UC Berkeley

The mosquito-borne dengue virus (DENV) is endemic in 100 countries and annually threatens over half of the worlds' population. Any of the four serotypes of DENV (DENV1-4) can cause a wide range of disease, ranging from inapparent illness to classic dengue fever to severe disease characterized by vascular permeability and hypotensive shock (dengue hemorrhagic fever/ dengue shock syndrome; DHF/DSS). The complex interactions between epidemiological, viral, host genetic, and immunological factors make dengue pathogenesis difficult to dissect. While both the humoral and cellular arms of the adaptive immune response have been reported to contribute to the development of severe dengue disease, antibodies alone are sufficient to enhance DENV disease in vitro, a concept referred to as "antibody-dependent enhancement". Under this hypothesis, antibodies generated against the first DENV infection recognize but do not neutralize the second DENV serotype and instead permit increased entry into Fc receptor (FcR)-bearing target cells. The development of small animal models to study DENV pathogenesis has greatly facilitated dissection of the role of individual immune components critical to both protection and enhancement of DENV disease. Here, we characterize a novel mouse model of lethal DENV disease and further demonstrate that antibodies alone are sufficient to enhance a sub-lethal DENV infection. Hallmarks of the antibody-enhanced, lethal phenotype include increased viral load, reduction in platelet counts, elevated pro-inflammatory cytokines and vascular permeability, all of which are considered defining features of severe human dengue disease. We further demonstrate that genetically modified non-FcR-binding anti-DENV monoclonal antibodies (mAbs) can be used to treat lethal disease caused by both high viral inoculum (virus-only) and enhancing amounts of antibody. While prophylactic ability appears dependent upon the neutralizing titer of the mAb, therapeutic efficacy following an antibody-enhanced, lethal infection is unique to highly avid mAbs that target the fusion loop and envelope protein Domain III (EDIII) A strand epitopes. Further investigation into the protective role of antibodies examined the role of anti-EDIII antibodies in mediating serotype-specific protection and enhancement in vivo. While previous studies of murine mAbs suggested that anti-EDIII antibodies were predominantly serotype-specific and highly neutralizing, our work demonstrated that they only constitute between 15% and 35% of the in vitro neutralizing potency of human and mouse serum, respectively, and are not required for protection in vivo. Further analysis of specific antibody subsets revealed that serotype-specific antibodies (targeting all E domains) contribute predominantly to maintaining serotype-specific protection, while cross-reactive antibodies drive disease enhancement. Taken together, our in vivo studies using a small animal model of DENV disease have validated in vitro observations and identified specific roles for antibody subsets in mediating both protection and enhancement.