefrontal cortex regulates lateral amygdala neuronal plasticity and responses to previously conditioned stimuli. Journal of Neuroscience, 23(35), 11054-11064.

Rosenzweig, M. R., & Bennett, E. L. (1996). Psychobiology of plasticity: effects of training and experience on brain and behavior. Behavioural Brain Research, 78(1), 57-65.

Rothbaum, B. O., & Davis, M. (2003). Applying Learning Principles to the Treatment of Post-Trauma Reactions. Annals of the New York Academy of Sciences, 1008(1), 112-121. doi:10.1196/annals.1301.012

Rudy, J. W., Huff, N. C., & Matus-Amat, P. (2004). Understanding contextual fear conditioning: insights from a two-process model. Neuroscience & Biobehavioral Reviews, 28(7), 675-685. doi:http://dx.doi.org/10.1016/j.neubiorev.2004.09.004

Ruegg, H., Yu, W.-Z., & Bodnar, R. J. (1997). Opioid-receptor subtype agonist-induced enhancements of sucrose intake are dependent upon sucrose concentration. Physiology & behavior, 62(1), 121-128.

Rumpel, S., LeDoux, J., Zador, A., & Malinow, R. (2005). Postsynaptic Receptor Trafficking Underlying a Form of Associative Learning. science, 308(5718), 83-88. doi:10.1126/science.1103944

Sah, P., Westbrook, R., & Lüthi, A. (2008). Fear Conditioning and Long‐term Potentiation in the Amygdala. Annals of the New York Academy of Sciences, 1129(1), 88-95.

Salamone, J. D., & Correa, M. (2012). The mysterious motivational functions of mesolimbic dopamine. Neuron, 76(3), 470-485.

Sama, P. R., Eapen, Z. J., Weinfurt, K. P., Shah, B. R., & Schulman, K. A. (2014). An evaluation of mobile health application tools. JMIR mHealth and uHealth, 2(2).

Sananbenesi, F., Fischer, A., Schrick, C., Spiess, J., & Radulovic, J. (2002). Phosphorylation of Hippocampal Erk-1/2, Elk-1, and p90-Rsk-1 during Contextual Fear Conditioning: Interactions between Erk-1/2 and Elk-1. Molecular and Cellular Neuroscience, 21(3), 463-476. doi:http://dx.doi.org/10.1006/mcne.2002.1188

415

Sánchez-Villegas, A., Toledo, E., de Irala, J., Ruiz-Canela, M., Pla-Vidal, J., & Martínez-González, M. A. (2012). Fast-food and commercial baked goods consumption and the risk of depression. Public health nutrition, 15(3), 424-432.

Sandweiss, A. J., & Vanderah, T. W. (2015). The pharmacology of neurokinin receptors in addiction: prospects for therapy. Substance abuse and rehabilitation, 6, 93.

Sangüesa, G., Cascales, M., Griñán, C., Sánchez, R. M., Roglans, N., Pallàs, M., . . . Alegret, M. (2018). Impairment of Novel Object Recognition Memory and Brain Insulin Signaling in Fructose-but Not Glucose-Drinking Female Rats. Molecular neurobiology, 1-16.

Santini, E., Ge, H., Ren, K., de Ortiz, S. P., & Quirk, G. J. (2004). Consolidation of fear extinction requires protein synthesis in the medial prefrontal cortex. Journal of Neuroscience, 24(25), 5704-5710.

Santos, C. J., Ferreira, A. V., Oliveira, A. L., Oliveira, M. C., Gomes, J. S., & Aguiar, D. C. (2018). Carbohydrate-enriched diet predispose to anxiety and depression-like behavior after stress in mice. Nutritional neuroscience, 21(1), 33-39.

Savage, S. W., Zald, D. H., Cowan, R. L., Volkow, N. D., Marks‐Shulman, P. A., Kessler, R. M., . . . Dunn, J. P. (2014). Regulation of novelty seeking by midbrain dopamine D2/D3 signaling and ghrelin is altered in obesity. Obesity, 22(6), 1452-1457.

Schafe, G. E., Atkins, C. M., Swank, M. W., Bauer, E. P., Sweatt, J. D., & LeDoux, J. E. (2000). Activation of ERK/MAP kinase in the amygdala is required for memory consolidation of pavlovian fear conditioning. The Journal of neuroscience, 20(21), 8177-8187.

Schafe, G. E., & LeDoux, J. E. (2000). Memory consolidation of auditory pavlovian fear conditioning requires protein synthesis and protein kinase A in the amygdala. Journal of Neuroscience, 20(18), RC96-RC96.

Schafe, G. E., Nader, K., Blair, H. T., & LeDoux, J. E. (2001). Memory consolidation of Pavlovian fear conditioning: a cellular and molecular perspective. Trends in neurosciences, 24(9), 540-546.

Schindelin, J., Arganda-Carreras, I., Frise, E., Kaynig, V., Longair, M., Pietzsch, T., . . . Schmid, B. (2012). Fiji: an open-source platform for biological-image analysis. Nature methods, 9(7), 676.

Schneider, L. H. (1989). Orosensory Self‐Stimulation by Sucrose Involves Brain Dopaminergic Mechanisms a. Annals of the New York Academy of Sciences, 575(1), 307-320.

Scholzen, T., & Gerdes, J. (2000). The Ki‐67 protein: from the known and the unknown. Journal of cellular physiology, 182(3), 311-322.

Schwartzbaum, J. (1988). Electrophysiology of taste, feeding and reward in lateral hypothalamus of rabbit. Physiology & behavior, 44(4-5), 507-526.

See, R. E., Fuchs, R. A., Ledford, C. C., & McLAUGHLIN, J. (2003). Drug addiction, relapse, and the amygdala. Annals of the New York Academy of Sciences, 985(1), 294-307.

Shariff, M., Klenowski, P., Morgan, M., Patkar, O., Mu, E., Bellingham, M., . . . Bartlett, S. E. (2017). Binge-like sucrose consumption reduces the dendritic length and complexity of principal neurons in the adolescent rat basolateral amygdala. PloS one, 12(8), e0183063.

416

Shariff, M., Quik, M., Holgate, J., Morgan, M., Patkar, O. L., Tam, V., . . . Bartlett, S. E. (2016). Neuronal nicotinic acetylcholine receptor modulators reduce sugar intake. PloS one, 11(3), e0150270.

Sharpe, M., Clemens, K., Morris, M., & Westbrook, R. (2016). Daily exposure to sucrose impairs subsequent learning about food cues: a role for alterations in ghrelin signaling and dopamine D2 receptors. Neuropsychopharmacology, 41(5), 1357.

Shearrer, G., Daniels, M., Toledo-Corral, C., Weigensberg, M., Spruijt-Metz, D., & Davis, J. (2016). Associations among sugar sweetened beverage intake, visceral fat, and cortisol awakening response in minority youth. Physiology & behavior, 167, 188-193.

Shearrer, G. E., Stice, E., & Burger, K. S. (2018). Adolescents at high risk of obesity show greater striatal response to increased sugar content in milkshakes. The American journal of clinical nutrition, 107(6), 859-866.

Shearrer, G. E., Stice, E., & Burger, K. S. (2018). Adolescents at high risk of obesity show greater striatal response to increased sugar content in milkshakes. The American journal of clinical nutrition, nqy050-nqy050. doi:10.1093/ajcn/nqy050

Shi, Z., Taylor, A. W., Wittert, G., Goldney, R., & Gill, T. K. (2010). Soft drink consumption and mental health problems among adults in Australia. Public health nutrition, 13(7), 1073-1079.

Shin, L. M., Orr, S. P., Carson, M. A., Rauch, S. L., Macklin, M. L., Lasko, N. B., . . . Cannistraro, P. A. (2004). Regional cerebral blood flow in the amygdala and medial prefrontalcortex during traumatic imagery in male and female vietnam veterans with ptsd. Archives of general psychiatry, 61(2), 168-176.

Shin, L. M., Rauch, S. L., & Pitman, R. K. (2006). Amygdala, medial prefrontal cortex, and hippocampal function in PTSD. Annals of the New York Academy of Sciences, 1071(1), 67-79.

Short, A., Fennell, K., Perreau, V., Fox, A., O’Bryan, M., Kim, J., . . . Hannan, A. (2016). Elevated paternal glucocorticoid exposure alters the small noncoding RNA profile in sperm and modifies anxiety and depressive phenotypes in the offspring. Translational psychiatry, 6(6), e837.

Sibille, K. T., Bartsch, F., Reddy, D., Fillingim, R. B., & Keil, A. (2016). Increasing neuroplasticity to bolster chronic pain treatment: a role for intermittent fasting and glucose administration? The Journal of Pain, 17(3), 275-281.

Silverman, M. N., Pearce, B. D., Biron, C. A., & Miller, A. H. (2005). Immune modulation of the hypothalamic-pituitary-adrenal (HPA) axis during viral infection. Viral immunology, 18(1), 41-78.

Simon, G. E., Von Korff, M., Saunders, K., Miglioretti, D. L., Crane, P. K., Van Belle, G., & Kessler, R. C. (2006). Association between obesity and psychiatric disorders in the US adult population. Archives of general psychiatry, 63(7), 824-830.

Sinha, R. (2008). Chronic stress, drug use, and vulnerability to addiction. Annals of the New York Academy of Sciences, 1141(1), 105-130.

Sinha, R. (2018). Role of addiction and stress neurobiology on food intake and obesity. Biological psychology, 131, 5-13.

Smith, D. M., & Mizumori, S. J. Y. (2006a). Hippocampal place cells, context, and episodic memory. Hippocampus, 16(9), 716-729.

Smith, D. M., & Mizumori, S. J. Y. (2006b). Learning-Related Development of Context-Specific Neuronal Responses to Places and Events: The

417

Hippocampal Role in Context Processing. The Journal of neuroscience, 26(12), 3154-3163. doi:10.1523/jneurosci.3234-05.2006

Smith, I. (1960). Carbohydrates. Chromatographic and electrophoretic techniques, 1, 246-260.

Smith, K. S., Bucci, D. J., Luikart, B. W., & Mahler, S. V. (2016). DREADDS: Use and application in behavioral neuroscience. Behavioral neuroscience, 130(2), 137.

Smith, S. M., & Vale, W. W. (2006). The role of the hypothalamic-pituitary-adrenal axis in neuroendocrine responses to stress. Dialogues in clinical neuroscience, 8(4), 383.

Sohn, J.-W., Elmquist, J. K., & Williams, K. W. (2013). Neuronal circuits that regulate feeding behavior and metabolism. Trends in neurosciences, 36(9), 504-512.

Soltys, Z., Orzylowska-Sliwinska, O., Zaremba, M., Orlowski, D., Piechota, M., Fiedorowicz, A., . . . Oderfeld-Nowak, B. (2005). Quantitative morphological study of microglial cells in the ischemic rat brain using principal component analysis. Journal of neuroscience methods, 146(1), 50-60.

Song, E. Y., Kim, Y. B., Kim, Y. H., & Jung, M. W. (2005). Role of active movement in place-specific firing of hippocampal neurons. Hippocampus, 15(1), 8-17.

Soto, M., Chaumontet, C., Even, P. C., Nadkarni, N., Piedcoq, J., Darcel, N., ... & Fromentin, G. . ( 2015). Intermittent access to liquid sucrose differentially modulates energy intake and related central pathways in control or high-fat fed mice. Physiology & behavior( 140), 44-53.

Sotres-Bayon, F., Cain, C. K., & LeDoux, J. E. (2006). Brain mechanisms of fear extinction: historical perspectives on the contribution of prefrontal cortex. Biological psychiatry, 60(4), 329-336.

Southgate, D. (1995). Digestion and metabolism of sugars. The American journal of clinical nutrition, 62(1), 203S-210S.

Souza, C., Moreira, J., Siqueira, I., Pereira, A., Rieger, D., Souza, D., . . . Perry, M. (2007). Highly palatable diet consumption increases protein oxidation in rat frontal cortex and anxiety-like behavior. Life sciences, 81(3), 198-203.

Spangler, R., Wittkowski, K. M., Goddard, N. L., Avena, N. M., Hoebel, B. G., & Leibowitz, S. F. (2004). Opiate-like effects of sugar on gene expression in reward areas of the rat brain. Molecular Brain Research, 124(2), 134-142.

Sparta, D. R., Smithuis, J., Stamatakis, A. M., Jennings, J. H., Kantak, P. A., Ung, R. L., & Stuber, G. D. (2014). Inhibition of projections from the basolateral amygdala to the entorhinal cortex disrupts the acquisition of contextual fear. Frontiers in Behavioral Neuroscience, 8, 129. doi:10.3389/fnbeh.2014.00129

Spielman, L. J., Little, J. P., & Klegeris, A. (2014). Inflammation and insulin/IGF-1 resistance as the possible link between obesity and neurodegeneration. Journal of neuroimmunology, 273(1-2), 8-21.

Squire, L. R., & Alvarez, P. (1995). Retrograde amnesia and memory consolidation: a neurobiological perspective. Current opinion in neurobiology, 5(2), 169-177.

Staley, J. K., & Mash, D. C. (1996). Adaptive increase in D3 dopamine receptors in the brain reward circuits of human cocaine fatalities. Journal of Neuroscience, 16(19), 6100-6106.

Stanciu, M., Radulovic, J., & Spiess, J. (2001). Phosphorylated cAMP response element binding protein in the mouse brain after fear conditioning:

418

relationship to Fos production. Molecular Brain Research, 94(1), 15-24. doi:https://doi.org/10.1016/S0169-328X(01)00174-7

Stanley, B. G., Lanthier, D., & Leibowitz, S. F. (1988). Multiple brain sites sensitive to feeding stimulation by opioid agonists: a cannula-mapping study. Pharmacology Biochemistry and Behavior, 31(4), 825-832.

Steensland, P., Simms, J. A., Nielsen, C. K., Holgate, J., Bito-Onon, J. J., & Bartlett, S. E. (2010). The neurokinin 1 receptor antagonist, ezlopitant, reduces appetitive responding for sucrose and ethanol. PloS one, 5(9), e12527.

Stence, N., Waite, M., & Dailey, M. E. (2001). Dynamics of microglial activation: A confocal time‐lapse analysis in hippocampal slices. Glia, 33(3), 256-266.

Stranahan, A. M., Norman, E. D., Lee, K., Cutler, R. G., Telljohann, R. S., Egan, J. M., & Mattson, M. P. (2008). Diet‐induced insulin resistance impairs hippocampal synaptic plasticity and cognition in middle‐aged rats. Hippocampus, 18(11), 1085-1088.

Straube, T., Mentzel, H.-J., & Miltner, W. H. (2005). Common and distinct brain activation to threat and safety signals in social phobia. Neuropsychobiology, 52(3), 163-168.

Strobel, C., Marek, R., Gooch, H. M., Sullivan, R. K., & Sah, P. (2015). Prefrontal and auditory input to intercalated neurons of the amygdala. Cell reports, 10(9), 1435-1442.

Stroebe, W., Papies, E. K., & Aarts, H. (2008). From homeostatic to hedonic theories of eating: Self‐regulatory failure in food‐rich environments. Applied Psychology, 57, 172-193.

Stunkard, A. J., Faith, M. S., & Allison, K. C. (2003). Depression and obesity. Biological psychiatry, 54(3), 330-337.

Stutzmann, G. E., & LeDoux, J. E. (1999). GABAergic antagonists block the inhibitory effects of serotonin in the lateral amygdala: a mechanism for modulation of sensory inputs related to fear conditioning. Journal of Neuroscience, 19(11), RC8-RC8.

Suzuki, T., Ishigooka, J., Watanabe, S., & Miyaoka, H. (2002). Enhancement of delayed release of dopamine in the amygdala induced by conditioned fear stress in methamphetamine-sensitized rats. European journal of pharmacology, 435(1), 59-65.

Swanton, T., Cook, J., Beswick, J. A., Freeman, S., Lawrence, C. B., & Brough, D. (2018). Is Targeting the Inflammasome a Way Forward for Neuroscience Drug Discovery? SLAS DISCOVERY: Advancing Life Sciences R&D, 2472555218786210.

Sweeney, P., & Yang, Y. (2017). Neural circuit mechanisms underlying emotional regulation of homeostatic feeding. Trends in Endocrinology & Metabolism, 28(6), 437-448.

Tabuchi, E., Yokawa, T., Mallick, H., Inubushi, T., Kondoh, T., Ono, T., & Torii, K. (2002). Spatio–temporal dynamics of brain activated regions during drinking behavior in rats. Brain research, 951(2), 270-279.

Tahara, A., Takasu, T., Yokono, M., Imamura, M., & Kurosaki, E. (2018). Antidiabetic and antiobesity effects of SGLT2 inhibitor ipragliflozin in type 2 diabetic mice fed sugar solution. European journal of pharmacology, 818, 545-553.

Takei, S., Morinobu, S., Yamamoto, S., Fuchikami, M., Matsumoto, T., & Yamawaki, S. (2011). Enhanced hippocampal BDNF/TrkB signaling in response to fear conditioning in an animal model of posttraumatic stress

419

disorder. Journal of psychiatric research, 45(4), 460-468. doi:https://doi.org/10.1016/j.jpsychires.2010.08.009

Tarren, J. R., Lester, H. A., Belmer, A., & Bartlett, S. E. (2017). Acute ethanol administration upregulates synaptic α4-subunit of neuronal nicotinic acetylcholine receptors within the nucleus accumbens and amygdala. Frontiers in molecular neuroscience, 10, 338.

Tarrier, N. (2001). What can be learned from clinical trials? Reply to Devilly and Foa (2001).

Teff, K. L., Elliott, S. S., Tschöp, M., Kieffer, T. J., Rader, D., Heiman, M., . . . Havel, P. J. (2004). Dietary fructose reduces circulating insulin and leptin, attenuates postprandial suppression of ghrelin, and increases triglycerides in women. The Journal of Clinical Endocrinology & Metabolism, 89(6), 2963-2972.

Tellez, L. A., Han, W., Zhang, X., Ferreira, T. L., Perez, I. O., Shammah-Lagnado, S. J., . . . De Araujo, I. E. (2016). Separate circuitries encode the hedonic and nutritional values of sugar. Nature neuroscience, 19(3), 465.

Torres-Platas, S. G., Comeau, S., Rachalski, A., Dal Bo, G., Cruceanu, C., Turecki, G., . . . Mechawar, N. (2014). Morphometric characterization of microglial phenotypes in human cerebral cortex. Journal of neuroinflammation, 11(1), 12.

Traverso, L. M., Ruiz, G., & De la Casa, L. G. (2012). Effect of the NMDA antagonist MK-801 on latent inhibition of fear conditioning. Pharmacology Biochemistry and Behavior, 102(4), 488-494. doi:http://dx.doi.org/10.1016/j.pbb.2012.06.014

Tremblay, M.-È., Lowery, R. L., & Majewska, A. K. (2010). Microglial interactions with synapses are modulated by visual experience. PLoS biology, 8(11), e1000527.

Trifilieff, P., Calandreau, L., Herry, C., Mons, N., & Micheau, J. (2007). Biphasic ERK1/2 activation in both the hippocampus and amygdala may reveal a system consolidation of contextual fear memory. Neurobiology of learning and memory, 88(4), 424-434.

Trifilieff, P., Herry, C., Vanhoutte, P., Caboche, J., Desmedt, A., Riedel, G., . . . Micheau, J. (2006). Foreground contextual fear memory consolidation requires two independent phases of hippocampal ERK/CREB activation. Learning & Memory, 13(3), 349-358. doi:10.1101/lm.80206

Trogrlic, L., Wilson, Y. M., Newman, A. G., & Murphy, M. (2011). Context fear learning specifically activates distinct populations of neurons in amygdala and hypothalamus. Learning & Memory, 18(10), 678-687. doi:10.1101/lm.2314311

Tronson, N. C., & Taylor, J. R. (2007). Molecular mechanisms of memory reconsolidation. Nature reviews neuroscience, 8(4), 262.

Tryon, M. S., Stanhope, K. L., Epel, E. S., Mason, A. E., Brown, R., Medici, V., . . . Laugero, K. D. (2015). Excessive sugar consumption may be a difficult habit to break: a view from the brain and body. The Journal of Clinical Endocrinology & Metabolism, 100(6), 2239-2247.

Tucker, D. M., Anderson, E., & Luu, P. (2019). Method for using electrical stimulation for cerebral network analysis: US Patent App. 14/195,593.

Tye, K. M., Stuber, G. D., De Ridder, B., Bonci, A., & Janak, P. H. (2008). Rapid strengthening of thalamo-amygdala synapses mediates cue–reward learning. Nature, 453(7199), 1253.

420

Tynan, R. J., Naicker, S., Hinwood, M., Nalivaiko, E., Buller, K. M., Pow, D. V., . . . Walker, F. R. (2010). Chronic stress alters the density and morphology of microglia in a subset of stress-responsive brain regions. Brain, Behavior, and Immunity, 24(7), 1058-1068. doi:https://doi.org/10.1016/j.bbi.2010.02.001

Ulrich-Lai, Y. M., Christiansen, A. M., Ostrander, M. M., Jones, A. A., Jones, K. R., Choi, D. C., . . . Davis, J. F. (2010). Pleasurable behaviors reduce stress via brain reward pathways. Proceedings of the National Academy of Sciences, 107(47), 20529-20534.

Ulrich-Lai, Y. M., Ostrander, M. M., & Herman, J. P. (2011). HPA axis dampening by limited sucrose intake: reward frequency vs. caloric consumption. Physiology & behavior, 103(1), 104-110.

Ulrich, M., Stauß, P., & Grön, G. (2018). Glucose Modulates Human Ventral Tegmental Activity in Response to Sexual Stimuli. The journal of sexual medicine, 15(1), 20-28.

Urcelay, G. P., & Miller, R. R. (2010). Two roles of the context in Pavlovian fear conditioning. Journal of Experimental Psychology: Animal Behavior Processes, 36(2), 268.

Ursano, R. J., Zhang, L., Li, H., Johnson, L., Carlton, J., Fullerton, C. S., & Benedek, D. M. (2009). PTSD and traumatic stress: from gene to community and bench to bedside. Brain research, 1293, 2-12.

Valencia-Torres, L., Olarte-Sánchez, C. M., Lyons, D. J., Georgescu, T., Greenwald-Yarnell, M., Myers Jr, M. G., . . . Heisler, L. K. (2017). Activation of Ventral Tegmental Area 5-HT 2C Receptors Reduces Incentive Motivation. Neuropsychopharmacology, 42(7), 1511.

Van der Borght, K., Köhnke, R., Göransson, N., Deierborg, T., Brundin, P., Erlanson-Albertsson, C., & Lindqvist, A. (2011). Reduced neurogenesis in the rat hippocampus following high fructose consumption. Regulatory peptides, 167(1), 26-30.

Van der Kolk, B. A. (1994). The body keeps the score: Memory and the evolving psychobiology of posttraumatic stress. Harvard review of psychiatry, 1(5), 253-265.

van Dooren, F. E., Nefs, G., Schram, M. T., Verhey, F. R., Denollet, J., & Pouwer, F. (2013). Depression and risk of mortality in people with diabetes mellitus: a systematic review and meta-analysis. PloS one, 8(3), e57058.

Van Strien, T., Frijters, J. E., Bergers, G. P., & Defares, P. B. (1986). The Dutch Eating Behavior Questionnaire (DEBQ) for assessment of restrained, emotional, and external eating behavior. International Journal of eating disorders, 5(2), 295-315.

Varghese, F. P., & Brown, E. S. (2001). The hypothalamic-pituitary-adrenal axis in major depressive disorder: a brief primer for primary care physicians. Primary care companion to the Journal of clinical psychiatry, 3(4), 151.

Varvel, N. H., Grathwohl, S. A., Baumann, F., Liebig, C., Bosch, A., Brawek, B., . . . Aguzzi, A. (2012). Microglial repopulation model reveals a robust homeostatic process for replacing CNS myeloid cells. Proceedings of the National Academy of Sciences, 201210150.

Vendruscolo, L. F., Gueye, A. B., Darnaudéry, M., Ahmed, S. H., & Cador, M. (2010). Sugar overconsumption during adolescence selectively alters motivation and reward function in adult rats. PloS one, 5(2), e9296.

Verhoeven, K. J., Simonsen, K. L., & McIntyre, L. M. (2005). Implementing false discovery rate control: increasing your power. Oikos, 108(3), 643-647.

421

Vertes, R. P. (2004). Differential projections of the infralimbic and prelimbic cortex in the rat. Synapse, 51(1), 32-58.

Verty, A. N., McGregor, I. S., & Mallet, P. E. (2004). The dopamine receptor antagonist SCH 23390 attenuates feeding induced by Δ9-tetrahydrocannabinol. Brain research, 1020(1-2), 188-195.

Vigano, D., Rubino, T., Di Chiara, G., Ascari, I., Massi, P., & Parolaro, D. (2003). μ opioid receptor signaling in morphine sensitization. Neuroscience, 117(4), 921-929.

Volkow, N. D., Fowler, J., & Wang, G.-J. (2002). Role of dopamine in drug reinforcement and addiction in humans: results from imaging studies. Behavioural pharmacology, 13(5-6), 355-366.

Volkow, N. D., & Fowler, J. S. (2000). Addiction, a disease of compulsion and drive: involvement of the orbitofrontal cortex. Cerebral cortex, 10(3), 318-325.

Volkow, N. D., Fowler, J. S., Wang, G.-J., & Swanson, J. M. (2004). Dopamine in drug abuse and addiction: results from imaging studies and treatment implications. Molecular psychiatry, 9(6), 557.

Volkow, N. D., & Wise, R. A. (2005). How can drug addiction help us understand obesity? Nature neuroscience, 8(5), 555.

Vollmayr, B., Bachteler, D., Vengeliene, V., Gass, P., Spanagel, R., & Henn, F. (2004). Rats with congenital learned helplessness respond less to sucrose but show no deficits in activity or learning. Behavioural Brain Research, 150(1-2), 217-221.

Vollmer, L. L., Ghosal, S., McGuire, J. L., Ahlbrand, R. L., Li, K.-Y., Santin, J. M., . . . Sah, R. (2016). Microglial Acid Sensing Regulates Carbon Dioxide-Evoked Fear. Biological psychiatry, 80(7), 541-551. doi:https://doi.org/10.1016/j.biopsych.2016.04.022

Vousden, D. A., Epp, J., Okuno, H., Nieman, B. J., van Eede, M., Dazai, J., . . . Lerch, J. P. (2015). Whole-brain mapping of behaviourally induced neural activation in mice. Brain Structure and Function, 220(4), 2043-2057.

Wade, T. R., de Wit, H., & Richards, J. B. (2000). Effects of dopaminergic drugs on delayed reward as a measure of impulsive behavior in rats. Psychopharmacology, 150(1), 90-101.

Wake, H., Moorhouse, A. J., Jinno, S., Kohsaka, S., & Nabekura, J. (2009). Resting microglia directly monitor the functional state of synapses in vivo and determine the fate of ischemic terminals. Journal of Neuroscience, 29(13), 3974-3980.

Walker, F. R., Beynon, S. B., Jones, K. A., Zhao, Z., Kongsui, R., Cairns, M., & Nilsson, M. (2014). Dynamic structural remodelling of microglia in health and disease: A review of the models, the signals and the mechanisms. Brain, behavior, and immunity, 37, 1-14. doi:https://doi.org/10.1016/j.bbi.2013.12.010

Wallace, D. L., Vialou, V., Rios, L., Carle-Florence, T. L., Chakravarty, S., Kumar, A., . . . Iniguez, S. D. (2008). The influence of ΔFosB in the nucleus accumbens on natural reward-related behavior. Journal of Neuroscience, 28(41), 10272-10277.

Wang, G.-J., Volkow, N. D., Logan, J., Pappas, N. R., Wong, C. T., Zhu, W., . . . Fowler, J. S. (2001). Brain dopamine and obesity. The Lancet, 357(9253), 354-357.

Wang, G.-J., Yang, J., Volkow, N. D., Telang, F., Ma, Y., Zhu, W., . . . Fowler, J. S. (2006). Gastric stimulation in obese subjects activates the hippocampus and

422

other regions involved in brain reward circuitry. Proceedings of the National Academy of Sciences, 103(42), 15641-15645.

Watkins, L. R., Milligan, E. D., & Maier, S. F. (2001). Glial activation: a driving force for pathological pain. Trends in neurosciences, 24(8), 450-455.

Watson, J. B., & Rayner, R. (1920). Conditioned emotional reactions. Journal of experimental psychology, 3(1), 1.

Weeks, J. R. (1962). Experimental morphine addiction: method for automatic intravenous injections in unrestrained rats. science, 138(3537), 143-144.

Wehry, A. M., Beesdo-Baum, K., Hennelly, M. M., Connolly, S. D., & Strawn, J. R. (2015). Assessment and treatment of anxiety disorders in children and adolescents. Current psychiatry reports, 17(7), 52.

Weihrauch, M., & Diehl, V. (2004). Artificial sweeteners—do they bear a carcinogenic risk? Annals of Oncology, 15(10), 1460-1465.

Weiss, D. S., Marmar, C. R., Metzler, T. J., & Ronfeldt, H. M. (1995). Predicting symptomatic distress in emergency services personnel. Journal of consulting and clinical psychology, 63(3), 361.

Weissman, M. M., Wolk, S., Wickramaratne, P., Goldstein, R. B., Adams, P., Greenwald, S., . . . Steinberg, D. (1999). Children with prepubertal-onset major depressive disorder and anxiety grown up. Archives of general psychiatry, 56(9), 794-801.

Weisstaub, N. V., Zhou, M., Lira, A., Lambe, E., Gonzalez-Maeso, J., Hornung, J. P., . . . Gingrich, J. A. (2006). Cortical 5-HT2A receptor signaling modulates anxiety-like behaviors in mice. science, 313(5786), 536-540. doi:10.1126/science.1123432

Welberg, L. A., & Seckl, J. R. (2001). Prenatal stress, glucocorticoids and the programming of the brain. Journal of neuroendocrinology, 13(2), 113-128.

Westover, A. N., & Marangell, L. B. (2002). A cross‐national relationship between sugar consumption and major depression? Depression and anxiety, 16(3), 118-120.

Westwater, M. L., Fletcher, P. C., & Ziauddeen, H. (2016). Sugar addiction: the state of the science. European Journal of Nutrition, 55(2), 55-69.

Wilensky, A. E., Schafe, G. E., Kristensen, M. P., & LeDoux, J. E. (2006). Rethinking the fear circuit: the central nucleus of the amygdala is required for the acquisition, consolidation, and expression of Pavlovian fear conditioning. Journal of Neuroscience, 26(48), 12387-12396.

Williams-Spooner, M., Richardson, R., & Baker, K. (2017). 835. A High-Fat High-Sugar Diet-Induced Impairment of Fear Inhibition and Place-Recognition Memory in Adolescent Rats. Biological psychiatry, 81(10), S338-S339.

Williamson, L. L., Sholar, P. W., Mistry, R. S., Smith, S. H., & Bilbo, S. D. (2011). Microglia and memory: modulation by early-life infection. Journal of Neuroscience, 31(43), 15511-15521.

Willins, D. L., Deutch, A. Y., & Roth, B. L. (1997). Serotonin 5-HT2A receptors are expressed on pyramidal cells and interneurons in the rat cortex. Synapse, 27(1), 79-82. doi:10.1002/(sici)1098-2396(199709)27:1<79::aid-syn8>3.0.co;2-a

Wilmouth, C. E., & Spear, L. P. (2009). Hedonic sensitivity in adolescent and adult rats: taste reactivity and voluntary sucrose consumption. Pharmacology Biochemistry and Behavior, 92(4), 566-573.

Wilson, G. T. (2010). Eating disorders, obesity and addiction. European Eating Disorders Review, 18(5), 341-351.

423

Wilson, Y. M., & Murphy, M. (2009). A discrete population of neurons in the lateral amygdala is specifically activated by contextual fear conditioning. Learning & Memory, 16(6), 357-361. doi:10.1101/lm.1361509

Wiltgen, B. J., Brown, R. A., Talton, L. E., & Silva, A. J. (2004). New circuits for old memories: the role of the neocortex in consolidation. Neuron, 44(1), 101-108.

Wise, R. A. (1996). Addictive drugs and brain stimulation reward. Annual review of neuroscience, 19(1), 319-340.

Wise, R. A. (2004). Dopamine, learning and motivation. Nature reviews neuroscience, 5(6), 483.

Wise, R. A. (2006). Role of brain dopamine in food reward and reinforcement. Philosophical Transactions of the Royal Society of London B: Biological Sciences, 361(1471), 1149-1158.

Wise, R. A., Leeb, K., Pocock, D., Newton, P., Burnette, B., & Justice, J. (1995). Fluctuations in nucleus accumbens dopamine concentration during intravenous cocaine self-administration in rats. Psychopharmacology, 120(1), 10-20.

Wohleb, E. S., Fenn, A. M., Pacenta, A. M., Powell, N. D., Sheridan, J. F., & Godbout, J. P. (2012). Peripheral innate immune challenge exaggerated microglia activation, increased the number of inflammatory CNS macrophages, and prolonged social withdrawal in socially defeated mice. Psychoneuroendocrinology, 37(9), 1491-1505.

Wojtalik, J. A., Eack, S. M., Smith, M. J., & Keshavan, M. S. (2018). Using cognitive neuroscience to improve mental health treatment: A comprehensive review. Journal of the Society for Social Work and Research, 9(2), 223-260.

Wolf, A., Bray, G., & Popkin, B. (2008). A short history of beverages and how our body treats them. obesity reviews, 9(2), 151-164.

Wong-Lin, K., Wang, D. H., Moustafa, A. A., Cohen, J. Y., & Nakamura, K. (2017). Toward a multiscale modeling framework for understanding serotonergic function. J Psychopharmacol, 31(9), 1121-1136. doi:10.1177/0269881117699612

Woods, J. S., & Leibowitz, S. F. (1985). Hypothalamic sites sensitive to morphine and naloxone: effects on feeding behavior. Pharmacology Biochemistry and Behavior, 23(3), 431-438.

Wu, J. (2009). Understanding of nicotinic acetylcholine receptors. Acta Pharmacologica Sinica, 30(6), 653.

Wyvell, C. L., & Berridge, K. C. (2001). Incentive sensitization by previous amphetamine exposure: increased cue-triggered “wanting” for sucrose reward. Journal of Neuroscience, 21(19), 7831-7840.

Xiong, L., Meng, Q., Sun, X., Lu, X., Fu, Q., Peng, Q., . . . Hu, Z. Z. (2018). CART peptide in the nucleus accumbens shell inhibits cocaine‐induced locomotor sensitization to transient overexpression of α‐Ca2+/Calmodulin‐dependent Protein Kinase II. Journal of neurochemistry.

Xu, C., Krabbe, S., Gründemann, J., Botta, P., Fadok, J. P., Osakada, F., . . . Lüthi, A. (2016). Distinct Hippocampal Pathways Mediate Dissociable Roles of Context in Memory Retrieval. Cell, 167(4), 961-972.e916. doi:https://doi.org/10.1016/j.cell.2016.09.051

Xu, T. J., & Reichelt, A. C. (2018). Sucrose or sucrose and caffeine differentially impact memory and anxiety-like behaviours, and alter hippocampal parvalbumin and doublecortin. Neuropharmacology, 137, 24-32.

424

Yamanaka, A., Kunii, K., Nambu, T., Tsujino, N., Sakai, A., Matsuzaki, I., . . . Sakurai, T. (2000). Orexin-induced food intake involves neuropeptide Y pathway. Brain research, 859(2), 404-409.

Yang, H., de Jong, J. W., Tak, Y., Peck, J., Bateup, H. S., & Lammel, S. (2018). Nucleus accumbens subnuclei regulate motivated behavior via direct inhibition and disinhibition of VTA dopamine subpopulations. Neuron.

Yehuda, R. (2001). Biology of posttraumatic stress disorder. The Journal of clinical psychiatry.

Yehuda, R. (2002). Post-traumatic stress disorder. New England journal of medicine, 346(2), 108-114.

Yeomans, M. R. (2017). Adverse effects of consuming high fat–sugar diets on cognition: implications for understanding obesity. Proceedings of the Nutrition Society, 76(4), 455-465.

Yokoyama, M., Suzuki, E., Sato, T., Maruta, S., Watanabe, S., & Miyaoka, H. (2005). Amygdalic levels of dopamine and serotonin rise upon exposure to conditioned fear stress without elevation of glutamate. Neuroscience letters, 379(1), 37-41.

Yoshida, H., Kanamaru, C., Ohtani, A., Li, F., Senzaki, K., & Shiga, T. (2011). Subtype specific roles of serotonin receptors in the spine formation of cortical neurons in vitro. Neurosci Res, 71(3), 311-314. doi:10.1016/j.neures.2011.07.1824

Yoshioka, M., Matsumoto, M., Togashi, H., & Saito, H. (1995). Effects of conditioned fear stress on 5-HT release in the rat prefrontal cortex. Pharmacology Biochemistry and Behavior, 51(2-3), 515-519.

Yu, J. H., & Kim, M.-S. (2012). Molecular mechanisms of appetite regulation. Diabetes & metabolism journal, 36(6), 391-398.

Zehra, A., Burns, J., Liu, C. K., Manza, P., Wiers, C. E., Volkow, N. D., & Wang, G.-J. (2018). Cannabis addiction and the brain: A review. Journal of Neuroimmune Pharmacology, 1-15.

Zelikowsky, M., Hersman, S., Chawla, M. K., Barnes, C. A., & Fanselow, M. S. (2014). Neuronal Ensembles in Amygdala, Hippocampus, and Prefrontal Cortex Track Differential Components of Contextual Fear. The Journal of Neuroscience, 34(25), 8462-8466. doi:10.1523/jneurosci.3624-13.2014

Zhang, C., Liu, X., Zhou, P., Zhang, J., He, W., & Yuan, T.-F. (2018). Cholinergic tone in ventral tegmental area: Functional organization and behavioral implications. Neurochemistry international.

Zhang, G., Ásgeirsdóttir, H. N., Cohen, S. J., Munchow, A. H., Barrera, M. P., & Stackman Jr, R. W. (2013). Stimulation of serotonin 2A receptors facilitates consolidation and extinction of fear memory in C57BL/6J mice. Neuropharmacology, 64, 403-413.

Zhang, S., Zhang, H., Ku, S. M., Juarez, B., Morel, C., Tzavaras, N., . . . Russo, S. J. (2018). Sex Differences in the Neuroadaptations of Reward-related Circuits in Response to Subchronic Variable Stress. Neuroscience, 376, 108-116.

Zheng, J., Luo, F., Guo, N.-n., Cheng, Z.-y., & Li, B.-m. (2015). β1-and β2-adrenoceptors in hippocampal CA3 region are required for long-term memory consolidation in rats. Brain research, 1627, 109-118. doi:http://dx.doi.org/10.1016/j.brainres.2015.08.035

Ziauddeen, H., Farooqi, I. S., & Fletcher, P. C. (2012). Obesity and the brain: how convincing is the addiction model? Nature reviews neuroscience, 13(4), 279.

425

Zimmerman, D. W., & Zumbo, B. D. (1993). Rank transformations and the power of the Student t test and Welch t'test for non-normal populations with unequal variances. Canadian Journal of Experimental Psychology/Revue canadienne de psychologie expérimentale, 47(3), 523.

Zuur, A., Ieno, E. N., & Smith, G. M. (2007). Analyzing ecological data: Springer Science & Business Media.

Zuur, A. F., & Ieno, E. N. (2016). A protocol for conducting and presenting results of regression‐type analyses. Methods in Ecology and Evolution, 7(6), 636-645.

Zuur, A. F., Ieno, E. N., Walker, N. J., Saveliev, A. A., & Smith, G. M. (2009). Zero-truncated and zero-inflated models for count data Mixed effects models and extensions in ecology with R (pp. 261-293): Springer.

INVESTIGATING THE NEUROPLASTICITY OF EMOTIONAL MEMORIES · INVESTIGATING THE NEUROPLASTICITY OF...

Documents

Transcript of INVESTIGATING THE NEUROPLASTICITY OF EMOTIONAL MEMORIES · INVESTIGATING THE NEUROPLASTICITY OF...