Defining the Development of Receptive Niches for in Utero Transplantation of Hematopoietic Stem Cells.

Abstract 3527

Poster Board III-464

In utero hematopoietic stem cell transplantation (IUSCT) is a promising therapeutic alternative to postnatal stem cell transplantation which could potentially provide successful treatment for many genetic and developmental diseases affecting the immune and hematopoietic systems. Advances in molecular biological techniques and improvements in obstetrical procedures such as chorionic villous sampling now permit the collection of fetal material and the diagnosis of genetic disorders early enough in gestation to allow the use of IUSCT to treat these diseases prior to the onset of irreversible organ damage. However, after almost 20 years of experimental work, the only clinical applications for which IUSCT has proven successful are diseases such as SCID in which there is a selective advantage of donor cell development over host cells. Thus, the future success of this promising approach depends upon a full understanding of the mechanisms of engraftment and differentiation of stem cells during the fetal period. The sheep fetus shares many important immuno-physiological and developmental characteristics with the human fetus and has served as an accurate model in which to study IUSCT. Therefore, in the present studies we used the sheep model to evaluate the evolution and maturation of the microenvironmental niches during fetal development with the goal of delineating the period of gestation during which a multilineage-supporting bone marrow environment is present. To achieve this objective, we performed histologic and immunofluorescence analyses on bone from fetal sheep at 5.7, 6.5, 7.2, and 9.2 gestational weeks (gw; term: 21 gw), using antibodies to markers expressed on various niche cells. At the earliest time point of 5.7gw, the bone rudiment consisted mainly of mesenchymal structures, and no cells expressing markers of endothelium (CD34, CD31) or osteoblasts (osteopontin, N-cadherin) were observed. Beginning at 6.5gw, an extremely limited number of osteoblasts were present, possibly indicating the onset of development of the osteoblastic niche at that point. Cells exhibiting a CD34+ALDH+ phenotype were also observed in specific areas lining the perichondrium and within some endochondral compartments. Larger numbers of osteoblasts were seen at 7.2gw, and this was the first point at which we began to observe the association of CD34+ALDH+ cells with these bone niches. By 9.2gw, the population of osteoblasts was well established and the population of CD34+ALDH+ cells closely interacted with the osteoblasts, suggesting that the osteoblastic niche began forming at 6.5gw, but only became fully established by 9.2gw. When similar analyses were performed with antibodies to CD34 and CD31 to detect endothelial cells, we found that CD34+CD31+ endothelial cells were not present in detectable numbers at 5.7, 6.5, or 7.2gw, but by 9.2gw, the CD34+CD31+ population had increased dramatically, indicating that the vascular niche develops fairly rapidly during between 7.2 and 9.2gw in the sheep. These data collectively indicate that the bone marrow osteoblastic niche commences development at 7.2gw and reaches relative maturity by 9.2gw, while the vascular niche develops relatively rapidly between 7.2 and 9.2gw. These findings have important implications for optimizing engraftment of HSC following transplantation in utero and may help to explain the limited clinical success that has thus far been achieved with this approach. Given that the sheep fetus begins to attain immune competence at around 9.2gw, we routinely perform IUSCT between 7.8-8.6gw to avoid potential complications associated with immune rejection of the transplanted HSC. Based on our present results, it appears that in so doing, we are actually transplanting the HSC at a time when neither the osteoblastic or vascular niches have fully developed, making it unlikely they can serve as receptive sites for engraftment of the transplanted HSC. In the light of these findings, it is thus crucial to further narrow down the window of development of both the vascular and osteoblastic niches to determine the precise developmental time for optimal HSC engraftment. Given that our present results suggest this may be near the time the fetus achieves immune-competence, future strategies for improving the success of IUSCT may need to incorporate immuno-modulation.