JVI Accepts, published online ahead of print on 20 August 2014 J. Virol. doi:10.1128/JVI.00692-14 Copyright © 2014, American Society for Microbiology. All Rights Reserved.
1
Immunodeficient Mouse Models with Different Disease Profiles by in vivo Infection with the Same
2
Clinical Isolate of Enterovirus 71
3 4
Running Title: NOD/SCID, stat-1 KO, and ifngr KO mice for EV71
5 6
Authors: Chun-Che Liao1,2,#, An-Ting Liou2,3,# , Ya-Shu Chang2, Szu-Yao Wu2,4 ,
7
Chih-Shin Chang2,4, Chien-Kuo Lee5, John T. Kung6, Pang-Hsien Tu2, Ya-Yen Yu7, Chi-Yung
8
Lin7, Jen-Shiou Lin7,and Chiaho Shih1, 2,3 *
9 10
1
Graduate Institute of Life Sciences, National Defense Medical Center, Taipei, Taiwan
11
2
Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan
12
3
Taiwan International Graduate Program (TIGP) in Molecular Medicine, National Yang-Ming University and Academia Sinica, Taipei, Taiwan
13 14
4
15
5
16
6
17
7
18
Institute of Microbiology and Immunology, National Yang-Ming University, Taipei, Taiwan Institute of Immunology, National Taiwan University, Taipei 100, Taiwan
Institute of Molecular Biology, Academia Sinica, Taipei, Taiwan Section of Clinical Virology and Molecular Diagnosis, Department of Laboratory Medicine, Changhua Christian Hospital, Changhua, Taiwan
19 20
*To whom correspondence should be addressed
21
#The first two authors made equal contribution to this paper.
22
Mailing address: Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan
23
Tel: 886-2-2652-3996; Fax: 8862-2652-3597; E-mail:
[email protected] 1
24
ABSTRACT
Like poliovirus, severe infection with enterovirus 71 (EV71) can cause neuropathology.
25
Unlike poliovirus, EV71 is often associated with hand, foot, and mouth disease (HFMD). Here, we
26
established three mouse models for experimental infection with the same clinical isolate of EV71. The
27
NOD/SCID mouse model is unique in developing skin rash, an HFMD-like symptom. While the
28
NOD/SCID model developed limb paralysis and death at near 100% efficiency, the interferon-gamma
29
receptor knockout (ifngr KO) and the stat-1 knockout mice exhibited paralysis and death rates near 78%
30
and 30%, respectively. Productive infection with EV71 depends on viral dose, host age, and inoculation
31
routes. Infectious EV71, and VP1-specific RNA and protein in muscle, brain, and spinal cord, were
32
compared side-by-side between NOD/SCID and stat-1 models before, during, and after disease onset.
33
Spleen fibrosis and muscle degeneration are common in NOD/SCID and stat-1models. Main differences
34
between these two models include their disease manifestations and cytokine/chemokine profiles. Pathology
35
of the NOD/SCID model includes 1) inflammation and expression of viral VP1 antigen in muscle; 2)
36
increased neutrophils and decreased eosinophils and lymphocytes; 3) hair loss and skin rash. Characteristic
37
pathology of the stat-1 model includes 1) a strong tropism of EV71 for central nervous system; 2) detection
38
of VP1 protein in the Purkinje layer of cerebellar cortex, pons, brainstem, and spinal cord; 3) amplification
39
of microglial cells; 4) dystrophy of intestinal villi. Our comparative studies on these new models by oral or
40
intraperitoneal (i.p.) infection underscored the contribution of host immunity, including interferon-gamma
41
receptor, to EV71 pathogenesis.
42 43 44 45 46 2
47 48 49
IMPORTANCE
50
Asia-Pacific regions. Disease manifestations include subclinical infection, common cold-like syndromes,
51
hand-foot-and-mouth disease (HFMD), uncomplicated brainstem encephalitis, severe dysregulated
52
autonomic nerve system, fatal pulmonary edema, and cardiopulmonary collapse. To date, no effective
53
vaccine or treatment is available. A user-friendly and widely accessible animal model for researching EV71
54
infection and pathogenesis is urgently needed by the global community, both academia and industry.
In the past decade, enterovirus 71 (EV71) has emerged as a major threat to the public health in the
3
55
INTRODUCTION
In the past decade, enterovirus 71 (EV71) has emerged as a major threat to the
56
public health in the Asia-Pacific regions (1, 2). As a member of the Picornaviridae, EV71 is related to
57
poliovirus (3), hepatitis A virus (EV72) (4), and coxsackieviruses (5). It is likely that both viral and host
58
factors could contribute to the diverse pathogenesis profiles, ranging from subclinical infection, common
59
cold-like syndromes, hand-foot-and-mouth disease (HFMD), to uncomplicated brainstem encephalitis,
60
severe dysregulated autonomic nerve system, fatal pulmonary edema, and cardiopulmonary collapse. To
61
date, no effective vaccine or treatment is available (6, 7). A user-friendly and widely accessible animal
62
model for EV71 infection will be most convenient to the research community.
63 64
By serial passages through mouse brain, Wang et al. (8) reported experimental infection of newborn ICR
65
mice with an EV71 mouse-adapted strain MP4. In general, newborn mice (1-2 day old) are difficult to
66
handle technically. A further adapted strain M2 can infect 12-14 day old ICR mice (9). However, it remains
67
unclear if adaptive mutations associated with the various mouse-adapted strains can be found in human
68
natural infection. For example, a G145E adaptive mutation at the VP1 of a mouse-adapted EV71 strain is
69
essential for the hind limb paralysis in 3- to 4-week-old NOD/SCID mice (10). Cynomolgus monkeys could
70
also serve as an animal model for studying EV71 pathogenesis (11). However, monkey is not a convenient
71
model, and not easily accessible to most laboratories. AG129 mice are known to lack type I and II
72
interferon receptors (IFNR), and therefore are severely defective in innate immunity. Khong et al. (12)
73
reported that AG129 mice were susceptible to non-mouse-adapted EV71 infection via i.p. and oral routes.
74
Infected mice developed limb paralysis and death. An adapted strain of EV71 which can infect adult AG129
75
mice has been reported (13). Yamayoshi et al. (14) reported that human Scavenger receptor B2 (SCARB2)
76
can function as a receptor for EV71 entry in vitro. Similarly, Nishimura et al. (15) reported that human
77
P-selectin glycoprotein ligand-1(PSGL-1) can also facilitate EV71 infection in vitro. Recently, it was 4
78
reported that human SCARB2 transgenic (Tg) mice can serve as an infectious animal model of EV71 (16,
79
17). While these receptor Tg mice can result in limb paralysis upon i.p. injection with EV71, no typical
80
symptoms of HFMD with skin rash were observed. In fact, acute flaccid paralysis clinically is a rare event
81
(0.31%) in human EV71 natural infection (1, 18). Furthermore, limb paralysis is not a unique phenotype of
82
EV71, many other viruses can also cause limb paralysis in animal models (19, 20). Another important issue
83
in the current SCARB2 receptor Tg model is its lack of oral infection (17). While oral infection was
84
observed in another SCARB2 Tg model, the efficiency appeared to be very low using a so-called Isehara
85
strain of EV71 (success rate only 2/50 in three experiments) (16).Therefore, it is urgent to develop an
86
animal model that can closely mimick human natural infection, such as producing the HFMD-like
87
phenotype via a more efficient oral infection.
88 89
In literature, the same animal model infected with different genotypes of EV71 was conducted (16, 17). In
90
contrast, using the same viral strain for infection of different animal models remains to be explored.
91
Considering the outbred human population, the effect of host genetic factors on viral pathogenesis needs to
92
be investigated.
93 94
In this study, by using the same clinical isolate of EV71 in three newly established animal models
95
susceptible to EV71 infection, we demonstrated that these models with different genetic backgrounds
96
exhibited different disease profiles. Interestingly, in the non-obese diabetic mice with severe combined
97
immunodeficiency disease (NOD/SCID), the infection and disease manifestations appeared to be more
98
systemic in multiple tissues than neurotropic. In particular, spleen fibrosis, muscle inflammation and
99
destruction, as well as abnormal hematological features, were evident in the NOD/SCID model. Notably, no
100
productive infection and disease were observed when UV-treated EV71 was inoculated. A robust oral 5
101
infection system was established using the NOD/SCID mouse model. In contrast, in the stat-1 KO model,
102
EV71 infection was highly neurotropic, with striking VP1 expression and massive microglial cell
103
amplification in CNS. Both Pons and cerebellar Purkinje cells were strongly targeted by EV71, suggesting a
104
retrograde axonal transport mechanism. While the strong neurotropism of EV71 in the stat-1 KO model
105
resembles what has been described in patients, our studies on the NOD/SCID model suggests that EV71
106
might also cause a systemic infection-like symptoms. Most likely, both muscle and CNS infections could
107
contribute to limb paralysis. We report here that skin rash in NOD/SCID mice can be observed by i.p. or
108
oral route of inoculation. Furthermore, by using an interferon-gamma receptor knockout mouse model, we
109
demonstrated that IFN-gamma receptor is likely to be critical for the protection from EV71 infection and
110
pathogenesis. Our new models provide a novel tool for both basic science research and development on
111
antivirals and vaccine candidates.
112
6
113
MATERIALS AND METHODS
114
Ethics Statement. All animal experiments were conducted under protocols approved by Academia Sinica
115
Institutional Animal Care & Utilization Committee (ASIACUC Protocol number 2010003). Research was
116
conducted in compliance with the principles stated in the Guide for the Care and Use of Laboratory Animals,
117
National Research Council, 1996.
118
Cells and virus preparation. Human rhabdomyosarcoma (RD) cells (ATCC CCL-136) were grown in
119
DMEM (Gibco) with 10% of FBS (Hyclone) and 1% of penicillin/streptomycin (Gibco). EV71 clinical
120
isolates were kindly provided by Section of Clincial Virology and Molecular Diagnosis, Department of
121
Laboratory Medicine, Changhua Christian Hospital. For virus preparation, RD cells were infected with
122
EV71 clinical isolate F23 with 0.01 M.O.I and maintained in DMEM with 0.2 % FBS. Virus was harvested
123
by three free-thaw cycles and clarified by centrifugation at 3000 ×g for 30 minutes at 4 °C. Virus in
124
clarified medium was concentrated by ultracentrifugation in Beckman SW28 rotors at 121896 ×g (26000
125
rpm) for 4 hours at 4 °C with 30% sucrose cushion and resuspended in PBS. The virus titer was determined
126
by plaque assay before animal infection. In brief, RD cell monolayers in 6-well plates (SPL Life Science)
127
were infected with 10-fold serial diluted virus in DMEM. After incubation at 37 °C with 5 % CO2 for 1 hour,
128
virus was removed and cells were covered 4ml of 0.3% agarose (Lonza) in DMEM with 0.2% FBS. After
129
incubation in 37 °C with 5 % CO2 for 3 days, the cells were fixed by 3.7% formalin (Merk) for 1 hour and
130
stained with 0.5% crystal violet. For UV-inactivation experiment, purified EV71 was irradiated by UV at
131
750 mJ/cm2 using a CL-1000 Ultraviolet Crosslinker (UVP). The residual viral titer after UV-treatment was
132
determined by plaque assay.
133
Experimental infection and antibody treatment.NOD/SCID mice were purchased from LASCO Co., Ltd
134
(Taiwan). The stat-1 KO mice and ifngr KO (129/Sv/Ev inbred) mice were provided by Dr. Chien-Kuo
135
Lee and Dr. John Kung, respectively. Mice were housed under specific pathogen-free condition in the 7
136
individual-ventilated cage. One-week-old NOD/SCID or stat-1 KO mice were infected via an
137
intraperitoneal (i.p.) or subcutaneous (s.c.) route using an EV71 clinical strain F23 at a dose of 105 to 108
138
pfu/mouse. For oral infection, 108 pfu of EV71 was injected into each NOD/SCID mouse (3-day-old) with a
139
24-guage feeding tube after 6 hours of fasting. Neutralizing antibodies against IFN-α receptor or
140
IFN-γ (Bio X Cell) were i.p. injected at a dose of 150 μg/mouse into one-week old wild type C57BL/6
141
mice (NAR Labs, Taiwan). EV71 infection (108 pfu/mouse) was via i.p. route. IFN-α receptor neutralizing
142
antibody was injected one day before EV71 inoculation, followed by repetitive antibody treatments on 1
143
and 3 dpi. For IFN-γ neutralizing antibody, it was coinjected with EV71, followed by repetitive antibody
144
treatments on 1 and 2 dpi. Disease manifestation was monitored daily.
145
Histopathological and immunohistochemical (IHC) stain. The euthanatized mice were perfused
146
transcardialiy with PBS followed by 10% neutral buffered formalin (CHIN I PAO CO.,LTD, Taiwan).
147
Tissues were fixed in 10% neutral buffered formalin overnight. Fixed tissues were paraffin embedded,
148
sliced, and stained with H&E or Masson's trichrome stain by Pathology Core Laboratory, IBMS,
149
AcademiaSinica. Xylene and ethanol were used for deparaffinization and rehydration. Retrieval buffer pH
150
6.0 (Dako) and endogenous enzyme blocker (Dako) were used for antigen retrieval and blocking steps. The
151
slides were washed with PBS containing 0.1% Tween 20 (PBST) and incubated with anti-VP1 (Abnova),
152
anti-F4/80 (Taiwan GeneTex), anti-Keratin 6, anti-Keratin 14 (Covance), anti-pan-collagen TypeⅠ(Millipore)
153
or anti-IBA1 (Taiwan GeneTex) antibodies. These slides were washed with PBST and incubated in
154
secondary antibody, anti-rabbit (Dako) or anti-rat (Dako) antibody. DAB system (Dako) was used to present
155
signal of each antigen. Sections were counterstained with haematoxylin (J.T. Baker), and mounted with
156
mounting reagent (MUTO Pure Chemicals).
157
Tissue sampling and virus titration. After euthanasia, blood samples were collected by heart puncture
158
prior to PBS perfusion. Organs and tissues were harvested, weighed and homogenized by a mechanical 8
159
homogenizer (Next Advance) in PBS. Virus in supernatants of clarified homogenate and blood serum were
160
routinely first amplified by passaging the tissue-derived virus through the RD cells for one round (24 hrs)
161
before reinfection with RD cells for plaque assay. Viral titer was determined by plaque assay and expressed
162
as PFU per gram (for tissue) or per ml (for sera).
163
Real time RT-PCR. Total RNA of tissue homogenates were extracted by WelPrep cell/tissue RNA kit
164
(Welgene), and used for reverse transcription by High Capacity cDNA Reverse Transcription kit (Applied
165
Biosystems). The synthetic cDNA was subjected to real time quantitative PCR analysis by ABI 7500
166
(Applied Biosystems) with Power SYBR Green PCR Master Kit (Applied Biosystems). The specific
167
primers for VP1: forward: CTAGAGGGTACCACCAATCC; reverse: AACCTGGCCAGTAGGAGT.
168
primer sequences of the internal control GAPDH: forward: GTTCCTACCCCCAATGTG; reverse:
169
CAACCTGGTCCTCAGTGTAG.
170
Cytokine quantification. Cytokines in tissue homogenates were measured by Procarta Immunoassay Kit
171
(Affymetrix) according to the manufacturer’s instructions. Detection sensitivity of various cytokines was as
172
described in manufacturer’s protocol.
173
Complete blood count (CBC). Blood samples from infected mice were collected with 1.5 mg/ml of
174
Na2EDTA. The complete blood count of mice were conducted by Abbott Cell-Dyn 3700.
175
Immunoblotting. RIPA buffer was used to extract protein from tissue homogenates. VP1 protein and
176
tubulin were detected by anti-VP1 antibody (Abnova) and anti-tubulin antibody (Taiwan GeneTex),
177
respectively.
178
Statistical analysis. All experimental results from two to six independent experiments were analyzed by
179
Student’s t-test. Two Way ANOVA was used to analyze the differences of HFMD-like syndromes and limb
180
paralysis between oral and non-oral routes of infection.
181 9
The
182
RESULTS
183
Host immunity plays an important role in resistance and clearance of viral infection. For example, AG129
184
mice deficient in interferon (IFN-α/β and -γ) signaling can be infected with EV71 in vivo (12). To address
185
the issue of immunity and host defense against EV71, we tested several immunodeficient mouse strains for
186
their susceptibility to infection with clinical isolates of EV71 (Fig. 1A). To our surprise, neither IFN
187
receptor-alpha knockout mice (ifnar KO) nor the complement C5-deficient FVB mice could support EV71
188
infection. Instead, we observed successful infection in stat-1 KO, NOD/SCID, and IFN receptor-gamma
189
knockout mice (ifngr KO). Because we succeeded in the ifngr KO model only very recently, our more
190
detailed characterizations focused primarily on the other two models. The stat-1 KO mice lacks the STAT1
191
gene and is defective in IFN signaling (21), while NOD/SCID mice are known to be defective in innate
192
immunity of NK cells as well as humoral immunity of T and B cells (22).
193 194
We i.p. injected EV71 (genotype B5, strain F23, Materials & Methods) into 1-week old NOD/SCID mice,
195
and monitored the time course of viral titer in the sera post-inoculation (Fig. 1B). On day 3 post inoculation
196
(dpi), viral titer reached a peak, suggesting that EV71 was replicating in the inoculated mice. However, viral
197
titer began to decline on day 4, and became undetectable on day 8. We noted occasional (36%) hair loss
198
(Fig. 1C) on day 2-6, and skin rash (Fig. 1D, Fig. S1A) after 9 dpi. In contrast, near 100% limb paralysis
199
was observed on 8-12 dpi (Fig. 1E, Fig. S1B). These phenomena were not detected in PBS-injected
200
NOD/SCID control. To test whether the disease manifestation could be caused by input virus without viral
201
proliferation in vivo, we performed UV-inactivation experiment (Materials and Methods). These
202
UV-inactivated virus produced neither plaques in RD cells (data not shown) nor death and clinical score in
203
NOD/SCID mice inoculated with UV-inactivated virus (Fig. 1F and 1G). This result strongly supports for
204
the correlation between viral replication (Fig. 1B) and pathogenesis (Fig. 1F and 1 G). 10
205
The survival rate and clinical score were dependent on viral dose (below 106 pfu/mouse vs. 107-108
206
pfu/mouse) (Fig. 1H-1I), host age (1 wk vs. 2-4 wk) (Fig. 1J-1K), and injection route (oral, s.c., and i.p.)
207
(Fig. 1L-1M).When mice were inoculated at a lower dose of EV71 (below 106 pfu), or at an older age (2-4
208
wks), no death and clinical score were ever observed (Fig. 1H-1K).Oral infection appeared to be less
209
efficient than infections via i.p. or s.c. route in a highly reproducible manner (Fig. 1N). Therefore, we
210
established a reproducible protocol of in vivo infection by i.p., s.c., and oral injection with an EV71 clinical
211
isolate (B5 genotype, strain F23) using the NOD/SCID mouse system. The i.p. protocol was used in most of
212
the following NOD/SCID studies in this paper.
213 214
Next, we examined the growth kinetics of the inoculated EV71 in different tissues at different time points
215
post-inoculation (Fig. 2A-2H). Viral RNA extracted from various tissues was measured by RT-qPCR
216
analysis using GAPDH as an internal control, and infectious virus titer was measured by standard plaque
217
formation assay using human RD rhabdomyosarcoma cells (Materials and Methods). Interestingly, we
218
observed two general phenomena in most tissues:
219
tissues examined, and viral RNA tended to decline on day 3 in most tissues, except for kidney and heart. In
220
kidney (Fig. 2C), the rapid drop of viral RNA occurred earlier on day 2, instead of day 3. In the heart tissue
221
(Fig. 2D), we observed no decline of viral RNA throughout the entire time course. In general, the decline of
222
infectivity of viral titer appeared to precede that of viral RNA. We also noted that in brain and muscle (Fig.
223
2F and 2G), there were small peaks of viral RNA at 2 dpi. Unlike the tissues of intestine, spleen, and kidney,
224
muscle and brain are anatomically most distal to the i.p. injection site. Therefore, the viral RNA titers in
225
brain and muscle at 1 dpi were very low. In the first round of “tug of war” between the host defense
226
mechanism and the replicating virus in brain and muscle, host defense of the immunodeficient NOD/SCID
227
mice enjoyed a transient victory over the invading virus at 2-3 dpi. However, in the second round of “tug of
1) Viral titer tended to decline on day 2 in all eight
11
228
war”, replicating virus began to take off after 3 dpi. Overall, such a “fall-and-rise” pattern of viral titer and
229
viral RNA in most tissues at earlier time points post-inoculation, strongly suggests host-attempted clearance
230
of the input virus (tissue culture-derived) at the initial phase of viral infection by an unknown host system.
231
The following trend of a steady increase of viral titer and RNA post-infection, lends a strong support for in
232
vivo EV71 replication in NOD/SCID mice. 2) At the later time points before death (8-11 dpi), infectious
233
viral titer dropped to a very low or undetectable level in most tissues, except for spleen (Fig. 2B) and
234
muscle (Fig. 2F). In sharp contrast, the viral RNA level at 8-11 dpi reduced only slightly in spleen and brain,
235
or remained to be increasing in other tissues. This discrepancy between infectious viral titer and viral RNA
236
seems to be paradoxical, and we interpret it as due to the significantly delayed clearance of non-infectious
237
viral RNA relative to the more rapid clearance of viral infectivity, as was observed in the earlier phase at
238
2-3 dpi. These non-infectious viral RNA could originate from infectious virus being inactivated by the
239
released cytokines during inflammation (See Discussion). We speculate here that the amplification of viral
240
titer before reaching the point of disease onset could explain for why the kinetics of clearance of viral RNA
241
lagged far behind that of viral infectivity. In other words, clearance of non-infectious viral RNA could be
242
observed, if the infected NOD/SCID mice could have survived for a longer period.
243 244
We examined sectioned tissues between EV71 infected and mock infected tissues by H&E staining. Most
245
apparent pathologies were noted in spleen, muscle, spinal cord and skin (Fig. 3A). In spinal cord, shrunken
246
neurons with dark nuclei were consistent with degenerated neurons, as highlighted by black arrows. In
247
contrast to the control, the infected muscle was atrophic and irregular in size. In skin section, we noted
248
vacuoles (arrowheads in Fig. 3A) in the dermis of infected mice with skin lesions, suggesting inflammation
249
related edema. The spleen of infected mice revealed extensive dense collagen bands on H&E stained
250
sections, indicative of fibrosis, which was confirmed by the Masson’s trichrome staining as the blue colored 12
251
deposits (Fig. 3B). Spleen atrophy was revealed by the reduced size (left panel, Fig. 3C) and weight (right
252
panel, Fig. 3C) of the dissected spleens of EV71 infected NOD/SCID mice. EV71-encoded VP1 protein can
253
be detected as brown color in spleen, muscle, and skin by immunohistochemical staining (IHC) using an
254
anti-VP1 polyclonal antibody (Fig. 3D). We also detected a cell surface marker of infiltrating macrophages
255
using anti-F4/80 monoclonal antibody (Fig. 3E). Using antibodies specific for keratin 6, keratin 14, and
256
pan-collagen, we performed IHC staining of skin sections, and detected no apparent difference in the
257
expression of keratin and collagen between infected mice and mock infected controls (Fig. S2).
258 259
We compared the hematological characteristics between infected mice and mock control by complete blood
260
counts (Materials and Methods), and detected no significant difference in platelets and monocytes (data not
261
shown), minor difference in MCV (mean cell volume), MCH (mean cell hemoglobin), and RDW (Red
262
blood cell volume distribution width), and more significant difference in the numbers of neutrophil,
263
eosinophil, and lymphocytes (Fig. 4). This reduction in lymphocytes is consistent with the interpretation
264
that EV71 in the NOD/SCID system could bias the differentiation of bone marrow cells toward the myeloid
265
over the lymphoid lineages. The significantly increased neutrophil and reduced eosinophil also suggest a
266
strong influence of EV71 on the myeloid differentiation pathways toward neutrophil over the eosinophil.
267 268
In addition to the NOD/SCID model, we succeeded in establishing stat-1 KO mice as another model
269
susceptible to in vivo infection with a clinical isolate of EV71 (genotype B5, strain F23). Although we did
270
not succeed in productive infection using the subcutaneous (s.c.) route, we observed significant clinical
271
score by i.p. injection of 108 pfu EV71 into 1-wk old stat-1 KO mice (Fig. 5A). Again, host age is a critical
272
factor, since we have observed no productive infection and disease manifestation in 2-wk-old mice.
273
Approximately 30% of the infected mice developed limb paralysis around 7 dpi (Fig. 5B), and no death was 13
274
observed before sacrifice up to three weeks post-inoculation (data not shown). As shown in Fig. 5C, we
275
observed evident pathology on H&E stained sections in EV71 infected mice, including loss of intestinal
276
villi, dis-organized muscle structure, and spleen fibrosis. Although there was no significant pathology in
277
brain stem and spinal cord by H&E staining (Fig. 5C), strong expression of VP1 protein was observed by
278
IHC staining in pons of brain stem, cerebellar Purkinje cells, superior colliculus, and grey matter of spinal
279
cord (Fig. 5D, 5E). Unlike the NOD/SCID model (Fig. 3D), we detected no VP1 protein and inflammation
280
in muscle of the stat-1 KO model, albeit the muscle exhibited a certain degree of dystrophy by H&E (Fig.
281
5C, 5E), which could be a consequence, rather than a cause, of limb paralysis. Injury of neuronal cells is
282
often associated with amplification of adjacent microglial cells (23). Indeed, strong expression of IBA1, a
283
microglial cell marker, was detected in brain and spinal cord by IHC using an anti-IBA1 antibody (Fig. 5F).
284 285
Elevated cytokine levels are known to be associated with inflammation. Previously, increased levels of
286
various combinations of IL-6, IL-10, IP-10, MCP-1, and TNF-αwere detected in serum or tissues in
287
EV71-infected mice or humans (12, 17, 24, 25). We examined the tissue cytokine profiles in spleen, muscle,
288
brain, and spinal cord in EV71-infected NOD/SCID model, and in brain and spinal cord in the stat-1 KO
289
model (Table 1, Fig. S3, Fig. S4). Interestingly, the cytokine profiles, from brain and muscle of
290
EV71-infected NOD/SCID mice, exhibited a high degree of similarity to the serum cytokine profile of
291
EV71-infected patients with pulmonary edema, such as IFN-γ, IL1-β, IL-10, IP-10, MCP-1, TNF-α
292
(muscle), and IL-6 (brain) (25). MCP-1 and IP-10 were reported to be associated with neuronal injury in
293
EV71 infected patients with pulmonary edema (25). IP10 might have a certain degree of protective effect,
294
since ip-10 KO mice infected with a mouse-adapted EV71 variant displayed more severe disease (26). The
295
biological significance of specific cytokines from the NOD/SCID and stat-1 KO models will be discussed
296
further in comparison with human patients and other mouse models. 14
297 298
To compare the growth kinetics of EV71 in these two mouse models more closely, we performed EV71
299
experimental infection of NOD/SCID and stat-1 KO mice side-by-side, and compared the viral RNA,
300
infectivity, and viral protein expression in muscle, brain, and spinal cord, at different time points
301
post-infection (before disease onset, onset, and after-onset) (Fig. 6). The viral RNA increased in copy
302
number, reached a peak around disease onset, and declined after disease onset in both models (Fig. 6A). In
303
contrast to the muscle and brain tissues, spinal cord of the stat-1 KO model contained a higher copy number
304
of viral RNA at disease onset. The infectious titers of EV71 in these three tissues were also compared
305
between these two models at different time points post-infection (Fig. 6B). Major differences in pfu activity
306
between these two models include: 1) at disease onset, no infectious titers could be detected in brain and
307
spinal cord in NOD/SCID mice; and 2) after-onset, while infectious EV71 could still be detected in the
308
muscle of NOD/SCID mice, no infectivity was detected in the muscle of stat-1 KO mice. We also compared
309
viral protein expression between these two models (Fig. 6C). At disease onset, while VP1 protein was
310
detected in the muscle of NOD/SCID mice, it was not detectable in stat-1 KO mice (upper panel). In
311
contrast, while VP1 protein was detected in the spinal cord of stat-1 KO mice, it was not detectable in
312
NOD/SCID mice (bottom panel). No VP1 protein was detected in the brain tissue in both models at disease
313
onset (middle panel).In summary, results in Figure 6 support the general notion that the same clinical isolate
314
of EV71 exhibited different tissue tropism in two different mouse models –muscle tropism in the
315
NOD/SCID model and neurotropism in the stat-1 KO model.
316 317
It is known that the stat-1 KO mice are defective in both IFN-alpha and IFN-gamma signaling (27, 28).
318
Since we did not observe any disease development by infecting the IFN-alpha receptor KO mice (ifnar KO)
319
with EV71 (Fig. 1A), we attempted to block the IFN signaling in vivo by administering antibodies specific 15
320
for both IFN-alpha receptor and IFN-gamma (Materials and Methods). One-week old wild type C57BL/6
321
mice were coinjected once with 150 ug/mouse of each anti-IFN-alpha receptor and anti-IFN-gamma
322
antibodies, one day before inoculation with EV71 (108 pfu/mouse). The antibody treatment was repeated
323
two more times after EV71 inoculation. We observed no disease development in this double antibody
324
experiment (data not shown). As an alternative approach, we tested the effect of IFN by using IFN-gamma
325
receptor knockout mice (ifngr KO) (29). As shown in Figure 7, we observed a more rapid onset of disease
326
and death in this model than those of NOD/SCID and stat-1 KO models (Fig. 7A and 7B). The death and
327
paralysis rates of the ifngr KO model are generally higher than that of the stat-1 KO model (78% vs. 31%).
328
Taken together, IFN-gamma appears to be important for protection against EV71 infection.
329 330 331
DISCUSSION
332
We established here three models for EV71 in vivo infection using NOD/SCID, stat-1 KO, and ifngr KO
333
mice. Since these mice are non-transgenic, and contain no putative human SCARB2 receptor for viral entry
334
(14, 15), it is puzzling that they could support EV71 infection and replication (Fig. 1B, 1F, 1G, Fig. 2). It is
335
possible that mouse SCARB2, which shares 99.9% amino acid sequence similarity with human SCARB2,
336
could serve as a surrogate receptor for viral entry, albeit at a much reduced efficiency (30). In addition, their
337
susceptibility to EV71 infection must be related to their immunodeficiency. NOD/SCID mice are deficient
338
not only in mature T and B cells, but also in NK cells (22). Unlike the stat-1 KO or AG129 mice,
339
NOD/SCID mice are not defective in IFN signaling. The pathogenesis profiles between NOD/SCID and
340
stat-1 KO mice, infected with the same EV71 clinical isolate F23, are very different and complementary to
341
each other (Table S1, Supplemental Material).
342 16
343
In vivo viral replication and tissue tropism in the animal models. The rapid “fall-and-rise” pattern of
344
viral RNA and infectious titer in most tissues in the NOD/SCID model lends a strong support for an active
345
replication of EV71 in vivo (Fig. 2). The clearance of the non-infectious and infectious viral RNAs
346
appeared to lag far behind the clearance of viral infectivity. By comparing the growth curves of the
347
inoculated virus in NOD/SCID mice and stat-1 mice side-by-side, it became clear that the viral RNA titer
348
reached the peak at the point of disease onset in both models (Fig. 6A). In contrast to the active viral growth
349
in Fig. 2 and 6, UV-inactivated virus produced no disease and clinical score in the NOD/SCID mice (Fig.
350
1F and 1G). Taken together, these results strongly suggest in vivo synthesis of viral RNA genome and
351
infectious virus titer before the disease onset. The presence of infectious virus titer in the muscle at 11 dpi
352
(Fig. 2F) and the strong presence of VP1 antigen by Western blot at disease onset (8 dpi) (Fig. 6C), favors
353
the notion of muscle-tropism of EV71 in the NOD/SCID model. In contrast, the detection of VP1 in the
354
brain and spinal cord by IHC (Fig. 5D and 5E) and in the spinal cord by Western blot (Fig. 6C), favors the
355
notion of neurotropism of EV71 in the stat-1 model.
356 357
HFMD-like skin rash. As shown in Table 2, we compared the characteristics of several currently available
358
mouse models for EV71 infection, including ICR mice or IP-10 knockout mice, infected with
359
mouse-adapted EV71 variants (8, 26), SCARB2 Tg mice(16, 17), and interferon-deficient AG129(12). To
360
our knowledge, the NOD/SCID system is the only model that mimicks cutaneous lesion or skin rash of
361
human HFMD (Table2), which was found in about 70% of EV71 infected children with disease
362
manifestation (1, 2, 31, 32). No skin rash was observed in previous mouse studies (8, 16, 17, 33, 34), or the
363
cynomolgusmonkey model (11, 35).In one hSCARB2 Tg model, hair loss, but no skin rash, was observed
364
(17). In our current NOD/SCID model, both hair loss and skin rash were detected in 36 % of mice
365
inoculated i.p., and 14% of mice via oral infection (Fig. 1N).We observed VP1 expression by IHC in hair 17
366
follicles (Fig. 3D), and edema-like vacuoles in dermis (Fig. 3A) in EV71 i.p. infected NOD/SCID model.
367 368
Spleen fibrosis, identified by the Masson’s trichrome staining, is common in both NOD/SCID andstat-1
369
KO mouse models (Fig. 3B, Fig. 5C). However, only spleen atrophy (2, 12, 36, 37), but no spleen fibrosis,
370
has ever been reported in mouse models and humans. Whether spleen fibrosis can be detected in human
371
natural infection deserves further investigation in the future.
372 373
Limb paralysis due to myositis or CNS involvement? EV71 infection presents two major clinical
374
features: HFMD and neurological disorder (2). In AG129 mouse model and hSCARB2 Tg model, viral
375
protein and injury were found in both limb muscle and CNS (12, 17). It is therefore difficult to definitively
376
sort out whether limb paralysis is primarily caused by muscle or CNS injury.As discussed below, we
377
addressed this issueby using the same virus strain in two complementary models of NOD/SCID and stat-1
378
KO mice.
379 380
In the stat-1 KO model, we observed limb paralysis (30%), as well as strong expression of viral VP1
381
antigen (Fig. 5D, 5E) and inflammation (microglial IBA1 protein in Fig. 5F), only in the brain stem and
382
spinal cord, but not in the limb muscle (Fig. 5C and 5E). In this regard, EV71 in thestat-1 KO model is
383
highly neurotropic than muscle-tropic, and limb paralysis is primarily caused by neurological disease. In
384
contrast, in the NOD/SCID model, we obtained no direct evidence for neurotropism of EV71. Although
385
near 100% of infected NOD/SCID mice developed limb paralysis, we detected no VP1 in the CNS of
386
infected NOD/SCID mice by IHC on 4, 8, and 12 dpi (data not shown). In addition, infectious virus
387
appeared to have been cleared from CNS before 8 dpi (Fig. 2G, Fig. 2H). Given the massive destruction and
388
strong expression of VP1 in muscle (Fig. 3A, Fig. 3D), but not in the CNS (data not shown), we interpreted 18
389
here that limb paralysis in the NOD/SCID model could be caused mainly by a muscle, rather than
390
neurological problem. Taken together, limb paralysis could be caused by either muscle or neurological
391
injury, depending on the immunogenetic background of the host animals.
392 393
Viral spread pathways. Based on the IHC staining of VP1 in the stat-1 KO model, VP1 was strongly
394
expressed in all levels of spinal cord, including anterior horn and all areas of grey matter (Fig. 5E). In
395
addition, we detected VP1 in pons of brain stem (Fig. 5D), but not in the cerebral cortex (data not shown).
396
Taken together, this result is in support for the retrograde axonal transport from peripheral nerves to spinal
397
cord ascending to CNS, in a manner similar to other enterovirus systems, including EV71 mouse-adapted
398
strains(34, 38, 39), and poliovirus(40). On the other hand, we also detected VP1 expression in a small area
399
of superior colliculus of mid-brain (Fig. 5D), which is not anatomically or directly connected with spinal
400
cord. IL-17, as detected in Table 1, has been reported to influence the permeability of the blood brain barrier
401
(41). It cannot be excluded that EV71 could somehow pass through the blood brain barrier, perhaps at a
402
lower efficiency, in infected mice.
Future investigation is needed to further address this issue.
403 404
Pulmonary edema. Pulmonary edema is associated with high mortality rate in EV71-infected patients with
405
brainstem encephalitis (31, 42-44). In cases of bulbar poliomyelitis, fatal pulmonary edema was also
406
reported (45). So far, no pulmonary edema has ever been observed in any EV71 animal models (12, 16, 17,
407
33), including our own NOD/SCID and stat-1 KO mice (data not shown). While pulmonary edema was
408
generally believed to be neurogenic, it cannot be excluded whether pulmonary failure could be contributed
409
in part by myositis involved in lung respiration. As discussed earlier, we observed no apparent injury or
410
viral protein in the CNS of our NOD/SCID model (data not shown); however, massive muscle damage is
411
evident (Fig. 3A), a phenomenon similar to a recent report on pulmonary hypoventilation in 2-week-old 19
412
ICR mice infected with a mouse-adapted strain of EV71 (46).
413 414
Oral infection via a natural enteric route. Inoculation via the oral route with mouse-adapted strains of
415
EV71 can result in productive infection and disease in mice (8, 33). In the human SCARB2 Tg mouse
416
model, neurological signs can be obtained when inoculated via the i.p. route. However, when inoculated via
417
the intragastric (i.g.) route at 107 TCID50, only one out of twenty (1/20) mice showed neurological signs
418
(16) . In a separate experiment using 108 TCID50, no incidence of neurological signs (0/10) was observed.
419
The reason for the very low efficiency of oral infection with EV71 in the hSCARB2 Tg model (16) remains
420
unclear. Using the IFN-α receptor KO model, we obtained no productive infection via the oral route either
421
(Fig. 1A). However, recently, we succeeded reproducibly in oral infection in the NOD/SCID mouse model
422
(Fig. 1L, 1M, 1N, Table 2).Similarly, AG129 mice defective in both IFN-α/β and IFN-γ signaling, can be
423
infected with EV71 via the oral route (12). While we succeeded in establishing an efficient in vivo infection
424
via the i.p. route using the ifngr KO model (Fig. 7), it remains unclear if we can establish oral infection
425
using this ifngr KO model. The potential role of IFN signaling in EV71 clearance and pathogenesis remains
426
to be elucidated in the future (47).
427 428
Cytokine profiles and infiltrating macrophages. We noted the striking similarity of cytokine profiles
429
between the NOD/SCID model (muscle, brain, and spinal cord) and patients (serum and cerebrospinal
430
fluid), including IFN-γ, IL-10, IP-10, IL-6, MCP-1, IL-1β, and TNF-α (25) (Table 1). Therefore, the
431
NOD/SCID model, despite its immunodeficiency, could serve as an animal model resembling human
432
EV71 infection, at least as far as cytokine profiles are concerned. In the AG129 model (12), serum IL-1β
433
was induced by EV71 only via the oral route, but not the i.p. route. In contrast, in our studies via the i.p.
434
route, we detected IL-1β in brain and muscle of the NOD/SCID model. Furthermore, it is interesting to 20
435
note that IL-17F was detected in the spinal cord of both EV71-infected NOD/SCID and stat-1 KO models
436
(Table 1). IL-17 cytokines, including IL-17F, are known to be involved in neuroinflammation in
437
experimental autoimmune encephalomyelitis (EAE) (48). We also detected IL-23 in the spinal cord of the
438
NOD/SCID model (Table 1). IL-23, a member of the IL-12 family of cytokines, can promote Th17 cell
439
development. The IL-23/IL-17 axis is well known to play an important role in various inflammatory
440
diseases (49). Finally, the fact that we detected macrophage inflammation protein 2 (MIP2) in CNS and
441
muscle (Table 1), as well as F4/80 in the muscle of NOD/SCID mice (Fig. 3E), suggests that infiltrating
442
macrophages could contribute to inflammation and cytokine release in this model.
443 444
Interferon-gamma signaling and viral infection. IFN-gamma is known to exercise pleiotropic effects on
445
the immune system. In addition to its antiviral effect, IFN-gamma plays an immunomodulatory role in
446
macrophages, dendritic cells, and lymphoid cells (50-52). Mice without IFN-gamma receptor exhibited
447
increased susceptibility to infection with Listeria monocytogenes, vaccinia virus (29), lymphocytic
448
choriomeningitis virus (LCMV) (53-55), herpes simplex virus type 1 (HSV-1) (51), and dengue
449
infection-induced paralysis (56). However, IFN-gamma and its receptor do not seem to be important for
450
infection with vesicular stomatitis virus (VSV) (55, 57, 58). In the case of EV71, we demonstrated here a
451
protective role of IFN-gamma receptor to EV71-induced paralysis and death (Fig. 7).
452 453
As shown in Table 1, we detected IFN-gamma in muscle, brain, and spinal cord of NOD/SCID mice as well
454
as in the muscle of the stat-1 mice. The absence of IFN-gamma in the spinal cord of the stat-1 mice is
455
consistent with the strong presence of VP1 antigen in the spinal cord by Western blot analysis (lower panel,
456
Fig. 6C). Similarly, the presence of IFN-gamma in the brain and spinal cord of the NOD/SCID mice (Table
457
1) could explain its absence of detectable viral antigen and pathology in the brain and spinal cord (Fig. 6C). 21
458
Although IFN-gamma can be detected in muscle of the NOD/SCID mice, viral RNA, viral protein, and
459
infectious titer can be detected in the muscle of NOD/SCID mice at disease onset (Fig. 6).It is possible that
460
the existing level of IFN-gamma is too low to clear the virus effectively.
461 462
It remains somewhat puzzling that the stat-1KO mice, defective in both IFN-alpha/beta and IFN-gamma
463
signalings, had a paralysis rate around 30% and no death; while the ifngr KO mice, defective in only the
464
IFN-gamma, but not IFN-alpha, displayed a paralysis and death rate near 78%.Because of the difference in
465
their respective genetic backgrounds (C57BL/6 for ifnar KO vs. 129/Sv/Ev for ifngr KO), it is impossible to
466
compare these two mouse systems at present. Another issue that may be worth discussion here is that, we
467
did not succeed in inducing paralysis in our anti-IFN-gamma antibody experiment (data not shown).
468
However, ifngr KO mice displayed paralysis upon infection with EV71. It is possible that the blockage of a
469
cytokine ligand may not have the same phenotype as the blockage of a cytokine receptor. For example,
470
there appeared to be no difference in the infection with LCMV between IFN-gamma KO mice and wild
471
type mice (59). However, increased susceptibility to LCMV infection was reported by using ifngr KO mice
472
(55).
473 474
Age-dependent susceptibility. In humans, young age is a major risk factor for natural infection with EV71
475
(1, 2).
476
and death. For example, in our NOD/SCID and stat-1 KO models, we routinely i.p. inoculated
477
one-week-old mice for most studies. In other models, mice younger than two or three weeks old were
478
susceptible to EV71 infection via various routes and viral strains (12, 16, 17). Similarly, in one poliovirus
479
receptor (PVR) transgenic model, 2-week-old mice were four orders of magnitude more susceptible to
480
paralysis than adult mice (60). It remains to be investigated what kind of viral or host factors could
In animal models, younger age is also a crucial determinant for productive infection, pathogenesis,
22
481
contribute to the age dependency, such as the internal ribosome entry site (61), and the neonatal
482
development of the nerve, gut, and/or immune systems.
483 484
Potential for translational research. One last question to be addressed here is whether these
485
immunodeficient mice are adequate for EV71 vaccine research? Our answer is positive since these models
486
can still provide an in vivo platform for evaluation of vaccine efficacy by passive transfer of neutralizing
487
antibodies. Such antibodies can be induced by vaccination in immuno-competent animals or humans.
488
Finally, these user-friendly mouse models can be infected with clinical isolates of EV71, and thus provide
489
an opportunity for in vivo testing of anti-virals.
490 491
Acknowledgments.We thank the service by the Pathology Core Laboratory, Institute of Biomedical
492
Sciences (IBMS), Academia Sinica, Taiwan. This work is supported by Academia Sinica and National
493
Science Council, Taiwan.
494
23
495
References
496
1.
Ho, M., E. R. Chen, K. H. Hsu, S. J. Twu, K. T. Chen, S. F. Tsai, J. R. Wang, and S. R.
497
Shih. 1999. An epidemic of enterovirus 71 infection in Taiwan. Taiwan Enterovirus Epidemic
498
Working Group. N Engl J Med 341:929-35.
499
2.
features, diagnosis, and management of enterovirus 71. The Lancet Neurology 9:1097-1105.
500 501
3.
Racaniello, V. R. 2007. Picornaviridae: the viruses and their replication., p. 796-839. In H. P. Knipe DM (ed.), Fields virology, 5th ed. Lippincott Williams & Wilkins, Inc., Philadelphia.
502 503
Ooi, M. H., S. C. Wong, P. Lewthwaite, M. J. Cardosa, and T. Solomon. 2010. Clinical
4.
Feng, Z., L. Hensley, K. L. McKnight, F. Hu, V. Madden, L. Ping, S. H. Jeong, C. Walker,
504
R. E. Lanford, and S. M. Lemon. 2013. A pathogenic picornavirus acquires an envelope by
505
hijacking cellular membranes. Nature 496:367-71.
506
5.
Curr Top Microbiol Immunol 323:89-121.
507 508
6.
7.
Lee, M. S., F. C. Tseng, J. R. Wang, C. Y. Chi, P. Chong, and I. J. Su. 2012. Challenges to licensure of enterovirus 71 vaccines. PLoS Negl Trop Dis 6:e1737.
511 512
Lee, M. S., and L. Y. Chang. 2010. Development of enterovirus 71 vaccines. Expert Rev Vaccines 9:149-56.
509 510
Sean, P., and B. L. Semler. 2008. Coxsackievirus B RNA replication: lessons from poliovirus.
8.
Wang, Y. F., C. T. Chou, H. Y. Lei, C. C. Liu, S. M. Wang, J. J. Yan, I. J. Su, J. R. Wang,
513
T. M. Yeh, S. H. Chen, and C. K. Yu. 2004. A mouse-adapted enterovirus 71 strain causes
514
neurological disease in mice after oral infection. J Virol 78:7916-24.
515
9.
Li, Z. H., C. M. Li, P. Ling, F. H. Shen, S. H. Chen, C. C. Liu, and C. K. Yu. 2008.
516
Ribavirin reduces mortality in enterovirus 71-infected mice by decreasing viral replication. J
517
Infect Dis 197:854-7. 24
518
10.
Arita, M., Y. Ami, T. Wakita, and H. Shimizu. 2008. Cooperative effect of the attenuation
519
determinants derived from poliovirus sabin 1 strain is essential for attenuation of enterovirus 71
520
in the NOD/SCID mouse infection model. J Virol 82:1787-97.
521
11.
Hashimoto, I., A. Hagiwara, and H. Kodama. 1978. Neurovirulence in cynomolgus monkeys
522
of enterovirus 71 isolated from a patient with hand, foot and mouth disease. Arch Virol
523
56:257-61.
524
12.
Khong, W. X., B. Yan, H. Yeo, E. L. Tan, J. J. Lee, J. K. Ng, V. T. Chow, and S. Alonso.
525
2012. A non-mouse-adapted enterovirus 71 (EV71) strain exhibits neurotropism, causing
526
neurological manifestations in a novel mouse model of EV71 infection. J Virol 86:2121-31.
527
13.
enterovirus 71 to adult interferon deficient mice. PLoS One 8:e59501.
528 529
14.
Yamayoshi, S., Y. Yamashita, J. Li, N. Hanagata, T. Minowa, T. Takemura, and S. Koike. 2009. Scavenger receptor B2 is a cellular receptor for enterovirus 71. Nat Med 15:798-801.
530 531
Caine, E. A., C. D. Partidos, J. D. Santangelo, and J. E. Osorio. 2013. Adaptation of
15.
Nishimura, Y., M. Shimojima, Y. Tano, T. Miyamura, T. Wakita, and H. Shimizu. 2009.
532
Human P-selectin glycoprotein ligand-1 is a functional receptor for enterovirus 71. Nat Med
533
15:794-7.
534
16.
Fujii, K., N. Nagata, Y. Sato, K. C. Ong, K. T. Wong, S. Yamayoshi, M. Shimanuki, H.
535
Shitara, C. Taya, and S. Koike. 2013. Transgenic mouse model for the study of enterovirus
536
71 neuropathogenesis. Proc Natl Acad Sci U S A 110:14753-8.
537
17.
Lin, Y. W., S. L. Yu, H. Y. Shao, H. Y. Lin, C. C. Liu, K. N. Hsiao, E. Chitra, Y. L. Tsou,
538
H. W. Chang, C. Sia, P. Chong, and Y. H. Chow. 2013. Human SCARB2 transgenic mice as
539
an infectious animal model for enterovirus 71. PLoS One 8:e57591.
540
18.
Huang, M. C., S. M. Wang, Y. W. Hsu, H. C. Lin, C. Y. Chi, and C. C. Liu. 2006. 25
541
Long-term cognitive and motor deficits after enterovirus 71 brainstem encephalitis in children.
542
Pediatrics 118:e1785-8.
543
19.
Petersen, L. R., A. C. Brault, and R. S. Nasci. 2013. West Nile virus: review of the literature. JAMA 310:308-15.
544 545
20.
Zompi, S., and E. Harris. 2012. Animal models of dengue virus infection. Viruses 4:62-82.
546
21.
Stark, G. R., and J. E. Darnell, Jr. 2012. The JAK-STAT pathway at twenty. Immunity 36:503-14.
547 548
22.
Shultz, L. D., P. A. Schweitzer, S. W. Christianson, B. Gott, I. B. Schweitzer, B. Tennent,
549
S. McKenna, L. Mobraaten, T. V. Rajan, D. L. Greiner, and et al. 1995. Multiple defects in
550
innate and adaptive immunologic function in NOD/LtSz-scid mice. J Immunol 154:180-91.
551
23.
hyperalgesia and spinal microglia reactivity in neonatal rats. Eur J Pain 17:1180-8.
552 553
24.
Weng, K. F., L. L. Chen, P. N. Huang, and S. R. Shih. 2010. Neural pathogenesis of enterovirus 71 infection. Microbes Infect 12:505-10.
554 555
Vega-Avelaira, D., J. J. Ballesteros, and J. A. Lopez-Garcia. 2013. Inflammation-induced
25.
Wang, S. M., H. Y. Lei, C. K. Yu, J. R. Wang, I. J. Su, and C. C. Liu. 2008. Acute
556
chemokine response in the blood and cerebrospinal fluid of children with enterovirus
557
71-associated brainstem encephalitis. J Infect Dis 198:1002-6.
558
26.
Shen, F. H., C. C. Tsai, L. C. Wang, K. C. Chang, Y. Y. Tung, I. J. Su, and S. H. Chen.
559
2013. Enterovirus 71 infection increases expression of interferon-gamma-inducible protein 10
560
which protects mice by reducing viral burden in multiple tissues. J Gen Virol 94:1019-27.
561
27.
the mouse Stat1 gene results in compromised innate immunity to viral disease. Cell 84:443-50.
562 563
Durbin, J. E., R. Hackenmiller, M. C. Simon, and D. E. Levy. 1996. Targeted disruption of
28.
Meraz, M. A., J. M. White, K. C. Sheehan, E. A. Bach, S. J. Rodig, A. S. Dighe, D. H. 26
564
Kaplan, J. K. Riley, A. C. Greenlund, D. Campbell, K. Carver-Moore, R. N. DuBois, R.
565
Clark, M. Aguet, and R. D. Schreiber. 1996. Targeted disruption of the Stat1 gene in mice
566
reveals unexpected physiologic specificity in the JAK-STAT signaling pathway. Cell
567
84:431-42.
568
29.
Huang, S., W. Hendriks, A. Althage, S. Hemmi, H. Bluethmann, R. Kamijo, J. Vilcek, R.
569
M. Zinkernagel, and M. Aguet. 1993. Immune response in mice that lack the
570
interferon-gamma receptor. Science 259:1742-5.
571
30.
important for enterovirus 71 binding and infection. J Virol 85:4937-46.
572 573
Yamayoshi, S., and S. Koike. 2011. Identification of a human SCARB2 region that is
31.
Lum, L. C., K. T. Wong, S. K. Lam, K. B. Chua, A. Y. Goh, W. L. Lim, B. B. Ong, G.
574
Paul, S. AbuBakar, and M. Lambert. 1998. Fatal enterovirus 71 encephalomyelitis. J Pediatr
575
133:795-8.
576
32.
Shindarov, L. M., M. P. Chumakov, M. K. Voroshilova, S. Bojinov, S. M. Vasilenko, I.
577
Iordanov, I. D. Kirov, E. Kamenov, E. V. Leshchinskaya, G. Mitov, I. A. Robinson, S.
578
Sivchev, and S. Staikov. 1979. Epidemiological, clinical, and pathomorphological
579
characteristics of epidemic poliomyelitis-like disease caused by enterovirus 71. J Hyg
580
Epidemiol Microbiol Immunol 23:284-95.
581
33.
involvement. Journal of General Virology 85:69-77.
582 583
Chen, Y. C. 2004. A murine oral enterovirus 71 infection model with central nervous system
34.
Ong, K. C., M. Badmanathan, S. Devi, K. L. Leong, M. J. Cardosa, and K. T. Wong. 2008.
584
Pathologic characterization of a murine model of human enterovirus 71 encephalomyelitis. J
585
Neuropathol Exp Neurol 67:532-42.
586
35.
Nagata, N., T. Iwasaki, Y. Ami, Y. Tano, A. Harashima, Y. Suzaki, Y. Sato, H. Hasegawa, 27
587
T. Sata, T. Miyamura, and H. Shimizu. 2004. Differential localization of neurons susceptible
588
to enterovirus 71 and poliovirus type 1 in the central nervous system of cynomolgus monkeys
589
after intravenous inoculation. J Gen Virol 85:2981-9.
590
36.
MENG Guo-zhao, L. M.-q., LI Yun-qian, WANG Xu-ming, CHEN Qiu-yue, WEI
591
Hua-sheng. 2011. Severe hand-foot-mouth disease:a report of 3 autopsied cases. Chinese
592
Journal of Clinical and Experimental Pathology 27:48-51,55.
593
37.
Khong, W. X., D. G. Foo, S. L. Trasti, E. L. Tan, and S. Alonso. 2011. Sustained high levels
594
of interleukin-6 contribute to the pathogenesis of enterovirus 71 in a neonate mouse model. J
595
Virol 85:3067-76.
596
38.
Chen, C. S., Y. C. Yao, S. C. Lin, Y. P. Lee, Y. F. Wang, J. R. Wang, C. C. Liu, H. Y. Lei,
597
and C. K. Yu. 2007. Retrograde axonal transport: a major transmission route of enterovirus 71
598
in mice. J Virol 81:8996-9003.
599
39.
Wong, K. T., B. Munisamy, K. C. Ong, H. Kojima, N. Noriyo, K. B. Chua, B. B. Ong, and
600
K. Nagashima. 2008. The distribution of inflammation and virus in human enterovirus 71
601
encephalomyelitis suggests possible viral spread by neural pathways. J Neuropathol Exp
602
Neurol 67:162-9.
603
40.
system by neural pathways. J Infect Dis 166:747-52.
604 605
Ren, R., and V. R. Racaniello. 1992. Poliovirus spreads from muscle to the central nervous
41.
Kebir, H., K. Kreymborg, I. Ifergan, A. Dodelet-Devillers, R. Cayrol, M. Bernard, F.
606
Giuliani, N. Arbour, B. Becher, and A. Prat. 2007. Human TH17 lymphocytes promote
607
blood-brain barrier disruption and central nervous system inflammation. Nat Med 13:1173-5.
608 609
42.
Chang, L.-Y., Y.-C. Huang, and T.-Y. Lin. 1998. Fulminant neurogenic pulmonary oedema with hand, foot, and mouth disease. The Lancet 352:367-368. 28
610
43.
Wang, S.-M., C.-C. Liu, H.-W. Tseng, J.-R. Wang, C.-C. Huang, Y.-J. Chen, Y.-J. Yang,
611
S.-J. Lin, and T.-F. Yeh. 1999. Clinical Spectrum of Enterovirus 71 Infection in Children in
612
Southern Taiwan, with an Emphasis on Neurological Complications. Clinical Infectious
613
Diseases 29:184-190.
614
44.
Chang, L. Y., T. Y. Lin, K. H. Hsu, Y. C. Huang, K. L. Lin, C. Hsueh, S. R. Shih, H. C.
615
Ning, M. S. Hwang, H. S. Wang, and C. Y. Lee. 1999. Clinical features and risk factors of
616
pulmonary oedema after enterovirus-71-related hand, foot, and mouth disease. Lancet
617
354:1682-6.
618
45.
Baker, A. B. 1957. Poliomyelitis. 16. A study of pulmonary edema. Neurology 7:743-51.
619
46.
Xiu, J. H., H. Zhu, Y. F. Xu, J. N. Liu, X. Z. Xia, and L. F. Zhang. 2013. Necrotizing
620
myositis causes restrictive hypoventilation in a mouse model for human enterovirus 71
621
infection. Virol J 10:215.
622
47.
Kjerrulf, M., D. Grdic, L. Ekman, K. Schon, M. Vajdy, and N. Y. Lycke. 1997.
623
Interferon-gamma receptor-deficient mice exhibit impaired gut mucosal immune responses but
624
intact oral tolerance. Immunology 92:60-8.
625
48.
Kang, Z., C. Wang, J. Zepp, L. Wu, K. Sun, J. Zhao, U. Chandrasekharan, P. E.
626
Dicorleto, B. D. Trapp, R. M. Ransohoff, and X. Li. 2013. Act1 mediates IL-17-induced
627
EAE pathogenesis selectively in NG2(+) glial cells. Nat Neurosci 16:1401-8.
628
49.
functions. Nat Rev Immunol 5:521-31.
629 630
50.
Sen, G. C. 2001. VIRUSES AND INTERFERONS. Annual Review of Microbiology 55:255-281.
631 632
Hunter, C. A. 2005. New IL-12-family members: IL-23 and IL-27, cytokines with divergent
51.
Bigley, N. J. 2014. Complexity of interferon gamma interactions with HSV-1. Frontiers in 29
Immunology 5.
633 634
52.
Interferon Response during Toxoplasma gondii Infection. Infect Immun 82:3090-3097.
635 636
Sturge, C. R., and F. Yarovinsky. 2014. Complex Immune Cell Interplay in the Gamma
53.
Leist, T. P., M. Eppler, and R. M. Zinkernagel. 1989. Enhanced virus replication and
637
inhibition of lymphocytic choriomeningitis virus disease in anti-gamma interferon-treated mice.
638
J Virol 63:2813-9.
639
54.
Klavinskis, L. S., R. Geckeler, and M. B. Oldstone. 1989. Cytotoxic T lymphocyte control of
640
acute lymphocytic choriomeningitis virus infection: interferon gamma, but not tumour necrosis
641
factor alpha, displays antiviral activity in vivo. J Gen Virol 70 ( Pt 12):3317-25.
642
55.
Muller, U., U. Steinhoff, L. F. Reis, S. Hemmi, J. Pavlovic, R. M. Zinkernagel, and M.
643
Aguet. 1994. Functional role of type I and type II interferons in antiviral defense. Science
644
264:1918-21.
645
56.
Shresta, S., J. L. Kyle, H. M. Snider, M. Basavapatna, P. R. Beatty, and E. Harris. 2004.
646
Interferon-dependent immunity is essential for resistance to primary dengue virus infection in
647
mice, whereas T- and B-cell-dependent immunity are less critical. J Virol 78:2701-10.
648
57.
infections of the central nervous system. Cytokine Growth Factor Rev 13:441-54.
649 650
58.
Komatsu, T., and C. S. Reiss. 1997. IFN-gamma is not required in the IL-12 response to vesicular stomatitis virus infection of the olfactory bulb. J Immunol 159:3444-52.
651 652
Chesler, D. A., and C. S. Reiss. 2002. The role of IFN-gamma in immune responses to viral
59.
Nansen, A., J. P. Christensen, C. Ropke, O. Marker, A. Scheynius, and A. R. Thomsen.
653
1998. Role of interferon-gamma in the pathogenesis of LCMV-induced meningitis: unimpaired
654
leucocyte recruitment, but deficient macrophage activation in interferon-gamma knock-out
655
mice. J Neuroimmunol 86:202-12. 30
656
60.
Crotty, S., L. Hix, L. J. Sigal, and R. Andino. 2002. Poliovirus pathogenesis in a new
657
poliovirus receptor transgenic mouse model: age-dependent paralysis and a mucosal route of
658
infection. J Gen Virol 83:1707-20.
659
61.
Kauder, S., S. Kan, and V. R. Racaniello. 2006. Age-dependent poliovirus replication in the
660
mouse central nervous system is determined by internal ribosome entry site-mediated
661
translation. J Virol 80:2589-95.
662
31
663
FIGURE LEGENDS
664
Figure. 1
665
mice infected with a clinical isolate of EV71 strain F23 (genotype B5). (A) Comparisons among various
666
immune-deficient mouse models infected with EV71 via different routes (*limb paralysis rate about 78%;
667
** limb paralysis rate about 31.3%). (B) One-week-old NOD/SCID mice were infected with 108 pfu of
668
EV71 strain F23. Serum samples were collected from a group of five mice on day 1, 2, 3, 4, 8, and 11,
669
respectively. Virus titer in serum was measured by plaque assay. One-week-old NOD/SCID mice were
670
infected with EV71 strain F23 (left panel) or normal saline (right panel). Disease manifestations, including
671
hair loss (C), skin rash (D), limb paralysis (E), and death, were monitored daily. Comparisons were made
672
between different parameters of infection, including with or without UV treatment (F and G), different viral
673
dose (105~108 pfu) (Hand I), host age (1~4weeks) (J and K), and inoculation route (i.p., s.c., or oral) (L and
674
M). (N) A summary of disease manifestations in NOD/SCID mice infected with EV71 through different
675
viral doses and routes. Clinical scores were defined as follows: 0: Healthy; 1: Hair loss, wasting, or ruffled
676
hair; 2: Limb weakness; 3: Only one-limb paralysis; 4: 2-4 limb paralysis; 5: Death. Error bars indicate the
677
standard deviations (sd) in 6 mice of each group.
Studies on various parameters which affect disease manifestations of immunodeficient
678 679
Figure 2. A “fall-and-rise” pattern of viral RNA and infectious titer of EV71 at earlier time points
680
post-inoculation in various tissues of EV71-infected NOD/SCID mice. One-week-old NOD/SCID mice
681
were infected i.p. with 108 pfu of EV71 strain F23. On different days post-infection, mice were sacrificed.
682
Viral RNA (left panel) and virus titer (right panel) of intestine(A), spleen (B), kidney (C), heart (D), lung
683
(E), muscle (F), brain (G), and spinal cord (H) were determined by quantitative RT-PCR and plaque assay.
684
Viral RNA was detected by VP1 primers and normalized by GAPDH primers. Virus titer in each tissue was
685
normalized by the gram weight of dissected tissues. These dash lines in various graphs in (A) to (H) were 32
686
drawn by connecting the average points of data at each time point. (see text for further discussion) *:
687
indicating the time point of disease onset.
688 689
Figure 3. Pathological changes of EV71-infected NOD/SCID mice. One-week-old NOD/SCID mice
690
were infected i.p. with 108 pfu of EV71 (left panel) or normal saline (right panel). Moribund mice were
691
sacrificed on day 12 post-infection. The paraffin sections were examined at 200X with H&E stain (A),
692
Masson’s stain (B), anti-EV71 VP1 (D), and anti-F4/80 (E). (A) Multiple blue stained cells in muscle could
693
represent infiltrated macrophages (see Fig. 3E below). The black arrows in spinal cord showed degenerated
694
neurons. The arrowheads in skin and hair follicle highlight vacuous edema at 400X magnification. (B) The
695
blue color showed collagen deposits and spleen fibrosis. (C) Spleen atrophy was observed on day 11
696
post-infection (left panel). The weights of each dissected spleen from infected mice were measured and
697
normalized with hearts (right panel). (D) IHC staining detected VP1 protein in spleen, muscle, and skin hair
698
follicles. (E) Muscle-infiltrated macrophages can be visualized using anti-F4/80 antibody by IHC. The
699
statistics was presented by mean ± standard deviation (sd) in 6 mice of each group. **p<0.01 by Student’s
700
t-test compared with saline control.
701 702
Figure 4. Hematological changes of NOD/SCID mice i.p. infected with EV71 (108 pfu) or normal
703
saline. Blood was collected on day 11 post-infection. Complete blood counts were recorded by Abbott
704
Cell-Dyn 3700. (MCV: Mean cell volume; MCH: Mean cell hemoglobin; RDW: Red blood cell volume
705
distribution width). The statistics was presented by mean ± standard deviation (sd) in 6 mice of each group.
706
*p < 0.05; **p