Methodology for the measurement of HIF-1α in bovine leukocytes and assessment of the molecule as a biomarker for bovine respiratory disease outcome

Sarah Gestier BVSc

A thesis submitted for the degree of Master of Philosophy at

The University of Queensland in 2014

School of Veterinary Science

ii

Abstract

Bovine respiratory disease (BRD) is a multi-factorial inflammatory respiratory disease complex and

is a significant problem for the beef industry. While the infectious agents involved are known and

vaccines exist for some, the host factors determining outcome of infections are still only partially

understood. The aim of this project was to explore the usefulness of an inflammation regulatory

molecule as a biomarker for susceptibility/resistance to the development of BRD in beef cattle.

The transcription factor HIF-1α is recognised for its importance in the development and

coordination of an immune response to hypoxia and inflammation caused by infectious agents. For

this reason, HIF-1α was identified as an attractive biomarker candidate for BRD susceptibility.

The initial objective of this study was to develop methods by which the level of HIF-1α stabilisation

in response to hypoxia could be measured in key immune cells. Once developed, the second

objective of the study was to use these methodologies to determine whether there was significant

positive or negative correlation between the regulation of HIF-1α and the development of clinical

BRD. The experimental hypothesis was that there is a significant difference in HIF-1α regulation

between animals with clinical BRD compared with other clinically healthy animals, and thus in

response to hypoxia there would be a difference in the lymphocyte and monocyte HIF-1α

expression between these two groups of cattle.

This project developed a methodology for hypoxic culture of bovine lymphocytes and monocytes to

induce HIF-1α stabilisation. Tube cultures for lymphocyte analysis were cultured with 1.25 x106

cells in 250 µl serum-free medium suspension treated with 200 µM cobalt chloride and 5 µg/ml

LPS; with a non-stimulated control tube containing only the cells suspended in serum-free medium.

Chamber slides for monocyte analysis were cultured containing 2.5 x 106 cells in 500µl serum-free

medium with or without stimulants. Tubes and chamber slides were cultured for 18 hours in 5%

CO2 at 37°C. Tube cultured cells underwent post-culture fluorescent staining using a fluorochrome-

conjugated HIF-1α antibody for measurement of HIF-1α fluorescence in B lymphocytes and T

lymphocytes using flow cytometry. T lymphocytes were labelled using a fluorochrome-conjugated

anti-CD3 antibody and B lymphocytes labelled using a fluorochrome-conjugated anti-CD20

antibody. An additional protocol was developed to measure monocyte HIF-1α expression using

immunocytochemical staining of chamber slide cultures. This alternative approach was necessitated

by the finding of poor monocyte survival during hypoxic culture and the the pre-flow cytometry

staining procedures. These methodologies for the stabilisation and measurement of HIF-1α in

bovine lymphocytes and monocytes were then applied to a sample population of 88 feedlot cattle,

34 of which had been treated in the feedlot for clinical BRD (cases), while the remainder was

untreated, clinically healthy controls. The groups were compared using two-sample t-tests for both

iii

flow cytometry and immunocytochemistry. No statistical significance difference was detected in the

HIF-1α expression of leukocytes when cases were compared to control animals. However, due to

time constraints only a relatively small number of animals were tested and the possibility remains

that with a larger cohort the HIF-1α response may still prove useful alone or in combination with

other biomarkers in the setting of BRD. Further work is needed to improve these methodologies, or

develop alternative methods. If this can be done successfully a higher powered study with a larger

cohort of animals could be undertaken. However the extent of the within-group variation of cases

and controls suggests that any difference between the groups would need to be very substantial for

such a difference to be statistically significant, even with a much larger sample size.

Based on this study, the expression of HIF-1α and its value as a biomarker for the development of

clinical BRD as a disease outcome could still be significant and remains to be established. The

expression of HIF-1α could still be shown as significant in the context of the bovine acute immune

responses to BRD viral pathogens if further investigations incorporated relevant and appropriate

viral challenges. The lack of a detected difference in HIF-1α could also mean that HIF-1α does not

have a significant role in the pathogenesis of clinical BRD, and further studies involving viral

challenge could also help to establish or refute this possibility.

iv

Declaration by author

(All candidates to reproduce this section in their thesis verbatim)

This thesis is composed of my original work, and contains no material previously published or

written by another person except where due reference has been made in the text. I have clearly

stated the contribution by others to jointly-authored works that I have included in my thesis.

I have clearly stated the contribution of others to my thesis as a whole, including statistical

assistance, survey design, data analysis, significant technical procedures, professional editorial

advice, and any other original research work used or reported in my thesis. The content of my thesis

is the result of work I have carried out since the commencement of my research higher degree

candidature and does not include a substantial part of work that has been submitted to qualify for

the award of any other degree or diploma in any university or other tertiary institution. I have

clearly stated which parts of my thesis, if any, have been submitted to qualify for another award.

I acknowledge that an electronic copy of my thesis must be lodged with the University Library and,

subject to the General Award Rules of The University of Queensland, immediately made available

for research and study in accordance with the Copyright Act 1968.

I acknowledge that copyright of all material contained in my thesis resides with the copyright

holder(s) of that material. Where appropriate I have obtained copyright permission from the

copyright holder to reproduce material in this thesis.

v

Publications during candidature

No publications.

Publications included in this thesis

No publications included.

vi

Contributions by others to the thesis

Dr Helle Bielefeldt-Ohmann contributed to the conception and design of the project, performed

reading of ICC slides as well as critical revision of written work. Dr Tamsin Barnes contributed to

the analysis and interpretation of experimental data and revision of written work. Dr Helen Owens

contributed to the revision of written work. Dr Nana Satake contributed flow cytometry technical

and interpretative expertise.

Statement of parts of the thesis submitted to qualify for the award of another degree

None.

vii

Acknowledgements

Acknowledgement and gratitude to Dr Helle Bielefeldt-Ohmann, Dr Tamsin Barnes and Dr Helen

Owens for their supportive roles as supervisors, to the Meat and Livestock Association for financial

support for a scholarship and research funds and to the technical staff at the School of Veterinary

Science, University of Queensland, Gatton.

viii

Keywords

hypoxia-inducible factor-1α, biomarker, bovine respiratory disease

Australian and New Zealand Standard Research Classifications (ANZSRC)

ANZSRC code: 070705, Veterinary Immunology, 100%

Fields of Research (FoR) Classification

FoR code: 0707, Veterinary Sciences, 100%

ix

Table of Contents

Abstract ii  Declaration by author iv  Publications during candidature v  Publications included in this thesis v  Contributions by others to the thesis vi  Statement of parts of the thesis submitted to qualify for the award of another degree vi  Acknowledgements vii  Keywords viii  Australian and New Zealand Standard Research Classifications (ANZSRC) viii  Fields of Research (FoR) Classification viii  Table of Contents ix  List of Figures xii  List of Tables xiv  Abbreviations xv  1   Introduction, Literature Review, Conclusions and Research Proposal 1  1.1   Introduction 1  1.2   Literature Review 1  1.2.1   HIF-1α regulation during normoxia, development and homeostasis 3  1.2.2   HIF-1α during hypoxic conditions 5  1.2.3   The importance of HIF-1α and HIF-1 in mammalian immunity 6  1.2.4   HIF-1α and BRD 12  1.3   Conclusions and Research Proposal 16  1.3.1   Experimental hypothesis 16  1.3.2   Null Hypothesis 16  1.3.3   Research Objectives 16  2   Thesis overview 18  3   Optimization of hypoxic cell culture and harvest 19  3.1   The production of HIF-1α in hypoxic conditions in vivo 19  3.1.1   Hypoxic cell culture 19  3.1.2   Bovine leukocytes chosen for hypoxic culture 19  3.2   Materials, Methods and Results 20  3.2.1   Bovine leukocytes used in development of hypoxic-mimetic cell culture methods 21  3.2.2   Procedure for mononuclear cell isolation 21  3.2.3   Preparation of frozen test samples for culture 21  3.2.4   Hypoxic mimetic culture, stimulant and culture medium 22  

x

3.2.5   Culture duration testing 24  3.2.6   Culture vessel testing, post-culture viability and HIF-1α degradation 24  3.3   Discussion 29  3.4   Conclusion – final cell culture and post-culture harvest protocol 33  

4   Optimization of flow cytometry protocol for measuring HIF 1α expression in lymphocyte subtypes 34  

4.1   Introduction 34  4.2   Materials, methods and results 34  4.2.1   Flow cytometer maintenance and flow cytometry controls 34  4.2.2   Cellular staining method for flow cytometry 35  4.2.3   Antibody titration 36  4.2.4   T lymphocyte CD3 antibody selection and titration 38  4.2.5   B lymphocyte CD20 antibody selection and titration 41  4.2.6   Monocyte CD14 antibody selection and titration 41  4.2.7   HIF-1α antibody selection 41  4.2.8   Triple staining of cultured cells for flow cytometry 45  4.3   Discussion 48  4.4   Conclusion - Final post-culture staining and flow protocol for 3-colour flow

cytometry of T lymphocytes, B lymphocytes and HIF-1α 49  

5   Development of a protocol to detect HIF-1α expression in monocytes 50  5.1   Introduction 50  5.2   Materials, methods and results 51  5.2.1   HIF-1α antibody selection for ICC 51  5.2.2   ICC protocol for HIF-1α expression in monocytes 51  5.3   Discussion 53  5.4   Conclusion 54  

6   Methodology application to assess HIF-1α expression as a biomarker of bovine respiratory disease 55  

6.1   Introduction 55  6.2   Materials and methods 56  6.2.1   Study population and animal selection 56  6.2.2   Baseline data 57  6.2.3   Blood sample collection 57  6.2.4   Mononuclear cell isolation from experimental samples 58  6.2.5   Cell culture and post-culture harvest of experimental samples 58  6.2.6   Flow cytometry protocol 58  6.2.7   ICC for monocyte expression of HIF-1α 60  6.3   Results 61  6.3.1   Mononuclear cell isolation 61  

xi

6.3.2   Comparison of baseline data 61  6.3.3   Flow cytometry results 64  6.3.4   ICC results 65  6.4   Discussion 69  6.5   Conclusion 71  7   General discussion and future directions 72  7.1   The importance of finding a biomarker for BRD 72  7.2   Methodology development issues 72  7.2.1   Hypoxic cell culture and harvest methods 72  7.2.2   Flow cytometry methods 73  7.2.3   ICC methods 75  7.3   Future directions for investigation of biomarkers for BRD 75  8   References 77  9   Appendices 90  9.1   Initial T lymphocyte CD3 antibody 90  9.2   CD20 antibody titration on flow cytometry 90  9.3   Initial CD14 antibody 91  

xii

List of Figures

Figure 1.1: Diagram of (a) degradation of HIF-1α during normoxia and (b) HIF -1α stabilization during hypoxia (right) (From Nizet 2009) .................................................................................................................... 4 Figure 2.1: Thesis overview showing the interconnected relationships between various aspects of the methodologies developed in chapters 3, 4 and 5. Flow cytometry results influenced the cell culture methodology and lead to the development of ICC. All developed methodology is applied to experimental samples in chapter 6, followed by a discussion of developed methodology and future directions in chapter 7. ......................................................................................................................................................................... 18 Figure 3.1: A: SS/Pacific Blue dot plot obtained from Kaluza of all events within the Cells gate. Colour gradients denote areas of lower (blue) to higher event density (red), percentages are of all events within the Cells gate. An uncultured aliquot of cells stained with pacific blue-labelled CD14 antibody shows a distinct CD14 positive population to the right of a larger group of negative cells. B: An aliquot of cells cultured without stimulants shows a similar but depleted CD14 positive population, and generalised increased SS of all cells. C: An aliquot of cells cultured with CoCl2 and LPS shows generalised increased SS of all cells but the CD14 positive population cannot be distinctly visualised. ........................................................................ 27 Figure 3.2: A: ICC of uncultured mononuclear cells with anti-activated caspase-3. There is moderate staining intensity within both cells with abundant and scant cytoplasm (putative monocytes and lymphocytes, respectively). B: ICC of mononuclear cells cultured with stimulants before undergoing staining procedures for flow cytometry. There is marked staining intensity of cells with abundant cytoplasm (putative monocytes). ...................................................................................................................................................... 29 Figure 4.1: A bivariate scatter plot of the forward scatter (FS) and side scatter (SS) of the uncultured unstained test cells as visualised on a linear scale. Debris usually has low FS and SS. Lymphocytes have moderate FS but relatively low amounts of SS in contrast to monocytes. Events with high FS and SS are postulated to be aggregates of cells. ................................................................................................................ 36 Figure 4.2: A: Peripheral blood mononuclear cells - histogram of CD3 fluorescence of events with the cells gate. A peak of cells positive for CD3 can be visualised to the right of the larger negative peak. The negative linear boundary (black line) was set using unstained cells. The positive linear boundary (black line) was set using the CD3 positive gate described in 4.2B. B: Scatter plot of CD3 fluoresence versus SS, log scale. A population positive for CD3 has been gated by eye (black border). This positive population is distinct from and far to the right of the negative population. ................................................................................................ 39 Figure 4.3 A: Peripheral blood mononuclear cells - histogram of CD20 fluorescence of events with the cells gate, log scale. A peak of cells positive for CD20 can be visualised however is not separate from the negative peak. The negative linear boundary (black line) was set using unstained cells. The positive linear boundary (black line) was set using the CD3 positive gate described in 4.2B. B: Scatter plot of CD20 fluorescence versus SS, log scale. A population of cells positive for CD20 is visualised and gated by eye (black border). 41 Figure 4.4: Graph of the mean fluorescence of HIF-1α antibody at 1µg/ml with different incubation lengths, unstimulated and stimulated cells. There is a small increase in mean fluorescence in unstimulated cells, while stimulated cell mean fluorescence largely remains the same. ......................................................................... 43 Figure 4.5: Graph of the mean fluorescence of HIF-1α antibody at 2µg/ml with different incubation lengths, unstimulated and stimulated cells. There is a small increase in mean fluorescence in unstimulated cells, while stimulated cell mean fluorescence largely remains the same. ......................................................................... 44 Figure 4.6: A. Scatter plot of HIF-1α fluorescence versus SS, log scale. Unstimulated cells were stained with 2µg/ml HIF-1α antibody for three hours. A single population with one central red (i.e. high density) focus is visible. B. Scatter plot of HIF-1α fluorescence versus SS, log scale. Stimulated cells were stained with 2µg/ml HIF-1α antibody for three hours. The mean HIF-1α fluorescence was consistently lower in stimulated cells however stimulation caused a repeatable change in population shape with respect to HIF-1α and SS. A larger population with greater SS is now visible (upper large red foci) situated above a smaller population with lower SS (lower red foci). ...................................................................................................... 45 Figure 4.7: Scatter plot of CD3 fluorescence and CD20 fluorescence, triple-stained stimulated cells, log scale. B and T cell populations are gated (solid black borders) with percentages of gated cells that are CD3 positive, CD20 positive or positive for both phenotypic markers. .................................................................. 47

xiii

Figure 5.1: Post-culture in situ ICC of adherent monocytes. Secondary controls Wells 1 and 2 contained unstimulated and stimulated cells, respectively, and did not receive primary antibody. The lower wells 3 and 4 contained duplicate cell populations to 1 and 2 and were treated with primary HIF-1α antibody. .............. 52 Figure 6.1 A: An example of the forward scatter/side scatter plot obtained from Kaluza. Events included in cell population lie within the gate “Cells” comprising a solid black line. This gate excludes debris and aggregate cells. B: An example of a T lymphocyte fluorescent marker/B lymphocyte fluorescent marker scatter plot obtained from Kaluza, based on the events included in the “Cells” gate. “T cells” and “B cells” populations are both gated (solid black lines). C: An example of a histogram of HIF-1α fluorescence detected within the T cell gate on flow cytometry with Kaluza-generated summary statistics. .................................... 59 Figure 6.2: ICC of stimulated cells showing several monocytes with scattered intracytoplasmic HIF-1α expression (arrow) and a central, activated monocyte with strong intracytoplasmic and intranuclear HIF-1α signal. ............................................................................................................................................................... 66 Figure 6.3. A: Percentage difference in ICC HIF-1αcytoplasmic signal between stimulated and unstimulated cells, comparison of case group versus control group, subset one. B: Percentage difference in ICC HIF-αcytoplasmic/nuclear “nuclear positive” signal between stimulated and unstimulated cells, comparison of case group versus control group, subset one. C: Percentage difference in ICC HIF-1αcytoplasmic signal between stimulated and unstimulated cells, comparison of case group versus control group, subset two. D: Percentage difference in ICC HIF-1αcytoplasmic/nuclear “nuclear positive” signal between stimulated and unstimulated cells, comparison of case group versus control group, subset two. ........................................... 68

xiv

List of Tables

Table 4.1: Detectors used for detection of cell characteristics and the cytometer parameters for each detector output as set using single and then triple-stained cells. ................................................................................... 46 Table 6.1: Baseline data for cases and controls for both subsets ..................................................................... 62 Table 6.2: Descriptive statistics for the absolute difference in mean, median and standard deviation of HIF-1α in B cells and T cells between unstimulated and stimulated cultures for case group and control groups. . 65 Table 6.3: ICC results for case and control groups from both subsets. The mean, median and standard deviation of the percentage of cells displaying cytoplasmic or combined cytoplasmic and nuclear HIF-1α positive ICC staining in stimulated and unstimulated cell cultures, as well as for the percentage difference of these two types of HIF-1α positive staining between stimulated and unstimulated cultures. ......................... 66

xv

Abbreviations

APC allophycocyanin

ARD1 arrest defective-1 acetyltransferase

ATP adenosine triphosphate

BHV-1 bovine herpesvirus-1

BRD bovine respiratory disease

BRSV bovine respiratory syncytial virus

BVDV bovine viral diarrhoea virus

CO2 carbon dioxide

CoCl2 cobalt chloride

DFO desferrioxamine

ELISA enzyme-linked immunosorbent assay

EPO erythropoietin

FIH factor-inhibiting HIF

FITC fluorescein isothiocyanate

FS forward scatter

HHV-8 human herpesvirus-8

HIF-1α human immunodeficiency virus-1

HIV-1 hypoxia-inducible factor-1α

HRE hypoxic response element

ICC immunocytochemistry

IFN interferon

IHC immunohistochemistry

IL interleukin

iNOS inducible nitric oxide synthase

LANA latency associated nuclear antigen

LMD list mode data file

LPS lipopolysaccharide

mRNA messenger ribonucleic acid

NaN3 sodium azide

NF-κB nuclear factor [kappa] B

NO nitric oxide

ODDD oxygen-dependent degradation domain

PBS phosphate-buffered saline

xvi

PE phycoerytherin

PHD prolyl hydroxylase domain-containing protein

PI-3 parainfluenza-3 virus

pVHL von Hippel-Lindau protein

ROS reactive oxygen species

RSV respiratory syncytial virus

SS side scatter

TLR toll-like receptor

TNF-α tumour necrosis factor-α

UQ The University of Queensland

VEGF vascular endothelial growth factor

VSV vesicular stomatitis virus

1

1 Introduction, Literature Review, Conclusions and Research Proposal

1.1 Introduction

Previously an experimental model of bovine respiratory disease (BRD) has shown that animals are

most susceptible to secondary bacterial infections at the height of the virus-induced inflammatory

response and will develop severe, even fatal, fibrinous bronchopneumonia (Bielefeldt-Ohmann et

al., 2008). The significance of hypoxia-inducible factor-1α (HIF-1α) in the development and

coordination of an immune response to hypoxic microenvironments caused by infectious agents

makes HIF-1α an attractive biomarker for BRD susceptibility. There is evidence in the current

literature to suggest that HIF-1α is an important regulator of innate immunity, that it is induced

during immune stress responses independent of hypoxia, and that both too vigorous and too little of

a HIF-1α response could be detrimental in the probability of overcoming a sophisticated infectious

challenge such as that inherent in the BRD complex.

To investigate the possibility of the cytokine/reactive oxygen species (ROS)- induced positive

feedback loop of HIF-1α induction contributing to the development of clinical BRD, this project

aimed to develop methods for the stabilization and measurement of HIF-1α in bovine immune cells.

Once developed, a further aim was to apply these methods experimentally to quantify leukocyte

HIF-1α responsiveness of a group of feedlot cattle with a history of clinical BRD, and compare with

leukocyte HIF-1α responsiveness of feedlot cattle without clinical respiratory disease.

The discovery of a relationship between HIF-1α expression and the development of clinical BRD

may help to better understand why certain animals develop clinical BRD while other animals do

not, and may establish HIF-1α as a potential biomarker to be used in the development of breeding

programs aimed at reducing BRD susceptibility in feedlot cattle.

1.2 Literature Review

Oxygen is crucial in the survival of most eukaryotic cells due to the oxygen requirements of

important cellular processes, most notably the oxidation of nutrients to produce adenosine

triphosphate (ATP) by oxidative phosphorylation (Lipmann, 1941, Kalckar, 1941). Tissue hypoxia

is defined as low oxygen within a tissue, organ or cell compared to the oxygen level normally

demanded within that tissue at sea level ([Anon], 1973). The rapid cellular and systemic response

mechanisms to adapt to hypoxic conditions are coordinated by Hypoxia-inducible factor-1 (HIF-1),

a heterodimeric helix-loop-helix transcription factor that is considered an important regulator of

cellular homeostasis, cellular adaptation to stress caused by hypoxia, and innate immunity (Nizet

and Johnson, 2009).

Hypoxia-inducible factor-1 was first identified when the molecule was found to be associated with

marked induction of erythropoietin (EPO) gene transcription that occurs under hypoxic conditions

2

(Goldberg et al., 1988, Semenza et al., 1991, Semenza and Wang, 1992). Subsequent study of HIF-

1 recognized its existence in all mammalian cells, not just those involved in EPO production (Wang

and Semenza, 1993c, Maxwell et al., 1993).

HIF-1 is composed of 2 subunit proteins, a hypoxia-inducible factor-1α (HIF-1α) subunit (Wang

and Semenza, 1993a, Wang and Semenza, 1995) which is oxygen-sensitive (Wang et al., 1995,

Jiang et al., 1996) and a constitutively expressed hypoxia-inducible factor-1β (HIF-ß) subunit

(Wang and Semenza, 1995).

When stabilised by hypoxia or other factors, HIF-1α translocates into the nucleus where it binds to

HIF-1ß (Wang et al., 1995). The resultant heterodimer HIF-1 binds to 50-base pair cis-acting

sequence termed a hypoxic response element (HRE) of target gene enhancer and promotor

sequences, and activates the expression of these target genes (Semenza et al., 1991, Semenza and

Wang, 1992, Wang et al., 1995). HIF-1 is known to regulate the expression of more than 100 target

genes (Nizet and Johnson, 2009) involved in systemic and cellular adaptive responses to hypoxia.

HIF-1 regulates genes important for erythropoiesis and iron metabolism (Semenza et al., 1991),

angiogenesis (Hertig, 1935), cell proliferation and survival, as well as responses supporting innate

immunity within the characteristic hypoxic microenvironment of inflamed tissues (Nizet and

Johnson, 2009). Many pro-inflammatory cytokines and ROS can stabilize HIF-1α even under

normoxic conditions, with subsequent HIF-1 induction of proteins that promote inflammation in a

seemingly positive feedback loop of inflammation regulation.

More recently, two other closely related forms of the HIF-α subunit have been identified and

characterized as HIF-2α (Tian et al., 1997) and HIF-3α (Gu et al., 1998). Both HIF-2α and HIF-3α

can also complex with HIF-1ß (Tian et al., 1997, Gu et al., 1998). HIF-1α and HIF-2α have a

similar domain structure, are inducible by hypoxia and bind HRE of target genes (Tian et al., 1997,

Wiesener et al., 1998, O'Rourke et al., 1999). HIF-2α expression has been found predominantly in

vascular structures, lung, endothelium and carotid body (Ema et al., 1997, Tian et al., 1997, Tian et

al., 1998), in contrast to the ubiquitous HIF-1α. HIF-3α is expressed in a variety of tissues (Gu et

al., 1998) but lacks the C-terminal activation domain required for co-activator binding, so cannot

recruit co-transcriptional regulators and transcriptional machinery to gene targets (Gu et al., 1998).

There is evidence to suggest that HIF-3α splice variants function as inhibitors of HIF-1 induced

gene expression, possibly to protect against hypoxic damage (Makino et al., 2001, Forristal et al.,

2010, Augstein et al., 2011).

3

1.2.1 HIF-1α regulation during normoxia, development and homeostasis

An appreciation for the metabolism of HIF-1α during normoxia is necessary to better understand

the currently known mechanisms of HIF-1α in the hypoxic microenvironment.

1.2.1.1 HIF-1α during normoxic conditions

Normoxia is currently defined for tissue culture as being the equivalent of atmospheric oxygen

pressure at sea level or as physiological oxygenation for well-vascularised and perfused tissue

(Nizet and Johnson, 2009). Reported oxygen pressure values for healthy tissue vary from 10-30

mm Hg (Intaglietta et al., 1996, Braun et al., 2001). During normoxic conditions HIF-1α is a short-

lived protein with a half-life of less than five minutes (Huang et al., 1996). In the absence of

stabilization, it is removed via ubiquitination followed by proteasomal degradation (Salceda and

Caro, 1997, Huang et al., 1998, Kallio et al., 1999).

The ubiquitination of HIF-1α is performed by an E3 ubiquitin ligase complex that also contains the

von Hippel-Lindau (pVHL) tumor suppressor protein, which acts as the recognition point for HIF-

1α during binding for ubiquitination (Maxwell et al., 1999). To promote VHL interaction with HIF-

1α, there must be hydroxylation of prolyl residues(Ivan et al., 2001, Jaakkola et al., 2001, Yu et al.,

2001) within identified oxygen-dependent degradation domains (ODDDs)(Huang et al., 1998) of

HIF-1α (Huang et al., 1998). Hydroxylation of the prolyl residues is carried out by three prolyl

hydroxylase domain-containing protein (PHD) enzymes, and all PHD enzymes require oxygen and

iron to complete this activity (Epstein et al., 2001, Jaakkola et al., 2001, Bruick and McKnight,

2001). This implicates PHD enzymes as important oxygen sensors within the HIF-1α pathway

(Schofield and Ratcliffe, 2004, Walshe and D'Amore, 2008). HIF-1α is subsequently tagged with

polyubiquitin and undergoes proteosomal proteolysis (Salceda and Caro, 1997, Kallio et al., 1999,

Masson et al., 2001) (see Figure 1.1).

4

Figure 1.1: Diagram of (a) degradation of HIF-1α during normoxia and (b) HIF -1α stabilization

during hypoxia (right) (From Nizet 2009) In addition to PHD enzyme activity, there are further oxygen-sensing mechanisms in the HIF-1α

pathway. HIF-1α contains a carboxy-terminal activation domain (Jiang et al., 1997) in which there

is hydroxylation of an asparagine (Lando et al., 2002b) by an asparaginyl hydroxylase called factor-

inhibiting HIF (FIH) (Lando et al., 2002a) during normoxia. This hydroxylation prevents binding of

the p300/cAMP- response element-binding protein, a co-activator which, during hypoxia, binds to

HIF-1α and enhances its transcriptional activity (Lando et al., 2002b)(see Figure 1.1). FIH also

interacts with VHL to form a ternary complex with HIF-1α in which pHVL inhibits HIF-1α

transactivation function (Mahon et al., 2001). FIH also has an absolute requirement for oxygen

(Koivunen et al., 2004, Ehrismann et al., 2007) and so is considered a second oxygen sensing

mechanism in the inhibitory pathway (Walshe and D'Amore, 2008).

A study by Jeong et al using HEK293 cells showed HIF-1α is also further negatively regulated by

acetylation performed by arrest defective 1acetyltransferase (ARD1) which binds in the ODDD and

enhances pVHL binding to HIF-1α (Jeong et al., 2002). This study concluded ARD1 functions as a

negative regulator of HIF-1α during normoxia, as there were decreased levels of ARD1 mRNA and

5

decreased acetylated HIF-1α during hypoxia (Jeong et al., 2002). However, another study used

several cell types to show that neither the mRNA nor the protein levels of ARD1 are regulated by

short term or long term hypoxia (Bilton et al., 2005), and that overexpression or silencing of the

ARD1 gene had no effect on HIF-1α stability in normoxic or hypoxic conditions (Bilton et al.,

2005). The relationship between normoxia, hypoxia and HIF-1α acetylation by ARD1 remains

unclear.

1.2.2 HIF-1α during hypoxic conditions

Several important pathological processes involve or induce hypoxic conditions in tissue including

inflammation and neoplasia (Najafipour and Ferrell, 1995, Helmlinger et al., 1997). The hypoxia

within acutely inflamed tissue is due to decreased perfusion from microvascular injury and

thrombosis, clogging of vessels with recruited inflammatory cells, and the metabolic activities of

recruited inflammatory cells and any infectious pathogen (Simmen et al., 1994, Saadi et al., 2002,

Sitkovsky and Lukashev, 2005, Kempf et al., 2005, Nizet and Johnson, 2009). Within the context of

the hypoxic inflammatory microenvironment and the immune response, HIF-1α is established as a

molecule of interest to this project about BRD susceptibility.

Under hypoxic conditions HIF-1α ubiquitination and degradation is decreased (Kallio et al., 1999)

due to reduced activity of oxygen-dependent PHD enzymes (Epstein et al., 2001, Jaakkola et al.,

2001). PHD are also actively degraded by the E3 ubiquitin ligase (Nakayama et al., 2004).

HIF-1α accumulates in the cytoplasm and translocates to the nucleus (Kallio et al., 1999) where it

binds to HIF-1β (Wang et al., 1995) and forms activated HIF-1 for transcription (Huang et al.,

1996, Kallio et al., 1997)}. Experiments in vitro and in vivo have also shown increased

transcription of HIF-1α mRNA in response to hypoxia (Wiener et al., 1996, Yu et al., 1998,

Semenza, 2000, Belaiba et al., 2007) and have replaced an original concept of solely

posttranslational hypoxic HIF-1α regulation.

1.2.2.1 Transcription of HIF-1 target genes in response to hypoxia

The HIF-1 transcription factor regulates the expression of more than 100 target genes (Nizet and

Johnson, 2009). In response to hypoxia, HIF-1 activates genes important for erythropoiesis and iron

metabolism, the products of which enhance tissue oxygenation. Gene products include

erythropoietin (EPO) for new erythrocyte formation (Semenza et al., 1991), transferrin that

transports ferric iron into cells (Rolfs et al., 1997) and ceruloplasmin, which oxidises ferrous iron to

ferric iron (Mukhopadhyay et al., 2000).

6

1.2.2.1.1 Angiogenesis

HIF-1 also promotes angiogenesis, defined as the formation of new vessels from existing

vasculature (Hertig, 1935) to increase vascularisation in response to hypoxia. The important

endothelial mitogen vascular endothelial growth factor (VEGF) (Leung et al., 1989, Keck et al.,

1989) is produced following HIF-1 gene transcription regulation (Forsythe et al., 1996). Vascular

endothelial growth factor causes recruitment to and proliferation of vascular endothelial cells within

hypoxic or avascular tissues, including during the development of tumours, in diseases with an

angiogenic pathogenesis and wound healing (Shweiki et al., 1992, Folkman and Shing, 1992,

Brown et al., 1992, Adamis et al., 1994).

There is evidence of inhibition of HIF-1 angiogenesis in specific tissues. The cornea is subject to

continuous hypoxia during sleep but vessel proliferation does not occur due to a splice variant of

the HIF-3α called inhibitory Per/Arnt/Sim protein, which inhibits HIF-1 controlled gene expression

(Makino et al., 2001). Other examples are vascular endothelium and smooth muscle (Augstein et

al., 2011), as well as the Purkinje cell layer of the cerebellum (Makino et al., 2001).

1.2.2.1.2 Cell proliferation and survival

HIF-1 can control cell proliferation and survival. Hypoxia and HIF-1 induce growth factors such as

insulin-like growth factor-2 and transforming growth factor-α that activate signal transduction

pathways leading to cell-specific proliferation and survival, as well as stimulating HIF-1α

expression. These events implicate HIF-1α in an autocrine-signalling pathway that can be crucial

for neoplastic disease (Semenza, 2003). Hypoxia and HIF-1 has also been shown to induce

apoptosis via several mechanisms (Greijer and van der Wall, 2004), as well as inhibit neutrophil

apoptosis (Walmsley et al., 2005).

1.2.3 The importance of HIF-1α and HIF-1 in mammalian immunity

The transcription of certain other target genes by the HIF-1 transcription factor has an impact on

immune function. This impact can come from the hypoxic stabilisation of HIF-1α, but also through

the stabilisation of HIF-1α by non-hypoxic pathways such as those initiated by pathogens and

proinflammatory stimuli, which can regulate HIF-1 at transcriptional, translational and

posttranslational levels.

1.2.3.1 HIF-1α and cellular energy production in immune cells

The HIF-1 transcription factor is involved in transcribing genes for glycolytic energy generation

inimmune cells. Hypoxia is a characteristic of inflamed tissue (Sitkovsky and Lukashev, 2005) due

7

to the rapid consumption of oxygen while blood vessels are clogged with phagocytes (Sitkovsky

and Lukashev, 2005). During hypoxic conditions, cellular utilisation of oxygen-dependent cellular

oxidative phosphorylation is decreased in favour of anaerobic glycolysis. This adaption was first

noted by Pasteur in the 19th century and termed the Pasteur effect. With an available glucose source,

glycolysis allows optimized cellular energy production, function and survival in low oxygen

environments, and the HIF-1 transcription factor has been shown as a necessary mediator in

performing this metabolic switch in mammalian cells (Seagroves et al., 2001). HIF-1 up-regulates

the expression of many of the glycolytic enzymes and glucose transporters allowing increased

glycolytic pathway activity to maintain ATP levels (Semenza et al., 1994, O'Rourke et al., 1996,

Iyer et al., 1998, Chen et al., 2001).

Myeloid cells such as neutrophils and macrophages are amongst the first responders to changes in

tissue integrity or entry of pathogenic organisms (Nizet and Johnson, 2009), constituting important

components of the innate immune inflammatory response to invading microbial organisms and

foreign bodies. These cells release many different anitimicrobial molecules and pro-inflammatory

mediators, localise and phagocytise pathogens and tissue debris; fight local infection and prevent

systemic spread (Nizet and Johnson, 2009). It has been shown that when HIF-1α is absent, the ATP

pool in macrophages and neutrophils is drastically reduced to 15-40% of wild-type levels with

concurrent impairment of myeloid immune function, even under normoxic culture conditions

(Cramer et al., 2003). A number of other studies have linked ATP levels in myeloid cells with the

capacity for inflammation (Kittlick, 1986). Such ATP depletion in the absence of HIF-1α suggests

that myeloid cells may rely on glycolysis as a main mechanism of energy production.

1.2.3.2 HIF-1α and immune cell function

The importance of HIF-1α in myeloid cell function has been thoroughly demonstrated in the

aforementioned, key study by Cramer et al (2003). This study employed conditional gene targeting

of myeloid cells in knock-out mice (silencing HIF-1α, its negative regulator pVHL and a known

downstream target, VEGF). These knock-out mice demonstrated no abnormalities under normal

conditions, however showed marked aberrations in experimental models of acute inflammation. The

knock-out mice did not develop severe joint swelling and cartilage destruction in a collagen induced

arthritis and showed no cutaneous redness or oedema, cardinal signs of inflammation, following

application of a chemical irritant to the skin, suggesting an impaired acute inflammatory response.

(Cramer et al., 2003). This study also demonstrated that loss of HIF-1α impaired key macrophage

cell activities such as aggregation, invasion and motility, and that HIF-1α is essential for myeloid

cell infiltration and activation in vivo through a mechanism independent of VEGF (Cramer et al.,

8

2003). There have been studies with conflicting results however. A 2005 study by Peyssonnaux

found that HIF-1α -null murine neutrophils were recruited to sites of microbial infection as

efficiently as wild type neutrophils. It is possible the presence of a bacterial stimulus in the latter

study and the use of different myeloid cell types may factor into the difference in myeloid cell

recruitment after loss of HIF-1α.

HIF-1α exerts its mechanism of action in acute inflammation through induction of ß2 integrin

adhesion receptors whose expression coordinates the recruitment of myeloid cells to sites of

inflammation (Kong et al., 2004), and HIF-1α activity delays neutrophil apoptosis to sustain a local

immune response (Walmsley et al., 2005). Overexpression of HIF-1α in macrophages has shown

surprising increases in phagocytosis, with macrophages better able to kill bacteria under hypoxic

conditions than normoxic conditions (Anand et al., 2007).

In contrast, HIF-1α appears to dampen more chronic inflammatory processes through negative

regulation of lymphocyte adaptive immune functions. Chimeric mice with HIF-1α deficient T and

B cells showed greatly increased autoimmune tissue damage (Kojima et al., 2002) indicating a

possible tissue-protecting role for HIF-1α. Increased production of HIF-1α in T cells induces a shift

from type 1 helper T-cell to type 2 helper T-cell regulated immune responses, which then inhibit

Th1-mediated microbicidal functions (Eltzschig and Carmeliet, 2011). HIF-1α also stimulates

proliferation of regulatory T-cells which are inhibitory T-cells (Ben-Shoshan et al., 2008). HIF-1α

transcriptional regulation negatively modulates proinflammatory cytokine production by CD4+ and

CD8+ T cells, as activated T cells with deletion of the Hif1a gene released greater levels of tumour

necrosis factor-α (TNF-α) and interferon- γ (IFN-γ) than wild type T cells (Lukashev et al., 2006).

Another study supported these findings by using a murine model of sepsis to show that activated

Hif1a gene-deficient T cells underwent increased cell proliferation and enhanced secretion of

interleukin-2 (IL-2), IFN-γ, TNFα and interleukin-6 (IL-6), blocking bacterial proliferation and

improving survival of mice (Thiel et al., 2007).

These immunosuppressive actions are thought to be the result of hypoxia-induced accumulation of

extracellular adenosine due to hypoxic regulation of enzymes involved in adenosine metabolism

(Synnestvedt et al., 2002). Adenosine accumulation is a rapid response to hypoxia, and subsequent

signaling via adenosine receptors results in accumulation of immunosuppressive intracellular cylic

AMP (cAMP) and thus inhibition of T-cell receptor activated T-cells (Sitkovsky, 2009).

Strict regulation of HIF-1 function in T cells is important to prevent cell death and suggests T cells

play a role in hypoxia-mediated and HIF-1 dependent cell death in hypoxic tissue (Biju et al.,

2004).

9

1.2.3.3 HIF-1α induction of pro-inflammatory molecules and further regulation of the HIF-1

pathway due to inflammation and pathogens

1.2.3.3.1 HIF-1α and inflammatory mediators

During the innate immune response, HIF-1α promotes the myeloid expression of antimicrobial

peptides, granule proteases, VEGF and pro-inflammatory cytokines such as TNF-α, interleukin-1α,

interleukin-1ß (IL-1ß) and interleukin-12. In addition, it upregulates toll-like receptor (TLR)

expression and activates production of nitric oxide (NO) by inducible nitric oxide synthase (iNOS)

(Peyssonnaux et al., 2005, Peyssonnaux et al., 2007). Increased VEGF and the released

proinflammatory cytokines recruit and activate more immune effector cells (Zinkernagel et al.,

2007).

TNF-α and IL-1ß are proinflammatory cytokines that have been shown to increase the activity of

HIF-1α via increased mRNA translation, as seen in tumour cells (Hellwig-Burgel et al., 1999), renal

proximal tubular cells (Zhou et al., 2003), epithelial cells(Sandau et al., 2001b) and vascular smooth

muscle cells (Richard et al., 2000). Many of these cell types also express pro-inflammatory

cytokines like IL-1β and TNF-α, thus creating an autocrine feedback loop (Frede et al., 2007). TNF-

α also increases HIF-1α protein expression in neutrophils and macrophages participating in wound

healing (Albina et al., 2001).

TNF- α induces HIF-1α via NO and ROS pathways (Haddad and Land, 2001, Brune and Zhou,

2007) and can also post-translationally induce HIF-α activity in a nuclear factor [kappa] B (NF-

κB)-dependent manner (Jung et al., 2003, van Uden et al., 2008). In addition, platelet-derived

growth factor (PDGF) is increasingly produced in response to hypoxia and can itself stimulate HIF-

1 dependent gene expression in an autocrine-paracrine way (Zhang et al., 2003).

NO is a microbicidal molecule produced by activated macrophages and granulocytes during

inflammation via the inducible NO synthetase (iNOS) (Thomas et al., 2008). It has been long

known that HIF-1 activates the expression of iNOS to produce NO (Melillo et al., 1997, Jung et al.,

2000, Hu et al., 2002) however more recent work indicates a contrast between classic pro-

inflammatory macrophages (M1macrophages) in which Th1 cytokines induce HIF-1α, ultimately

upregulating NO, to that of alternatively activated macrophages (M2 macrophages) in which Th2

cytokines induce HIF-2α and ultimately suppress NO (Takeda et al., 2010). In the context of

bacterial infection, iNOS expression in wild-type macrophages is principally regulated by HIF-1α

(Peyssonnaux et al., 2005). It has also been demonstrated that in the same circumstances, inhibition

of iNOS blocks the expected generation of increased levels of HIF-1α, indicating an amplification

10

loop where NO plays an important role in further stabilization of HIF-1α in macrophages (Sandau

et al., 2001a, Peyssonnaux et al., 2005).

NO has been shown to stabilize HIF-1α by redistributing cellular oxygen, inhibiting PHD activity

and preventing HIF-1α ubiquination during normoxia (Sandau et al., 2001a, Hagen et al., 2003,

Peyssonnaux et al., 2005, Brune and Zhou, 2007). However, under hypoxic conditions it has been

shown that NO can also have an inhibitory affect on HIF-1α stabilization and activity by

reactivation of PHD (Callapina et al., 2005, Zhou and Brune, 2006).

During acute inflammation, myeloid cells produce a burst of ROS that is critical for the bactericidal

action of phagocytes, but may result in damage to host tissues if ROS formation proceeds

uncontrolled (Dehne and Brune, 2009). This oxidative burst occurs equally well in wild-type and

HIF-1α deficient macrophages (Peyssonnaux et al., 2005). During hypoxia, subsequent ROS

generated within mitochondria stabilise HIF-1α through the same mechanism of inhibition of PHD

enzyme function (Chandel et al., 2000, Guzy et al., 2005).

Ascorbate is an antioxidant that keeps PHD enzymes in a reduced form and therefore functioning to

tag HIF-1α for degradation. An ascorbate deficiency has been shown to limit PHD function in

human tumours, resulting in increased HIF-1α levels (Knowles et al., 2003). Other work involving

deletion of a transcriptional regulator responsible for increasing antioxidant levels resulted in

increased HIF-1α levels due to increased free ROS, which in turn cause direct oxidation of Fe (II) at

the catalytic site of PHD, which ultimately limits PHD enzyme activity (Gerald et al., 2004).

In addition, a non-hypoxic pathway mediating the effect of TNF-α in regulating the stabilization,

translocation and activation of HIF-1α in a ROS-sensitive mechanism has also been reported

(Haddad, 2002). Taken together, these findings indicate that an increase in ROS during acute

inflammation and/or hypoxia may result in further HIF-1α stabilization.

One may speculate that HIF-1α may be situated at the centre of an amplification loop mechanism

for innate immune activation: stimulation of HIF-1α by hypoxia and bacteria induces the production

of NO and TNF-α, functioning not only to generate inflammation and control bacterial infection but

also to further stabilize HIF-1α in recruited myeloid cells (Peyssonnaux et al., 2005). To complete

the loop, increases in ROS/nitrogen species in the inflammatory process have been shown to

stabilize HIF-1α (Pouyssegur and Mechta-Grigoriou, 2006). Transgenic mice expressing

constitutively active HIF-1α in the epidermis showed increased VEGF and capillary formation, but

no evidence of vascular leakage or inflammation (Elson et al., 2001), indicating that the effects of

increased constitutive HIF-1α (instead as part of a positive feedback loop) within non-inflamed

11

epidermis may be controlled by the physiochemical microenvironment. The effects of increased

HIF-1α within an inflammatory process as induced by hypoxia and inflammatory mediators may be

different.

1.2.3.3.2 HIF-1α and the immune response to viral pathogens

The relationship between oxygen tension and viral replication is more complex than previously

thought, with recent studies collectively highlighting the major but variable role that HIF-1α plays

in modulating viral infections in both normoxic and hypoxic conditions (Morinet et al., 2013).

For example hepatitis C virus replication is enhanced under hypoxic conditions, although the

mechanism appears to be HIF-1α –independent (Vassilaki et al., 2013). The nuclear antigens of

Epstein-Barr virus bind to PHD, inhibiting HIF-1α degredation (Darekar et al., 2012). The

respiratory tract pathogen respiratory syncytial virus (RSV) induces HIF-1α in human bronchial

cells via a NO-dependent pathway, and the resultant increased VEGF also enhances monolayer

permeability, a pathway which may contribute to airway oedema of acute RSV infection (Kilani et

al., 2004). Hypoxia-independent stabilization of HIF-1α in vitro and in vivo has also been observed

in another study of RSV (Haeberle et al., 2008), further supporting HIF-1α stabilisation occurring

as a result of a viral infection.

Human immunodeficiency virus-1 (HIV-1) viral protein vpr induces ROS, which contribute to HIF-

1α expression, with the subsequent interesting mechanism of HIV-1 gene transcription brought

about by HIF-1α forming a heterodimer with vpr (Deshmane et al., 2009).

Hypoxia induces the lytic replication cycle of human herpesvirus-8 (HHV-8, also known as

Kaposi’s sarcoma-associated herpesvirus) in vitro via a functional HRE (Davis et al., 2001, Haque

et al., 2003). HHV-8 also contains a latency associated nuclear antigen (LANA) that enhances the

transcriptional level of HIF-1α mRNA but also induces the lytic replication cycle during hypoxia

(Dalton-Griffin et al., 2009). LANA also stabilises HIF-1α during normoxia by targeting pVHL

(Cai et al., 2007). However, a more complex relationship between HHV-8 and HIF-1α has emerged

with evidence that the HHV-8 viral protein ORF34, previously transcribed via HIF-1, was then able

to induce HIF-1α degradation via the proteasome (Haque and Kousoulas, 2013).

Reduction of viral replication and protein expression of an adenovirus has also been reported

(Pipiya et al., 2005). The replication of vesicular stomatitis virus (VSV) is inhibited in vitro by HIF-

1α (Hwang et al., 2006). Renal carcinoma cells with constitutive HIF-1 activity during nomoxia

demonstrate enhanced cellular resistance to the cytoltic effect of VSV when compared to wild-type

with the expected low HIF-1α activity (Hwang et al., 2006).

12

1.2.3.3.3 HIF-1α and the immune response to bacterial pathogens

Bacteria lipopolysaccharide (LPS) is a component of the cell wall of gram-negative bacteria (Quinn

et al., 2011) and has been shown to induce stabilisation of HIF-1α mRNA in resident lung

macrophages under normoxic conditions (Blouin et al., 2004, Eitam et al., 2010).

HIF-1α expression increased four-fold in wild type macrophages at normoxia following exposure to

gram-positive or gram-negative bacteria with resulting HIF-1α transcriptional gene activity levels

similar to or greater than levels achieved with hypoxia (Peyssonnaux et al., 2005). In this study by

Peyssonnaux et al, mice lacking HIF-1α in their myeloid lineage showed decreased myeloid

bactericidal activity, with 2-fold decrease in intracellular killing under normoxia and 3-fold

decrease intracellular killing of a Group A Streptococcus under hypoxia, with macrophage killing

of the gram negative P. aeruginosa likewise impaired. Conversely, activation of the HIF-1α

pathway by deletion of the VHL protein or by pharmacologic inducers improved myeloid cell

bactericidal activity (Peyssonnaux et al., 2005).

1.2.4 HIF-1α and BRD

1.2.4.1 HIF-1α and bovine pulmonary immune responses

Pulmonary diseases associated with intra-alveolar exudates, edematous septal thickening, infection

and inflammation can often result in decreased mucosal oxygenation and ultimately blood and

tissue hypoxia (Schaible et al., 2010). As with other tissues, hypoxia and immune signaling

pathways in the lung are closely connected and so subsequent activation of various hypoxia

sensitive pathways may contribute to the development of infectious and inflammatory lung disease

(Schaible et al., 2010) hypoxic induction of HIF-1 in ferrets has been shown to occur in many

pulmonary cell types including pulmonary arterial endothelium, smooth muscle, bronchial

epithelium, alveolar epithelium alveolar macrophages and microvascular endothelium (Yu et al.,

1998).

In the bovine respiratory tract there are similar innate immune responses as identified in other

mammalian species, and given the association of hypoxia with lung disease it is reasonable to

expect that HIF-1α plays a similar role in regulating these responses. Activation of HIF-1α in

macrophages and neutrophils in the bovine lung likely enhances the innate immunological functions

of these cells via increased bacterial killing, phagocytosis, cell survival and cytokine production

(Cramer et al., 2003, Schaible et al., 2010). Alveolar epithelial cells, resident alveolar macrophages

and dendritic cells recognize various microbial components via pattern recognition receptors (PRR)

with resultant production of proinflamamtory cytokines via the NF-kB pathway, directly affecting

expression of genes involved in innate immunity as well as indirectly causing amplification of the

HIF pathway (Schaible et al., 2010).

13

The anti-inflammatory actions of HIF, brought on by concurrent local hypoxia and accumulation of

adenosine that negatively regulate T-cells (see section 1.2.3.2) also participate in negative

regulation of the pulmonary innate immune response in vivo, as demonstrated by a study of

neutrophil-mediated lung inflammation in mice (Thiel et al., 2005). In addition, HIF-1α is involved

in the regulation of several barrier-protective genes that are thought to facilitate a protective role of

HIF-1α activation in a model of mucosal inflammation using murine colitis (Robinson et al., 2008)

a role which, given similar structural features and a common embryonic origin of the foregut

(Gilbert, 2010), may also apply to respiratory mucosal inflammation.

1.2.4.2 HIF-1α as a potential biomarker for development of clinical BRD

Of diseases in beef cattle, respiratory disease has the greatest economic impact. Economic losses

occur due to morbidity, mortality, treatment and prevention costs and loss of production (Griffin,

1997). Due to the high incidence of pneumonia in cattle, the ubiquity of the respiratory pathogens,

and the various anatomical factors predisposing cattle to respiratory infection, investigations are

focused on ways to enhance effective and non-injurious immune responses to these pathogens

(Ackermann et al., 2010b) in addition to combating environmental and managerial issues that may

predispose cattle to respiratory disease.

BRD as a specific entity is a multi-factorial, inflammatory respiratory disease complex which poses

a significant problem for the beef industry. The disease complex involves both viral and bacterial

pathogens and invokes both innate and adaptive immune responses as well as the development of

clinical pneumonia in certain feedlot cattle. The main viruses of this disease complex are bovine

herpesvirus-1 (BHV-1), bovine respiratory syncytial virus (BRSV), bovine viral diarrhoea virus

(BVDV) and parainfluenza-3 virus (PI-3)(Srikumaran et al., 2007) and the prominent bacterial

pathogens are Mannheimia haemolytica, Histophilus somni and Mycoplasma bovis (Srikumaran et

al., 2007). BRD involves complex interactions amongst these viral and bacterial pathogens with

viral infection predisposing cattle to secondary bacterial pneumonia, resulting in intense pulmonary

inflammation (Hodgson et al., 2005). An experimental model of this viral-bacterial synergy has

been described which utilizes primary BHV-1 respiratory challenge followed by aerosol challenge

with M. haemolytica (Bielefeldt Ohmann et al., 1991).

BHV-1, PI-3, BRSV and BVDV can infect lung epithelial cells, and can signal through TLR 3, 7, 8

and 9 (Ackermann et al., 2010b). Following BHV-1 infection, innate immune responses include the

antiviral action of and the recruitment of lymphoid cells, macrophages, neutrophils and natural

killer cells to the site of infection (Bielefeldt Ohmann et al., 1991, Babiuk et al., 1996). Bovine

14

bronchial cells that are infected with BHV-1 will release cytokines, which subsequently promote

neutrophil recruitment to the lung (Rivera-Rivas et al., 2009). In vitro, TNF-α, IL-1β and IFN-γ are

secreted in response to BHV-1 infection, causing the increased expression of the β2 integrin on

bovine alveolar macrophages and neutrophils (Czuprynski, 2009). The β2 integrin is the receptor

for the leukotoxin (LKT) of M. haemolytica, and LKT binding causes neutrophils and macrophages

to release cytokines and reactive oxygen intermediates that might further exacerbate the

inflammatory response in the respiratory tract (Czuprynski, 2009).

Although BHV-1 is a cytopathic virus, it is not this effect, but rather the virus-induced release of

cytokines and their subsequent effects, which constitute the basic mechanisms in the inflammatory

process and lung pathology (Bielefeldt Ohmann et al., 1991). Although vigorous inflammatory

response initiated by the virus infection can be beneficial in host defense, it also establishes

conditions that predispose to poorly regulated inflammation in the lung. Activation of the HIF-1a

pathway by microenvironmental hypoxia in myeloid cells will enhance the innate immune response

to microbial pathogens via cytokine production, microbial killing, increased phagocytosis, NO

production and TLR expression as already reviewed (Schaible et al., 2010). However this enhanced

inflammatory response may also contribute to tissue damage and further lung disease.

The increased production of pro-inflammatory cytokines that occurs during a primary BHV-1

infection is of significance to this project. It is in the context of this vigorous inflammatory response

that is characteristic of BRD viral-bacterial synergy, that the possibility of HIF-1α as a biomarker is

raised. The previously described involvement of HIF-1α in a veritable amplification loop of

cytokine-ROS innate inflammation (see section 1.2.3.3) may make HIF-1α an interesting molecule

to study in the context of the cytokine-involved inflammatory process culminating in BRD.

Work on H5N1 human influenza virus in cynomolgus macaques has found a pronounced

upregulation of HIF-1α expression in peripheral blood leukocytes and tissue macrophages in the

context of severe pneumonia (Tolnay et al., 2010), along with profound and sustained upregulation

of mRNA for IFNs, IL-1, IL-6 and TNF-α proteins (Baskin et al., 2009). Alternatively a lack of

HIF-1α response could also increase susceptibility to BRD. A sluggish HIF-1α response could

decrease the hypoxia-triggered HIF-1α adenosinergic suppression of proinflammatory cytokines in

CD4+ and CD8+ lymphocytes, an important immunosuppressive mechanism which downregulates

the immune response to protect local tissue and maximize the anti-pathogen cellular response

during pathogen encounter (Sitkovsky and Lukashev, 2005, Zinkernagel et al., 2007, Sitkovsky,

2009). However, the variable interactions of HIF-1α and viruses during the establishment of viral

15

infections (see section 1.2.3.3.2) suggests that many other mechanisms or indeed complex

combination of mechanisms may be at play.

1.2.4.3 Previous BRD biomarker studies

There have been previous studies investigating possible biomarkers for BRD outcome (Aich et al.,

2009, Eitam et al.). Aich et al 2009 established the value of using combinatorial “omics”

approaches to identify candidate biomarkers and used multimodal analysis of these biomarkers to

predict disease outcome. Twenty calves seronegative on enzyme-linked immunosorbent assay

(ELISA) for BHV-1 and leukotoxin of M. haemolytica were challenged on day 0 with BHV-1, with

serum samples collected prior to this infection. All calves were then challenged with M.

haemolytica on day 4 post infection when susceptibility to fatal secondary bacterial infection is

greatest, with serum samples collected days 1-4. Eight infected calves survived and twelve died.

Comparing samples from control animals on day 0 with samples from BHV-1 infected calves on

days 1-4 identified differentially expressed and statistically significant biomarkers. Multivariate

analysis was used to look for significant differences when animals were grouped by BRD outcome

of survival versus mortality. Proteomic studies identified a group of acute phase proteins,

haptoglobin and apolipoprotein, which could be linked to the primary viral respiratory infection, but

there was no significant association observed with fatal BRD. In contrast, metabolomic and

elemental analyses identified biomarkers for a primary viral infection (glucose, low-density

lipoprotein, valine, phosphorous, and iron) and also the disease outcome (lactate, glucose, iron).

While multivariate analysis of proteomic and metabolite profiles did not discriminate between

animals that survived or died post-synergic viral–bacterial infection by analyzing preinfection

serum samples, analysis of pre-infection serum elemental profiles, did allow some discrimination

(Aich et al., 2009). This study employed high-throughput methodologies but only challenged

twenty calves, and the number of control animals used is unclear. This study was designed to

discriminate between the complex biological responses induced by the primary viral infection and

the secondary bacterial infection resulting in mortality. While death due to BRD is a significant

disease outcome, the production loss in animals with less severe disease is also important. Using

this animal model achieving 60% mortality, there was little scope to assess any relationship of these

biomarkers to the varying severity of non-fatal BRD. It remains to be determined whether identified

biomarker profiles as predictors for mortality are only associated with the maximal viral-bacterial

synergy on day 4, or whether these biomarkers could be predictive of mortality with earlier or later

bacterial challenges.

Eitam et al 2010 aimed to detect individual variations in the stress responses of 13 newly received

16

young calves through measurement of their leukocyte heat shock protein response, selected

neutrophil-related gene expression and oxidative stress, and relate these variations to pulmonary

adhesions at slaughter as an indicative sign of clinical and subclinical episodes of BRD at an early

age. This study developed a discriminant analysis model, based on levels of linoleoyl tyrosine

oxidative products and β-glycan, according to which vulnerable individuals may be predicted at

near 100% probability after seven days arrival at the feedlot. The design and results of this study are

intriguing, but do address the problem of early detection of subclinical signs. However, as the

authors note, a larger scale experiment (i.e. n > 13) should be carried out to confirm this model

(Eitam et al.).

1.3 Conclusions and Research Proposal

To investigate the possibility of the cytokine/ROS- induced positive feedback loop of HIF-1α

induction contributing to the development of clinical BRD, or a possible lack of HIF-1α

stabilisation contributing to the development of clinical BRD, the following experiment, null

hypothesis and alternative hypotheses have been established.

1.3.1 Experimental hypothesis

The experimental hypothesis is that there is a significant difference in HIF-1α regulation between

animals with clinical BRD compared with other clinically healthy animals.

1.3.2 Null Hypothesis

The null hypothesis is that there is no significant difference in HIF-1α regulation in bovine

leukocytes of animals with clinical BRD compared with other clinically healthy animals.

1.3.3 Research Objectives

1. Isolate bovine peripheral blood leukocytes from previously observable clinically ill and

healthy animals in large enough concentrations for cell culture.

2. Develop a methodology for short-term culture of isolated leukocytes with an agent that will

stimulate HIF-1α expression; successful culture also entails achieving good post-culture cell

percentage viability and adequate total numbers of viable cells.

3. Develop a methodology for successful molecular tagging of cells with a fluoresecent marker

for the HIF-1α molecule and other phenotypic markers to identify leukocyte population sub-

types using flow cytometry or an alternative technique.

4. Investigate the value of measuring HIF-1α expression as a predictor for the outcome of

BRD by implementing developed methodologies experimentally to compare the HIF-1α

17

expression in a group of clinically healthy beef feedlot cattle with the HIF-1α expression in

feedlot cattle that developed clinical respiratory disease while in the feedlot.

5. Perform statistical analysis of acquired flow data and immunohistochemical data to

determine if there is a statistically significant difference of acute HIF-1α expression in

stimulated leukocytes from animals with BRD compared to clinically healthy animals.

18

2 Thesis overview

The next five chapters cover the development of methodologies of i) a hypoxic cell culture, ii) a

flow cytometry protocol for detection of HIF-1α in lymphocytes and iii) an immunocytochemistry

(ICC) protocol for detection of HIF-1α in monocytes, followed by the application of developed

methodologies and a general discussion. The development of these methods were interconnected

and often conducted concurrently. Each method involved iterative cycles of trial and error to

optimise different aspects (see figure 2.1).

Figure 2.1: Thesis overview showing the interconnected relationships between various aspects of the methodologies developed in chapters 3, 4 and 5. Flow cytometry results influenced the cell culture methodology and lead to the development of ICC. All developed methodology is applied to experimental samples in chapter 6, followed by a discussion of developed methodology and future directions in chapter 7.

19

3 Optimization of hypoxic cell culture and harvest

3.1 The production of HIF-1α in hypoxic conditions in vivo

When a cell encounters a low oxygen environment, intracellular HIF-1α protein is stabilized by a

decrease in HIF-1α ubiquitination and degradation (Kallio et al., 1999) due to reduced activity of

oxygen-dependent PHD enzymes (Epstein et al., 2001, Jaakkola et al., 2001)}. PHDs are also

actively degraded by the E3 ubiquitin ligase (Nakayama et al., 2004). Activating this pathway by

exposing the cell to true or simulated hypoxia will thus potentially allow measurement of cellular

HIF-1α.

3.1.1 Hypoxic cell culture

The hypoxic cellular environment required for HIF-1α stabilization is created experimentally by

one of two cell culture methods. Firstly, a hypoxic cell culture can be achieved by use of a modular

incubator chamber (Wu and Yotnda, 2011) to create a hypoxic atmosphere in which cell cultures

are conducted. In tissue culture, hypoxia is routinely defined as levels that are equivalent to between

0.5% and 3% oxygen by volume of air that perfuses the growth medium (Linder et al., 2003, Nizet

and Johnson, 2009). The second common method is to use a hypoxic-mimetic agent to simulate

hypoxia in cell culture. Common hypoxic-mimetic agents that induce HIF-1α are the iron chelator

desferrioxamine (DFO) (Wang and Semenza, 1993b) and cobalt chloride (CoCl2) (Wang and

Semenza, 1993c).

DFO is thought to induce HIF- 1α by substitution or removal of the Fe2+ ion (Wang and Semenza,

1993b, Schofield and Ratcliffe, 2004), inhibiting PHD enzyme activity. Studies have shown CoCl2

may inhibit HIF-1α degradation though a variety of other mechanisms including binding directly to

HIF-1α, thus hindering recognition by the pVHL protein, and may also be dependent on ROS

formation (Yuan et al., 2003, Chachami et al., 2004, Salnikow et al., 2004).

Hypoxic-mimetic agents are commonly used in hypoxic cell culture with human cell-lines to

stabilize and measure HIF-1α (Wu and Yotnda, 2011). Based on the prevalence and convenience of

hypoxic-mimetic cell culture utilisation, a hypoxic mimetic was chosen to stimulate stabilisation of

HIF-1α (hereafter referred to as stimulation) within selected bovine leukocytes. The aim was to

develop a protocol for hypoxic cell culture of selected bovine leukocytes that provided an optimal

yield of the target cells under conditions that promoted stabilisation of HIF-1α.

3.1.2 Bovine leukocytes chosen for hypoxic culture

Following BHV-1 infection, innate immune responses include the antiviral action of IFN and the

recruitment of lymphoid cells, macrophages, neutrophils and natural killer cells to the site of

infection (Bielefeldt Ohmann et al., 1991, Babiuk et al., 1996). Although neutrophils are phagocytes

20

involved in these responses, the neutrophil populations from the test animal or experimental

animals were not included in this study because granulocytes are typically poorly preserved by

freezing (Boonlayangoor et al., 1980, Nishimura et al., 2001) and use of fresh cells for culture was

not practical. The importance of HIF-1α in myeloid cell function has been thoroughly demonstrated

(Cramer et al., 2003) and HIF-1α has also been shown to be critically involved in the regulation of

the antimicrobial functions of lymphocytes (Thiel et al., 2007). Based on the involvement of these

cells in BHV-1 infection and the functional importance of cellular utilisation of HIF-1α, bovine

monocytes and lymphocytes were chosen for development of method for a hypoxic cell culture.

Monocytes and lymphocytes also survive frozen storage and were a practical choice for this project.

HIF-1α protein in leukocytes has traditionally been measured using immunoblot assays and

immunohistochemistry (IHC)(Yu et al., 1998, Semenza, 2005, Peyssonnaux et al., 2005) or ICC

(Jantsch et al., 2008). Most studies involving measurement of HIF-1α protein have been performed

in human or mouse cells, and the transferability of the findings to bovine leukocytes was uncertain.

Western blot analysis of HIF-1α has been undertaken in cultured bovine luteal cells (Nishimura

and Okuda, 2010) and corneal stromal cells (Xing and Bonanno, 2009).

This hypoxic cell culture aimed to (i) induce intracellular HIF-1α stabilisation, with cell harvest

focused on minimizing degradation of stabilized HIF-1α and (ii) facilitate subsequent measurement

of HIF-1α in these cells. This project explored the use of flow cytometry to measure the stabilised

HIF-1α protein as a relatively novel technique with the hopes that development of a successful

methodology would allow enhanced sensitivity in detecting differences in HIF-1α expression

between individuals. In the wake of methodology development issues, ICC was also implemented

as a more established method of HIF-1α measurement.

3.2 Materials, Methods and Results

To maximize potential for successful HIF-1α stabilization via hypoxic-mimetic culture with

minimal post-culture HIF-1α degradation, many aspects of the hypoxic-mimetic cell culture and

harvest methods required concurrent optimization. These aspects included the choice of hypoxic

mimetic agent and other stimulants, choice of culture duration, choice of culture vessels and cell

harvest procedures. These methods and results for optimization of these aspects are described in

sections 3.2.4, 3.2.5 and 3.2.6 below. Pre-culture sample preparation methods did not require

further optimization and are described in sections 3.2.1, 3.2.2 and 3.2.3 below.

21

3.2.1 Bovine leukocytes used in development of hypoxic-mimetic cell culture methods

Bovine blood, used for methodology testing and development was collected in 450 ml blood bags

from clinically normal female, Holstein Fresian teaching cows residing at the University of

Queensland (UQ) Gatton Campus dairy facility.

3.2.2 Procedure for mononuclear cell isolation

Mononuclear leukocytes were isolated from 300 ml of whole blood. Using an Eppendorf Centrifuge

5810R (Eppendorf South Pacific, Australia), 50 ml aliquots of peripheral blood were centrifuged at

2000 g, 4 °C for 25 minutes to separate out the erythrocytes from the leukocytes (“buffy coat”

layer). The buffy coat layer was collected from each aliquot and transferred into a fresh 50 ml tube,

and phosphate-buffered saline solution (PBS, pH 7.2) added to achieve a 20 ml total volume. The

resulting cellular suspension was layered onto 25 ml of a solution containing polysucrose and

sodium diatrizoate, adjusted to a density of 1.077 g/ml (Histopaque-1077; Sigma Aldrich, St Louis,

MO) and centrifuged at 2000 g, 20 °C for 40 minutes. The mononuclear cell layer was then

collected and transferred into a fresh 50 ml tube, filled to 50 ml with PBS and centrifuged at 1000 g,

4 °C for 15 minutes. The supernatant was poured off, and the cell pellet resuspended in 50 ml of

fresh PBS. This last step was repeated twice, centrifuging for 10 and 5 minutes, respectively.

Ten µl of cell suspension was mixed with 90 µl trypan blue 0.4% (Life Technologies, Victoria,

Australia) and a manual cell count of viable and non-viable cells was performed for each animal

using a haemocytometer (Neubauer, Germany). Cell viability was assessed on exclusion of trypan

blue (Strober, 2001). Percentage viability was calculated as the number of viable cells divided by

the number of total cells counted x 100. The number of cells/ml of sample was calculated using the

following method: #Viable cells counted within the 1mm x 1mm haemocytometer counting grid x

105 = cells/ml; cells/ml x sample volume in ml = total cell number in sample. These methods were

used for all cell viability assessments and counts conducted during the project. Cells in PBS were

spun down and resuspended at a concentration of 1 x 107 cells/ml and 0.9 ml aliquots were

transferred into 1.8ml sterile round-bottom cryotubes (Nunc, Roskilde, Denmark) mixed at a 1:1

ratio with freezing solution comprising 40 % Roswell Park Memorial Institute medium (Sigma

Aldrich, St Louis, MO), 40 % horse serum (Life Technologies, New Zealand) and 20 % dimethyl

sulphoxide (Sigma Aldrich, Missouri, USA).

3.2.3 Preparation of frozen test samples for culture

All cell cultures described hereafter took place in a Sanyo CO2 Incubator MCO-19AIC (Sanyo,

Sakata, Japan) at 5% CO2 and 37°C. The pre-culture setup procedures took place in a Biological

22

Safety Cabinet Class II (Email Westinghouse Pty Ltd, New South Wales, Australia) after 15

minutes of ultraviolet light sterilization and treatment of all items and surfaces with 70% ethanol.

Each frozen 1.8ml cryotube theoretically contained approximately 9 x 106 viable cells. Eight

cryotubes frozen at -81°C were retrieved for each animal and thawed at 37°C in a Grant W14 water

bath (Grant, Cambridge, UK). The contents of the cryotubes were pooled and immediately washed

three times in 15 ml PBS, centrifuged at 1000 RPM for 7 minutes at 4°C during each wash. The

cells were recounted and post-freezing cellular viability ranged from 65% to 86%, with total cell

yields (out of a theoretically possible 7.2 x 107 cells from 8 cryotubes) ranging from 1 x107 to 6 x

107 cells.

3.2.4 Hypoxic mimetic culture, stimulant and culture medium

3.2.4.1 Culture medium

The culture medium used in all cultures was X-Vivo 15 serum-free medium with gentamicin and

phenol red (Lonza, Maryland, USA). Gentamicin was considered essential for this project because

cells isolated from non-sterile blood collections were being cultured.

3.2.4.2 Hypoxic-mimetic culture and comparing hypoxic mimetics

For cell culture, DFO (Sigma Aldrich, St Louis, MO, USA) and CoCl2 (Sigma Aldrich, St Louis,

MO, USA) were tested for suitability as a hypoxic-mimetic agent to stimulate bovine leukocytes.

The criteria for suitability were adequate cellular viability and adequate total cell numbers

harvestable post-culture.

DFO was the first available hypoxic mimetic to be tested. Fresh mononuclear cells were washed

three times in PBS, and then re-suspended in serum-free medium at an approximate concentration

of 1 x 107 cells/ml. Five cultures were seeded with 250 µl cell suspension (estimated 2.5 x 106 cells)

and 250 µl of 200 µM DFO, achieving a final concentration 100 µM DFO. Five negative control

cultures were seeded with 250 µl of cell suspension and 250 µl serum-free medium without DFO.

All culture plates were incubated in 5% CO2 at 37°C. After 42 hours, culture fluid was collected

and plates rinsed twice with PBS to collect lymphocytes. The percentage viability of lymphocytes

in DFO-treated cultures ranged from 55 to 83%, and harvestable cell counts ranged from 6 x 105 to

1.2 x 106 cells. Control cultures had 63 to 91% viability and cell yields ranging from 7 x 105 to 1.6 x

106 cells.

23

Further cell cultures were set up to compare the effects of DFO to CoCl2 on cell viability and total

cell yields post-culture. Duplicates of approximately 5 x 106 cells were treated with 100 µM DFO,

100 µM CoCl2 or 200 µM CoCl2 and cultured along with untreated control cultures. One duplicate

well was harvested after 18 hours of culture and the other at 42 hours. After 18 hours, only a minor

difference in total cell yield for the two CoCl2 concentrations was recorded (3.9 x106 cells for 100

µM and 3.2 x 106 cells for 200 µM), with cellular viability of 90% for 100µM and 94% for 200µM

CoCl2. In contrast, DFO culture had only 1.2 x106 cells and 63% viability and the 18-hour control

total cell count was 1.1 x 107 cells with 95% viability. Cultures harvested at 42-hours had total cell

yields of 1.7 x106 cells for both 100µM and 200µM CoCl2 with less than 50% cellular viability.

The DFO culture cell yield at 42-hours was 1.6 x 106 cells with 55% viability whereas the control

culture contained 5.5 x 106 cells with 91% viability. Thus, at the tested concentrations CoCl2 and

DFO gave similar cell viability and yields at 42 hours, while at 18 hours both concentrations of

CoCl2 yielded greater cell viability and total cell numbers in comparison to the DFO culture. The

18-hour higher cell viability of CoCl2 compared to culture of the same duration using half the

concentration of DFO indicated that CoCl2 could be used at higher concentrations to maximize

potential hypoxia while minimizing cytotoxicity. CoCl2 was therefore the hypoxic mimetic of

choice for further hypoxic cell culture development.

3.2.4.3 Lipopolysaccaride (LPS) as an additional culture stimulant

CoCl2 was titrated further with or without the addition of lipopolysaccharide (LPS) from

Escherichia coli (Sigma Aldrich, Missouri, USA). The inclusion of LPS as an additional stimulant

in these hypoxic cell cultures to stabilize the HIF-1α molecule was based on the previously verified

ability of bacteria or LPS to stabilize and contribute to accumulation of HIF-1α under both hypoxic

and normoxic conditions (Rius et al., 2008). LPS raises HIF-1α levels through pathways involving

MAPK and NF-kB (Peyssonnaux et al., 2007). Maximal induction of HIF-1α in alveolar

macrophages has been observed using an LPS concentration of 1µg/ml in 6-hour cultures of murine

alveolar macrophages (Blouin et al., 2004) but LPS concentrations in this project were increased to

test the upper limits of using this stimulant. CoCl2 was therefore further tested in six-hour cultures.

Each culture contained 2.5 x 106 cells exposed to either 200 µM, 600 µM or 1.8 mM CoCl2 in

permutations with relatively high LPS concentrations of 5 µg/ml, 15 µg/ml and 45 µg/ml. Results

showed that cellular viability decreased with increased CoCl2 dose; culture with 600 µM and 1.8

mM CoCl2 resulted in less than 50% cellular viability after 6 hours, with total cell counts of 1.8 x

24

106 and 9.7 x 105 cells for 600 µM and 1.8 mM CoCl2 respectively where CoCl2 was used without

LPS. Cultures with 600 µM and 1.8 mM CoCl2 in combination with all concentrations of LPS also

resulted in less than 50% cellular viability. The addition of LPS at 5 µg/ml to 200µM CoCl2 gave

the best cellular viability at 73% and the highest total cell count at 2.15 x 106 cells. LPS was

subsequently used at 5 µg/ml as this concentration gave the best cellular viability while being

closest to the previously published 1µg/ml.

3.2.5 Culture duration testing

Stabilisation and measurement of HIF-1α has been recorded with a culture time as short as three

hours (Pisani and Dechesne, 2005) or four hours (Bove et al., 2008). Forty-two-hour cultures were

initially tested; these cultures often resulted in extremely low post-culture viability and experiment

discontinuation due to lack of cells. Shorter culture times (6 to 8 hours) resulted in acceptable post-

culture viability of cells (refer to section 3.2.4.2 and 3.2.4.3).

3.2.6 Culture vessel testing, post-culture viability and HIF-1α degradation

3.2.6.1 Plate cultures

The selection of culture vessel had various far-reaching effects on the outcome of the culture.

Initially, individual, 6-well or 24-well plates (Nunc, Roskilde, Denmark) were trialed. The larger

numbers of cells and culture medium volume needed were issues in addition to the important issue

of adherent monocytes. Single, 6-well or 24-well culture plates provided large surface areas for

activated monocytes to adhere to during culture. The outcome to this adherence was that a great

proportion of activated monocytes would then remain on the plate surface after repeated washing,

remaining unavailable for harvest into suspension for subsequent analysis by flow cytometry.

Lidocaine may be used to detach adherent monocytes from plate surfaces (Ohmann et al., 1983,

Nielsen, 1987). To test the feasibility of using lidocaine to retrieve adherent monocytes from plate

cultures, four trials using round culture plates were performed. After a four hour culture, the cell

culture plates were placed on ice after washing with PBS to retrieve non-adherent cells, and each

plate surface was treated with 2ml of 3.5mg/ml Ilium Lidocaine 20 (Troy Laboratories,

Glendenning, Australia). After 15-20 minutes the lidocaine reduced monocyte adherence to the

plate surface and subsequent gentle physical removal of the monocyte population by light abrasion

of the culture surface with a 20µl pipette tip was possible. Total cell yields from this harvesting

technique were on average 2 x 106 cells from an original total of 1 x 107 cells spread across four

culture plates. Unfortunately the original percentage of monocytes in the pre-culture population was

not determined at the time, but was estimated off later aliquots using manual differential counts to

25

be approximately 39%, meaning that approximately half of cultured monocytes could be retrieved

when using lidocaine during harvest.

3.2.6.2 Culturing in tubes

Five ml polypropylene sterile ‘snap cap’ tubes (Beckton Dickinson, NJ, USA) were initially chosen

for culture to discourage the adhesion of activated monocytes to the internal tube surface, and

similar total cell yields were achieved when comparing this technique to culturing and treating

plates post-culture with lidocaine. Stimulated tube cultures contained 1.25 x106 cells in 250µl

serum-free medium suspension treated with 200 µM CoCl2 and 5µg/ml LPS. A tube containing

non-stimulated cells suspended in serum-free medium was used as a control. The use of 5ml tubes

in an 18-hour culture in 5% CO2 at 37 °C gave an average 20 % cellular survival rate.

3.2.6.3 Preservation of HIF-1α by placing post-culture cells on ice, with cellular fixation in

presence of CoCl2

Tube cultures were immediately placed on crushed ice (temperature approximately 4°C) with the

intention to slow cellular processes including potential degradation of the HIF-1α molecule by the

ubiquitin proteasome (Walshe and D'Amore, 2008). The chilled cells then remained in the presence

of CoCl2 until undergoing formalin fixation. Manufacturer’s recommendations stated that the

fixative solution Medium A (Life Technologies, Victoria, Australia) should be used at room

temperature, however, it was successfully used at 4°C after consultation with the manufacturer.

Post-culture cell counts and cell washings in PBS were omitted to keep cells exposed to CoCl2 up

until the point of cellular fixation to minimise potential HIF-1α degradation, Due to reduced

binding of the B cell surface marker anti-CD20 antibody to fixed B cells, a 20-minute incubation of

cells with this antibody took place while on ice prior to fixation. However there was no further

washing step needed to remove this antibody before use of Medium A to fix the cells.

3.2.6.4 Tube cultures: monocyte viability during harvest and cellular staining

After culture, cells were harvested, fixed and stained with antibodies for surface markers of B cells,

T cells, monocytes and HIF-1α for analysis by flow cytometry (refer to chapter 4 for further detail

of staining and flow cytometry). It was observed that the cell populations harvested from all tube

cultures showed a repeatable and problematic drop in the proportions of CD14 positive cells

(monocytes) detectable by flow cytometry by detection of a pacific blue fluorochrome attached to

the CD14 antibody. To assess the relationship between cell culture and the apparent absence of the

monocyte population, aliquots of cell suspension isolated from the same individual animal were

thawed, washed and left on ice without undergoing culture, thawed and cultured as negative

26

controls, or thawed and cultured with CoCl2 and LPS. After 18 hours, a post-culture 100µl of cell

suspension was removed for cytospin slide preparation and Wright’s Giemsa staining to facilitate a

manual differential cell count. All remaining cells were routinely stained with CD14 antibody

labelled with a pacific blue fluorochrome (refer to chapter 4 for further detail of staining). A post-

staining 100µl sample of cell suspension was taken for cytospin slide preparation. A post-thawing

cytospin slide was also prepared from the uncultured cells.

Analysis software (Kaluza, Beckman Coulter) was used to gate cell populations to exclude debris

and aggregated cells (refer to Chapter 4 for further detail). Side scatter (SS) is a measure of the

internal complexity of the cell and can be useful in determining populations of monocytes (Givan,

2001). For each sample, a dot plot of cells was created by plotting SS against the signal of the

pacific blue fluorochrome of the CD14 antibody (figure 3.1). The thawed sample had a well-

demarcated CD14 positive population, considered distinct due to the relatively narrow range of SS

of this population as shown in figure 3.1A. This population had distinct positive staining for pacific

blue as shown by the population’s location as furthest to the right on the increasing log scale of

pacific blue fluorescent signal (figure 3.1A). Based on the lower end of this positive signal, a

quadrant gate entitled ‘CD14 positive cells’ was applied to set the positivity cut-off point for pacific

blue fluorescence, and the percentage of CD14 positive cells in the uncultured sample was

calculated by Kaluza to be 22.0%. These gates were replicated on the graphs for the cultured

samples. The sample cultured without stimulants had a CD14 positive population of 14.4%, as well

as a general increase in SS in both negative and positive cells (figure 3.1B). The sample cultured

with CoCl2 and LPS had a CD14 positive population of only 1.6%, as well as increased SS, and the

distinct population of CD14 positive cells was no longer visible (figure 3.1C).

Manual cell differential counts of monocytes and lymphocytes were performed on the post-

thawing/post-culture, and post-staining cytospin slides. Counts were performed by the author and

an experienced clinical pathologist. Post-thaw uncultured cells comprised 61% lymphocytes and

39% monocytes, cultured cells without stimulants comprised 70% lymphocytes and 30%

monocytes, while cells cultured with CoCl2 and LPS comprised 96% lymphocytes and only 4%

monocytes. The post-staining cytospins displayed poor cellular preservation and manual cell

differential was deemed too unreliable, however it was concluded that manual cell differential

counts showed decreased proportions of monocytes present after exposure to culture and/or

stimulants, in agreement with flow cytometry results. These findings were likely to be due to

cellular fragility and monocyte apoptosis. Poor cellular preservation on post-staining cytospin slides

27

and the overall lower percentages on monocytes detected during flow cytometry suggested the

possibility that staining procedures also contributed to monocyte depletion.

Figure 3.1: A: SS/Pacific Blue dot plot obtained from Kaluza of all events within the Cells gate. Colour gradients denote areas of lower (blue) to higher event density (red), percentages are of all events within the Cells gate. An uncultured aliquot of cells stained with pacific blue-labelled CD14 antibody shows a distinct CD14 positive population to the right of a larger group of negative cells. B: An aliquot of cells cultured without stimulants shows a similar but depleted CD14 positive population, and generalised increased SS of all cells. C: An aliquot of cells cultured with CoCl2 and LPS shows generalised increased SS of all cells but the CD14 positive population cannot be distinctly visualised.

28

3.2.6.4.1 Immunocytochemical (ICC) staining for activated caspase-3 as measure for apoptosis

Eighteen-hour cell cultures were repeated to collect post-culture and post-staining samples to

prepare further cytospin slides for ICC staining for activated caspase-3, a molecule which is

involved in programmed cellular death and can be used as a measure of apoptosis. Treatments were

i) cells that were thawed but put back on ice for 18 hours without culture, ii) cells that were cultured

without CoCl2 and LPS, iii) cells that were cultured with CoCl2 only, and iv) cells that were

cultured with CoCl2 and LPS. After culture and again after staining with CD14 antibody, 150µl of

cell suspension was taken to create three cytospin slides for each time-point. Cytospin slides were

heated at 95 °C in Target Retrieval Citrate Solution pH 6.0 (DAKO, Carpentaria, USA) for 25

minutes, followed by a 20 minute cooling period. For the detection of activated caspase-3, a rabbit

anti-activated caspase-3 polyclonal antibody (Abcam) was used in combination with the DAKO

Envision kit for detection of rabbit antibody binding. The chromogen was AEC. ICC staining of

mononuclear cells for activated caspase-3 confirmed the majority of mononuclear cells were

expressing this intracytoplasmic molecule and were likely undergoing apoptosis. Uncultured cells

exhibited moderate staining intensity for activated caspase-3 (Figure 3.2A) with a slight increased

intensity noted in uncultured cells in the post-staining slide. The most intense staining was evident

in cultured samples in both the post-culture and post-staining slides (Figure 3.2B). It was concluded

that cells that had undergone cell culture had increased levels of intracellular activated caspase-3

compared to similar uncultured populations that had been thawed and kept chilled before staining. It

was concluded the cell culture conditions were therefore likely causing increased cellular apoptosis,

while permeabilisation during staining (refer to chapter 4) might further deplete cell numbers by

causing the rupture of already fragile cells. Taken together, these findings indicated that the

detection of HIF-1α in cultured monocytes using flow cytometry was currently unachievable, and

that an alternative plan was necessary.

29

Figure 3.2: A: ICC of uncultured mononuclear cells with anti-activated caspase-3. There is moderate staining intensity within both cells with abundant and scant cytoplasm (putative monocytes and lymphocytes, respectively). B: ICC of mononuclear cells cultured with stimulants before undergoing staining procedures for flow cytometry. There is marked staining intensity of cells with abundant cytoplasm (putative monocytes).

3.2.6.5 Chamber slide cultures for monocyte HIF-1α

Nunc Lab-Tek chamber slides (Nunc, Roskilde, Denmark) were selected as an alternative to

plates/tubes to study the expression of HIF-1α in bovine monocytes. With four chambers on each

slide, this layout allowed for two treated (i.e. stimulated with 200 µM CoCl2 and 5µg/ml LPS) and

two untreated negative control chambers (refer to Chapter 5, Figure 5.1).

Each chamber was seeded with 2.5 x 106 cells in 500 µl serum-free medium (with or without

stimulants) and cultured for 18 hours in 5% CO2 at 37°C. Immediately post-culture the culture fluid

containing lymphocytes was discarded and the chamber slides immersed in 10% buffered formalin

for 5 minutes to fix the remaining adherent monocytes and minimize the potential for HIF-1α

degradation within the adherent cells. The chamber slides had detachable chamber walls that were

removed after monocyte fixation, leaving a short (3mm) plastic gasket designed to neatly facilitate

subsequent ICC staining of the monocyte population. Slides were air-dried and stored in clean slide

containers that were sealed in zip-lock plastic bags and then stored at 25°C away from light.

3.3 Discussion

The work up of factors such as variable numbers of cells cultured, choice of hypoxic mimetic,

culture duration, culture vessels and post-culture harvesting processes led to a number of key

decisions regarding the cell culture process and the subsequent methods used to detect the HIF-1α

protein. It was determined that a total of 2 x 107 cells were required for the monocyte cell cultures

for ICC and 2.5 x 106 cells were required for lymphocyte cultures to be used for flow cytometry.

30

Eight cryotubes were required from each animal. This was expected to provide sufficient cells for

ICC and flow cytometry considering the range of cellular viability observed.

CoCl2 was used in subsequent test cultures at no less than 200 µM, to attempt to maximize

stimulation with the view that cytotoxicity was going to be a consistent feature with this cell

culture. The results indicate likely increased cellular apoptosis in cultures with DFO when

compared to CoCl2, but that both hypoxic mimetics result in apoptosis. These results were

consistent with a previous study that found that DFO and CoCl2 treatment of cell preparations for

HIF-1α stabilisation resulted in a dose dependent reduction in cell viability via apoptosis in

leukemic cells with DFO being the more cytotoxic (Guo et al., 2006). The observation of a dose-

dependent reduction in cell viability is consistent with another study in which human pancreatic

cancer cells were cultured with different doses of CoCl2 for 72 hours, and HIF-1α protein

expression measured using western blot. Apoptotic cells were identified by ultrastructural changes

using electron microscopy at 48-hours and by measurement of Annexin V and propidium iodide

using flow cytometry after 72-hours (Dai et al., 2012). In addition to a dose-dependent increase in

HIF-1α protein expression, this study also reported a dose-dependent increase in apoptosis evident

from flow cytometry, with the highest dose of CoCl2 at 200 µM resulting in 52.3% of cells

identified as apoptotic after a 72-hour culture.

LPS as an additional stimulant appeared to improve post-culture cellular viability and was added to

the final protocol. It is possible this effect may be due to LPS-induced stimulation of monocytes

with IL-6 production, which in turn stimulates B lymphocytes (Mangan and Wahl, 1991) although

this seems unlikely considering the relatively short culture time of 6 hours used during LPS

titration.

The poor post-culture viability that required this optimisation has been shown to be a significant

issue with frozen-thawed cells. Regrettably, due to the ongoing trouble-shooting and work-up of

protocols and the restricted nature of sample collections, the opportunity to culture fresh cells

immediately after collection did not arise. Using fresh cells remains an avenue for future

investigation into hypoxic culture of bovine leukocytes for stabilisation of HIF-1α.

An 18-hour culture was used to maximise the culture time and thus potential for HIF-1α

stabilization by the cells of the bovine immune system while still retaining adequate numbers of

viable cells for post-culture analysis. A study subjecting human glioblastoma cells to in vitro

hypoxia for 1 hour, 6 hours or 18 hours with oxygen concentrations between 2% and <0.02%,

demonstrated, using western blotting and flow cytometry, that there was a clear pattern of

31

increasing HIF-1α protein expression in the nuclear extracts of these cells (Vordermark and Brown,

2003). Another study recommended a culture duration of 24 hours when culturing mammalian cells

in 100 µM CoCl2 in 5% CO2 and 37 °C (Wu and Yotnda, 2011).This project demonstrated that

drug-related cytotoxicity in bovine leukocytes with use of hypoxic mimetics was compounded by

length of culture, however as these other studies have shown that HIF-1α stabilisation increases

with culture time, a balance needs to be achieved. There does not appear to be any published data

on previous attempts of hypoxic culture for HIF-1α in bovine leukocytes in which to refer.

The use of lidocaine to retrieve adherent monocytes was not pursued once it was realised that the

cultured monocytes were not durable enough for flow cytomtery. The use of lidocaine to harvest

monocytes from plates resulted in percentages of harvested cells comparable to those using tubes

for culture. However, the prolonged time required for the lidocaine to take effect also precluded

further use as during this time the unfixed monocytes were re-exposed to normoxia after rinsing

away the CoCl2, resulting in degradation of HIF-1α to basal levels in all cultures due to a 5-10

minute half-life (Wang et al., 1995).

Post-culture monocyte viability investigations of tube cultures indicated that monocyte apoptosis

was likely being induced by cell culture stimulation to the degree that these cells would not tolerate

the staining process for flow cytometry.

The percentage of cells identified as monocytes in the thawed uncultured cell population and the

unstimulated cultured cell population was higher than expected. The possibility that some of these

cells were large activated lymphocytes cannot be discounted, thus meaning the drop in the

proportion of monocytes after culture may have been less than was apparent from the manual

differential counts. However, the practical reality remained that hypoxic culture of this cell

population resulted in the loss of an appreciable CD14 positive population on flow cytometry,

whereas within an uncultured cell population a distinct CD14 positive subset was clearly visible,

indicating a loss of monocytes or a loss of CD14 labelling by the CD14 antibody. As CD14 is a

receptor for LPS(Kielian and Blecha, 1995), there is a possibility that LPS binding could prevent

the CD14 labelled antibody from binding to CD14 via several possible mechanisms. LPS has been

shown to bind to CD14 without the presence of LPS-binding protein (LPB) in serum-free medium,

although LPB accelerates this process(Hailman et al., 1994). A short term 1-hour culture with LPS

followed by CD14 labelling could have investigated this idea further.

The flow cytometry results and discussion of this issue are included in this chapter as these

consequences for the flow experiment fundamentally influenced the final protocol for culture; tube

cultures were therefore used exclusively for lymphocyte culture and analysis and not for monocyte

32

analysis. The 20% cellular survival rate of cells in tube cultures suggested that the reduced

medium/CO2 interface (compared to traditional plates) did not adversely affect cellular viability.

The approximately 20% post-culture survival rate was consistent, and so post-culture manual cell

counts were discontinued as another way to further minimise harvest-to-fixation interval and

preserve the HIF-1α protein.

The use of chamber slide culture followed by immediate slide fixation was chosen as the best

alternative way to assess the monocytes. Chamber slide culture and immediate fixation would

address both the post-culture monocyte apoptosis issue by temporally minimizing the opportunity

for viable monocytes to undergo apoptosis in the post-culture period, as well as minimize HIF-1α

degradation in the adherent monocytes. The monocytes would be assessed using ICC to take

advantage of the monocyte adherence to slides, the ability to quickly fix the cells and due to time

constraints of the project.

Due to the short 5-10 minute half-life and consequent rapid degradation of HIF-1α under aerobic

conditions (Wang et al., 1995), it was deemed vital to perform the initial steps of tube culture

harvest on ice and to employ use of ice-cold reagents; similar techniques were employed

successfully in another published flow cytometry protocol used to assess HIF-1α as an intrinsic

marker of tumour hypoxia (Vordermark and Brown, 2003).

The rapid degradation of HIF-1α under normoxic conditions is mediated by the ubiquitin-

proteasome system (Walshe and D'Amore, 2008) and a number of proteasome inhibitors are

commercially available to halt protein degredation by these proteasomes within a cell. A study that

tested specific proteasome inhibitors, including Lactacystin, MG-132, calpain inhibitor I and

calpain inhibitor II. Using gel shift assay measurement of nuclear extracts it found that 100 µg/ml

calpain inhibitor I, when added during the last 15 min of a 4-hour, 0.5% O2 hypoxic culture of Hep

3B cells, protected the degradation of the HIF-1 complex in cells transferred from hypoxia to

normoxia (Salceda and Caro, 1997). The use of calpain inhibitor I in the post-culture harvest

protocol was decided against because the same study cited demonstrated that the same

concentration of calpain inhibitor I will also induce stabilization of HIF-1α and formation of the

HIF-1 complex in cells under normoxic conditions and that there was no readily apparent way to

distinguish this from stabilization induced by the hypoxic mimetic agent (Salceda and Caro, 1997).

An additional reason is that proteasome inhibitors have also been shown to impair hypoxia-

dependent translocation of HIF-1α from the cytoplasm to the nucleus (Kallio et al., 1999). A

proteasome inhibitor was not used to avoid removing this valuable evidence of HIF-1α

functionality from the ICC results.

33

3.4 Conclusion – final cell culture and post-culture harvest protocol

Ten mM CoCl2 stock solution was made up fresh each time and frozen 100 µg/ml LPS aliquots,

thawed once only, were used to ensure consistent stability of both stimulants. Frozen cells were

thawed in a 37°C waterbath and washed three times in PBS. Cells were counted and resuspended in

serum-free medium at 5 x106 cells/ml. Tube cultures for lymphocyte analysis were cultured with

1.25 x106 cells in 250 µl serum-free medium suspension treated with 200 µM cobalt chloride and 5

µg/ml LPS; with a non-stimulated control tube containing only the cells suspended in serum-free

medium. Chamber slides for monocyte analysis were cultured containing 2.5 x 106 cells in 500µl

serum-free medium with or without stimulants. Tubes and chamber slides were cultured for 18

hours in 5% CO2 at 37°C, with chamber slides set up just prior to the tube cultures. At harvest,

chamber slides were harvested first as described in section 3.2.6.5. While the chamber slides were

air-drying after fixation, the tube cultures were harvested and placed straight into wet ice for 60

seconds before incubation with the anti-CD20 antibody for 20 minutes while remaining on ice

(refer to Chapter 4 for further detail). One hundred µl fixative solution Medium A was then added

to each tube and incubated for a further 20 minutes on ice. The subsequent steps are covered in

chapter 4.

34

4 Optimization of flow cytometry protocol for measuring HIF 1α expression in lymphocyte subtypes

4.1 Introduction

HIF-1α protein in leukocytes has traditionally been measured using immunoblot assays and IHC

(Yu et al., 1998, Semenza, 2005, Peyssonnaux et al., 2005) or ICC (Jantsch et al., 2008). Western

blot analysis of HIF-1α has been undertaken in cultured bovine luteal cells (Nishimura and Okuda,

2010) and corneal stromal cells (Xing and Bonanno, 2009). Flow cytometry is able to provide a

sensitive and semi-quantitative measurement of protein expression such as HIF-1α. This quality,

together with the ability to simultaneously phenotype the cell population in the same assay made

flow cytometry an attraction option for investigating the aims of this project. Phenotyping would

indicate if specific populations of leukocytes were responsible for any detected differences in HIF-

1α expression.

This chapter covers the work-up of the specific antibodies for flow cytometry that were investigated

during this project but also discusses basic concepts of antibody titration for flow and interpreting

graphical flow data as relevant. 4.2 Materials, methods and results

4.2.1 Flow cytometer maintenance and flow cytometry controls

The machine in use during this project was a Gallios flow cytometer (Beckman Coulter, Australia).

This cytometer contains a blue laser (488nm), red laser (638nm) and a violet laser (405nm) with ten

available channels (or detectors) for fluorescence detection. Given this wide range of available

excitation and emission detection capability, fluorochromes were chosen to minimise spectral

overlap.

The fluorochromes fluorescein isothiocyanate (FITC) and Dylight-488 are excited by the blue laser

with Dylight-488 having an excitation wavelength of 493nm and a emission wavelength of 518nm.

Allophycocyanin (APC), APC-Cy7 and Alexa Fluor® 647 are excited by the red laser with Alexa

Fluor® 647 having an excitation wavelength of 652nm and an emission wavelength of 668nm.

Lastly, the Pacific Blue fluorochrome is excited by the violet laser, with an excitation wavelength of

403nm and an emission wavelength of 455nm.

In a case of spectral overlap, each fluorochrome will contribute a signal to several detectors,

therefore the contribution in detectors not assigned to that fluorochrome must be subtracted fromt

the total signal in those detectors. This process is called compensation, by which the spectral

overlap between different fluorochromes is mathematically eliminated. (Baumgarth and Roederer,

2000).

35

Instrument controls are used to check the set up of the instrument, including the PMT voltage, gain

and compensation (Maecker and Trotter, 2006) The instrument controls used in this project were

Flow-check Pro beads (Beckman Coulter, Sydney, Australia) for laser alignment and the use of a

master control sample during the running of experimental samples. Cleansing and priming was also

performed between runs to avoid machine blockage. Specificity controls are used to distinguish

specific from non-specific binding, and biological controls are to provide relevant comparison

conditions for a marker stimulated in vitro (Maecker and Trotter, 2006). In this project the cells

cultured without any CoCl2 or LPS (unstimulated cells) but stained with the same antibodies served

as both a specificity and biological control for HIF-1α. Unstained samples served as specificity

controls for the titration of CD3, CD20 and CD14 antibodies as immunophenotypic markers.

Phenotypic antibody titration occasionally also incorporated an isotype control to look for any

obvious affects of non-specific staining of an antibody of a particular isotype, but there are apparent

limitations of such isotype controls to truly match the background staining of the antibody being

tested (Maecker and Trotter, 2006).

4.2.2 Cellular staining method for flow cytometry

Developing an antibody staining protocol for flow cytometry involved optimising the types,

combinations and amounts of antibody used, the length of antibody incubation, incubation

temperature and the omittance or inclusion of wash steps. With the exception of the earliest

cultures, the consistent features of the staining method following hypoxic culture were as follows.

Tube cultures (both stimulated and unstimulated) for flow cytometry were harvested from the CO2

incubator and placed straight into wet ice for 60 seconds before addition of the anti-CD20 antibody.

One hundred µl of fixative solution medium A (Life Technologies, Vic, Australia) was then added

to each tube which was vortexed and incubated on ice for 20 minutes in the dark. Tubes were half-

raised from ice and gently vortexed again at mid-incubation. Two ml of wash medium comprising

3% horse serum (Life Technologies) and 1% sodium azide (NaN3) (BDH Lab Supplies, UK) in

PBS was added to each tube, and cells were pelleted by centrifugation for five minutes at 350 g at

25°C. The supernatant was removed followed by a gentle vortex to resuspend the pellet. One

hundred µl of permeabilisation solution medium B (Life Technologies) was added to each tube

along with CD3, CD14 and HIF-1α antibodies as appropriate to the specific experiment. Each tube

was gentled vortexed and incubated at room temperature. Cells were washed and centrifuged again

as described for the medium A-step. Supernatant was removed and cells were resuspended in 500µl

of 0.1% formalin in PBS for short-term storage in the dark at 4°C. For analysis by flow cytometry,

samples were resuspended in Isoflow sheath fluid (Beckman Coulter).

36

4.2.3 Antibody titration

Flow cytometry can be broadly defined as a system of identifying cells and measuring molecules of

interest by analyzing the scatter of light that result as particles flow in a stream of liquid through a

beam of light (Givan, 2001). Both forward scatter (FS) and the side scatter (SS) of light are

measured with FS providing information on cell size and volume of an event, while SS is an

indication of internal complexity of a cell, taken together, these parameters can be used to identify

cell populations (Givan, 2001).

For each single phenotypic antibody titration discussed, aliquots of 5 x 105 cells were stained with

increasing amounts of antibody starting with neat amounts of 1µl (1/200), 2µl (1/100), 4µl (1/50),

8µl (1/25) and 10µl (1/20 dilution) 12µl (approximately 1/16) and 16µl (approximately 1/12) in a

total of 200µl cell suspension. Two aliquots of cells were fixed and permeabilised without antibody

as unstained cells. All cells used during single antibody titrations were uncultured as it was easier to

visually differentiate populations on FS/SS scatter plot of uncultured cells. A typical FS/SS scatter

plot of the uncultured test cells is shown in figure 4.1, depicting the location of signal clusters of

debris, lymphocytes, monocytes and larger events termed ‘other’ (putative aggregates of cells or

large debris). This scatter plot is consistent with a typical scatter plot of peripheral blood

mononuclear cells (Givan, 2001).

Figure 4.1: A bivariate scatter plot of the forward scatter (FS) and side scatter (SS) of the uncultured unstained test cells as visualised on a linear scale. Debris usually has low FS and SS. Lymphocytes have moderate FS but relatively low amounts of SS in contrast to monocytes. Events with high FS and SS are postulated to be aggregates of cells. To then determine if a phenotypic antibody had stained any of the cells, the stained cells were first

compared to unstained control cells. At least 10,000 events were analysed for every sample. The

37

unstained cells were analysed, with machine voltage and gain of the FS, SS and relevant

fluorochrome channel adjusted such that these unstained cells appeared on the scale in the FS/SS

scatter plot, and in the first decade of a 4-decade logarithmic scale of a histogram for the

fluorochrome being measured (Maecker and Trotter, 2006) and thus becoming the initial negative

boundary. Using the same instrument settings, the aliquots of stained cells were then analysed

starting with the lowest concentration of antibody. The univariate histogram of the fluorochrome

was assessed for the emergence of a distinct positive peak (see figure 4.2B) and the bivariate scatter

plot of SS and the fluorochrome was assessed for an emerging distinctly positive population (figure

4.2B). The instrument voltage and gain was minimally adjusted to keep the entire putative positive

and negative populations visible on the histogram, with the negative population kept to the left half

of the histogram (first two decades) where possible. After samples were run, the resulting machine

settings were saved as a protocol to be used later during the multi-antibody protocol setup.

The light signals created as FS, SS and fluorochrome fluorescence are converted by photodetectors

into electronic signals, the intensity of which is related to the intensity of the light. Photodetectors

have voltages applied so that the electronic signal is of a large enough current to be measured

(Givan, 2001). Changing the voltage is a way of decreasing or increasing the current response of

that detector. The other method to change the response to light is to alter the amplification of that

current after it leaves the detector by using linear or log amplification, with choice of the precise

gain applied (Givan, 2001). Applying discrimination to a detector limits events detected – in this

project only discrimination used was to restrict the FS so that very small events (debris and dead

cells) were excluded, and this was minimal (see Table 4.1). Once an electronic signal has been

created and amplified, the intensity of the signal is analysed and grouped (binned) by an analogue-

to-digital converter (ADC) into 1 of 1024 specific signal ranges called channels (Givan, 2001).

These 1024 equal channels are plotted along the x-axis of the histograms and scatter plots used in

this chapter. In this project, log amplification was used to look at light signals over a wide range of

intensity as this is a common amplification used for immunofluorescence studies (Givan, 2001).

The x-axes are thus divided into four log decades so that each log decade represents a quarter of

available channels. Numerical values from the flow cytometry analysis software Kaluza (Beckman

Coulter, Sydney, Australia) that are given in this chapter (and in the experimental chapter) are in

reference to these channels, with differences in fluorescence intensity expressed as a channel shift

(e.g. channel shift from 5 to 8 is channel shift of 3) rather than relating back to a linear change in

fluorescence intensity.

A list mode data file (LMD) was created for each analysis, which described each cell in the

sequence in which it passed through the cytometer laser beam (Givan, 2001) measuring the

parameters that were selected during the analysis (e.g. FS, SS, and detected fluorescence in

38

channels of interest). LMDs were then used for future computer analysis of data collected during

flow cytometry.

4.2.4 T lymphocyte CD3 antibody selection and titration

The multichain CD3 molecule is located within and the cellular membrane of normal and neoplastic

T cells (Campana et al., 1989, Janossy et al., 1989) and in virtually no other cell type, although it

does appear to be present in small amounts in Purkinje cells (Garson et al., 1982). The expression of

the epsilon chain of CD3 appears to be stable regardless of the activation and differentiation stage

of the T cells and therefore constituted the most robust immunophenotypic marker for T lineage

cells in flow cytometry and IHC. 4.2.4.1 Initial T lymphocyte CD3 antibody

Please refer to appendix (section 9.1) for further details. This antibody was not used in the final

protocol.

4.2.4.2 Alternative T lymphocyte CD3 antibody

An alternative antibody, a commercial rat anti-human CD3 IgG1 (clone CD3-12) conjugated to the

fluorochrome Alexa Fluor® 647 (MCA1477A647, Abd Serotec, Oxford UK) was researched and

selected for testing. The Alexa Fluor® 647 is excited by the red laser (638nm) and is detected in the

FL6 channel on the Gallios (see first figure). This clone has reported cross-reactivity on IHC in a

variety of species (Jones et al., 1993) and stated bovine species cross-reactivity in the manufacturer

datasheet. Subsequent to this experimental work, others have published in the use of this CD3

antibody to identify bovine T cells by flow cytometry (Yang et al., 2012).

4.2.4.2.1 IHC staining to confirm T cell positivity

Reactivity of this antibody with bovine T cells was confirmed by immunohistochemical staining of

a bovine lymph node at 1/50 dilution. This showed positive signal confined to the cells within the

paracortex, a T lymphocyte zone (data not shown).

4.2.4.2.2 CD3 antibody titration on flow cytometry

The CD3 antibody was raised against a synthetic peptide of an intracytoplasmic epitope of the

transmembrane part of the CD3ε molecule and so cellular membrane permeabilisation was required

to allow the antibody to enter the cytoplasm and bind.

Titration showed a distinct population of positive events within the ‘cells’ gate. This population

appeared at 2µl (20.5% of gated events) and was considered best visualised using 8µl of neat

39

antibody where 19.8% of the cells were positive for CD3 (Figure 4.2B). This percentage was

relatively steady within increasing amounts of antibody and was considered to represent a

consistent population of T cells.

The previously tested CD3-APC antibody was used as an isotype control. Based on titration, 8µl

was considered the most optimal amount of this antibody when used as a single stain.

Figure 4.2: A: Peripheral blood mononuclear cells - histogram of CD3 fluorescence of events with the cells gate. A peak of cells positive for CD3 can be visualised to the right of the larger negative peak. The negative linear boundary (black line) was set using unstained cells. The positive linear boundary (black line) was set using the CD3 positive gate described in 4.2B. B: Scatter plot of CD3 fluoresence versus SS, log scale. A population positive for CD3 has been gated by eye (black border). This positive population is distinct from and far to the right of the negative population.

40

4.2.5 B lymphocyte CD20 antibody selection and titration

The CD20 is a transmembrane protein expressed on mature B cells but not on plasma cells, and is a

commonly used B cell marker (Nadler et al., 1981).

4.2.5.1 Initial B lymphocyte CD20 antibody

The initial B lymphocyte antibody selected was an APC-Cy7 conjugated mouse anti-human CD20

antibody (ab82017, Abcam). Following the titration process described previously for CD3, this

antibody stained too weakly to be considered for further testing (data not shown).

4.2.5.2 Alternative B lymphocyte CD20 antibody

The next B lymphocyte antibody chosen for investigation was a mouse monoclonal IgG1 anti-

CD20, clone MEM-97 (Abcam, Cambridge, UK). This was recommended for use in flow

cytometry applications and also reported to cross-react with bovine species by the manufacturer.

4.2.5.2.1 IHC staining to confirm B cell positivity

Immunohistochemical staining at a 1/50 dilution of bovine lymph node showed good positive

signal, which was confined to the cells of follicles contained in the outer cortex (B lymphocyte

zones).

4.2.5.2.2 CD20 antibody titration on flow cytometry

Please refer to appendix (section 9.3) for details of an attempt to use a secondary antibody with this

CD20 antibody.

Given that the monocyte population was no longer going to be assessed using flow cytometry, the

Pacific Blue fluorophore (originally allocated for monocyte staining) became available for use with

the CD20 antibody instead. The CD20 antibody was conjugated to a Pacific Blue fluorochrome

using an Apex IgG Antibody Labelling Kit (Life Technologies, Vic, Australia) according to

manufacturer’s instructions. The conjugation process diluted the CD20 antibody from 1mg/ml to a

final concentration of 160µg/ml. Titration showed the best staining at 16 µl (see figure 4.3B)

however the positive population had increased to 31.6% of events within the cells gate. The positive

population peak was not yet fully separate from the negative peak on histogram (Figure 4.3A).

However when gated by eye on the CD20/SS scatter plot (Figure 4.3B) this CD20 positive

population sat within the typical lymphocyte region.

41

Figure 4.3 A: Peripheral blood mononuclear cells - histogram of CD20 fluorescence of events with the cells gate, log scale. A peak of cells positive for CD20 can be visualised however is not separate from the negative peak. The negative linear boundary (black line) was set using unstained cells. The positive linear boundary (black line) was set using the CD3 positive gate described in 4.2B. B: Scatter plot of CD20 fluorescence versus SS, log scale. A population of cells positive for CD20 is visualised and gated by eye (black border). 4.2.6 Monocyte CD14 antibody selection and titration

Please refer to appendix (section 9.3) for development details. This antibody was not used in the

final protocol.

4.2.7 HIF-1α antibody selection

4.2.7.1 Initial HIF-1α antibody

A rabbit monoclonal antibody for HIF-1α (Abcam, Cambridge, United Kingdom) was selected for

the initial development of flow cytometry methods. This antibody has been used in ICC and IHC

for human HIF-1α (Kim et al., 2010, Ameri et al., 2010) and was considered likely to react with

bovine HIF-1α due to a 95% homology of deduced amino-acid sequences between human and

bovine HIF-1α (Hara et al., 1999) and its successful use in bovine tissue protocols in other studies

(Bielefeldt-Ohmann et al., 2012). A donkey anti-rabbit polyclonal IgG antibody (ab7698 Abcam)

was selected as a secondary antibody for use in indirect staining of this unlabelled primary

antibody.

This initial work demonstrated an overall issue with non-specific binding of the chosen secondary

antibody however as some of the unstimulated samples stained only with the FITC polyclonal

antibody had fluorescent signals greater than some of the stimulated samples containing both the

HIF-1α primary antibody and the secondary FITC antibody. Bovine serum albumin was employed

42

as a blocking agent to attempt to decrease non-specific binding of the secondary FITC antibody but

a detectable pattern in FITC fluorescence failed to emerge. The use of a secondary antibody was

concluded to be too problematic, especially for labelling an induced molecule of interest with no

positive control available. 4.2.7.1.1 Initial hypoxic cultures for testing use of initial HIF-1α antibody on flow cytometry

Initial hypoxic cell cultures using CoCl2 and LPS for testing of different dilutions of HIF-1α

primary antibody and secondary FITC antibody were of 4-hour, 6-hour and 8-hour duration (data

not shown). These cultures were performed to see if stabilization of HIF-1α could be detected using

flow cytometry after a short-term culture as described in one instance in literature (Vordermark and

Brown, 2003).

4.2.7.2 Alternative HIF-1α antibody

A mouse IgG1 monoclonal HIF-1α antibody (ESEE122 clone) commercially conjugated to Dylight

488 (Novus Biologicals, Littleton, CO, USA) was chosen for alternative investigation. The neat

concentration of this antibody was 1mg/ml. This HIF-1α antibody had stated cross-reactivity with

bovine cells by the manufacturer and was also listed for use in many immunostaining procedures

such as IHC, ICC and immunofluorescence, with a recommended dilution range of 1/100 to 1/5000.

No specific references of the use of this antibody in bovines were available but the manufacturer

guaranteed the reactivity. Dylight-488 is spectrally similar to FITC and so was not expected to

cause any issues of spectral overlap.

4.2.7.2.1 Hypoxic cultures for testing of alternative HIF-α antibody on flow cytometry

Two sessions of 18-hour hypoxic cell culture were undertaken using established methods (detailed

in chapter 3) including placing cells on ice post-culture with the fixation of cells in the presence of

CoCl2. The two sessions compared staining of stimulated and unstimulated cultures with HIF-1α

antibody, initially at 1µg/ml concentration and 2µg/ml dilution, as well as unstained samples of

stimulated and unstimulated cells. For each session HIF-1α antibody was incubated for a maximum

time of three hours. Every 30 minutes the incubation of an aliquot of cells was stopped and the

staining process finished for those cells in order to assess any affect of antibody incubation time on

subsequent HIF-1α antibody fluorescence measured on flow cytometry.

With increasing incubation time, both 1µg/ml and 2µg/ml staining showed a small increase in mean

HIF-1α antibody fluorescence in unstimulated cells while cells stimulated by hypoxic cell culture

showed very little increase in mean fluorescence (Figure 4.1 and 4.2). Unstained stimulated cells

had a mean fluorescence of 0.54 and a standard deviation of 0.32, and unstained, unstimulated cells

43

had a mean fluorescence of 0.82 with a standard deviation of 0.58. The mean fluorescence of the

unstained cells was therefore, quite close to stimulated stained cells which might indicate a lack of

antibody binding. However, the standard deviation of stained samples (stimulated or unstimulated)

was at least 10.5, demonstrating a broader variability of HIF-1α antibody fluorescence and the

possibility that some staining had taken place.

Figure 4.4: Graph of the mean fluorescence of HIF-1α antibody at 1µg/ml with different incubation lengths, unstimulated and stimulated cells. There is a small increase in mean fluorescence in unstimulated cells, while stimulated cell mean fluorescence largely remains the same.

0.00  

0.50  

1.00  

1.50  

2.00  

2.50  

3.00  

0.5   1   1.5   2   2.5   3  

Mean  flu

orescence  (cha

nnel)  

Incuba2on  2me  (hours)  

HIF-­‐1α  an2body  staining  1µg/ml          

 Uns+mulated  cells    

Cells  s+mulated  by  hypoxic  culture                

44

Figure 4.5: Graph of the mean fluorescence of HIF-1α antibody at 2µg/ml with different incubation lengths, unstimulated and stimulated cells. There is a small increase in mean fluorescence in unstimulated cells, while stimulated cell mean fluorescence largely remains the same. Examination of scatter plots of HIF-1α fluorescence versus SS from all incubation lengths showed

the emergence of two distinct populations in the stimulated cells, one being a larger population with

higher SS and the other a smaller population with lower SS (Figure 4.8B). These populations would

consistently appear with stimulation but were not discernable in unstimulated cells (Figure 4.8A). It

was concluded that the channel shift of mean fluorescence of HIF-1α of only 1-2 channels, when

comparing unstimulated to stimulated cells, was a small change, and a doubtful result in terms of

real meaning.

There was, however, a subtle but consistent shift in cell population SS distribution with hypoxic cell

culture. As this subtle shift was the only repeatable response detected during flow method work-up,

and further work-up was not possible during the remaining project time-frame, it was decided to

finish the multicolour flow cytomtery protocol by combining the phenotypic antibodies with the

HIF-1α antibody for use on experimental samples, in the hopes that further information may be

gleaned from looking at cells from other individuals. No further work-up was performed, and it was

decided to use HIF-1α antibody at concentration of 2µg/ml as this equated to approximately 1/500

(the middle of the range recommended by the manufacturer).

0.00  

0.50  

1.00  

1.50  

2.00  

2.50  

3.00  

3.50  

0.5   1   1.5   2   2.5   3  

Mean  flu

orescence    (cha

nnel)  

Incuba2on  2me  (hours)  

HIF-­‐1α  an2body  staining  2µg/ml        

Uns+mulated  cells      

Cells  s+mulated  by  hypoxic  culture              

45

Figure 4.6: A. Scatter plot of HIF-1α fluorescence versus SS, log scale. Unstimulated cells were stained with 2µg/ml HIF-1α antibody for three hours. A single population with one central red (i.e. high density) focus is visible. B. Scatter plot of HIF-1α fluorescence versus SS, log scale. Stimulated cells were stained with 2µg/ml HIF-1α antibody for three hours. The mean HIF-1α fluorescence was consistently lower in stimulated cells however stimulation caused a repeatable change in population shape with respect to HIF-1α and SS. A larger population with greater SS is now visible (upper large red foci) situated above a smaller population with lower SS (lower red foci). 4.2.8 Triple staining of cultured cells for flow cytometry

The process of adding antibodies to create a triple-stained sample was informed by empirical

knowledge gained from the single antibody titrations. Samples of stimulated and unstimulated cells

were cultured using the established method and then stained with a combination of 8 µl CD3, 16 µl

CD20 and HIF-1α at 2µg/ml dilution. These volumes were considered optimal based on single

antibody titration results (see previous sections), with additional samples stained singly with each

antibody and unstained samples.

The voltage, gain and discrimination values for channels FL1, FL6 and FL9 were set up based on

the values determined during single antibody titration of HIF-1α , CD3 and CD20, respectively (see

Table 4.2). These values were minimally adjusted during triple-stained sample analysis to ensure all

positively stained populations detected by each channel were on the scale covered by the

histograms and scatter plots (see table 4.2 below for original and adjusted values).

The fluorochrome combination used in the development of this multicolour experiment was chosen

in an attempt to minimize spectral overlap and the need for compensation. After removal of CD14

from the flow cytometry protocol, the fluorochromes that were ultimately used in the developed

protocol were Dylight-488 (for HIF-1α) Pacific Blue (B cell marker) and Alexa Fluor® 647 (T cell

marker). The excitation and emission wavelengths for these fluorochromes are listed in the table

46

below, as well as the signal channel used for detection, the voltage, gain and discrimination values

determined by single antibody titrations and the adjusted values after triple-stained samples were

analysed.

Table 4.1: Detectors used for detection of cell characteristics and the cytometer parameters for each detector output as set using single and then triple-stained cells. Detector Detected cell

characteristic Labelled antibody

Single voltage

Triple voltage

Single gain

Triple gain

Descrim Comp

FS Size n/a 72 72 2 2 8 No SS Complexity n/a 80 80 2 2 OFF No FL1 Dylight-488

fluorescence HIF-1α 400 420 1 1 OFF 2%

with FL9

FL6 Alexor Fluor 647 fluorescence

CD3 477 519 2 2 OFF No

FL9 Pacific Blue fluorescence

CD20 410 424 2 2 OFF 2% with FL1

Using this triple stain in stimulated cells, the T cell population had a mean fluorescence (X-mean)

of 24.61, and the B cell population had a X-mean of 13.64. Both positive populations were visible

graphically (Figure 4.4).

In a repeat experiment of the culture and triple-stain flow analysis lower amounts of CD3 and CD20

antibody were used. Triple-staining with 4µl CD3, 10µl CD20, but with identical HIF-1α antibody

concentrations, gave similar resolution of the T and B cell populations, with the X-mean of the T

cell population mean fluorescence at 23.22 and the B cell at 15.37. The unstimulated cells had a

mean HIF-1α fluorescence of 4.27 and stimulated cells had a mean HIF-1α fluorescence of 5.86.

These final amounts of antibody were used as standard for the rest of the project. Instrument

settings were saved as a triple-stain protocol for use with experimental samples.

47

Figure 4.7: Scatter plot of CD3 fluorescence and CD20 fluorescence, triple-stained stimulated cells, log scale. B and T cell populations are gated (solid black borders) with percentages of gated cells that are CD3 positive, CD20 positive or positive for both phenotypic markers.

4.2.8.1 Master Control samples

Analysis of the experimental samples (see chapter 6) was expected to require more than one session

of flow cytometry. Standardisation of such longitudinal flow studies can be facilitated by use of a

known control cell population, aliquots of which can be thawed, stained and run with each session

(Maecker and Trotter, 2006). In this case, a ‘master control’ population was created based on advice

from a consultant flow cytometrist Dr Nana Satake (personal communication, August 2011).

Five dairy cows residing at the university campus were selected based on their clinical health and

availability to donate 450ml of blood for this project. Three of the animals were clinically healthy at

the time of blood collection, while the remaining two animals had pyrexia at time of sampling and

considered immunologically challenged due to recent surgery. These animals were unfortunately

not beef cattle but their variation in health status was an attempt to account for any phenotypic

variation in the leukocytes of experimental samples due to immune status. Peripheral blood

mononuclear cells were isolated from each animal using methods previously described (see chapter

3). Aliquots of cells from each individual were cultured in both stimulated and unstimulated

conditions as previously described (see chapter 3).

The master control cells varied only slightly from the established staining protocol in that equal

aliquots of cultured cells from all five animals were mixed to create the master control population.

Immediately post-culture the samples were put on ice, followed by rapid combination and vortexing

of stimulated cells from all five animals and 250µl aliquots (representing a mix of cells from the 5

animals combined) were removed as master control samples. This process was repeated for the

unstimulated samples. Staining then proceeded as per the established protocol. Initially, master

control samples were stained with each antibody singly, in addition to a triple-stained sample and

48

an unstained sample. Single-antibody samples were included to facilitate troubleshooting of any

issues arising in the triple-stained samples of this new population being tested, however no obvious

issues arose and subsequent batches of master control cells were triple-stained only (in addition to

the usual unstained samples). All master control samples were analysed using the same triple-stain

flow protocol established for use in the experimental samples (see chapter 6).

4.3 Discussion

Both the CD3 and CD20 antibodies showed positive staining on IHC, and with distinct positive

populations on flow cytometry were considered successful titrations.

HIF-1α expression was measured by the detection of Dylight-488 fluorescence by the flow

cytometer and post-collection data analysis software was used to investigate the presence of

statistically significant trends or differences of HIF-1α across and between sample populations and

between different leukocyte phenotypes. Flow cytometry is not commonly used to measure HIF-1α

due the rapid degradation of the molecules under aerobic conditions, however use of another

technique such as western blotting would remove the phenotypic data gained through flow

cytometry.

The attempt to use a secondary FITC antibody for the HIF-1α primary antibody during flow

cytometry method development yielded confusing results that did not resolve using further

troubleshooting procedures. In this case, the use of 2% horse serum or foetal bovine serum could

have been used to block Fc receptors in an attempt to decreased non-specific binding of the

secondary FITC antibody.

It was decided that secondary antibodies were too problematic and were not used for further method

development. Indirect staining can increase fluorescence intensity as the usually polyclonal

secondary antibody binds to multiple site on the primary antibody, enhancing what might be

otherwise weak fluorescent signal (Givan, 2001). However the use of a secondary antibody

increases the time needed to complete staining and requires the use of extra controls (Givan, 2001).

A relevant disadvantage of indirect staining is that it increases the potential for non-specific staining

(Givan, 2001), which is theorised to have occurred with the secondary labelling of the initial HIF-

1α primary antibody. Indirect staining is also problematic in a multicolour experiment due to

potential cross-reactivity between other primary antibodies and the secondary antibody (Givan,

2001).

To take advantage of the amplification of fluorescent signal by an indirect method, a primary

antibody may be conjugated to biotin, followed by staining with avidin or streptavidin conjugated to

a fluorochrome (Givan, 2001) which provides a specific binding reaction. A biotinylated primary

49

HIF-1α antibody could have been considered in further investigation of a potentially weak HIF-1α

signal.

4.4 Conclusion - Final post-culture staining and flow protocol for 3-colour flow cytometry of T lymphocytes, B lymphocytes and HIF-1α

Immediately following removal from the incubator, tube cultures were placed on ice and allowed to

sit for 60 seconds to facilitate culture medium cooling before antibody incubation. After 60 seconds

the lymphocytes were incubated on ice for 20 minutes with 10 µl mouse IgG1 monoclonal antibody

to CD20 (MEM-97, Abcam, UK), which had previously undergone covalent conjugation to a

Pacific Blue fluorophore using an Apex IgG Antibody Labelling Kit. Tubes were half-raised from

ice and gently vortexed manually at the start and again at mid-incubation.

All tubes were treated with 100 µl of Medium A fixative solution, vortexed manually and allowed

to fix on ice for 20 minutes, with 100 µl fixative adequate for fixation of 1 x 106 cells (total cell

counts at this point during test tube cultures were approximately 2.5 x 105 cells). The CD20

antibody, fixative and CoCl2 were then washed from cells by addition of 3ml of flow wash

composed of 0.1% NaN3 and 3% horse serum in PBS, then centrifuged at 350 g for 5 minutes at

25°C. This temperature allowed cells time to warm in preparation for permeabilisation. The

supernatant was carefully aspirated to 250 µl and the tube gently vortexed to resuspend the cellular

pellet.

All tubes were treated with 100 µl Medium B Permeabilisation medium (Life Technologies, Vic,

Australia) and gently vortexed, then immediately incubated with 4 µl of rat anti-human IgG1

monoclonal CD3 antibody (CD3-12 clone) which was commercially conjugated to Alexa Fluor®

647 and 2 µg/ml of mouse IgG1 monoclonal HIF-1α antibody (ESEE122 clone) commercially

conjugated to Dylight 488. Cells were incubated at room temperature in the dark for 3 hours, then

were again washed, centrifuged and the resultant supernatant aspirated and the pellet resuspended

as previously described. Cells were analysed by flow cytometry using the triple-stain flow protocol

within 2 hours of staining, with refrigeration at 4°C in the dark during this time.

50

5 Development of a protocol to detect HIF-1α expression in monocytes

5.1 Introduction

Flow cytometry was found unsuitable for detecting HIF-1α expression in monocytes using the

developed cell culture protocol. An alternative technique to measure HIF-1α in monocytes was

therefore necessary.

There are a number of approaches to detection and measurement of HIF-1α stabilisation and

subsequent HIF-1 activity within cells, with several of these considered before the final technique

was chosen. HIF-1α protein levels in cell extracts are commonly detected using western blot (Wang

et al., 1995, Jiang et al., 1996, Carmeliet et al., 1998, Jin et al., 2011). An ELISA has also been

developed for measurement of HIF-1α in cultured tumour cell lines (Formento et al., 2005).

Methods have been developed to measure HIF-1 activity by the introduction of transgenes with

hypoxic response elements as promoter sequences coupled to reporter genes (Shibata et al., 2000).

DNA-binding activity of HIF-1 has been measured using electrophoretic mobility shift assays

(Wang and Semenza, 1993b). Expression of HIF-1α mRNA can be detected using quantitative

reverse transcription PCR (Dai et al., 2007) and northern blot (Wiener et al., 1996) as HIF-1α

mRNA has been shown to be upregulated under hypoxic conditions (see chapter 1). HIF-1

transcription activation can also be detected by measurement of downstream gene products such as

VEGF using a variety of the above methods.

HIF-1α expression within cells is also routinely detected using immunolabelling techniques such as

ICC and IHC (Koukourakis et al., 2002, Jin et al., 2011). ICC is the demonstration within individual

cells of a cellular constituent in situ by detecting specific antibody-antigen interactions where the

antibody has been tagged with a visible label (Sternberger, 1979), while IHC performs this

demonstration within whole tissues. ICC using direct labelling is the single layer application of a

labelled antibody to the cell antigen, however, this is less common due to insensitivity and the

requirement for directly-labelled antibodies (Noorden, 2000) and for these reasons was not used in

this project. Indirect labelling is more common and involves application of a primary unlabelled

antibody followed by a labelled secondary antibody raised to the immunoglobulin of the species

providing the primary antibody. Indirect labelling is more sensitive because this allows

accumulation of label at the original reaction site (Noorden, 2000). Commonly used labels include

fluorescent compounds, enzymes and colloidal gold (Noorden, 2000). An enzyme label is an

attached enzyme complex such as horseradish peroxidase (HRP) which activates a subsequently

applied chromagen substrate, producing a coloured reaction at the antigen site that is visible with

light microscopy (Fetsch, 2005).

51

5.2 Materials, methods and results

5.2.1 HIF-1α antibody selection for ICC

A rabbit monoclonal antibody for HIF-1α (Abcam, Cambridge, United Kingdom) was selected.

This antibody has been used for ICC and IHC for human HIF-1α (Kim et al., 2010, Ameri et al.,

2010) and was considered likely to react with bovine HIF-1α due to a 95% homology of deduced

amino-acid sequences between human and bovine HIF-1α (Hara et al., 1999). This antibody has

worked in several different test-modes including IHC and Western Blot, and has been used

previously for IHC on bovine tissue (Bielefeldt-Ohmann et al., 2012). A previous study using

another rabbit polyclonal HIF-1α antibody featuring the same immunogen and human reactivity has

also shown reactivity with bovine HIF-1α (Bielefeldt-Ohmann et al., 2008).

5.2.2 ICC protocol for HIF-1α expression in monocytes

5.2.2.1 Fixation of monocytes

Monocytes fixed in 10% buffered formalin appeared to have overall better cell morphology than

samples fixed in acetone when several slides using each of these fixative solutions were examined

under light microscopy. As previously discussed in chapter 4, the cultured monocytes were

subjected to ICC in-situ after culture. The immunocytochemical protocol was based on use of the

EnVision+ System-HRP rabbit kit (DAKO, Carpentaria, USA).

5.2.2.2 Target antigen retrieval, anti-cross-contamination measures and blocking steps

Prior to ICC, slides were examined by light microscopy to confirm the presence of cells.

The formalin-fixed slides were then immersed in Target Retrieval Solution pH 9.0 (DAKO) at 95°C

for 25 minutes, for target antigen retrieval. Target retrieval using citrate pH 6.0 (DAKO) and

proteinase K Ready-to-use (DAKO) were also tested but resulted in no detectable signal. Slides

were allowed to cool for 20 minutes while still immersed in target retrieval solution, then were

thoroughly rinsed with deionised water.

During harvest and monocyte fixation the detachable walls were removed from the chamber slides.

Doing this without disturbing the gasket for ICC proved, in reality, very difficult to achieve

consistently, even when following the instructions for removal. Using test slides it was quickly

discovered that although a gasket appeared well adhered to the slide, application of a stained

solution (e.g. Methyl Green) to an individual well indicated that occasional points of gasket ‘micro-

detachments’ existed (invisible to gross inspection) that allowed reagents to diffuse from one well

into an adjacent well. Gaskets were subsequently discarded from every slide and each of the 4 wells

per slide was carefully drawn around with a delimiting pen (DAKO) creating a water-repellent

52

barrier between wells. During this process slides were immersed or covered in deionised water to

prevent drying out.

Slides were then treated with 0.35% hydrogen peroxide in water-solution for 15 minutes to block

endogenous peroxidase activity and rinsed with a solution of Tris-buffered-saline with Tween 20

(TBST) at a concentration of 0.05 mol/L Tris-HCl and 0.3 mol/L NaCl, with 0.1% Tween 20 and

0.01% preservative (DAKO). Slides were then treated with glycine (glycine binds to and blocks

aldehydes left over from the fixation step) for 15 minutes, and then rinsed briefly with TBST.

5.2.2.3 Primary HIF-1α antibody application

Antibody Diluent solution (DAKO) was applied for 30 minutes. Excess antibody diluent was then

tapped off, followed by application of 150µl of the primary anti-HIF-1α antibody at a titrated 1/100

dilution (diluted in the same Antibody Diluent solution) to wells 3 and 4 only (well 3 containing

cells from the unstimulated culture and well 4 containing cells stimulated by CoCl2 – see Figure

5.1). The primary antibody was incubated for 2 hours. The cell populations of wells 1 and 2

mirrored the populations in wells 3 and 4 (Figure 5.1) but did not receive primary HIF-1α to serve

as negative primary controls for detection of any non-specific staining of the secondary antibody.

These upper 2 wells were deliberately assigned as secondary controls to avoid any contact with

primary antibody during vertical rinsing steps (direction of rinse indicated by the arrow in Figure

5.1). After primary antibody incubation, all wells were rinsed 5 times with TBST.

Figure 5.1: Post-culture in situ ICC of adherent monocytes. Secondary controls Wells 1 and 2 contained unstimulated and stimulated cells, respectively, and did not receive primary antibody. The lower wells 3 and 4 contained duplicate cell populations to 1 and 2 and were treated with primary HIF-1α antibody.

53

5.2.2.4 Secondary antibody application

Real Envision HRP goat anti-rabbit immunoglobulin (DAKO) was applied to all wells including

controls, and incubated for 30 minutes followed by repeated rinsing with TBST. Slides were

incubated with 3-amino-9-ethylcarbazole (AEC+) high sensitivity, ready-to-use substrate

chromagen (DAKO) for 15 minutes duration for development of a detectable signal.

3,3'-Diaminobenzidine (DAB) diluted 1:50 in substrate buffer (DAKO) was tested as a possible

substrate chromagen (due to a greater stability than AEC+), however there were consistently higher

levels of background staining and stain precipitate which interfered with production of a distinct

positive signal. Excess substrate was gently tapped off, immediately followed by application of a

methyl green counterstain for 1 minute before a final rinse with deionised water. Methyl green

consistently appeared to contrast better with AEC than Mayer’s Haematoxylin as it improved

visibility of any HIF-1α signal within the nucleus.

5.2.2.5 Slide mounting and slide examination

Slides were quickly coverslipped with Faramount aqueous mounting medium (DAKO) and left to

dry overnight before examination by light microscopy. The primary controls showed no staining

that would have indicated non-specific binding of the secondary antibody (results not shown) and

background staining was minimal.

5.3 Discussion

The primary HIF-1α antibody used has been previously used in IHC on bovine cells (Bielefeldt-

Ohmann et al., 2012) and has been previously successfully implemented using isotype controls (H.

Bielefeldt-Ohmann, personal communication, November 2012). Other controls to consider for

future work include cells from a knock-out animal for HIF- 1α stained with the primary HIF-1α

antibody, on an absorption control which involves the incubation of the primary antibody with the

antigen used to generate the antibody such that the absorbed antibody can no longer bind to the

antigen present on the slide (Burry, 2011). Given more time and funds, the work-up of such

additional absorption controls for this antibody for use in ICC would further help strengthen the

results of this protocol.

The major issue encountered with this protocol was unreliability of the slide gaskets, requiring

extensive use of a delimiter pen to recreate water-proof barriers between the cells. Although time-

consuming, creation of this barrier was vital to ensure no cross contamination of reagents, given the

close physical proximity of the positive and control wells, while facilitating uniform ICC staining

and a reduction in required reagent volume.

54

The most apparent disadvantages of ICC are that this method produces what can be still only be

considered semi-quantitative results, even when performed with the application of morphometric

measurements (Oberholzer et al., 1987, Hershkovitz et al., 2013). The method can provide

information as to the presence of absence of HIF-1α, but can be argued as relatively insensitive for

distinguishing between any subtle differences in the level of HIF-1α expression between different

animals or culture conditions. There is potential for batch variation when performing the same ICC

multiple times and proper assessment relies on the experience of the slide reader. However using

ICC allows the direct identification and examination of cells expressing HIF-1α. In the context of

this project, this technique was considered additionally useful for specifically detecting intranuclear

HIF-1α due to nuclear translocation occurring subsequent to intracytoplasmic stabilisation in

response to hypoxia. Positive intranuclear HIF-1α can be interpreted as evidence of a more

advanced HIF-1α response occurring in these cells compared to negative or cells with

intracytoplasmic staining only (see section1.2.2.1). ICC was considered a good choice for this

project given its common use for HIF-1α, the prior experience gained in immunostaining (for flow

cytometry) during this project, easy access to indirect enzymatic commercial ICC kits, and a readily

available ICC protocol provided by H. Bielefeldt-Ohmann (personal communication, June 2010). In

addition to these reasons, ICC was ultimately chosen over other potential methods due to prior ICC

experience of this researcher, as a practical way of addressing remaining time-constraints of the

project. Western blot was also considered as an alternative method due to available supervisor

expertise and facilities. Western blot was not used as the large numbers of cells required for

adequate protein yield would be problematic due to the low total cell numbers in many of the

intended experimental samples (see 6.2.1), as well as the low retrievable cell population proportion

and low total cell counts of harvested hypoxic cultures.

5.4 Conclusion

A protocol for detecting HIF-1α in bovine monocytes was successfully developed as an alternative

to measurement of HIF-1α in monocytes using flow cytometry.

55

6 Methodology application to assess HIF-1α expression as a biomarker of bovine respiratory disease

6.1 Introduction

BRD is a multi-factorial inflammatory respiratory disease complex. Due to the high incidence of

pneumonia in cattle, investigations are currently focused on ways to enhance effective and non-

injurious immune responses to these pathogens (Ackermann et al., 2010a), in addition to addressing

the environmental and managerial issues that may predispose cattle to respiratory disease. A way to

enhance effective immune responses is to identify candidate molecules that are expressed

differently between animals that have an effective response to an infectious challenge and other

animals that develop disease. A consistent difference in expression means this molecule could be

involved in and also act as biomarker for a particular immune response and disease outcome. For

BRD, identification of a biomarker would allow potential targeted breeding programs to enhance

resistance to this complex disease.

There have been previous studies investigating possible biomarkers for BRD outcome (Aich et al.,

2009, Eitam et al.). Aich et al 2009 established the value of using combinatorial “omics”

approaches to identify candidate biomarkers and used multimodal analysis of these biomarkers to

predict disease outcome. Eitam et al 2010 aimed to detect individual variations in the stress

responses of 13 newly received young calves through measurement of their leukocyte heat shock

protein response, selected neutrophil-related gene expression and oxidative stress, and relate these

variations to pulmonary adhesions at slaughter as an indicative sign of clinical and subclinical

episodes of BRD at an early age.

The increased production of proinflammatory cytokines that occurs during a primary BHV-1

infection (Bielefeldt Ohmann et al., 1991) suggests a potential role of a HIF-1α positive feedback

loop in the subsequent development of BRD. Alternatively, a sluggish HIF-1α response could

decrease the hypoxia-triggered HIF-1α adenosinergic suppression of proinflammatory cytokines in

CD4+ and CD8+ lymphocytes, an important immunosuppressive mechanism which down-regulates

the immune response to protect local tissue and maximize anti-pathogen cellular response during

pathogen encounter (Sitkovsky and Lukashev, 2005, Zinkernagel et al., 2007, Sitkovsky, 2009).

If an association between HIF-1α levels and case/control status existed and this association was also

present prior to cases actually developing BRD, it might be possible to use HIF-1α as a biomarker

for BRD susceptibility. This information could be used to identify susceptible animals prior to

56

developing clinical disease and assist in the creation of targeted breeding programs to reduce the

incidence of BRD.

The aims of this study were to investigate the value of measuring the HIF-1α expression response

as a predictive biomarker for the outcome of BRD by implementing developed methodologies

experimentally to compare the HIF-1α expression response in a group of clinically healthy beef

feedlot cattle with the HIF-1α expression in feedlot cattle that developed and were treated for

clinical respiratory disease while in the feedlot. The application also tested how practical and user-

friendly the developed methodology is in practice.

6.2 Materials and methods

6.2.1 Study population and animal selection

This study was conducted on two small subsets of animals enrolled in a parent project aimed at

identifying and quantifying risk factors for BRD in feedlot cattle. Blood samples were collected

from cattle in the parent project on the day of induction at the feedlot and after approximately 42

days on feed, induction being the process of tagging, weighing, treating and entering their data

(animal identifiers) into the feedlot computer system. Subset one was selected prospectively from

November 2011 to March 2012. Animals eligible for inclusion were slaughtered at a combined

feedlot/abattoir facility in South East Queensland from November 2011 to March 2012, and came

from five cohorts (pens of cattle assembled at induction to the feedlot and kept together during time

on feed) from two of the 14 feedlots participating in the parent study. Using the following methods,

individual animals were then selected from the eligible cohorts. Cattle slaughtered during the study

period were eligible for selection as cases if they had been pulled from the cohort and treated for

respiratory disease by feedlot staff (treatment records sourced from feedlot database). In contrast,

cattle were eligible for selection as controls if they had not been pulled and treated for respiratory

disease whilst on feed. Eligibility for inclusion also depended on blood samples taken at both

induction and approximately 42 days on feed being of sufficient quality and quantity for serological

analyses.

As a maximum of 15 samples could be processed on any single collection day, controls were

selected from the eligible list using numbers randomly generated by Stata 12.1 (StataCorp LP,

Texas, USA). For most cohorts the selection process used for controls was also applied to cases,

although some cohorts had low case numbers so all eligible cases were sampled. The overall

case:control ratio in subset one was approximately 1:2, although within individual cohorts this ratio

varied considerably due to the variable number of case animals in some cohorts. The overall sample

57

size was small due to time, budget and labour constraints. The project was considered to be a pilot

project for these reasons.

The second subset (subset two) comprised 42 animals previously sampled by another researcher

also working within the parent project (Mr. Andrew Ferguson, School of Veterinary Science, UQ,

Australia). These cattle were slaughtered at the same facility from May to July 2010 and were

sourced from one of the two feedlots included in the prospective sampling. Cattle were eligible as

cases if they had been pulled and treated for respiratory disease, and showed an increase in

serological status of 1 or more to BHV-1, BRSV, BVDV or PI3 virus. Controls were animals that

were not pulled but showed an increase in serological status of 1 or more to BHV-1, BRSV, BVDV

or PI3 virus. Cases in subset two varied from the cases in subset one in that this serological status

information was part of case eligibility. Samples from subset two contained fewer cells and were

only used in the ICC analyses.

6.2.2 Baseline data

Baseline data for each selected animal in both subsets was sourced from participating feedlots to

check for possible confounding variables. These data included each animal’s cohort, breed, gender,

dentition and induction weight. Serology results, using the BIO-K 284 multiplex-ELISA (Bio-X

Diagnostics) for antibodies to BHV-1, BRSV, BVDV, PI3 virus and M. bovis for induction and day

42 samples, were also made available by the parent project. Serological results were expressed on a

0-5 scale, with higher numbers indicative of higher antibody titre.

Within each subset of animals, Fisher’s exact test was used to compare the distribution of

categorical variables between case and control populations and the Mann-Whitney U test was used

to compare induction weights, as the distribution was not normal.

6.2.3 Blood sample collection

Blood samples were obtained from selected cattle immediately post mortem. Post mortem

collection was used due to the large amount of blood required and the need to avoid additional

stress to the animals in the immediate pre-slaughter period. Blood collection was undertaken by

trained abattoir workers under the direct supervision of the primary researcher. Approximately 500

ml of whole blood was collected via gravity from each animal within two minutes of initial jugular

cut; blood samples were thus non-sterile. Blood was collected into sterile, pre-chilled 500 ml

containers (Interpath, Victoria, Australia) containing 50 ml of 6 % sodium citrate tribasic dihydrate

58

(Sigma Alrich, St Louis MO, USA) in deionised, sterile water. Samples were immediately chilled

and transported for 1.5 hours to the lab for processing to purify mononuclear leukocytes.

6.2.4 Mononuclear cell isolation from experimental samples

Refer to section 3.2.2 for this previously described protocol. This procedure was only undertaken

within the confines of this project on the 46 experimental samples of subset one.

6.2.5 Cell culture and post-culture harvest of experimental samples

Refer to section 3.4 for this previously described protocol. Stimulated cultures contained CoCl2 and

LPS and unstimulated cultures did not. A stimulated and unstimulated tube culture was set up for

each animal from subset one. Chamber slide culture was performed on all experimental samples (88

individual animals), with two chamber slides containing stimulated and stimulated wells cultured

for each animal (see Figure 5.1 in chapter 5).

6.2.6 Flow cytometry protocol

Refer to Chapter 4 for previously described protocols for cell staining and flow cytometry.

A stimulated and unstimulated sample was run for each animal. At the start of each session, the

cytometer was calibrated using fluorescent beads as sold and recommended by the manufacturer.

A ‘master control’ sample comprised of leukocytes isolated from five different adult dairy cows

located on the UQ Gatton Campus was also run during each of the flow cytometry sessions to detect

any technical machine-related problems or other session-specific issues causing differences that

would have an impact on the ability to compare results from different sessions during flow

cytometry studies. The use of cells from multiple individuals that were either clinically well or

showing clinical disease was to cover the anticipated phenotypic range of immune cells to be

encountered in the experiment (refer to Chapter 4 for further details). These five animals were

selected based on their clinical health and availability to donate 450ml of blood. Three of the

animals were clinically healthy at the time of blood collection, while the remaining two animals had

pyrexia at time of sampling and considered immunologically challenged due to recent surgery.

6.2.6.1 Gating of cell populations

Analysis software (Kaluza, Beckman Coulter) was used to gate cell populations for stimulated and

unstimulated samples from each animal. The first gate (“Cells”) was applied to the forward

scatter/side scatter plot to select the cell population (Figure 6.1A), excluding debris, stain

precipitate and large events (presumed aggregates of cells). Only events included in the “Cells” gate

were included in subsequent gating procedures. The next scatter plot was generated with the axes

59

CD 3 (T lymphocyte) versus CD20 (B lymphocyte) fluorescent signal (Figure 6.1B). The labelled

“B lymphocytes” and “T lymphocytes” populations were gated individually. For each of the gated

B and T lymphocyte populations, expression of HIF-1α fluorescent signal was then visualized as a

histogram (Figure 6.1C).

Figure 6.1 A: An example of the forward scatter/side scatter plot obtained from Kaluza. Events included in cell population lie within the gate “Cells” comprising a solid black line. This gate excludes debris and aggregate cells. B: An example of a T lymphocyte fluorescent marker/B lymphocyte fluorescent marker scatter plot obtained from Kaluza, based on the events included in the “Cells” gate. “T cells” and “B cells” populations are both gated (solid black lines). C: An example of a histogram of HIF-1α fluorescence detected within the T cell gate on flow cytometry with Kaluza-generated summary statistics. 6.2.6.2 Flow data analysis methods

The difference in HIF-1α fluorescence between stimulated and unstimulated samples in case animal

T and B lymphocyte populations was compared with that of the controls. For each sample

60

(stimulated and unstimulated for each animal) the HIF-1α signal from the B lymphocyte population

and the T lymphocyte population was summarized by the sample mean, median and standard

deviation as provided by the Kaluza flow cytometry analysis software. These summary statistics

were selected to assess central tendency and spread of the HIF-1α fluorescence of each cell

population, and were exported from Kaluza into Stata 12.1 for further analysis. The difference

between the stimulated and unstimulated values was then calculated for each of these summary

statistics for each animal. This resulted in 18 summary statistics for each animal (stimulated B cell

mean, median and standard deviation, unstimulated B cell mean, median and standard deviation,

difference between stimulated B cell and unstimulated B cell mean, median and standard deviation,

stimulated T cell mean, median and standard deviation, unstimulated T cell mean, median and

standard deviation, difference between stimulated T cell and unstimulated T cell mean, median and

standard deviation). The animals were grouped by case/control status and the group mean, median

and standard deviation of each of these summary statistics was calculated for all the stimulated and

all the unstimulated samples in the case group and again for the control group. The distribution of

the differences in sample mean, median and standard deviation within case and control groups were

roughly normally distributed and generally the distributions of the case and control groups were of

equal variance. These differences were then compared across the case and control groups using a 2-

sample t-test assuming equal variance, except for the standard deviation of the B lymphocyte

populations, where a 2-sample t-test not assuming variance was used.

6.2.7 ICC for monocyte expression of HIF-1α

Refer to Chapter 5 for this previously described protocol.

6.2.7.1 ICC slide reading

ICC slides were read by this project’s primary supervisor and experienced IHC expert, Dr Helle

Bielefeldt-Ohmann. At least 200 cells were counted on each slide and each cell was assigned to one

of three categories. These categories were 1) no HIF-1α signal, 2) cytoplasmic HIF-1α signal only

and 3) cytoplasmic and nuclear HIF-1α signal.

6.2.7.2 ICC slide reading data analysis methods

The ICC results from the two subsets of cattle were analysed separately (see section 6.3.4) using

Stata 12.1. For each animal, the percentage of cells in each HIF-1α signal category was calculated

for both the unstimulated and stimulated culture, after which the absolute difference between these

two percentages was calculated. The percentage differences for the case group were compared to

61

those for the control group using a t-test not assuming equal variance for subset one, and a t-test

assuming equal variance for the subset two.

6.3 Results

6.3.1 Mononuclear cell isolation

From subset one, blood was collected from a total of 46 animals – 19 cases and 27 controls.

Peripheral blood mononuclear cell yields varied with total counts ranging from 1.7 x 108 to 2.17 x

109 cells isolated from 200 ml blood processed during the three sessions. Based on this range, the

amount of whole blood processed per animal was reduced to 100ml thereafter to expedite

processing while still isolating ample numbers of cells for culture, as the experimental design only

required 2.5 x106 cells for tube cell cultures and 2 x 107 cells for chamber cell cultures per animal.

In subset two, mononuclear leukocytes were isolated from 15 case animals and 27 control animals.

The low cellularity of the subset two samples meant these 42 animals were only included in the

chamber cultures for HIF-1α ICC.

6.3.2 Comparison of baseline data

Baseline data for the case and control animals is presented in table 6.1. There was no significant

difference in cohort, breed, gender, dentition or serological categories between cases and controls in

either of the two subsets of animals (table 6.1). Case animals were significantly lighter than controls

in the subset one (controls: mean (sd) 451 (30.7) kg; cases: mean (sd) 431 (21.7) kg (p= 0.028).

However, there was no significant difference in induction weight between cases and controls for

subset two (controls: mean (sd) 467 (27.1) kg; cases: mean (sd) 465 (33.4) kg (p= 0.753).

62

Table 6.1: Baseline data for cases and controls for both subsets

Variable

First Subset Second Subset Case

numbers (%)

Control numbers

(%) P-Value

Case numbers

(%)

Control numbers

(%) P-Value Cohort

0.7

1.00

Cohort 1 6 (31.6) 4 (14.8)

N/A N/A Cohort 2 5 (26.3) 8 (29.6)

N/A N/A

Cohort 3 1 (5.2) 2(7.4)

N/A N/A Cohort 4 6 (31.6) 9 (33.3)

N/A N/A

Cohort 5 1 (5.2) 4 (14.8)

N/A N/A Cohort 6 N/A N/A

7 (46.7) 12 (44.4)

Cohort 7 N/A N/A

8 (53.3) 15 (55.5) Breed

0.19

0.45

Angus 11 (57.9) 19 (70.4)

8 (53.3) 10 (37.0) Angus/Hereford 1 (5.2) 0 (0.0)

0 (0.0) 3 (11.1)

Angus/Santa Gertrudis 1 (5.2) 0 (0.0)

1 (6.6) 4 (14.8) Angus/Shorthorn 0 (0.0) 0 (0.0)

1 (6.6) 3 (11.1)

Brangus 1 (5.2) 0 (0.0)

0 (0.0) 0 (0.0) Charbray 1 (5.2) 1 (3.7)

0 (0.0) 0 (0.0)

Droughtmaster 0 (0.0) 1 (3.7)

0 (0.0) 0 (0.0) Hereford 4 (21.1) 2 (7.4)

4 (26.7) 2 (7.4)

Hereford/Santa Getrudis 0 (0.0) 1 (3.7)

0 (0.0) 1 (3.7) Santa Gertrudis 0 (0.0) 3 (11.1)

1 (6.6) 4 (14.8)

Gender Male 19 (100.0) 27 (100.0)

15 (100.0) 27 (100.0)

Dentition

0.25

0.84 0 14 (73.6) 23 (85.2)

11 (73.3) 22 (81.5)

2 5 (10.5) 3 (11.1)

3 (20.0) 3 (11.1) 4 0 (0.0) 1 (3.7)

1 (6.7) 2 (7.4)

Induction BHV-1 Titre

0.83

1.00 0 15 (79.0) 22 (81.5)

8 (53.3) 15 (55.6)

1 2 (10.5) 2 (7.4)

6 (40.0) 9 (33.3) 2 0 (0.0) 2 (7.4)

0 (0.0) 1 (3.7)

3 1 (5.3) 1 (3.7)

1 (6.7) 1 (3.7) 4 0 (0.0) 0 (0.0)

0 (0.0) 1 (3.7)

5 0 (0.0) 0 (0.0)

0 (0.0) 0 (0.0) Missing 1 (5.3) 0 (0.0)

0 (0.0) 0 (0.0)

Day 42 BHV-1 Titre

0.58

0.17 0 1 (5.3) 5 (18.5)

0 (0.0) 4 (14.8)

1 3 (15.8) 3 (11.1)

4 (26.7) 4 (14.8) 2 2 (10.5) 2 (7.4)

3 (20.0) 5 (18.5)

3 12 (63.2) 12 (44.4)

8 (53.3) 9 (33.3) 4 0 (0.0) 1 (3.7)

0 (0.0) 5 (18.5)

5 1 (5.3) 4 (14.8)

0 (0.0) 0 (0.0) Missing 0 (0.0) 0 (0.0)

0 (0.0) 0 (0.0)

63

Table 6.1: Baseline data for cases and controls for both subsets (continued)

Variable

First Subset Second Subset Case

numbers (%)

Control numbers

(%) P-Value

Case numbers

(%)

Control numbers

(%) P-Value Induction BVDV Titre

0.36

0.78

0 2 (10.5) 6 (22.2)

10 (66.7) 19 (70.4) 1 0 (0.0) 0 (0.0)

2 (13.3) 1 (3.7)

2 2 (10.5) 0 (0.0)

0 (0.0) 0 (0.0) 3 2 (10.5) 2 (7.4)

0 (0.0) 1 (3.7)

4 7 (36.8) 8 (29.6)

3 (20.0) 6 (22.2) 5 5 (26.3) 11 (40.7)

0 (0.0) 0 (0.0)

Missing 1 (5.3) 0 (0.0)

0 (0.0) 0 (0.0) Day 42 BVDV Titre

0.83

0.85

0 0 (0.0) 1 (3.7)

0 (0.0) 0 (0.0) 1 0 (0.0) 0 (0.0)

1 (6.7) 1 (3.7)

2 0 (0.0) 0 (0.0)

0 (0.0) 0 (0.0) 3 3 (15.9) 3 (11.1)

2 (13.3) 6 (22.2)

4 8 (42.1) 9 (33.3)

12 (80.0) 20 (74.1) 5 8 (42.1) 14 (51.9)

0 (0.0) 0 (0.0)

Missing 0 (0.0) 0 (0.0)

0 (0.0) 0 (0.0) Induction BSRV Titre

0.21

0.28

0 2 (10.5) 1 (3.7)

4 (26.7) 10 (37.0) 1 8 (42.1) 7 (25.9)

1 (6.7) 7 (25.9)

2 0 (0.0) 5 (18.5)

4 (26.7) 3 (11.1) 3 4 (21.1) 8 (29.6)

5 (33.3) 4 (14.8)

4 4 (21.1) 4 (14.8)

1 (6.7) 3 (11.1) 5 0 (0.0) 2 (7.4)

0 (0.0) 0 (0.0)

Missing 1 (5.3) 0 (0.0)

0 (0.0) 0 (0.0) Day 42 BSRV Titre

0.95

0.15

0 0 (0.0) 1 (3.7)

0 (0.0) 0 (0.0) 1 1 (5.3) 1 (3.7)

0 (0.0) 0 (0.0)

2 2 (10.5) 4 (14.8)

1 (6.7) 5 (18.5) 3 5 (26.3) 4 (14.8)

6 (40.0) 16 (59.3)

4 8 (42.1) 11 (40.7)

8 (53.3) 6 (22.2) 5 3 (15.8) 6 (22.2)

0 (0.0) 0 (0.0)

Missing 0 (0.0) 0 (0.0)

0 (0.0) 0 (0.0) Induction PI3 Titre

0.19

0.25

0 3 (15.8) 0 (0.0)

1 (6.7) 0 (0.0) 1 3 (15.8) 2 (7.41)

0 (0.0) 5 (18.5)

2 3 (15.8) 4 (14.8)

5 (33.3) 4 (14.8) 3 2 (10.5) 7 (25.9)

4 (26.7) 10 (37.0)

4 7 (36.8) 13 (48.2)

4 (26.7) 6 (22.2) 5 0 (0.0) 1 (3.7)

1 (6.7) 2 (7.4)

Missing 1 (5.3) 0 (0.0)

0 (0.0) 0 (0.0)

64

Table 6.1: Baseline data for cases and controls for both subsets (continued)

Variable

First Subset Second Subset Case

numbers (%)

Control numbers

(%) P-Value

Case numbers

(%)

Control numbers

(%) P-Value Day 42 PI3 Titre

0.79

0.77

0 0 (0.0) 0 (0.0)

0 (0.0) 0 (0.0) 1 1 (5.3) 1 (3.7)

0 (0.0) 3 (11.1)

2 2 (10.5) 2 (7.4)

2 (13.3) 4 (14.8) 3 5 (26.3) 7 (25.9)

3 (20.0) 6 (22.2)

4 10 (52.6) 12 (44.4)

10 (66.7) 13 (48.2) 5 1 (5.3) 5 (18.5)

0 (0.0) 1 (3.7)

Missing 0 (0.0) 0 (0.0)

0 (0.0) 0 (0.0)

6.3.3 Flow cytometry results

The mean and standard deviation of the HIF-1α fluorescent signal of B and T cells from stimulated

and unstimulated cultures for the master control samples from each session were compared, with

session one being different from the next two sessions. For example, sessions two and three resulted

in similar stimulated master control HIF-1α T-cell fluorescence, with a mean (sd) of 2.91 (1.83) and

2.94 (1.86) for these sessions, respectively. However, the session one stimulated master control

HIF-1α T-cell fluorescence mean (sd) was 3.43 (2.32). Session one samples contained larger

amounts of non-specific debris than previously encountered, and this along with the difference in

the master controls suggests that there were some differences in sample processing during this first

session. These differences may have arisen during the culture or subsequent flow procedures. The

fifteen animals from session one of flow cytometry were therefore excluded from further analysis.

Summarized analysis results of the remaining 31 animals from session two and session three are

shown in Table 6.2.

65

Table 6.2: Descriptive statistics for the absolute difference in mean, median and standard deviation of HIF-1α in B cells and T cells between unstimulated and stimulated cultures for case group and control groups.

Cell Type - Summary statistic of HIF-1α fluorescence

First Subset - Sessions two and three Number of

animals Mean Median SD P-Value1 B Cells - Mean

0.97

Difference Cases 16 -0.39 -0.42 0.42 Difference Controls 15 -0.39 -0.35 0.36 B Cells - Median

0.66

Difference Cases 16 -0.42 -0.34 0.44 Difference Controls 15 -0.35 -0.40 0.39 B Cells - Standard Deviation 15

0.162

Difference Cases 16 0.63 -0.05 2.84 Difference Controls 15 -0.43 -0.33 0.70 T Cells - Mean

0.50

Difference Cases 16 -0.44 -0.29 1.02 Difference Controls 15 -0.67 -0.64 0.81 T Cells - Median

0.76

Difference Cases 16 -0.44 -0.40 0.83 Difference Controls 15 -0.51 -0.50 0.27 T Cells - Standard Deviation

0.30

Difference Cases 16 -0.21 -0.07 0.88 Difference Controls 15 -0.58 -0.49 1.06

1 The p-value refers to the result of a 2-sample t-test comparing the difference between unstimulated and stimulated cultures of the case group to those of control group. 2 A t-test for unequal variances was used. 6.3.3.1 Statistical analysis of flow cytometry results

The difference in HIF-1α expression between simulated and unstimulated populations was not

significantly different between cases and controls when the mean, median and standard deviation of

the HIF-1α fluorescence for each of the B and T lymphocyte populations were compared (Table 6.2

above). There was also no significant difference present when data were reanalyzed including

results from the session one animals (data not shown).

6.3.4 ICC results

Six animals (all from subset one) were removed from ICC analysis due to concerns about the

quality of ICC staining for those animals. ICC of slides made from unstimulated cultures of subset

two had consistently greater numbers of cells with positive staining for HIF-1α than the

unstimulated cells of subset one. Due to this difference in baseline activation despite lack of

stimulation, the two subsets were analysed separately (Table 6.3). There was no staining present in

66

the control wells that had not received primary antibody, whereas some staining was detected in all

stimulated (Figure 6.2) and all unstimulated cultures immunolabeled for HIF-1α.

Table 6.3: ICC results for case and control groups from both subsets. The mean, median and standard deviation of the percentage of cells displaying cytoplasmic or combined cytoplasmic and nuclear HIF-1α positive ICC staining in stimulated and unstimulated cell cultures, as well as for the percentage difference of these two types of HIF-1α positive staining between stimulated and unstimulated cultures.

Cell culture status-Case/control status

First Subset Second Subset Number

of animals Mean Median SD Number of

animals Mean Median SD % Cells with Cytoplasmic HIF-1α signal

Unstimulated Cases 15 3.68 3.54 2.85 16 4.13 3.75 3.47 Unstimulated Controls 27 3.28 2.50 2.51 23 6.39 5.00 6.17 Stimulated Cases 14 3.00 2.75 2.14 16 3.97 3.25 3.19 Stimulated Controls 27 3.11 2.50 2.66 24 4.71 3.50 5.31 Difference Cases 14 -0.87 0.00 2.78 16 -0.16 0.00 3.45 Difference Controls 27 -0.17 -0.12 2.99 23 -1.50 1.00 7.63 % Cells with Cytoplasmic and Nuclear HIF-1α signal

Unstimulated Cases 15 24.80 28.00 19.07 16 9.50 0.75 16.50 Unstimulated Controls 27 21.69 17.00 18.20 23 3.02 1.00 4.74 Stimulated Cases 14 27.61 32.00 19.35 16 8.09 4.00 9.62 Stimulated Controls 27 26.93 29.00 18.37 24 3.63 2.50 3.49 Difference Cases 14 3.18 -0.28 13.15 16 -1.41 0.50 8.62 Difference Controls 27 5.24 5.50 13.28 23 0.72 1.00 4.05

                 

Figure 6.2: ICC of stimulated cells showing several monocytes with scattered intracytoplasmic HIF-1α expression (arrow) and a central, activated monocyte with strong intracytoplasmic and intranuclear HIF-1α signal.

67

6.3.4.1 Statistical analysis of ICC results

For neither subset of animals was there any significant difference in the percentage of HIF-1α

positive cells (either cytoplasmic ‘cytoplasmic positive’ or cytoplasmic/nuclear ‘nuclear positive’)

between the stimulated and unstimulated cultures when the results of case group was compared to

the results from the control group. Subset one had a p-value = 0.47 for difference in HIF-1α

cytoplasmic positive staining and p-value = 0.37 for difference in HIF-1α nuclear positive staining,

and the second subset had a p-value = 0.47 for difference in HIF-1α cytoplasmic positive staining

and a p-value = 0.64 for difference in HIF-1α nuclear positive staining (Figure 6.4 A-D).

68

Figure 6.3. A: Percentage difference in ICC HIF-1αcytoplasmic signal between stimulated and unstimulated cells, comparison of case group versus control group, subset one. B: Percentage difference in ICC HIF-αcytoplasmic/nuclear “nuclear positive” signal between stimulated and unstimulated cells, comparison of case group versus control group, subset one. C: Percentage difference in ICC HIF-1αcytoplasmic signal between stimulated and unstimulated cells, comparison of case group versus control group, subset two. D: Percentage difference in ICC HIF-1αcytoplasmic/nuclear “nuclear positive” signal between stimulated and unstimulated cells, comparison of case group versus control group, subset two.

0.0

5.1

.15

-40 -20 0 20 40 -40 -20 0 20 40

Controls Cases

Den

sity

Percentage difference of cytoplasmic positive cells

A

0.0

5.1

.15

-40 -20 0 20 40 -40 -20 0 20 40

Controls Cases

Den

sity

Percentage difference of cytoplasmic positive cells

C

0.0

5.1

.15

-40 -20 0 20 40 -40 -20 0 20 40

Controls Cases

Den

sity

Percentage difference of nuclear positive cells

B

0.0

5.1

.15

-40 -20 0 20 40 -40 -20 0 20 40

Controls Cases

Den

sity

Percentage difference of nuclear positive cells

D

69

6.4 Discussion

The cell culture, flow cytometry and ICC protocols described in previous chapters were

successfully applied to the test samples in this experiment. Using these protocols, the results

showed no significant difference in HIF-1α expression between stimulated and unstimulated bovine

leukocytes of animals that were treated for clinical BRD, when compared to that of control animals.

This outcome does not refute the null hypothesis that there is no significant difference in HIF-1α

regulation in bovine leukocytes of animals with clinical BRD compared with other clinically

healthy animals. These results do not support the idea of a direct effect of increased or decreased

HIF-1α expression in macrophages and lymphocytes contributing to the development of BRD

through cytokine production by upregulation or lack of suppression, and so do not support the

experimental hypothesis. The lack of a significant difference in HIF-1α expression between case

and control groups does not support the potential use of HIF-1α as a biomarker for BRD.

The lack of a detected difference in HIF-1α does not refute the theory of an overactive or

underactive immune response involving HIF-1α contributing to the development of clinical BRD.

This experiment did not expose any bovine cells to actual viral pathogens to account for any

possible HIF-1α modulation of the primary viral infection, in order to assess how this interaction

may alter HIF-1α expression and therefore any role HIF-1α may ultimately play in the pathogenesis

of BRD. Viral agents have a variety of mechanisms by which to utilise HIF-1α during infection.

Viruses can undergo HIF-1α –dependent or HIF-1α-enhanced viral replication (Davis et al., 2001,

Haque et al., 2003). VSV can induce HIF-1α stabilisation in bronchial cells, potentially contributing

to the clinical airway oedema of acute VSV infection (Kilani et al., 2004). Other viruses also induce

HIF-1α stabilisation (see 1.2.3.3.2). In some viruses, viral replication is inhibited by HIF-1α (Pipiya

et al., 2005, Hwang et al., 2006). Following BHV-1 infection, innate immune responses include the

antiviral action of IFN and the recruitment of lymphoid cells, macrophages, neutrophils and natural

killer cells to the site of infection (Bielefeldt Ohmann et al., 1991, Babiuk et al., 1996). Bovine

bronchial cells that are infected with BHV-1 will release cytokines, which subsequently promote

neutrophil recruitment to the lung (Rivera-Rivas et al., 2009). In vitro, TNF-α, IL-1β and IFN-γ are

secreted in response to BHV-1 infection, causing the increased expression of β2 integrin on bovine

alveolar macrophages and neutrophils (Czuprynski, 2009). The expression of HIF-1α and therefore

its value as a biomarker could still be significant in the context of these acute immune responses if

further investigations incorporated relevant and appropriate viral challenges. In this case, the lack of

a detected difference in HIF-1α could also mean that HIF-1α does not have a significant role in the

pathogenesis of clinical BRD, and further studies involving viral challenge could help to establish

or refute this possibility as well.

70

The overall small sample size tested in this experiment resulted in a low power to detect a possible

difference between the groups, so it is unlikely that a true but subtle difference would have been

statistically identified using such low numbers of animals. Testing larger numbers of animals would

have enhanced the power of the study. However the extent of the within-group variation suggests

that any difference between the groups would need to be even more substantial for such a difference

to be statistically significant, even with a much larger sample size.

Of interest, however, is the large variability in the differences of HIF-1α expression between

stimulated and unstimulated cells within both the cases and control groups. This variability in HIF-

1α expression of cells may be due to individual animal genetics not related to overall BRD

susceptibility, sample quality variation and the influences of the remaining insufficiencies in the

methodologies that will be covered further in the general discussion chapter. Between two

experimental sessions, the stimulated master control mean only varied by 0.04. In one session

however, the master control variation was larger, showing a similar variation in channel shift

between different replicates of stimulated cells as was seen between experimental stimulated and

unstimulated cell cultures. This larger variation in the stimulated master control indicates that

variation between stimulated and unstimulated cells may be due to a deficiency in methodology

rather than true variation. Conducting greater numbers of smaller flow sessions would have allowed

more replicates of the master control to be run. This would have allowed collection of more data

points of the master control values to establish the true repeatability, limits of acceptable variation

and the overall usefulness of the master control. This control, once properly established, could then

be used with revised cell culture and flow cytometry methodologies to more confidently detect true

variation in HIF-1α expression. Given the small changes detected in HIF-1α fluorescence, a more

complex analysis using raw data from flow cytometry LMD files (see Chapter 4) might be more

appropriate, instead of using only the summary results generated by cytometer analysis software.

Given additional resource availability, the use of the raw data LMD files during statistical analysis

would have been implemented.

ICC staining was observed either within the monocytic cytoplasm only, or within both the

cytoplasm and nucleus. This pattern of staining is consistent with the expected two types of

intracellular distribution of HIF-1α before and after translocation into the nucleus. There was no

significant difference in either nuclear positive or cytoplasmic positive staining upon stimulation

between cases and controls, however, within subset one there was consistently greater amounts of

nuclear positive staining in concurrence with slightly less cytoplasmic positive staining in

71

stimulated cells, when compared to the unstimulated cells. These results are consistent with

successful stabilisation and translocation of HIF-1α, however, without a known true positive

control, this cannot be confirmed.

6.5 Conclusion

The methodologies developed in the preceding chapters were applied to experimental samples.

There was no evidence to reject the null hypothesis within the framework of the current study.

However, the low power and methodological issues mean that a definitive statement cannot be

made as to whether HIF-1α can be used as a predictive biomarker for BRD. If a significant

difference in HIF-1α had been detected, there remains an issue regarding the practicality of this

protocol. This will be expanded upon in chapter 7.

72

7 General discussion and future directions

7.1 The importance of finding a biomarker for BRD

BRD `is a multi-factorial, inflammatory respiratory disease complex which poses a significant

problem for the beef industry. The disease complex involves both viral and bacterial pathogens and

invokes both innate and adaptive immune responses as well as the development of clinical

pneumonia in certain feedlot cattle. Investigations are focused on ways to enhance effective and

non-injurious immune responses to these pathogens (Ackermann et al., 2010b) in addition to

combating environmental and managerial issues that may predispose cattle to respiratory disease.

An initial step in the ultimate aim of enhancing effective immune responses is to identify biomarker

molecules that are expressed differently in animals that have an effective response to an infectious

challenge when compared to other animals that develop disease. Such a biomarker would allow the

implementation of targeted breeding programs to enhance resistance to this common yet complex

disease. This project proposed HIF-1α as a potential biomarker for development of clinical disease,

and endeavoured to develop methods to induce and detect HIF-1α expression in bovine

lymphocytes and monocytes. This was followed by implementation of developed methods

experimentally to compare the HIF-1α expression in a group of clinically healthy beef feedlot cattle

(controls) with the HIF-1α expression in beef feedlot cattle that were treated for clinical respiratory

disease (cases) while in the feedlot.

The lack of a significant difference in HIF-1α expression between case and control groups did not

support the potential use of HIF-1α as a biomarker for BRD. However, with sample size dictated by

time and resource constraints, the study only had the power to detect a large difference in HIF-1α

expression between cases and controls. Furthermore, this finding is not considered conclusive due

to a number of weaknesses and limitations in the developed methods. These weaknesses,

limitations, as well as potential solutions and additional ideas for further development will now be

discussed in turn.

7.2 Methodology development issues

7.2.1 Hypoxic cell culture and harvest methods

The inherent weakness of the hypoxic cell culture method is that it was not possible to confirm

successful HIF-1α stabilisation by means other than the detection methods simultaneously being

developed. The stabilisation protocol was developed based on similar studies in the literature,

however, ultimately its efficacy could not be confirmed. Future work could involve use of a better-

established measurement technique such as Western blot and the use of commercially produced

CoCl2-treated cell lysate as a known positive control to confirm accumulation of the protein

73

(Lidgren et al., 2005). A positive control could also aid in further troubleshooting of culture length

and hypoxic mimetic concentration. Additional stimulants such as TNF- α could also be used as an

inducing signal to further increase HIF-1α stimulation during a shorter culture time (Albina et al.,

2001). As discussed in chapter 6, use of BRD viral agents during hypoxic culture would better

mimic the in vivo situation and could also contribute to a better understanding of HIF-1α’s role in

the pathogenesis of clinical BRD. A different cell type could also be considered, given that BHV-1,

PI-3, BRSV and BVDV can infect lung epithelial cells (Ackermann et al., 2010b) and that bovine

bronchial cells that are infected with BHV-1 will release cytokines (Rivera-Rivas et al., 2009). Use

of bovine lung cells for culture in conjunction with these viruses could be an avenue for testing the

potential role of HIF-1α in BRD susceptibility.

A further shortcoming in the protocol was the poor post-culture monocyte viability that meant this

cell population had to be assessed with an alternative technique. This negates, to an extent, one of

the reasons for using flow cytometry, i.e. allowing for simultaneous assessment of multiple cell

types, and also decreases the attractiveness of the protocol for more widespread use, as it would

require multiple sets of reagents, equipment and expertise. A lower concentration of CoCl2 may

improve monocyte viability to the point where these cells could once again be included in flow

cytometry analysis.

Poor post-culture viability was a significant issue with frozen-thawed cells. Unfortunately, due to

the ongoing trouble-shooting and work-up of protocols and the restricted nature of sample

collections, the opportunity to be in the position to culture fresh cells immediately after collection

did not arise. Using fresh cells remains an avenue for future investigation into hypoxic culture of

bovine leukocytes for stabilisation of HIF-1α.

7.2.2 Flow cytometry methods

This project explored the use of flow cytometry to measure HIF-1α protein as a relatively novel

technique that could potentially allow enhanced sensitivity in detecting differences in HIF-1α

expression between individuals. Flow cytometry has the capacity to make individual measurements

from a large number of discrete particles, and can be used for the analysis of any cellular structure

or function as long as an appropriate probe is available (Jaroszeski and Radcliff, 1999). Given the

large potential for variation inherent in the multiple steps of this protocol, a positive control with

clearly defined levels of HIF-1α would have been of great use. Unfortunately such a control was not

available. The difference in HIF-1α fluorescence between unstimulated and stimulated cells was not

pronounced and did not show a consistent increased or decreased fluorescence after stimulation.

The HIF-1α fluorescence was also quite close to the negative controls (1 to 2 channel shifts only) in

74

contrast to the phenotypic marker fluorescence, which often was over 15 channel shifts above the

negative controls. This suggests the fluorochrome used for HIF-1α detection was not bright enough

in the context of the relative lower expression of the molecule. Alternatively the phenotypic

fluorochromes were too bright in comparison, or it could be a combination of the two options, with

subtle changes in HIF-1α possibly masked as a result. The brightness of the HIF-1α signal is

affected not only by the intensity with which the positive cells are stained, but also the background

of the negative cells (Baumgarth and Roederer, 2000). Use of a brighter fluorochrome such as PE,

Cy5PE or APC could be used to investigate what was essentially an unknown level of a molecule of

interest (Baumgarth and Roederer, 2000). In addition, discontinuation of cell phenotyping until

measurement of HIF-1α protein has been established should have been considered, if time was not a

limitation.

It is possible that the significant degree of apoptosis occurring in the monocyte population was also

affecting other cell types, causing aberrant events. Large amounts of cell death in this project during

method development could have influenced flow cytometry results. Ethidium monoazide could

have been used to detect dead cells in a separate sample.

With the developed method there still existed flow cytometry events that were not labelled with

either phenotypic marker. These may represent either remaining monocytes, B cells undergoing

differentiation and therefore down-regulate CD20, dead cells, large debris or contaminating

granulocytes, although the latter is unlikely considering the poor survival of freezing-preservation

of granulocytes. Events that were labelled by both CD20 and CD3 were also noted. The nature of

such events remains uncertain. Neoplastic lymphocytes can show true CD20 and CD3 positivity on

the same cell (Sun et al., 2004), however, in this case the events are considered more likely to be

dead cells with non-specific binding of antibody. Several methods were employed to minimize

variation due to electronic factors. These methods included first establishing and then using fixed

machine settings during the experiment sessions, the diligent use of the recommended quality

assurance beads to assess laser alignment prior to each session and the use of a tailored master

control with each session. Unstained samples (which did not contain signal) addressed possible

cellular auto-fluorescence and variation due to non-specific binding of antibody was minimised by

frequent washing of the cells during preparation and omitting use of problematic secondary

antibodies in the protocol. Gating of populations was performed manually on 2-dimensional dot

plots, the subjectivity of which is a potential source of variation during routine post-processing of

flow data, and is acknowledged as a possible source of data variation in this experiment.

75

7.2.3 ICC methods

As a consequence of methodology development issues, ICC was also implemented as a more

established method of HIF-1α measurement. The method can provide information as to the

presence of absence of HIF-1α andthe nuclear translocation of the molecule, but can be argued as

relatively insensitive for distinguishing between any subtle differences in the level of HIF-1α

expression between different animals or culture conditions.

A possible source of variation in ICC is the subjectivity of human interpretation of staining. This

could be mitigated by applying morphometric computer-based interpretation of the

immunolabelling. Lack of appropriate facilities for this was, however, and issue not resolved within

the timeframe of the project. In this study the ICC was carried out manually and if the approach was

to be applied to a much larger number of animals automated staining should be considered in

combination with a morphometric read-out.

required.

7.3 Future directions for investigation of biomarkers for BRD

In the future, in addition to possible modifications of the methodologies used in this study, other

methodologies for the evaluation of HIF-1α in bovines should be considered, such as Western blot

for HIF-1α protein or real-time polymerase chain reaction for HIF-1α mRNA. Western blot was

also considered as a method in this study due to available supervisor expertise and facilities.

However, the large numbers of cells required for adequate protein yield would be problematic due

to the low total cell numbers in many of the intended experimental samples as well as the low

retrievable cell population proportion and low total cell counts of harvested hypoxic cultures.

Performing Western blot instead of flow cytometry would remove information gained through

phenotyping, and would therefore require isolation of lymphocyte subsets prior to hypoxic culture

conditions. This is unlikely to be practical on a large scale unless cell sorting facilities are available.

Nevertheless, Western blot detection of HIF-1α protein is more established method than flow

cytometry and may represent a more practical alternative for large scale assessment.

Another possibility to consider in further investigations of the role the HIF-1α pathway in BRD is to

measure molecules induced downstream of HIF-1α such as VEGF, TNFα, other inflammatory

mediators and ROS. It is also worthwhile considering other molecules in hypoxic inflammation in

the lungs as potential biomarkers, such as the transcription factor NF-kB. Alveolar epithelial cells

and resident alveolar macrophages recognize various microbial components via pattern recognition

receptors (PRR) with resultant production of proinflamamtory cytokines via the NF-kB pathway,

directly affecting expression of genes involved in innate immunity as well as indirectly causing

amplification of the HIF1α pathway (Schaible et al., 2010).

76

Based on this study, the expression of HIF-1α and its value as a biomarker for the development of

clinical BRD as a disease outcome could still be significant and remains to be established. Further

work is needed to improve these methodologies, or develop alternative methods and undertake a

study in a much larger population of cattle in order to achieve more conclusive results. The

expression of HIF-1α could still be shown as significant in the context of the bovine acute immune

responses to BRD viral pathogens if further investigations incorporated relevant and appropriate

viral challenges.

In this study, the lack of a detected difference in HIF-1α could also mean that HIF-1α does not have

a significant role in the pathogenesis of clinical BRD, and further studies involving viral challenge

could also help to establish or refute this possibility.

77

8 References

 [ANON] 1973. Glossary on respiration and gas exchange. Journal of Applied Physiology, 34, 549-58.

ACKERMANN, M. R., DERSCHEID, R. & ROTH, J. A. 2010a. Innate immunology of bovine respiratory disease. The Veterinary clinics of North America. Food animal practice, 26, 215-28.

ACKERMANN, M. R., DERSCHEID, R. & ROTH, J. A. 2010b. Innate immunology of bovine respiratory disease. Vet Clin North Am Food Anim Pract, 26, 215-28.

ADAMIS, A. P., MILLER, J. W., BERNAL, M. T., D'AMICO, D. J., FOLKMAN, J., YEO, T. K. & YEO, K. T. 1994. Increased vascular endothelial growth factor levels in the vitreous of eyes with proliferative diabetic retinopathy. Am J Ophthalmol, 118, 445-50.

AICH, P., BABIUK, L. A., POTTER, A. A. & GRIEBEL, P. 2009. Biomarkers for prediction of bovine respiratory disease outcome. OMICS, 13, 199-209.

ALBINA, J. E., MASTROFRANCESCO, B., VESSELLA, J. A., LOUIS, C. A., HENRY, W. L., JR. & REICHNER, J. S. 2001. HIF-1 expression in healing wounds: HIF-1alpha induction in primary inflammatory cells by TNF-alpha. Am J Physiol Cell Physiol, 281, C1971-7.

AMERI, K., LUONG, R., ZHANG, H., POWELL, A. A., MONTGOMERY, K. D., ESPINOSA, I., BOULEY, D. M., HARRIS, A. L. & JEFFREY, S. S. 2010. Circulating tumour cells demonstrate an altered response to hypoxia and an aggressive phenotype. Br J Cancer, 102, 561-9.

ANAND, R. J., GRIBAR, S. C., LI, J., KOHLER, J. W., BRANCA, M. F., DUBOWSKI, T., SODHI, C. P. & HACKAM, D. J. 2007. Hypoxia causes an increase in phagocytosis by macrophages in a HIF-1alpha-dependent manner. J Leukoc Biol, 82, 1257-65.

AUGSTEIN, A., POITZ, D. M., BRAUN-DULLAEUS, R. C., STRASSER, R. H. & SCHMEISSER, A. 2011. Cell-specific and hypoxia-dependent regulation of human HIF-3alpha: inhibition of the expression of HIF target genes in vascular cells. Cell Mol Life Sci, 68, 2627-42.

BABIUK, L. A., VAN DRUNEN LITTEL-VAN DEN HURK, S. & TIKOO, S. K. 1996. Immunology of bovine herpesvirus 1 infection. Vet Microbiol, 53, 31-42.

BASKIN, C. R., BIELEFELDT-OHMANN, H., TUMPEY, T. M., SABOURIN, P. J., LONG, J. P., GARCIA-SASTRE, A., TOLNAY, A. E., ALBRECHT, R., PYLES, J. A., OLSON, P. H., AICHER, L. D., ROSENZWEIG, E. R., MURALI-KRISHNA, K., CLARK, E. A., KOTUR, M. S., FORNEK, J. L., PROLL, S., PALERMO, R. E., SABOURIN, C. L. & KATZE, M. G. 2009. Early and sustained innate immune response defines pathology and death in nonhuman primates infected by highly pathogenic influenza virus. Proc Natl Acad Sci U S A, 106, 3455-60.

BAUMGARTH, N. & ROEDERER, M. 2000. A practical approach to multicolor flow cytometry for immunophenotyping. J Immunol Methods, 243, 77-97.

BELAIBA, R. S., BONELLO, S., ZAHRINGER, C., SCHMIDT, S., HESS, J., KIETZMANN, T. & GORLACH, A. 2007. Hypoxia up-regulates hypoxia-inducible factor-1alpha transcription by involving phosphatidylinositol 3-kinase and nuclear factor kappaB in pulmonary artery smooth muscle cells. Mol Biol Cell, 18, 4691-7.

BEN-SHOSHAN, J., MAYSEL-AUSLENDER, S., MOR, A., KEREN, G. & GEORGE, J. 2008. Hypoxia controls CD4+CD25+ regulatory T-cell homeostasis via hypoxia-inducible factor-1alpha. Eur J Immunol, 38, 2412-8.

BIELEFELDT OHMANN, H., BABIUK, L. A. & HARLAND, R. 1991. Cytokine synergy with viral cytopathic effects and bacterial products during the pathogenesis of respiratory tract infection. Clin Immunol Immunopathol, 60, 153-70.

78

BIELEFELDT-OHMANN, H., SMIRNOVA, N. P., TOLNAY, A. E., WEBB, B. T., ANTONIAZZI, A. Q., VAN CAMPEN, H. & HANSEN, T. R. 2012. Neuro-invasion by a 'Trojan Horse' strategy and vasculopathy during intrauterine flavivirus infection. Int J Exp Pathol, 93, 24-33.

BIELEFELDT-OHMANN, H., TOLNAY, A. E., REISENHAUER, C. E., HANSEN, T. R., SMIRNOVA, N. & VAN CAMPEN, H. 2008. Transplacental infection with non-cytopathic bovine viral diarrhoea virus types 1b and 2: viral spread and molecular neuropathology. J Comp Pathol, 138, 72-85.

BIJU, M. P., NEUMANN, A. K., BENSINGER, S. J., JOHNSON, R. S., TURKA, L. A. & HAASE, V. H. 2004. Vhlh gene deletion induces Hif-1-mediated cell death in thymocytes. Mol Cell Biol, 24, 9038-47.

BILTON, R., MAZURE, N., TROTTIER, E., HATTAB, M., DERY, M. A., RICHARD, D. E., POUYSSEGUR, J. & BRAHIMI-HORN, M. C. 2005. Arrest-defective-1 protein, an acetyltransferase, does not alter stability of hypoxia-inducible factor (HIF)-1alpha and is not induced by hypoxia or HIF. J Biol Chem, 280, 31132-40.

BLOUIN, C. C., PAGE, E. L., SOUCY, G. M. & RICHARD, D. E. 2004. Hypoxic gene activation by lipopolysaccharide in macrophages: implication of hypoxia-inducible factor 1alpha. Blood, 103, 1124-30.

BOONLAYANGOOR, P., TELISCHI, M., BOONLAYANGOOR, S., SINCLAIR, T. F. & MILLHOUSE, E. W. 1980. Cryopreservation of human granulocytes: study of granulocyte function and ultrastructure. Blood, 56, 237-45.

BOVE, P. F., HRISTOVA, M., WESLEY, U. V., OLSON, N., LOUNSBURY, K. M. & VAN DER VLIET, A. 2008. Inflammatory levels of nitric oxide inhibit airway epithelial cell migration by inhibition of the kinase ERK1/2 and activation of hypoxia-inducible factor-1 alpha. J Biol Chem, 283, 17919-28.

BRAUN, R. D., LANZEN, J. L., SNYDER, S. A. & DEWHIRST, M. W. 2001. Comparison of tumor and normal tissue oxygen tension measurements using OxyLite or microelectrodes in rodents. Am J Physiol Heart Circ Physiol, 280, H2533-44.

BROWN, L. F., YEO, K. T., BERSE, B., YEO, T. K., SENGER, D. R., DVORAK, H. F. & VAN DE WATER, L. 1992. Expression of vascular permeability factor (vascular endothelial growth factor) by epidermal keratinocytes during wound healing. J Exp Med, 176, 1375-9.

BRUICK, R. K. & MCKNIGHT, S. L. 2001. A conserved family of prolyl-4-hydroxylases that modify HIF. Science, 294, 1337-40.

BRUNE, B. & ZHOU, J. 2007. Nitric oxide and superoxide: interference with hypoxic signaling. Cardiovasc Res, 75, 275-82.

BURRY, R. W. 2011. Controls for immunocytochemistry: an update. J Histochem Cytochem, 59, 6-12.

CAI, Q., MURAKAMI, M., SI, H. & ROBERTSON, E. S. 2007. A potential alpha-helix motif in the amino terminus of LANA encoded by Kaposi's sarcoma-associated herpesvirus is critical for nuclear accumulation of HIF-1alpha in normoxia. J Virol, 81, 10413-23.

CALLAPINA, M., ZHOU, J., SCHMID, T., KOHL, R. & BRUNE, B. 2005. NO restores HIF-1alpha hydroxylation during hypoxia: role of reactive oxygen species. Free Radic Biol Med, 39, 925-36.

CAMPANA, D., COUSTAN-SMITH, E. & JANOSSY, G. 1989. "Cytoplasmic" expression of nuclear antigens. Leukemia, 3, 239-40.

CARMELIET, P., DOR, Y., HERBERT, J. M., FUKUMURA, D., BRUSSELMANS, K., DEWERCHIN, M., NEEMAN, M., BONO, F., ABRAMOVITCH, R., MAXWELL, P., KOCH, C. J., RATCLIFFE, P., MOONS, L., JAIN, R. K., COLLEN, D. & KESHERT, E. 1998. Role of HIF-1alpha in hypoxia-mediated apoptosis, cell proliferation and tumour angiogenesis. Nature, 394, 485-90.

CHACHAMI, G., SIMOS, G., HATZIEFTHIMIOU, A., BONANOU, S., MOLYVDAS, P. A. & PARASKEVA, E. 2004. Cobalt induces hypoxia-inducible factor-1alpha expression in airway smooth muscle cells by a reactive oxygen species- and PI3K-dependent mechanism. Am J Respir Cell Mol Biol, 31, 544-51.

79

CHANDEL, N. S., MCCLINTOCK, D. S., FELICIANO, C. E., WOOD, T. M., MELENDEZ, J. A., RODRIGUEZ, A. M. & SCHUMACKER, P. T. 2000. Reactive oxygen species generated at mitochondrial complex III stabilize hypoxia-inducible factor-1alpha during hypoxia: a mechanism of O2 sensing. J Biol Chem, 275, 25130-8.

CHEN, C., PORE, N., BEHROOZ, A., ISMAIL-BEIGI, F. & MAITY, A. 2001. Regulation of glut1 mRNA by hypoxia-inducible factor-1. Interaction between H-ras and hypoxia. J Biol Chem, 276, 9519-25.

CRAMER, T., YAMANISHI, Y., CLAUSEN, B. E., FORSTER, I., PAWLINSKI, R., MACKMAN, N., HAASE, V. H., JAENISCH, R., CORR, M., NIZET, V., FIRESTEIN, G. S., GERBER, H. P., FERRARA, N. & JOHNSON, R. S. 2003. HIF-1alpha is essential for myeloid cell-mediated inflammation. Cell, 112, 645-57.

CZUPRYNSKI, C. J. 2009. Host response to bovine respiratory pathogens. Anim Health Res Rev, 10, 141-3.

DAI, Y., XU, M., WANG, Y., PASHA, Z., LI, T. & ASHRAF, M. 2007. HIF-1alpha induced-VEGF overexpression in bone marrow stem cells protects cardiomyocytes against ischemia. J Mol Cell Cardiol, 42, 1036-44.

DAI, Z. J., GAO, J., MA, X. B., YAN, K., LIU, X. X., KANG, H. F., JI, Z. Z., GUAN, H. T. & WANG, X. J. 2012. Up-regulation of hypoxia inducible factor-1alpha by cobalt chloride correlates with proliferation and apoptosis in PC-2 cells. J Exp Clin Cancer Res, 31, 28.

DALTON-GRIFFIN, L., WILSON, S. J. & KELLAM, P. 2009. X-box binding protein 1 contributes to induction of the Kaposi's sarcoma-associated herpesvirus lytic cycle under hypoxic conditions. J Virol, 83, 7202-9.

DAREKAR, S., GEORGIOU, K., YURCHENKO, M., YENAMANDRA, S. P., CHACHAMI, G., SIMOS, G., KLEIN, G. & KASHUBA, E. 2012. Epstein-Barr virus immortalization of human B-cells leads to stabilization of hypoxia-induced factor 1 alpha, congruent with the Warburg effect. PLoS One, 7, e42072.

DAVIS, D. A., RINDERKNECHT, A. S., ZOETEWEIJ, J. P., AOKI, Y., READ-CONNOLE, E. L., TOSATO, G., BLAUVELT, A. & YARCHOAN, R. 2001. Hypoxia induces lytic replication of Kaposi sarcoma-associated herpesvirus. Blood, 97, 3244-50.

DEHNE, N. & BRUNE, B. 2009. HIF-1 in the inflammatory microenvironment. Exp Cell Res, 315, 1791-7.

DESHMANE, S. L., MUKERJEE, R., FAN, S., DEL VALLE, L., MICHIELS, C., SWEET, T., ROM, I., KHALILI, K., RAPPAPORT, J., AMINI, S. & SAWAYA, B. E. 2009. Activation of the oxidative stress pathway by HIV-1 Vpr leads to induction of hypoxia-inducible factor 1alpha expression. J Biol Chem, 284, 11364-73.

EHRISMANN, D., FLASHMAN, E., GENN, D. N., MATHIOUDAKIS, N., HEWITSON, K. S., RATCLIFFE, P. J. & SCHOFIELD, C. J. 2007. Studies on the activity of the hypoxia-inducible-factor hydroxylases using an oxygen consumption assay. Biochem J, 401, 227-34.

EITAM, H., VAYA, J., BROSH, A., ORLOV, A., KHATIB, S., IZHAKI, I. & SHABTAY, A. 2010. Differential stress responses among newly received calves: variations in reductant capacity and Hsp gene expression. Cell stress & chaperones, 15, 865-76.

ELSON, D. A., THURSTON, G., HUANG, L. E., GINZINGER, D. G., MCDONALD, D. M., JOHNSON, R. S. & ARBEIT, J. M. 2001. Induction of hypervascularity without leakage or inflammation in transgenic mice overexpressing hypoxia-inducible factor-1alpha. Genes Dev, 15, 2520-32.

ELTZSCHIG, H. K. & CARMELIET, P. 2011. Hypoxia and inflammation. N Engl J Med, 364, 656-65.

EMA, M., TAYA, S., YOKOTANI, N., SOGAWA, K., MATSUDA, Y. & FUJII-KURIYAMA, Y. 1997. A novel bHLH-PAS factor with close sequence similarity to hypoxia-inducible factor 1alpha regulates the VEGF expression and is potentially involved in lung and vascular development. Proc Natl Acad Sci U S A, 94, 4273-8.

80

EPSTEIN, A. C., GLEADLE, J. M., MCNEILL, L. A., HEWITSON, K. S., O'ROURKE, J., MOLE, D. R., MUKHERJI, M., METZEN, E., WILSON, M. I., DHANDA, A., TIAN, Y. M., MASSON, N., HAMILTON, D. L., JAAKKOLA, P., BARSTEAD, R., HODGKIN, J., MAXWELL, P. H., PUGH, C. W., SCHOFIELD, C. J. & RATCLIFFE, P. J. 2001. C. elegans EGL-9 and mammalian homologs define a family of dioxygenases that regulate HIF by prolyl hydroxylation. Cell, 107, 43-54.

FETSCH, P. 2005. Immunocytochemistry. In: WALKER, J. & RAPLEY, R. (eds.) Medical Biomethods Handbook. Humana Press.

FOLKMAN, J. & SHING, Y. 1992. Angiogenesis. J Biol Chem, 267, 10931-4.

FORMENTO, J. L., BERRA, E., FERRUA, B., MAGNE, N., SIMOS, G., BRAHIMI-HORN, C., POUYSSEGUR, J. & MILANO, G. 2005. Enzyme-linked immunosorbent assay for pharmacological studies targeting hypoxia-inducible factor 1alpha. Clin Diagn Lab Immunol, 12, 660-4.

FORRISTAL, C. E., WRIGHT, K. L., HANLEY, N. A., OREFFO, R. O. & HOUGHTON, F. D. 2010. Hypoxia inducible factors regulate pluripotency and proliferation in human embryonic stem cells cultured at reduced oxygen tensions. Reproduction, 139, 85-97.

FORSYTHE, J. A., JIANG, B. H., IYER, N. V., AGANI, F., LEUNG, S. W., KOOS, R. D. & SEMENZA, G. L. 1996. Activation of vascular endothelial growth factor gene transcription by hypoxia-inducible factor 1. Mol Cell Biol, 16, 4604-13.

FREDE, S., BERCHNER-PFANNSCHMIDT, U. & FANDREY, J. 2007. Regulation of hypoxia-inducible factors during inflammation. Methods Enzymol, 435, 405-19.

GARSON, J. A., BEVERLEY, P. C., COAKHAM, H. B. & HARPER, E. I. 1982. Monoclonal antibodies against human T lymphocytes label Purkinje neurones of many species. Nature, 298, 375-7.

GERALD, D., BERRA, E., FRAPART, Y. M., CHAN, D. A., GIACCIA, A. J., MANSUY, D., POUYSSEGUR, J., YANIV, M. & MECHTA-GRIGORIOU, F. 2004. JunD reduces tumor angiogenesis by protecting cells from oxidative stress. Cell, 118, 781-94.

GILBERT, S. F. 2010. Developmental biology, Sunderland, Mass., Sinauer Associates.

GIVAN, A. L. 2001. Flow cytometry : first principles, New York ; Chichester, Wiley.

GOLDBERG, M. A., DUNNING, S. P. & BUNN, H. F. 1988. Regulation of the erythropoietin gene: evidence that the oxygen sensor is a heme protein. Science, 242, 1412-5.

GREIJER, A. E. & VAN DER WALL, E. 2004. The role of hypoxia inducible factor 1 (HIF-1) in hypoxia induced apoptosis. Journal of clinical pathology, 57, 1009-14.

GRIFFIN, D. 1997. Economic impact associated with respiratory disease in beef cattle. Vet Clin North Am Food Anim Pract, 13, 367-77.

GU, Y. Z., MORAN, S. M., HOGENESCH, J. B., WARTMAN, L. & BRADFIELD, C. A. 1998. Molecular characterization and chromosomal localization of a third alpha-class hypoxia inducible factor subunit, HIF3alpha. Gene Expr, 7, 205-13.

GUO, M., SONG, L. P., JIANG, Y., LIU, W., YU, Y. & CHEN, G. Q. 2006. Hypoxia-mimetic agents desferrioxamine and cobalt chloride induce leukemic cell apoptosis through different hypoxia-inducible factor-1alpha independent mechanisms. Apoptosis, 11, 67-77.

GUZY, R. D., HOYOS, B., ROBIN, E., CHEN, H., LIU, L., MANSFIELD, K. D., SIMON, M. C., HAMMERLING, U. & SCHUMACKER, P. T. 2005. Mitochondrial complex III is required for hypoxia-induced ROS production and cellular oxygen sensing. Cell Metab, 1, 401-8.

HADDAD, J. J. 2002. Antioxidant and prooxidant mechanisms in the regulation of redox(y)-sensitive transcription factors. Cell Signal, 14, 879-97.

81

HADDAD, J. J. & LAND, S. C. 2001. A non-hypoxic, ROS-sensitive pathway mediates TNF-alpha-dependent regulation of HIF-1alpha. FEBS letters, 505, 269-74.

HAEBERLE, H. A., DURRSTEIN, C., ROSENBERGER, P., HOSAKOTE, Y. M., KUHLICKE, J., KEMPF, V. A., GAROFALO, R. P. & ELTZSCHIG, H. K. 2008. Oxygen-independent stabilization of hypoxia inducible factor (HIF)-1 during RSV infection. PLoS One, 3, e3352.

HAGEN, T., TAYLOR, C. T., LAM, F. & MONCADA, S. 2003. Redistribution of intracellular oxygen in hypoxia by nitric oxide: effect on HIF1alpha. Science, 302, 1975-8.

HAILMAN, E., LICHENSTEIN, H. S., WURFEL, M. M., MILLER, D. S., JOHNSON, D. A., KELLEY, M., BUSSE, L. A., ZUKOWSKI, M. M. & WRIGHT, S. D. 1994. Lipopolysaccharide (LPS)-binding protein accelerates the binding of LPS to CD14. J Exp Med, 179, 269-77.

HAQUE, M., DAVIS, D. A., WANG, V., WIDMER, I. & YARCHOAN, R. 2003. Kaposi's sarcoma-associated herpesvirus (human herpesvirus 8) contains hypoxia response elements: relevance to lytic induction by hypoxia. J Virol, 77, 6761-8.

HAQUE, M. & KOUSOULAS, K. G. 2013. The Kaposi's sarcoma-associated herpesvirus ORF34 protein binds to HIF-1alpha and causes its degradation via the proteasome pathway. J Virol, 87, 2164-73.

HARA, S., KOBAYASHI, C. & IMURA, N. 1999. Molecular cloning of cDNAs encoding hypoxia-inducible factor (HIF)-1alpha and -2alpha of bovine arterial endothelial cells. Biochim Biophys Acta, 1445, 237-43.

HELLWIG-BURGEL, T., RUTKOWSKI, K., METZEN, E., FANDREY, J. & JELKMANN, W. 1999. Interleukin-1beta and tumor necrosis factor-alpha stimulate DNA binding of hypoxia-inducible factor-1. Blood, 94, 1561-7.

HELMLINGER, G., YUAN, F., DELLIAN, M. & JAIN, R. K. 1997. Interstitial pH and pO2 gradients in solid tumors in vivo: high-resolution measurements reveal a lack of correlation. Nat Med, 3, 177-82.

HERSHKOVITZ, T., SHENHAV, A., SABO, E., BEN-IZHAK, O. & HERSHKOVITZ, D. 2013. Development of a computerized morphometry application for assessment of the tumor fraction in colon carcinoma tissue samples. Appl Immunohistochem Mol Morphol, 21, 54-8.

HERTIG, A. T. 1935. Angiogenesis in the early human chorion and in the primary placenta of the macaque monkey, Washington,.

HODGSON, P. D., AICH, P., MANUJA, A., HOKAMP, K., ROCHE, F. M., BRINKMAN, F. S., POTTER, A., BABIUK, L. A. & GRIEBEL, P. J. 2005. Effect of stress on viral-bacterial synergy in bovine respiratory disease: novel mechanisms to regulate inflammation. Comp Funct Genomics, 6, 244-50.

HU, R., DAI, A. & TAN, S. 2002. Hypoxia-inducible factor 1 alpha upregulates the expression of inducible nitric oxide synthase gene in pulmonary arteries of hyposic rat. Chin Med J (Engl), 115, 1833-7.

HUANG, L. E., ARANY, Z., LIVINGSTON, D. M. & BUNN, H. F. 1996. Activation of hypoxia-inducible transcription factor depends primarily upon redox-sensitive stabilization of its alpha subunit. J Biol Chem, 271, 32253-9.

HUANG, L. E., GU, J., SCHAU, M. & BUNN, H. F. 1998. Regulation of hypoxia-inducible factor 1alpha is mediated by an O2-dependent degradation domain via the ubiquitin-proteasome pathway. Proc Natl Acad Sci U S A, 95, 7987-92.

HWANG, II, WATSON, I. R., DER, S. D. & OHH, M. 2006. Loss of VHL confers hypoxia-inducible factor (HIF)-dependent resistance to vesicular stomatitis virus: role of HIF in antiviral response. J Virol, 80, 10712-23.

INTAGLIETTA, M., JOHNSON, P. C. & WINSLOW, R. M. 1996. Microvascular and tissue oxygen distribution. Cardiovasc Res, 32, 632-43.

82

IVAN, M., KONDO, K., YANG, H., KIM, W., VALIANDO, J., OHH, M., SALIC, A., ASARA, J. M., LANE, W. S. & KAELIN, W. G., JR. 2001. HIFalpha targeted for VHL-mediated destruction by proline hydroxylation: implications for O2 sensing. Science, 292, 464-8.

IYER, N. V., KOTCH, L. E., AGANI, F., LEUNG, S. W., LAUGHNER, E., WENGER, R. H., GASSMANN, M., GEARHART, J. D., LAWLER, A. M., YU, A. Y. & SEMENZA, G. L. 1998. Cellular and developmental control of O2 homeostasis by hypoxia-inducible factor 1 alpha. Genes Dev, 12, 149-62.

JAAKKOLA, P., MOLE, D. R., TIAN, Y. M., WILSON, M. I., GIELBERT, J., GASKELL, S. J., KRIEGSHEIM, A., HEBESTREIT, H. F., MUKHERJI, M., SCHOFIELD, C. J., MAXWELL, P. H., PUGH, C. W. & RATCLIFFE, P. J. 2001. Targeting of HIF-alpha to the von Hippel-Lindau ubiquitylation complex by O2-regulated prolyl hydroxylation. Science, 292, 468-72.

JANOSSY, G., COUSTAN-SMITH, E. & CAMPANA, D. 1989. The reliability of cytoplasmic CD3 and CD22 antigen expression in the immunodiagnosis of acute leukemia: a study of 500 cases. Leukemia, 3, 170-81.

JANTSCH, J., CHAKRAVORTTY, D., TURZA, N., PRECHTEL, A. T., BUCHHOLZ, B., GERLACH, R. G., VOLKE, M., GLASNER, J., WARNECKE, C., WIESENER, M. S., ECKARDT, K. U., STEINKASSERER, A., HENSEL, M. & WILLAM, C. 2008. Hypoxia and hypoxia-inducible factor-1 alpha modulate lipopolysaccharide-induced dendritic cell activation and function. J Immunol, 180, 4697-705.

JAROSZESKI, M. J. & RADCLIFF, G. 1999. Fundamentals of flow cytometry. Mol Biotechnol, 11, 37-53.

JEONG, J. W., BAE, M. K., AHN, M. Y., KIM, S. H., SOHN, T. K., BAE, M. H., YOO, M. A., SONG, E. J., LEE, K. J. & KIM, K. W. 2002. Regulation and destabilization of HIF-1alpha by ARD1-mediated acetylation. Cell, 111, 709-20.

JIANG, B. H., SEMENZA, G. L., BAUER, C. & MARTI, H. H. 1996. Hypoxia-inducible factor 1 levels vary exponentially over a physiologically relevant range of O2 tension. Am J Physiol, 271, C1172-80.

JIANG, B. H., ZHENG, J. Z., LEUNG, S. W., ROE, R. & SEMENZA, G. L. 1997. Transactivation and inhibitory domains of hypoxia-inducible factor 1alpha. Modulation of transcriptional activity by oxygen tension. J Biol Chem, 272, 19253-60.

JIN, W. S., KONG, Z. L., SHEN, Z. F., JIN, Y. Z., ZHANG, W. K. & CHEN, G. F. 2011. Regulation of hypoxia inducible factor-1alpha expression by the alteration of redox status in HepG2 cells. J Exp Clin Cancer Res, 30, 61.

JONES, M., CORDELL, J. L., BEYERS, A. D., TSE, A. G. & MASON, D. Y. 1993. Detection of T and B cells in many animal species using cross-reactive anti-peptide antibodies. J Immunol, 150, 5429-35.

JUNG, F., PALMER, L. A., ZHOU, N. & JOHNS, R. A. 2000. Hypoxic regulation of inducible nitric oxide synthase via hypoxia inducible factor-1 in cardiac myocytes. Circ Res, 86, 319-25.

JUNG, Y., ISAACS, J. S., LEE, S., TREPEL, J., LIU, Z. G. & NECKERS, L. 2003. Hypoxia-inducible factor induction by tumour necrosis factor in normoxic cells requires receptor-interacting protein-dependent nuclear factor kappa B activation. Biochem J, 370, 1011-7.

KALCKAR, H. M. 1941. The Nature of Energetic Coupling in Biological Syntheses. Chemical Reviews, 28, 71-178.

KALLIO, P. J., PONGRATZ, I., GRADIN, K., MCGUIRE, J. & POELLINGER, L. 1997. Activation of hypoxia-inducible factor 1alpha: posttranscriptional regulation and conformational change by recruitment of the Arnt transcription factor. Proc Natl Acad Sci U S A, 94, 5667-72.

KALLIO, P. J., WILSON, W. J., O'BRIEN, S., MAKINO, Y. & POELLINGER, L. 1999. Regulation of the hypoxia-inducible transcription factor 1alpha by the ubiquitin-proteasome pathway. J Biol Chem, 274, 6519-25.

83

KECK, P. J., HAUSER, S. D., KRIVI, G., SANZO, K., WARREN, T., FEDER, J. & CONNOLLY, D. T. 1989. Vascular permeability factor, an endothelial cell mitogen related to PDGF. Science, 246, 1309-12.

KEMPF, V. A., LEBIEDZIEJEWSKI, M., ALITALO, K., WALZLEIN, J. H., EHEHALT, U., FIEBIG, J., HUBER, S., SCHUTT, B., SANDER, C. A., MULLER, S., GRASSL, G., YAZDI, A. S., BREHM, B. & AUTENRIETH, I. B. 2005. Activation of hypoxia-inducible factor-1 in bacillary angiomatosis: evidence for a role of hypoxia-inducible factor-1 in bacterial infections. Circulation, 111, 1054-62.

KIELIAN, T. L. & BLECHA, F. 1995. CD14 and other recognition molecules for lipopolysaccharide: a review. Immunopharmacology, 29, 187-205.

KILANI, M. M., MOHAMMED, K. A., NASREEN, N., TEPPER, R. S. & ANTONY, V. B. 2004. RSV causes HIF-1alpha stabilization via NO release in primary bronchial epithelial cells. Inflammation, 28, 245-51.

KIM, J. O., KIM, J. Y., KWACK, M. H., HONG, S. H., KIM, M. K., KIM, J. C. & SUNG, Y. K. 2010. Identification of troglitazone responsive genes: induction of RTP801 during troglitazone-induced apoptosis in Hep 3B cells. BMB Rep, 43, 599-603.

KITTLICK, P. D. 1986. Inflammation, glycolytic metabolism, and glycosaminoglycans. Exp Pathol, 30, 1-19.

KNOWLES, H. J., RAVAL, R. R., HARRIS, A. L. & RATCLIFFE, P. J. 2003. Effect of ascorbate on the activity of hypoxia-inducible factor in cancer cells. Cancer Res, 63, 1764-8.

KOIVUNEN, P., HIRSILA, M., GUNZLER, V., KIVIRIKKO, K. I. & MYLLYHARJU, J. 2004. Catalytic properties of the asparaginyl hydroxylase (FIH) in the oxygen sensing pathway are distinct from those of its prolyl 4-hydroxylases. J Biol Chem, 279, 9899-904.

KOJIMA, H., GU, H., NOMURA, S., CALDWELL, C. C., KOBATA, T., CARMELIET, P., SEMENZA, G. L. & SITKOVSKY, M. V. 2002. Abnormal B lymphocyte development and autoimmunity in hypoxia-inducible factor 1alpha -deficient chimeric mice. Proc Natl Acad Sci U S A, 99, 2170-4.

KONG, T., ELTZSCHIG, H. K., KARHAUSEN, J., COLGAN, S. P. & SHELLEY, C. S. 2004. Leukocyte adhesion during hypoxia is mediated by HIF-1-dependent induction of beta2 integrin gene expression. Proc Natl Acad Sci U S A, 101, 10440-5.

KOUKOURAKIS, M. I., GIATROMANOLAKI, A., SIVRIDIS, E., SIMOPOULOS, C., TURLEY, H., TALKS, K., GATTER, K. C. & HARRIS, A. L. 2002. Hypoxia-inducible factor (HIF1A and HIF2A), angiogenesis, and chemoradiotherapy outcome of squamous cell head-and-neck cancer. International journal of radiation oncology, biology, physics, 53, 1192-1202.

LANDO, D., PEET, D. J., GORMAN, J. J., WHELAN, D. A., WHITELAW, M. L. & BRUICK, R. K. 2002a. FIH-1 is an asparaginyl hydroxylase enzyme that regulates the transcriptional activity of hypoxia-inducible factor. Genes Dev, 16, 1466-71.

LANDO, D., PEET, D. J., WHELAN, D. A., GORMAN, J. J. & WHITELAW, M. L. 2002b. Asparagine hydroxylation of the HIF transactivation domain a hypoxic switch. Science, 295, 858-61.

LEUNG, D. W., CACHIANES, G., KUANG, W. J., GOEDDEL, D. V. & FERRARA, N. 1989. Vascular endothelial growth factor is a secreted angiogenic mitogen. Science, 246, 1306-9.

LIDGREN, A., HEDBERG, Y., GRANKVIST, K., RASMUSON, T., VASKO, J. & LJUNGBERG, B. 2005. The expression of hypoxia-inducible factor 1alpha is a favorable independent prognostic factor in renal cell carcinoma. Clin Cancer Res, 11, 1129-35.

LINDER, N., MARTELIN, E., LAPATTO, R. & RAIVIO, K. O. 2003. Posttranslational inactivation of human xanthine oxidoreductase by oxygen under standard cell culture conditions. Am J Physiol Cell Physiol, 285, C48-55.

LIPMANN, F. 1941. Metabolic Generation and Utilization of Phosphate Bond Energy. Advances in Enzymology and Related Areas of Molecular Biology.

84

LUKASHEV, D., KLEBANOV, B., KOJIMA, H., GRINBERG, A., OHTA, A., BERENFELD, L., WENGER, R. H. & SITKOVSKY, M. 2006. Cutting edge: hypoxia-inducible factor 1alpha and its activation-inducible short isoform I.1 negatively regulate functions of CD4+ and CD8+ T lymphocytes. J Immunol, 177, 4962-5.

MAECKER, H. T. & TROTTER, J. 2006. Flow cytometry controls, instrument setup, and the determination of positivity. Cytometry A, 69, 1037-42.

MAHON, P. C., HIROTA, K. & SEMENZA, G. L. 2001. FIH-1: a novel protein that interacts with HIF-1alpha and VHL to mediate repression of HIF-1 transcriptional activity. Genes Dev, 15, 2675-86.

MAKINO, Y., CAO, R., SVENSSON, K., BERTILSSON, G., ASMAN, M., TANAKA, H., CAO, Y., BERKENSTAM, A. & POELLINGER, L. 2001. Inhibitory PAS domain protein is a negative regulator of hypoxia-inducible gene expression. Nature, 414, 550-4.

MANGAN, D. F. & WAHL, S. M. 1991. Differential regulation of human monocyte programmed cell death (apoptosis) by chemotactic factors and pro-inflammatory cytokines. J Immunol, 147, 3408-12.

MASSON, N., WILLAM, C., MAXWELL, P. H., PUGH, C. W. & RATCLIFFE, P. J. 2001. Independent function of two destruction domains in hypoxia-inducible factor-alpha chains activated by prolyl hydroxylation. EMBO J, 20, 5197-206.

MAXWELL, P. H., PUGH, C. W. & RATCLIFFE, P. J. 1993. Inducible operation of the erythropoietin 3' enhancer in multiple cell lines: evidence for a widespread oxygen-sensing mechanism. Proc Natl Acad Sci U S A, 90, 2423-7.

MAXWELL, P. H., WIESENER, M. S., CHANG, G. W., CLIFFORD, S. C., VAUX, E. C., COCKMAN, M. E., WYKOFF, C. C., PUGH, C. W., MAHER, E. R. & RATCLIFFE, P. J. 1999. The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature, 399, 271-5.

MELILLO, G., TAYLOR, L. S., BROOKS, A., MUSSO, T., COX, G. W. & VARESIO, L. 1997. Functional requirement of the hypoxia-responsive element in the activation of the inducible nitric oxide synthase promoter by the iron chelator desferrioxamine. J Biol Chem, 272, 12236-43.

MORINET, F., CASETTI, L., FRANCOIS, J. H., CAPRON, C. & PILLET, S. 2013. Oxygen tension level and human viral infections. Virology, 444, 31-6.

MUKHOPADHYAY, C. K., MAZUMDER, B. & FOX, P. L. 2000. Role of hypoxia-inducible factor-1 in transcriptional activation of ceruloplasmin by iron deficiency. J Biol Chem, 275, 21048-54.

NADLER, L. M., RITZ, J., HARDY, R., PESANDO, J. M., SCHLOSSMAN, S. F. & STASHENKO, P. 1981. A unique cell surface antigen identifying lymphoid malignancies of B cell origin. J Clin Invest, 67, 134-40.

NAJAFIPOUR, H. & FERRELL, W. R. 1995. Comparison of synovial PO2 and sympathetic vasoconstrictor responses in normal and acutely inflamed rabbit knee joints. Exp Physiol, 80, 209-20.

NAKAYAMA, K., FREW, I. J., HAGENSEN, M., SKALS, M., HABELHAH, H., BHOUMIK, A., KADOYA, T., ERDJUMENT-BROMAGE, H., TEMPST, P., FRAPPELL, P. B., BOWTELL, D. D. & RONAI, Z. 2004. Siah2 regulates stability of prolyl-hydroxylases, controls HIF1alpha abundance, and modulates physiological responses to hypoxia. Cell, 117, 941-52.

NIELSEN, H. 1987. Isolation and functional activity of human blood monocytes after adherence to plastic surfaces: comparison of different detachment methods. Acta Pathol Microbiol Immunol Scand C, 95, 81-4.

NISHIMURA, M., MITSUNAGA, S. & JUJI, T. 2001. Frozen-stored granulocytes can be used for an immunofluorescence test to detect granulocyte antibodies. Transfusion, 41, 1268-72.

NISHIMURA, R. & OKUDA, K. 2010. Hypoxia is important for establishing vascularization during corpus luteum formation in cattle. J Reprod Dev, 56, 110-6.

85

NIZET, V. & JOHNSON, R. S. 2009. Interdependence of hypoxic and innate immune responses. Nat Rev Immunol, 9, 609-17.

NOORDEN, S. V. 2000. Immunocytochemistry. Methods in molecular medicine, 40, 415-37.

O'ROURKE, J. F., PUGH, C. W., BARTLETT, S. M. & RATCLIFFE, P. J. 1996. Identification of hypoxically inducible mRNAs in HeLa cells using differential-display PCR. Role of hypoxia-inducible factor-1. Eur J Biochem, 241, 403-10.

O'ROURKE, J. F., TIAN, Y. M., RATCLIFFE, P. J. & PUGH, C. W. 1999. Oxygen-regulated and transactivating domains in endothelial PAS protein 1: comparison with hypoxia-inducible factor-1alpha. J Biol Chem, 274, 2060-71.

OBERHOLZER, M., ETTLIN, R. A., KLOPPEL, G., LOHR, M., BUSER, M. W., MIHATSCH, M. J., CHRISTEN, H., MARTI, U. & HEITZ, P. U. 1987. Morphometry and immunocytochemistry. Anal Quant Cytol Histol, 9, 123-32.

OHMANN, H. B., FILION, L. G. & BABIUK, L. A. 1983. Bovine monocytes and macrophages: an accessory role in suppressor-cell generation by Con A and in lectin-induced proliferation. Immunology, 50, 189-97.

PEYSSONNAUX, C., CEJUDO-MARTIN, P., DOEDENS, A., ZINKERNAGEL, A. S., JOHNSON, R. S. & NIZET, V. 2007. Cutting edge: Essential role of hypoxia inducible factor-1alpha in development of lipopolysaccharide-induced sepsis. J Immunol, 178, 7516-9.

PEYSSONNAUX, C., DATTA, V., CRAMER, T., DOEDENS, A., THEODORAKIS, E. A., GALLO, R. L., HURTADO-ZIOLA, N., NIZET, V. & JOHNSON, R. S. 2005. HIF-1alpha expression regulates the bactericidal capacity of phagocytes. J Clin Invest, 115, 1806-15.

PIPER, E. K., JONSSON, N. N., GONDRO, C., LEW-TABOR, A. E., MOOLHUIJZEN, P., VANCE, M. E. & JACKSON, L. A. 2009. Immunological profiles of Bos taurus and Bos indicus cattle infested with the cattle tick, Rhipicephalus (Boophilus) microplus. Clin Vaccine Immunol, 16, 1074-86.

PIPIYA, T., SAUTHOFF, H., HUANG, Y. Q., CHANG, B., CHENG, J., HEITNER, S., CHEN, S., ROM, W. N. & HAY, J. G. 2005. Hypoxia reduces adenoviral replication in cancer cells by downregulation of viral protein expression. Gene Ther, 12, 911-7.

PISANI, D. F. & DECHESNE, C. A. 2005. Skeletal muscle HIF-1alpha expression is dependent on muscle fiber type. J Gen Physiol, 126, 173-8.

POUYSSEGUR, J. & MECHTA-GRIGORIOU, F. 2006. Redox regulation of the hypoxia-inducible factor. Biol Chem, 387, 1337-46.

QUINN, P. J., MARKEY, B. K., LEONARD, F. C., FITZPATRICK, E. S., FANNING, S. & HARTIGAN, P. J. 2011. Veterinary microbiology and microbial disease, Chichester, West Sussex, UK, Wiley-Blackwell.

RICHARD, D. E., BERRA, E. & POUYSSEGUR, J. 2000. Nonhypoxic pathway mediates the induction of hypoxia-inducible factor 1alpha in vascular smooth muscle cells. J Biol Chem, 275, 26765-71.

RIUS, J., GUMA, M., SCHACHTRUP, C., AKASSOGLOU, K., ZINKERNAGEL, A. S., NIZET, V., JOHNSON, R. S., HADDAD, G. G. & KARIN, M. 2008. NF-kappaB links innate immunity to the hypoxic response through transcriptional regulation of HIF-1alpha. Nature, 453, 807-11.

RIVERA-RIVAS, J. J., KISIELA, D. & CZUPRYNSKI, C. J. 2009. Bovine herpesvirus type 1 infection of bovine bronchial epithelial cells increases neutrophil adhesion and activation. Vet Immunol Immunopathol, 131, 167-76.

ROBINSON, A., KEELY, S., KARHAUSEN, J., GERICH, M. E., FURUTA, G. T. & COLGAN, S. P. 2008. Mucosal protection by hypoxia-inducible factor prolyl hydroxylase inhibition. Gastroenterology, 134, 145-55.

86

ROLFS, A., KVIETIKOVA, I., GASSMANN, M. & WENGER, R. H. 1997. Oxygen-regulated transferrin expression is mediated by hypoxia-inducible factor-1. J Biol Chem, 272, 20055-62.

SAADI, S., WRENSHALL, L. E. & PLATT, J. L. 2002. Regional manifestations and control of the immune system. FASEB J, 16, 849-56.

SAGER, H., DAVIS, W. C., DOBBELAERE, D. A. & JUNGI, T. W. 1997. Macrophage-parasite relationship in theileriosis. Reversible phenotypic and functional dedifferentiation of macrophages infected with Theileria annulata. J Leukoc Biol, 61, 459-68.

SALCEDA, S. & CARO, J. 1997. Hypoxia-inducible factor 1alpha (HIF-1alpha) protein is rapidly degraded by the ubiquitin-proteasome system under normoxic conditions. Its stabilization by hypoxia depends on redox-induced changes. J Biol Chem, 272, 22642-7.

SALNIKOW, K., DONALD, S. P., BRUICK, R. K., ZHITKOVICH, A., PHANG, J. M. & KASPRZAK, K. S. 2004. Depletion of intracellular ascorbate by the carcinogenic metals nickel and cobalt results in the induction of hypoxic stress. J Biol Chem, 279, 40337-44.

SANDAU, K. B., FANDREY, J. & BRUNE, B. 2001a. Accumulation of HIF-1alpha under the influence of nitric oxide. Blood, 97, 1009-15.

SANDAU, K. B., ZHOU, J., KIETZMANN, T. & BRUNE, B. 2001b. Regulation of the hypoxia-inducible factor 1alpha by the inflammatory mediators nitric oxide and tumor necrosis factor-alpha in contrast to desferroxamine and phenylarsine oxide. J Biol Chem, 276, 39805-11.

SCHAIBLE, B., SCHAFFER, K. & TAYLOR, C. T. 2010. Hypoxia, innate immunity and infection in the lung. Respir Physiol Neurobiol, 174, 235-43.

SCHOFIELD, C. J. & RATCLIFFE, P. J. 2004. Oxygen sensing by HIF hydroxylases. Nat Rev Mol Cell Biol, 5, 343-54.

SEAGROVES, T. N., RYAN, H. E., LU, H., WOUTERS, B. G., KNAPP, M., THIBAULT, P., LADEROUTE, K. & JOHNSON, R. S. 2001. Transcription factor HIF-1 is a necessary mediator of the pasteur effect in mammalian cells. Mol Cell Biol, 21, 3436-44.

SEMENZA, G. L. 2000. Surviving ischemia: adaptive responses mediated by hypoxia-inducible factor 1. J Clin Invest, 106, 809-12.

SEMENZA, G. L. 2003. Targeting HIF-1 for cancer therapy. Nat Rev Cancer, 3, 721-32.

SEMENZA, G. L. 2005. Pulmonary vascular responses to chronic hypoxia mediated by hypoxia-inducible factor 1. Proceedings of the American Thoracic Society, 2, 68-70.

SEMENZA, G. L., NEJFELT, M. K., CHI, S. M. & ANTONARAKIS, S. E. 1991. Hypoxia-inducible nuclear factors bind to an enhancer element located 3' to the human erythropoietin gene. Proc Natl Acad Sci U S A, 88, 5680-4.

SEMENZA, G. L., ROTH, P. H., FANG, H. M. & WANG, G. L. 1994. Transcriptional regulation of genes encoding glycolytic enzymes by hypoxia-inducible factor 1. J Biol Chem, 269, 23757-63.

SEMENZA, G. L. & WANG, G. L. 1992. A nuclear factor induced by hypoxia via de novo protein synthesis binds to the human erythropoietin gene enhancer at a site required for transcriptional activation. Mol Cell Biol, 12, 5447-54.

SHIBATA, T., GIACCIA, A. J. & BROWN, J. M. 2000. Development of a hypoxia-responsive vector for tumor-specific gene therapy. Gene Ther, 7, 493-8.

SHWEIKI, D., ITIN, A., SOFFER, D. & KESHET, E. 1992. Vascular endothelial growth factor induced by hypoxia may mediate hypoxia-initiated angiogenesis. Nature, 359, 843-5.

87

SIMMEN, H. P., BATTAGLIA, H., GIOVANOLI, P. & BLASER, J. 1994. Analysis of pH, pO2 and pCO2 in drainage fluid allows for rapid detection of infectious complications during the follow-up period after abdominal surgery. Infection, 22, 386-9.

SITKOVSKY, M. & LUKASHEV, D. 2005. Regulation of immune cells by local-tissue oxygen tension: HIF1 alpha and adenosine receptors. Nat Rev Immunol, 5, 712-21.

SITKOVSKY, M. V. 2009. T regulatory cells: hypoxia-adenosinergic suppression and re-direction of the immune response. Trends Immunol, 30, 102-8.

SOPP, P. & HOWARD, C. J. 1997. Cross-reactivity of monoclonal antibodies to defined human leucocyte differentiation antigens with bovine cells. Vet Immunol Immunopathol, 56, 11-25.

SRIKUMARAN, S., KELLING, C. L. & AMBAGALA, A. 2007. Immune evasion by pathogens of bovine respiratory disease complex. Anim Health Res Rev, 8, 215-29.

STERNBERGER, L. A. 1979. Immunocytochemistry, New York, Wiley.

STROBER, W. 2001. Trypan blue exclusion test of cell viability. Curr Protoc Immunol, Appendix 3, Appendix 3B.

SUN, T., AKALIN, A., RODACKER, M. & BRAUN, T. 2004. CD20 positive T cell lymphoma: is it a real entity? J Clin Pathol, 57, 442-4.

SYNNESTVEDT, K., FURUTA, G. T., COMERFORD, K. M., LOUIS, N., KARHAUSEN, J., ELTZSCHIG, H. K., HANSEN, K. R., THOMPSON, L. F. & COLGAN, S. P. 2002. Ecto-5'-nucleotidase (CD73) regulation by hypoxia-inducible factor-1 mediates permeability changes in intestinal epithelia. J Clin Invest, 110, 993-1002.

TAKEDA, N., O'DEA, E. L., DOEDENS, A., KIM, J. W., WEIDEMANN, A., STOCKMANN, C., ASAGIRI, M., SIMON, M. C., HOFFMANN, A. & JOHNSON, R. S. 2010. Differential activation and antagonistic function of HIF-{alpha} isoforms in macrophages are essential for NO homeostasis. Genes Dev, 24, 491-501.

THIEL, M., CALDWELL, C. C., KRETH, S., KUBOKI, S., CHEN, P., SMITH, P., OHTA, A., LENTSCH, A. B., LUKASHEV, D. & SITKOVSKY, M. V. 2007. Targeted deletion of HIF-1alpha gene in T cells prevents their inhibition in hypoxic inflamed tissues and improves septic mice survival. PLoS One, 2, e853.

THIEL, M., CHOUKER, A., OHTA, A., JACKSON, E., CALDWELL, C., SMITH, P., LUKASHEV, D., BITTMANN, I. & SITKOVSKY, M. V. 2005. Oxygenation inhibits the physiological tissue-protecting mechanism and thereby exacerbates acute inflammatory lung injury. PLoS Biol, 3, e174.

THOMAS, D. D., RIDNOUR, L. A., ISENBERG, J. S., FLORES-SANTANA, W., SWITZER, C. H., DONZELLI, S., HUSSAIN, P., VECOLI, C., PAOLOCCI, N., AMBS, S., COLTON, C. A., HARRIS, C. C., ROBERTS, D. D. & WINK, D. A. 2008. The chemical biology of nitric oxide: implications in cellular signaling. Free Radic Biol Med, 45, 18-31.

TIAN, H., HAMMER, R. E., MATSUMOTO, A. M., RUSSELL, D. W. & MCKNIGHT, S. L. 1998. The hypoxia-responsive transcription factor EPAS1 is essential for catecholamine homeostasis and protection against heart failure during embryonic development. Genes Dev, 12, 3320-4.

TIAN, H., MCKNIGHT, S. L. & RUSSELL, D. W. 1997. Endothelial PAS domain protein 1 (EPAS1), a transcription factor selectively expressed in endothelial cells. Genes Dev, 11, 72-82.

TOLNAY, A. E., BASKIN, C. R., TUMPEY, T. M., SABOURIN, P. J., SABOURIN, C. L., LONG, J. P., PYLES, J. A., ALBRECHT, R. A., GARCIA-SASTRE, A., KATZE, M. G. & BIELEFELDT-OHMANN, H. 2010. Extrapulmonary tissue responses in cynomolgus macaques (Macaca fascicularis) infected with highly pathogenic avian influenza A (H5N1) virus. Arch Virol, 155, 905-14.

VAN UDEN, P., KENNETH, N. S. & ROCHA, S. 2008. Regulation of hypoxia-inducible factor-1alpha by NF-kappaB. Biochem J, 412, 477-84.

88

VASSILAKI, N., KALLIAMPAKOU, K. I., KOTTA-LOIZOU, I., BEFANI, C., LIAKOS, P., SIMOS, G., MENTIS, A. F., KALLIAROPOULOS, A., DOUMBA, P. P., SMIRLIS, D., FOKA, P., BAUHOFER, O., POENISCH, M., WINDISCH, M. P., LEE, M. E., KOSKINAS, J., BARTENSCHLAGER, R. & MAVROMARA, P. 2013. Low oxygen tension enhances hepatitis C virus replication. J Virol, 87, 2935-48.

VORDERMARK, D. & BROWN, J. M. 2003. Evaluation of hypoxia-inducible factor-1alpha (HIF-1alpha) as an intrinsic marker of tumor hypoxia in U87 MG human glioblastoma: in vitro and xenograft studies. Int J Radiat Oncol Biol Phys, 56, 1184-93.

WALMSLEY, S. R., PRINT, C., FARAHI, N., PEYSSONNAUX, C., JOHNSON, R. S., CRAMER, T., SOBOLEWSKI, A., CONDLIFFE, A. M., COWBURN, A. S., JOHNSON, N. & CHILVERS, E. R. 2005. Hypoxia-induced neutrophil survival is mediated by HIF-1alpha-dependent NF-kappaB activity. J Exp Med, 201, 105-15.

WALSHE, T. E. & D'AMORE, P. A. 2008. The role of hypoxia in vascular injury and repair. Annu Rev Pathol, 3, 615-43.

WANG, G. L., JIANG, B. H., RUE, E. A. & SEMENZA, G. L. 1995. Hypoxia-inducible factor 1 is a basic-helix-loop-helix-PAS heterodimer regulated by cellular O2 tension. Proc Natl Acad Sci U S A, 92, 5510-4.

WANG, G. L. & SEMENZA, G. L. 1993a. Characterization of hypoxia-inducible factor 1 and regulation of DNA binding activity by hypoxia. J Biol Chem, 268, 21513-8.

WANG, G. L. & SEMENZA, G. L. 1993b. Desferrioxamine induces erythropoietin gene expression and hypoxia-inducible factor 1 DNA-binding activity: implications for models of hypoxia signal transduction. Blood, 82, 3610-5.

WANG, G. L. & SEMENZA, G. L. 1993c. General involvement of hypoxia-inducible factor 1 in transcriptional response to hypoxia. Proc Natl Acad Sci U S A, 90, 4304-8.

WANG, G. L. & SEMENZA, G. L. 1995. Purification and characterization of hypoxia-inducible factor 1. J Biol Chem, 270, 1230-7.

WIENER, C. M., BOOTH, G. & SEMENZA, G. L. 1996. In vivo expression of mRNAs encoding hypoxia-inducible factor 1. Biochem Biophys Res Commun, 225, 485-8.

WIESENER, M. S., TURLEY, H., ALLEN, W. E., WILLAM, C., ECKARDT, K. U., TALKS, K. L., WOOD, S. M., GATTER, K. C., HARRIS, A. L., PUGH, C. W., RATCLIFFE, P. J. & MAXWELL, P. H. 1998. Induction of endothelial PAS domain protein-1 by hypoxia: characterization and comparison with hypoxia-inducible factor-1alpha. Blood, 92, 2260-8.

WU, D. & YOTNDA, P. 2011. Induction and testing of hypoxia in cell culture. J Vis Exp.

XING, D. & BONANNO, J. A. 2009. Hypoxia preconditioning protection of corneal stromal cells requires HIF1alpha but not VEGF. Mol Vis, 15, 1020-7.

YANG, J., FU, Z., FENG, X., SHI, Y., YUAN, C., LIU, J., HONG, Y., LI, H., LU, K. & LIN, J. 2012. Comparison of worm development and host immune responses in natural hosts of Schistosoma japonicum, yellow cattle and water buffalo. BMC Vet Res, 8, 25.

YU, A. Y., FRID, M. G., SHIMODA, L. A., WIENER, C. M., STENMARK, K. & SEMENZA, G. L. 1998. Temporal, spatial, and oxygen-regulated expression of hypoxia-inducible factor-1 in the lung. Am J Physiol, 275, L818-26.

YU, F., WHITE, S. B., ZHAO, Q. & LEE, F. S. 2001. HIF-1alpha binding to VHL is regulated by stimulus-sensitive proline hydroxylation. Proc Natl Acad Sci U S A, 98, 9630-5.

YUAN, Y., HILLIARD, G., FERGUSON, T. & MILLHORN, D. E. 2003. Cobalt inhibits the interaction between hypoxia-inducible factor-alpha and von Hippel-Lindau protein by direct binding to hypoxia-inducible factor-alpha. J Biol Chem, 278, 15911-6.

89

ZHANG, S. X., GOZAL, D., SACHLEBEN, L. R., JR., RANE, M., KLEIN, J. B. & GOZAL, E. 2003. Hypoxia induces an autocrine-paracrine survival pathway via platelet-derived growth factor (PDGF)-B/PDGF-beta receptor/phosphatidylinositol 3-kinase/Akt signaling in RN46A neuronal cells. FASEB J, 17, 1709-11.

ZHOU, J. & BRUNE, B. 2006. Cytokines and hormones in the regulation of hypoxia inducible factor-1alpha (HIF-1alpha). Cardiovasc Hematol Agents Med Chem, 4, 189-97.

ZHOU, J., FANDREY, J., SCHUMANN, J., TIEGS, G. & BRUNE, B. 2003. NO and TNF-alpha released from activated macrophages stabilize HIF-1alpha in resting tubular LLC-PK1 cells. Am J Physiol Cell Physiol, 284, C439-46.

ZINKERNAGEL, A. S., JOHNSON, R. S. & NIZET, V. 2007. Hypoxia inducible factor (HIF) function in innate immunity and infection. J Mol Med, 85, 1339-46.

90

9 Appendices

9.1 Initial T lymphocyte CD3 antibody

Given the relative rarity of bovine-specific antibodies, or antibodies with reliable reports of bovine

cross-reactivity, the selection of ideal antibodies was always going to be a challenge. The initial

antibody selected as a T-cell phenotypic marker was an Allophycocyanin (APC) conjugated mouse

anti-human monoclonal CD3 IgG1 antibody (MCA463APC AbD Serotec, Oxford, UK) APC was a

bright fluorochrome and had minimal spectral overlap with FITC, the original fluorochrome choice

for labelling the HIF-1α antibody.

Titrations with increasing concentrations of antibody on flow cytometry unfortunately failed to

show an emerging distinct population of CD3-APC positive cells with all cells being contained

within one peak on the histogram. With titration of this CD3 antibody there was only a gradual,

slight, generalised increase in fluorescence across all cells (i.e., events); an increase which rose in

proportion with increasing antibody concentration. This pattern was suggestive of non-specific

binding. This apparent absence of true staining was confirmed by examining cells stained with

moderate CD3-APC antibody concentration mixed with aliquot of unstained cells from the same

animal at a ratio of 1:1 using the flow cytometer. The absence of a second, slightly higher distinct

peak of APC positive cells, or at least a change in shape of the upper contour of the peak

demonstrated that a known unstained population of cells had the same CD3-APC fluorescence

profile as a known stained population, thereby confirming that the ‘stained’ population was likely

background staining rather than target staining, i.e., the CD3-APC antibody was not binding to any

specific subpopulation of cells present.

9.2 CD20 antibody titration on flow cytometry

Originally, this antibody was used in unconjugated form as a primary antibody, followed by

application of a secondary goat anti-mouse polyclonal IgG antibody (Biolegend, San Diego, CA,

USA) conjugated commercially to the fluorochrome APC-Cy7. However, such use of a secondary

antibody was discontinued due to on-going non-specific staining results when titrating both the

primary and secondary antibodies. Bovine serum albumin was employed as a blocking agent to

block Fc receptors on the cell surface to attempt to decrease non-specific binding of the secondary

antibody in the staining process in an attempt to improve specificity, but results did not improve. It

became apparent that use of secondary conjugated antibodies was too problematic when attempting

to create a multicolour flow cytometry protocol.

91

9.3 Initial CD14 antibody

The initially chosen phenotypic antibody for monocytes was a Pacific Blue conjugated mouse anti-

human monoclonal CD14 IgG2a antibody (clone TÜK4) (AbD Serotec, Oxford, UK). This clone

was selected as it had recorded cross-reactivity with bovine CD14 for flow cytometry (Sopp and

Howard, 1997). Initial titration showed an emerging positive sub-population of cells however

further dual-staining with the already confirmed CD3 - Alexa Fluor® 647 antibody showed that

80% of the cells positive for CD14-Pac Blue were also positive for CD3 - Alexa Fluor® 647 (figure

4.4).

In order to identify the constituent cells in the population, Cells positive for both CD3 and CD14

were gated and plotted on a FS/SS scatter plot in linear scale. The majority of events within the

lymphocyte region were positive, with lesser numbers of cells within the monocyte area staining, as

well as debris (graph not shown). Given that the CD3 antibody had already been titrated

successfully, had specifically stained T lymphocytes on IHC, and that the majority of cells in the

sample were expected to be lymphocytes, a specificity problem with the CD14 antibody seemed the

most likely explanation. This was further supported by lack of positive staining in the monocyte

region. Further IHC testing using this CD14 TÜK4 clone at varying concentrations and following

use of EDTA, citrate and proteinase K target retrieval systems failed to detect clear positivity in

monocytes or other cells.

Figure 4.4. Bivariate scatter plot of CD3 fluorescent signal versus CD14 fluorescence, divided into quadrants, log scale. The negative quadrant was set based on unstained cells. The majority of events fall within the negative quadrant or within the ‘both CD14/CD3’ quadrant, indicating double-staining of these events with both CD14 and CD3 antibody. 9.3.1.1 Alternative monocyte CD14 antibody

An alternative mouse monoclonal IgG1 antibody clone MM61A (VMRD, Pullman WA, USA) had

been used successfully to label bovine monocytes for flow cytometry (Piper et al., 2009) (Sager et

92

al., 1997). Titration on uncultured cells labeled with a Pacific Blue fluorophore (using the same

labeling kit as for CD20 for a labeled antibody concentration of 15 µg/ml) showed a positive peak

emerging at 50 µl of antibody (figure 4.5A) and a relatively distinct positive population on the

SS/CD14 scatter plot (figure 4.5B). Workup was discontinued at this point due to removal of

monocytes from the flow analysis.

Figure 4.5. A: Histogram of CD14 fluorescence of events with the cells gate, log scale. A peak of cells positive for CD14 can be visualised however is not separate from the negative peak. The negative linear boundary (black line) was set using unstained cells. The positive linear boundary (black line) was set using the CD3 positive gate described in 4.5B. B: Scatter plot of CD14 fluorescence versus SS, log scale. A population of cells positive for CD14 is visualised and gated by eye (black border).