Bone Marrow CD34+ Cells Expanded On Human Brain Endothelial Cells Reconstitutes Lethally Irradiated Baboons in a Variable Manner.

Abstract

Abstract 3214Poster Board III-151Increased cell dose has a positive impact on the therapeutic outcome of bone marrow (BM) transplantation. However, methods to successfully expand hematopoietic stem cells (HSC) from BM have yet to be achieved. It has been shown previously that ex vivo expansion of BM cells using porcine microvascular endothelial cells can rescue a baboon from a lethal dose of radiation (Brandt et al. Blood 1999). However, in prior studies baboons that received CD34+ cell doses less than 4 × 106 cells/kg body weight failed to achieve hematopoietic engraftment. In our current studies we have used human brain endothelial cells (HUBECs) and cytokines to expand BM cells and examined their ability to provide hematopoietic reconstitution in three lethally irradiated baboons following autologous transplantation. After ex vivo culture, the grafts represented a 1.8- to 2.1-fold expansion of CD34+ cells, a 3.7-fold to 13.2-fold increase of colony-forming cells (CFC), and a 1.9-fold to 3.2-fold increase of cobblestone area-forming cells (CAFC) in comparison to the input cell numbers. The animal (PA6873) which received expanded product of a suboptimal dose of CD34+ cells (1.6 × 106/kg) achieved only myeloid engraftment (day 24). Out of 3 baboons transplanted two displayed myeloid engraftment, one animal achieved both myeloid and platelet engraftment and the third animal (PA6888) failed to achieve engraftment. The animal (PA6893) which received the expanded product of 4.9 × 106/kg CD34+ cells achieved myeloid engraftment (WBC > 500/ml blood) by day 8 and platelet engraftment (>20,000/μl) by day 39. The WBC recovery of this baboon was comparable but the platelet recovery was delayed (31 day vs. 39 day) in comparison to that experienced by an animal that received a large number of unexpanded CD34+ cells (26 × 106 /kg). Interestingly, despite the grafts of all three animals having a similar degree of CD34+ cell expansion, similar progenitor cell (CFC, CAFC) expansion, a similar pattern of cell adhesion molecule expression and similar migration capacity across an SDF1 gradient, the hematopoietic reconstitution capacity of each graft differed greatly. Prior studies using human BM cells in non-contact HUBEC co-cultures in NOD/SCID mice demonstrated no such variability. Our current data indicates three possibilities for the variations in hematopoietic reconstitution observed in a baboon model. One possibility is that there might be cell autonomous biological variation of HUBEC expanded BM grafts. The second possibility is the expanded graft may have contained variable numbers of contaminating endothelial cells (HUBEC) contributing to variations in hematopoietic reconstitution. Transplantation of vascular endothelial cells without HSC rescue has been shown to enhance hematopoietic recovery following radiation in mice (Chute et al. Blood 2007). Biological variation amongst the hosts is unlikely since identical pre-transplant conditioning and post-transplant care was provided to all three baboons. In addition, since only non-adherent cells were harvested from the co-culture, it is possible that more primitive repopulating HSCs embedded within the endothelial monolayer might have been excluded from the graft. Taken together, our findings highlight inherent differences in the hematopoietic reconstitution capacity of expanded BM grafts in xenotransplantation studies and large animal models. DisclosuresNo relevant conflicts of interest to declare.