A humanized mouse model of chronic COVID-19

Sefik, Esen; Israelow, Benjamin; Mirza, Haris; Zhao, Jun; Qu, Rihao; Kaffe, Eleanna; Song, Eric; Halene, Stephanie; Meffre, Eric; Kluger, Yuval; Nussenzweig, Michel; Wilen, Craig B.; Iwasaki, Akiko; Flavell, Richard A.

doi:10.1038/s41587-021-01155-4

Download PDF

Article
Published: 17 December 2021

A humanized mouse model of chronic COVID-19

Nature Biotechnology volume 40, pages 906–920 (2022)Cite this article

36k Accesses
63 Citations
520 Altmetric
Metrics details

Subjects

Abstract

Coronavirus disease 2019 (COVID-19) is an infectious disease that can present as an uncontrolled, hyperactive immune response, causing severe immunological injury. Existing rodent models do not recapitulate the sustained immunopathology of patients with severe disease. Here we describe a humanized mouse model of COVID-19 that uses adeno-associated virus to deliver human ACE2 to the lungs of humanized MISTRG6 mice. This model recapitulates innate and adaptive human immune responses to severe acute respiratory syndrome coronavirus 2 infection up to 28 days after infection, with key features of chronic COVID-19, including weight loss, persistent viral RNA, lung pathology with fibrosis, a human inflammatory macrophage response, a persistent interferon-stimulated gene signature and T cell lymphopenia. We used this model to study two therapeutics on immunopathology, patient-derived antibodies and steroids and found that the same inflammatory macrophages crucial to containing early infection later drove immunopathology. This model will enable evaluation of COVID-19 disease mechanisms and treatments.

Genetically diverse mouse models of SARS-CoV-2 infection reproduce clinical variation in type I interferon and cytokine responses in COVID-19

Article Open access 25 July 2023

Genetically modified mouse models to help fight COVID-19

Article 26 October 2020

SARS-CoV-2 immunity in animal models

Article Open access 18 January 2024

Main

COVID-19, caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)¹, is a heterogenous disease with few therapeutic options. Although anti-viral immunity mediates viral clearance in mild COVID-19, robust inflammatory cytokines, decreased circulating lymphocytes and dysregulated myeloid and lymphocyte compartments characterize immunopathology in severe COVID-19 (refs. ^{2,3,4,5,6,7,8,9,10,11,12,13}). Accurate model systems are essential to evaluate promising discoveries, but most available rodent and non-human primate models do not reveal the immunopathology seen in patients and are not well suited to test therapeutics in the context of severe COVID-19 or post-acute sequelae of COVID-19 (refs. ^{14,15,16,17,18,19}).

ACE2 is a SARS-CoV-2 receptor^20,21,22. SARS-CoV-2 does not infect standard laboratory mice owing to differences between mouse and human ACE2 (hACE2) that limit viral entry^{14,15,20,21,22}. Mice with transgenic or transient hACE2 expression can be infected with SARS-CoV-2 (refs. ^{14,16,17,18,19}). Although acute viral response, transmission and vaccine efficacy can be tested in these models, they lack severe, chronic disease (Extended Data Fig. 1). We hypothesized that a functional human immune system would model innate and adaptive human immunity during SARS-CoV-2 infection and confer chronicity and pathology seen in patients. Mice with a human immune system (humanized mice), generated via transplantation of human hematopoietic stem and progenitor cells (HSPCs), are invaluable tools to study the human immune system in vivo^23,24. MISTRG6 (ref. ²⁵) mice were engineered by a human/mouse homolog gene replacement strategy to provide physiological expression with regard to quantity, location and time of M-CSF (monocytes and tissue macrophage development)²⁶, GM-CSF/IL3 (lung alveolar macrophages)²⁷, SIRPα (macrophage tolerance to human cells)²⁸, ThPO (hematopoiesis and platelets)²⁹ and IL6 (improved engraftment and antibody responses)^25,30,31 in a Rag2/γ common chain deleted background. When engrafted with human HSPCs, these mice have a comprehensive immune system similar to humans^25,32. Delivering hACE2 via an adeno-associated virus (AAV) vector to the lungs allows SARS-CoV-2 infection of HSPC-engrafted MISTRG6 mice. Thus, we created a humanized mouse model of COVID-19 that recapitulates the human innate and adaptive immune systems that is amenable to the mechanistic study of COVID-19 and its myriad of complications.

Results

MISTRG6 humanized mice that transiently express hACE2 can be infected with SARS-CoV-2

We delivered AAV-hACE2 (ref. ¹⁷) to the lungs of immune-reconstituted MISTRG6 mice (MISTRG6-hACE2) that were engrafted with human HSPCs (Fig. 1a). MISTRG6-hACE2 mice were then infected with SARS-CoV-2, which yielded similar viral RNA levels and viral titers to wild-type mice¹⁷ but with prolonged high viral titers and RNA for at least 35 days post-infection (dpi) (Fig. 1b,c, and Supplementary Fig. 1a,b). In contrast, MISTRG6 mice lacking AAV-hACE2 expression did not have detectable viral titers, confirming the necessity of hACE2 for infection (Fig. 1c). Of note, MISTRG6-hACE2 mice that were infected, but not engrafted, with human HSPCs (and thus lack human immune cells) had lower viral titers than infected mice with human immune cells (Fig. 1c). The presence of human immune cells in MISTRG6-hACE2 mice caused persistent disease with severe, chronic weight loss for at least 35 dpi (Fig. 1d and Supplementary Fig. 1c). MISTRG6-hACE2 humanized mice also exhibited more severe lung pathology compared to other animal models^{17,18,33,34,35}. This lung pathology followed the three stages described in human patients with COVID-19: exudative, organizing and fibrotic phases^{4,36,37,38,39}. At 2 dpi, MISTRG6-hACE2 mice showed prominent collection of proteinaceous debris in the alveolar spaces, a hallmark of the exudative phase (Fig. 1e and Supplementary Fig. 1d). Lungs at 4 dpi transitioned into organizing phase, highlighted by infiltration of alveolar spaces by lymphocytes, macrophages and fibroblasts (Fig. 1e and Supplementary Fig. 1d). Between 14 dpi and 28 dpi, expanding infiltration of inter-alveolar spaces (Fig. 1e and Supplementary Fig. 1d–f) and increasing deposition of collagen, highlighted by thickening of collagen bundles in the lung parenchyma (Fig. 1f and Supplementary Fig. 1d–f), further obliterated alveolar architecture. These changes highlighted the transition from the organizing to the fibrotic phase. The architectural remodeling and collagen deposition was more pronounced at 28 dpi and 35 dpi, resulting in replacement of normal alveolar architecture by dilated cystic spaces, separated by thick interstitial tissue bands (Fig. 1e,f and Supplementary Fig. 1d–f). Thus, these data suggest that humanized MISTRG6-hACE2 mice model chronic disease and recapitulate the severe, persistent lung pathology seen in patients with severe COVID19.

Inflammatory macrophages and monocytes characterize the immune landscape in SARS-CoV-2-infected MISTRG6-hACE2 mice

We next characterized human and mouse immune cells in SARS-CoV-2-infected and uninfected MISTRG6-hACE2 mice by flow cytometry to evaluate the cells that drive immunopathology and viral clearance (Supplementary Fig. 2a,b). The degree of humanization was similar in blood between uninfected and infected mice, but SARS-CoV-2 infection substantially increased recruitment of human immune cells to the lower respiratory tract and lung parenchyma, as quantified in whole lung homogenates and bronchioalveolar lavage (BAL) (Fig. 2a,b). The infiltrates present in lungs and BAL at 2–4 dpi consisted of human monocytes, macrophages and T cells (Fig. 2c), again reminiscent of human COVID-19 lung pathology⁷. In our model, human neutrophils are lacking, so, instead, we measured mouse neutrophils, which are the most abundant mouse immune cells in these mice (Supplementary Fig. 2b). In contrast to persistent human immune cell infiltration, mouse immune cell and neutrophil infiltration was transient, peaking at 4 dpi and returning to baseline by 14 dpi (Supplementary Fig. 2b). Variability in neutrophils is in line with human histopathological findings in lungs⁴⁰. However, the cause of such variability and whether this is a result of superimposed bacterial infections are not established^41,42.

**Fig. 2: Immune landscape in *MISTRG6-hACE2* mice infected with SARS-CoV-2 is characterized by inflammatory macrophages and monocytes.**

The lung monocytes of infected mice comprised all three human monocyte subsets (CD14⁺ classical, CD14⁺CD16⁺ intermediate and CD16⁺ non-classical; Fig. 2c). By contrast, and in line with healthy human lung monocytes⁴³, uninfected humanized mouse lungs harbored only classical CD14⁺ monocytes, which expanded in response to SARS-CoV-2 infection (Fig. 2c). Intermediate and non-classical monocytes infiltrated infected lungs at high frequencies as early as 2 dpi, peaking at 4 dpi. Macrophage numbers increased at 4 dpi and peaked at 14 dpi, remaining high until 28 dpi (Fig. 2c,d and Supplementary Fig. 2c,d). By 2 dpi, the macrophage compartment was enriched for inflammatory and monocyte-derived macrophages, which outnumbered alveolar macrophages, suggesting that macrophages recruited early from the circulation are the long-term contributors of immunopathology (Fig. 2e–g). Plasmacytoid dendritic cells (pDCs), generally known for their contribution to early anti-viral response and main producers of type I interferon alpha (IFNα), were enriched substantially later in infection (14 dpi in Fig. 2h). Infected mice at 14 dpi also showed activated pDCs (Fig. 2i and Supplementary Fig. 2e) marked by CCR7 and CD83 expression⁴⁴. This coincided with chronically sustained interferon response and sustained inflammatory macrophages, as seen in patients with severe COVID-19 (refs. ^7,8,45,46).

Adaptive immune responses of patients with COVID-19 are recapitulated in MISTRG6-hACE2 mice infected with SARS-CoV-2

Human COVID-19 shows profound T cell lymphopenia^10,12,18,47 that is recapitulated in our humanized mice. Infected MISTRG6-hACE2 humanized mice presented with a profound loss of T cells, especially CD8⁺ T cells, in blood and spleens (Fig. 3a,b and Extended Data Fig. 2a–c). Moreover, lung T cells of MISTRG6-hACE2 mice were activated (positive for HLA-DR, CXCR3, ICOS and PD1) and the main producers of IFN-gamma (IFNγ) (Figs. 3c,d and 4b,d), as in patients with COVID-19 (refs. ^10,11,12). T cell populations comprised both TCRalpha/beta T cells (Tαβ), enriched for CD4⁺, and TCRgamma/delta T cells (Tγδ) in both uninfected and infected mice (Fig. 3e and Extended Data Fig. 2d). However, during the course of infection, as the circulating T cell numbers decline, the total number of T cells in lungs increased with higher representation of Tγδ cells compared to uninfected mice (Fig. 3e). The early T cell response in infected lungs showed an increase in resident and infiltrating Tγδ cells and bystander-activated memory Tαβ cells as determined by the lung transcriptional profile (Fig. 4b,c and Extended Data Fig. 4h).

**Fig. 3: Adaptive immune responses in *MISTRG6-hACE2* mice infected with SARS-CoV-2 are characterized by virus-specific activated T cells, IgG⁺ B cells and systemic lymphopenia.**

As the human cells in these mice develop within a mouse thymus, we sought to determine whether T cells were responsive to virus in the context of major histocompatibility complex developed in our system. To test this, we stained human CD8⁺ T cells from infected mice that had been engrafted with an HLA-A2-positive source of HSPCs with viral antigen-bearing HLA-A2 tetramers (Extended Data Fig. 2e). Virus-specific CD8⁺Tαβ cells against membrane glycoprotein (M) and spike (S) peptides were detectable as early as 4 dpi, peaking at 28 dpi (Fig. 3f). To test if these T cells could attenuate infection, we transferred sorted lung T cells and total splenocytes from infected mice to unengrafted MISTRG6-hACE2 mice before infection with SARS-CoV-2. Viral titers at 4 dpi in mice that received prophylactic lung T cells were clearly lower than mice that did not receive human T cells (Extended Data Fig. 2f). Thus, a functional anti-SARS-CoV-2 T cell response develops in infected MISTRG6-hACE2 mice.

B cells gradually increased in response to infection, peaking late at 14–28 dpi in BAL and lungs (Fig. 3g and Extended Data Fig. 2g). Although the germinal center B cell response was reported to be suboptimal in reconstituted MISTRG6 mice³⁰, a high proportion of IgM⁺ B cells were seen early in infection that were subsequently replaced by IgG⁺ B cells (14 dpi and 28 dpi) and thus defined the anti-viral B cell response (Fig. 3h,i and Extended Data Fig. 2h). B cells, particularly late in infection, expressed high levels of CD11c (Fig. 3j and Extended Data Fig. 2I), implicating an inflammatory, systemic lupus erythematosus (SLE)-like extrafollicular B cell response^48,49 in humanized COVID-19, reminiscent of the human disease⁵⁰. In contrast to T cell lymphopenia, B cell numbers in the humanized mice were normal or even increased (Fig. 3g and Extended Data Fig. 2j), as in most patients⁵¹. A robust memory B cell expansion is detected early in human infection, with serum IgM and IgA antibodies detected earlier than IgG²⁶. Initial serum IgM and IgA titers in human SARS-CoV-2 infection decline (~28 d) as IgG titers peak (~49 d)⁵². We observed a similar dynamic for IgM⁺ and IgG⁺ B cells in humanized lungs, with IgM⁺ B cells declining over time and IgG⁺ B cells emerging at 14 dpi and reaching high levels at 28 dpi, at which point serum IgG levels also peak (Fig. 3h,i and Extended Data Fig. 2h,k). Thus, MISTRG6-hACE2 humanized mice recapitulate T and B cell responses induced by SARS-CoV-2 infection in patients.

SARS-CoV-2 lungs sustain ISGs and display SLE-like features

We evaluated the transcriptional landscape in uninfected and SARS-CoV-2 infected lungs of humanized mice at 2, 4, 7, 14 and 28 dpi. Mapping of transcripts to the human genome or the mouse genome separately identified 285 human genes and 516 mouse genes that were over-represented in infected lungs (Fig. 4a and Supplementary Table 1). This approach served as a proxy for identifying changes in human versus mouse immune cells. Although there was marked heterogeneity in the strength of the response, the anti-viral response was sustained throughout the course of infection, suggesting that early anti-viral responses were either maintained or amplified late in infection (Fig. 4a). Corresponding pathway analysis of these differentially expressed mouse genes (mouse DEGs) identified cellular responses to interferons, cytokine production, ribonuclease activity and neutrophil activation as top biological processes induced during SARS-CoV-2 infection (Supplementary Table 2). Human DEGs were enriched for extracellular matrix assembly, opsonization and complement activation with a focus on immune phenotypes in monocytes, activated T cells and B cells; this further corroborated our findings from flow cytometric analysis (Supplementary Table 2). Similar circuitries of monocytes, macrophages, activated T cells and extrafollicular B cells were described in humans with SARS-CoV-2 pneumonia^50,53.

We performed single-cell RNA sequencing to better evaluate the phenotype of human and mouse cells in lungs of infected MISTRG6-hACE2 mice. Our transcriptome analysis of human cells at the single-cell level also revealed that human macrophages, monocytes and human T cells were abundant at 4 dpi (Fig. 4b,c). Human monocytes infiltrating infected lungs were already differentiating into macrophages at this stage of infection. These infiltrating monocytes and macrophages (clusters 0, 1, 2 and 3) were the main producers of inflammatory cytokines IL1α, IL1β, TNF and IL6, as in patients with severe COVID-19 (refs. ^6,9,10,11,13) (Fig. 4b,d). As in recent studies characterizing human COVID-19 (ref. ⁵³), alveolar macrophages in infected MISTRG6-hACE2 mice were the main producers of T cell chemoattractants, such as CXCL10 (Fig. 4d). We also monitored mouse immune cells, predominantly enriched for neutrophils (clusters 1 and 2), few mouse monocytes (clusters 5 and 6) and macrophages (cluster 3) at 4 dpi (Extended Data Fig. 3a,b). Mouse neutrophils, which peak at 4 dpi (Supplementary Fig. 1B), were the main producers of mouse inflammatory cytokines and chemokines (Il1a, Il1b, TNF, Il18 and Cxcl10; Extended Data Fig. 3c).

Notably, most of the human and mouse DEGs in lungs were type I or type II interferon-stimulated genes (ISGs) (Fig. 4e and Supplementary Tables 1–3). ISGs were sustained at high levels throughout infection, persisting even as late as 14 dpi and 28 dpi, recapitulating the interferon-dependent phenotypes identified in patients with COVID-19 (Fig. 4a,e and Supplementary Tables 2 and 3)^13,45,46,54. Despite their fundamental role in many infectious and inflammatory diseases, the direct quantitation of type I interferons has been challenging in health and disease⁵⁵. Despite a significant ISG signature, minimal amounts of type I interferons have been detected in the blood or lungs of patients with severe COVID-19 (refs. ^56,57,58,59). This contrasts with those seen in patients infected with highly pathogenic influenza viruses⁶⁰. We first quantified type I interferons by qPCR. Of the many IFNA gene transcripts we tested (IFNA1, IFNA2, IFNA6, IFNA8, IFNA14 and IFNA21), only IFNA1 and IFNA6 mRNA levels were reliably detected and increased upon infection with SARS-CoV-2, peaking at 14 dpi and coincident with pDC numbers in the lung tissue (Fig. 4f). We observed a similar, but not significant, trend for IFNB levels (Extended Data Fig. 3d). Next, we quantified IFNα levels by ELISA in infected and uninfected mice. IFNα levels in serum and BAL, although low, mirrored transcript levels primarily produced by lung pDCs (Extended Data Fig. 3e,f). Interferons during the recovery phase of influenza infection have been shown to prevent epithelial cell proliferation and differentiation, hence interfering with lung repair⁶¹. Likewise, our findings suggest that sustained type I interferon signaling might also contribute to persistent lung pathology in COVID-19, as supported by histopathological assessment of 14–28-dpi lungs, when interferon levels are particularly high (Fig. 1e,f and Supplementary Table 3). Human and mouse DEGs were also enriched for type II ISGs (Fig. 4e). Type II interferon, IFNG, was mainly produced by T cells (Fig. 4d) as early as 2 dpi and sustained until 28 dpi. (Fig. 4g and Extended Data Fig. 3g). In addition, elevated levels of various pro-inflammatory cytokines and chemokines (IL6, IL8, TNF, IL1B, CXCL10 and IFNG) peaked or remained high late in infection, further supporting delayed immune resolution (Fig. 4g). An inflammatory cytokine signature (particularly IL6 (refs. ^6,10) and IL8 (ref. ⁶²)), recapitulated in our model, was closely correlated with COVID-19 severity in patients^6,63. Elevated serum TNFα levels were not as predictive of severity as other cytokines (IL6 (refs. ^6,10) and IL8 (ref. ⁶²)) in patients but correlated with viral load^6,10. TNFα levels in MISTRG6-hACE2 mice were also higher early in infection with a similar but weaker correlation (Pearson’s r = 0.25) between TNF and viral N transcript, suggesting a more pronounced role for TNFα during the acute phase of disease.

Given the role of type I interferons on bystander activation of T and B cells^64,65, we first focused on the genes that typify bystander activation of memory T cells⁶⁴. This suggested that, early in infection (2–4 dpi), T cell activation might be an antigen-independent, interferon-driven response (Extended Data Fig. 3h). We sought to identify the origin of genes upregulated in patients by validating their expression in our infected mice and then identifying their cellular source⁸. B cell response in humanized lungs was particularly enriched for genes that are upregulated in patients with moderate and severe COVID-19 (Extended Data Fig. 3i,j). Lack of germinal center formation coupled with extrafollicular B cell responses correlated with poor clinical outcomes in patients with COVID-19 (refs. ^50,66). In line with these observations, unbiased pathway analysis of the 28-dpi lung transcriptome (Fig. 4a, Extended Data Fig. 3I and Supplementary Table 2), and a more focused look at the SLE gene signatures⁶⁷ identified a particular enrichment of SLE-like extrafollicular responses in lungs of infected mice at 28 dpi (Fig. 4h). Furthermore, B cell responses at 28 dpi exhibited the features of activated, bystander, clonal B cells that are not vaccine-specific in the context of influenza vaccine response, previously characterized in humans⁶⁵ (Extended Data Fig. 3k), further suggesting a highly inflammatory, bystander B cell response in humanized COVID-19 (refs. ^8,10,11,50). Taken together, our lung transcriptome analysis identifies a central role of monocyte-derived macrophage interferon response that persists and is amplified later in infection and suggests that interferons regulate both the innate and adaptive immune response in COVID-19.

Human monoclonal recombinant antibodies as prophylactic and therapeutic interventions affect disease outcome

We tested whether MISTRG6-hACE2 mice could be used to evaluate patient-derived human antibodies as modulators of infection. MISTRG6-hACE2 mice were treated with convalescent plasma⁶⁸ before infection with SARS-CoV-2 (Extended Data Fig. 4a). The treated mice had significantly lower viral titers in lungs at 4 dpi, showing that the plasma was partially effective (Fig. 5a). However, prophylactic convalescent plasma did not prevent weight loss or histopathological transition into the organizing phase, highlighted by immune cell infiltration, particularly inflammatory macrophages, into the alveolar spaces (Fig. 5b,c and Extended Data Fig. 4b,c) These findings highlight the limited efficacy of prophylactic administration of convalescent plasma in controlling lung pathology^69,70,71. Monoclonal recombinant antibodies (mAbs) cloned from these convalescent patients had high neutralizing activity against SARS-CoV-2 in vitro and in vivo^68,72. Thus, we tested two complementary mAb clones in vivo for prophylactic and therapeutic treatments of SARS-CoV-2 infection in humanized mice (Extended Data Fig. 4a,i). Mice were treated either with individual mAbs before infection or with both mAbs combined at two time points after infection (11 or 35 hours post-infection (hpi)) and analyzed at 4 dpi (Extended Data Fig. 4a,i). As measured by viral titers and viral RNA in lungs, prophylactic treatment with mAbs prevented SARS-CoV-2 infection (Fig. 5d). Prophylactic antibody administration also attenuated immune infiltration, yielding fewer infiltrating immune cells, particularly macrophages in lungs and BAL (Fig. 5e–h and Extended Data Fig. 5d). Although T cell activation in lungs was similar, possibly due to bystander activation of T cells, treatment with mAb clone-144 reduced infiltrating Tγδ cells in lungs (Extended Data Fig. 4e,f) and prevented both systemic T cell lymphopenia (Extended Data Fig. 4g,h) and weight loss (Fig. 5i). Next, we tested therapeutic mAb treatment in MISTRG6-hACE2 mice. Although therapeutic treatment with mAbs similarly prompted viral clearance at both early and late time points, this failed to prevent immune infiltration or significantly alter the composition of the immune infiltrate in lungs in contrast to prophylaxis (Fig. 5j–n and Extended Data Fig. 4j,l,m). Humanized mice treated with both mAbs early (11 hpi) had fewer immune cells in BAL at 4 dpi compared to untreated and late (35 hpi) treated groups, suggesting that the inflammatory responses are attenuated by early treatment, but less so by late treatment, with mAbs (Fig. 5l). Although neither therapeutic intervention prevented weight loss, early treatment prevented systemic T cell lymphopenia, whereas later mAb administration had little effect and a similar infiltration profile as untreated mice at 4 dpi (Fig. 5n and Extended Data Fig. 4k–m). These findings highlight the efficacy of mAb treatment in controlling viral infection and viral titers but underline the need for early treatment to control immunopathology, as noted clinically^69,70,71.

Accurate timing of corticosteroids is necessary to balance viral clearance and prevent immunopathology

Our transcriptome analysis revealed glucocorticoids as possible upstream regulators of DEGs that are induced in infected lungs (Fig. 4a and Supplementary Table 2). Moreover, given that dexamethasone is, thus far, the only therapeutic treatment that has strongly affected recovery and reduced mortality in patients with severe disease⁷³, we hypothesized that dexamethasone treatment might attenuate immunopathology in mice infected with SARS-CoV-2. To test this hypothesis, we treated mice with dexamethasone for 3 d starting at 7 dpi, when the immune infiltration is established but viral titers had declined in lungs (Extended Data Fig. 5a). Mice treated with dexamethasone recovered weight rapidly by 14 dpi and resumed weight gain similar to their uninfected counterparts (Figs. 1d and 6a). Dexamethasone treatment reduced human immune infiltration and reversed immune activation (Fig. 6b–i). Mouse neutrophils in BAL were fewer in dexamethasone-treated mice (Extended Data Fig. 5b), and inflammatory macrophages were largely absent from dexamethasone-treated mouse lungs (Fig. 6c–f). Alveolar macrophages were restored after therapy to a frequency similar to uninfected animals (Fig. 6e,f). Dexamethasone also blocked accumulation of pDCs (Fig. 6g) and reduced T cell activation in lungs (Fig. 6h,i and Extended Data Fig. 5c). Interestingly, dexamethasone treatment also blocked IgG-specific, but not IgM-specific, B cell responses (Extended Data Fig. 5d,e). It was notable that lack of immune cells in dexamethasone-treated lungs also correlated with lower viral RNA levels by 28 dpi (Extended Data Fig. 5f). As the immune infiltrate is established early (by 4 dpi), we investigated the timing of dexamethasone-mediated control of immunopathology for COVID-19. We treated mice early with dexamethasone for 3 d starting at 3 dpi once the immune infiltration was established and viral titers were still high (Extended Data Fig. 5g). In stark contrast to late treatment, early dexamethasone-treated mice became moribund by 7 dpi with rapidly declining weights compared to untreated mice (Fig. 6j). Dexamethasone-treated mice had significantly fewer immune cells infiltrating the lungs and lacked inflammatory macrophages (Fig. 6k and Extended Data Fig. 5h). Notably, the disabled anti-viral response in these mice led to significantly higher viral loads in the lungs (Fig. 6l). These deleterious consequences of early dexamethasone treatment highlight the importance of the early anti-viral response to contain the viral infection. Careful exploration of differential timing of dexamethasone treatment distinguishes the early protective anti-viral response and the subsequent pathological immune response.

Discussion

Here we present a humanized COVID-19 mouse model that combines vector-based delivery of hACE2 and a comprehensive human immune system that recapitulates both innate and adaptive human immunity during SARS-CoV-2 infection. Our model supports chronic disease and enables characterization of the effects of patient-derived antibodies and identification of the sources of immunopathology, in particular the aberrant macrophage response.

Limitations of our model include substantial variability between individual animals and, to a lesser degree, between cohorts. Many parameters might contribute to variability. Although each cohort is engrafted with HSPCs from the same donor, the degree of engraftment varies between individual animals, and, thus, each individual animal tested is unique in its baseline human immune system. Different cohorts are engrafted with cells from different donors; thus, there is genetic variability between different cohorts, which can be exploited in the future. In this study, we used 11 different donors as sources of HSPCs. Further variability can result from experimental differences with the amount of AAV administration and efficiency of cellular infection. Transcriptome analysis revealed differences between individual animals in the strength of the inflammatory response, which might also explain the variable outcome observed in disease outcome in human SARS-CoV-2 infection⁷⁴. However, regardless of such heterogeneity, sustained interferon response, as has been postulated in humans^13,45,46,54, was a common theme that shaped both early anti-viral innate response and late adaptive immune response in humanized COVID-19.

Sustained viral RNA and gross clinical features, including failure to recover body weight and severe lung pathology even at very late time points after infection, are unique among animal models to our MISTRG6 model, with human immune cells and hACE2 expression, which suggests a potentially unique contribution of human immune cells to pathology and/or viral RNA persistence. Human immune cells could produce various factors that might affect infected cells and infection dynamics or be targets of infection themselves. In human lung autopsies, SARS-CoV-2 genomes have been detected in alveolar macrophages⁵³, which express ACE2. We also have detected viral RNA in sorted human immune cells (hCD45⁺), specifically in sorted hCD45⁺CD11b⁺ cells, enriched for human macrophages, but it is not clear whether RNA detected here is the result of viral replication or a residue of phagocytosed infected epithelial cells. Viral titers measured using sensitive cell culture systems were detectable throughout the duration of disease in engrafted MISTRG6-hACE2 mice. Although lung sampling of viable virus is largely lacking in humans, a few patients with severe disease had positive SARS-CoV-2 culture beyond the acute 10-d window after symptom onset^{75,76,77,78,79,80,81,82}. Chronic disease manifestations reflected in histopathological abnormalities (significant cellular infiltrates, thickened septa and collagen deposition) of infected humanized lungs late in infection point to lack of recovery and fibrosis, recapitulating severe human COVID-19 lung pathology. To our knowledge, no other COVID-19 disease model recapitulates the chronic weight loss, sustained high viral RNA and chronic histopathology with pulmonary fibrosis as seen in human patients (Extended Data Fig. 1)^{4,14,16,17,18,19,33,34,35,37,38,39,83,84}. Nonetheless, chronic, humanized COVID-19 is not a lethal disease, which will stimulate further investigation of variables such as age, pre-existing health conditions and comorbidities that contribute to high case fatality rate in humans.

Systemic T cell lymphopenia was another common feature of humanized COVID-19 and human disease^11,12,50. Systemic, cyclical lymphopenia that follows rapid, strong activation of T cells in the infected lungs suggest that T cells are continuously recruited, activated and consumed in the lungs of infected MISTRG6-hACE2 mice. In matched samples, we looked for genes in the lung transcriptome that correlate with spleen T cell frequencies to link systemic lymphopenia to transcriptional changes in lungs. Frequencies of splenic T cells negatively correlated with human expression of genes such as HIF1A, UBAP2L, MIF and FABP7 (Supplementary Table 4), suggesting that inflammation and stress response in lungs, possibly mediated by lung macrophages, might affect systemic lymphopenia. Genes that correlated with systemic lymphopenia were also enriched for ISGs (30/50 top correlating human genes and 26/50 top correlating mouse genes) and suggest some potential therapeutic targets to improve lymphopenia in patients.

MISTRG6-hACE2 mice allowed the study of two aspects of the immune response: infectious viral clearance and immunopathology. We first evaluated patient-derived human antibodies to study this separation. Despite efficient viral clearance, as measured by sharply reduced viral titers, and lack of antibody-mediated enhancement of disease, mAbs, when given late, did not help prevent the potentially pathological inflammatory response, suggesting that the players of immune pathology, such as the inflammatory macrophages, are recruited very early in infection. Our preclinical data are similar to the clinical experience of more than 20 SARS-CoV-2-specific monoclonal antibodies currently in different stages of clinical trial testing^69,70,71. Results so far also support a reduction in viral load upon mAb treatment⁶⁹. Notably, the antibodies prevent hospitalization and death if administered early. Our model could be particularly useful in evaluating efficacy and timing for these antibodies⁷¹. When administered in a timely manner, mAbs could be particularly useful in protecting uninfected individuals and preventing transmission from an infected person by rapid clearance of infectious virus.

Separating the two aspects of the immune response—infectious viral clearance and immunopathology—recapitulated in our model might prove useful in the control of COVID-19. Patients might benefit from early mAb treatment coupled with dexamethasone later in infection. Our data suggest that glucocorticoids, specifically dexamethasone (decoded as upstream regulators of the inflammatory gene signature in humanized COVID-19), should be considered only when viral titers are undetectable, suggesting that timing is crucial in promoting immune protection while preventing pathology and highlighting the importance of accurate and frequent viral detection methods. When applied to chronic disease, dexamethasone effectively controls immunopathology and reverses COVID-19 morbidity. As expected, dexamethasone treatment has broad effects and reverses many aspects of immune activation. However, non-specific suppression of the immune system using dexamethasone early in infection was catastrophic.

Even in the absence of previously described high-risk criteria, there are debilitating effects among certain COVID-19 patient groups. Although socio-economic factors might be responsible for some or all of these effects, perturbation of the proposed system should allow testing of direct medical effects. Our humanized mouse system can be completely personalized by matching patient HSPCs with antibodies and medical history, allowing the testing of novel therapeutics to address conflicting reports in pre-clinical models and to predict efficacy in patients.

Methods

Mice

MISTRG6 was generated by the Flavell laboratory by combining mice generated by this laboratory, the laboratory of Markus Manz and Regeneron Pharmaceuticals based on the Rag2^−/−IL2rg^−/−129xBalb/c background, supplemented with genes for human M-CSF, IL3, SIRPα, thrombopoietin, GM-CSF and IL6 knocked into their respective mouse loci^25,30. MISTRG6 mice are deposited at Jackson Laboratories and made available to academic, non-profit and governmental institutions under a Yale-Regeneron material transfer agreement (already approved and agreed to by all parties). Instructions on obtaining the material transfer agreement for this mouse strain will be available along with strain information and upon reasonable request. Unconstituted MISTRG6 mice were maintained with cycling treatment with enrofloxacin in the drinking water (Baytril, 0.27 mg ml⁻¹). CD1 strain of mice acquired from Charles River Laboratories were used for cross-fostering of MISTRG6 pups upon birth to stabilize healthy microbiota. C57Bl/6 strain of mice (available at Jackson Laboratories) were used as controls for infection with SARS-CoV-2. All mice were maintained under specific pathogen-free conditions in our animal facilities (Biosafety Level (BSL) 1, 2 or 3) under our Animal Studies Committee-approved protocol. Mice were housed on a 14-h light and 10-h dark cycle maintained at 40–60% humidity and 72 °F ± 2 °F. All animal experimentations were performed in compliance with Yale Institutional Animal Care and Use Committee protocols. For SARS-CoV-2-infected mice, all procedures were performed in a BSL-3 facility with approval from the Yale Institutional Animal Care and Use Committee and Yale Environmental Health and Safety.

Transplantation of human CD34⁺ hematopoietic progenitor cells into mice

Fetal liver samples were cut in small fragments, treated for 45 min at 37 °C with collagenase D (Roche, 200 μg ml⁻¹) and prepared into a cell suspension. Human CD34⁺ cells were purified by performing density gradient centrifugation (Lymphocyte Separation Medium, MP Biomedicals), followed by positive immunomagnetic selection with EasySep Human CD34 Positive Selection Kit (STEMCELL Technologies). For intra-hepatic engraftment, newborn 1–3-day-old pups were injected with 20,000 human fetal liver CD34⁺ cells in 20 μl of PBS into the liver with a 22-gauge needle (Hamilton Company). All use of human materials was approved by the Yale University Human Investigation Committee.

AAV-hACE2 administration

AAV9 encoding hACE2 was purchased from Vector Biolabs (AAV9-CMV-hACE2). Animals were anaesthetized using isoflurane. The rostral neck was shaved and disinfected. A 5-mm incision was made, and the trachea was visualized. Using a 32-gauge insulin syringe, a 50-µl injection dose of 10¹¹ genomic copies per milliliter of AAV-CMV-hACE2 was injected into the trachea. The incision was closed with 4–0 Vicryl suture and/or 3M Vetbond tissue adhesive. After administration of analgesic, animals were placed in a heated cage until full recovery. Mice were then moved to BSL-3 facilities for acclimation.

SARS-CoV-2 infection

SARS-CoV-2 isolate USA-WA1/2020 was obtained from the BEI reagent repository. All infection experiments were performed in a BSL-3 facility, licensed by the State of Connecticut and Yale University. Mice were anesthetized using 20% v/v isoflurane diluted in propylene glycol. Using a pipette, 50 µl of SARS-CoV-2 (1–3 × 10⁶ plaque-forming units (PFU)) was delivered intranasally.

Therapeutics

MISTRG6-hACE2 mice received prophylactic treatment of convalescent plasma (5 ml kg⁻¹) or monoclonal antibodies at 10 mg kg⁻¹ (clone 135-m135 or clone 144-m144) 8 h before infection. SARS-CoV-2-infected MISTRG6-hACE2 mice received a mixed cocktail of monoclonal antibodies clone 135 (m135) and clone 144 (m144) at 20 mg kg⁻¹ at 11 hpi for early treatment and 35 hpi for late treatment. SARS-CoV-2-infected MISTRG6-hACE2 were treated with dexamethasone (Sigma-Aldrich) at 10 mg kg⁻¹ for 3 d starting at 7 dpi.

Viral titers

Mice were euthanized in 100% isoflurane. Approximately half of the right lung lobe was placed in a bead homogenizer tube with 1 ml of PBS + 2% FBS. After homogenization, 300 μl of this mixture was placed in 1-ml TRIzol (Invitrogen) for RNA extraction and analysis. Remaining volume of lung homogenates was cleared of debris by centrifugation (3,900g for 10 min). Infectious titers of SARS-CoV-2 were determined by plaque assay in Vero E6 (standard) or Vero ACE2⁺TMPRSS2⁺ (sensitive) cells in DMEM 4% FBS and 0.6% Avicel RC-581¹⁷. Plaques were resolved at 48 hpi by fixing in 10% formaldehyde for 1 h followed by staining for 1 h in 0.5% crystal violet in 20% ethanol. Plates were rinsed in water to visualize plaques. Multiple dilutions of lung homogenates were used to quantify infectious titers (minimum number of plaques that can be quantified per milliliter of lung homogenate = 10).

Viral RNA analysis

RNA was extracted with the RNeasy Mini Kit (Qiagen) per the manufacturer’s protocol. SARS-CoV-2 RNA levels were quantified using the Luna Universal Probe One-Step RT–qPCR kit (New England Biolabs) and US Centers for Disease Control and Prevention real-time RT–PCR primer/probe sets for 2019-nCoV_N1. For each sample, 1 ug of RNA was used.

Histology

Members of the Yale University Department of Pathology kindly provided assistance with embedding and sectioning of lung tissue. A pulmonary pathologist reviewed the slides blinded and identified immune cell infiltration and other related pathologies.

Isolation of cells and flow cytometry

All mice were analyzed at approximately 9–11 weeks of age. Single-cell suspensions were prepared from blood, spleen BAL and lung. Mice were euthanized with 100% isoflurane. BAL was performed using standard methods with a 22-gauge catheter (BD Biosciences). Blood was collected either retro-orbitally or via cardiac puncture after euthanasia. BAL was performed using standard methods with a 22-gauge catheter⁸⁵. Lungs were harvested, minced and incubated in a digestion cocktail containing 1 mg ml⁻¹ of collagenase D (Sigma-Aldrich) and 30 µg ml⁻¹ of DNase I (Sigma-Aldrich) in RPMI at 37 °C for 20 min. Tissue was then filtered through a 70-µm filter. Cells were treated with ammonium chloride potassium buffer and resuspended in PBS with 1% FBS. Mononuclear cells were incubated at 4 °C with human (BD Biosciences) and mouse (Bio X Cell, BE0307) Fc block for 10 min. After washing, primary antibody staining was performed at 4 °C for 20 min. After washing with PBS, cells were fixed using 4% paraformaldehyde. For intracellular staining, cells were washed with BD permeabilization buffer and stained in the same buffer for 45 min at room temperature. Samples were analyzed on an LSRII flow cytometer (BD Biosciences). Data were analyzed using FlowJo software version 3.2 and plotted using Prism 9 version 9.2.0 (283).

Adoptive T cell transfer

Infected lungs (14 dpi) were processed as described before. Cells were stained with antibodies against human CD45, human/mouse CD11B, human CD3 and mouse CD45. Lung T cells (humanCD45⁺, mouseCD45⁻ and CD11B⁻CD3⁺cells) were sorted using a BD Aria Cell Sorter in a BSL-3 facility. Sorted T cells (10,000) were mixed with total splenocytes (10 million) and injected intravenously to unengrafted MISTRG6-hACE2 mice. These mice were then infected as described before and analyzed at 4 dpi.

Antibodies

Antibodies against the following antigens were used:

Mouse antigens: CD45 (Clone: 30-F11), Ly6G (1A8), Ly6C (HK1.4), CD31(MEC13.3) and CD326 (G8.8).

Human antigens: CD45 (HI30), CD3 (UCHT1), CD14 (HCD14), CD16 (3G8), CD19 (HIB19), CD33 (WM53), CD20 (2H7), CD206 (15-2), CD86 (BU63), CD123 (6H6), IGM (MHM-88), IGG (M1310G05) CD163 (GHI/61), CD169 (7-239) CD68 (Y1/82A), CD11B (M1/70), CD11C (3.9), HLA-DR (LN3), CD183 (G025H7), ICOS (C398.4A), PD1 (A17188B), NKp46 (9E2), CD56(MEM-188), CD4 (OKT4), CD8 (SK1), TCRGD (B1), CCR7 (G043H7) and CD83 (HB15e). All antibodies were obtained from BioLegend, unless otherwise specified.

Details on antibodies were provided as Antigen(clone)/ Fluorochrome/ Supplier/ Catalog number/ Concentration. Mouse antigens: CD45 (30-F11) PerCP BioLegend 103130 1:200; CD45 (30-F11) FITC BioLegend 103108 1:200; CD45 (30-F11) Brilliant Violet (BV) 711 BioLegend 103147 1:400; Ly6G (1A8) APC/Cyanine7 BioLegend 127624 1:100; Ly6C (HKl.4) BV605 BioLegend 128036 1:400; CD31 (MEC13.3) APC/Cyanine7 BioLegend 102507 1:400; and CD326 (G8.8) APC/Cyanine7 BioLegend 118218 1:200.

Human antigens: CD45 (Hl30) BV605 BioLegend 304044 1:100; CD45 (Hl30) Pacific Blue BioLegend 304029 1:100; CD3 (UCHT1) PE BioLegend 300408 1:200; CD14(HCD14) APC/Cyanine7 BioLegend 325620 1:100; CD16 (3G8) PerCP BioLegend 302030 1:200; CD16(3G8) FITC BioLegend 302006 1:100; CD19 (HIB19) APC/Cyanine7 BioLegend 302218 1:100; CD19 (HIB19) Alexa Fluor 700 BioLegend 302226 1:100; CD33 (WM53) APC BioLegend 983902 1:100; CD20 (2H7) APC/Cyanine7 BioLegend 302313 1:100; CD20 (2H7) Alexa Fluor 700 BioLegend 302322 1:100; CD206 (15-2) PE BioLegend 321106 1:100; CD206 (15-2) APC BioLegend 321109 1:100; CD86 (BU63) PE/Cyanine7 BioLegend 374210 1:100; CD123 (6H6) PE BioLegend 306006 1:100; IGM (MHM-88) APC BioLegend 314510 1:100; IGM (MHM-88) APC/Cyanine7 BioLegend 314520 1:100; IGG (M1310G05) APC BioLegend 410712 1:100; IGG (M1310G05) APC/Cyanine7 BioLegend 410732 1:100; CD169 (7-239) APC BioLegend 346007 1:100; CD11B (M1/70) BV711 BioLegend 101242 1:100; CD11C(3.9) PE/Cyanine7 BioLegend 301608 1:200; HLA-DR (LN3) Alexa Fluor 700 BioLegend 327014 1:100; HLA-DR (LN3) PerCP BioLegend 327020 1:400; HLA-DR (LN3) FITC BioLegend 327005 1:100; CD183 (G025H7) PE/Cyanine7 BioLegend 353720 1:100; ICOS (C398.4A) APC BioLegend 313510 1:100; CD279-PD1 (A17188B) APC/Cyanine7 BioLegend 1:100; CD335-NKp46 (9E2) PE/Cyanine7 BioLegend 331916 1:100; CD56(MEM-188) APC BioLegend 304610 1:100; CD56 (MEM-188) Pacific Blue BioLegend 304629 1:50; CD4 (OKT4) BV711 BioLegend 317440 1:100; CD8 (SK1) Pacific Blue BioLegend 344718 1:100; CD8 (SK1) BV421 BioLegend 344748 1:200; TCRGD (B1) FITC BioLegend 331208 1:100; CD163 (GHl/61) APC BioLegend 333610 1:100; CD68 (Yl/82A) BV421 BioLegend 333828 1:200; CD83 (HB15) APC BioLegend 305311 1:100; and CCR7 (G043H7) PE/Cyanine7 BioLegend 353226 1:100. Convalescent plasma and monoclonal antibodies (clones 135 and 144) were acquired from M. Nussenzweig, as previously described⁶⁸.

Bulk whole tissue lung RNA sequencing

RNA isolated from homogenized lung tissue used for viral RNA analysis was also used for whole tissue transcriptome analysis. Libraries were made with the help of the Yale Center for Genomic Analysis. In brief, libraries were prepared with an Illumina rRNA depletion kit and sequenced on a NovaSeq. Raw sequencing reads were aligned to the human–mouse combined genome with STAR (https://doi.org/10.1093/bioinformatics/bts635), annotated and counted with HTSeq (https://doi.org/10.1093/bioinformatics/btu638), normalized using DESeq2 (https://doi.org/10.1186/s13059-014-0550-8) and graphed using the Broad Institute Morpheus web tool. Differential expression analysis was also performed with DESeq2. For ISG identification, http://www.interferome.org was used with parameters -In Vivo, -Mus musculus or Homo sapiens -fold change up 2 and down 2. Corresponding pathway analysis of differentially expressed mouse and human genes was achieved using multiple platforms (Gene Ontology, gene set enrichment analysis and Ingenuity).

ELISA

IFNα (Sigma-Aldrich) ELISAs were performed based on manufacturer instructions using serum and BAL fluid from uninfected and infected mice. In short, Triton X-100 and RNase A were added to serum and BAL samples at final concentrations of 0.5% and 0.5 mg ml⁻¹, respectively, and incubated at room temperature for at least 1 h before use to reduce risk from any potential virus in serum and BAL. For serum IFNα levels, 20 ul of serum was used, and, for BAL IFNα levels, 100 ul of BAL fluid was used.

Gene expression

RNA from whole tissue samples was extracted with the RNeasy Mini Kit (Qiagen) per the manufacturer’s protocol. RNA from sorted human immune cells (human CD45⁺), pDCs (CD123^hi and CD11B⁻), macrophages and monocytes (CD11B⁺) and rest of human immune cells (CD123⁻CD11B⁻CD45⁺) was extracted using TRIzol reagent. Expression was normalized to human ACTB or HPRT1. Individual values and means were plotted. High-Capacity cDNA Reverse Transcription Kit was used to make cDNA. qRT–PCR was performed using a SYBR FAST universal qPCR kit (KAPA Biosystems). Predesigned KiCqStart primers for IFNA1, IFNA2, IFNA6, IFNA8, IFNA14, IFNA21, IFNB, IFNG, HPRT1, ACE2, Ace2 and Hprt1 were purchased from Sigma-Aldrich.

Single-cell RNA sequencing (10x Genomics)

Single-cell suspensions from digested lungs were processed for droplet-based single-cell RNA sequencing, and 10,000 cells were encapsulated into droplets using 10x Chromium GEM technology. Libraries were prepared in-house using Chromium Next GEM Single Cell 3′ Reagent Kit version 3.1 (10x Genomics). Single-cell RNA sequencing libraries were sequenced using NovaSeq. Raw sequencing reads were processed with Cell Ranger 3.1.0 using a human–mouse combined reference to generate a gene cell count matrix. To distinguish human and mouse cells, we counted the number of human genes (nHuman) and mouse genes (nMouse) with non-zero expression in each cell and selected cells with nHuman >20 × nMouse as human cells. The count matrix of human cells and human genes was used in the downstream analysis with Seurat 3.2 (ref. ⁸⁶). Specifically, this matrix was filtered, retaining cells with more than 200 and fewer than 5,000 genes and less than 20% mitochondria transcripts. We then log-transformed each entry of the matrix by computing log (CPM/100 + 1), where CPM stands for counts per million. To visualize the cell subpopulations in two dimensions, we applied principal component analysis followed by t-distributed stochastic neighbor embedding (t-SNE), a non-linear dimension reduction method, to the log-transformed data. Graph-based clustering was then used to generate clusters that were overlaid on the t-SNE coordinates to investigate cell subpopulations. Marker genes for each cluster of cells were identified using the Wilcoxon test with Seurat. For the adjusted P values, the Bonferroni correction was used.

Statistical analysis

Unpaired or paired t-test was used to determine statical significance for changes in immune cell frequencies and numbers when comparing infected mice to uninfected control mice. For experiments where mice were treated with monoclonal antibodies or steroids, one-way ANOVA was used to determine statistical significance across groups, and adjusted P values were reported. To determine whether the viral RNA quantification was statistically significant across treatment groups or time points, the Mann–Whitney two-tailed test was used. A one-sample Wilcoxon signed-rank test was used as a non-parametric alternative to the one-sample t-test to determine whether the viral titer quantification of the untreated condition was significantly different from an undetectable viral load (that is, viral titer = 0). Note that our data cannot be assumed to be normally distributed. The results indicate that there is a significant difference between the untreated and mAb-treated mice (for example, early, late or prophylactic treatment). For these analyses, we also calculated the effect size (for example, r = 0.90 (large effect size), V = 21 and P = 0.03).

Additional information on research design is available in the Nature Research Reporting Summary linked to this article.

Reporting Summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.

Data availability

All data that support the findings of this study are available within the paper and its Supplementary Information files. The data supporting this publication are available at Figshare.com under project ‘A humanized mouse model of COVID-19’ (https://figshare.com/projects/A_humanized_mouse_model_of_COVID-19/125659).

All 10x Genomics single-cell RNA sequencing and bulk RNA sequencing data that support the findings of this study were deposited in the Gene Expression Omnibus repository with accession code GSE186794.

References

Letko, M., Marzi, A. & Munster, V. Functional assessment of cell entry and receptor usage for SARS-CoV-2 and other lineage B betacoronaviruses. Nat. Microbiol. 5, 562–569 (2020).
Article CAS PubMed Google Scholar
Huang, C. et al. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. Lancet 395, 497–506 (2020).
Article CAS PubMed PubMed Central Google Scholar
Mehta, P. et al. COVID-19: consider cytokine storm syndromes and immunosuppression. Lancet 395, 1033–1034 (2020).
Xu, Z. et al. Pathological findings of COVID-19 associated with acute respiratory distress syndrome. Lancet Respir. Med. 8, 420–422 (2020).
Article CAS PubMed PubMed Central Google Scholar
Schulte-Schrepping, J. et al. Severe COVID-19 is marked by a dysregulated myeloid cell compartment. Cell 182, 1419–1440 (2020).
Article CAS PubMed PubMed Central Google Scholar
Del Valle, D. M. et al. An inflammatory cytokine signature predicts COVID-19 severity and survival. Nat. Med. 26, 1636–1643 (2020).
Article PubMed PubMed Central CAS Google Scholar
Liao, M. et al. Single-cell landscape of bronchoalveolar immune cells in patients with COVID-19. Nat. Med. 26, 842–844 (2020).
Zhang, J. Y. et al. Single-cell landscape of immunological responses in patients with COVID-19. Nat. Immunol. 21, 1107–1118 (2020).
Article CAS PubMed Google Scholar
Chua, R. L. et al. COVID-19 severity correlates with airway epithelium–immune cell interactions identified by single-cell analysis. Nat. Biotechnol. 38, 970–979 (2020).
Article CAS PubMed Google Scholar
Lucas, C. et al. Longitudinal analyses reveal immunological misfiring in severe COVID-19. Nature 584, 463–469 (2020).
Article CAS PubMed PubMed Central Google Scholar
Chen, Z. & Wherry, E. J. T cell responses in patients with COVID-19. Nat. Rev. Immunol. 20, 529–536 (2020).
Mathew, D. et al. Deep immune profiling of COVID-19 patients reveals distinct immunotypes with therapeutic implications. Science 369, eabc8511 (2020).
Delorey, T. M. et al. COVID-19 tissue atlases reveal SARS-CoV-2 pathology and cellular targets. Nature 595, 107–113 (2021).
Article CAS PubMed PubMed Central Google Scholar
Yang, X. H. et al. Mice transgenic for human angiotensin-converting enzyme 2 provide a model for SARS coronavirus infection. Comp. Med. 57, 450–459 (2007).
CAS PubMed Google Scholar
Li, W. et al. Efficient replication of severe acute respiratory syndrome coronavirus in mouse cells is limited by murine angiotensin-converting enzyme 2. J. Virol. 78, 11429–11433 (2004).
Article CAS PubMed PubMed Central Google Scholar
McCray, P. B. et al. Lethal infection of K18-hACE2 mice infected with severe acute respiratory syndrome coronavirus. J. Virol. 81, 813–821 (2007).
Article CAS PubMed Google Scholar
Israelow, B. et al. Mouse model of SARS-CoV-2 reveals inflammatory role of type I interferon signaling. J. Exp. Med. 217, e20201241 (2020).
Bao, L. et al. The pathogenicity of SARS-CoV-2 in hACE2 transgenic mice. Nature 583, 830–833 (2020).
Article CAS PubMed Google Scholar
Hassan, A. O. et al. A SARS-CoV-2 infection model in mice demonstrates protection by neutralizing antibodies. Cell 182, 744–753 (2020).
Article CAS PubMed PubMed Central Google Scholar
Xu, H. et al. High expression of ACE2 receptor of 2019-nCoV on the epithelial cells of oral mucosa. Int. J. Oral. Sci. 12, 8–8 (2020).
Article CAS PubMed PubMed Central Google Scholar
Yan, R. et al. Structural basis for the recognition of SARS-CoV-2 by full-length human ACE2. Science 367, 1444–1448 (2020).
Article CAS PubMed PubMed Central Google Scholar
Hoffmann, M. et al. SARS-CoV-2 cell entry depends on ACE2 and TMPRSS2 and is blocked by a clinically proven protease inhibitor. Cell 181, 271–280 (2020).
Article CAS PubMed PubMed Central Google Scholar
Theocharides, A. P., Rongvaux, A., Fritsch, K., Flavell, R. A. & Manz, M. G. Humanized hemato-lymphoid system mice. Haematologica 101, 5–19 (2016).
Article CAS PubMed PubMed Central Google Scholar
Rongvaux, A. et al. Human hemato-lymphoid system mice: current use and future potential for medicine. Annu Rev. Immunol. 31, 635–674 (2013).
Article CAS PubMed PubMed Central Google Scholar
Rongvaux, A. et al. Development and function of human innate immune cells in a humanized mouse model. Nat. Biotechnol. 32, 364–372 (2014).
Article CAS PubMed PubMed Central Google Scholar
Rathinam, C. et al. Efficient differentiation and function of human macrophages in humanized CSF-1 mice. Blood 118, 3119–3128 (2011).
Willinger, T. et al. Human IL-3/GM-CSF knock-in mice support human alveolar macrophage development and human immune responses in the lung. Proc. Natl Acad. Sci. USA 108, 2390–2395 (2011).
Article CAS PubMed PubMed Central Google Scholar
Strowig, T. et al. Transgenic expression of human signal regulatory protein alpha in Rag2^−/−γ_c^−/− mice improves engraftment of human hematopoietic cells in humanized mice. Proc. Natl Acad. Sci. USA 108, 13218–13223 (2011).
Article CAS PubMed PubMed Central Google Scholar
Rongvaux, A. et al. Human thrombopoietin knockin mice efficiently support human hematopoiesis in vivo. Proc. Natl Acad. Sci. USA 108, 2378–2383 (2011).
Article CAS PubMed PubMed Central Google Scholar
Yu, H. et al. A novel humanized mouse model with significant improvement of class-switched, antigen-specific antibody production. Blood 129, 959–969 (2017).
Article CAS PubMed PubMed Central Google Scholar
Das, R. et al. Microenvironment-dependent growth of preneoplastic and malignant plasma cells in humanized mice. Nat. Med. 22, 1351–1357 (2016).
Article CAS PubMed PubMed Central Google Scholar
Sippel, T. R., Radtke, S., Olsen, T. M., Kiem, H.-P. & Rongvaux, A. Human hematopoietic stem cell maintenance and myeloid cell development in next-generation humanized mouse models. Blood Adv. 3, 268 (2019).
Article CAS PubMed PubMed Central Google Scholar
Sun, S.-H. et al. A mouse model of SARS-CoV-2 infection and pathogenesis. Cell Host Microbe 28, 124–133 (2020).
Rockx, B. et al. Comparative pathogenesis of COVID-19, MERS, and SARS in a nonhuman primate model. Science 368, 1012–1015 (2020).
Article CAS PubMed PubMed Central Google Scholar
Roberts, A. et al. A mouse-adapted SARS-coronavirus causes disease and mortality in BALB/c mice. PLoS Pathog. 3, e5 (2007).
Article PubMed PubMed Central CAS Google Scholar
Polak, S. B., Van Gool, I. C., Cohen, D., von der Thusen, J. H. & van Paassen, J. A systematic review of pathological findings in COVID-19: a pathophysiological timeline and possible mechanisms of disease progression. Mod. Pathol. 33, 2128–2138 (2020).
Article CAS PubMed PubMed Central Google Scholar
Tian, S. et al. Pathological study of the 2019 novel coronavirus disease (COVID-19) through postmortem core biopsies. Mod. Pathol. 33, 1007–1014 (2020).
Menter, T. et al. Postmortem examination of COVID‐19 patients reveals diffuse alveolar damage with severe capillary congestion and variegated findings in lungs and other organs suggesting vascular dysfunction. Histopathology 77, 198–209 (2020).
Article PubMed PubMed Central Google Scholar
Barton, L. M., Duval, E. J., Stroberg, E., Ghosh, S. & Mukhopadhyay, S. Covid-19 autopsies, Oklahoma, USA. Am. J. Clin. Pathol. 153, 725–733 (2020).
Article CAS PubMed Google Scholar
Nathan, C. Neutrophils and COVID-19: Nots, NETs, and knots. J. Exp. Med. 217, e20201439 (2020).
Shafran, N. et al. Secondary bacterial infection in COVID-19 patients is a stronger predictor for death compared to influenza patients. Sci. Rep. 11, 12703 (2021).
Article CAS PubMed PubMed Central Google Scholar
Russell, C. D. et al. Co-infections, secondary infections, and antimicrobial use in patients hospitalised with COVID-19 during the first pandemic wave from the ISARIC WHO CCP-UK study: a multicentre, prospective cohort study. Lancet Microbe 2, e354–e365 (2021).
Baharom, F. et al. Dendritic cells and monocytes with distinct inflammatory responses reside in lung mucosa of healthy humans. J. Immunol. 196, 4498–4509 (2016).
Article CAS PubMed Google Scholar
Villani, A. C. et al. Single-cell RNA-seq reveals new types of human blood dendritic cells, monocytes, and progenitors. Science 356, eaah4573 (2017).
Zhou, Z. et al. Heightened innate immune responses in the respiratory tract of COVID-19 patients. Cell Host Microbe 27, 883–890 (2020).
Nienhold, R. et al. Two distinct immunopathological profiles in autopsy lungs of COVID-19. Nat. Commun. 11, 5086 (2020).
Tan, L. et al. Lymphopenia predicts disease severity of COVID-19: a descriptive and predictive study. Signal Transduct. Target. Ther. 5, 33 (2020).
Wang, S. et al. IL-21 drives expansion and plasma cell differentiation of autoreactive CD11c^hi T-bet⁺ B cells in SLE. Nat. Commun. 9, 1758 (2018).
Article PubMed PubMed Central CAS Google Scholar
Jenks, S. A. et al. Distinct effector B cells induced by unregulated toll-like receptor 7 contribute to pathogenic responses in systemic lupus erythematosus. Immunity 49, 725–739 (2018).
Article CAS PubMed PubMed Central Google Scholar
Woodruff, M. C. et al. Extrafollicular B cell responses correlate with neutralizing antibodies and morbidity in COVID-19. Nat. Immunol. 21, 1506–1516 (2020).
Article PubMed PubMed Central CAS Google Scholar
Schultheiß, C. et al. Next-generation sequencing of T and B cell receptor repertoires from COVID-19 patients showed signatures associated with severity of disease. Immunity 53, 442–455 (2020).
Article PubMed PubMed Central CAS Google Scholar
Stephens, D. S. & McElrath, M. J. COVID-19 and the path to immunity. JAMA 324, 1279–1281 (2020).
Article CAS PubMed Google Scholar
Grant, R. A. et al. Circuits between infected macrophages and T cells in SARS-CoV-2 pneumonia. Nature 590, 635–641 (2021).
Dolan, M. E. et al. Investigation of COVID-19 comorbidities reveals genes and pathways coincident with the SARS-CoV-2 viral disease. Sci. Rep. 10, (2020).
Rodero, M. P. et al. Detection of interferon alpha protein reveals differential levels and cellular sources in disease. J. Exp. Med. 214, 1547–1555 (2017).
Article CAS PubMed PubMed Central Google Scholar
Peiffer-Smadja, N. & Yazdanpanah, Y. Nebulised interferon beta-1a for patients with COVID-19. Lancet Resp. Med. 9, 122–123 (2021).
Davoudi-Monfared, E. et al. A randomized clinical trial of the efficacy and safety of interferon β-1a in treatment of severe COVID-19. Antimicrob. Agents Chemother. 64, e01061-20 (2020).
Acharya, D., Liu, G. & Gack, M. U. Dysregulation of type I interferon responses in COVID-19. Nat. Rev. Immunol. 20, 397–398 (2020).
Article CAS PubMed PubMed Central Google Scholar
Blanco-Melo, D. et al. Imbalanced host response to SARS-CoV-2 drives development of COVID-19. Cell 181, 1036–1045 (2020).
McNab, F., Mayer-Barber, K., Sher, A., Wack, A. & O’Garra, A. Type I interferons in infectious disease. Nat. Rev. Immunol. 15, 87–103 (2015).
Article CAS PubMed PubMed Central Google Scholar
Major, J. et al. Type I and III interferons disrupt lung epithelial repair during recovery from viral infection. Science 369, 712–717 (2020).
Article CAS PubMed PubMed Central Google Scholar
Huang, W. et al. The inflammatory factors associated with disease severity to predict COVID-19 progression. J. Immunol. 206, 1597–1608 (2021).
Article CAS PubMed Google Scholar
Leisman, D. E. et al. Cytokine elevation in severe and critical COVID-19: a rapid systematic review, meta-analysis, and comparison with other inflammatory syndromes. Lancet Respir. Med. 8, 1233–1244 (2020).
Low, J. S. et al. Tissue-resident memory T cell reactivation by diverse antigen-presenting cells imparts distinct functional responses. J. Exp. Med. 217, e20192291 (2020).
Horns, F., Dekker, C. L. & Quake, S. R. Memory B cell activation, broad anti-influenza antibodies, and bystander activation revealed by single-cell transcriptomics. Cell Rep. 30, 905–913 (2020).
Article CAS PubMed PubMed Central Google Scholar
Kaneko, N. et al. Loss of Bcl-6-expressing T follicular helper cells and germinal centers in COVID-19. Cell 183, 143–157 (2020).
Article CAS PubMed PubMed Central Google Scholar
Hutcheson, J. et al. Combined deficiency of proapoptotic regulators Bim and Fas results in the early onset of systemic autoimmunity. Immunity 28, 206–217 (2008).
Article CAS PubMed Google Scholar
Davide, F. R. et al. Convergent antibody responses to SARS-CoV-2 in convalescent individuals. Nature 584, 437–442 (2020).
Cruz-Teran, C. et al. Challenges and opportunities for antiviral monoclonal antibodies as COVID-19 therapy. Adv. Drug Deliv. Rev. 169, 100–117 (2020).
Casadevall, A. & Pirofski, L.-A. The convalescent sera option for containing COVID-19. J. Clin. Invest. 130, 1545–1548 (2020).
Chen, P. et al. SARS-CoV-2 neutralizing antibody LY-CoV555 in outpatients with Covid-19. N. Engl. J. Med. 384, 229–237 (2020).
Schäfer, A. et al. Antibody potency, effector function, and combinations in protection and therapy for SARS-CoV-2 infection in vivo. J. Exp. Med. 218, e20201993 (2020).
RECOVERY Collaborative Group et al. Dexamethasone in hospitalized patients with Covid-19. N. Engl. J. Med. 384, 693–704 (2020).
Ware, L. B. Physiological and biological heterogeneity in COVID-19-associated acute respiratory distress syndrome. Lancet Respir. Med. 8, 1163–1165 (2020).
Singanayagam, A. et al. Duration of infectiousness and correlation with RT–PCR cycle threshold values in cases of COVID-19, England, January to May 2020. Euro. Surveill. 25, 2001483 (2020).
Arons, M. M. et al. Presymptomatic SARS-CoV-2 infections and transmission in a skilled nursing facility. N. Engl. J. Med. 382, 2081–2090 (2020).
Article CAS PubMed Google Scholar
Basile, K. et al. Cell-based culture of SARS-CoV-2 informs infectivity and safe de-isolation assessments during COVID-19. Clin. Infect. Dis. 73, e2952–e2959 (2020).
Decker, A. et al. Prolonged SARS-CoV-2 shedding and mild course of COVID-19 in a patient after recent heart transplantation. Am. J. Transpl. 20, 3239–3245 (2020).
Article CAS Google Scholar
Folgueira, M. D., Luczkowiak, J., Lasala, F., Pérez-Rivilla, A. & Delgado, R. Prolonged SARS-CoV-2 cell culture replication in respiratory samples from patients with severe COVID-19. Clin. Microbiol. Infect. 27, 886–891 (2021).
Article CAS PubMed PubMed Central Google Scholar
Jeong, H. W. et al. Viable SARS-CoV-2 in various specimens from COVID-19 patients. Clin. Microbiol. Infect. 26, 1520–1524 (2020).
Article CAS PubMed PubMed Central Google Scholar
Liu, W. D. et al. Prolonged virus shedding even after seroconversion in a patient with COVID-19. J. Infect. 81, 318–356 (2020).
PubMed PubMed Central Google Scholar
van Kampen, J. J. A. et al. Duration and key determinants of infectious virus shedding in hospitalized patients with coronavirus disease-2019 (COVID-19). Nat. Commun. 12, 267 (2021).
Article PubMed PubMed Central CAS Google Scholar
Bussani, R. et al. Persistence of viral RNA, pneumocyte syncytia and thrombosis are hallmarks of advanced COVID-19 pathology. EBioMedicine 61, 103104 (2020).
Article PubMed PubMed Central Google Scholar
Wei, J. et al. Genome-wide CRISPR screens reveal host factors critical for SARS-CoV-2 infection. Cell 184, 76–91 (2020).
Sun, F., Xiao, G. & Qu, Z. Murine bronchoalveolar lavage. Bio. Protoc. 7, e2287 (2017).
PubMed PubMed Central Google Scholar
Stuart, T. et al. Comprehensive integration ofsingle-cell data. Cell 177, 1888–1902. e1821.
Imai, M. et al. Syrian hamsters as a small animal model for SARS-CoV-2 infection and countermeasure development. Proc. Natl. Acad. Sci. USA 117, 16587–16595 (2020).
Article CAS PubMed PubMed Central Google Scholar
Dinnon, K. H. et al. A mouse-adapted model ofSARS-CoV-2 to test COVID-19 countermeasures. Nature 586, 560–566 (2020).
Article PubMed PubMed Central CAS Google Scholar
Song, E. et al. Neuroinvasion of SARS-CoV-2 inhuman and mouse brain. J. Exp. Med. 218, e20202135 (2021).
Article CAS PubMed PubMed Central Google Scholar
Ter Meulen, J. et al. Human monoclonal antibody asprophylaxis for SARS coronavirus infection in ferrets. Lancet 363, 2139–2141 (2004).
Article CAS PubMed PubMed Central Google Scholar
Oladunni, F. S. et al. Lethality of SARS-CoV-2 infection in K18 human angiotensin-converting enzyme 2 transgenic mice. Nat.Commun. 11, 6122 (2020).
Article CAS PubMed PubMed Central Google Scholar

Download references

Acknowledgements

The generation of the original MISTRG6 model was supported by the Bill and Melinda Gates Foundation. We thank G. Yancopoulos, D. Valenzuela, A. Murphy and W. Auerbach at Regeneron Pharmaceuticals who generated, in collaboration with our groups, the individual knock-in alleles combined in MISTRG. We thank H. Steach, M. Chiorazzi, I. Odell, E. Eynon and W. Philbrick and all the other members of the Flavell laboratory for discussions and comments; J. Alderman, C. Lieber, B. Cadugan and E. Hughes-Picard for administrative assistance; P. Ranney and C. Hughes for mouse colony management; D. Urbanos for human CD34⁺ cell isolation; and L. Devine for help with cell sorting. E.S. is a Howard Hughes Medical Institute Fellow of the Damon Runyon Cancer Research Foundation (DRG-2316-18). This work was funded by the Howard Hughes Medical Institute (R.A.F., M.N. and A.I.). This study was also supported, in part, by National Institute of Health grants R01AI157488 (A.I.), F30CA239444 (E.S.), 2T32AI007517, K08AI163493 (B.I.), AI061093 (E.M.), AI118855 (E.M.), CA016359 (E.M.), K08 AI128043 (C.B.W.) and U01 CA260507 (S.H.), the Burroughs Wellcome Fund (C.B.W.), the Patterson Foundation (C.B.W.), a Fast Grant from Emergent Ventures at the Mercatus Center (A.I., E.S. and C.B.W.), the Mathers Foundation (A.I., C.B.W. and E.M.) and the Ludwig Family Foundation (A.I. and C.B.W.).

Author information

Authors and Affiliations

Department of Immunobiology, Yale University School of Medicine, New Haven, CT, USA
Esen Sefik, Benjamin Israelow, Haris Mirza, Jun Zhao, Rihao Qu, Eleanna Kaffe, Eric Song, Eric Meffre, Craig B. Wilen, Akiko Iwasaki & Richard A. Flavell
Department of Medicine, Section of Infectious Diseases, Yale University School of Medicine, New Haven, CT, USA
Benjamin Israelow
Department of Pathology, Yale University School of Medicine, New Haven, CT, USA
Haris Mirza, Jun Zhao, Rihao Qu & Yuval Kluger
Section of Hematology, Yale Cancer Center and Department of Internal Medicine, Yale University School of Medicine, New Haven, CT, USA
Stephanie Halene
Laboratory of Molecular Immunology, The Rockefeller University, New York, NY, USA
Michel Nussenzweig
Department of Laboratory Medicine, Yale University School of Medicine, New Haven, CT, USA
Craig B. Wilen
Howard Hughes Medical Institute, Yale University School of Medicine, New Haven, CT, USA
Akiko Iwasaki & Richard A. Flavell

Authors

Esen Sefik
View author publications
You can also search for this author in PubMed Google Scholar
Benjamin Israelow
View author publications
You can also search for this author in PubMed Google Scholar
Haris Mirza
View author publications
You can also search for this author in PubMed Google Scholar
Jun Zhao
View author publications
You can also search for this author in PubMed Google Scholar
Rihao Qu
View author publications
You can also search for this author in PubMed Google Scholar
Eleanna Kaffe
View author publications
You can also search for this author in PubMed Google Scholar
Eric Song
View author publications
You can also search for this author in PubMed Google Scholar
Stephanie Halene
View author publications
You can also search for this author in PubMed Google Scholar
Eric Meffre
View author publications
You can also search for this author in PubMed Google Scholar
Yuval Kluger
View author publications
You can also search for this author in PubMed Google Scholar
Michel Nussenzweig
View author publications
You can also search for this author in PubMed Google Scholar
Craig B. Wilen
View author publications
You can also search for this author in PubMed Google Scholar
Akiko Iwasaki
View author publications
You can also search for this author in PubMed Google Scholar
Richard A. Flavell
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

E.S. conceived the project, performed experiments, analyzed the data and wrote the manuscript. B.I. performed experiments and helped establish the model in Biosafety Level 3. H.M. preformed histopathological assessment of lung pathology, offered vital conceptual insights and aided in writing of the manuscript. J.Z. and R.Q. performed bioinformatics analysis. E.K. prepared samples for histopathological assessment. E.S. helped establish the model in Biosafety Level 3. M.N. provided monoclonal antibodies used in the study. S.H., E.M., Y.K., M.N., C.B.W. and A.I. offered vital conceptual insights, contributed to the overall interpretation of this work and aided in writing of the manuscript. R.A.F. co-conceived and supervised the project, helped interpret the work and supervised writing of the manuscript.

Corresponding author

Correspondence to Richard A. Flavell.

Ethics declarations

Competing interests

R.A.F. is an advisor to GlaxoSmithKline, Zai Lab and Ventus Therapeutics.

Additional information

Peer review information Nature Biotechnology thanks the anonymous reviewers for their contribution to the peer review of this work.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

Extended Data Fig. 1 Comparison of COVID-19 parameters and response to therapeutics in human patients, humanized mice and other animal models.

Although substantial anti-viral immunity mediates viral clearance in non-severe COVID-19², robust inflammatory cytokine production, decreased circulating lymphocytes and failure to generate germinal centers characterizes the immunopathology in severe COVID-19^2,3,4. Dysregulated myeloid^5,6,7,8 and lymphocyte^{9,10,11,12,53} compartments, including monocytes, macrophages, neutrophils, NK cells and antibody secreting B cells, have been described as players in the observed lung immunopathology of COVID-19. Accurate model systems are essential to rapidly evaluate promising discoveries but most models currently available in mice, ferrets and hamsters do not recapitulate sustained immunopathology described in COVID-19 patients. The following table summarizes various aspects of the human disease and how well these are recapitulated in existing animal models of COVID-19 and in our humanized mouse model of COVID-19. Additional work cited: refs. ^{87,88,89,90,91}.

Extended Data Fig. 2 (matched to Fig. 3).

a. Numbers of human T cells in the blood pre- and post-infection (2,4,7,14,28dpi). Unpaired, two-tailed t-test. P-values<0.05 are plotted. Uninfected n=5, 2dpi n=4, 4dpi n=6, 7dpi n=5, 14dpi n=5, 28dpi=4, biologically independent mice examined over at least 3 independent experiments. Individual values for each mouse and means are plotted. b. Numbers of human T cells in the spleens of uninfected or infected mice (2,4,7,14,28dpi). Unpaired, two-tailed t-test. P-values<0.05 are plotted. Unpaired, two-tailed t-test. Only p-values<0.05 are plotted. Uninfected n=7, 2dpi n=4, 4dpi n=6, 7dpi n=4, 14dpi n=8, 28dpi n=4, biologically independent mice examined over at least 3 independent experiments. Individual values for each mouse and means are plotted. c. Frequencies of human CD4+ or CD8+ T cells within hCD3+ population in the blood and spleens of uninfected or infected mice at 2,4,7,14,28 dpi. Blood: n=4 biologically independent mice examined over at least 2 independent experiments. Means with SD are plotted. Spleen: Uninfected n=5, 2dpi n=4, 4dpi n=6, 7dpi n=4, 14dpi n=4 biologically independent mice examined over at least 3 independent experiments. Individual values for each mouse and means are plotted. d. Frequencies of human CD4+ or CD8+ T cells within hCD3+ population in the lungs of uninfected or infected mice at 2,4,7,14,28 dpi. Uninfected n=4, 2dpi n=4, 4dpi n=6, 7dpi n=4, 14dpi n=4 biologically independent mice examined over 3 independent experiments. Individual values for each mouse and means are plotted. e. Representative flow cytometry plots of CD8+ T cells stained with viral antigen bearing HLA-A2 tetramers. HLA-A2 tetramers bound to peptides of Nucleoprotein(N), Membrane protein(M) and Spike protein were used to characterize T cells from uninfected and infected (2,4,7,14,28dpi) mice. Representative of uninfected n=3, 2dpi n=2, 4dpi n=3, 7dpi n=3, 14dpi n=3, 28dpi n=3 biologically independent mice examined over 2 independent experiments. f. Lung T cell numbers (per lobe) and viral titers following prophylactic T cell transfer in infected mice at 4dpi. Splenocytes and sorted T cells from lungs of 14dpi mice where the contribution of the antigen specific B cell response was minimal were transferred to unengrafted MISTRG-hACE2 (no human immune cells prior to transfer). Control MISTRG-hACE2 mice received splenocytes but not sorted lung T cells (n=3 biologically independent mice examined over 2 independent experiments). These mice were subsequently infected and analyzed at 4dpi for viral titers and human immune cells in lungs. In mice that received sorted lung T cells, human immune compartment in lungs comprised of T cells and few CD16+ monocytes and macrophages but no B cells (n=4 biologically independent mice examined over 2 independent experiments). g. Numbers of human B cells in the BAL of uninfected or infected mice at 2, 4, 7, 14, 28 dpi. Unpaired, two-tailed t-test. P-values<0.05 are plotted. Uninfected n=3, 2dpi n=3, 4dpi n=4, 7dpi n=3, 14dpi n=7, 28dpi n=4 biologically independent mice examined over at least 3 independent experiments. Individual values for each mouse and means are plotted. h. Representative flow cytometry plots of IgG+ B cells in the BAL of uninfected and infected mice 14 and 28dpi. Representative of n=4 biologically independent mice examined over 3 independent experiments. i. CD11C and CD19 expression on human immune cells from infected lungs at 14 and 28 dpi. Representative of n=5 (14dpi) and n=4 (28dpi) biologically independent mice examined over at least 3 independent experiments. j. Frequencies of human B cells marked by CD19 and CD20 expression within hCD45+ population in spleen and blood of uninfected or infected mice at 2, 4, 7, 14, 28 dpi. Blood: Uninfected n=6, 2dpi n=5, 4dpi n=8, 7dpi n=5, 14dpi n=6, 28dpi n=6 biologically independent mice examined over at least 3 independent experiments. Spleen: Uninfected n=6, 2dpi n=4, 4dpi n=8, 7dpi n=5, 14dpi n=6, 28dpi n=7 biologically independent mice examined over at least three independent experiments. Individual values for each mouse and means are plotted. k. IgG levels measured by ELISA in serum of uninfected or infected mice at 7, 14, 28 dpi. Uninfected n=5, 7dpi n=2, 14dpi n=2, 28dpi n=5 biologically independent mice examined over 2 independent experiments. Individual values for each mouse and means are plotted. Unpaired, two tailed t-test.

Extended Data Fig. 3 (matched to Fig. 4).

a. t-distributed stochastic neighbor embedding (t-SNE) plot with clustering results of single cell RNA sequencing of mouse cells from lungs at 4dpi. Single cell suspensions from whole infected lung at 4dpi were processed and sequenced. There were 731 cells identified as mouse cells. b. Dot plot of cluster identifying genes for mouse cell clusters described in S4A. c. Cluster distribution and expression profile of mouse inflammatory cytokines for clusters described in 4B. d. Normalized expression of human IFNB measured by qPCR in homogenized lung tissue of uninfected and infected (2,4,7,14,28dpi) mice. Expression was normalized to human HPRT1. Unpaired t-test, one-tailed since the distribution is expected to be one-sided. Uninfected n=3, 2dpi n=3, 4dpi n=6, 7dpi n=7, 14dpi n=10, 28dpi n=2 biologically independent mice examined over at least 3 independent experiments. Individual values for each mouse and means are presented. P-values<0.05 are plotted. e. IFNA levels measured by ELISA in serum and BAL of uninfected or infected mice. Serum: Uninfected n=6, 4dpi n=3, 7dpi n=2, 14dpi n=5, 28dpi n=3; BAL: Uninfected n=3, 14dpi n=4 biologically independent mice examined over at least 2 independent experiments for all time points. Individual values (average of duplicates) for each mouse and means are plotted. f. Type I interferon levels (IFNA6 and IFNB) per cell measured by qPCR in sorted human immune cells (human CD45+): plasmacytoid dendritic cells (PDCs; CD123^hi CD11B-), macrophages and monocytes (CD11B+) and rest of human immune cells (CD123-CD11B-CD45+). Expression was normalized to human ACTB. Individual values and means are plotted. n=3 biologically independent mice examined over 2 independent experiments. Individual values with means are plotted. g. Normalized expression of human IFNG measured by qPCR in homogenized lung tissue of uninfected and infected (2,4,7,14,28dpi) mice. Expression was normalized to human HPRT1. Unpaired t-test, one tailed since distribution is expected to be one-sided. Uninfected n=10, 2dpi n=3, 4dpi n=8, 7dpi n=3, 14dpi n=15, 28dpi n=5 biologically independent mice examined over at least 3 independent experiments for all time points. h. Heat map of normalized counts for bystander activated memory T cell signature genes⁶⁴ (based on Low et al., JEM 2020) in lungs of MISTRG6-hACE2 mice infected with SARS-CoV-2. Row min and max of transformed values, calculated by subtracting row mean and diving by STD for each gene across all samples, are visualized. Uninfected n=5, 2dpi n=4, 4dpi n=7, 7dpi n=3, 14dpi n=2, 28dpi n=2 biologically independent mice examined over at least 2 independent experiments for all time points. i. Volcano plot showing foldchange and adjusted p-values (by Wald test using DESeq2) of differentially regulated genes at 28dpi compared to uninfected lungs. Genes with FC(Log2)>1 and p value<0.05 are highlighted in red. Replicates of at least 2 mice. Uninfected n=5, 28dpi n=2 biologically independent mice examined over 2 independent experiments. j. Heatmap of normalized counts for genes that are induced in B cells of patients with moderate or severe COVID-19 in comparison with healthy controls. Normalized counts in lungs of uninfected or infected MISTRG6-hACE2 were plotted over the course of infection. Row min and max of transformed values, calculated by subtracting row mean and diving by STD for each gene across all samples, are visualized. Uninfected n=5, 2dpi n=4, 4dpi n=7, 7dpi n=3, 14dpi n=2, 28dpi n=2 biologically independent mice examined over at least 2 independent experiments for all time points. k. Heatmap of normalized counts for bystander activated memory B cell signature genes⁶⁵ (based on Horns et al., 2020) in lungs of MISTRG6-hACE2 mice infected with SARS-CoV-2. Row min and max of transformed values, calculated by subtracting row mean and diving by STD for each gene across all samples, are visualized. Uninfected n=5, 2dpi n=4, 4dpi n=7, 7dpi n=3, 14dpi n=2, 28dpi n=2 biologically independent mice examined over at least 2 independent experiments for all time points.

Extended Data Fig. 4 (matched to Fig. 5).

a. Schematic of experimental design of prophylactic antibody treatment MISTRG6-hACE2 mice received prophylactic treatment of convalescent plasma (5ml/kg) or monoclonal antibodies at 10mg/kg (clone 135 -m135 or clone 144-m144) 8 h prior to infection or left untreated (untd). Mice were euthanized 4dpi. Convalescent plasma samples from the top 30 neutralizers in a cohort of 148 individuals were pooled to create a mixture with an NT50 titer of 1597 against HIV-1 pseudotyped with SARS-CoV-2 S protein⁶⁸. Sequencing the antibody genes from infected humans has revealed the expansion of closely related Receptor Binding Domain of the Spike protein (RBD)-specific B cell antibody clones in different SARS-CoV-2 infected individuals. Although most convalescent plasma samples obtained from patients who recovered from COVID-19 did not contain high levels of neutralizing activity, RBD-specific antibodies with potent antiviral activity were found in all individuals tested⁶⁸. Monoclonal recombinant antibodies (mAbs) used in this study were cloned from these convalescent patients and had high neutralizing activity against SARS-CoV-2 in vitro and in vivo in mouse adapted SARS-CoV-2 infection^68,72. b. Weight change in convalescent plasma treated mice at 2 and 4dpi plotted as percent change compared with original weight measured just before inoculation with SARS-CoV-2. Unpaired t-test, two-tailed comparing 4dpi values. N=4 biologically independent mice examined over 2 independent experiments for all time points. Individual values per each mouse plotted. c. Representative images (10x magnification) of H&E staining of lungs (4dpi) from mice treated prophylactically with convalescent plasma or left untreated. The histopathological findings mark transition into the organizing phase, highlighted by infiltration of the inter- and intra-alveolar space by lymphocytes macrophages, and fibroblasts. Representative of N=4 biologically independent mice examined over 2 independent experiments for all time points. Individual values for each mouse are plotted. d. Frequencies of human monocytes (CD14+ classical; CD14+CD16+ intermediate, CD16+ non classical) within human CD45+ cells in the lungs of MISTRG6-hACE2 mice which received a prophylactic treatment of monoclonal antibody clone 135 (m135) or clone 144(m144) 8 h prior to infection or left untreated (untd). Unpaired, two-tailed t-test. P-values<0.05 are plotted. Untreated N=5, treated n=6 biologically independent mice examined over 2 independent experiments for all time points. Individual values for each mouse and means are plotted. e. Frequencies of HLA-DR+CXCR3+ T cells within human CD3+CD45+ cells in lungs of MISTRG6-hACE2 mice which received a prophylactic treatment of monoclonal antibody clone 135 (m135) or clone 144(m144) 8 h prior to infection or left untreated (untd). Unpaired, two-tailed t-test. P-values<0.05 are plotted. Untreated N=5, treated n=6 N=4 biologically independent mice examined over 2 independent experiments for all time points. Individual values for each mouse and means are plotted. f. Frequencies of Tγδ cells within human CD3+CD45+ cells in lungs of MISTRG6-hACE2 mice which received prophylactic treatment of monoclonal antibody clone 135 (m135) or clone 144(m144) 8 h prior to infection or left untreated (untd). Unpaired, two-tailed t-test. N=5-6. P-values<0.05 are plotted. Untreated N=5, treated n=6 biologically independent mice examined over 2 independent experiments for all time points. Individual values for each mouse and means are plotted. g. Frequencies of human T cells within human CD45+ cells in spleens of MISTRG6-hACE2 mice received a prophylactic treatment of monoclonal antibody clone 135 (m135) or clone 144(m144) 8 h prior to infection or left untreated (untd). Unpaired, two-tailed t-test. P-values<0.05 are plotted. Untreated N=5, treated n=64 biologically independent mice examined over two independent experiments for all time points. Individual values for each mouse and means are plotted. h. Frequencies of human CD3+ T cells within human CD45+ population in the blood pre- and post-infection (2, 4, 7, 14,2 8dpi). Lines connect pre- and post-infection values for the same mouse. MISTRG6-hACE2 mice received a prophylactic treatment of monoclonal antibody clone 135 (m135) or clone 144(m144) 8 h prior to infection or left untreated (untd). Paired, two-tailed t-test. P-values<0.05 are plotted. Untreated n=5, treated n=6 biologically independent mice examined over 2 independent experiments for all time points. Individual values for each mouse are plotted. i. Schematic of experimental design of post-infection mAb treatment. MISTRG6-hACE2 mice received a mixed cocktail of monoclonal antibodies clone 135 (m135) and clone 144(m144) at 20mg/kg or left untreated (untd). Early treatment groups were treated 11 h post-infection and late treatment 35 h post-infection. Monoclonal recombinant antibodies (mAbs) used in this study were cloned from the convalescent patients (whose plasma was used prophylactically before) and had high neutralizing activity against SARS-CoV-2 in vitro and in vivo in mouse adapted SARS-CoV-2 infection^68,72. j. Viral titers measured by PFU in homogenized lung tissue following early or late therapeutic mAb at 4dpi using Vero E6 (standard assay) or Vero-ACE2+TMPRSS2+ (sensitive assay) cells. Treated n=4 biologically independent mice examined over 2 independent experiments for measurement with Vero-ACE2+TMPRSS2+ (sensitive assay) cells. Individual values for each mouse are plotted. k. Frequencies of human T cells within human CD45+ cells in spleens of MISTRG6-hACE2 mice that received early, late or no treatment of monoclonal antibody mix. Unpaired, two-tailed t-test. Untreated n=7, treated n=5 biologically independent mice examined over 3 independent experiments for all time points. Pooled, infection-matched mice are presented. P-values<0.05 are plotted. Individual values representing each mouse are plotted. Individual values for each mouse are plotted. l. Number of CD16+ human monocytes in lungs of treated and untreated mice at 4dpi. Mice were either MISTRG6-hACE2 mice that received a mixed cocktail of monoclonal antibodies clone 135 (m135) and clone 144(m144) or left untreated (untd). Early treatment group was treated 11 h post-infection (hpi) and late treatment 35 hpi. Unpaired, two-tailed t-test. Untreated n=7, treated n=5 biologically independent mice examined over 3 independent experiments for all time points. Pooled, infection-matched mice are presented. P-values<0.05 are plotted. Individual values representing each mouse are plotted. m. Number of human macrophages in lungs of treated and untreated mice at 4dpi. MISTRG6-hACE2 mice that received a mixed cocktail of monoclonal antibodies clone 135 (m135) and clone 144(m144) or left untreated (untd). Early treatment group was treated 11 h post-infection and late treatment 35 h post-infection. Unpaired, two-tailed t-test. Untreated n=7, treated n=5 biologically independent mice examined over 3 independent experiments for all time points. Pooled, infection-matched mice are presented. P-values<0.05 are plotted. Individual values representing each mouse are plotted.

Extended Data Fig. 5 (matched to Fig. 6).

a. Schematic of experimental design of dexamethasone treatment: SARS-CoV-2 infected MISTRG6-hACE2 mice were either treated with dexamethasone on days 7, 8, 9 post-infection or left untreated. Mice were analyzed either at 14 dpi or 28 dpi. Mice were treated intraperitoneally (i.p.) with dexamethasone at 10mg/kg dose. b. Representative flow cytometry plots of Ly6G expressing SSC^hi cells (mouse neutrophils) within the mouse immune cell population (mouse CD45+) in the BAL of dexamethasone treated or control untreated mice. Representative of 4 biologically independent mice examined over at least 2 independent experiments. c. HLA-DR expression on lung T cells 28 dpi in dexamethasone treated or control mice. Representative of 3 biologically independent mice examined over at least 2 independent experiments. d. Representative flow cytometry plots of Surface IgG and CD19 expression on human immune cells gated on hCD45+ cells in lungs of untreated or dexamethasone treated mice at 28 dpi. N=3 of biologically independent mice examined over at least 2 independent experiments. e. Frequencies of IgM+ and IgG+ B cells in lungs of untreated or dexamethasone treated mice at 28 dpi. Mice were treated with dexamethasone on days 7,8,9 post-infection. N=3 biologically independent mice examined over at least 2 independent experiments. Unpaired t-test, two-tailed. f. Viral RNA in the lung homogenates of dexamethasone treated or control untreated mice at 28dpi. Mann-Whitney, two-tailed test. N=5 biologically independent mice examined over 3 independent experiments. g. Schematic of experimental design of SARS-CoV-2 infected MISTRG6-hACE2 mice either treated with dexamethasone on days 3, 4, 5 dpi or left untreated. h. CD206 and CD68 expression in lung human immune cells in mice treated with dexamethasone or left untreated at 7dpi. CD206hi+CD68+ cells are alveolar macrophages. Treated N=4, untreated N=6 biologically independent mice examined over 2 independent experiments. Alveolar macrophages are marked by high CD206 expression. Inflammatory macrophages are enriched in the CD206 negative population which also express CD86 (not shown).

Supplementary information

Supplementary Figs. 1 and 2.

Supplementary Fig. 1 (matched to Fig. 1): a, Comparison of viral titers measured by PFU in homogenized lung tissue at 4 dpi using Vero E6 cells or Vero-ACE2⁺TMPRSS2⁺ cells. Vero-ACE2⁺TMPRSS2⁺ cells overexpress ACE2 (refs. ^20,21) and TMPRSS2 (ref. ²²), two receptors involved in SARS-CoV-2 infection. Vero-ACE2⁺TMPRSS2⁺ cells allow a more sensitive cell-culture-based assay for quantification of SARS-CoV-2 titers. b, Viral titers measured by PFU using Vero-ACE2⁺TMPRSS2⁺ cells and viral RNA in homogenized lung tissue at 2, 4, 7, 14, 28 and 35 dpi. Significance in viral titers (compared to 4 dpi) was determined by two-tailed unpaired t-test. Viral titers: 2 dpi, n = 2; 4 dpi, n = 5; 7 dpi, n = 2; 14 dpi, n = 8; 28 dpi, n = 7; and 35 dpi, n = 6 biologically independent mice examined over at least three independent experiment for 14–35 dpi. Significance in viral RNA (compared to 4 dpi) was determined by the Mann–Whitney two-tailed test. Viral RNA: 2 dpi, n = 4; 4 dpi, n = 22; 7 dpi, n = 6; 14 dpi, n = 7; 28 dpi, n = 8; 35 dpi, n = 6 biologically independent mice examined over at least three independent experiment for 14–35 dpi. Individual values for each mouse and means are plotted. c, Weight change of individual mice and reconstituted MISTRG6-hACE2 mice on 2, 4, 7, 10, 14, 21, 28 and 35 dpi. 2 dpi, n = 37; 4 dpi, n = 37; 7 dpi, n = 23; 10 dpi, n =16; 14 dpi, n = 15; 21 dpi, n = 12; 28 dpi, n = 10; 35 dpi, n = 4 biologically independent mice examined over at least three independent experiment for 14–35 dpi. Individual values for each mouse are plotted. d, Representative H&E images (×2 and ×10 magnification) of all lungs used in comparison of histopathological scores (Fig. 1e). Each pair of images (×2 and ×10 magnification) represents an individual mouse. At 2 dpi, MISTRG6-hACE2 mice showed prominent collection of proteinaceous debris in the alveolar spaces, which is the hallmark of the exudative phase of human COVID-19 pulmonary pathology. The debris was a mixture of necrotic pneumocytes and vasculopathy associated fibrin. Lungs at 4 dpi appeared to transition into the organizing phase, highlighted by infiltration of inter- and intra-alveolar space by lymphocytes macrophages and fibroblasts (4 dpi in Fig. 1e,f and Supplementary Fig 2d). The inter-alveolar spaces continued to expand between 14 dpi and 28 dpi by increasing accumulation of immune and stromal cells. Uninfected, n = 6; 2 dpi, n = 7; 4 dpi, n = 16; 7 dpi, n = 5; 14 dpi, n = 10; 28 dpi, n =6; 35 dpi, n = 4 biologically independent mice examined over at least four independent experiments. e, Histopathology by H&E staining of infected lungs at higher magnifications at 28 dpi. The inter-alveolar spaces continued to expand by increasing accumulation of immune and stromal cells. Representative of six mice examined over at least four independent experiments. f, Representative images (×40) of trichrome staining of all lungs used in comparison of histopathological scores (Fig. 1f). Each image set represents an individual mouse. During the 14–28-dpi period, there was increasing deposition of collagen, highlighted by thickening of collagen bundles in the lung parenchyma and continued obliteration of alveolar architecture (Fig. 1e,f and Supplementary Fig. 1d–g). These changes highlighted the transition from organizing phase to fibrotic phase. The architectural remodeling and collagen deposition was more pronounced at 28 dpi and 35 dpi, resulting in replacement of normal alveolar architecture by dilated cystic spaces, separated by thick interstitial tissue bands (also in Fig. 1f). Uninfected, n = 6; 4 dpi, n = 4; 14 dpi, n = 9; 28 dpi, n = 7, 35 dpi, n = 5 biologically independent mice examined over at least three independent experiments. Supplementary Fig. 2 (matched to Fig. 2): a, Representative gating strategy of human immune cells in the lungs of infected MISTRG-hACE2 mice. b, Representative gating strategy and quantification of mouse immune cells and mouse neutrophils in the lungs of infected MISTRG-hACE2 mice. Mouse immune cells: uninfected, n = 6; 2 dpi, n = 6; 4 dpi, n = 17; 7 dpi, n = 5; 14 dpi, n = 6; 28 dpi, n = 6, biologically independent mice examined over at least three independent experiments. Mouse neutrophils: uninfected, n = 3; 2 dpi, n = 4; 4 dpi, n = 11; 7 dpi, n = 3; 14 dpi, n = 6; 28 dpi, n = 7 biologically independent mice examined over at least three independent experiments. Individual values for each mouse and means are plotted. c, Immunohistochemistry and H&E staining of matched lungs. Lungs were stained against human CD68, CD3 and CD20 staining. Uninfected, n = 2; 4 dpi, n = 3; 14 dpi, n = 3; 28 dpi, n = 3, biologically independent mice examined over at least two independent experiments. d, Side-by-side quantification of T cells (CD3), B cells (CD20) and macrophages (CD68) by immunohistochemistry and flow cytometry of matched samples normalized to human immune cells (T+B+macrophages). B cells were characterized with combined CD19 and CD20 expression by flow cytometry. B cells quantified by CD20 expression alone might be underreported by immunohistochemistry compared to the quantitation by flow cytometry. Uninfected, n = 2; 4 dpi, n = 3; 14 dpi, n = 3; 28 dpi, n = 3 biologically independent mice examined over at least two independent experiments. Means with s.d. are presented. e, Representative gating strategy for mature (CCR7⁺) and activated (CD83⁺) pDCs. Representative of n = 5 biologically independent mice examined over two independent experiments.

Reporting Summary

Supplementary Table 1

DEGs in SARS-CoV-2-infected lungs. Normalized counts and fold changes in infected lungs are compared with respect to uninfected mice, and adjusted P values are presented for 516 mouse and 285 human differentially regulated genes. The following numbers of replicates were used in the study: Human genes: uninfected, n = 5; 2 dpi, n = 4; 4 dpi, n = 7; 7 dpi, n = 3; 14 dpi, n = 2; 28 dpi, n = 2 biologically independent mice examined over at least two independent experiments for all time points. The following numbers of replicates were used in the study: Human genes: uninfected, n = 5; 2 dpi, n = 4; 4 dpi, n = 7; 7 dpi, n = 3; 14 dpi, n = 2; 28 dpi, n = 2 biologically independent mice examined over at least two independent experiments for all time points. Mouse genes: uninfected, n = 5; 2 dpi, n = 4; 4 dpi, n = 9; 7 dpi, n = 3; 14 dpi, n = 4; 28 dpi, n = 2 biologically independent mice examined over at least two independent experiments for all time points. Whole tissue lung samples with less than 1% human reads were excluded from human DEG analysis (pre-established criterion). Differential expression analysis was performed with DESeq2, and statistical significance was deemed using the Wald test.

Supplementary Table 2

Pathway analysis of mouse and human DEGs in SARS-CoV-2-infected lungs. Gene Ontology, gene set enrichment analysis and Ingenuity pathway analysis platforms were used. The following numbers of replicates were used in the study: Human genes: uninfected, n = 5; 2 dpi, n = 4; 4 dpi, n = 7; 7 dpi, n = 3; 14 dpi, n = 2; 28 dpi, n = 2 biologically independent mice examined over at least two independent experiments for all time points. Mouse genes: uninfected, n = 5; 2 dpi, n = 4; 4 dpi, n = 9; 7 dpi, n = 3; 14 dpi, n = 4; 28 dpi, n = 2 biologically independent mice examined over at least two independent experiments for all time points. Whole tissue lung samples with less than 1% human reads were excluded from human DEG analysis (pre-established criterion). Differential expression analysis was performed with DESeq2, and statistical significance was deemed using the Wald test.

Supplementary Table 3

Distribution of type I ISGs in SARS-CoV-2-infected lungs during the course of infection. Normalized counts for type I ISGs within human DEGs are presented. The following numbers of replicates were used in the study: human genes: uninfected, n = 5; 2 dpi, n = 4; 4 dpi, n = 7; 7 dpi, n = 3; 14 dpi, n = 2; 28 dpi, n = 2 biologically independent mice examined over at least two independent experiments for all time points. Mouse genes: uninfected, n = 5; 2 dpi, n = 4; 4 dpi, n = 9; 7 dpi, n = 3; 14 dpi, n = 4; 28 dpi, n = 2 biologically independent mice examined over at least two independent experiments for all time points. Whole tissue lung samples with less than 1% human reads were excluded from human DEG analysis (pre-established criterion). Differential expression analysis was performed with DESeq2, and statistical significance was deemed using the Wald test. For ISG identification, http://www.interferome.org was used with parameters -In Vivo and Homo sapiens -fold change up 2 and down 2.

Supplementary Table 4

Genes that correlate with systemic lymphopenia. Top 50 negatively and positively correlating genes were ranked using Pearson’s r comparing normalized counts for each gene and the corresponding splenic T cell frequency out of total human immune cells. The following numbers of replicates were used in the comparison. Human genes: uninfected, n = 5; 2 dpi, n = 4; 4 dpi, n = 7; 7 dpi, n = 3; 14 dpi, n = 2; 28 dpi, n = 2 biologically independent mice examined over at least two independent experiments for all time points. Whole tissue lung samples with less than 1% human reads were excluded from correlation analysis (pre-established criterion).

Rights and permissions

Reprints and permissions

About this article

Cite this article

Sefik, E., Israelow, B., Mirza, H. et al. A humanized mouse model of chronic COVID-19. Nat Biotechnol 40, 906–920 (2022). https://doi.org/10.1038/s41587-021-01155-4

Download citation

Received: 26 February 2021
Accepted: 05 November 2021
Published: 17 December 2021
Issue Date: June 2022
DOI: https://doi.org/10.1038/s41587-021-01155-4

This article is cited by

SARS-CoV-2 immunity in animal models
- Zhao Chen
- Yaochang Yuan
- Jing Sun
Cellular & Molecular Immunology (2024)
Emerging and reemerging infectious diseases: global trends and new strategies for their prevention and control
- Shen Wang
- Wujian Li
- Sandra Chiu
Signal Transduction and Targeted Therapy (2024)
Comorbidities, multimorbidity and COVID-19
- Clark D. Russell
- Nazir I. Lone
- J. Kenneth Baillie
Nature Medicine (2023)
Accelerating antiviral drug discovery: lessons from COVID-19
- Annette von Delft
- Matthew D. Hall
- Alpha A. Lee
Nature Reviews Drug Discovery (2023)
Animal models to study the neurological manifestations of the post-COVID-19 condition
- Carla Usai
- Lourdes Mateu
- Joaquim Segalés
Lab Animal (2023)

Subjects

Abstract

Similar content being viewed by others

Main

Results

MISTRG6 humanized mice that transiently express hACE2 can be infected with SARS-CoV-2

Inflammatory macrophages and monocytes characterize the immune landscape in SARS-CoV-2-infected MISTRG6-hACE2 mice

Adaptive immune responses of patients with COVID-19 are recapitulated in MISTRG6-hACE2 mice infected with SARS-CoV-2

SARS-CoV-2 lungs sustain ISGs and display SLE-like features

Human monoclonal recombinant antibodies as prophylactic and therapeutic interventions affect disease outcome

Accurate timing of corticosteroids is necessary to balance viral clearance and prevent immunopathology

Discussion

Methods

Mice

Transplantation of human CD34+ hematopoietic progenitor cells into mice

AAV-hACE2 administration

SARS-CoV-2 infection

Therapeutics

Viral titers

Viral RNA analysis

Histology

Isolation of cells and flow cytometry

Adoptive T cell transfer

Antibodies

Bulk whole tissue lung RNA sequencing

ELISA

Gene expression

Single-cell RNA sequencing (10x Genomics)

Statistical analysis

Reporting Summary

Data availability

References

Acknowledgements

Author information

Authors and Affiliations

Contributions

Corresponding author

Ethics declarations

Competing interests

Additional information

Extended data

Extended Data Fig. 2 (matched to Fig. 3).

Extended Data Fig. 3 (matched to Fig. 4).

Extended Data Fig. 4 (matched to Fig. 5).

Extended Data Fig. 5 (matched to Fig. 6).

Supplementary information

Rights and permissions

About this article

Cite this article

Share this article

This article is cited by

Search

Quick links

Transplantation of human CD34⁺ hematopoietic progenitor cells into mice