A Human-Immune-System (HIS) humanized mouse model (DRAGA: HLA-A2. HLA-DR4. Rag1 KO.IL-2Rγc KO. NOD) for COVID-19

We report the first Human Immune System (HIS)-humanized mouse model (“DRAGA”: HLA-A2.HLA-DR4.Rag1KO.IL-2RγcKO.NOD) for COVID-19 research. This mouse is reconstituted with human cord blood-derived, HLA-matched hematopoietic stem cells. It engrafts human epi/endothelial cells expressing the human ACE2 receptor for SARS-CoV-2 and TMPRSS2 serine protease co-localized on lung epithelia. HIS-DRAGA mice sustained SARS-CoV-2 infection, showing deteriorated clinical condition, replicating virus in the lungs, and human-like lung immunopathology including T-cell infiltrates, microthrombi and pulmonary sequelae. Among T-cell infiltrates, lung-resident (CD103+) CD8+ T cells were sequestered in epithelial (CD326+) lung niches and secreted granzyme B and perforin, indicating cytotoxic potential. Infected mice also developed antibodies against the SARS-CoV-2 viral proteins. Hence, HIS-DRAGA mice showed unique advantages as a surrogate in vivo human model for studying SARS-CoV-2 immunopathology and for testing the safety and efficacy of candidate vaccines and therapeutics.

(which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprint this version posted January 29, 2021. ; https://doi.org/10.1101/2020.08.19.251249 doi: bioRxiv preprint Figure S1. Quantification of hACE2 protein in HIS-DRAGA lungs. Human ACE2 levels in immunoprecipitates obtained from non-infected HIS-DRAGA and human lung lysates using S1(RBD)-mFc 2a protein + rat anti-mouse IgG2a-magnetic beads were quantified by ELISA. Of note, the OD450nm values for protein immunoprecipitated from a pool of 10 non-infected, non-HIShumanized DRAGA mouse lung lysates (negative control) fell below the limit of detection (OD450nm <0.05). Insert shows Western blot detection of hACE2 protein in the concentrated immunoprecipitates probed with a mouse monoclonal anti-human ACE2 IgG followed by goat antimouse IgG-HRP with ECL detection. Lane 1, human lung immunoprecipitate; lane 2, HIS-DRAGA mouse lungs immunoprecipitate; lane 3, DRAGA mouse lungs immunoprecipitate (note this sample did not contain detectable hACE2). Lower panel shows the experimental conditions for immunoprecipitation of hACE2, quantification by ELISA, and the ratio of hACE2 in human versus HIS-DRAGA mouse lung samples. Figure S2. Co-localization of hACE2 receptor with hCD326 + alveolar ECs in a non-infected HIS-DRAGA mouse. a,b. Co-localization of hACE2 with alveolar hCD326 + ECs (orange) revealed by costaining of S1(RBD)-mFc 2a protein + goat anti-mouse IgG-FITC (green) and anti-hCD326-PE (red) in representative HIS-DRAGA female and male mice. c. Lack of hCD326 + ECs and negligible binding of S1(RBD)-mFc 2a protein to a representative lung section from a non-HIS-reconstituted DRAGA female mouse. Figure S3. Binding of SARS-CoV-2 S1(RBD) protein to the endothelium of liver cholangiocytes in non-infected HIS-DRAGA mice. S1(RBD) binding to the liver cholangiocytes from representative non-infected HIS-DRAGA female (panel a) and male (panel b) mice. Merged images and enlargements show binding of S1(RBD)-mFc 2a revealed by a goat anti-mouse IgG-FITC conjugate (green) and nuclei (DAPI, blue). Lower panels, representative images showing minimal background binding of the goat anti-mouse IgG-FITC secondary antibody (green) and nuclei (DAPI, blue) in tissues from the same mice in panels a and b.
105 and is also made available for use under a CC0 license.

Figure S5. Binding of SARS-CoV-2 S1(RBD) protein to intestinal epithelia of infected HIS-DRAGA mice.
Tissue sections from the small intestine of HIS-DRAGA mouse #F1 (upper and middle panels) and #F2 (lower panel) survivors of SARS-CoV-2 infection with 2.8x10 4 pfu and 2.8x10 3 pfu, respectively, at the experimental endpoint (14 dpi). Sections were co-stained with DAPI (nuclei, blue) and S1(RBD)-mFc 2a protein+ goat anti-mouse IgG-FITC (green). Shown is the S1(RBD) protein bound to the columnar epithelial cells (cEC) of the absorptive intestinal villi (V). Figure S6. Binding of SARS-CoV-2 S1(RBD) protein to the brain epithelia in infected HIS-DRAGA mice. Tissue sections of cerebellum cortex from HIS-DRAGA mice #F1 (upper panel) and #F2 (lower panel) survivors of SARS-CoV-2 infection with 2.8x10 4 pfu and 2.8x10 3 pfu, respectively, at the experimental endpoint (14 dpi) co-stained with DAPI (nuclei, blue) and S1(RBD)-mFc 2a protein + goat anti-mouse IgG-FITC (green). Shown is the S1(RBD) protein bound to the white matter (WM), granular layer (GL), and Purkinje cells (PC), but not to the outer neuronal layer in the grey matter (GM). Figure S7. Identification of hACE2 on tissue sections from organs of a non-infected HIS-DRAGA mouse, detected with an anti-hACE2 specific antibody. a. Merged images of tissue sections from a representative non-infected HIS-DRAGA mouse stained with mouse anti-hACE2 followed by goat anti-mouse IgG-FITC (green) and DAPI (blue). b. Minimal background binding of the secondary antibody (goat anti-mouse IgG-FITC) (left) overlapped with DAPI staining (right) of tissue sections adjacent to those shown in panel a.

Figure S8. Lung pathology of a HIS-DRAGA mouse recovering from SARS-CoV-2 infection.
Representative H&E-stained lung section from HIS-DRAGA mouse #F2 that recovered its initial body weight at 9 days after infection with SARS-CoV-2 at 2.8x10 3 pfu. Interstitial and intra-alveolar infiltrates are indicated by arrows. Figure S9. CD61 + intra-alveolar microthrombi in a SARS-CoV-2 infected HIS DRAGA mouse. Intraalveolar microthrombi in lung section from a HIS-DRAGA mouse infected with 2.8x10 4 pfu of SARS-CoV-2, which had not recovered its initial body weight by 14 dpi. Left panel, staining with anti-CD61-PE (red). Right panel with enlargement: merged anti-CD61-PE co-staining with DAPI (blue). The non-nucleated CD61 + cluster indicates this is a platelet micro-thrombus.

Figure S10. Large intra-alveolar thrombi in a SARS-CoV-2 infected HIS-DRAGA mouse. a.
Representative lung section from HIS-DRAGA mouse #F1 infected with 2.8x10 4 pfu, which had not recovered its initial body weight by 14 dpi. Co-staining with anti-CD61-PE (red) + DAPI (blue). b. Representative lung section from a non-infected HIS-DRAGA mouse stained as in panel a, with no evidence of intra-alveolar thrombi. Figure S11. Pulmonary sequelae in SARS-CoV-2 infected DRAGA mice. a. Masson's Trichrome staining of lung sections from HIS-DRAGA mice #F3 (upper panel) and #F4 (lower panel) infected with 10 3 pfu SARS-CoV-2, which recovered their initial body weights by 9 and 25 dpi, respectively. Shown are small peri-alveolar infiltrates building collagen fibers (blue, arrows). b. Masson's Trichrome staining of lung sections from HIS-DRAGA mice #F5 (upper panel) and #F6 (lower panel) infected with 10 3 pfu SARS-CoV-2, which had not recovered their initial body weights by 25 dpi. Shown are peri-alveolar and intra-alveolar infiltrated areas building collagen fibers (blue). Figure S12. Antibody serum titers to SARS-CoV-2 viral proteins. Titers of hIgM and hIgG antibodies to S1(RBD) protein (a) S-trimer protein (b) and N protein (c) in sera (diluted 1:20) from 8 HIS-DRAGA mice infected with SARS-CoV-2 (10 3 pfu), as measured by ELISA at 25 dpi. An anti-S1(RBD) antibody provided in the kit (Bethyl Laboratories) was included as a positive control in panels a and b (C+). The antibody titers against the N protein in serum from a non-infected mouse served as a negative control in panel c. OD450nm values were corrected by subtracting values (ranging from 0.045-0.067) of serum samples from the same mice prior to infection. Standard deviations (+/-SD) for each serum sample in duplicate wells were determined at 99% interval of confidence by SigmaPlot v.14 software.