Neuron-intrinsic NF-κB Signaling Mediates Reovirus Virulence...2020/08/05 · 25 692-8071. E-mail...

1

Neuron-intrinsic NF-κB Signaling Mediates Reovirus 1

Virulence 2

3

Andrea J. Pruijssers1,2§, Gwen Taylor3,4§, Pamela Brigleb4,5, Pengcheng Shang3,4, Kelly 4

Urbanek3,4, Judy J. Brown2,6, and Terence S. Dermody3,4,5* 5

6

1 Department of Pediatrics, Vanderbilt University School of Medicine, Nashville, 7

Tennessee, USA 8

2 Elizabeth B. Lamb Center for Pediatric Research, Vanderbilt University Medical 9

Center, Nashville, Tennessee, USA 10

3 Department of Pediatrics, University of Pittsburgh School of Medicine, Pittsburgh, 11

Pennsylvania, USA 12

4 Center for Microbial Pathogenesis, UPMC Children’s Hospital of Pittsburgh, 13

Pittsburgh, Pennsylvania, USA 14

5 Department of Microbiology and Molecular Genetics, University of Pittsburgh School of 15

Medicine, Pittsburgh, Pennsylvania, USA 16

6 Department of Pathology, Microbiology, and Immunology, Vanderbilt University 17

Medical Center, Nashville, Tennessee, USA 18

19

§These authors contributed equally. 20

21

Corresponding author: 22

.CC-BY-NC-ND 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made

The copyright holder for this preprintthis version posted August 6, 2020. ; https://doi.org/10.1101/2020.08.05.238220doi: bioRxiv preprint

https://doi.org/10.1101/2020.08.05.238220

http://creativecommons.org/licenses/by-nc-nd/4.0/

2

*Terence S. Dermody, UPMC Children’s Hospital of Pittsburgh, Administrative Office 23

Building Suite 5300, 4401 Penn Avenue, Pittsburgh, PA 15224. Phone number: (412) 24

692-8071. E-mail address: [email protected] 25

26

Running title: 27

Role of p65 in reovirus neuropathogenesis 28

29

Keywords: 30

Reovirus, NF-B, p65, RelA, encephalitis, neurovirulence 31

32

Conflict of Interest: 33

The authors declare that no conflicts of interest exist. 34



mailto:[email protected]

https://doi.org/10.1101/2020.08.05.238220


3

ABSTRACT 35

Pathological effects of apoptosis associated with viral infections of the central nervous 36

system are an important cause of morbidity and mortality. Reovirus is a neurotropic 37

virus that causes apoptosis in neurons, leading to lethal encephalitis in newborn mice. 38

Reovirus-induced encephalitis is diminished in mice with germline ablation of NF-κB 39

subunit p50. It is not known whether the pro-apoptotic function of NF-κB is mediated by 40

neuron-intrinsic processes, NF-κB-regulated cytokine production by inflammatory cells, 41

or a combination of both. To determine the contribution of cell type-specific NF-κB 42

signaling in reovirus-induced neuronal injury, we established mice that lack NF-κB p65 43

expression in neurons using the Cre/loxP recombination system. Following intracranial 44

inoculation of reovirus, 50% of wild-type (WT) mice succumbed to infection, whereas 45

more than 90% of mice lacking neural NF-κB p65 (Nsp65-/-) mice survived. While viral 46

loads in brains of WT and Nsp65-/- were comparable, histological analysis revealed that 47

reovirus antigen-positive areas in the brain of WT mice displayed enhanced cleaved 48

caspase-3 immunoreactivity, a marker of apoptosis, compared with Nsp65-/- mice. 49

These data suggest that neuron-intrinsic NF-κB-dependent factors are essential 50

mediators of reovirus neurovirulence. RNA sequencing analysis of reovirus-infected 51

cortices of WT and Nsp65-/- mice suggests that NF-κB activation in neurons upregulates 52

genes involved in innate immunity, inflammation, and cell death following reovirus 53

infection. A better understanding of the contribution of cell type-specific NF-κB-54

dependent signaling to viral neuropathogenesis could inform development of new 55

therapeutics that target and protect highly vulnerable cell populations 56

57



https://doi.org/10.1101/2020.08.05.238220


4

INTRODUCTION 58

Many neurotropic viruses activate the NF-B pathway during infection, which can 59

contribute to cell survival and evasion of the immune response or elicit apoptosis and 60

mediate viral spread (1). The role of NF-B signaling in the central nervous system 61

(CNS) is cell-type dependent. In neurons, NF-B signaling maintains overall neuronal 62

homeostasis, synapse growth and function, and neuroprotection under disease 63

conditions. In glial cells, NF-B signaling also contributes to maintaining neuronal cell 64

health, although chronic NF-B activation is neurotoxic (2). The factors that determine 65

cell fate following viral induced-NF-B activation in the CNS are not well understood. 66

Mammalian orthoreoviruses (reoviruses) are nonenveloped viruses with a 67

segmented, double-stranded RNA genome that infect a variety of mammalian species 68

but cause disease only in the very young. Reovirus is transmitted via the fecal-oral route 69

and disseminates from the intestine to virtually all organs via hematogenous or neural 70

routes. Infection of newborn mice with serotype 3 reovirus strains causes neuronal 71

apoptosis, which leads to lethal encephalitis. Following cell entry, reovirus activates NF-72

B using a mechanism that requires IKK and IKK ((3)) and subsequent nuclear 73

translocation of the canonical p50 and p65 NF-B subunits ((3, 4)). NF-B activation 74

results in the expression of interferon-stimulated genes (ISGs), cytokines, chemokines, 75

and other genes involved in antiviral responses and cell proliferation (5). Mouse embryo 76

fibroblasts (MEFs) lacking either p50 or p65 display reduced levels of apoptosis 77

following reovirus infection compared with wild-type (WT) MEFs (4), indicating that NF-78

B subunits p50 and p65 mediate reovirus-induced apoptosis. 79



https://doi.org/10.1101/2020.08.05.238220


5

The role of p50 in reovirus virulence in newborn mice is organ-specific. In mice 80

with germline ablation of NF-κB subunit p50 (p50-/- mice), reovirus neurovirulence and 81

neuronal apoptosis are diminished relative to WT mice. In contrast, p50-/- mice succumb 82

to severe myocarditis coupled with increased viral replication in the heart (6). These 83

results suggested that NF-κB activation produces different outcomes depending on the 84

cell type infected, a pro-apoptotic function in neurons versus an anti-apoptotic function 85

in the heart. It is unclear whether the pro-apoptotic function of NF-κB is mediated by 86

neuron-intrinsic processes, a result of NF-κB-regulated cytokine production by 87

inflammatory cells, or a combination of both. 88

In this study, we established mice lacking NF-κB in neurons and investigated 89

reovirus neuropathogenesis in these animals and WT mice. We discovered that mice 90

with a neuron-specific deletion in the NF-B p65-encoding gene (Nsp65-/- mice) display 91

increased survival following reovirus infection relative to WT mice and are likewise 92

protected from brain injury. Brain regions in WT mice that harbor reovirus antigen also 93

contain the cleaved (activated) form of the executioner apoptotic caspase, caspase-3. 94

However, cleaved caspase-3 staining is markedly reduced in brains of Nsp65-/- mice. To 95

elucidate how NF-κB signaling in the CNS contributes to reovirus neuropathogenesis, 96

we compared reovirus-induced transcriptional changes in brains of WT and Nsp65-/- 97

mice. Genes involved in innate immunity, inflammation, and cell death were identified as 98

NF-κB pathways upregulated in the murine brain upon reovirus infection. These data 99

identify potential therapeutic targets for development of antiviral agents to limit virus-100

induced neural injury. 101

102



https://doi.org/10.1101/2020.08.05.238220


6

RESULTS 103

Establishment of neuron-specific p65-deficient mice. To determine the 104

contribution of neuron specific NF-κB signaling in reovirus-induced encephalitis in vivo, 105

we crossed mice expressing Cre recombinase under control of the rat nestin promoter 106

and enhancer to mice with loxP sites flanking exons 5 - 8 of p65/RelA (p65 f/f) (7). 107

Nestin is expressed in neuronal and glial precursor cells. Mice that carry the nestin Cre 108

transgene and homozygous for the p65 floxed allele (neuron-specific p65-deficient 109

[Nsp65-/-] mice) were present in the third generation (Fig. 1 A). Immunoblot analysis 110

revealed that p65 expression was diminished in brains of Nsp65-/- but not in WT p65 f/f 111

mice (Fig. 1 B). Expression of p65 was not diminished in tissues that do not express 112

nestin, demonstrating that p65 expression is conditionally ablated in brain tissue of 113

Nsp65-/- mice. 114

115

Apoptosis induced by reovirus infection is reduced in Nsp65-/- cortical 116

neurons. To determine whether neuron-intrinsic NF-κB-dependent factors contribute to 117

reovirus-induced apoptosis, we infected cultures of WT, p50-deficient (p50-/-), and 118

Nsp65-/- cortical neurons with potently apoptotic reovirus strain T3SA+ and quantified 119

the percentage of apoptotic neurons following acridine orange staining (Fig. 2 A, B). A 120

significantly higher percentage of WT neurons were observed to undergo apoptosis 121

following reovirus infection compared with p50-/- and Nsp65-/- neurons, suggesting that 122

the pro-apoptotic function of NF-κB in the CNS is at least in part determined by neuron-123

intrinsic factors. 124



https://doi.org/10.1101/2020.08.05.238220


7

Expression of NF-B in neurons is required for reovirus virulence. To 125

determine whether neuron-specific p65 contributes to reovirus neurovirulence, newborn 126

WT and Nsp65-/- mice were inoculated intracranially with T3SA+ and monitored for 127

survival for 21 d (Fig. 3 A). While more than 90 percent of Nsp65-/- mice survived T3SA+ 128

infection, only 50 percent of WT mice survived, suggesting that p65 expression in 129

neurons enhances reovirus neurovirulence. 130

To determine whether neuron-specific NF-κB influences reovirus replication and 131

pathology, newborn WT and Nsp65-/- mice were inoculated intracranially with T3SA+. At 132

6, 8, and 10 d post-inoculation, mice were euthanized, and brains were resected. Viral 133

loads detected in the left hemispheres of WT and Nsp65-/- mice were comparable at all 134

time points tested (Fig. 3 B). Serial coronal sections from embedded, titer-matched right 135

hemispheres at 6 d post-inoculation were stained with hematoxylin and eosin (H&E) or 136

immunostained with reovirus polyclonal antiserum or an antibody specific for the 137

cleaved form of caspase-3 to monitor for cell death (Fig. 3 C). Areas in the brains of WT 138

mice positive for reovirus antigen displayed enhanced cleaved caspase-3 139

immunoreactivity, whereas reovirus antigen-positive areas in the brains of Nsp65-/- mice 140

displayed low levels of cleaved caspase-3 immunoreactivity. Together, these data 141

suggest that neuron-intrinsic NF-κB-dependent factors are not required for reovirus 142

replication in the brain but are essential mediators of reovirus neurovirulence. 143

144

NF-κB-dependent changes in mRNA expression in neurons following 145

reovirus infection. To identify mediators of NF-κB-dependent, reovirus-induced 146

apoptosis in neurons, we inoculated two-day-old WT and Nsp65-/- mice intracranially 147



https://doi.org/10.1101/2020.08.05.238220


8

with reovirus T3SA+ or PBS (mock). At 2 and 6 d post-inoculation, mice were 148

euthanized, and cortices were micro-dissected, homogenized, and processed for viral 149

titer via plaque assay and RNA purification. Transcript abundance in samples from 150

infected WT and Nsp65-/- mice, matched for viral titer, as well as those from PBS-151

inoculated controls was determined by mRNA sequencing. Three samples of RNA for 152

each genotype per time point were subjected to paired-end RNA sequencing. We 153

identified genes that were significantly up or down regulated in a reovirus- (897 genes), 154

p65- (140 genes), and a reovirus- and p65- (814 genes) dependent manner and ranked 155

the genes for abundance using DESeq2 (see methods) (Fig. 4A). The top 790 genes 156

were uploaded to Ingenuity Software (Qiagen, Inc), which considers the adjusted p 157

values of the analyses, and gene networks containing more than one candidate in the 158

analysis were identified (Fig. 4B). Examples of gene networks that contained multiple 159

candidates were related to inflammation, antimicrobial response, immune cell 160

development, and cell death (Fig. 4B). Representative genes under direct control of p65 161

that were differentially induced in infected brain tissue were predominantly in groups 162

involved in innate immunity, inflammation and cell death (Fig. 4C). 163

164

Reovirus infection in the brain induces a p65-dependent upregulation of 165

proinflammatory cytokines. To validate at the protein level the pro-inflammatory 166

cytokines that were upregulated at the transcript level in the presence of neuronal p65 167

(Fig. 4C), three-day-old WT and Nsp65-/- mice were inoculated intracranially with T3SA+ 168

or PBS. At 6 d post-inoculation, brains were harvested, homogenized, and processed 169

for Luminex and plaque assay. Brain homogenates from WT and Nsp65-/- mice, 170



https://doi.org/10.1101/2020.08.05.238220


9

matched for viral titer, were analyzed for CCL2, CCL4, CCL5, CCL7, CCL12, and 171

CXCL11 protein levels (Fig. 5). Levels of CCL2, CCL4, CCL5, CCL7, CCL12, and 172

CXCL11 were upregulated in a reovirus- and p65-dependent manner, with statistically 173

significant differences between WT and Nsp65-/- mice observed for CCL4, CCL5, and 174

CXCL11. These data validate our gene expression data and demonstrate that cytokine 175

production during reovirus infection is dependent on NF-B expression in neurons. 176

177

Proinflammatory cytokines are upregulated in reovirus-infected primary 178

cortical neurons in an NF-B dependent manner. While p65 expression is 179

diminished in cells containing nestin in the brains of Nsp65-/- mice, including neurons 180

and microglia, p65 is expressed in cells that lack nestin, raising the possibility that 181

surrounding cells in the brain contribute to the observed differences in cytokine 182

production. To determine whether reovirus infection of neurons is sufficient to induce an 183

increase in cytokine transcript levels in an NF-B-dependent manner, we infected WT 184

and Nsp65-/- primary cortical neurons with T3SA+ infectious subvirion particles (ISVPs), 185

a reovirus disassembly intermediate, that efficiently infects primary neuronal cultures 186

(8). Transcript levels were detected by RT-PCR at 3, 6, 16, and 24 h post adsorption 187

(Fig. 6). The level of S4 gene transcripts was used as a surrogate to quantify viral 188

replication. At 6, 16, and 24 h post-adsorption, S4 gene transcript levels were 189

significantly greater in infected Nsp65-/- cortical neurons relative to WT cortical neurons, 190

suggesting an NF-B-dependent decrease in viral replication. However, we observed an 191

NF-B-dependent increase in transcript levels of Ccl2, Ccl4, Ccl7, and Cxcl9. 192

Interestingly the increases in cytokine transcript levels can be divided into three groups 193



https://doi.org/10.1101/2020.08.05.238220


10

based on kinetics of expression: Ccl7 increased at 3 h post-adsorption and rapidly 194

declined by 6 h post-adsorption (immediate early); Ccl2 and Ccl4 increased at 3 h post-195

adsorption but slowly declined over 13 h (early), and Cxcl9 increased at 16 to 24 h post-196

adsorption (late). These data demonstrate that reovirus infection of primary cortical 197

neurons upregulates cytokine gene expression in an NF-B-dependent manner, 198

suggesting that differences in cytokine levels observed in the brains of infected mice are 199

at least in part attributable to NF-B activation in neurons. 200

201



https://doi.org/10.1101/2020.08.05.238220


11

DISCUSSION 202

The NF-B pathway is central to a variety of cellular processes including the 203

regulation of cell cycle and cell fate, initiation of innate immune responses, and 204

production of cytokines and chemokines. Several neurotropic viruses including 205

enterovirus 71 (9, 10), herpes simplex virus-1 (HSV-1) (11), human immunodeficiency 206

virus-1 (12), reovirus (4, 13), Sindbis virus (14), Venezuelan equine encephalitis virus 207

(VEEV) (15), vesicular stomatitis virus (16), and West Nile virus (WNV) (17) activate 208

NF-B, which can lead to cell survival or apoptosis depending on the virus strain and 209

cell type infected. Apoptosis induced by serotype 3 reovirus contributes to 210

neuropathogenesis and is dependent on NF-B activation (6). However, genes 211

activated by NF-B in response to reovirus infection in the CNS are unknown. In this 212

study, we identified neuron-specific, reovirus-induced changes in the expression of 213

genes under NF-B control by establishing mice in which NF-B p65 expression is 214

ablated in neurons. We found that reovirus infection in the brain leads to upregulation of 215

NF-B-dependent host pathways involved in innate immunity, inflammation, and cell 216

death. 217

Our previous studies using mice with germline deletion of NF-B p50 indicate 218

that NF-B is required for reovirus to cause encephalitis but protects against reovirus 219

myocarditis (6). To determine whether reovirus-induced neural injury is mediated by the 220

activation of NF-B in neurons, immune cells, or a combination of both, we established 221

Nsp65-/- mice in which NF-B p65 is deleted in neurons (Fig 1). Nsp65-/- mice displayed 222

increased survival following reovirus infection relative to WT mice and were protected 223



https://doi.org/10.1101/2020.08.05.238220


12

from severe brain injury (Fig. 3). These data suggest that neuron-intrinsic NF-κB-224

dependent factors are essential for reovirus neurovirulence. 225

An NF-B-dependent gene network induced by reovirus infection in HeLa cells 226

revealed temporal expression of distinct gene clusters, a majority of which are involved 227

in innate immune responses and IFN signaling (5). However, genes activated by NF-B 228

can differ depending on the agonist or cell type (2, 18). We defined an NF-B-229

dependent gene network induced by reovirus in the brain by RNA sequencing and 230

observed that a majority of the genes identified are involved in innate immune 231

responses and cell death (Fig. 4), similar to the findings made using HeLa cells (5). 232

In studies of a variety of neurotropic viruses, including dengue virus (19), 233

Japanese encephalitis virus (JEV) (20), reovirus (21), VEEV (22), and WNV (20), the 234

expression of genes encoding components of IFN signaling is most commonly altered 235

(20). For example, JEV, reovirus, and WNV induce the expression of 86 common ISGs 236

in the brain (20). These genes function in inflammation, immune signaling, virus 237

recognition, and the antiviral response. Of these 86 ISGs common to neurotropic viral 238

infections, we identified 72 that are upregulated during reovirus infection in WT mice. 239

However, only 43 of these genes are expressed in response to NF-B activation (Fig. 240

4). Interestingly, genes we identified under NF-B control function in inflammation and 241

immune signaling, while those that serve roles in viral recognition and the antiviral 242

response are not NF-B-dependent. 243

Neurotropic viral infections also induce elaboration of chemokines and 244

recruitment of immune cells, which contribute to neuropathology (20, 23). We identified 245

several chemokine genes that are expressed in response to reovirus infection and NF-246



https://doi.org/10.1101/2020.08.05.238220


13

B activation, including Ccl2, Ccl4, Ccl5, Ccl7, Ccl12, CxclL9, and Cxcl11 (Fig. 4). 247

Furthermore, we found that chemokine levels also are upregulated in response to 248

reovirus infection and NF-B activation, with statistically significant increases observed 249

for CCL4, CCL5, and CXCL11 (Fig. 5) Reovirus infection also upregulates CXCL10 250

both at the transcript and protein level, but this effect is independent of NF-B 251

expression (data not shown). Since transcript and protein levels were quantified at 6 d 252

post-inoculation, it is possible that peak levels of CCL2, CCL7, and CCL12 are 253

produced earlier in the infectious course and thus were not significantly increased in our 254

analysis. Chemokines upregulated in the brain following reovirus infection also are 255

upregulated by other neurotropic viruses. Following CNS infection, HSV-1 induces 256

CCL2, CCL3, CCL5, CXCL9, and CXCL10, and JEV and WNV induce CCL2, CCL7, 257

CXCL9, and CXCL10 (24). The similarities in chemokine responses elicited by 258

neurotropic viruses suggest similar underlying pathological mechanisms and potentially 259

common therapeutic targets. 260

Chemokines released during neurotropic viral infections can be produced by 261

resident CNS cells or infiltrating immune cells (20, 25). Neurons are the primary source 262

of chemokines following WNV infection, while astrocytes and microglia are the primary 263

source following infection with HIV-1 (26) or HSV-1 (27). We hypothesized that neurons 264

would be the primary source of chemokine production during reovirus infection, as these 265

cells are heavily infected by reovirus in the brain. Using primary cortical neuron cultures 266

prepared from WT and Nsp65-/- mice, we found that transcript levels of Ccl2, Ccl4, Ccl7, 267

Ccl12, and Cxcl9 were increased in an NF-B-dependent manner during reovirus 268

infection, while those of Ccl5 and Cxcl11 were not. These data suggest that while 269



https://doi.org/10.1101/2020.08.05.238220


14

reovirus-infected neurons contribute to the observed increase in chemokines in the 270

brain, only a subset of these chemokines are dependent on NF-B expression in 271

neurons. Therefore, other cell types appear to modulate chemokine responses in the 272

brain during reovirus infection. Since nestin is expressed in both neurons and microglia, 273

NF-B p65 is ablated in neurons and microglia in Nsp65-/- mice. We think it possible that 274

expression of Ccl5 and Cxcl11 in the CNS following reovirus infection is dependent on 275

NF-B activation in microglia rather than neurons. 276

Neurotropic viruses induce expression of highly orchestrated and distinct 277

patterns of chemokines that recruit specific immune cell populations to sites of infection 278

and contribute to disease. We observed that enhanced expression of the chemokine 279

genes could be grouped into immediate-early (Ccl7), early (Ccl2, Ccl4, Ccl12), and late 280

(Cxcl9) classes dependent on the kinetics of expression following reovirus infection of 281

cortical neurons (Fig. 6). This time-dependent gene expression observed in vitro raises 282

the possibility that expression of chemokines in vivo during reovirus infection of the CNS 283

also is kinetically regulated. Understanding how this chemokine expression pattern is 284

involved in immune cell infiltration and neuronal cell death is essential to understanding 285

the role of NF-B in reovirus neuropathogenesis. 286

Our studies enhance an understanding of the function of NF-B in viral 287

neuropathogenesis. We identified NF-B-dependent pathways induced by reovirus 288

infection in the CNS and discovered that expression of genes involved in immune 289

signaling also are induced by other neurotropic viruses. Additionally, we identified a 290

pattern of chemokine expression activated by reovirus in the brain that is in part 291

recapitulated by reovirus-infected neurons. Collectively, these findings suggest that 292



https://doi.org/10.1101/2020.08.05.238220


15

therapeutic interruption of the chemokine networks activated by reovirus may have 293

broad utility against neurotropic viruses. 294



https://doi.org/10.1101/2020.08.05.238220


16

MATERIALS AND METHODS 295

Mice. Control p50+/+ (B6129PF1/J-AW-J/AW) and p50-/- (B6129P-Nfkb1tm1Bal) mice 296

(6, 28) and nestin Cre (B6.Cg-Tg(Nes-cre)1Kln/J) mice were obtained from Jackson 297

Laboratory. RelAF/F mice were provided by Dr. Albert Baldwin (29). Experiments were 298

conducted using animal biosafety 2 (ABSL2) facilities and guidelines. All animal 299

husbandry and experimental procedures were performed in accordance with U.S. Public 300

Health Service policy and approved by the Institutional Animal Care and Use 301

Committees at Vanderbilt University and the University of Pittsburgh. 302

303

Establishment of neuron-specific p65-deficient mice. RelAF/F (p65 f/f) mice 304

containing loxP sites flanking exons 5-8 of RelA, were interbred with mice expressing 305

Cre recombinase under control of the rat nestin promoter and enhancer (Stock No: 306

003771: B6.Cg-Tg(Nes-cre)1Kln/J (30) to establish mice heterozygous for floxed RelA 307

and the nestin Cre transgene (RelA+/F; nestin-Cre+/-). Mice carrying the nestin Cre 308

transgene and homozygous for the RelA floxed allele (neuron-specific p65-deficient 309

[Nsp65-/-] mice) were present in the third generation. For experimental breeding, Nsp65-310

/- males were crossed with RelAF/F females to establish litters with 50% of pups carrying 311

the nestin Cre transgene (Nsp65-/-) and 50% without the nestin Cre transgene (WT). All 312

experimental mice were genotyped with respect to RelA and Cre following necropsy. 313

314

Cells and viruses. Spinner-adapted murine L929 (L) cells were maintained in 315

Joklik’s modified Eagle’s minimal essential medium (US Biological, M3867) 316

supplemented to contain 5% fetal bovine serum (FBS; VWR, 97068-085), 2 mM L-317



https://doi.org/10.1101/2020.08.05.238220


17

glutamine, 100 U/ml penicillin, 100 µg/ml streptomycin, and 25 ng/ml amphotericin B. 318

Primary neuronal cultures were derived from cortices of p50+/+ (B6129PF1/J-AW-J/AW), 319

p50-/- (B6129P-Nfkb1tm1Bal), Nsp65-/- (RelAF/F; nestin-Cre+/-), or WT (RelAF/F; nestin-Cre-/-) 320

day 15 embryos as described (31). Neurons were plated at a density of 105 cells/well in 321

24-well plates (Greiner) pre-treated with 10 mg/ml poly-D-lysine hydrobromide (Sigma-322

Aldrich, P0899) and 1.64 mg/ml laminin (Corning, 354232) diluted in Neurobasal 323

medium (Gibco, 21103049). Cultures were incubated for 24 h in Neurobasal medium 324

supplemented to contain 10% FBS, 0.6 mM GlutaMAX, 50 U/ml penicillin, and 50 µg/ml 325

streptomycin. Cultures thereafter were maintained in Neurobasal medium supplemented 326

to contain 1 x B-27 (Gibco, 35050079), 0.6 mM GlutaMAX, 50 U/ml penicillin, and 50 327

µg/ml streptomycin. One-half of the medium volume was replaced with fresh medium 328

every 3 to 4 d. Neurons were cultivated for 5 to 7 d prior to use. 329

Purified reovirus strain T3SA+ was prepared from second- or third-passage L-cell 330

lysate stocks (32). Viral particles were Vertrel XF-extracted from infected cell lysates, 331

layered onto 1.2- to 1.4-g/cm3 CsCl gradients, and centrifuged at 62,000 × g for 16 h. 332

Bands corresponding to virions (1.36 g/cm3) were collected and dialyzed in virion-333

storage buffer (150 mM NaCl, 15 mM MgCl2, 10 mM Tris-HCl [pH 7.4]) (33). The 334

concentration of reovirus virions in purified preparations was determined from an 335

equivalence of one OD unit at 260 nm equals 2.1 × 1012 virions (33). Viral titer was 336

determined by plaque assay using L cells (34). ISVPs were prepared by treating 2 × 337

1011 virion particles with 20 μg of α-chymotrypsin (Sigma-Aldrich) in a 100-μl volume of 338

virion storage buffer at 35°C for 60 min (35). Reactions were terminated by the addition 339



https://doi.org/10.1101/2020.08.05.238220


18

of 2 mM phenylmethylsulphonyl fluoride (Sigma-Aldrich). ISVP titer was determined by 340

plaque assay using L cells. 341

342

SDS-PAGE and immunoblotting. Samples for SDS-PAGE were diluted in 5 x 343

Laemmli sample buffer (BioRad) containing 10% -mercaptoethanol, incubated at 95°C 344

for 5 min, and resolved in 10% Mini-Protean TGX gels (Bio-Rad). Proteins were 345

transferred to nitrocellulose membranes (Bio-Rad) at 100 V at 4°C for 1 h and 346

immunoblotted using rabbit anti-p65 (Abcam, ab16502), cleaved caspase-3, (Cell 347

Signaling Technology, 9664), and mouse anti-GAPDH (Sigma-Aldrich, G8795). 348

Secondary antibodies IRDye® 680RD goat anti-rabbit IgG and IRDye® 800CW goat 349

anti-mouse IgG (Li-Cor) were used for detection. Membranes were scanned using an 350

Odyssey CLx imaging system (Li-Cor), and pixel intensities of the bands were quantified 351

using Image Studio Software (Li-Cor). 352

353

Quantification of apoptosis by acridine orange (AO) staining. Neurons were 354

adsorbed with a multiplicity of infection (MOI) of 1000 reovirus ISVPs per cell diluted in 355

PBS at 37°C for 1 h. The inoculum was removed, fresh medium was added, and 356

neurons were incubated at 37°C for 24 or 48 h. The percentage of apoptotic cells was 357

determined using AO staining as described (36). Images were collected from > 600 cells 358

in two to three fields of view per well by epi-illumination fluorescence microscopy using 359

a fluorescein filter set (Zeiss Photomicroscope III). 360

361



https://doi.org/10.1101/2020.08.05.238220


19

Infection of mice. Two-to-three-day-old WT or Nsp65-/- mice weighing 1.6 to 2.3 362

g were inoculated intracranially in the right cerebral hemisphere with 5 l of 2 plaque-363

forming units (PFU)/l of purified T3SA+ diluted in PBS using a 30-gauge needle and 364

syringe (Hamilton). Titers of virus in the inoculum were determined to confirm the 365

number of infectious particles in the final dilution. The results of these assays are 366

reported as the inoculating viral dose. For analysis of virulence, mice were monitored for 367

symptoms of disease for 21 d post-inoculation and euthanized when moribund, as 368

defined by severe lethargy or seizures, paralysis, or loss of 25% of body weight. Death 369

was not used as an endpoint in our experiments. 370

Viral replication and immunohistopathology were analyzed at 6 d post-371

inoculation. Mice were euthanized, and brains were removed and hemisected along the 372

longitudinal fissure. The left hemisphere was collected in 1 ml PBS, frozen and thawed 373

once, and homogenized using a TissueLyser (Qiagen, Inc.). Viral titers in brain 374

homogenates were determined by plaque assay using L929 cells (37). The right 375

hemisphere was submerged in 10% buffered formalin at room temperature for 24 to 120 376

h, embedded in paraffin, and cut into consecutive 6-μm sections. Consecutive sections 377

were processed as described (37) for evaluation of histopathologic changes by H&E 378

staining or for immunohistochemical detection of reovirus antigen (1:45,000 dilution) or 379

the cleaved (active) form of caspase-3 (Cell Signaling, 9664; 1:300 dilution). Images 380

were captured at a magnification of x20 and a resolution of 0.5 m/pixel using a high-381

throughput Leica SCN400 slide scanner automated digital imaging system. 382

383



https://doi.org/10.1101/2020.08.05.238220


20

RNA sequencing. Two-to-three-day-old WT or Nsp65-/- mice were inoculated 384

intracranially with T3SA+ and euthanized at 2 and 6 d post-inoculation. Cortices were 385

microdissected, homogenized, and processed for plaque assay and RNA purification. 386

Tails were harvested for genotyping. Total RNA was extracted from brain homogenates 387

using TRIzol, digested with DNase, and further purified using the High Pure RNA 388

Isolation kit (Roche). RNA quantity was assessed using the Quant-iT RiboGreen RNA 389

Assay kit (Invitrogen), and RNA quality was assessed using an Agilent bioanalyzer. A 390

total of 1 mg of RNA was used for library preparation. Transcript abundance in samples 391

from infected WT and Nsp65-/- mice matched for viral titer as well as those from PBS-392

inoculated controls were determined by mRNA sequencing using a HiSeq 2000 393

instrument (Illumina). Three samples of RNA for each genotype per time point were 394

subjected to paired-end RNA sequencing. Statistical analysis identified genes that were 395

significantly increased or decreased in a reovirus-, p65-, and a reovirus- and p65-396

dependent manner. 397

398

Data analysis. Quality-controlled FASTQ files were aligned to the Ensemble 399

Mus musculus genome (GRCm38) using STAR aligner (version 2.5.1). HTSeq-count 400

was used to quantify counts of reads uniquely mapped to annotated genes using the 401

GRCm38 annotation gtf file (38). Differential gene expression by age, brain regions, p65 402

expression, and infection was analyzed using DESeq2 (39) using a model based on the 403

negative binomial distribution. The resulting P values were adjusted using the Benjamini 404

and Hochberg approach for controlling the false discovery rate, and differentially 405



https://doi.org/10.1101/2020.08.05.238220


21

expressed genes were identified at the 5% threshold. Gene-set enrichment analysis 406

was used to assess the statistical enrichment of gene ontologies and pathways (40). 407

408

Luminex assays. Brain homogenates prepared from mice 6 d post-inoculation 409

were prepared as described for viral titer. Lysates were analyzed by Luminex profiling 410

(Luminex Corporation) using the Bio-Plex Pro Mouse Chemokine 31-plex panel (Bio-411

Rad; 12009159) at a 1:2 dilution following the manufacturer’s instructions. Cytokine 412

abundance was determined using the laboratory multianalyte profiling system (LabMAP, 413

Luminex). 414

415

Infection of primary cortical neurons. Neurons were adsorbed with T3SA+ 416

ISVPs at an MOI of 1000 PFU/cell at 37°C for 1 h, washed with PBS, and incubated at 417

37°C for various intervals. Cells were washed with PBS and lysed in 1 mL TRIzol for 418

RNA isolation. 419

420

qPCR and complementary DNA synthesis. Total RNA was extracted from 421

infected cortical neurons at various intervals using TRIZOL and an RNeasy Mini kit 422

(Qiagen) according to manufacturer’s instructions, followed by DNase digestion. RNA 423

was reverse-transcribed with the SuperScript IV First-Strand Synthesis system 424

(ThermoFisher) using 0.5 to 1 g RNA. RT-qPCR was conducted using SsoAdvanced 425

Universal Probes Supermix (BioRad) with an Applied Biosystems ViiA7. Gene 426

expression was determined by the Cq method in which cDNA abundance was 427

normalized to mouse HPRT. Primers from Thermo Fisher include: CCL2 428



https://doi.org/10.1101/2020.08.05.238220


22

(Mm00441242_m1), CCL4 (Mm00443111_m1), CCL5 (Mm01302427_m1), CCL7 429

(Mm00443113_m1), CCL12 (Mm01617100_m1), CXCL9 (Mm00434946_m1), CXCL11 430

(Mm00444662_m1), TNF (Mm00443258_m1), and HPRT (Mm00446968_m1). 431

432

Statistical analysis. Experiments were performed in triplicate and repeated at 433

least twice. Representative results of single experiments are shown. Mean values were 434

compared using Student’s t test, Mann-Whitney test, Fisher’s Combined Probability test, 435

2-way ANOVA with Sidak’s multiple comparisons, or Log-rank test. Error bars denote 436

the range of data or standard deviation (S. D.). P values of < 0.05 were considered to 437

be statistically significant. 438



https://doi.org/10.1101/2020.08.05.238220


23

Acknowledgments 439

We thank members of the Dermody laboratory for insightful discussions 440

regarding this work. We thank Dr. Albert Baldwin at the University of North Carolina 441

Chapel Hill for sharing with us the strain of mice containing floxed p65/RelA developed 442

in his laboratory. We acknowledge the staff of the Vanderbilt Translational Pathology 443

Shared Resource and the Research Pathology/Health Sciences Tissue Bank at UPMC 444

Hillman Cancer Center. 445

This work was supported by the U.S. Public Health Service awards T32 AI60525 446

(P.B.), T32 HL007751 and F31 DK108562 (J.J.B.), and R01 AI038296 (T.S.D.) and the 447

Elizabeth B. Lamb Center for Pediatric Research. The funders had no role in study 448

design, data collection and analysis, decision to publish, or preparation of the 449

manuscript. 450



https://doi.org/10.1101/2020.08.05.238220


24

REFERENCES 451

1. Hiscott J, Kwon H, Genin P. 2001. Hostile takeovers: viral appropriation of the 452

NF-kappaB pathway. J Clin Invest 107:143-51. 453

2. Dresselhaus EC, Meffert MK. 2019. Cellular Specificity of NF-κB Function in the 454

Nervous System. Front Immunol 10:1043. 455

3. Hansberger MW, Campbell JA, Danthi P, Arrate P, Pennington KN, Marcu KB, 456

Ballard DW, Dermody TS. 2007. IkB kinase subunits a and g are required for 457

activation of NF-kB and induction of apoptosis by mammalian reovirus. Journal of 458

Virology 81:1360-1371. 459

4. Connolly JL, Rodgers SE, Clarke P, Ballard DW, Kerr LD, Tyler KL, Dermody TS. 460

2000. Reovirus-induced apoptosis requires activation of transcription factor NF-461

kB. Journal of Virology 74:2981-2989. 462

5. O'Donnell SM, Holm GH, Pierce JM, Tian B, Watson MJ, Chari RS, Ballard DW, 463

Brasier AR, Dermody TS. 2006. Identification of an NF-kB-dependent gene 464

network in cells infected by mammalian reovirus. Journal of Virology 80:1077-465

1086. 466

6. O'Donnell SM, Hansberger MW, Connolly JL, Chappell JD, Watson MJ, Pierce 467

JM, Wetzel JD, Han W, Barton ES, Forrest JC, Valyi-Nagy T, Yull FE, Blackwell 468

TS, Rottman JN, Sherry B, Dermody TS. 2005. Organ-specific roles for 469

transcription factor NF-kappaB in reovirus-induced apoptosis and disease. J Clin 470

Invest 115:2341-50. 471



https://doi.org/10.1101/2020.08.05.238220


25

7. Steinbrecher KA, Harmel-Laws E, Sitcheran R, Baldwin AS. 2008. Loss of 472

epithelial RelA results in deregulated intestinal proliferative/apoptotic 473

homeostasis and susceptibility to inflammation. J Immunol 180:2588-2599. 474

8. Pruijssers AJ, Hengel H, Abel TW, Dermody TS. 2013. Apoptosis induction 475

influences reovirus replication and virulence in newborn mice. J Virol 87:12980-9. 476

9. Tung WH, Lee IT, Hsieh HL, Yang CM. 2010. EV71 induces COX-2 expression 477

via c-Src/PDGFR/PI3K/Akt/p42/p44 MAPK/AP-1 and NF-kappaB in rat brain 478

astrocytes. J Cell Physiol 224:376-86. 479

10. Hsiao HB, Chou AH, Lin SI, Chen IH, Lien SP, Liu CC, Chong P, Liu SJ. 2014. 480

Toll-like receptor 9-mediated protection of enterovirus 71 infection in mice is due 481

to the release of danger-associated molecular patterns. Journal of Virology 482

88:11658-70. 483

11. Gregory D, Hargett D, Holmes D, Money E, Bachenheimer SL. 2004. Efficient 484

replication by herpes simplex virus type 1 involves activation of the IkappaB 485

kinase-IkappaB-p65 pathway. Journal of Virology 78:13582-90. 486

12. Liu R, Tan J, Lin Y, Jia R, Yang W, Liang C, Geng Y, Qiao W. 2013. HIV-1 Vpr 487

activates both canonical and noncanonical NF-κB pathway by enhancing the 488

phosphorylation of IKKα/β. Virology 439:47-56. 489

13. O'Donnell SM, Hansberger MW, Connolly JL, Chappell JD, Watson MJ, Pierce 490

JM, Wetzel JD, Han W, Barton ES, Forrest JC, Valyi-Nagy T, Yull FE, Blackwell 491

TS, Rottman JN, Sherry B, Dermody TS. 2005. Organ-specific roles for 492



https://doi.org/10.1101/2020.08.05.238220


26

transcription factor NF-kB in reovirus-induced apoptosis and disease. Journal of 493

Clinical Investigation 115:2341-2350. 494

14. Lin KI, Lee SH, Narayanan R, Baraban JM, Hardwick JM, Ratan RR. 1995. Thiol 495

agents and Bcl-2 identify an alphavirus-induced apoptotic pathway that requires 496

activation of the transcription factor NF-kappa B. Journal of Cell Biology 497

131:1149-61. 498

15. Amaya M, Voss K, Sampey G, Senina S, de la Fuente C, Mueller C, Calvert V, 499

Kehn-Hall K, Carpenter C, Kashanchi F, Bailey C, Mogelsvang S, Petricoin E, 500

Narayanan A. 2014. The role of IKKβ in Venezuelan equine encephalitis virus 501

infection. PLoS ONE 9:e86745. 502

16. Boulares AH, Ferran MC, Lucas-Lenard J. 1996. NF-kappaB activation Is 503

delayed in mouse L929 cells infected with interferon suppressing, but not 504

inducing, vesicular stomatitis virus strains. Virology 218:71-80. 505

17. Kesson AM, King NJ. 2001. Transcriptional regulation of major histocompatibility 506

complex class I by flavivirus West Nile is dependent on NF-kappaB activation. J 507

Infect Dis 184:947-54. 508

18. Hiscott J, Kwon H, Génin P. 2001. Hostile takeovers: viral appropriation of the 509

NF-kappaB pathway. J Clin Invest 107:143-51. 510

19. Bordignon J, Probst CM, Mosimann AL, Pavoni DP, Stella V, Buck GA, 511

Satproedprai N, Fawcett P, Zanata SM, de Noronha L, Krieger MA, Duarte Dos 512

Santos CN. 2008. Expression profile of interferon stimulated genes in central 513

nervous system of mice infected with dengue virus Type-1. Virology 377:319-29. 514



https://doi.org/10.1101/2020.08.05.238220


27

20. Clarke P, Leser JS, Bowen RA, Tyler KL. 2014. Virus-induced transcriptional 515

changes in the brain include the differential expression of genes associated with 516

interferon, apoptosis, interleukin 17 receptor A, and glutamate signaling as well 517

as flavivirus-specific upregulation of tRNA synthetases. mBio 5:e00902-14. 518

21. Tyler KL, Leser JS, Phang TL, Clarke P. 2010. Gene expression in the brain 519

during reovirus encephalitis. J Neurovirol 16:56-71. 520

22. Sharma A, Bhattacharya B, Puri RK, Maheshwari RK. 2008. Venezuelan equine 521

encephalitis virus infection causes modulation of inflammatory and immune 522

response genes in mouse brain. BMC Genomics 9:289. 523

23. Chandwani MN, Creisher PS, O'Donnell LA. 2019. Understanding the Role of 524

Antiviral Cytokines and Chemokines on Neural Stem/Progenitor Cell Activity and 525

Survival. Viral Immunol 32:15-24. 526

24. Klein DDaR. 2011. Pathogenesis of Encephalitis: Chemokines and viral 527

infections of the CNS. 528

25. Hosking MP, Lane TE. 2010. The role of chemokines during viral infection of the 529

CNS. PLoS Pathog 6:e1000937. 530

26. van Marle G, Henry S, Todoruk T, Sullivan A, Silva C, Rourke SB, Holden J, 531

McArthur JC, Gill MJ, Power C. 2004. Human immunodeficiency virus type 1 Nef 532

protein mediates neural cell death: a neurotoxic role for IP-10. Virology 329:302-533

18. 534



https://doi.org/10.1101/2020.08.05.238220


28

27. Aravalli RN, Hu S, Rowen TN, Palmquist JM, Lokensgard JR. 2005. Cutting 535

edge: TLR2-mediated proinflammatory cytokine and chemokine production by 536

microglial cells in response to herpes simplex virus. J Immunol 175:4189-93. 537

28. Sha WC, Liou HC, Tuomanen EI, Baltimore D. 1995. Targeted disruption of the 538

p50 subunit of NF-kappa B leads to multifocal defects in immune responses. Cell 539

80:321-30. 540

29. Steinbrecher KA, Harmel-Laws E, Sitcheran R, Baldwin AS. 2008. Loss of 541

epithelial RelA results in deregulated intestinal proliferative/apoptotic 542

homeostasis and susceptibility to inflammation. J Immunol 180:2588-99. 543

30. Luo L, Ambrozkiewicz MC, Benseler F, Chen C, Dumontier E, Falkner S, Furlanis 544

E, Gomez AM, Hoshina N, Huang WH, Hutchison MA, Itoh-Maruoka Y, Lavery 545

LA, Li W, Maruo T, Motohashi J, Pai EL, Pelkey KA, Pereira A, Philips T, Sinclair 546

JL, Stogsdill JA, Traunmüller L, Wang J, Wortel J, You W, Abumaria N, Beier KT, 547

Brose N, Burgess HA, Cepko CL, Cloutier JF, Eroglu C, Goebbels S, Kaeser PS, 548

Kay JN, Lu W, Luo L, Mandai K, McBain CJ, Nave KA, Prado MAM, Prado VF, 549

Rothstein J, Rubenstein JLR, Saher G, Sakimura K, Sanes JR, Scheiffele P, 550

Takai Y, et al. 2020. Optimizing Nervous System-Specific Gene Targeting with 551

Cre Driver Lines: Prevalence of Germline Recombination and Influencing 552

Factors. Neuron 106:37-65.e5. 553

31. Antar AAR, Konopka JL, Campbell JA, Henry RA, Perdigoto AL, Carter BD, 554

Pozzi A, Abel TW, Dermody TS. 2009. Junctional adhesion molecule-A is 555

required for hematogenous dissemination of reovirus. Cell Host Microbe 5:59-71. 556



https://doi.org/10.1101/2020.08.05.238220


29

32. Furlong DB, Nibert ML, Fields BN. 1988. Sigma 1 protein of mammalian 557

reoviruses extends from the surfaces of viral particles. Journal of Virology 558

62:246-56. 559

33. Smith RE, Zweerink HJ, Joklik WK. 1969. Polypeptide components of virions, top 560

component and cores of reovirus type 3. Virology 39:791-810. 561

34. Virgin HW, IV, Bassel-Duby R, Fields BN, Tyler KL. 1988. Antibody protects 562

against lethal infection with the neurally spreading reovirus type 3 (Dearing). 563

Journal of Virology 62:4594-4604. 564

35. Baer GS, Dermody TS. 1997. Mutations in reovirus outer-capsid protein s3 565

selected during persistent infections of L cells confer resistance to protease 566

inhibitor E64. Journal of Virology 71:4921-4928. 567

36. Tyler KL, Squier MK, Rodgers SE, Schneider SE, Oberhaus SM, Grdina TA, 568

Cohen JJ, Dermody TS. 1995. Differences in the capacity of reovirus strains to 569

induce apoptosis are determined by the viral attachment protein s1. Journal of 570

Virology 69:6972-6979. 571

37. Sutherland DM, Aravamudhan P, Dietrich MH, Stehle T, Dermody TS. 2018. 572

Reovirus neurotropism and virulence are dictated by sequences in the head 573

domain of the viral attachment protein. J Virol 92:e00974-18. 574

38. Anders S, Pyl PT, Huber W. 2015. HTSeq--a Python framework to work with 575

high-throughput sequencing data. Bioinformatics 31:166-9. 576

39. Love MI, Huber W, Anders S. 2014. Moderated estimation of fold change and 577

dispersion for RNA-seq data with DESeq2. Genome Biol 15:550. 578



https://doi.org/10.1101/2020.08.05.238220


30

40. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, 579

Paulovich A, Pomeroy SL, Golub TR, Lander ES, Mesirov JP. 2005. Gene set 580

enrichment analysis: a knowledge-based approach for interpreting genome-wide 581

expression profiles. Proc Natl Acad Sci U S A 102:15545-50. 582

583



https://doi.org/10.1101/2020.08.05.238220


31

FIGURE LEGENDS 584

585

FIG 1 Neuron-specific p65-null mice lack p65 expression in the brain. (A) Breeding 586

scheme for establishment of neuron-specific p65-null mice (Nsp65-/-). Mice with loxP 587

sites (gray arrows) flanking exons 5 – 8 of the p65 allele (p65 f/f) were interbred with 588

mice expressing Cre recombinase under control of the rat nestin promoter and 589

enhancer to establish mice heterozygous for the nestin Cre transgene and homozygous 590

for the p65 floxed allele (Nsp65-/-). (B) Tissue homogenates from wild-type (WT) and 591

Nsp65-/- (KO) mice were resolved by SDS-PAGE and immunoblotted using antibodies 592

specific for p65 and GAPDH. 593

594

FIG 2 Reovirus-induced apoptosis is diminished in cultures of cortical neurons that lack 595

either NF-B p50 or p65. Wild-type (WT), p50-/- (A), and Nsp65-/- (B) cortical neurons 596

were adsorbed with T3SA+ reovirus at the MOIs shown. At 24 h post-adsorption, the 597

percentage of apoptotic nuclei were quantified following AO staining. Staurosporine (ST, 598

10 mM) was used as positive control. Each point is expressed as the mean percentage 599

of apoptotic cells for triplicate wells. Error bars indicate SD. *, P < 0.05; **, P < 0.01; ***, 600

P < 0.001 as determined by Student’s t test compared with results for Nsp65-/- at the 601

same MOI. 602

603

FIG 3 Neuron-specific NF-B p65 is required for reovirus neurovirulence but not virus 604

replication. (A) Survival following intracranial inoculation. Three-day-old wild-type (WT) 605

and Nsp65-/- mice (n = 14 WT and 16 Nsp65-/-) were inoculated intracranially with 10 606



https://doi.org/10.1101/2020.08.05.238220


32

PFU of reovirus T3SA+ and monitored for disease for 21 d. Mice were euthanized when 607

moribund. Values that differ from WT by log rank test are indicated (*, P < 0.05). (B) 608

Viral titers and (C) immunohistopathology in the brain following intracranial inoculation. 609

An independent cohort of identically inoculated mice (n = 5 to 6 mice for each strain) 610

were euthanized at 6, 8, and 10 d post-inoculation. (B) Viral titers of the homogenized 611

left brain hemisphere were determined by plaque assay. Data are log transformed and 612

displayed with a linear x axis scale. (C) Titer-matched right brain hemispheres at 6 d 613

post-inoculation were fixed in formalin and embedded in paraffin. Coronal sections were 614

either stained with H&E or immunostained with reovirus polyclonal antiserum or 615

antibody specific for cleaved caspase-3. Representative sections are shown. 616

617

FIG 4 Expression of genes influenced by reovirus infection and NF-B p65 in the brain. 618

Wild-type (WT) and Nsp65-/- newborn mice were inoculated intracranially with 10 PFU of 619

reovirus T3SA+ or PBS. Cortices were removed at 6 d post-inoculation and processed 620

for RNA purification and mRNA sequencing. RNA samples prepared from infected mice 621

were matched for viral titer. (A) Venn diagram for differential expression analysis using 622

DeSeq2 package in R between WT infected vs. mock (red) and Nsp65-/- infected vs. 623

mock (blue). (B) Networks with the most differentially expressed genes were identified 624

using Ingenuity Pathway Analysis. (C) Heat map depicting expression levels. 625

626

FIG 5 Reovirus-induced cytokine expression in the brain is dependent on NF-B p65 627

expression in neurons. Wild-type (WT) and Nsp65-/- (KO) newborn mice were inoculated 628

intracranially with 10 PFU of reovirus T3SA+ or PBS. Cortices were removed at 6 d 629



https://doi.org/10.1101/2020.08.05.238220


33

post-inoculation and processed for Luminex analysis and plaque assay. Samples were 630

matched for viral titer. Data are presented as the mean values of 8 to 10 mice. *, P < 631

0.05; **, P < 0.005, as determined by unpaired Student’s t test compared with results for 632

Nsp65-/- mice. 633

634

FIG 6 Reovirus-induced upregulation of cytokine transcripts in cultured cortical neurons 635

is dependent on NF-B p65 expression. Wild-type (WT) and Nsp65-/- cortical neurons 636

were adsorbed with 1000 PFU of reovirus T3SA+ ISVPs or PBS. Cells were harvested 637

at times indicated, RNA was isolated, and transcript levels were determined by RT-638

qPCR. Data are presented as the mean values of two replicates. *, P < 0.05; **, P < 639

0.005; ****, P < 0.0001, as determined by two-way ANOVA with Sidak’s multiple 640

comparisons. 641

642



https://doi.org/10.1101/2020.08.05.238220


34

Xp65

allele

LoxP

site

LoxP

site

p65 f/f

transgenic mouse

Nestin Cre

transgenic mouse

Nsp65-/-

transgenic mouse

A

WTKOB WTKO WTKO WTKO

brain heart spleen thymus

4 5 6 7 8 9

Nestin

promoter

Cre

recombinase

FIG 1 Neuron-specific p65-null mice lack p65 expression in the brain. (A) Breeding scheme for establishment of neuron-specific p65-null mice (Nsp65-/-). Mice with loxP sites (gray arrows) flanking exons 5 – 8 of the p65 allele (p65 f/f) were interbred with mice expressing Cre recombinase under control of the rat nestin promoter and enhancer to establish mice heterozygous for the nestin Cre transgene and homozygous for the p65 floxed allele (Nsp65-/-). (B) Tissue homogenates from wild-type (WT) and Nsp65-/- (KO) mice were resolved by SDS-PAGE and immunoblotted using antibodies specific for p65 and GAPDH.



https://doi.org/10.1101/2020.08.05.238220


35

A B

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5 WT

**

Mock ST

T3SA+ (PFU/cell)

***

*

*

10 100 1000

p50-/-

Ap

op

totic n

ucle

i (r

ela

tive

to

mo

ck)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0 WT

**

Mock ST

T3SA+ (PFU/cell)

***

****

10 100 1000

Nsp65-/-

Ap

op

totic n

ucle

i (r

ela

tive

to

mo

ck)

FIG 2 Reovirus-induced apoptosis is diminished in cultures of cortical neurons that lack

either NF-B p50 or p65. Wild-type (WT), p50-/- (A), and Nsp65-/- (B) cortical neurons were adsorbed with T3SA+ reovirus at the MOIs shown. At 24 h post-adsorption, the percentage of apoptotic nuclei were quantified following AO staining. Staurosporine (ST, 10 mM) was used as positive control. Each point is expressed as the mean percentage of apoptotic cells for triplicate wells. Error bars indicate SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001 as determined by Student’s t test compared with results for Nsp65-/- at the same MOI.



https://doi.org/10.1101/2020.08.05.238220


36

A

H&E Reovirus Caspase-3

WT

Nsp65-/-

C

B

0

20

40

60

80

100

Nsp65 -/-

WT

Time post-infection (d)

Perc

en

t su

rviv

al

1 3 5 7 9 11 13 15 17 19 210

2

4

6

8

10

Time post-infection (d)

6 8 10

Nsp65-/-WT

Vir

al ti

ter

(Lo

g10 P

FU

/bra

in)

FIG 3 Neuron-specific NF-B p65 is required for reovirus neurovirulence but not virus replication. (A) Survival following intracranial inoculation. Three-day-old wild-type (WT) and Nsp65-/- mice (n = 14 WT and 16 Nsp65-/-) were inoculated intracranially with 10 PFU of reovirus T3SA+ and monitored for disease for 21 d. Mice were euthanized when moribund. Values that differ from WT by log rank test are indicated (*, P < 0.05). (B) Viral titers and (C) immunohistopathology in the brain following intracranial inoculation. An independent cohort of identically inoculated mice (n = 5 to 6 mice for each strain) were euthanized at 6, 8, and 10 d post-inoculation. (B) Viral titers of the homogenized left brain hemisphere were determined by plaque assay. Data are log transformed and displayed with a linear x axis scale. (C) Titer-matched right brain hemispheres at 6 d post-inoculation were fixed in formalin and embedded in paraffin. Coronal sections were either stained with H&E or immunostained with reovirus polyclonal antiserum or antibody specific for cleaved caspase-3. Representative sections are shown.



https://doi.org/10.1101/2020.08.05.238220


37

0

7

Fold

-change

57814 83

WT Nsp65-/-

C

A B

Ontology:

Regulation:

Gene ID: B2M

C2

C4B

H2-T

23

LY

6A

PS

MB

8

PS

MB

9

TA

P2

CC

L2

CC

L4

CC

L5

CC

L7

CC

L12

CX

CL9

CX

CL11

IL1

b

TN

Fa

OA

S1G

BC

L2A

1a

NA

IP

TN

FS

F1

0

InflammationImmunity Cell Fate

Focus

molecules

Infectious disease, organismal injury and abnormalities

organismal injury and abnormalities

DNA replication, recombination and repair,

infectious disease

Cell death and survival, cell compromise,

infectious disease

46 31

41 29

39 28

Networks Score

FIG 4 Expression of genes influenced by reovirus infection and NF-B p65 in the brain. Wild-type (WT) and Nsp65-/- newborn mice were inoculated intracranially with 10 PFU of reovirus T3SA+ or PBS. Cortices were removed at 6 d post-inoculation and processed for RNA purification and mRNA sequencing. RNA samples prepared from infected mice were matched for viral titer. (A) Venn diagram for differential expression analysis using DeSeq2 package in R between WT infected vs. mock (red) and Nsp65-/- infected vs. mock (blue). (B) Networks with the most differentially expressed genes were identified using Ingenuity Pathway Analysis. (C) Heat map depicting expression levels.



https://doi.org/10.1101/2020.08.05.238220


38

0

10000

20000

30000C

CL

2 (

pg

/bra

in)

WT KO WT KO

PBS SA+

0

500

1000

1500

2000

CC

L4

(p

g/b

rain

)

WT KO WT KO

PBS SA+

*

0

500

1000

1500

2000

CC

L5

(p

g/b

rain

)

WT KO WT KO

PBS SA+

*

0

2000

4000

6000

CC

L7

(p

g/b

rain

)

WT KO WT KO

PBS SA+

0

2000

4000

6000

8000C

CL

12

(p

g/b

rain

)

WT KO WT KO

PBS SA+

0

2000

4000

6000

8000

CX

CL

11 (

pg

/bra

in)

WT KO WT KO

PBS SA+

**

FIG 5 Reovirus-induced cytokine expression in the brain is dependent on NF-B p65 expression in neurons. Wild-type (WT) and Nsp65-/- (KO) newborn mice were inoculated intracranially with 10 PFU of reovirus T3SA+ or PBS. Cortices were removed at 6 d post-inoculation and processed for Luminex analysis and plaque assay. Samples were matched for viral titer. Data are presented as the mean values of 8 to 10 mice. *, P < 0.05; **, P < 0.005, as determined by unpaired Student’s t test compared with results for Nsp65-/- mice.



https://doi.org/10.1101/2020.08.05.238220


39

0 3 6 16 24

0

100000

200000

300000

400000

S4

Time post adsorption (h)

mR

NA

(F

old

ch

an

ge)

WT

Nsp65-/-

****

****

0 3 6 16 24

0

5

10

15

20

25

CCL2


mR

NA

(F

old

ch

an

ge)

WT

Nsp65-/-

**

0 3 6 16 24

0

5

10

15

CCL4


mR

NA

(F

old

ch

an

ge)

WT

Nsp65-/-****

0 3 6 16 24

0

10

20

30

40

50

CCL5


mR

NA

(F

old

ch

an

ge)

WT

Nsp65-/-

*

0 3 6 16 24

0

2

4

6

CCL7


mR

NA

(F

old

ch

an

ge)

WT

Nsp65-/-

*

0 3 6 16 24

0.0

0.5

1.0

1.5

2.0

CCL12


mR

NA

(F

old

ch

an

ge)

WT

Nsp65-/-******

**

0 3 6 16 24

0

5

10

15

CXCL9


mR

NA

(F

old

ch

an

ge)

WT

Nsp65-/-****

****

*

0 3 6 16 24

0

10

20

30

CXCL11


mR

NA

(F

old

ch

an

ge)

WT

Nsp65-/-

FIG 6 Reovirus-induced upregulation of cytokine transcripts in cultured cortical

neurons is dependent on NF-B p65 expression. Wild-type (WT) and Nsp65-/- cortical neurons were adsorbed with 1000 PFU of reovirus T3SA+ ISVPs or PBS. Cells were harvested at times indicated, RNA was isolated, and transcript levels were determined by RT-qPCR. Data are presented as the mean values of two replicates. *, P < 0.05; **, P < 0.005; ****, P < 0.0001, as determined by two-way ANOVA with Sidak’s multiple comparisons.



https://doi.org/10.1101/2020.08.05.238220


Neuron-intrinsic NF-κB Signaling Mediates Reovirus Virulence...2020/08/05 · 25 692-8071. E-mail...

Documents

Transcript of Neuron-intrinsic NF-κB Signaling Mediates Reovirus Virulence...2020/08/05 · 25 692-8071. E-mail...