Therapeutic administration of a recombinant human monoclonal antibody reduces the severity of chikungunya virus disease in rhesus macaques

Abstract

Chikungunya virus (CHIKV) is a mosquito-borne virus that causes a febrile syndrome in humans associated with acute and chronic debilitating joint and muscle pain. Currently no licensed vaccines or therapeutics are available to prevent or treat CHIKV infections. We recently isolated a panel of potently neutralizing human monoclonal antibodies (mAbs), one (4N12) of which exhibited prophylactic and post-exposure therapeutic activity against CHIKV in immunocompromised mice. Here, we describe the development of an engineered CHIKV mAb, designated SVIR001, that has similar antigen binding and neutralization profiles to its parent, 4N12. Because therapeutic administration of SVIR001 in immunocompetent mice significantly reduced viral load in joint tissues, we evaluated its efficacy in a rhesus macaque model of CHIKV infection. Rhesus macaques that were treated after infection with SVIR001 showed rapid elimination of viremia and less severe joint infiltration and disease compared to animals treated with SVIR002, an isotype control mAb. SVIR001 reduced viral burden at the site of infection and at distant sites and also diminished the numbers of activated innate immune cells and levels of pro-inflammatory cytokines and chemokines. SVIR001 therapy; however, did not substantively reduce the induction of CHIKV-specific B or T cell responses. Collectively, these results show promising therapeutic activity of a human anti-CHIKV mAb in rhesus macaques and provide proof-of-principle for its possible use in humans to treat active CHIKV infections.

Fig 1. Characterization of neutralization, escape, and therapeutic efficacy of SVIR001 in mice.

(A) SVIR001 and parent 4N12 mAbs were evaluated by neutralization assay in Vero cells. Virus was pre-incubated with indicated concentrations of mAb for 1 h and used to inoculate Vero cells. Data are reported as the relative infection normalized to a no mAb control. Results are representative of one of three independent experiments performed in duplicate. (B) The escape variant virus, which was generated by serial passage in the presence of SVIR001, was subjected to neutralization with SVIR001 and compared to a virus passaged in the absence of SVIR001 (control passaged virus). Results are representative of one of three independent experiments performed in duplicate. (C-D) Confirmation of SVIR001 escape phenotype with engineered six nucleotide deletion (E2 del 734–739). WT or deletion CHIKV-181/25 or CHIKV-LR viruses were incubated with indicated mAb for 1 h. Virus-mAb mixture was added to Vero cells. Data was normalized to a no mAb control. Each graph is two independent experiments done in triplicate and the mean ± SD are shown. (E-G) WT mice were inoculated subcutaneously with 10³ FFU of CHIKV-LR in the footpad and treated with 50 μg (E-F) or 300 μg (F-G) of anti-CHIKV mAb SVIR001 or control mAb SVIR002 at 1 (E), 3 (F), or 3 and 10 (G) dpi. At 3 (E), 5 (F), or 28 (G) dpi, virus was quantified by infectious focus (E-F) or qRT-PCR (G) assays from the ipsilateral and contralateral ankle to determine therapeutic efficacy. The median value is shown with the limit of detection indicated by the dotted line. Statistics were calculated on log-transformed data using the Mann-Whitney test (**, P < 0.01, ***, P < 0.001, ****, P < 0.0001). Each data point represents an individual animal. The data (E-G) were pooled from 2 independent experiments.

Fig 2. Plasma antibody levels and viral load following CHIKV mAb therapy.

Rhesus macaques were inoculated subcutaneously in both arms with 1 x 10⁷ PFU of CHIKV LR. On day 1 and 3 after infection, rhesus macaques were administered 5 or 15 mg/kg SVIR001 (human anti-CHIKV mAb) or 15 mg/kg SVIR002 (human anti-lysozyme mAb), n = 4/group. Blood was collected on 0, 1, 2, 3, 4, 5, and 7 dpi. (A) Human mAb concentration in the plasma was measured by ELISA with lysozyme or CHIKV virions, and mAb concentration was calculated using a standard curve. Statistical significance was calculated using Tukey’s multiple comparison test (n = 4; ***, P < 0.0005, **, P < 0.01, *, P < 0.05). (B) Virus was quantified from plasma by qRT-PCR. Statistically significant differences are reported on the log-transformed data using Dunnett’s multiple comparison test (n = 4; ****, P < 0.0005).

Fig 3. Tissue viral load following CHIKV mAb therapy.

Animals were euthanized at day 7 post-infection, and viral RNA was isolated from tissues and quantified by qRT-PCR. The viral load in (A) arm joints and muscles, (B) leg joints and muscles, and (C) lymphoid tissues, heart and kidney are reported. Statistical significance was determined on the log-transformed data using Dunnett’s multiple comparison test, and multiplicity-adjusted P values are reported (n = 4; ** P < 0.005, * P < 0.05).

Fig 4. Histological images of joint-associated tissue from CHIKV-infected animals at 7 dpi.

At 7 dpi, joint-associated tissue from the wrist and finger of each animal were fixed, paraffin embedded, sectioned, and stained with hematoxylin and eosin. Shown are representative images of stained sections from joint-associated soft tissue of four animals treated with control antibody SVIR002 or CHIKV mAb SVIR001. In the SVIR002 treated animals, there was abundant inflammation surrounding multiple vessels. In the SVIR001 treated animals (5 or 15 mg/kg dose), there was limited or no perivascular inflammation.

Fig 5. CHIKV mAb therapy reduced plasma cytokines and chemokine activation at 2 dpi.

Heat map comparing average fold-change in plasma cytokine profiles for NHP treated inoculated with CHIKV and treated with control antibody SVIR002 or CHIKV mAb SVIR001. Clustering was performed using the Broad Institute’s webtool Morpheus. A 29-plex-cytokine magnetic bead assay was performed on plasma from rhesus macaques isolated at day 0, 1, 2, 3, 5, and 7 post-infection.

Fig 6. CHIKV mAb therapy reduced activation of peripheral blood monocytes/macrophages, Myeloid DCs, and NK cells.

Total peripheral blood mononuclear cells from 0, 1, 2, 3, 4 and 5 dpi were stained with antibodies directed against HLA-DR, CD14, CD169, CD11c, CD20, CD3, CD8, and CD16 to assess changes in the activation of macrophage/monocyte, DC, and NK cell subsets. The percent of activated, CD169⁺ (A) monocyte/macrophages, (B) plasmacytoid DCs, (C) myeloid DCs, or (D) NK cells within the population are reported (n = 4).

Fig 7. CHIKV mAb treatment did not cause significant changes in CD4⁺ or CD8⁺ T cell proliferation.

Rhesus macaques were inoculated with CHIKV and treated with control antibody SVIR002 or CHIKV mAb SVIR001. Blood was drawn daily 0–7 dpi, and PBMCs were examined for proliferative responses of different (A-C) CD4⁺ and (D-F) CD8⁺ T cell subsets. T cell subsets were defined in S2 Fig as Naïve (NV), Central Memory (CM), and Effector Memory (EM). The Ki67⁺ proliferative status was plotted as a percentage of the total population. (G) IFNγ ELISpot analysis was performed on PBMCs from rhesus macaques at 7 dpi. PBMCs from animals treated with SVIR001 (5 mg/kg or 15 mg/kg) or SVIR002 (15 mg/kg) were stimulated with CHIKV peptide pools (10 μg/well), inactivated CHIKV (iCHIKV) (10 μg/well), or PMA/Ionomycin as a positive control. DMSO was used as a negative control to establish the baseline number of IFNγ-producing T cells for each animal. Spots were quantified on an AID ELISpot plate reader (n = 4/group).