Antibody potency, effector function, and combinations in protection and therapy for SARS-CoV-2 infection in vivo

Abstract

SARS-CoV-2, the causative agent of COVID-19, has been responsible for over 42 million infections and 1 million deaths since its emergence in December 2019. There are few therapeutic options and no approved vaccines. Here, we examine the properties of highly potent human monoclonal antibodies (hu-mAbs) in a Syrian hamster model of SARS-CoV-2 and in a mouse-adapted model of SARS-CoV-2 infection (SARS-CoV-2 MA). Antibody combinations were effective for prevention and in therapy when administered early. However, in vitro antibody neutralization potency did not uniformly correlate with in vivo protection, and some hu-mAbs were more protective in combination in vivo. Analysis of antibody Fc regions revealed that binding to activating Fc receptors contributes to optimal protection against SARS-CoV-2 MA. The data indicate that intact effector function can affect hu-mAb protective activity and that in vivo testing is required to establish optimal hu-mAb combinations for COVID-19 prevention.

Figure 1.

Antibody potency against the SARS-CoV-2 MA spike. (A) Diagram of the MA SARS-CoV-2 pseudovirus luciferase assay. SARS-CoV-2 S-MA pseudotyped HIV-1 particles carrying the nanoluc gene are used to infect muAce2-expressing HT1080 cells, which will express nanoluc luciferase upon infection, while wtS pseudotyped particles are unable to infect muAce2-expressing cells. Yellow stars indicate visible light released by the nanoluciferase (nLUC) reaction. (B) RLU reads from lysates of muAce2 and human (hu) Ace2–expressing HT1080 cells infected with increasing amounts of SARS-CoV-2 S-MA and wtS pseudovirus, as well as nonpseudotyped control virus (w/o S). Data are mean ± standard deviation of triplicates. One representative experiment is shown. nluc, nanoluc. (C) % neutralization measured for cell lysates of HT1080_muAce2 cells 48 h after infection with SARS-CoV-2 S-MA pseudovirus in the presence of increasing concentrations of mAbs. n = 8 samples and 1 isotype control. Data points are shown as circles and curve fits as lines. Data are mean of duplicates, and one representative experiment is shown. ctr, control. (D) IC₉₀ values detected in the SARS-CoV-2 S-MA pseudovirus neutralization assay (IC₉₀-murine) plotted against those detected in the wtS SARS-CoV-2 pseudovirus neutralization assay (IC₉₀-human). R² = 0.8095; P < 0.02178. Mean values of at least two experiments are shown. ns, not significant.

Figure 2.

In vivo potency does not uniformly correlate with in vitro potency. (A) SARS-CoV-2 MA lung titer following antibody prophylaxis. Antibodies (8 mg/kg) were delivered intraperitoneally 12 h before infection with 10⁵ PFU of SARS-CoV-2 MA. Combined data from two independent experiments are shown. All groups are n = 10 mice/group except for C119 and C135-LS. The line is at the geometric mean, and each symbol represents the titer for a single animal. Asterisks indicate statistical differences compared with isotype control by Kruskal-Wallis test with a Dunn’s multiple comparison test (****, P < 0.0001; ***, P = 0.0003; **, P = 0.007–0.004). (B) PFU/lobe values plotted against IC₉₀ values detected in the SARS-CoV-2 S-MA pseudovirus neutralization assay. R² = 0.1585; P = 0.788. The R and P values in A and B were determined by two-tailed Spearman correlations. ctr, control; LOD, limit of detection; ns, not significant.

Figure 3.

The variable requirement of Fc-effector function for SARS-CoV-2 neutralization. (A) Antibody potency curves for WT and GRLR mutant antibodies. The % neutralization for cell lysates of HT1080_muAce2 cells 48 h after infection with SARS-CoV-2 S-MA pseudovirus in the presence of increasing concentrations of WT (solid lines) or GRLR-modified (dashed lines) mAbs. Data points are shown as closed (WT) or open (GRLR) circles, and corresponding curve fits are shown as continuous (WT) or dashed (GRLR) lines. Data are mean of duplicates, and one representative experiment is shown. ctr, control. (B) IC₉₀ values for antibodies shown in A. Bars represent mean values of two experiments (shown as open circles). (C) SARS-CoV-2 MA lung titer following antibody prophylaxis. WT or GRLR antibodies (8 mg/kg) were delivered intraperitoneally 12 h before infection with 10⁵ PFU of SARS-CoV-2 MA. Combined data from two independent experiments are shown, and all groups are n = 10 mice/group. The line is at the geometric mean, and each symbol represents the titer for a single animal. Asterisks indicate statistical differences compared with isotype control by Mann-Whitney test (**, P < 0.004; ***, P = 0.0001). (D) Antibody potency curves for WT and Fc mutant antibodies performed and displayed similarly to those in A. (E) IC₉₀ values for the Fc mutant antibodies shown in D. Bars represent mean values of two experiments (shown as open circles). (F) SARS-CoV-2 MA lung titer following antibody prophylaxis with chimeric mAb comprised of the variable domains of C104 grafted onto constant regions of mouse IgG1, IgG2b, and IgG_D265A. WT C104– and isotype control antibody–treated groups were controls. Antibodies (8 mg/kg) were delivered intraperitoneally 12 h before infection with 10⁵ PFU of SARS-CoV-2 MA. One independent experiment is shown. All groups are n = 5 mice/group. The line is at the geometric mean, and each symbol represents the titer for a single animal. Asterisks indicate statistical differences compared with isotype control by one-way ANOVA with Dunnett’s multiple comparison test or Mann-Whitney test (*, P = 0.022; ***, P = 0.0001–0.0003; ****, P < 0.0001). LOD, limit of detection; ns, not significant.

Figure 4.

Antibody combinations increase in vivo potency. SARS-CoV-2 MA lung titer following antibody prophylaxis with isotype control (16 mg/kg) or combinations of C135-LS + C121-LS or C135-LS + C144-LS mixed at a ratio of 1:1 for combined dose levels of 16, 5.3, or 1.8 mg/kg. Antibodies were delivered intraperitoneally 12 h before infection with 10⁵ PFU of SARS-CoV-2 MA. Single antibodies C121-LS, C135-LS, or C144-LS were similarly administered at 8 mg/kg (n = 8/antibody). Combined data from two independent experiments are shown. For 16-mg/kg groups, n = 14 or 15 mice, and all other groups were 9 or 10 mice/group. The line is at the geometric mean, and each symbol represents the titer for a single animal. Asterisks indicate statistical differences compared with isotype control by one-way ANOVA with a Dunnett’s multiple comparison test (****, P < 0.0001) or Kruskal-Wallis test with a Tukey’s multiple comparison test (*, P = 0.02–0.04; ***, P = 0.0002; ****, P < 0.0001). LOD, limit of detection.

Figure 5.

Prevention and therapy with C135-LS + C144-LS antibody combination in Syrian hamsters. Hamsters were infected intranasally with 2.6 × 10⁴ PFU SARS-CoV-2, and viral lung titers 3 d after infection were determined by plaque assay on Vero-E6 cells. Antibodies were administered either 24 h before (prophylactic) or 12 h after (treatment) infection. The line is at the geometric mean, and each symbol represents the titer for a single animal. Dotted line indicates the limit of detection. Asterisks indicate statistical differences compared with isotype control by one-way ANOVA with a Dunnett’s multiple comparison test (**, P < 0.01). LOD, limit of detection.