A humanised ACE2, TMPRSS2, and FCGRT mouse model reveals the protective efficacy of anti-receptor binding domain antibodies elicited by SARS-CoV-2 hybrid immunity

Abstract

Background: Despite the importance of vaccination- and infection-elicited antibodies (Abs) to SARS-CoV-2 immunity, current mouse models do not fully capture the dynamics of Ab-mediated immunity in vivo, including potential contributions of the neonatal Fc receptor, encoded by FCGRT.

Methods: We generated triple knock-in (TKI) mice expressing human ACE2, TMPRSS2, and FCGRT; and evaluated the protective efficacy of anti-SARS-CoV-2 monoclonal Abs (mAbs) and plasma from individuals with immunity elicited by vaccination alone plus SARS-CoV-2 infection-induced (hybrid) immunity.

Findings: A human anti-SARS-CoV-2 mAb harbouring a half-life-extending mutation, but not the wild-type mAb, exhibited prolonged half-life in TKI mice and protected against lung infection with Omicron BA.2, validating the utility of these mice for evaluating therapeutic Abs. Pooled plasma from individuals with hybrid immunity to Delta, but not from vaccinated-only individuals, cleared infectious Delta from the lungs of TKI mice (P < 0.01), even though the two plasma pools had similar Delta-binding and -neutralising Ab titres in vitro. Similarly, plasma from individuals with hybrid Omicron BA.1/2 immunity, but not hybrid Delta immunity, decreased lung infection (P < 0.05) with BA.5 in TKI mice, despite the plasma pools having comparable BA.5-binding and -neutralising titres in vitro. Depletion of receptor-binding domain-targeting Abs from hybrid immune plasma abrogated their protection against infection.

Interpretation: These results demonstrate the utility of TKI mice as a tool for the development of anti-SARS-CoV-2 mAb therapeutics, show that in vitro neutralisation assays do not accurately predict in vivo protection, and highlight the importance of hybrid immunity for eliciting protective anti-receptor-binding domain Abs.

Funding: This work was funded by grants from the e-Asia Joint Research Program (N10A650706 and N10A660577 to MLM, in collaboration with SS); the NIH (U19 AI142790-02S1 to EOS and SS and R44 AI157900 to KJ); the GHR Foundation (to SS and EOS); the Overton family (to SS and EOS); the Arvin Gottlieb Foundation (to SS and EOS), the Prebys Foundation (to SS); and the American Association of Immunologists Fellowship Program for Career Reentry (to FASB).

Keywords: Delta; Hybrid immunity; Monoclonal antibodies; Mouse model; Omicron BA.1; Omicron BA.2; Omicron BA.5; Polyclonal antibodies; SARS-CoV-2; hACE2; hFCGRT; hTMPRSS2.

Fig. 1

hACE2-hTMPRSS2-hFCGRT TKI mice support SARS-CoV-2 Delta infection. (a, b) RT-PCR analysis of human (h) or mouse (m) ACE2, TMPRSS2, and FCGRT mRNA in tissues from (a) TKI and (b) C57BL/6J (B6) mice. Mouse glyceraldehyde 3-phosphate dehydrogenase (mGapdh) mRNA was probed as a loading control. (c) Western blot analysis of human or mouse proteins in the lung and nasal turbinate (nt) tissues from TKI and B6 mice. (d) Experimental protocol for the data shown in (e). TKI mice were inoculated intranasally with SARS-CoV-2 Delta (50 × 10³ PFU), and nt and lung tissues were harvested on days 2, 3, and 4 post-infection. (e) RT-qPCR analysis of Delta E genomic (g) and 7a subgenomic (sg) RNA, and plaque assay of infectious virus in the nt and/or lungs on days 2, 3, and 4 post-infection. (f) Experimental protocol for the data shown in (g). TKI mice were inoculated intranasally with SARS-CoV-2 Delta (0.5, 5, or 50 × 10³ PFU), and nt, lung, and brain tissues were harvested on day 2 post-infection. (g) RT-qPCR analysis of Delta 7a sgRNA and plaque assay of infectious virus in the nt, lung, and brain tissues on day 2 post-infection. Data are presented as the mean ± SEM of 5 mice/group in (e) and 7–8 mice/group in (g). Circles represent individual mice and dotted lines indicate the limit of detection. Group means were compared by the Kruskal–Wallis test. ∗P < 0.05, ∗∗P < 0.01.

Fig. 2

Functionality of hFcRn in hACE2-TMPRSS2-FCGRT TKI mice. (a) Pharmacokinetics of 1C3-YTE mAb, 1C3-WT mAb, and polyclonal human IgG (pAb) in TKI and DKI mice. Groups of treatment-naïve mice were injected intraperitoneally with 1 mg/kg mAbs or pAbs and bled at 3, 24, 72, 168, 336, and 504 h post-dose. Each mouse was bled at a maximum of two time points. Data are presented as the mean ± SEM of 3–8 mice/group/time point, pooled from two independent experiments. Group means at day 21 were compared by mixed-effect analysis and Šídák's correction. The open and filled bars represent DKI and TKI, respectively. (b) Half-life of 1C3-YTE, 1C3-WT, and pAbs in TKI and DKI mice. Group means were compared by mixed-effect analysis and Šídák's correction. The open and filled bars represent DKI and TKI, respectively. (c) Experimental protocol for the data shown in (d and e). TKI and DKI mice were injected intraperitoneally (d) or intranasally (e) with 0.5 mg/kg of 1C3-YTE, 1C3-WT, or an anti-SARS-CoV-1 hIgG (negative control). One day later, mice were inoculated intranasally with Omicron BA.2 (10⁴ PFU in 30 μl) and nasal turbinate (nt) and lungs were harvested on day 2 post-infection. (d) and (e) RT-qPCR analysis of Omicron BA.2 subgenomic (sg) RNA and plaque assay of infectious virus in the nt and/or lungs of TKI and DKI mice on day 2 post-infection. Data are presented as the mean ± SEM of 6–12 mice/group, pooled from two independent experiments. Circles represent individual mice and dotted lines indicate the limit of detection. Group means were compared by the Kruskal–Wallis and Dunn's multiple comparisons tests. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

Fig. 3

Differential efficacy of human plasma Abs elicited by SARS-CoV-2 vaccination or vaccination plus infection in protecting TKI mice against Delta infection. (a) Timelines of SARS-CoV-2 vaccination and infection for plasma donors. Plasma samples were collected on the indicated days from the three donor groups: 2-dose + inf^Delta (n = 7; two doses of BNT162b2 vaccine followed by infection with Delta); 3-dose (n = 10; three doses of BNT162b2 vaccine); and 3-dose + inf^Omicron (n = 5; three doses of BNT162b2 vaccine followed by infection with Omicron BA.1 or BA.2). Values in green represent the median ± interquartile range (IQR) days between the 2nd or 3rd vaccination and symptom onset; values in red represent the median ± IQR days between natural exposure or vaccination and plasma collection. Infection windows for 2-dose + inf^Delta and 3-dose + inf^Omicron groups were deduced from known symptom onset dates. (b) Anti-Delta spike (S) protein IgG ELISA, anti-Delta receptor-binding domain (RBD) IgG ELISA, pVSV-S^Delta neutralisation assay, or PRNT^Delta of plasma collected from donors in the 2-dose + inf^Delta group or the 3-dose group. Data are presented as the geometric mean ± error from one experiment. White and red circles represent individual and pooled samples, respectively. (c) Experimental protocol for the data shown in (d). TKI mice were injected intraperitoneally with 500 μl of pooled plasma from the 2-dose + inf^Delta or 3-dose groups or from SARS-CoV-2-naïve and unvaccinated donors (control). One day later, mice were inoculated intranasally with Delta (10³ PFU), and nasal turbinate (nt), lungs, and blood samples were harvested on day 2 post-infection. (d) RT-qPCR analysis of Delta subgenomic (sg) RNA and plaque assay of infectious virus in the nt and/or lung tissues of TKI mice on day 2 post-infection. Circles represent individual mice and dotted lines indicate the limit of detection. Data are presented as the mean ± SEM of 8–10 mice/group, pooled from two independent experiments. Group means were compared by the Mann–Whitney test (b) or the unpaired Kruskal–Wallis and Dunn's multiple comparison tests (d). ∗P < 0.05, ∗∗P < 0.01.

Fig. 4

Differential efficacy of human plasma Abs elicited by SARS-CoV-2 vaccination and Delta or Omicron BA.1/2 infection in protecting TKI mice against Omicron BA.5 infection. (a) Anti-Omicron BA.5 spike (S) protein IgG ELISA, anti-Omicron BA.5 receptor-binding domain (RBD) IgG ELISA, pVSV-S^BA.5 neutralisation assay, or PRNT^BA.5 of plasma collected from donors in the 2-dose + inf^Delta or 3-dose + inf^Omicron groups indicated in Fig. 3a. Data are presented as the geometric mean ± error from one experiment. White and red circles represent individual and pooled plasma samples, respectively. (b) Experimental protocol for the data shown in (c). TKI mice were injected intraperitoneally with 500 μl of pooled plasma from the 2-dose + inf^Delta or 3-dose groups or from SARS-CoV-2-naïve and unvaccinated donors (control). One day later, mice were inoculated intranasally with Omicron BA.5 (5 × 10⁴ PFU), and nasal turbinate (nt) and lungs were harvested on day 2 post-infection. (c) RT-qPCR analysis of Omicron BA.5 subgenomic (sg) RNA and plaque assay of infectious virus in the nt and/or lung tissues of TKI mice on day 2 post-infection. Data are presented as the mean ± SD of 7–8 mice/group, pooled from two independent experiments. Circles represent individual mice and dotted lines indicate the limit of detection. Group means were compared by Kruskal–Wallis and Dunn's multiple comparison tests. ∗P < 0.05.

Fig. 5

Human SARS-CoV-2 RBD-binding Abs from vaccinated and Omicron-infected donors protect TKI mice against Delta infection. (a) PRNT of 3-dose + inf^Omicron plasma samples after depletion of RBD-binding Abs or mock-depletion. Data are presented as the mean of two duplicates. (b) Experimental protocol for the data shown in (c). TKI mice were injected intraperitoneally with 500 μl of pooled plasma from SARS-CoV-2-naïve and unvaccinated donors (control) or 3-dose + inf^Omicron plasma depleted of RBD-binding Abs or mock depleted. One day later, mice were inoculated intranasally with Delta (10³ PFU), and nasal turbinate (nt) and lungs were harvested on day 2 post-infection. (c) RT-qPCR analysis of Delta subgenomic (sg) RNA and plaque assay of infectious virus in the nt and/or lung tissues of TKI mice on day 2 post-infection. Data are presented as the mean ± SD of 10–12 mice/group, pooled from two independent experiments. Circles represent individual mice and dotted lines indicate the limit of detection. Group means were compared by unpaired Kruskal–Wallis and Dunn's multiple comparison tests. ∗P < 0.05, ∗∗∗P < 0.001.