To study the consequences of the interaction of respiratory syncytial virus (RSV) with dendritic cells in vitro, we established a model of the primary immune response using dendritic cells, autologous naive T cells and the superantigen toxic shock syndrome toxin 1 (TSST 1). About 10% of the naive T cells express the T cell receptor chain Vbeta2. These cells were stimulated by TSST 1 and could be analysed by flow cytometry. Cultures infected with RSV produced significantly less interferon-gamma compared to uninfected cultures. In a first set of experiments we evaluated whether this culture model using isolated CD4(+) CD45RA(+) T cells, in fact, reflects the primary immune response. In a prospective study, cells were isolated from 13 children at birth, at 1 year of age and at 4 years of age. RSV reduced interferon-gamma production at all the age groups analysed and the results were stable over time within a given individual. In a second set of experiments, we asked whether clinical differences in the course of RSV infection are due to variations in the cellular immune response. At the age of 1 year (5-9 months after the RSV epidemic) dendritic cells and naive T cells were obtained from 27 children with a history of bronchiolitis, from 15 children with a benign course of RSV infection and from 26 controls without RSV infection. The frequency of interferon-gamma-producing cells in RSV infected cultures was significantly lower (P < 0.001) in cultures from children with a history of RSV bronchiolitis compared to children with mild RSV infection. Cultures from children without infection displayed a wide range of results. Overall, interferon-gamma generation in this group was still lower (P < 0.05) than in the group with mild RSV infection. Because we have ruled out that memory cells play a role in the experiments performed, the most likely explanation for our results is that a high generation of interferon-gamma in the primary immune response protects from severe RSV mediated disease.