Lethal Infection of Human ACE2-Transgenic Mice Caused by SARS-CoV-2-related Pangolin Coronavirus GX_P2V(short_3UTR)

Abstract

SARS-CoV-2-related pangolin coronavirus GX_P2V(short_3UTR) can cause 100% mortality in human ACE2-transgenic mice, potentially attributable to late-stage brain infection. This underscores a spillover risk of GX_P2V into humans and provides a unique model for understanding the pathogenic mechanisms of SARS-CoV-2-related viruses.

Full text

Lethal Infection of Human ACE2-Transgenic Mice Caused by SARS-CoV-2-
1
related Pangolin Coronavirus GX_P2V(short_3UTR)
2
3
Abstract
4
SARS-CoV-2-related pangolin coronavirus GX_P2V(short_3UTR) can cause 100%
5
mortality in human ACE2-transgenic mice, potentially attributable to late-stage brain
6
infection. This underscores a spillover risk of GX_P2V into humans and provides a
7
unique model for understanding the pathogenic mechanisms of SARS-CoV-2-related
8
viruses.
9
10
Dear Editor,
11
Two SARS-CoV-2-related pangolin coronaviruses, GD/2019 and GX/2017, were
12
identified prior to the COVID-19 outbreak (1, 2). The respective isolates, termed pCoV-
13
GD01 and GX_P2V, were cultured in 2020 and 2017, respectively (2, 3). The infectivity
14
and pathogenicity of these isolates have been studied (4-6). The pCoV-GD01 isolate,
15
which has higher homology with SARS-CoV-2, can infect and cause disease in both
16
golden hamsters and hACE2 mice (4). In contrast, while GX_P2V can also infect both
17
species, it does not appear to cause obvious disease in these animals (5, 6). We
18
previously reported that the early passaged GX_P2V isolate was actually a cell culture-
19
adapted mutant, named GX_P2V(short_3UTR), which possesses a 104-nucleotide
20
deletion at the 3’-UTR (6). In this study, we cloned this mutant, considering the
21
propensity of coronaviruses to undergo rapid adaptive mutation in cell culture, and
22
.CC-BY-NC-ND 4.0 International licensemade available 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(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.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
The copyright holder for this preprintthis version posted January 4, 2024. ; https://doi.org/10.1101/2024.01.03.574008doi: bioRxiv preprint

assessed its pathogenicity in hACE2 mice. We found that the GX_P2V(short_3UTR)
23
clone can infect hACE2 mice, with high viral loads detected in both lung and brain
24
tissues. This infection resulted in 100% mortality in the hACE2 mice. We surmise that
25
the cause of death may be linked to the occurrence of late brain infection.
26
The GX_P2V(short_3UTR) mutant, initially isolated from the early passages of
27
the GX_P2V sample (6), and the GX_P2V virus itself, have not been studied in terms
28
of their adaptive mutations in cell cultures. To obtain a genetically homogenous clone
29
for animal experiments, we cloned the passaged mutant through two successive plaque
30
assays. Eight viral clones were chosen for next-generation sequencing (National
31
Genomics Data Center of China, GSA: CRA014225). These clones, when compared
32
with the genome of the original mutant (6), all shared four identical mutations:
33
ORF1ab_D6889G, S_T730I, S_K807N, and E_A22D (Supporting Information, Table
34
S1). Clone 7, named as GX_P2V C7, was randomly selected for the evaluation of viral
35
pathogenicity in hACE2 mice (Figure 1A). The hACE2 mouse model expressing
36
human ACE2 under control of the CAG promoter was developed using random
37
integration technology by Beijing SpePharm Biotechnology Company.
38
We initially assessed whether GX_P2V C7 could cause disease in hACE2 mice by
39
monitoring daily weight and clinical symptoms. A total of four 6 to 8-week-old hACE2
40
mice were intranasally infected with a dosage of 5×105 plaque-forming units (pfu) of
41
the virus. Four mice inoculated with inactivated virus and four mock-infected mice
42
were used as controls. Surprisingly, all the mice that were infected with the live virus
43
succumbed to the infection within 7-8 days post-inoculation, rendering a mortality rate
44
.CC-BY-NC-ND 4.0 International licensemade available 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
The copyright holder for this preprintthis version posted January 4, 2024. ; https://doi.org/10.1101/2024.01.03.574008doi: bioRxiv preprint

of 100% (Figure 1B). The mice began to exhibit a decrease in body weight starting from
45
day 5 post-infection, reaching a 10% decrease from the initial weight by day 6 (Figure
46
1C). By the seventh day following infection, the mice displayed symptoms such as
47
piloerection, hunched posture, and sluggish movements, and their eyes turned white.
48
The criteria for clinical scoring of the mice and the daily clinical scores post-infection
49
with GX_P2V C7 are provided in the Supporting Information, Figure S1.
50
We then evaluated the tissue tropism of GX_P2V C7 in hACE2 mice. Using the
51
infection method described above, eight hACE2 mice were infected, eight mice were
52
inoculated with inactivated virus, and another eight mock-infected mice were used as
53
controls. The organs of four randomly selected mice in each group were dissected on
54
days 3 and 6 post-infection for quantitative analysis of viral RNA and titer. We detected
55
significant amounts of viral RNA in the brain, lung, turbinate, eye, and trachea of the
56
GX_P2V C7 infected mice (Figure 1D), whereas no or a low amount of viral RNA was
57
detected in other organs such as the heart, liver, spleen, kidneys, tongue, stomach, and
58
intestines. Specifically, in lung samples, we detected high viral RNA loads on days 3
59
and 6 post-infection, with no significant difference between these two time points (~
60
6.3 versus ~ 5.8 Log10[copies/mg]). In brain samples, on day 3 post-infection, viral
61
RNA was detected in all four infected mice, with an average value of 5.4
62
Log10[copies/mg]. Notably, by day 6 post-infection, we detected exceptionally high
63
viral RNA loads (~ 8.5 Log10[copies/mg]) in the brain samples from all four infected
64
mice (Figure 1D). On days 3 and 6 post-infection, the viral RNA loads in the turbinate
65
were similar, approximately 4.1 and 3.9 Log10[copies/mg], respectively. The viral RNA
66
.CC-BY-NC-ND 4.0 International licensemade available 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
The copyright holder for this preprintthis version posted January 4, 2024. ; https://doi.org/10.1101/2024.01.03.574008doi: bioRxiv preprint

loads in the trachea and eyes of the mice surpassed the limit of detection only on day 6
67
post-infection, with values of 2.6 and 3.8 Log10[copies/mg], respectively. Regarding the
68
infectious viral titers, lung tissues at day 3 post-infection had a value of ~ 1.8
69
Log10[pfu/mg], which decreased to ~ 0.5 Log10[pfu/mg] by day 6. Importantly, the
70
highest infectious titers were detected in the brain on day 6, which was significantly
71
greater than that on day 3 (~ 0.9 vs ~ 4.8 Log10[pfu/mg]) (Figure 1E). Additionally,
72
there were no significant differences in the infectious titers in the turbinate between day
73
3 (~ 0.9 Log10[pfu/mg]) and day 6 (~ 1.2 Log10[pfu/mg]) (Figure 1E). By day 6,
74
approximately 2.0 Log10[pfu/mg] was detected in the eyes of two mice. Neither
75
inactivated GX_P2V C7 nor mock infection caused death or any clinical symptoms in
76
the mice (Figure 1B-C and Supporting Information, Figure S2). In summary, in the mice
77
infected with live virus, the viral load in the lungs significantly decreased by day 6;
78
both the viral RNA loads and viral titers in the brain samples were relatively low on
79
day 3, but substantially increased by day 6. This finding suggested that severe brain
80
infection during the later stages of infection may be the key cause of death in these mice.
81
To determine the mechanisms underlying GX_P2V C7-induced death in hACE2
82
mice, we examined the pathological changes, presence of viral antigens, and cytokine
83
profiles in the lung and brain tissues of the mice on days 3 and 6 post-infection(Figure
84
1F-G, and Supporting Information, Figure S3 and S4). On both days, compared to those
85
of control mice, the lungs of infected mice showed no significant pathological
86
alterations, with only minor inflammatory responses due to slight granulocyte
87
infiltration (Figure 1F). On day 3 post-infection, shrunken neurons were visible in the
88
.CC-BY-NC-ND 4.0 International licensemade available 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
The copyright holder for this preprintthis version posted January 4, 2024. ; https://doi.org/10.1101/2024.01.03.574008doi: bioRxiv preprint

cerebral cortex of the mice. By day 6, in addition to the shrunken neurons, there was
89
focal lymphocytic infiltration around the blood vessels, although no conspicuous
90
inflammatory reaction was observed (Figure 1G). Upon staining for viral nucleocapsid
91
protein via immunohistochemistry, viral antigens were detected in both the lungs and
92
brains on days 3 and 6 post-infection, with extensive viral antigens notably present in
93
the brain on day 6 (Figure 1F-G). These findings align with the viral RNA load
94
assessments in the lung and brain tissues (Figure 1D). We also performed a Luminex
95
cytokine assay to detect 31 cytokines/chemokines in the lung and brain tissues of the
96
mice (Supporting Information, Figure S3 and S4). Consistent with the pathological
97
findings, there were slight increases or decreases in the levels of many
98
cytokines/chemokines in lung and brain tissues compared to those in control tissues,
99
but the levels of key inflammatory factors, such as IFN-γ, IL-6, IL-1β, and TNF-α, did
100
not significantly change. In brief, these analyses revealed that GX_P2V C7 infection in
101
hACE2 mice did not lead to severe inflammatory reactions, a finding that aligns with
102
previous reports by Zhengli Shi’s group using GX_P2V infection in two different
103
hACE2 mouse models (5), as well as our own findings in the golden hamster model (6).
104
To the best of our knowledge, this is the first report showing that a SARS-CoV-2-
105
related pangolin coronavirus can cause 100% mortality in hACE2 mice, suggesting a
106
risk for GX_P2V to spill over into humans. Our findings are evidently inconsistent with
107
those of Zhengli Shi et al. (5), who tested the virulence of GX_P2V in two different
108
hACE2 mouse models. It is important to note that we did not isolate the wild-type
109
GX_P2V strain. The study by Zhengli Shi et al tested the GX_P2V(short_3UTR)
110
.CC-BY-NC-ND 4.0 International licensemade available 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
The copyright holder for this preprintthis version posted January 4, 2024. ; https://doi.org/10.1101/2024.01.03.574008doi: bioRxiv preprint

variant that we reported. However, the adaptative evolutionary changes of this variant
111
during their laboratory culture remain understudied. In fact, according to additional
112
infection experiments, the uncloned GX_P2V(short_3UTR) also resulted in 100%
113
mortality in hACE2 mice. Due to the propensity of coronaviruses to undergo adaptive
114
mutation during passage culture, we cloned and analyzed mutations in
115
GX_P2V(short_3UTR), focusing specifically on the pathogenicity of the cloned strains.
116
The high pathogenicity mechanism of GX_P2V C7 in hACE2 mice, in the absence of
117
the wild-type GX_P2V control, requires further investigation. Compared to the original
118
sequence of GX_P2V(short_3UTR), GX_P2V C7 has two amino acid mutations in the
119
spike protein. Given the close relationship between coronavirus virulence and spike
120
protein mutations (7), it is possible that GX_P2V C7 has undergone a virulence-
121
enhancing mutation. However, it is important to note that our hACE2 mouse model
122
may be relatively unique. The company has not yet published a paper on this hACE2
123
mouse model, but our results suggest that hACE2 may be highly expressed in the mouse
124
brain. Additionally, according to the data provided by the company, these hACE2 mice
125
have abnormal physiology, as indicated by relatively reduced serum triglyceride,
126
cholesterol, and lipase levels, compared to those of wild-type C57BL/6J mice. In
127
summary, our study provides a unique perspective on the pathogenicity of GX_P2V
128
and offers a distinct alternative model for understanding the pathogenic mechanisms of
129
SARS-CoV-2-related coronaviruses.
130
131
Lai Wei1,#, Shuiqing Liu1,#, Shanshan Lu1,#, Shengdong Luo2, Xiaoping An1, Huahao
132
.CC-BY-NC-ND 4.0 International licensemade available 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
The copyright holder for this preprintthis version posted January 4, 2024. ; https://doi.org/10.1101/2024.01.03.574008doi: bioRxiv preprint

Fan1, Weiwei Chen2, Erguang Li3,*, Yigang Tong1,*, Lihua Song1,*
133
1 Beijing Advanced Innovation Center for Soft Matter Science and Engineering,
134
College of Life Science and Technology, Beijing University of Chemical Technology,
135
Beijing, China. 2Research Center for Clinical Medicine, The Fifth Medical Center of
136
PLA General Hospital, Beijing, China. 3State Key Laboratory of Pharmaceutical
137
Biotechnology, Medical School, Nanjing University, China
138
#Contributed equally.
139
*email: erguang@nju.edu.cn; tong.yigang@gmail.com; songlihua@gmail.com
140
141
REFERENCES
142
1. Liu P, Chen W, Chen JP. Viral Metagenomics Revealed Sendai Virus and
143
Coronavirus Infection of Malayan Pangolins (Manis javanica). Viruses. 2019 Oct
144
24;11(11).
145
2. Lam TT, Jia N, Zhang YW, Shum MH, Jiang JF, Zhu HC, et al. Identifying SARS-
146
CoV-2-related coronaviruses in Malayan pangolins. Nature. 2020 Jul;583(7815):282-5.
147
3. Xiao K, Zhai J, Feng Y, Zhou N, Zhang X, Zou JJ, et al. Isolation of SARS-CoV-
148
2-related coronavirus from Malayan pangolins. Nature. 2020 Jul;583(7815):286-9.
149
4. Huang XY, Chen Q, Sun MX, Zhou HY, Ye Q, Chen W, et al. A pangolin-origin
150
SARS-CoV-2-related coronavirus: infectivity, pathogenicity, and cross-protection by
151
preexisting immunity. Cell Discov. 2023 Jun 17;9(1):59.
152
5. Liu MQ, Lin HF, Li J, Chen Y, Luo Y, Zhang W, et al. A SARS-CoV-2-Related
153
Virus from Malayan Pangolin Causes Lung Infection without Severe Disease in Human
154
ACE2-Transgenic Mice. J Virol. 2023 Feb 28;97(2):e0171922.
155
.CC-BY-NC-ND 4.0 International licensemade available 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
The copyright holder for this preprintthis version posted January 4, 2024. ; https://doi.org/10.1101/2024.01.03.574008doi: bioRxiv preprint

6. Lu S, Luo S, Liu C, Li M, An X, Li M, et al. Induction of significant neutralizing
156
antibodies against SARS-CoV-2 by a highly attenuated pangolin coronavirus variant
157
with a 104nt deletion at the 3'-UTR. Emerg Microbes Infect. 2023 Dec;12(1):2151383.
158
7. Roberts A, Lamirande EW, Vogel L, Jackson JP, Paddock CD, Guarner J, et al.
159
Animal models and vaccines for SARS-CoV infection. Virus Res. 2008 Apr;133(1):20-
160
32.
161
162
ACKNOWLEDGEMENTS
163
This work was supported by NSFC-MFST project (China–Mongolia) (grant number
164
32161143027), National Key R&D Program of China (2021YFC2301804) and
165
Biosafety Special Program (No. 19SWAQ 13).
166
ETHICS STATEMENT
167
All animals involved in this study were housed and cared for in an AAALAC
168
(Association for Assessment and Accreditation of Laboratory Animal Care) accredited
169
facilities. The procedure for animal experiments (IACUC-2019-0027) was approved by
170
the Institutional Animal Care and Use Committee of the Fifth Medical Center, General
171
Hospital of the Chinese People's Liberation Army, and complied with IACUC standards.
172
AUTHOR CONTRIBUTIONS
173
L.Song conceived and designed the study and wrote the manuscript. L.W., S.Liu, S.Lu.,
174
and S.Luo. performed the experiments and analyzed the data. X.A., H.F., W.C., E.L.
175
and Y.T. analyzed the data and edited the manuscript. L.W. and L.Song wrote the
176
manuscript and all the authors approved the manuscript.
177
.CC-BY-NC-ND 4.0 International licensemade available 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
The copyright holder for this preprintthis version posted January 4, 2024. ; https://doi.org/10.1101/2024.01.03.574008doi: bioRxiv preprint

CONFLICT OF INTERESTS
178
The authors declare no competing interests.
179
SUPPORTING INFORMATION
180
Additional Supporting Information for this article can be found online at
181
DATA AVAILABILITY
182
All the data supporting the findings of this study are available within the article and
183
the Supporting Information, or from the corresponding author upon reasonable
184
request.
185
ORCID
186
Lihua Song, https://orcid.org/0000-0002-7299-5719
187
.CC-BY-NC-ND 4.0 International licensemade available 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
The copyright holder for this preprintthis version posted January 4, 2024. ; https://doi.org/10.1101/2024.01.03.574008doi: bioRxiv preprint

Figure 1: Characterization of a lethal infection model in human ACE2-transgenic
188
mice caused by the attenuated SARS-CoV-2-related pangolin coronavirus
189
GX_P2V C7. A Mutations in GX_P2V C7 compared to the GX_P2V(short_3UTR)
190
isolate (NCBI accession number: MW532698). The four identical mutations are shown
191
in bold. B Survival of hACE2 transgenic mice following intranasal infection with
192
GX_P2V C7 (n = 4), inactivated GX_P2V C7 (i-C7, n = 4), and mock infection (n = 4).
193
The number of deceased mice on each specific day is annotated on the left of the
194
survival curve. C Percentage of initial weight of hACE2 transgenic mice after intranasal
195
infection with GX_P2V C7 (n = 4), i-C7 (n = 4), and mock infection (n = 4). The
196
statistical significance of the differences between mock-infected (n = 4, blue dots) and
197
GX_P2V C7-infected (n = 4, red dots) or i-C7-infected mice (n = 4, orange dots) at 6
198
or 7 dpi are shown. The error bars represent the means ± SDs. D Quantification of
199
GX_P2V N gene copies in heart, liver, spleen, lung, kidney, tongue, intestine, stomach,
200
trachea, brain, eye, and turbinate homogenates at 3- and 6-day post-infection (dpi) (n =
201
4 per group). The limit of detection (LOD) for viral RNA loads in the original samples
202
was Log10[102 copies/mg]. The error bars represent the means of Log10[copies/mg] ±
203
SDs. The significances of the comparisons in the lung, brain, and turbinate are shown.
204
E Infectious viral titers in lung, brain, eye, and turbinate homogenates were measured
205
by plaque forming assay at 3 and 6 dpi (n = 4 per group). The statistical significance of
206
the differences in the lung, brain, and turbinate are shown. The error bars represent
207
means of Log10[pfu/mL] ± SDs. F, G Hematoxylin and eosin (H&E) staining and
208
immunohistochemical (IHC) staining with an anti–SARS-CoV-2 N-specific antibody
209
.CC-BY-NC-ND 4.0 International licensemade available 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
The copyright holder for this preprintthis version posted January 4, 2024. ; https://doi.org/10.1101/2024.01.03.574008doi: bioRxiv preprint

(SARS-CoV-2) revealed viral antigen–positive cells (brown) in the lung (F) and brain
210
(G), as shown at high magnification in the inset. Scale bars, 500 μm (F) and 1 mm (G),
211
respectively. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, P > 0.05, not
212
significant (ns); two-way ANOVA followed by Sidak’s multiple comparison test.
213
.CC-BY-NC-ND 4.0 International licensemade available 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
The copyright holder for this preprintthis version posted January 4, 2024. ; https://doi.org/10.1101/2024.01.03.574008doi: bioRxiv preprint

.CC-BY-NC-ND 4.0 International licensemade available 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
The copyright holder for this preprintthis version posted January 4, 2024. ; https://doi.org/10.1101/2024.01.03.574008doi: bioRxiv preprint