Chikungunya viruses that escape monoclonal antibody therapy are clinically attenuated, stable, and not purified in mosquitoes

Abstract

Chikungunya virus (CHIKV) is a reemerging mosquito-transmitted alphavirus that causes epidemics of debilitating polyarthritis in humans. A prior study identified two anti-CHIKV monoclonal antibodies ([MAbs] CHK-152 and CHK-166) against the E2 and E1 structural proteins, which had therapeutic efficacy in immunocompetent and immunocompromised mice. Combination MAb therapy was required as administration of a single MAb resulted in the rapid selection of neutralization escape variants and treatment failure in mice. Here, we initially evaluated the efficacy of combination MAb therapy in a nonhuman primate model of CHIKV infection. Treatment of rhesus macaques with CHK-152 and CHK-166 reduced viral spread and infection in distant tissue sites and also neutralized reservoirs of infectious virus. Escape viruses were not detected in the residual viral RNA present in tissues and organs of rhesus macaques. To evaluate the possible significance of MAb resistance, we engineered neutralization escape variant viruses (E1-K61T, E2-D59N, and the double mutant E1-K61T E2-D59N) that conferred resistance to CHK-152 and CHK-166 and tested them for fitness in mosquito cells, mammalian cells, mice, and Aedes albopictus mosquitoes. In both cell culture and mosquitoes, the mutant viruses grew equivalently and did not revert to wild-type (WT) sequence. All escape variants showed evidence of mild clinical attenuation, with decreased musculoskeletal disease at early times after infection in WT mice and a prolonged survival time in immunocompromised Ifnar1(-/-) mice. Unexpectedly, this was not associated with decreased infectivity, and consensus sequencing from tissues revealed no evidence of reversion or compensatory mutations. Competition studies with CHIKV WT also revealed no fitness compromise of the double mutant (E1-K61T E2-D59N) neutralization escape variant in WT mice. Collectively, our study suggests that neutralization escape viruses selected during combination MAb therapy with CHK-152 plus CHK-166 retain fitness, cause less severe clinical disease, and likely would not be purified during the enzootic cycle.

Importance: Chikungunya virus (CHIKV) causes explosive epidemics of acute and chronic arthritis in humans in Africa, the Indian subcontinent, and Southeast Asia and recently has spread to the New World. As there are no approved vaccines or therapies for human use, the possibility of CHIKV-induced debilitating disease is high in many parts of the world. To this end, our laboratory recently generated a combination monoclonal antibody therapy that aborted lethal and arthritogenic disease in wild-type and immunocompromised mice when administered as a single dose several days after infection. In this study, we show the efficacy of the antibody combination in nonhuman primates and also evaluate the significance of possible neutralization escape mutations in mosquito and mammalian cells, mice, and Aedes albopictus vector mosquitoes. Our experiments show that escape viruses from combination antibody therapy cause less severe CHIKV clinical disease, retain fitness, and likely would not be purified by mosquito vectors.

FIG 1

Anti-CHIKV therapy reduces viremia in rhesus macaques. Rhesus macaques were infected with 10⁷ PFU of CHIKV-LR and injected intravenously at 1 and 3 dpi with anti-CHIKV antibodies or a control WNV MAb (15 mg/kg). At 0, 1, 2, 3, 4, 5, and 7 dpi, peripheral blood samples were collected and processed for plasma. (A) Plasma anti-CHIKV mouse MAb titers were measured by ELISA (n = 3/group). The numbers under each treatment refer to the individual rhesus macaques used in this study. (B) Levels of CHIKV in plasma were measured by limiting-dilution plaque assay. Antibodies directed against CHIKV but not WNV lowered viremia to undetectable levels at 2 dpi, and this effect was maintained until the study endpoint (7 dpi) (n = 3 per group; P < 0.01 at 2 dpi, two-way ANOVA).

FIG 2

Anti-CHIKV therapy reduces CHIKV dissemination in rhesus macaques. Rhesus macaques were treated at 1 and 3 dpi with anti-CHIKV antibodies or control anti-WNV antibody (15 mg/kg). Necropsy occurred at 7 dpi, and monkey tissue samples were processed for total RNA by the TRIzol method (n = 6 per group). Quantitative RT-PCR was used to detect CHIKV loads in joints, muscle, and other organs and lymph nodes. Virus dissemination to peripheral muscles and joints (leg), organs (lung and kidney), and lymph nodes ([LN] inguinal and mesenteric) was greatly reduced by anti-CHIKV treatment. Viral loads in the arm muscles and joints (site of infection) as well as spleen and draining lymph nodes (axillary) were not affected by the antibody treatment. Asterisks indicate statistically significant differences (*, P < 0.05; **, P < 0.01; ***, P < 0.001) as judged by a Mann-Whitney test.

FIG 3

Growth kinetics of WT and mutant CHIKV in insect and mammalian cells. (A to D) C6/36 A. albopictus cells were infected with P0 BHK cell-derived CHIKV WT (A), CHIKV E1-K61T (B), CHIKV E2-D59N (C), or CHIKV E1-K61T E2-D59N (D) virus. For some experiments, CHIKVs (WT or mutants) were preincubated with 10 μg/ml of CHK-166, CHK-152, or both CHK-166 and CHK-152 for 1 h at 37°C, as indicated. Virus or virus-MAb complexes were added to C6/36 cells for 1 h at 37°C. After samples were washed to remove free virus and antibody, supernatants were harvested at 1, 24, 48, and 72 h postinfection for titration by FFU assay. The results are the average of three independent experiments performed in triplicate. Statistically significant differences are indicated (*, P < 0.05; **, P < 0.01, ***, P < 0.001). For the mutant viruses, none of the differences with and without MAb treatment were significantly different. (E to H) African green monkey Vero cells were infected with P0 C6/36 cell-derived CHIKV WT (E), CHIKV E1-K61T (F), CHIKV E2-D59N (G), or CHIKV E1-K61T E2-D59N (H) virus. For some experiments, CHIKVs (WT or mutants) were preincubated with 10 μg/ml of CHK-166, CHK-152, or both CHK-166 and CHK-152 for 1 h at 37°C, as indicated. Virus or virus-MAb complexes were added to Vero cells for 1 h at 37°C. After samples were washed to remove free virus and antibody, supernatants were harvested at 1, 12, 24, and 36 h postinfection for titration by FFU assay. The results are the average of three independent experiments performed in triplicate. Statistically significant differences are indicated (*, P < 0.05; **, P < 0.01, ***, P < 0.001). For the mutant viruses, none of the differences with and without MAb treatment were significantly different.

FIG 4

Virulence of WT and mutant CHIKV strains in immunocompromised Ifnar1^−/− or Rag1^−/− mice. (A to D) Six- to eight-week-old Ifnar1^−/− mice were passively transferred saline or 50 μg of each antibody, CHK-166 and CHK-152, via an intraperitoneal injection 1 day before infection with 10 FFU of CHIKV WT (A), CHIKV E1-K61T (B), CHIKV E2-D59N (C), or CHIKV E1-K61T E2-D59N (D) virus via a subcutaneous route. The survival curves were constructed from data of at least two independent experiments with between 7 and 10 mice per group. (E to J) Six week-old Ifnar1^−/− mice were infected with 10 FFU of CHIKV WT, CHIKV E1-K61T, CHIKV E2-D59N, or CHIKV E1-K61T E2-D59N (DM) virus via a subcutaneous route. At day 2 after infection, serum, spleen, liver, muscle (right leg), and brain were harvested from individual mice for virus titration by focus-forming assay. The data are the average ± standard deviation from 4 to 6 mice per group. The differences in viral burden did not attain statistical significance (P > 0.1, Mann-Whitney test). (K) Rag1^−/− mice were infected with 10³ FFU of CHIKV WT or CHIKV E1-K61T E2-D59N (DM) via a subcutaneous route. Blood samples were obtained from individual animals at days 14 and 28 after infection, and CHIKV RNA was measured by qRT-PCR. No statistically significant differences were observed.

FIG 5

Virulence of WT and neutralization escape variant viruses in WT C57BL/6 mice. Three-week-old WT C57BL/6 mice were infected with 10³ FFU of CHIKV WT, CHIKV E1-K61T, CHIKV E2-D59N, or CHIKV E1-K61T E2-D59N virus via a subcutaneous route. (A) At days 3 and 7 after infection, measurements of joint swelling were made. Statistically significant differences are indicated (***, P < 0.001). (B and C) At day 3, the ipsilateral (left) and contralateral (right) ankle to the site of injection were harvested, and virus was titrated by qRT-PCR (B) or plaque assay (C, for serum). (D to U) At day 3 after infection with CHIKV WT or CHIKV E1-K61T E2-D59N virus, serum was harvested and processed for the indicated cytokines and chemokines. The results are displayed as a scatter plot from three independent experiments from a total of nine mice. Statistically significant differences are indicated (*, P < 0.05). (V) At day 28, the indicated tissues were harvested, and yield was analyzed by qRT-PCR. None of the values obtained with the mutant viruses was statistically different from that of the WT virus. These results were pooled from two independent experiments with a total of 4 to 8 mice per group.