Dengue virus-specific human t cell clones. Serotype crossreactive proliferation, interferon γ production, and cytotoxic activity

Abstract

The severe complications of dengue virus infections, hemorrhagic manifestation and shock, are much more commonly observed during secondary infections caused by a different serotype of dengue virus than that which caused the primary infections. It has been speculated, therefore, that dengue hemorrhagic fever (DHF) and dengue shock syndrome (DSS) are caused by serotype crossreactive immunopathological mechanisms. We analyzed clones of dengue serotype crossreactive T lymphocytes derived from the PBMC of a donor who had been infected with dengue 3 virus. These PBMC responded best to dengue 3 antigen, but also responded to dengue 1, 2, and 4 antigens, in bulk culture proliferation assays. 12 dengue antigen-specific clones were established using a limiting dilution technique. All of the clones had CD3+ CD4+ CD8- phenotypes. Eight clones responded to dengue 1, 2, 3, and 4 antigens and are crossreactive, while four other clones responded predominantly to dengue 3 antigen. These results indicate that the serotype crossreactive dengue-specific T lymphocyte proliferation observed in bulk cultures reflects the crossreactive responses detected in the clonal level. Serotype crossreactive clones produced high titers of IFN-γ after stimulation with dengue 3 antigens, and also produced IFN-γ to lower levels after stimulation with dengue 1, 2, and 4 antigens. The crossreactive clones lysed autologous lymphoblastoid cell line (LCL) pulsed with dengue antigens, and the crossreactivity of CTL lysis by T cell clones was consistent with the crossreactivity observed in proliferation assays. Epidemiological studies have shown that secondary infections with dengue 2 virus cause DHF/DSS at a higher rate than the other serotypes. We hypothesized that the lysis of dengue virus-infected cells by CTL may lead to DHF/DSS; therefore, the clones were examined for cytotoxic activity against dengue 2 virus-infected LCL. All but one of the serotype crossreactive clones lysed dengue 2 virus-infected autologous LCL, and they did not lyse uninfected autologous LCL. The lysis of dengue antigen-pulsed or virus-infected LCL by the crossreactive CTL clones that we have examined is restricted by HLA DP or DQ antigens. These results indicate that primary dengue virus infections induce predominantly crossreactive memory CD4+ T lymphocytes. These crossreactive T lymphocytes proliferate and produce IFN-γ after stimulation with a virus strain of another serotype, and demonstrate crossreactive cytotoxic activity against autologous cells infected with heterologous dengue virus. Based on these results we hypothesize that dengue serotype crossreactive T lymphocytes may contribute to the pathogenesis of DHF/DSS; (a) by producing IFN-γ, which increases the number of Fcγ receptors and subsequently increases the number of dengue-infected cells by enhanced uptake of dengue virus-antibody complexes; and (b) by lysing dengue virus-infected cells during secondary infections with a virus of the heterologous serotype.