Abstract
In utero hematopoietic stem cell transplantation (IUHCT) is a potential alternative to postnatal HSCT for treatment of congenital hematologic disorders. We have recently documented in the murine model that, in contrast to the results for congenic IUHCT where 100% of recipients maintain permanent mixed hematopoietic chimerism, 70% of recipients of allogeneic IUHCT lose their chimerism by four weeks of age. We hypothesized that there was an adaptive immune barrier that was differentially activated in non-chimeric versus chimeric mice. To directly test this hypothesis, we compared the frequency of allospecific T-cells by the quantitative in vivo MLR assay in chimeric versus non-chimeric mice, as well as immunized and non-injected control mice. After IUHCT of GFP+ B6 BM cells into E14 Balb/c fetuses, lymphocytes were harvested from 4 week old Balb/c recipients, CFSE stained, and injected into F1 recipients, and F1 spleens were then FACS analyzed for frequency of alloreactive T cells at 96hrs. The frequency of alloreactive T cells in chimeric mice was 2.034 ± 0.562 percent of CD4+, CFSE+ input lymphocytes as compared to 6.323 ± 1.185 percent in non-chimeric mice (p=0.010). We also compared allospecific humoral response in the same groups by flow cytometric assay for anti-donor IgG. Serum was isolated from 4 week old Balb/c IUHCT recipients and incubated with B6 target cells and anti-IgG secondary antibody for detection of alloantibodies. Anti-IgG immunofluorescence was increased 9.512 ± 3.183 fold as compared to 1.097 ± 0.113 fold over controls in non-chimeric and chimeric mice respectively (p=0.013), and showed dose dependence with regard to serum concentration. These results confirm that loss of chimerism after allogeneic IUHCT in the murine model is associated with activation of an allospecific adaptive immune response. We further hypothesized that the source of immune activation was the mature maternal immune system, rather than the immature fetal immune system. To test this hypothesis, we performed IUHCT and used non-injected foster moms as the source of breast milk. Remarkably, we found that 17/17 pups raised by foster moms maintained long-term (> 4 months) donor cell chimerism, as compared to 15/52 pups raised by injected moms (p<0.0001). We then analyzed the injected moms for the presence of alloantibodies at P1 (one week after IUHCT), P8 and P28. At P1, 10/10 were negative for the presence of alloantibodies, while 8/10 were positive for the presence of alloantibodies at P8 as well as P28 (p=0.0007). We have subsequently identified a positive correlation between the number of aborted fetuses and the magnitude of the maternal humoral response (correlation coefficient=0.6, n=15, p=0.018). In fact, the absence of a maternal anti-donor humoral response is only observed in pregnancies with no aborted fetuses and the injected pups from these pregnancies are all chimeric. These results indicate that IUHCT at E14 in the murine model results in maternal immunization through fetal loss and production of alloantibodies which are then transferred in breast milk to pups, triggering an adaptive immune response and resulting in loss of chimerism in 70% of pups by four weeks of age. These findings explain the apparent contradiction of activation of an alloimmune response by IUHCT in the pre-immune fetal recipient and once again validate the concept of fetal immune tolerance as permissive for allogeneic engraftment after IUHCT. The importance of this mechanism of loss of chimerism after IUHCT in large animal or clinical settings is unknown and will be the focus of future investigations. In addition, our observations may have broad implications for autoimmune disease and the immune consequences of maternal-fetal cellular trafficking.
Mutual inhibition between Prkd2 and Bcl6 controls T follicular helper cell differentiation
