In utero transplantation offers the possibility of early treatment of serious diseases and a chance to circumvent transplantation barriers by treating the fetal patient during a period of natural immunodeficiency. In this issue of Blood, Durkin and colleagues measure the levels of donor engraftment required to reliably yield long-term engraftment in fetal mice and further demonstrate that host NK cells play a role in graft rejection and immune tolerance.
Stem cell transplantation before birth has been conceptualized as an approach to treat a fetus with a birth defect prior to the development of the immune system and to avoid the requirement for tissue compatibility matching. A window of opportunity is believed to exist between genetic testing using chorionic villus sampling, feasible at the end of the first trimester of pregnancy, and the anatomical maturation of the thymus at 14 weeks' gestation that marks a sharp rise in circulating T cells in the fetus.1,2 Hematopoiesis resides primarily in the liver during this period of development, but the bone marrow is being seeded and CD34+ cells are found in high numbers in the fetal circulation.3 It was hoped that unmatched stem cells transplanted into this environment would engraft without the need for immune suppression or cytoablation. Unfortunately, prenatal transplantation has almost always failed to produce sufficient levels of donor cell chimerism in patients suffering from various immunodeficiencies or hemoglobinopathies. The home-field advantage enjoyed by the fetal hematopoietic system points to the difficulties in engrafting a large bolus of stem cells and a potential knockout role for the nascent immune system.
The study by Durkin et al is a new addition to the literature showing that increasing the number of cells transplanted is a viable method for boosting donor chimerism levels and achieving durable engraftment. Long-term engraftment requires both successful grafting of stem cells and establishment of donor tolerance. Increasing the stem cell dose is important, but other effects and cells also appear to have an impact. Higher cell doses of minimally purified grafts introduce a large number of mature leukocytes and progenitors into the fetal hematopoietic system that can spar with the developing immune system and affect the development of tolerance. How much chimerism is enough to induce tolerance? The answer from Durkin et al is that a threshold of 1.8% engraftment measured at 3 weeks of age predicts long-term stable engraftment. Likewise, Chen et al recently observed a threshold of 2% chimerism at 1 month of age that was predictive of long-term blood chimerism and an even chance at donor-specific tolerance for allogeneic skin grafts.4 These findings underscore the importance of chimerism levels in the development of tolerance and the importance of the makeup and size of the initial graft in the eventual establishment of stable chimerism.
The focus on thymic development and the rise of T cells in the fetus has neglected the potentially important role of natural killer (NK) cells in fetal immunity and graft rejection. The phenomenon of hybrid resistance in mice has proven NK cells capable of graft rejection.5 Enhanced engraftment following in utero transplantation in nonobese diabetic/severe combined immunodeficiency (NOD/SCID) mice, with impaired NK-cell function, compared with SCID mice with functional NK cells, suggested a role for NK cells in impeding engraftment.6 Durkin and colleagues now show a clear role for NK cells in mediating graft rejection following in utero transplantation. NK-cell inhibitory receptor (Ly49) expression correlated with chimerism levels, yielding NK-cell tolerance above the threshold chimerism level associated with durable engraftment. NK-cell depletion could also rescue donor chimerism that fell below the threshold for stable engraftment. These results suggest modulating NK-cell numbers or activity as a way to boost engraftment, possibly by tuning grafts to best affect the maturation of tolerant NK cells.
A difficulty in studying in utero transplantation in the murine system is that the development of the immune system differs significantly between mice and humans. In mice, mature NK cells arise after birth, whereas in humans, NK cells appear in the liver at 6 weeks' gestation7 and represent about a third of leukocytes in the fetal blood at 13 weeks' gestation8 —the time around which most fetal transplantations were performed. The presence of fetal NK cells may have contributed to the lack of engraftment in many prenatal transplantation efforts. Indeed, the early development of NK cells may serve to protect the fetus from invasion by maternal cells that often pass into the fetal circulation. Transplanted donor cells, particularly small grafts of highly purified stem cells, may simply be recognized as more invaders to be knocked out of the ring before establishing themselves in the fetus.
Conflict-of-interest disclosure: The author declares no competing financial interests. ■