Of note, previous studies had been done in preparation of peripheral blood mononuclear cells and thus could not attribute changes in cytokine production to individually polarized T cells. In line with these previous reports our study in CD4+ T cell Trichostatin A clones confirmed a conspicuous direct effect of ribavirin on TH1 cells, which may be linked to altered expression of interferon regulatory factors. On the other hand, our data suggest that changes in TH2 functions are unlikely to contribute much to the overall anti-HCV effect of ribavirin, because we observed only slightly reduced production of IL-10 in our TH2 clones. CD4+ Tregs are an alternative source of immunomodulatory cytokines such as IL-10. Thus far, however, frequencies of Tregs in liver and blood have not shown a clear pattern to treatment outcomes. In two recent studies numbers of Tregs remained unchanged in peripheral blood, whereas exposure to PEG-IFN/ribavirin resulted in increased frequencies of Tregs in the liver. None of the previous studies, however, has studied functions of single Tregs and specifically addressed the potential role of ribavirin. Here, our in vitro analysis of Treg clones revealed the novel finding that ribavirin markedly reduced production of IL-10 in Tregs, but did not affect their overall low proliferative capacity. Beyond that, our co-culture assays showed that ribavirinmediated inhibition of IL-10 production of Treg clones clearly resulted in improved proliferation of T effector cells irrespective from their antigen-specificity and functional polarization. Our data suggest that ribavirin specifically affects Tregs in patients with chronic hepatitis C, because we observed only negligible effects in Treg clones from patients with self-limited HCV infection and Tregs from healthy controls. Nevertheless we did not check explicitly antigen-specificity of the T cell clones in the current experiments. However, we have shown previously that our Treg clones recognize epitopes on HCV core in an antigenspecific, HLA-DR-restricted fasion. Thus, it is quite likely that the newly discovered action of ribavirin on Tregs holds also true for Tregs in patients with chronic hepatitis C. Moreover, we demonstrated inhibitory activity of ribavirin on Treg function at concentration levels which are typically achieved in vivo during ribavirin treatment. Moreover, the ribavirin effect could also be demonstrated at therapeutic concentrations of IFN-alpha suggesting that our in vitro findings might have relevance for the in vivo situation under interferon/ribavirin combination therapy. Overall, we did not find that ribavirin modulate transcription factors regulating TH1/TH2 cell responses. However, Brenndo¨rfer et al. recently demonstrated that ribavirin can reverse the HCV NS3/4A-mediated impairment of antiviral signalling pathways in vivo. In summary our analysis of T cell clones supports the concept of immune-mediated activity of ribavirin and suggests that this compound affects the T cell balance by two supplementary mechanisms: importantly also reverses Tregmediated inhibition of T effector cells by inhibiting IL-10 release of HCV-specific Tregs. Beyond that, in vivo reduced IL-10 production under ribavirin may also result in down-regulated expression of ICOS, altered costimulatory signalling and increased priming of T cells towards TH1 differentiation. Thus, in chronic hepatitis C ribavirin is likely to shift the functional balance in the immune system between Tregs and T effector cells towards more efficous effector responses, thus strengthening the antiviral activity of PEG-IFN/ribavirin combination therapy.