Stable mixed chimerism has been considered the most robust tolerance strategy. However, rejection of solid donor tissues by chimeras has been observed, a state termed split tolerance. Since new non-myeloablative mixed chimerism approaches are being actively pursued, we sought to determine whether they lead to full tolerance or split tolerance and to define the mechanisms involved. Fully mismatched mixed chimeras generated by induction with various lymphocyte-depleting antibodies along with either low-dose irradiation or busulfan and temporary sirolimus, maintained stable mixed chimerism but nevertheless rejected donor skin grafts. Generation of stable mixed chimerism using antibody targeting CD40L, but not depleting antibodies to CD4 and CD8, could prevent split tolerance when skin grafts were given together with donor bone marrow. Minor antigen matching abrogated the ability of effector T cells to reject donor skin grafts. A CFSE killing assay indicated that chimeras were both directly and indirectly tolerant of donor hematopoietic cell antigens, suggesting that minor mismatches triggered a tissue-specific response. Thus, split tolerance due to tissuerestricted polymorphic antigens prevents full tolerance in a number of nonmyeloablative mixed chimerism protocols and a 'tolerizing' agent is required to overcome split tolerance. A model of the requirements for split tolerance is presented.Supporting information for this article is available at http://www.wiley-vch.de/contents/jc_2040/2007/36938_s.pdf
IntroductionMuch of the effort to develop donor-specific transplantation tolerance has been focused on inducing peripheral tolerance through costimulation blockade. While initially promising, with more extensive tests of this approach the success has been somewhat limited, particularly when translated to larger animal models and to the clinic [1]. In hindsight this may not be surprising given that tolerance naturally occurs primarily in the thymus and only secondarily in the periphery [2]. In contrast, the approach of generating hematopoietic chimerism via bone marrow transplantation (BMT) takes advantage of the thymic central tolerance mechanisms, and is considered the most robust method of inducing donor-specific tolerance [3][4][5][6]. The chimer- ism approach is limited clinically by the harsh recipient conditioning needed to establish chimerism and the possibility of graft-vs.-host disease [7]. More recently less toxic strategies have been developed that establish mixed allogeneic chimerism, where substantial levels of donor and recipient hematopoietic cells co-exist in the recipient, and have raised hope that robust transplantation tolerance will soon be routine clinically [6]. Mixed chimerism has been considered to induce tolerance to all other donor tissues. If true, mixed chimerism could be a solution for both solid organ transplantation and cellular transplants, such as allogeneic islets used to treat type-1 diabetes. However, studies in full chimeras, where virtually all hematopoietic cells in the recipien...