Autoimmunity is fundamentally a continuously evolving process. The autoimmune responses shift, drift and diversify with time not only to other epitopes in the original antigen but also to other related and sometimes to unrelated antigens. We have described a form of immune diversification--reciprocal T-B epitope spreading--where the activation of first T cells by epitopes from an autoantibody molecule could lead to help provided to a variety of B cells displaying a cross-reactive version of the original epitope. The response spreads in this way until large cohorts of T and B cells have expanded in lupus-prone mice. Such reciprocal T-B cell response can also be induced in normal animals, its extent is limited by the emergence of inhibitory T cells. The induction of such inhibitory T cells is generally impaired in lupus mice. The delivery of T cell epitopes via plasmid DNA vectors, however, can overcome this impairment in lupus mice. The inhibitory T cells thus induced can suppress autoantibody production and lupus disease by ablating or inhibiting autoreactive B cells. Thus, T-B diversification that develops spontaneously in lupus mice could be curtailed in normal animals by inhibitory T cells that emerge whenever there is an impending 'danger' of pathologic autoimmunity. We have successfully exploited this regulatory potential of the normal immune response to inhibit clinical autoimmunity. Understanding the mechanisms of autoimmune diversification in lupus mice and of its down-regulation in normal animals may pave the way for developing novel treatments for autoantibody-mediated diseases such as lupus.