In mice of the Igha immunoglobulin allotypic haplotype we found, the presence of T lymphocytes with an inherent inhibitory activity against the expression of the IgG2ab allotype (IgG2a of the Ighb immunoglobulin allotypic haplotype). This constitutive anti-IgG2ab T lymphocyte activity can be enhanced in vivo by what we called 'sensitization', which usually consists of one or two intravenous injections of B splenocytes from Ighb congenic mice. When injected at birth, the resulting anti-IgG2ab T splenocytes induce, with 100% success, total, specific and chronic (but experimentally reversible) suppression of IgG2ab in Igha/b F1 hybrid mice prepared by mating Igh congenic mice. Even if restricted to IgG2ab expression, this experimental model, which deals with an unambiguous case of T cell-mediated down-regulation of immunoglobulin production, provides a clear and powerful tool to dissect finely the behavior of the partners (T and B lymphocytes) intervening in regulation within the immune system. For example, we observed that CD4 T lymphocytes were necessary to obtain full recruitment of anti-IgG2ab CD8 T lymphocytes during the sensitization, that suppression induction in anti-IgG2ab T splenocytes of newborn recipients required cooperation between CD4 and CD8 T lymphocytes, and that CD8 T lymphocytes were essential for suppression maintenance. We showed that this suppression was not characterized by an accumulation of B lymphocytes containing the allotype they could not secrete or Cgamma2ab mRNA they could not translate. The recipient's immune system was not involved in the suppession maintenance; this was done by donor T lymphocytes, which ensured the chronicity of IgG2ab suppression throughout the recipient's life. We demonstrated that the mechanism of this suppression implied an MHC-restricted presentation by target B lymphocytes of Cgamma2ab peptides to the T cell receptor (TCR) of anti-IgG2ab T lymphocytes. Notwithstanding the requirement of a CD4-CD8 T lymphocyte cooperation during the induction phase, we functionally determined that the suppression induction implicated an MHC class I-, but not class II-restricted interaction. We also demonstrated the existence in vivo of alternative or concomitant use of perforine- and Fas-mediated cytotoxicity pathways in this T cell-induced IgG2ab suppression. Thus this suppression did not imply silencing IgG2ab production, but B lymphocyte destruction by CD8 T lymphocytes. Always using our suppression model, we demonstrated that an agonistic anti-CD40 treatment helps in recruiting CD8 cytotoxic T lymphocytes, involved in immune regulatory functions and that CD40 expression on Ighb B lymphocytes confronted with CD8 T lymphocyte effectors only operating via the Fas pathway was involved in the total suppression of IgG2ab expression. The selection and maintenance of such normal T cell activity against the IgG2ab allotype in mice of different genetic backgrounds remain somewhat enigmatic. Indeed, we did not observe any similar activity against other immunoglobulin allotypes or isotypes. The intestinal flora had no influence on the emergence of this anti-IgG2ab T lymphocyte activity, as it was untouched in germ-free Igha mice when compared with normal Igha mice. More recently, this model offered an opportunity to study problems pertaining to immune tolerance. For instance, we showed that the genetic elements involved in the building of anti-IgG2ab TCR were available in Igha and Ighb mice of different genetic backgrounds, but that somatic constraints, namely the perinatal presence of IgG2ab, effectively prevented their acquisition, while its absence led to their spontaneous emergence. Consequently, we were able to induce anti-IgG2ab T lymphocytes into a tolerance state by injecting Igha mice with soluble IgG2ab during the perinatal period. However, the full T lymphocyte tolerance obtained in this manner was not definitively acquired, as it had reversed spontaneously when investigated 3 to 6 months after the end of tolerogen treatment, even when this treatment had been prolonged from the perinatal period to 9 months of age. The mechanisms (induction and reversion) of this tolerance involves the physical elimination or the irreversible inactivation of the natural anti-IgG2ab T lymphocyte clones and their resurgence, from bone-marrow precursors, as long as the thymus remains operational, but not the establishment of a reversible, functional unresponsiveness (anergy) or an active, cell-mediated inhibition of anti-IgG2ab T clones. We attempted to elucidate, in Ighb mice, whether the natural T lymphocyte unresponsiveness to IgG2ab involved a central tolerance mechanism and to identify the type of tolerogen implicated in this tolerogenesis. The experiments principally showed that this natural T lymphocyte tolerance to IgG2ab was mediated by a thymic mechanism; that the capacity to induce it was gradually acquired by Ighb thymuses and was most probably due to potentially IgG2ab-producing/presenting cells, progressively colonizing the developing thymus; and that a significantly decreased postnatal Cgamma2ab gene transcription correlated with the emergence of anti-IgG2ab T lymphocytes in Igha/b F1 (postnatally deprived of their B lymphocyte compartment), which subjected them to autoimmune IgG2ab-allotype suppression.