Research ArticleANTIBODIES

Optimal therapeutic activity of monoclonal antibodies against chikungunya virus requires Fc-FcγR interaction on monocytes

See allHide authors and affiliations

Science Immunology  22 Feb 2019:
Vol. 4, Issue 32, eaav5062
DOI: 10.1126/sciimmunol.aav5062

Engaging monocytes to battle chikungunya

Antibody-binding receptors, including Fc receptors and complement receptors, play a central role in mediating antibody-dependent immune activation. Here, Fox et al. have examined the role of Fcγ receptors and complement component 1q (C1q) in meditating the therapeutic effects of monoclonal IgG antibodies targeting chikungunya virus. Using antibody engineering in conjunction with mouse strains lacking C1q or Fcγ receptors, they report that the therapeutic effects of these antibodies are dependent on expression of Fcγ receptors. Furthermore, by depleting distinct immune cell types, they found that engagement of Fc receptors on monocytes is central in driving antibody-dependent clearance of chikungunya virus.

Abstract

Chikungunya virus (CHIKV) is an emerging mosquito-borne virus that has caused explosive outbreaks worldwide. Although neutralizing monoclonal antibodies (mAbs) against CHIKV inhibit infection in animals, the contribution of Fc effector functions to protection remains unknown. Here, we evaluated the activity of therapeutic mAbs that had or lacked the ability to engage complement and Fcγ receptors (FcγR). When administered as post-exposure therapy in mice, the Fc effector functions of mAbs promoted virus clearance from infected cells and reduced joint swelling—results that were corroborated in antibody-treated transgenic animals lacking activating FcγR. The control of CHIKV infection by antibody-FcγR engagement was associated with an accelerated influx of monocytes. A series of immune cell depletions revealed that therapeutic mAbs required monocytes for efficient clearance of CHIKV infection. Overall, our study suggests that in mice, FcγR expression on monocytes is required for optimal therapeutic activity of antibodies against CHIKV and likely other related viruses.

INTRODUCTION

Chikungunya virus (CHIKV) is a mosquito-transmitted, single-stranded, positive-sense enveloped RNA virus belonging to the Alphavirus genus of the Togaviridae family. CHIKV was first isolated from an outbreak in Tanzania in 1952 and historically caused infections in Africa and Asia (1, 2). In 2013, CHIKV emerged in the Caribbean and spread into South and Central America, causing more than 1.7 million cases including locally acquired infections in Florida (3). Although CHIKV is rarely fatal, individuals infected with CHIKV develop fever, rash, myositis, and debilitating polyarthritis that can last for weeks. A subset of infected individuals suffers persistent joint pain and inflammation that endures for months to years (4, 5). Currently, there are no licensed vaccines or therapies to combat the acute or chronic phases of disease.

The CHIKV genome encodes four nonstructural proteins (nsP1 to nsP4) and five structural proteins (capsid, E3, E2, 6K, and E1) from two open reading frames. During infection, heterodimers of p62 (E3 and E2) and E1 assemble in the endoplasmic reticulum and form trimers. The E3 protein is cleaved by furin in the trans-Golgi compartment, and the E2-E1 heterodimer is transported to the plasma membrane where virion assembly and budding occur (6, 7). The mature virion displays 240 copies of the E2-E1 heterodimer assembled into 80 trimeric spikes (7, 8), which facilitate virus attachment and internalization through its cognate receptor, Mxra8 (911).

Multiple animal studies have highlighted the protective effect of antibodies against CHIKV infection. Passive transfer of CHIKV-immune human γ-globulin protects immunocompromised mice from lethal infection (12). Several candidate vaccines also elicit strongly neutralizing antibody responses (1316). Mouse and human anti-CHIKV monoclonal antibodies (mAbs) with potent neutralizing activity also have been identified; many inhibit CHIKV infection by blocking fusion or viral egress (1722). Therapeutic administration of these neutralizing mAbs increased survival in immunocompromised mice and reduced viral burden and disease in immunocompetent mice and nonhuman primates (17, 23, 24). Although antibodies can limit CHIKV disease, these studies did not address the contribution of antibody effector functions to protection. Because anti-CHIKV mAbs can interact with both free virus and the E2-E1 heterodimer on the cell surface, immune-mediated clearance mechanisms, such as antibody-dependent cellular cytotoxicity (ADCC), antibody-dependent cellular phagocytosis (ADCP), and complement activation, could contribute to clinical and virological protection.

Here, we evaluated the impact of Fc effector functions for antibody therapeutic efficacy in an immunocompetent mouse model of CHIKV-induced arthritis (25) that more closely approximates human disease compared with a lethal infection model in immunocompromised Ifnar1−/− mice (18). Intact versions of anti-CHIKV mAbs were compared with variants lacking appreciable binding to Fcγ receptors (FcγRs) and complement component 1q (C1q). Fc effector functions were required to reduce foot swelling, accelerate clearance of viral RNA, limit infiltration of immune cells, and reduce accumulation of proinflammatory chemokines. Fc effector function analysis showed that the intact mAbs facilitated phagocytosis of beads coated with CHIKV proteins by mouse monocytes and neutrophils. Cellular depletion studies in mice identified monocytes as a key cell type contributing to mAb-mediated clinical protection and clearance of viral RNA. Overall, these studies indicate that specific Fc effector functions of antibody and engagement of FcγR on monocytes are necessary for developing an optimal therapeutic response against CHIKV infection.

RESULTS

Fc effector functions of mAbs decrease CHIKV disease

We previously showed that combination therapy of two neutralizing anti-CHIKV mAbs [CHK-152 (anti-E2) and CHK-166 (anti-E1)] in the lethal, immunocompromised (Ifnar1−/− C57BL/6) mouse model for CHIKV improved survival and reduced resistance compared with single or isotype control mAbs (18, 24). Using this established mAb combination, we assessed protection in an immunocompetent wild-type (WT C57BL/6) murine model of CHIKV-induced musculoskeletal disease where animals develop a biphasic pattern of foot swelling, with separate peaks at 3 and 7 days post-infection (dpi) corresponding to tissue edema and immune cell infiltration, respectively (25, 26). We administered a single injection of the mouse CHK-152 + CHK-166 mAb [both immunoglobulin G2c (IgG2c) subclass] combination (250 μg each, 500 μg total) at 3 dpi, which correlates with the onset of disease signs. mAb treatment reduced the second phase of foot swelling, indicating that CHK-152 and CHK-166 could protect after the onset of disease signs (Fig. 1A).

Fig. 1 Clinical protection after mAb therapy.

(A) Four-week-old WT C57BL/6J mice were administered mouse anti-CHIKV mAbs [CHK-152 + CHK-166 (250 μg per mAb; 500 μg total)] or an isotype control (WNV E60; 500 μg) on 3 dpi with 103 FFU of CHIKV. Foot swelling was measured before infection and for 10 dpi (n = 8 per group, two experiments). Graphs show means ± SEM (***P < 0.001, ****P < 0.0001, two-way ANOVA with Sidak’s post-test). (B) mAbs (CHK-166 human IgG1, CHK-152 human IgG1, CHK-166 human IgG1 N297Q, and CHK-152 human IgG1 N297Q) were preincubated with 102 FFU of CHIKV and added to Vero cells for 18 hours. Viral foci were measured and compared with a no mAb control to determine relative infection. WNV hE16 is an isotype control mAb. Each graph represents the mean ± SD (two or three experiments). (C to F) Four-week-old mice were inoculated with CHIKV and then administered a (C and D) cocktail [CHK-152 + CHK-166 (250 μg per mAb; 500 μg total)] or (E and F) monotherapy [CHK-152 or CHK-166 (250 μg total)] of intact or N297Q variants of humanized mAbs or an isotype control (WNV hE16; 500 or 250 μg) on 3 dpi. (C, E, and F) Foot swelling was measured [(C) n = 8 to 10 per group, three experiments; (E) n = 7 per group, two experiments; (F) n = 7 per group, two experiments]. Graphs show means ± SEM (*intact versus isotype mAb, ¢intact versus N297Q, N297Q versus isotype mAb; two-way ANOVA with Tukey’s post-test: *P < 0.05, **P < 0.01, ****P < 0.0001, ¢P < 0.05, ¢¢P < 0.01, ¢¢¢P < 0.001, ¢¢¢¢P < 0.0001, P < 0.05). (D) Human IgG levels in the ipsilateral ankle were determined by ELISA at 5 dpi (n = 8 to 9 per group, two experiments). Bars indicate mean values (ns, not significant; Student’s t test).

To begin to evaluate the mechanism of antibody protection (e.g., neutralization versus effector function), we produced N297Q recombinant variants of CHK-152 and CHK-166 of a human IgG1 heavy chain subclass; this mutation abolishes an N-linked glycosylation site in the CH2 domain and creates an aglycosylated antibody that does not interact appreciably with FcγRs or C1q (27). The human IgG1 subclass was chosen because it has similar Fc and complement binding specificities as mouse IgG2c to mouse FcγRs (28). We first confirmed that intact and N297Q variants of CHK-152 and CHK-166 neutralized CHIKV equivalently in cell culture (Fig. 1B). Next, in vivo studies were performed by administering mAbs at 3 dpi and evaluating foot swelling. Whereas the intact human CHK-152 and CHK-166 mAbs reduced foot swelling markedly on days 6, 7, and 8 after infection, the N297Q mAb variants diminished swelling only slightly (only at 7 dpi) compared with the isotype control (Fig. 1C). To confirm that the differences in foot swelling were not due to disparate levels of intact and N297Q human antibody, we measured concentrations in the foot by enzyme-linked immunosorbent assay (ELISA) at 5 dpi; similar levels of antibodies were detected (Fig. 1D). To determine whether both mAbs in the cocktail required effector functions for therapeutic efficacy, we treated mice with either intact or N297Q CHK-152 or CHK-166 at 3 dpi and measured foot swelling. Whereas the individual intact IgG reduced foot swelling, the N297Q variants did not (Fig. 1, E and F). These results indicate that effector functions are needed to reduce foot swelling with anti-CHIKV mAb therapy.

To confirm that the requirement of Fc-mediated effector function was not unique to the CHK-152 and CHK-166 combination, we tested IM-CKV063, a neutralizing human mAb against CHIKV E2 (20, 21), and used different mutations [L234A-L235A (LALA)] in the Fc region that abrogate binding to FcγRs and C1q (29). We paired IM-CKV063 with humanized CHK-166, to limit virus escape, and then assessed therapeutic efficacy. Similar to results with CHK-152 and CHK-166, the intact IgG versions of IM-CKV063 and CHK-166 reduced foot swelling at 6 and 7 dpi compared with the isotype control or the LALA/N297Q group (fig. S1A). Administration of intact IM-CKV063 alone at 3 dpi also reduced foot swelling compared with the LALA variant or isotype control (fig. S1B), albeit to a lesser extent than CHK-152 or CHK-166 alone.

Fc effector functions of mAbs reduce CHIKV infection

High titers of CHIKV accumulate in the ipsilateral and contralateral ankles of WT mice by 3 dpi (19), the time of therapeutic mAb administration. We evaluated how the intact and N297Q versions of anti-CHIKV mAbs affected viral clearance, which could affect foot swelling. The ipsilateral and contralateral ankles were harvested 5, 7, or 28 dpi from mice receiving the intact or N297Q combination of humanized CHK-152 + CHK-166 therapy or an isotype control at 3 dpi, and amounts of CHIKV RNA were measured by quantitative reverse transcription polymerase chain reaction (qRT-PCR). The intact mAbs reduced the levels of CHIKV RNA in the ipsilateral ankle compared with the N297Q mAb–treated or isotype control mAb-treated mice at 5 and 7 dpi but not at 28 dpi, which reflects the persistent phase of infection (Fig. 2A) (30). The intact, but not N297Q, mAbs also reduced the amount of viral RNA in the contralateral ankle compared with the isotype control at both 7 and 28 dpi (fig. S2). Analogously, administration of the intact IM-CKV063 and CHK-166 mAb combination at 3 dpi reduced viral RNA levels at 7 dpi in the ipsilateral ankle compared with the LALA/N297Q or isotype control groups (fig. S1C).

Fig. 2 Intact mAb therapy reduces viral RNA levels.

WT mice were inoculated with 103 FFU of CHIKV and administered a cocktail of intact or N297Q variants of humanized anti-CHIKV mAbs or an isotype control mAb at 3 dpi. (A) Ipsilateral ankles were harvested at indicated days, and viral RNA was determined by qRT-PCR (5 and 7 dpi, n = 8 to 9 per group; 28 dpi, n = 7 to 9 per group, two experiments; one-way ANOVA with Tukey’s post-test: **P < 0.01, ***P < 0.001). (B) RNA in situ hybridization of ipsilateral feet using CHIKV-specific probes (479501) from tissues at 5 or 7 dpi. Images show low magnification (top; scale bars, 100 μm) and medium magnification (middle and bottom; scale bars, 100 μm) with a high-magnification inset (scale bars, 10 μm) (representative images from n = 6 per group, two experiments).

To corroborate these findings, we visualized CHIKV RNA by in situ hybridization in the ipsilateral foot 5 or 7 dpi. On 5 dpi, there was intense viral RNA staining in the muscle of the midfoot in mice receiving the N297Q cocktail or the isotype control; animals receiving the intact humanized CHK-152 + CHK-166 mAbs had less staining (Fig. 2B). By 7 dpi, CHIKV RNA was faintly detectable in tissue sections from mice treated with the intact mAbs, whereas abundant staining was apparent in the N297Q or isotype-treated groups (Fig. 2B). These results suggest that compared with mAbs lacking Fc effector functions, intact anti-CHIKV mAbs accelerate viral clearance and limit dissemination.

Fc effector function mediated immune cell recruitment during acute CHIKV disease

CD4+ T cells, macrophages, and monocyte recruitment contribute to CHIKV-associated musculoskeletal disease (26, 31, 32). Because the intact anti-CHIKV mAbs reduced the levels of viral RNA (and presumably viral nucleic acid pathogen-associated molecular patterns), we assessed whether treatment affected proinflammatory cytokine and chemokine expression and immune cell infiltration. CHIKV-infected mice were treated with intact or N297Q humanized mAbs or an isotype control at 3 dpi. After harvesting the ipsilateral ankle at 4 or 7 dpi, proinflammatory cytokines and chemokines were analyzed by a Bio-Plex assay, and the cellular infiltrate was analyzed by flow cytometry (fig. S3). At 4 dpi (1 day after mAb administration), the ipsilateral feet from mice treated with the intact mAbs had higher levels of proinflammatory chemokines and cytokines [e.g., CCL2, CCL3, CCL4, CCL5, CCL11, CXCL1, TNF-α (tumor necrosis factor–α), IFN-γ (interferon-γ), IL-1α (interleukin-1 α), IL-6, and IL-12 p70] and increased percentages of neutrophils compared with the N297Q mAb– or isotype-treated groups (Fig. 3, A to E, and table S1). Treatment with intact mAbs was also associated with greater numbers of CD45+ cells, monocytes, and neutrophils in the foot compared with N297Q mAb or isotype treatment at 4 dpi (Fig. 3F). However, the number of major histocompatibility complex (MHC) class II+ (activated) monocytes was not different between the groups (Fig. 3G).

Fig. 3 Fc effector functions of antibody affect infiltration of immune cells.

WT mice were inoculated with 103 FFU of CHIKV and administered a cocktail of intact or N297Q variants of humanized anti-CHIKV mAbs or an isotype control at 3 dpi. (A to D) Ipsilateral ankles were collected at 4 dpi and analyzed for chemokines (n = 9 to 10 per group, two experiments). Bars indicate median values (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, Kruskal-Wallis ANOVA with Dunn’s post-test). (E and F) At 4 dpi, cells from ipsilateral feet were stained for monocytes (CD11b+CD11cLy6GLy6C+), neutrophils (CD11b+CD11cLy6G+), NK cells (CD3NK1.1+), or (G) MHC class II+ monocytes (CD11b+CD11cLy6GLy6C+MHCII+) and analyzed by flow cytometry. (E) Percentage of indicated cell populations out of CD45+ cells and (F and G) number of viable cells of indicated populations [(E to G) n = 8 per group, two experiments]. Bars indicate mean values (*P < 0.05, **P < 0.01, ***P < 0.001, one-way ANOVA with Bonferroni’s post-test). (H to K) Ipsilateral ankles were collected at 7 dpi and analyzed for chemokines (n = 8 to 9 per group, two experiments). Bars indicate median values (*P < 0.05, **P < 0.01, Kruskal-Wallis ANOVA with Dunn’s post-test). (L and M) At 7 dpi, cells from ipsilateral feet were stained for monocytes, moDCs (CD11b+CD11c+Ly6GLy6C+MHCII+), neutrophils, NK cells, CD4+ T cells (CD3+CD4+), CD8+ T cells (CD3+CD8+), B cells (CD3CD19+), or (N) MHC class II+ monocytes and analyzed by flow cytometry. (L) Percentage of indicated cell populations out of CD45+ cells and (M and N) number of viable cells of indicated populations [(L to N) n = 6 to 10 per group, two or three experiments]. All comparisons in (L) to (N) were not statistically significant (one-way ANOVA with Bonferroni’s post-test). (O to S) Indicated cell numbers were compared between 4 and 7 dpi. Bars indicate mean ± SD (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, Student’s t test). The color of asterisks denotes significance for matching group (red, intact combo; blue, N297Q combo; black, isotype).

When proinflammatory cytokines and chemokines were analyzed in the ipsilateral feet at 7 dpi, the intact mAb-treated mice showed lower levels of CCL2, CCL3, CCL4, and CCL5 compared with animals treated with N297Q or isotype control mAbs (Fig. 3, H to K). Other cytokines and chemokines (e.g., CCL11, CXCL1, TNF-α, IFN-γ, IL-1α, IL-6, IL-10, IL-12 p40, and IL-12 p70) were measured but were not statistically different (table S1). Cellular analysis of the ipsilateral feet at 7 dpi revealed no significant differences in the frequency of the immune cell populations and a modest reduction in the number of total CD45+ cells, monocytes, monocyte-derived dendritic cells (moDCs), neutrophils, natural killer (NK) cells, CD4+ T cells, CD8+ T cells, B cells, and MHC class II+ monocytes with intact mAb treatment (Fig. 3, L to N). Consistent with this observation, pathological analysis of the ipsilateral foot at 7 dpi showed similar immune cell infiltrates in the dorsal midfoot and joint space compared with uninfected mice (fig. S4).

We analyzed the flux of proinflammatory cytokines and chemokines or immune cells in greater detail after antibody treatment by comparing their concentrations or cell numbers present at 4 and 7 dpi. For CCL2, CCL3, CCL4, CCL5, and IFN-γ, the intact mAb therapy reduced the levels from 4 to 7 dpi, whereas the N297Q and isotype-treated mice maintained or increased these proinflammatory molecules (fig. S5). When immune cells were compared between 4 and 7 dpi, all mAb treatment groups showed increased numbers of MHC class II+ monocytes (Fig. 3O). However, the intact mAb-treated mice sustained only a modest increase in CD45+ cells, monocytes, and NK cells compared with N297Q mAb or isotype-treated mice (Fig. 3, P to R). Moreover, the intact mAb-treated CHIKV-infected mice showed reduced numbers of neutrophils at 7 dpi compared with 4 dpi, whereas levels were increased slightly in animals treated with N297Q or isotype control mAbs (Fig. 3S). Collectively, these data suggest that virus engagement by intact mAbs promotes accelerated immune cell infiltration at the early phase of swelling (e.g., 4 dpi), but this ultimately results in reduced subsequent cell recruitment into the foot, which correlates with less swelling during the second phase (e.g., 7 dpi).

Clinical and virological protection of anti-CHIKV mAbs is mediated through Fc-FcγR interactions

The N297Q mutation in the human heavy chain abrogates binding to both FcγRs and C1q (27). To determine which of these effector molecules was associated with the clinical and virological phenotypes observed with mAb therapy, we performed analogous treatment studies at 3 dpi with intact humanized CHK-152 + CHK-166 in mice lacking C1q (C1q−/−) or the Fc receptor common γ chain (FcRγ−/−), which abrogates expression of all activating mouse FcγRs (FcγRI, FcγRIII, and FcγRIV). The intact CHIKV mAbs still reduced foot swelling in C1q−/− mice between days 6 and 8 after infection (Fig. 4A). In comparison, combination mAb therapy in FcRγ−/− mice did not diminish swelling at days 6 and 7 after infection, although some effect was observed at 8 dpi (Fig. 4B). At 5 dpi, the ipsilateral ankle was analyzed for viral RNA. Anti-CHIKV mAbs did not decrease viral RNA in FcRγ−/− mice, whereas levels were reduced (10-fold; P < 0.0001) in C1q−/− mice (Fig. 4C). To verify that this phenotype was not specific to heterologous humanized mAbs in mice, we repeated the studies using the mouse versions of CHK-152 and CHK-166, so the mAbs would be homologous to the host species. Similar to the humanized mAbs, the intact mouse mAbs reduced foot swelling in C1q−/− mice but not in FcRγ−/− mice at 7 dpi (Fig. 4, D and E). The intact mouse mAbs also decreased viral RNA in the WT and C1q−/− mice but not in the FcRγ−/− mice at 7 dpi (Fig. 4F). In addition, FcγR interactions were necessary to reduce viral RNA levels in other peripheral organs (spleen; Fig. 4G). These data suggest that antibody-mediated reductions in foot swelling and viral RNA levels principally are facilitated through activating FcγRs.

Fig. 4 Fc-FcγR interactions mediate clinical and virological protection.

C1q−/− (A, C, D, and F to H) or FcRγ−/− (B, C, E to G, and I) mice were inoculated with 103 FFU of CHIKV and administered a cocktail of humanized (A to C, H, and I) or mouse (D to G) intact anti-CHIKV mAbs or an isotype control mAb at 3 dpi. (A, B, D, and E) Foot swelling was measured before infection and for 10 or 7 dpi [(A) n = 9 to 10 per group; (B) n = 11 to 12 per group; (D) n = 6 to 8 per group; (E) n = 11 to 13 per group, two or three experiments]. Graphs show means ± SEM (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, two-way ANOVA with Sidak’s post-test). (C and F) The ipsilateral ankle (I. ankle) or (G) spleen was harvested 5 dpi (C) or 7 dpi (F and G), and levels of viral RNA were determined. (C, F, and G) Student’s t test (n = 6 to 13, two or three experiments; **P < 0.01, ***P < 0.001, ****P < 0.0001). (C) Tissue titers from WT mice are from Fig. 2A (5 dpi) and shown for comparison. (H and I) The ipsilateral feet from antibody-treated CHIKV-inoculated C1q−/− (H) or FcRγ−/− (I) mice were harvested at 4 dpi and stained for CD45+ cells, monocytes, neutrophils, and NK cells and analyzed for number of viable cells of indicated populations by flow cytometry [(H) n = 11 to 12 per group, three experiments; (I) n = 9 per group, three experiments]. Bars indicate mean values (*P < 0.05, unpaired t test).

We analyzed cellular recruitment into the ipsilateral foot of FcRγ−/− or C1q−/− mice to determine whether the increased cellularity observed in WT mice administered the intact mAbs (Fig. 3F) was promoted by activating FcγRs or C1q. CHIKV-infected FcRγ−/− or C1q−/− mice were treated with intact humanized CHK-152 and CHK-166 or an isotype control mAb at 3 dpi. The ipsilateral foot was collected at 4 dpi, and cellular infiltrates were analyzed (fig. S6). In the absence of C1q, treatment with the intact mAbs resulted in increased numbers of CD45+ cells and neutrophils and a trend toward more monocytes compared with the isotype control mAb–treated mice (Fig. 4H), similar to WT mice (Fig. 3, E and F). Similarly, in the absence of FcγRs, treatment with intact humanized CHK-152 and CHK-166 also resulted in increased numbers of neutrophils compared with the isotype control (Fig. 4I). In contrast, in FcRγ−/− mice treated with the intact CHK-152 and CHK-166 mAbs, we observed similar numbers of monocytes or CD45+ leukocytes in the ipsilateral foot relative to isotype control mAbs (Fig. 4I). Collectively, these data suggest that anti-CHIKV mAb engagement with FcγRs has a dominant effect on monocyte recruitment into the musculoskeletal tissues of the infected foot.

Monocytes are necessary for intact mAbs to reduce CHIKV RNA levels in mice

Fc engagement of FcγRs activates innate immune cells, including NK cells, neutrophils, and monocytes, to clear infected cells through ADCC or ADCP (33, 34). To begin to determine which innate immune cell contributed to the protection observed in vivo, we performed analysis of Fc effector functions with beads coated with recombinant CHIKV p62 (E3-E2)–E1 protein, mouse innate immune cells, and murine or humanized intact or N297Q variants of CHK-152 and CHK-166. Compared with the isotype control and N297Q mAbs, the mouse and humanized intact CHK-152 and CHK-166 mAbs facilitated greater phagocytosis of CHIKV antigen-coated beads in primary murine monocytes (Fig. 5, A and B) and neutrophils (Fig. 5, C and D). As expected, the murine mAbs were more potent than the human mAbs at promoting phagocytosis in mouse cells. These ex vivo data suggest that in mice, monocytes and/or neutrophils may be important for anti-CHIKV mAb effector function.

Fig. 5 Anti-CHIKV mAbs enhance phagocytosis with mouse monocytes and neutrophils.

Mouse, humanized intact, and humanized N297Q variants of anti-CHIKV mAbs or an isotype control were evaluated for mouse (A and B) monocyte-directed or (C and D) neutrophil-directed phagocytosis of CHIKV p62-E1–functionalized fluorescent beads (ADCP, antibody-directed cellular phagocytosis; ADNP, antibody-dependent neutrophil phagocytosis; n = 3 donors, two experiments). The dotted line indicates the no antibody control. Graphs show means ± SEM.

To determine which cell type was the primary mediator of mAb-dependent clinical protection and viral RNA reduction, using a series of antibodies, we depleted NK cells (NK1.1+) (fig. S7), monocytes and neutrophils (Ly6G+/Ly6C+) (Fig. 6, A to D), neutrophils alone (Ly6G+) (Fig. 6, E to H), or monocytes alone (CCR2+) (Fig. 6, I to L) throughout the course of CHIKV infection. As before, humanized anti-CHIKV mAbs or an isotype control was administered 3 dpi, and tissues were harvested at 7 dpi. Depletion was confirmed in peripheral blood on the day of harvest by flow cytometry (Fig. 6, A, B, E, F, I, and J, and fig. S7, A and B). Notably, when neutrophils were depleted, we observed a mild compensatory increase in circulating monocytes (Fig. 6F), and reciprocally, when monocytes were depleted, we detected an increase in circulating neutrophils (Fig. 6J), which has been reported previously in the context of CHIKV infection (35). Despite the large specific reduction in individual cell types, the flow cytometric analysis revealed that, for each depletion, a small (0.3 to 2%) residual population was present.

Fig. 6 Monocytes reduce CHIKV infection in the context of mAb therapy.

WT mice were inoculated with 103 FFU of CHIKV and administered a cocktail of (A to L) humanized or (M to P) mouse intact anti-CHIKV mAbs or an isotype control mAb at 3 dpi. (A to D) Monocytes and neutrophils, (E to H) neutrophils, or (I to P) monocytes were depleted using anti-Ly6G/Ly6C, anti-Ly6G, or anti-CCR2, respectively. (A and B) Monocyte and neutrophil, (E and F) neutrophil, or (I, J, M, and N) monocyte depletion was confirmed by flow cytometry analysis [(A) anti-Ly6G/Ly6C, (E) anti-Ly6G, or (I or M) anti-CCR2 in each set; the top is the specific cell depletion and the bottom is the nondepleting isotype]. (C, G, K, and O) Foot swelling was measured before infection and for 7 dpi [(C, G, K, and O) n = 6, two experiments]. Bars indicate mean ± SEM (two-way ANOVA with Tukey’s post-test: aanti-CHIKV mAb + depleting mAb versus isotype mAb + depleting mAb (open circle versus open triangle); banti-CHIKV mAb + isotype nondepleting mAb versus isotype mAb + isotype nondepleting mAb (closed circle versus closed triangle); cisotype mAb + depleting mAb versus isotype mAb + isotype nondepleting mAb (open triangle versus closed triangle); danti-CHIKV mAb + depleting mAb versus anti-CHIKV mAb + isotype nondepleting mAb (open circle versus closed circle); a, b, or dP < 0.05, aa, bb, or ddP < 0.01, bbb or cccP < 0.001, bbbb, cccc, or ddddP < 0.0001). (D, H, L, and P) Ipsilateral ankles were collected at 7 dpi, and viral RNA levels were measured. Bars indicate mean values, and significance was determined by Student’s t test between either the depleted or isotype nondepleted ankles [(D, H, L, and P) n = 6, two experiments; **P < 0.01, ***P < 0.001, ****P < 0.0001]. Open symbols denote mice depleted of indicated immune cells, and closed symbols denote mice that receive isotype nondepleting control mAbs.

NK cell depletion did not affect foot swelling in the presence of anti-CHIKV mAbs or an isotype control mAb (fig. S7C). Because anti-CHIKV mAb therapy still reduced viral RNA levels in the ipsilateral ankle in the absence of NK cells, these cells likely did not mediate clearance of infected cells (fig. S7D). In comparison, depletion of monocytes and neutrophils (Ly6G+/Ly6C+ cells) reduced foot swelling starting at 3 dpi compared with the isotype-depleted animals (Fig. 6C). Because this decrease in swelling occurred regardless of administration of anti-CHIKV mAbs, we cannot conclude that monocytes and neutrophils are required for the beneficial clinical effect of mAb therapy. Depletion of monocytes and neutrophils, however, resulted in a substantial loss in anti-CHIKV mAb antiviral activity in the ankles of mice (Fig. 6D); these data suggest that one or both of these cell types mediate mAb-dependent clearance of CHIKV. In comparison, depletion of neutrophils (Ly6G+ cells) alone did not alter foot swelling after CHIKV infection with or without mAb therapy (Fig. 6G) or affect the antiviral activity of anti-CHIKV mAbs in the ankle (Fig. 6H). However, in the absence of monocytes alone, anti-CHIKV mAb therapy failed to reduce foot swelling at 7 dpi compared with isotype-treated mice (Fig. 6K), and the antiviral activity of anti-CHIKV mAbs was lost (Fig. 6L). To confirm that the requirement of monocytes was not specific to humanized mAbs in mice, we repeated CCR2 depletion using the mouse versions of CHK-152 and CHK-166. As observed with the humanized anti-CHIKV mAbs, there was a compensatory increase in neutrophils with monocyte depletion (Fig. 6, M and N). Consistent with the humanized mAbs results, in the absence of monocytes, the mouse anti-CHIKV mAbs did not reduce foot swelling or viral RNA levels (Fig. 6, O and P). These results suggest that monocytes are the primary cell type responsible for clinical protection and clearance of CHIKV infection in the affected foot via Fc-FcγR interactions.

DISCUSSION

Although previous studies have established the efficacy of mAb therapy during CHIKV infection, the mechanisms of protection in vivo have not been definitively identified. We examined the contribution of Fc effector functions for clinical and virological protection in a murine model of CHIKV-induced arthritis. Using a combination of neutralizing anti-CHIKV mAbs that bind to distinct epitopes of the viral surface glycoproteins, we showed that a functional Fc region is required for optimal protection against foot swelling and CHIKV infection during the acute phase of disease. Consistent with these results, clinical and virological protection with intact mAb therapy was diminished in mice lacking expression of activating FcγRs. Although intact anti-CHIKV mAb therapy augmented monocyte and neutrophil recruitment and proinflammatory cytokine production during the first phase of disease, these cellular and soluble inflammatory mediators were reduced during the second phase of disease, which correlated with reduced foot swelling at 7 dpi. Cell depletion studies identified monocytes as a key cell type for mAb-dependent reductions in CHIKV RNA and clinical disease in mice.

The significance of antibody Fc effector functions has been described in other viral infections, including human immunodeficiency (HIV), influenza, Ebola, West Nile, and hepatitis B viruses (3641), although the specific cells mediating this effect in vivo are less well characterized. Broadly neutralizing mAbs targeting various epitopes on HIV gp120 clear infected cells through FcγR interactions, and modifications that enhance Fc-FcγR affinity increase antibody efficacy (42). For influenza virus, mAbs targeting the conserved stalk region of the hemagglutinin protect in vivo, in part through Fc-FcγR interaction on NK cells and neutrophils (43). We tested combinations of three different anti-CHIKV mAbs targeting spatially distinct epitopes on the virion with a range of neutralization capacity. CHK-152 binds across domains A and B of the E2 protein, CHK-166 binds to domain II of the E1 protein, and IM-CKV063 binds to an intersubunit epitope in domain A of the E2 protein (8, 18, 21). Thus, at least for three distinct epitopes, we observed a requirement of Fc-FcγR interactions for optimal antibody protection after CHIKV challenge. Our data are consistent with a study showing reduced efficacy of the CHK-152 N297Q variant in protecting against foot swelling in mice when administered 18 hours after infection (18). In a historical study with Semliki Forest virus (SFV), a distantly related alphavirus, cross-reactive anti–Sindbis virus γ-globulin failed to neutralize SFV in vitro but was protective in vivo, possibly through ADCC (44). Analogously, prophylaxis with non-neutralizing anti-CHIKV mAbs partially protected Ifnar1−/− mice from lethal CHIKV challenge, which suggested a role for Fc effector functions, although this was not directly tested (17).

CHIKV RNA persists for months in musculoskeletal tissues, including the joints of infected mice, nonhuman primates, and humans (4547). Although administration of the intact mAb combination reduced residual viral RNA levels in the contralateral ankle at day 28 more so than the N297Q variants or isotype control mAbs, in all cases, residual viral RNA still remained. Moreover, no difference in viral RNA clearance among treatment groups was observed in the ipsilateral ankle. The failure to clear viral RNA completely was not due to insufficient levels of antibody, because even weekly 500-μg dosing of anti-CHIKV mAbs did not improve clearance. It is possible that once joint-associated tissues are seeded with CHIKV, complete elimination may be difficult to accomplish with mAb therapy. The basis for clearance failure remains uncertain, although CHIKV RNA might be expressed in cells in the joint space that are immune privileged and inaccessible to antibodies or immune cells. Alternatively, the cells that persistently replicate CHIKV RNA may not express sufficient amounts of viral structural proteins on their surface to be targeted by antibody-mediated clearance mechanisms.

Previous studies have shown that monocytes and macrophages can contribute to foot swelling (26, 48). Intact anti-CHIKV mAb therapy at 3 dpi increased infiltration of monocytes and neutrophils 1 day later, and this was associated with a small but consistent concomitant increase in foot swelling. Notwithstanding this occurrence, the increased frequency of the neutrophils associated with intact mAb therapy at 4 dpi ultimately resulted in reduced accumulation of myeloid cells and diminished foot swelling at 7 dpi. Consistent with the temporal changes in recruitment of monocytes and neutrophils, mice treated with the intact mAbs had higher levels of the chemokines CCL2, CCL3, CCL4, and CCL5 at 4 dpi yet reduced levels at 7 dpi compared with the N297Q or isotype control mAbs. Thus, the differences in flux of immune cells in the ankle after intact mAb treatment could explain the improved clinical phenotype. Alternatively, differences in the activation state of infiltrating cells could modulate tissue inflammation and edema. Fc-FcγR, but not Fc-C1q, interactions were important for the anti-CHIKV mAb-dependent increase in accumulation of CD45+ cells and monocytes. One limitation in explaining these results is that the mechanistic basis for the different phases of swelling and the relationship to viral burden after CHIKV infection is still not fully understood. The impact of Fc-FcγR and C1q interaction on immune cell activation, subset composition, and the tissue microenvironment during CHIKV infection warrants further study to define precisely how antibody therapy modulates joint swelling.

NK cells, neutrophils, and monocytes express activating FcγRs that engage the Fc region of mAbs and induce ADCC and/or ADCP to remove virally infected cells (34). A recent study with influenza A virus used clodronate treatment to identify alveolar macrophages as a key cell type for mAb-mediated viral clearance and survival (43). In our experiments, depletion of NK cells or neutrophils did not change foot swelling or antibody-mediated reduction of CHIKV RNA. However, clearance of CHIKV RNA in mice was compromised by the combined absence of monocytes and neutrophils or depletion of monocytes alone, indicating that the monocytes contribute to clearing viral infection in a mAb-dependent manner. Notwithstanding these results, and independent of anti-CHIKV mAb therapy, depletion of monocytes and neutrophils or monocytes alone, but not neutrophils alone, resulted in increased viral RNA in the ipsilateral ankle compared with isotype control mAb–depleted animals. Previous studies that depleted monocytes with clodronate or using CCR2-DTR+ mice also observed increased CHIKV burden in the contralateral ankle and serum (26, 32). These data suggest that monocytes are important for controlling CHIKV infection or recruit immune cell subsets that facilitate viral clearance.

In the absence of monocytes alone, anti-CHIKV mAb therapy failed to protect against clinical disease. Monocyte depletion, regardless of mAb therapy, resulted in increased foot swelling at 7 dpi and was associated with more circulating neutrophils. Analogously, Ccr2−/− mice developed more severe CHIKV-induced disease because of compensatory recruitment of neutrophils and eosinophils (35). Consistent with this idea, when we depleted both monocytes and neutrophils, foot swelling was reduced after CHIKV infection. However, when others have depleted monocytes with clodronate or bindarit, an inhibitor of CCL2 production, diminished foot swelling was observed, which suggests that monocyte perturbations at different phases of CHIKV infection may result in distinct clinical disease outcomes (26, 49).

Antibodies can limit CHIKV infection and disease, especially when given before or shortly after virus infection (18, 20, 23, 50). Our study identifies specific Fc-FcγR interactions as key determinants of mAb-mediated clearance and clinical protection once infection is established in target tissues. Accordingly, antibody-based therapies against viruses that display structural glycoproteins on the surface of infected cells should likely be optimized for specific Fc effector functions to enhance clearance and minimize disease pathogenesis.

MATERIALS AND METHODS

Study design

The goal of this study was to determine whether antibody effector functions contributed to protection against infection and clinical disease caused by CHIKV and arthritogenic alphavirus. Using the mouse model of CHIKV arthritis, we performed passive antibody transfers, qRT-PCR for viral RNA, in situ hybridization, flow cytometry, cytokine analysis, in vitro functional assays, and in vivo antibody-based cellular depletions to evaluate how antibody effector functions modulate infection and disease. The sample size and number of independent experiments are indicated in each of the figure legends.

Cells, viruses, and antibodies

Murine mAbs against CHIKV were described previously (18), purified from hybridoma supernatants by sequential Protein A Sepharose and size exclusion chromatography, and buffer-exchanged into phosphate-buffered saline (PBS). mAbs with human Fc domains were produced recombinantly as previously described (51) in ExpiCHO-S cells at 32°C following the manufacturer’s instructions. All mAbs used were tested free of endotoxins using the limulus amebocyte lysate test cartridge in the Endosafe Portable Testing System (Charles River). Vero, BHK21, and C6/36 cells were cultured as described (18). The CHIKV La Reunion OPY1 strain was a gift of S. Higgs (Kansas State University) and produced from an infectious complementary DNA (cDNA) clone (52, 53). Focus reduction neutralization tests (FRNTs) were performed as previously described using CHK-11 as the detection antibody (19).

Mouse studies

Experiments were performed in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health after approval by the Institutional Animal Care and Use Committee at the Washington University School of Medicine (assurance number: A3381-01). All injections with virus were performed under anesthesia with ketamine hydrochloride (80 mg/kg) and xylazine (15 mg/kg).

Four-week-old WT C57BL/6J mice were purchased from The Jackson Laboratory. Four-week-old congenic C1q−/− or FcRγ−/− mice were bred at the Washington University Animal Facility. mAbs CHK-152 mouse IgG2c, CHK-166 mouse IgG2c, WNV E60 (murine isotype control; IgG2c), CHK-152 human IgG1, CHK-166 human IgG1, CHK-152 human IgG1 N297Q, CHK-166 human IgG1 N297Q, IM-CKV063 human IgG1, IM-CKV063 human IgG1 LALA, or WNV hE16 (human isotype control; IgG1) were administered to mice by intraperitoneal injection 3 days after virus infection. Mice were inoculated subcutaneously in the left footpad with 103 focus-forming units (FFU) of CHIKV in Hank’s balanced salt solution supplemented with 1% heat-inactivated fetal bovine serum (HI-FBS). Ipsilateral foot swelling was monitored via measurements (width × height) using digital calipers. Mice were sacrificed and perfused with PBS, and tissues were collected at 5, 7, or 28 dpi. Tissues were titered by qRT-PCR using RNA isolated from viral stocks as a standard curve to determine FFU equivalents, as previously described (19).

Quantification of human IgG in tissues

Unlabeled goat anti-human κ antibody that was cross-adsorbed to mouse IgG (SouthernBiotech) was bound overnight at 4°C on MaxiSorp immunocapture ELISA plates (Thermo Fisher Scientific) in a sodium bicarbonate buffer (pH 9.3). Wells were washed with PBS + 0.05% Tween 20 (Fisher Scientific) and blocked with blocking buffer [PBS + 2% bovine serum albumin (BSA) (Sigma)] for 1 hour at 37°C. Ankles were homogenized in Dulbecco’s modified Eagle’s medium + 2% HI-FBS using a MagNA Lyser instrument (Roche). Ankle homogenates were clarified (12,000 rpm for 5 min), heat-inactivated at 56°C for 1 hour, serially diluted in blocking buffer, and then added to wells for 1 hour at 4°C. Plates were rinsed and then incubated for 1 hour at 4°C with a horseradish peroxidase–conjugated goat anti-human Fc antibody that was cross-adsorbed against multiple animal species (SouthernBiotech). Plates were washed and developed with the TMB One-Step Substrate System (Dako). The reaction was stopped with 1 M H2SO4, and absorbance was monitored at 450 nm. A standard curve was run in parallel and analyzed using nonlinear regression to determine the concentration of the humanized antibodies in tissues.

Histology and viral RNA in situ hybridization

WT mice were inoculated with 103 FFU of CHIKV and treated with indicated anti-CHIKV or isotype control mAbs by intraperitoneal injection at 3 dpi. At 7 dpi, animals were perfused sequentially with PBS and 4% paraformaldehyde (PFA). Ipsilateral feet were collected, and hair was removed using Nair (Church & Dwight). Tissue was fixed for 24 hours in 4% PFA, rinsed with PBS and water, and then decalcified for 10 days in 14% EDTA-free acid (Sigma) in water at pH 7.2. Decalcified tissue was rinsed, dehydrated, embedded in paraffin, sectioned, and stained with hematoxylin and eosin.

Viral RNA in situ hybridization was performed on tissues from mice at 5 and 7 dpi. Tissue and slides were prepared as described above. In situ hybridization was performed using RNAscope 2.5 [Advanced Cell Diagnostics (ACD)] according to the manufacturer’s instructions and as previously described using the CHIKV probe (479501) designed and synthesized by ACD (54). Images were acquired on a Nikon Eclipse E400 microscope.

Cytokine and chemokine analysis

Mice were inoculated with 103 FFU of CHIKV and treated with indicated anti-CHIKV mAbs or an isotype control by intraperitoneal injection on 3 dpi. Ipsilateral ankles were collected at 4 or 7 dpi from PBS-perfused mice. Tissue was homogenized in PBS with 0.1% BSA and analyzed for cytokines and chemokines using a Bio-Plex Pro Mouse Cytokine 23-Plex Assay kit (Bio-Rad) following the manufacturer’s instructions.

Flow cytometry

Mice were inoculated with 103 FFU of CHIKV and treated with indicated anti-CHIKV mAbs or an isotype control by intraperitoneal injection on 3 dpi. At 4 or 7 dpi, mice were perfused with PBS, the ipsilateral feet were disarticulated, and the skin was everted. Tissue was digested in RPMI supplemented with 10% HI-FBS, HEPES, collagenase (Sigma), and deoxyribonuclease I (Sigma) for 1 hour at 37°C with agitation, strained through a 70-μm filter, and resuspended in RPMI supplemented with 10% HI-FBS. Single-cell suspensions were blocked for FcγR binding (BioLegend; clone 93; 1:50) and then stained with the following antibodies: CD45 BUV395 (BD Biosciences; clone 30-F11; 1:200), CD3 allophycocyanin (APC)–Cy7, phycoerythrin (PE)–Dazzle594, or BV421 (BioLegend; clone 145-2C11; 1:100), CD4 FITC (fluorescein isothiocyanate) (BioLegend; clone RM4-5; 1:200), CD8α APC (BioLegend; clone 53-6.3; 1:200), NK1.1 PE-Cy7 or PE (BioLegend; clone PK136; 1:200), CD11b PerCP-Cy5.5 (BioLegend; clone M1/70; 1:200), CD19 BV605 (BioLegend; clone 6D5; 1:200), Ly6C Pacific Blue (BioLegend; clone HK1.4; 1:200), Ly6G PE-Cy7 (BioLegend; clone 1A8; 1:200), CD11c APC (BioLegend; clone N418; 1:200), MHC class II (I-A/I-E) A700 or PE (BioLegend; clone M5/114.15.2; 1:200), or Ly6B FITC (Abcam; clone 7/4; 1:100). Viability was determined through exclusion of a fixable viability dye (eBioscience; eFluor 506; 1:500). Samples were processed on a BD LSRFortessa X20 or BD-LSRII flow cytometer and analyzed using FlowJo version 10 (FlowJo, LLC).

Preparation of recombinant CHIKV p62-E1

CHIKV p62-E1 [E3-E2-E1: residues S1-R64 of E3, S1-E161 of E2, and Y1-Q411 of E1 including a (GGGS)4 polylinker between E2 and E1] of the CHIKV-LR strain was cloned into the mammalian expression vector pFM1.2 (55) with a C-terminal octa-histidine tag. The resulting plasmid was transiently expressed in FreeStyle 293-F cells using 293fectin reagent (Thermo Fisher Scientific). Cell supernatants were harvested (72 and 120 hours after transfection), and soluble CHKV p62-E1 protein eluted from nickel agarose beads (GoldBio) was purified by Superdex 200 gel filtration chromatography in 20 mM HEPES (pH 7.4) and 150 mM NaCl, 0.01% NaN3 at 4°C.

Antibody-dependent cellular phagocytosis

Recombinant CHIKV p62-E1 was biotinylated and conjugated to streptavidin-coated Alexa Fluor 488 beads (Life Technologies). CHIKV p62-E1–coated beads were incubated with fivefold dilutions of antibodies (mAbs, 1 to 0.0016 μg/ml) in culture medium for 2 hours at 37°C. Monocytes harvested from the bone marrow of C57BL/6 mice and purified using a CD14+ Enrichment kit (STEMCELL Technologies) were added at a concentration of 2.5 × 104 cells per well and incubated for 4 hours at 37°C in low-adherence 96-well plates. After incubation, monocytes were incubated with the following antibodies: CD11b APC (BioLegend; clone M1/70), CD11c Alexa Fluor 700 (BioLegend; clone N418), and Ly6G BV421 (BioLegend; clone 1A8). Cells were fixed with 4% PFA and analyzed by flow cytometry on a BD LSR II cytometer using DIVA and FlowJo analysis software. The phagocytic score was determined using the following calculation: (% of Alexa Fluor 488+ cells) × (Alexa Fluor 488 geometric mean fluorescent intensity of Alexa Fluor 488+ cells)/10,000.

Antibody-dependent neutrophil phagocytosis

Recombinant CHIKV p62-E1 was biotinylated and conjugated to streptavidin-coated Alexa Fluor 488 beads. CHIKV p62-E1–coated beads were incubated with fivefold dilutions of antibodies (mAbs: 1 to 0.0016 μg/ml) in culture medium for 2 hours at 37°C. Bone marrow cells were harvested from C57BL/6 mice. Cells were washed with PBS, and 5.0 × 104 cells per well were added to bead-antibody immune complexes and incubated for 1 hour at 37°C. Cells were stained with the following antibodies: CD11b APC (BioLegend; clone M1/70), CD11c Alexa Fluor 700 (BioLegend; clone N418), and Ly6G BV421 (BioLegend; clone 1A8). Cells were fixed with 4% PFA and analyzed on a BD LSR II flow cytometer. A phagocytic score was determined as described above.

Antibody depletion of immune cell subsets

For NK cell depletion, anti-NK1.1 (BioXCell; clone PK136; 200 μg) or an isotype control (BioXCell; clone C1.18.4; 200 μg) was administered to mice by intraperitoneal injection 1 day before infection and 3 dpi. For monocyte and neutrophil depletion, anti-Ly6G/Ly6C (Gr-1; BioXCell; clone RB6-8C5; 500 μg) or an isotype control (BioXCell; clone LTF-2; 500 μg) was administered to mice by intraperitoneal injection 1 day before infection and 1, 3, and 5 dpi. For neutrophil depletion, anti-Ly6G (BioXCell; clone 1A8; 250 μg) or an isotype control (BioXCell; clone 2A3; 250 μg) was administered to mice by intraperitoneal injection 1 day before infection and 1, 3, and 5 dpi. For monocyte depletion, anti-CCR2 (clone MC-21; 25 μg) (56) or an isotype control mAb (BioXCell; clone LTF-2; 25 μg) was administered to mice by intraperitoneal injection on 1, 3, and 5 dpi. Anti-CHIKV mAbs or an isotype control was administered to mice by intraperitoneal injection on 3 dpi. Foot swelling was monitored using digital calipers. After extensive perfusion, ipsilateral ankles were collected, homogenized, and processed for CHIKV infection as described above.

For analysis of immune cell depletion, peripheral blood was collected on the day of harvest. Red blood cells were lysed with ammonium chloride potassium (ACK) lysis buffer (Gibco) and resuspended in RPMI supplemented with 10% HI-FBS. For NK1.1 depletion, single-cell suspensions were blocked for FcγR binding and stained with CD45 BUV395, CD3 APC-Cy7, NKp46 BV421 (BioLegend; clone 29A1.4; 1:50), and fixable viability dye (eFluor 506). For Ly6G/Ly6C, Ly6G, or CCR2 depletion, single-cell suspensions were blocked for FcγR binding and stained with antibodies against CD45 BUV395, CD11b PerCP-Cy5.5, Ly6C Pacific Blue, Ly6B FITC, Ly6G PE-Cy7, CD11c APC, and fixable viability dye (eFluor 506).

Statistical analysis

Statistical significance was assigned with P values of <0.05 using GraphPad Prism version 7.0 (La Jolla, CA). The specific statistical test for each dataset is indicated in respective figure legends and was selected on the basis of the number of comparison groups and variance of the data. For foot swelling analysis, significance was determined by a two-way analysis of variance (ANOVA) with Tukey’s post-test (more than two groups) or Sidak’s post-test (between two groups). For some virological and immune cell analysis, significance was determined by one-way ANOVA with Tukey’s post-test or Student’s t test. For other viral burden experiments, as well as cytokine/chemokine analysis, if data points were of unequal variance, a Kruskal-Wallis ANOVA with Dunn’s post-test or Mann-Whitney test was used, depending on the number of comparison groups.

SUPPLEMENTARY MATERIALS

immunology.sciencemag.org/cgi/content/full/4/32/eaav5062/DC1

Fig. S1. IM-CKV063 in combination with CHK-166 reduces clinical disease and viral RNA.

Fig. S2. Viral burden in contralateral ankle with anti-CHIKV mAb therapy.

Fig. S3. Gating scheme for infiltrating immune cells in WT mice.

Fig. S4. Similar levels of cellular infiltration in the ipsilateral feet at 7 dpi.

Fig. S5. Fc effector functions of antibody affect proinflammatory cytokine and chemokine expression.

Fig. S6. Gating scheme for infiltrating immune cells in FcRγ−/− and C1q−/− mice.

Fig. S7. NK cell depletion does not affect mAb-mediated protection.

Table S1. Proinflammatory chemokine and cytokine expression in joint tissue homogenates.

Table S2. Raw data in Excel file.

REFERENCES AND NOTES

Funding: This work was supported by NIH grants R01 AI089591 (M.S.D.), R01 AI114816 (M.S.D.), and T32 AI007172 (J.M.F.) and NIH contract HHSN272201400058C (B.J.D.). Authors contributions: J.M.F., B.M.G., and V.R. performed experiments. J.M.F., B.M.G., G.A., and M.S.D. designed the experiments and analyzed the data. L.H., B.J.D., M.A.E., D.H.F., and S.J. contributed key reagents and methodology. J.M.F. and M.S.D. wrote the first draft of the manuscript, and all authors provided editorial comments. Competing interests: M.S.D. is a consultant for InBios and Sanofi Pasteur and is on the Scientific Advisory Board of Moderna. S.J. and L.H. are employees of MacroGenics and have equity. B.J.D. is an employee and shareholder of Integral Molecular. The authors have no additional competing interests. Data and materials availability: Mouse strains and antibodies used are available from the corresponding author upon request.
View Abstract

Stay Connected to Science Immunology

Navigate This Article