Scott Alan Smith, M.D., Ph.D.
Dengue Virus, Vaccines, Immunology, Viral Pathogenesis
Symptomatic dengue virus infection ranges in disease severity from an influenza-like illness to life-threatening shock. One model of the mechanism underlying severe disease proposes that weakly cross-reactive antibodies induced during a primary infection facilitate virus entry into Fc receptor-bearing cells during a subsequent secondary infection, increasing viral replication and release of cytokines and vasoactive mediators, culminating in shock. This unique process, known as antibody-dependent enhancement of infection, has significantly hindered vaccine development. There is a concern that potent neutralizing antibodies must be generated to all four dengue virus serotypes, as a vaccine that induce weakly cross-reactive non-neutralizing antibodies may increase the likelihood of developing severe disease. Much of our understanding of this process has come from studies using mouse mAbs, however antibody responses in mice typically exhibit less complexity than those in humans. A better understanding of the humoral immune response to natural dengue virus infection in humans is sorely needed. Using a high-efficiency human hybridoma technology developed in our laboratory, it is now possible to generate human hybridomas reliably using B cells from the peripheral blood of individuals who have recovered from an infection. Employing this technology, we have generated over 250 hybridomas secreting human mAbs to dengue virus from subjects who had recovered from primary or secondary infection. The vast majority of these antibodies are broadly serotype cross-reactive, directed against either E or prM protein, and capable of antibody-mediated enhancement of infection in vitro; very few exhibited serotype-specific binding or potent neutralizing activity. By employing a flow cytometric neutralization screening approach, we were able to enrich in the numbers of potently neutralizing mAbs that can be identified. This tactic resulted in the identification of several major phenotypic groups of human mAbs, each exhibiting unique neutralization and functional properties. Interestingly, mAbs that display serotype specific, virion-only binding, and exhibit potent serotype specific neutralizing activity with little enhancing properties also are produced naturally by humans in response to infection. Understanding the epitopes and activity of these naturally-occurring human antibodies is critical for vaccine development. Ideally, the reactivity of epitopes bound by enhancing antibodies should be reduced or eliminated in candidate antigens during the rational development of a dengue vaccine, so as to discourage such dominant recognition of these antigenic features by the humoral immune response. The long-term goal is to use such molecular information in the rational design of dengue vaccines that enhance the induction of protective neutralizing antibodies and reduce the risk of development of severe disease.