B-cell delivered gene therapy for tolerance induction: role of autoantigen-specific B cells.

Antigen-specific tolerance induction using autologous B-cell gene therapy is a potential treatment to eliminate undesirable immune responses. For example, we have shown that experimental autoimmune encephalomyelitis (EAE) and type 1 diabetes in NOD mice can be ameliorated using antigen-Ig fusion protein transduced B cells. However, it is well established that auto-reactive antigen-specific B cells are activated in many autoimmune diseases and can contribute to pathogenesis. While syngeneic B cells from immunized or autoimmune mice can serve as tolerogenic antigen-presenting cells (APC), this observation begs the question of whether the antigen-specific B cells per se can be transduced as tolerogenic APC. To test this, we employed two model systems employing B cell receptor (BCR) transgenic or wild type (wt) mice as B-cell donors. While adoptively transferred MOG-Ig transduced wt C57Bl/6 B cells were highly tolerogenic and ameliorated EAE, MOG-Ig transduced anti-MOG B cells from BCR transgenic mice were not. This phenomenon was reproduced in the NOD diabetes model in which pro-insulin-Ig transduced polyclonal wt NOD B cells were protective, whereas similarly transduced anti-insulin BCR B cells were not. Since the frequency of antigen-specific B cells in an immunized animal is quite low, we wished to determine the threshold numbers of BCR transgenic B cells that could be present in an effective transduced population. Therefore, we "spiked" polyclonal wt C57Bl/6 B cells with different numbers of anti-MOG BCR transgenic B cells. In the EAE model, we found protection when BCR B cells were present at 1%, but they prevented tolerance induction at 10%. Antigen-specific B cells expressed normal levels of co-stimulatory molecules and were tolerogenic when transduced with an irrelevant antigen (OVA). Thus, the presence of a BCR specific for the target autoantigen may interfere with the tolerogenic process to that antigen, but BCR-specific B cells are not intrinsically defective as tolerogenic APC. Taken together, these data suggest that antigen-specific tolerance induction can be achieved in the presence of a limited number of antigen-specific B cells, but higher numbers of pathogenic B cells may mask this induction. This observation should guide future development of therapies using autologous B cells to treat patients with autoimmune diseases.