Mixed xenogeneic chimerism is a promising approach to achieving tolerance across allogeneic and xenogeneic barriers.  T cell tolerance, which is achieved through mixed chimerism, assures the absence of an induced antibody response to the donor. However, natural antibodies (NAb) limit the success of xenogeneic transplantation. In addition to T cell tolerance^1,2, which develops largely through a deletional mechanism^2,3, mixed chimerism leads to tolerance of anti-carbohydrate (αGal) Nab-producing B cells in Gal transferase (GalT) knockout mice^4,5. Even the anti-Gal B cells in presensitized GalT KO mice can be rendered tolerant when high doses of Gal+ bone marrow are given⁶. Ratàmouse mixed xenogeneic chimeras prepared with non-myeloablative conditioning are tolerant at the level of NAb-producing B cells, both for αGal (in GalT KO mice) and for non-Gal specificities^7-10. Anti-Gal-reactive cells consist of B-1b B cells that are enriched in the peritoneal cavity and only produce antibody following activation in vitro^11,12 .  Upon antigen exposure in vivo, these cells can lose CD11b expression and migrate to the spleen^11,12.  Pre-existing anti-Gal-producing B cells are rapidly tolerized by an anergy mechanism that is later followed by deletion of donor-reactive B cells when mixed chimerism is induced with Gal+ donors^5,6,9.  This tolerance is dependent on complement receptor expression by non-hematopoietic cells of the recipient¹³, which we hypothesize are splenic follicular dendritic cells (FDCs) that pick up immune complexes via their complement receptors. New studies on the role of FDCs, complement and secreted antibody in tolerizing these anti-Gal-producing B cells will be presented. Our data collectively suggest a model wherein, upon binding to Gal antigen, Gal-binding B cells fix complement on the cell surface and interact with complement receptors expressed on FDCs.  These interactions may be critical both for anti-Gal antibody responses and tolerance induction by mixed chimerism induction.

In mice that are presensitized to alloantigens and therefore contain preformed anti-donor alloantibodies, engraftment of bone marrow expressing the same alloantigens is markedly impaired.  When B cell-deficient mice were used as recipients, presensitization could be overcome with a non-myeloablative regimen involving high doses of donor marrow, permitting durable mixed chimerism and acceptance of second skin grafts by mice that had previously rejected donor skin¹⁴. However, this method did not achieve mixed chimerism in wild-type mice that were presensitized more than one year earlier and in which anti-donor alloantibody had disappeared.  The bone marrow transplant did not induce a secondary antibody response but was instead associated with an increase in anti-donor cytokine-producing T cells¹⁴.  Thus, in addition to producing alloantibodies that resist engraftment, B cells play an important role in driving a memory T cell alloresponse that is particularly resistant to conditioning with T cell depletion and costimulatory blockade. It will be essential to overcome this memory response in order to achieve mixed chimerism and allograft tolerance in allopresensitized recipients. Moreover, the ability to overcome presensitized anti-Gal responses with non-myeloablative conditioning by giving large doses of allogeneic Gal+ marrow suggest that a similar memory T cell population is not induced against carbohydrate antigens, making it possible to achieve mixed chimerism in the presence of high levels of these antibodies.