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.CC-BY-NC-ND 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.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
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
.CC-BY-NC-ND 4.0 International licenseavailable under a(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.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
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
.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
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
.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
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
.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
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
.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
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
.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
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
.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
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
.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
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
.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
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
.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
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
.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
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
.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
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
.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
15
therapeutic interruption of the chemokine networks activated by reovirus may have 293
broad utility against neurotropic viruses. 294
.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
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
.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
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
.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
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
.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
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
.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
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
.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
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
.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
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
.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
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
.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
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
.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
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
.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
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
.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
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
.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
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
.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
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
.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
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
.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
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
.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
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
.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
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
.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
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.
.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
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.
.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
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.
.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
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.
.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
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.
.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
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
Time post adsorption (h)
mR
NA
(F
old
ch
an
ge)
WT
Nsp65-/-
**
0 3 6 16 24
0
5
10
15
CCL4
Time post adsorption (h)
mR
NA
(F
old
ch
an
ge)
WT
Nsp65-/-****
0 3 6 16 24
0
10
20
30
40
50
CCL5
Time post adsorption (h)
mR
NA
(F
old
ch
an
ge)
WT
Nsp65-/-
*
0 3 6 16 24
0
2
4
6
CCL7
Time post adsorption (h)
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
Time post adsorption (h)
mR
NA
(F
old
ch
an
ge)
WT
Nsp65-/-******
**
0 3 6 16 24
0
5
10
15
CXCL9
Time post adsorption (h)
mR
NA
(F
old
ch
an
ge)
WT
Nsp65-/-****
****
*
0 3 6 16 24
0
10
20
30
CXCL11
Time post adsorption (h)
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.
.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
Top Related