key: cord-312305-ll29frwc
authors: Sun, Shihui; Gu, Hongjing; Cao, Lei; Chen, Qi; Yang, Guan; Li, Rui-Ting; Fan, Hang; Ye, Qing; Deng, Yong-Qiang; Song, Xiaopeng; Qi, Yini; Li, Min; Lan, Jun; Feng, Rui; Guo, Yan; Qin, Si; Wang, Lei; Zhang, Yi-Fei; Zhou, Chao; Zhao, Lingna; Chen, Yuehong; Shen, Meng; Cui, Yujun; Yang, Xiao; Wang, Xinquan; Wang, Hui; Wang, Xiangxi; Qin, Cheng-Feng
title: Characterization and structural basis of a lethal mouse-adapted SARS-CoV-2
date: 2020-11-11
journal: bioRxiv
DOI: 10.1101/2020.11.10.377333
sha: 
doc_id: 312305
cord_uid: ll29frwc

The ongoing SARS-CoV-2 pandemic has brought an urgent need for animal models to study the pathogenicity of the virus. Herein, we generated and characterized a novel mouse-adapted SARS-CoV-2 strain named MASCp36 that causes acute respiratory symptoms and mortality in standard laboratory mice. Particularly, this model exhibits age and gender related skewed distribution of mortality akin to severe COVID-19, and the 50% lethal dose (LD50) of MASCp36 was ∼100 PFU in aged, male BALB/c mice. Deep sequencing identified three amino acid mutations, N501Y, Q493H, and K417N, subsequently emerged at the receptor binding domain (RBD) of MASCp36, which significantly enhanced the binding affinity to its endogenous receptor, mouse ACE2 (mACE2). Cryo-electron microscopy (cryo-EM) analysis of mACE2 in complex with the RBD of MASCp36 at 3.7-angstrom resolution elucidates molecular basis for the receptor-binding switch driven by amino acid substitutions. Our study not only provides a robust platform for studying the pathogenesis of severe COVID-19 and rapid evaluation of coutermeasures against SARS-CoV-2, but also unveils the molecular mechanism for the rapid adaption and evolution of SARS-CoV-2 in mice. One sentence summary A mouse adapted SARS-CoV-2 strain that harbored three amino acid substitutions in the RBD of S protein showed 100% mortality in aged, male BALB/c mice.

Coronavirus disease 2019 caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has resulted in a public health crisis (1) . The symptoms of COVID-19 are similar to those of SARS-CoV and MERS-CoV infections, ranging from fever, fatigue, dry cough and dyspnea, and mild pneumonia to acute lung injury (ALI) and the acute respiratory distress syndrome (ARDS) in severe cases. In fatal cases, multi-organ failures accompanied by a dysregulated immune response have been observed (2) (3) (4) . Numerous studies have highlighted age and gender related discrepancies in the distribution of COVID-19 cases where the elderly and men tend to have a higher case-fatality ratio when compared to the young and females, suggesting that aged man are more likely to succumb to COVID-19 (5, 6) .

Coronaviridae family, and is an enveloped, single stranded positive-sense RNA virus.

Human angiotensin-converting enzyme 2 (ACE2), has been demonstrated as the functional receptor for SARS-CoV-2 (7, 8) . SARS-CoV-2 cannot infect wild-type laboratory mice due to inefficient interactions between its S protein and the ACE2 receptor of mouse (mACE2). (9) . So, several hACE2 expressing mouse models such as hACE2 transgenic mice (10) , AAV-hACE2 transduced mice (11) and Ad5-hACE2 transduced mice (12) have been developed. Furthermore, mouse adapted strains of SARS-CoV-2 have also been developed via either in vivo passaging or reverse genetics (13) (14) (15) . However, all these models cause only mild to moderate lung damage in mice. A small animal model capable of recapitulating the severe respiratory symptoms and high case fatality ratio of COVID-19 remains to be established.

In this study, we have developed and characterized a new lethal strain of SARS-CoV-2 named MASCp36 by using the in vivo passaging method. Previously, we reported the development of MASCp6, a mouse-adapted SARS-CoV-2 strain using a similar strategy (16) . Remarkably, intranasal injection of MASCp36 caused 100% fatality in aged BALB/c mice. Prior to death, all infected animals developed severe malfunctions of the respiratory system, including acute respiratory distress syndrome 

In our previous study, we generated a mouse-adapted strain of SARS-CoV-2 (MASCp6) by 6 serial passages of a SARS-CoV-2 in the lung of aged BALB/c mice, which caused moderate lung damage and no fatality in mice. Herein, we further serially passaged for additional 30 times to generate a more virulent SARS-CoV-2, and the resulting virus at passage 36 ( named as MASCp36) was used for stock preparation and titration.

To characterize the pathogenicity of MASCp36, groups of BALB/c were subjected to intranasal injection of varying doses of MASCp36. Strikingly, survival curve analysis showed that high doses of MASCp36 caused almost 100% mortality within ten days in all aged mice, while young mice were resistant to MASCp36 challenge and no animals developed disease and died in this group (Fig. 1A-D) . Aged mice challenged with high dose of MASCp36 developed typical respiratory symptoms and exhibited features like ruffled fur, hunched back, and reduced activity. Of particular note, tachypnea was common in all moribund animals (supplemental video). For the aged mice, male animals were more susceptible to MASCp36 in comparison to female ones, and the LD50 was calculated to 58 PFU and 690 PFU, respectively (Fig. 1A, B) . Additionally, this unique gender-dependent mortality was also recorded in aged C57BL/6 mice challenged with MASCp36 ( fig. S1 ) where 80% of the males challenged with the virus died from respiratory symptoms. Thus, serial passaging of MASCp6 generated a more virulent MASCp36, which was sufficient to cause 100% mortality in aged BALB/c mice at low dose (~1200 PFU).

We further characterized the in vivo replication dynamics of MASCp6 in both young and aged mice, and the results from qRT-PCR showed that high levels of SARS-CoV-2 subgenomic RNAs were persistent in the lung and tracheas till 4 day post infection (dpi) in aged mice (Fig. 1E) . Marginal viral RNAs were also detected in the intestine, heart, liver, spleen, brain, and kidney. The young mice had a similar tissue distribution as the aged ones upon MASCp36 challenge, and lung and tracheas represented the major tissues supporting viral replication (Fig. 1F ). Multiplex immuno staining showed that MASCp36 predominantly targeted the airway CC10+ club cells and SPC+ AT2 cells, while FOXJ1+ ciliated cells and PDPN+ alveolar type 1 (AT1) cells detected few positive signals (Fig. 1G ). More SARS-CoV-2-infected cells were detected in the lung from aged mice than those from young animals at 1 dpi (Fig. 1H) , which was in agreement to the high ratio of SPC+ AT2 cells that co-express ACE2 in aged mice ( fig. S2 ). Consequently, a striking loss of SPC+ AT2 cells with apoptosis were observed in the lung from aged mice ( fig. S3 ). Collectively, aged male mice that developed severe respiratory symptoms and 100% fatality serve as the most suitable animal host for MASCp36 and were chosen for subsequent analysis.

To further characterize the pathogical outcome in MASCp36 infected aged male mice, the lung as well as other organs were collected at 4 dpi and subjected to histopathological and immunostaining analysis. When observed by the naked eye, lung tissue samples from MASCp36 infected animals showed visible lung injury characterized with biolateral cardinal red appearance. Furthermore, a lot of sticky secretion was seen at the lung surfaces when compared with the control animals ( Fig.   2A ). According to the metrics of acute lung injury (ALI) laid out by the American Thoracic Society (ATS), MASCp36 infection induced necrotising pneumonia and extensive diffuse alveolar damages (DAD) and even ARDS on day 4 (17) . The microscopic observation showed large quantities of desquamative epithelial cells in bronchiole tubes (yellow arrow) and a large area of necrotic alveoli epithelial cells, fused alveoli walls with inflammatory cells infiltration especially neutrophils in alveolar septae or alveolar airspace, serious edema around vessels (cyan arrow) and scattered hemorrhage (blue arrow) ( Fig. 2A ). In addition, foamy cells, polykaryocytes, fibrin cluster deposition, and hyaline membrane formation were common in the MASCp36 infected animals (Fig. 2B) , indicative of acute respiratory distress syndrome (ARDS), which is well characterized in severe COVID-19 patients (4). Interestingly, typical viral inclusion bodies were also observed occasionally in lungs of infected mice (Fig. 2B) . Comparatively, lung damage in young mice infected with MASCp36 were much milder than those of the aged mice; less visual lung damage were seen in gross necropsy, and only thickened alveolar septa and activated inflammatory cell infiltration (19, 20) . These pathological damages caused by MASCp36 in mice recapitulated most spectrums of seriously ill COVID-19 patients caused by SARS-CoV-2 infection. Additionally, immunostaining of lung sections showed a significant infiltration of CD68 + macrophages and Ly-6G + neutrophils in MASCp36 infected mice ( fig. S7 ).

Interestingly, more CD68 + macrophages and Ly-6G + neutrophils were detected in young mice than that in aged mice 1 dpi, and reversed on 4 dpi, which indicated rapid and short-lived immune response to limite viral replication in young mice.

To test the utility of MASCp36 infected mouse model for evaluation of antiviral candidates, H014, a known human monoclonal antibody targeting the RBD of SARS-CoV-2 (21), was examined for its ability to confer benefit during the infection. 

To deduce the genetic basis for the lethal phenotype of MASCp36, deep sequencing was performed to identify the mutations emerged during the in vivo passaging history. and fig. S8B ). To clarify the potential role of these mutations, the RBD of these different adaptive strains were expressed to assay their binding affinities to mACE2 ( fig. S9 ). As expected, the WT RBD presented no detectable binding, but RBDs from mouse-adapted strains (RBDMASCp6, RBDMASCp25 and RBDMASCp36) gain gradually enhanced binding abilities to mACE2 with affinities ranging from ~500 μM to 2 μM (Fig. 4B ). The increased affinity between mACE2 and the RBD of mouse-adapted strains probably contribute to the enhanced virulence in mice.

To further elucidate the molecular basis for the gradual change in specificity of MASCp36, structural investigations of the mACE2 in complex with RBDMASCp25 or RBDMASCp36 were carried out. Two non-competing Fab fragments that recognize the RBD beyond the mACE2 binding sites were used to increase the molecular weight of this complex for pursuing an atomic resolution by cryo-EM reconstruction (fig. S9-S12). Interestingly, cryo-EM characterization of the mACE2 in complex with RBDMASCp25 revealed that the complex adopts three distinct conformational states, corresponding to tight binding (state 1), loose binding (state 2) and no binding modes (state 3) (fig. S13), indicative of a quick association and quick dissociation interaction manner between the mACE2 and RBDMASCp25. However, only the tight binding conformation was observed in the mACE2-RBDMASCp36 complex structure, reflecting a more stable/mature binding mode for the RBDMASCp36 to mACE2, akin to that of the RBDWT and hACE2. We determined asymmetric cryo-EM reconstructions of the mACE2-RBDMASCp36 complex at 3.7 Å and three states of the mACE2-RBDMASCp25 complex at 4.4 to 8.2 Å (figs. S11-S12 and table S1). The map quality around the mACE2-RBDMASCp36 interface was of sufficient quality for a reliable analysis of the interactions ( Fig. 4C and fig. S9 ).

The overall structure of the mACE2-RBDMASCp36 complex resembles that of the RBDWT-hACE2 complex with a root mean square deviation of 1.0 Å (Fig. 4C ). The RBDMASCp36 recognizes the helices (α1 and α2) located at the apical region of the mACE2 via its receptor binding motif (RBM) (Fig 4C-4E ). The interaction area on the mACE2 could be primarily divided into three patches (PI, PII and PIII), involving extensive hydrophilic and hydrophobic interactions with three regions separately clustered by three adaptation-mediated mutated residues (K417N, corresponding to Clus1; Q493H, corresponding to Clus2; and N501Y, corresponding Clus3) in the RBM ( Fig 4C-4E ). Coincidentally, a number of amino acid substitutions, such as Q493K, Q498Y and P499T, in the RBM identified in other reported mouse-adapted SARS-CoV-2 isolates (15, 22, 23) were included either in the Clus2 or Clus3, underlining the putative determinants for cross-transmission (Fig 4F-4G ). An extra Clus1 is further accumulated in the MASCp36 to gain utmost binding activity and infection efficacy ( Fig 4F-4G ). The extensive hydrophobic interactions in Clus3 constructed by Y501 (or Y498 or H498 in other mouse-adapted SARS-CoV-2 isolates), Y505 in the RBDMASCp36 and Y41, H353 in the mACE2, hydrogen bonds in Clus2 formed H493 (K493 in other mouse-adapted strain) in the RBDMASCp36 and N31, E35 in the mACE2 and hydrophilic contacts constituted by N417 in the RBDMASCp36 and N30, Q34 in the mACE2 contribute to the tight binding of the MASCp36 to mACE2. Contrarily, structural superimposition of the RBDWT over the mACE2-RBDMASCp36 complex reveals the loss of these interactions, leading to the inability of the RBDWT to bind mACE2 (Fig 4G) .

These analysis pinpoints key structure-infectivity correlates, unveiling the molecular basis for adaptation-mediated evolution and cross-transmission of SARS-CoV-2.

Clinically, the severe COVID-19 disease onset might result in death due to massive alveolar damage and progressive respiratory failure (24) (25) (26) . Distinct from all currently reported animal models which mimic the mild to moderate clinical symptoms of COVID-19, the MASCp36 infected mouse model could manifest many of the severe clinical syndromes associated with COVID-19 disease such as pulmonary oedema, fibrin plugs in alveolar, hyaline membrane, and scattered hemorrhage (25, 27) . The As the functional receptor of SARS-CoV and SARS-CoV-2, ACE2, is highly expressed on vascular endothelial cells and smooth muscle cells in multiple organs, which probably leads to the observed viral tropism contributing to cellular injury.

In this model, closely correlated with the higher ratio of ACE2-positive cells in type II pneumocytes in aged mice when compared to those in young mice, massive injury of (AT2 cells) type II pneumocytes was observed in aged mice. Therefore, age-related ACE2 expression pattern in lungs might contribute to the severe phenotype observed in aged mice. Although SARS-CoV-2 viral antigen has been detected in kidney of postmortem specimens (28) , no viral antigen or viral RNA were detected in our model. So in this MASCp36 infected mouse model, the kidney injury may arise due to secondary endothelial injury leading to proteinuria. In addition, although SARS-CoV-2 has also been implicated to have neurotropic potential in COVID-19 (29), we did not find typical characteristics of viral encephalitis in this model. Importantly, the imbalanced immune response with high-levels of proinflammatory cytokines, increased neutrophils and decreased lymphocytes, which were in line with SARS and MERS infections (30), playing a major role in the pathogenesis of COVID-19 (31) , were also observed in this model.

The skewed age distribution of COVID-19 disease was reproduced in the MASCp36 infected mouse model where more severe symptoms were observed in aged mice when compared to young mice. Different from H1N1 pandemic(32), COVID-19 appears to have a mild effect on populations under 30 years, and the elderly are more likely to progress to severe disease and are admitted to intensive care unit (ICU) worldwide (33) . ACE2, the functional receptor of SARS-CoV-2, expressed increasingly in the lungs with age, which might provide an explanation to the higher disease severity observed in older patients with COVID-19 (34) . More importantly, the host immune response may determine the outcome of the disease. Our immune system is composed of innate immunity and adaptive immunity. The innate immunity comprises of the first line of defense against pathogens and is acute as well as short lived. However, aging is linked with insufficient, prolonged and chronic activation of innate immunity associated with low-grade and systemic increases in inflammation (inflamm-aging) which can be detrimental for the body (35) . The delicate co-operation and balance are interrupted by the chronic activation of innate immunity and declined adaptive immune responses with increasing age in COVID-19 (36) . In the MASCp36 infected mouse model, the young mice presented acute inflammatory response with more innate immune cells infiltration on day 1, while lagged and sustained immune response in aged mice. The different immune response in mice model may be vital in limiting virus replication at early times and contribute to different outcome on day 4 in young or aged mice.

In addition to the age-related skewed distribution of COVID-19, gender-related differences in distribution of COVID-19 disease is also recapitulated in this MASCp36 infected mouse model with increased susceptibility and enhanced pathogenicity observed in male mice when compared to their female counterparts. Biological sex is an important determinant of COVID-19 disease severity (37) . In China, the death rate among confirmed cases is 2.8% for women and 4.7% for men (34) . In Italy, half of the confirmed COVID-19 cases are men which account for 65% of all deaths (38) . This pattern is generally consistent around the world. The skewed distribution of COVID-19 suggests that physiological differences between male and female may cause differential response to infection. So the hypothesis that females display reduced susceptibility to viral infections may be due to the stronger immune responses they mount than males (39) . It has been studied that androgens may lower and estrogens may enhance several aspects of host immunity. In addition, androgens facilitate and estrogens suppress lymphocyte apoptosis. Furthermore, genes on the X chromosome important for regulating immune functions, and androgens may suppress the expression of disease resistance genes such as the immunoglobulin superfamily (40) . In the MASCp36 infected mouse model, we found out that it presented higher mortality of the male than the female infected with the same dose of virus, indicating the successful recapitulation of COVID-19 and also its potential application in the study of the pathogenesis of the disease.

Learning from SARS-CoV MA15 with increased virulence in mice, multiple gene products may contribute independently to the virulence. Unlike the SARS-CoV MA15, three subsequent emerged mutations (N501Y, K417N, and Q493H) in the MASCp36 were located in RBD, which breaks a barrier for cross-species transmissions of SARS-CoV-2, enabling gradually adapted recognition of SARS-CoV-2 to mACE2 during the in vivo passages in mice. The serially increased affinities between mACE2 and the RBD of mouse-adapted strains confer to enhanced infections in mice. Cryo-EM structures of the mACE2 in complex with RBDMASCp25 and RBDMASCp36 define preciously the atomic determinants of the receptor-binding switch, providing novel insights into adaptationmediated evolution and cross-transmission of SARS-CoV-2. In addition, there are also 9 amino acid substitutions outside the S protein of MASCp36 (Fig. S8) . At present, we cannot rule out the contribution of these mutations, and further validation with reverse genetic tools will help understand the biological function of each single mutation (41) . 

Figs. S1 to S13 

Center, Beijing Institute of Microbiology and Epidemiology (approval number: IACUC-DWZX-2020-002).

Mouse adapted strain of SARS-CoV-2 (MASCp6) was developed in our previous study (16) . Additional serial passage of 30 times was performed as previously described (16) . Multiplex immunofluorescent assay. The multiplex immunofluorescence assay was conducted as previously described (16) . Briefly, the retrieved sections were incubated with primary antibody for 2 h followed by detection using the HRP-conjugated secondary antibody and TSA-dendronfluorophores (NEON 7-color Allround Discovery Kit for FFPE, Histova Biotechnology, NEFP750).

Afterwards, the primary and secondary antibodies were thoroughly eliminated by heating the slides in retrieval/elution buffer (Abcracker®, Histova Biotechnology, ABCFR5L) for 10 sec at 95°C 

Lung homogenates from MASCp36-infected mice or mock treated mice were processed as previously described (16) and subjected to RNA-Seq.

Total RNA from lung tissue were extracted using TRIzol (Invitrogen, USA) and treated with DNase I (NEB, USA). Sequencing libraries were generated using NEBNext® UltraTM RNA Library Prep Kit for Illumina® (NEB, USA) following the manufacturer's recommendations and index codes were added to attribute sequences to each sample. The clustering of the index-coded samples was performed on a cBot cluster generation system using HiSeq PE Cluster Kit v4-cBot-HS (Illumina) according to the manufacturer's instructions. After cluster generation, the libraries were sequenced on Illumina Novaseq6000 platform and 150 bp paired-end reads were generated.

After sequencing, perl script was used to filter the original data (Raw Data) to clean reads by removing contaminated reads for adapters and low-quality reads. Clean reads were aligned to the mouse genome (Mus_musculus GRCm38.99) using Hisat2 v2.1.0. The number of reads mapped to each gene in each sample was counted by HTSeq v0.6.0 and TPM (Transcripts Per Kilobase of exon model per Million mapped reads) was then calculated to estimate the expression level of genes in each sample. DESeq2 v1.6.3 was used for differential gene expression analysis. Genes with padj≤0.05 and |Log2FC| > 1 were identified as differentially expressed genes (DEGs). DEGs were used as query to search for enriched biological processes (Gene ontology BP) using Metascape.

Heatmaps of gene expression levels were constructed using heatmap package in R (https://cran.rstudio.com/web/packages/pheatmap/index.html). Dot plots and volcano plots were constructed using ggplot2 (https://ggplot2.tidyverse.org/) package in R. Production of Fab fragment. The B8 and D14 Fab fragments (43) were generated using a Pierce FAB preparation Kit (Thermo Scientific). Briefly, the antibody was mixed with immobilized-papain and then digested at 37 ˚C for 3-4 h. The Fab was separated from the Fc fragment and undigested IgGs by protein A affinity column and then concentrated for analysis.

Surface plasmon resonance. mACE2 was immobilized onto a CM5 sensor chip surface using the NHS/EDC method to a level of ~600 response units (RUs) using BIAcore® 3000 (GE Healthcare) and PBS as running buffer (supplemented with 0.05% Tween-20). wtRBD, RBDMASCp25 and RBDMASCp36, which were purified and diluted, were injected in concentration from high to low. The binding responses were measured, and chip surfaces were regenerated with 10 mM Glycine, pH 1.5 (GE Healthcare). The apparent binding affinity (KD) for individual antibody was calculated using BIAcore® 3000 Evaluation Software (GE Healthcare).

For the competitive binding assays, the first sample flew over the chip at a rate of 20 ul/min for 120 s, after which the second sample was injected at the same rate for another 120s. All antibodies were evaluated at a saturation concentration of 500 nM, while mACE2 was applied at 1000 nM concentration. All surfaces of chips were regenerated with 10 mM Glycine, pH 1.5 (GE Healthcare).

The response units were recorded at room temperature and analyzed using the same software as mentioned above.

For Cryo-EM sample preparation, the quaternary complex (RBDMASCp25/RBDMASCp36-FabB8-FabD14-mACE2) was diluted to 0.8 mg/ml.

Holy-carbon gold grid (Quantifoil R0.6/1.0 mesh 300) were freshly glow-discharged with a Solarus 950 plasma cleaner (Gatan) for 60s. A 3 μL aliquot of the mixture complex was transferred onto the grids, blotted with filter paper at 22℃ and 100% humidity, and plunged into the ethane using a Vitrobot Mark IV (FEI). For RBDMASCp25/RBDMASCp36-FabB8-FabD14-mACE2 complex, micrographs were collected at 300 kV using a Titan Krios microscope (Thermo Fisher), equipped with a K2 detector (Gatan, Pleasanton, CA), using SerialEM automated data collection software (43) Movies (32 frames, each 0.2 s, total dose 60 e − Å −2 ) were recorded at final pixel size of 1.04 Å with a defocus of between -1.25 and -2.7 μm.

Image processing. For RBDMASCp25-FabB8-FabD14-mACE2 complex, a total of 2,109 micrographs were recorded. For RBDMASCp36-FabB8-FabD14-mACE2 complex, a total of 2,982 micrographs were recorded. Both sets of the data were processed in the same way. Firstly, the raw data were processed by MotionCor2, which were aligned and averaged into motion-corrected summed images. Then, the defocus value for each micrograph was determined using Gctf. Next, particles were picked and extracted for two-dimensional alignment. The partial well-defined particles were selected for initial model reconstruction in Relion. The initial model was used as a reference for three-dimensional classification. After the refinement and postprocessing, the overall resolution of RBDMASCp36-FabB8-FabD14-mACE2 complex was up to 3.69Å, on the basis of the gold-standard Fourier shell correlation (threshold = 0.143) (44) . For RBDMASCp25-FabB8-FabD14-mACE2 complex, the Class I complex the resolution achieved was 7.89 Å, ClassII complex had a resolution of 8.17 Å, while Class III complex was reconstructed to a resolution of 4.4 Å. The quality of the local resolution was evaluated by ResMap (45) .

Model building and refinement. The wtRBD-hACE2 (PDB ID: 6M0J) structures were manually docked into the refined maps of RBDMASCp36-FabB8-FabD14-mACE2 complex using UCSF Chimera (46) and further corrected manually by real-space refinement in COOT (47) . The atomic models were further refined by positional and B-factor refinement in real space using Phenix (48) .

Validation of the final model was performed with Molprobity (49) . The data sets and refinement statistics are shown in table S1.

Statistical analyses were carried out using Prism software (GraphPad). All data are presented as means ± standard error of the means (SEM). Statistical significance among different groups was calculated using the Student's t test, Fisher's Exact test, Two-way ANOVA or Mann-Whitney test *, **, and *** indicate P < 0.05, P < 0.01, and P < 0.001, respectively. 

Dying with SARS-CoV-2 infection-an autopsy study of the first consecutive

Autopsy in suspected COVID-19 cases

The pathological autopsy of coronavirus disease 2019 (COVID-2019) in China: a review. Pathogens and disease

Considering how biological sex impacts immune responses and COVID-19 outcomes

Clinical features of COVID-19 in elderly patients: A comparison with young and middle-aged patients

Structure of the SARS-CoV-2 spike receptor-binding domain bound to the ACE2 receptor

Characteristics of SARS-CoV-2 and COVID-19

Comparative analysis reveals the species-specific genetic determinants of ACE2 required for SARS-CoV-2 entry

The pathogenicity of SARS-CoV-2 in hACE2 transgenic mice

Mouse model of SARS-CoV-2 reveals inflammatory role of type I interferon signaling. bioRxiv

A SARS-CoV-2 Infection Model in Mice Demonstrates Protection by Neutralizing Antibodies

A mouse-adapted model of SARS-CoV-2 to test COVID-19 countermeasures

Adaptation of SARS-CoV-2 in BALB/c mice for testing vaccine efficacy

A Mouse-Adapted SARS-CoV-2 Induces Acute Lung Injury and Mortality in Standard Laboratory Mice

Adaptation of SARS-CoV-2 in BALB/c mice for testing vaccine efficacy

An official American Thoracic Society workshop report: features and measurements of experimental acute lung injury in animals

COVID-19 Makes B Cells Forget, but T Cells Remember

The Novel Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) Directly Decimates Human Spleens and Lymph Nodes. medRxiv

Human Kidney is a Target for Novel Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) Infection. medRxiv

Structural basis for neutralization of SARS-CoV-2 and SARS-CoV by a potent therapeutic antibody

Mouse-adapted SARS-CoV-2 replicates efficiently in the upper and lower respiratory tract of BALB/c and C57BL/6J mice

A mouse-adapted model of SARS-CoV-2 to test COVID-19 countermeasures

Histopathology and ultrastructural findings of fatal COVID-19 infections in Washington State: a case series

Pathological study of the 2019 novel coronavirus disease (COVID-19) through postmortem core biopsies. Modern pathology : an official journal of the United States and Canadian Academy of Pathology

Clinical Features of COVID-19 and Factors Associated with Severe Clinical Course: A Systematic Review and Meta-Analysis. Ssrn

Pathological findings of COVID-19 associated with acute respiratory distress syndrome. The Lancet

Identification of a potential mechanism of acute kidney injury during the COVID-19 outbreak: a study based on singlecell transcriptome analysis

Neurological associations of COVID-19

Lung pathology of fatal severe acute respiratory syndrome

A new coronavirus associated with human respiratory disease in China

Pandemic preparedness and response--lessons from the H1N1 influenza of

Report of the WHO-China Joint Mission on Coronavirus Disease 2019 (COVID-19)

SARS-CoV-2: virus dynamics and host response

Chronic inflammation in the etiology of disease across the life span

Disparities in Age-specific Morbidity and Mortality From SARS-CoV-2 in China and the Republic of Korea

Characteristics of COVID-19 patients dying in Italy Report based on available data on

The lethal sex gap: COVID-19

Sex differences in immune responses

Spike mutation D614G alters SARS-CoV-2 fitness

Acknowledgements: This work was in memory of Prof

This work was supported by the National Key Plan for Scientific Research and Development of China

Quantifying the local resolution of cryo-EM density maps

visualization system for exploratory research and analysis

Tools for macromolecular model building and refinement into electron cryo-microscopy reconstructions

Towards automated crystallographic structure refinement with phenix.refine

MolProbity: all-atom structure validation for macromolecular crystallography

RELION: implementation of a Bayesian approach to cryo-EM structure determination

Automated electron microscope tomography using robust prediction of specimen movements

ChAdOx1 nCoV-19 vaccine prevents SARS-CoV-2 pneumonia in rhesus macaques

Data and materials availability: All requests for resources and reagents should be directed to C.-F.Q. (qincf@bmi.ac.cn and qinlab313@163.com) and will be fulfilled after completion of a materials transfer agreement.