High binding potency overcomes the requirement of Fc effector functions for broadly reactive anti-alphavirus antibodies

Abstract

Alphaviruses are emerging public health threats. Broadly reactive anti-alphavirus monoclonal antibodies (mAbs) have been shown to be protective in mouse models of infection. However, the antibody characteristics that promote in vivo efficacy and dependency on Fc effector functions remain ill defined. Here, we used two vaccine-elicited, broadly reactive, anti-alphavirus mAbs, SKT05 and SKT20, to establish correlates of mAb-mediated protection during Venezuelan equine encephalitis virus (VEEV) challenge. SKT20 required Fc effector functions to prevent lethality. The necessity of Fc effector functions for SKT20 was dose dependent and related to mAb binding potency and pseudovirus neutralization rather than epitope specificity. In contrast, survival mediated by SKT05 when given prophylactically was independent of Fc effector functions and rather was linked to early viral control through egress inhibition. However, control of virus replication and spread with SKT05 at later time points was Fc dependent. Therapeutic administration of SKT05 required Fc effector functions only at 3 days after infection. These findings extended to additional in vivo infection models with alternative VEEV subtypes and with chikungunya virus. Collectively, this study identified binding potency and pseudoviral neutralization as correlates for in vivo efficacy of mAbs and demonstrated that Fc-dependent mechanisms can be leveraged for development of therapeutic mAbs against emerging alphaviruses.

Figure 1.. Broadly reactive anti-alphavirus mAbs protect against lethal VEEV challenge independent of neutralization.

(A to C) BALB/c mice were administered 200 µg of SKT05, SKV09, a positive control antibody (1A3B-7), or isotype control antibody at 1 h post-aerosol challenge with 10³ PFU of VEEV strain Trinidad Donkey (TrD). Mice were monitored for 14 days for survival (A), weight loss (B), and clinical score (C) (n = 10 mice/group; two independent experiments; mean ±SEM). (D to F) C3H/HeN mice received 200 µg of indicated mAb 1 day prior to intranasal challenge with 10⁷ FFU of VEEV strain TC-83 (n = 10 mice/group; two independent experiments). (D) Survival was followed for 14 dpi in one cohort. Brains were harvested from a separate group of mice at 1, 3, and 6 dpi and viral RNA (E) or infectious virus (F) was determined by RT-qPCR or FFA, respectively. Bars represent the median. (G) Focus reduction neutralization test (FRNT) with indicated mAbs and TC-83 is representative of two independent experiments (mean ± SD) conducted in duplicate. Non-linear regression without constraints was used to determine the IC₅₀ values. (H and I) C3H/HeN mice were pre-treated with 200 µg of indicated mAb one day before infection with TC-83. (H) At 5 dpi, skulls with brains intact were harvested, fixed, then decalcified before paraffin embedding and sectioning. Representative images of sagittal skull and brain sections (100x) stained with anti-mouse IgG. Images are representative of two independent experiments (n = 3 mice/group at each time-point; n = 3 mice for naïve group). The black oval indicates areas of colocalization of mouse IgG, parenchymal loss, and cellular uptake of mouse IgG. Scale bars denote 100 µm. (I) Brains were harvested at 6 dpi and the concentration of cytokines and chemokines were determined using a Bio-plex assay. The box and whisker plots indicate the min to max of all representative data points and the center line represents the median of the interquartile range (IQR) between 25–75% (n = 10 mice/group; two independent experiments). The min and max concentration of the analyte detected in the naïve brain homogenate is represented by the shaded regions within the graphs. If naïve samples were at the limit of detection (LOD), only a dotted line is shown. Statistical significance was determined by a Log-rank test (A, D), one-way ANOVA with a Dunnett’s post-test of area under the curve (AUC) analysis from 0–5 dpi comparing each treatment to the control group (B, C), and Kruskal-Wallis with a Dunn’s post-test comparing all groups at each time point (E, F, I). The dotted line indicates starting body weight (B) or LOD (E, F, I).

Figure 2.. Fc effector functions are required for SKT20 protection but dispensable for SKT05-mediated survival when provided in prophylaxis to VEEV challenge.

(A) Binding of LALA-PG variants to VEEV VLP was assessed by ELISA. The data represents one independent experiment. (B) Binding of indicated mAbs to purified human (h) FcγRI was assessed by ELISA. The data is represented as the mean ± SD of two independent experiments performed in duplicate. (C to F) SKT05 variants (C, E) or SKT20 variants (D, F) (200 µg) were administered to C3H/HeN mice 1 day prior to inoculation with TC-83. A cohort of mice were followed for weight loss (mean ± SD) (C and D) and survival (E and F) for 13–14 days (n = 8 mice/group; two independent experiments). Statistical significance was determined by a one-way ANOVA with a Tukey’s post-test of AUC analysis from 0–7 dpi comparing all groups (C and D) or a Log-rank test (E and F). (G) Hematoxylin and eosin (H&E) staining of skulls from separate mice with brains intact at 5 dpi (40x and 100x). The yellow arrows indicate meningitis, and the circles indicate perivascular cuffing. The black boxes indicate regions magnified and shown in 100x images. Data are representative images of two independent experiments (n = mice 6 to 8/group). Scale bars for 40x images denote 200 µm and 100x images denote 100 µm. (H) Slides were blinded prior to pathological scoring of H&E images in (G). The average score is shown from two independent experiments (n = mice 6 to 8/group).

Figure 3.. SKT05 limits neuroinvasion and spread into the brain through inhibition of viral egress.

(A to C) C3H/HeN mice were treated with 200 µg of indicated mAb one day prior to infection with TC-83. (A) At 1 and 6 dpi, viral RNA loads were determined in the brains by RT-qPCR (n = mice 8/group; two independent experiments; Kruskal-Wallis with a Dunn’s post-test comparing all groups at each time point. The median is represented, and the dotted line indicates the LOD of the assay. RNA in situ hybridization of skulls with brains intact using VEEV-specific probes (red) at 1 (B) and 5 dpi (C). The red arrow points out focal viral RNA staining. Data are representative images of two independent experiments (n = mice 6 to 8/group per time point; n = 2 mice for naïve group). Scale bars for 40x images denote 200 µm and 100x images denote 100 µm. (D) A single-cycle, viral entry inhibition assay was performed on LUHMES. The relative infection compared to a no mAb control was determined by flow cytometry at 4 hpi (represented mean ± SD of two independent experiments conducted in duplicate). The IC₅₀ was determined by non-linear regression. (E) Viral egress inhibition by indicated mAbs was evaluated in LUHMES. Supernatants were collected at 1 and 6 hpi to quantify viral RNA by RT-qPCR (mean ± SD of three independent experiments performed in duplicate; two-way ANOVA with Dunnett’s post-test comparing all groups to the isotype control).

Figure 4.. SKT20 alters the pro-inflammatory response and immune cell infiltrates in the brain through Fc effector functions.

C3H/HeN mice were administered an isotype control, SKT20, or SKT20 LALA-PG (200 µg) at 1 day prior to infection with TC-83. (A and B) At 6 dpi, brains were harvested. (A) Chemokine and cytokine concentrations were determined using a Bio-plex assay. The box and whisker plots indicate the min to max of all representative data points and the center line represents the median of the IQR between 25–75% (n = 10 mice/group, two independent experiments). The shaded bar represents the min and max concentration for naïve animals. The dotted line represents the LOD of the assay. If naïve samples were at the LOD, only a dotted line is shown. (B) Viral RNA in brain homogenate was determined by RT-qPCR (n = 6 to 8 mice/group; two independent experiments; bars represent median). For (A and B), statistical significance was determined by Kruskal-Wallis with a Dunn’s post-test comparing all groups. (C and D) At 6 dpi, brains were collected and digested. Single cell suspensions were stained, and the percentage of activated (CD69⁺) CD4⁺ and CD8⁺ T cells (C) and activated (MHCII⁺) monocytes and macrophages (D) was determined by flow cytometry (n = 6 to 8 mice/group; bars represent the mean; one-way ANOVA with Tukey’s post-test).

Figure 5.. SKT05 can induce Fc effector functions.

(A and B) Antibody dependent cellular cytotoxicity (ADCC) reporter assay with TC-83 infected Vero cells pre-incubated with mAbs. Engagement of mAb-infected cells with Jurkat cells expressing mFcyRIV (A) or hFcyRIV-V158 (B) is represented as fold-induction over no mAb. One representative experiment is displayed from two independent experiments. (C and D) Antibody dependent complement deposition (ADCD) was determined by the presence of complement component 3 (C3) on the surface of TC-83-infected Vero cells pre-incubated with mAbs (C) and the corresponding percentage of dead cells (D) as determined by flow cytometry [two independent experiments performed in duplicate; mean ±SD; one-way ANOVA with Sidak’s post-test of AUC analysis from all dilutions comparing WT mAbs to the control mAb (colored stars) or SKT05 to SKT20 (black stars)]. FITC, fluorescein isothiocyanate; MFI, median fluorescence intensity. (E to H) C3H/HeN mice were administered 200 µg of indicated mAb one day prior to infection with TC-83. Mice administered SKT20, SKT20 K322A, or a control were followed for weight loss (E) and survival (F) for 14 days (n = 6 mice/group; two independent experiments). Mice provided SKT05, SKT05 K322A, or a control mAb were followed for weight loss (G) for six days and then euthanized to determine viral RNA loads in the brain (H) (n = 6 mice/group; two independent experiments). Statistical significance was determined by a one-way ANOVA with a Dunnett’s post-test of AUC analysis from 0–7 dpi (E) or 0–6 dpi (G), a Log-rank test (F), or a Kruskal-Wallis with a Dunn’s post-test (H). In (E and G), data shows mean ± SD. In (H), bars represent the median.

Figure 6.. Fc effector function necessity is dependent on binding potency rather than epitope specificity.

(A) mAb avidity to VEEV p62-E1 (subtype IAB) was assessed using a chaotropic ELISA (mean ± SD; representative data from one of two independent experiments conducted in duplicate). The avidity index (AI) is the average of two independent experiments. (B) Binding of indicated mAbs to the surface of live TC-83-infected Vero cells was determined by flow cytometry (median ± SD of two independent experiments performed in duplicate). EC₅₀ values were determined by nonlinear regression of log transformed data. (C to H) C3H/HeN mice were administered 200, 60, 20, or 2 µg of SKT05, SKT20, or 200 µg of a control mAb one day prior to infection with TC-83. One cohort of mice were followed for weight loss (C, F) and survival (D, G) for 14 days. (E, H) Viral loads were determined in the brains of a second cohort of mice at 5 dpi (n = 8 mice/group; two independent experiments). In (C and F), data shows mean ± SEM. In (E and H), bars represent the median. Statistical significance was determined by a one-way ANOVA with a Dunnett’s post-test of AUC analysis from 0–8 dpi comparing each group to the 200 µg group (C and F), Log-rank test compared to 200 µg group (D and G), or Kruskal-Wallis with a Dunn’s post-test comparing each group to the 200 µg dose (E and H). (I to M) C3H/HeN mice were administered 200 µg of SKT14, SKT14 LALA-PG, or a control mAb one day prior to infection with TC-83. One cohort of mice were followed for weight loss (I and L) and survival (J and M). Additional mice were euthanized at 5 dpi to determine viral RNA loads (K) in brain tissues by RT-qPCR (data related to SKT05 and SKT20 is the same as the 200 µg doses of in E and H) (n = 8 mice/group; two independent experiments). In (I and L), data shows mean ± SEM. In (K), bars represent the median. Statistical significance was determined by a Log-rank test compared to the SKT14 group (J and M) or Kruskal-Wallis with a Dunn’s post-test compared to the control group (K).

Figure 7.. The requirement of Fc effector functions for mAb-mediated protection is dose-dependent.

(A to D) C3H/HeN mice were administered 200 µg or 2 mg of SKT14 LALA-PG or SKT20 LALA-PG, 200 µg of SKT14 or SKT20, or 2 mg of a control mAb one day prior to infection with TC-83. Mice were followed for weight loss (A, C) and survival (B, D) for 14 days (n = 8 mice/group; two independent experiments). (E and F) C3H/HeN mice were administered 200 µg, 60 µg, 20 µg of SKT05 LALA-PG, or 200 µg of a control mAb one day prior to infection with TC-83. Mice were followed for weight loss (E) and survival (F) for 14 days (n = 6 mice/group; two independent experiments). In (A, C, and E), data shows mean ± SEM. Statistical significance was determined by a one-way ANOVA with a Dunnett’s post-test of AUC analysis from 0–6 dpi (A, C, and E) or a Log-rank test (B, D, and F) comparing each group to either 200 µg SKT14, 200 µg SKT20, or 200 µg SKT05 LALA-PG.

Figure 8.. SKT05 reduces clinical disease independent of Fc effector functions during therapeutic administration and against other alphaviruses.

(A and B) C3H/HeN mice were infected with TC-83 then administered 200 µg of indicated mAb at 1, 2, or 3 dpi. Mice were followed for survival (A) and weight loss (B) for 14 days (n = 8 mice/group; two independent experiments). In (B), data shows mean ± SEM. Statistical significance was determined by a Log-rank test between SKT05 to SKT05 LALA-PG at each time of administration (A), one-way ANOVA with a Tukey’s post-test of AUC analysis from 0–7 dpi (B). No significant difference was observed for any comparison in (B). (C to J) BALB/C mice were treated with 200 µg of SKT05, SKT05 LALA-PG, a positive control mAb (1A3B-7), or a control mAb at 1 h post-subcutaneous challenge with 10³ PFU of VEEV INH-9813 (IC subtype; C to F) or ZPC-738 (ID subtype; G to J). Mice were followed for 14 days for survival (C, G), weight loss (D, H), and clinical score (E, I) (n = 10 mice/group; two independent experiments). Viremia was assessed on 1, 2, and 3 dpi (F, J) (n = 3 mice/group; one independent experiment; bars represent median). In (D, E, H, and I), data shows mean ± SEM. Statistical significance was determined by a Log-rank test between treatment groups to control (#; symbol color) and SKT05 to SKT05 LALA-PG (C and G), one-way ANOVA with a Sidak’s post-test of AUC analysis from 0–5 dpi (D and E) or 0–6 dpi (H and I) comparing treatment groups to control mAb (#; symbol color) and SKT05 to SKT05 LALA-PG, or Kruskal Wallis with a Dunn’s post-test at each time point between treatment groups to control and SKT05 to SKT05 LALA-PG at each time point (F and J). (K to N) C57BL/6J mice were treated with 200 µg of indicated mAb 1 day prior to infection with 10³ FFU of CHIKV in the ipsilateral footpad. (K) One cohort of mice were monitored for foot swelling. Data shows the mean ± SEM of three independent experiments (n = 14 mice/group). Statistical significance was determined by one-way ANOVA with Tukey’s post-test of AUC analysis from all time points. (L to N) Additional mice were euthanized at 1 or 3 dpi and viral RNA load was determined by RT-qPCR in the ipsilateral ankle (L), contralateral ankle (M), and serum (N) (n = 8 mice/group; two independent experiments; Kruskal-Wallis with a Dunn’s post-test). Bars represent the median.