Inducible SBDS Deficiency and Germline FANCC/FANCG Deficiency Differentially Impact the Capacity of Bone Marrow Niches to Engraft Hematopoietic Stem Cells after Transplantation

Strategies to optimize interactions between the recipient bone marrow (BM) microenvironment and donor hematopoietic stem cells (HSC) could improve hematopoietic stem cell transplantation (HSCT) outcomes by reducing graft failure rates, improving blood and immune reconstitution efficiency, and reducing intensive cytotoxic conditioning requirements. Such methods are particularly relevant for patients with inherited bone marrow failure syndromes (iBMFs), rare genetic disorders for which HSCT is required to cure bone marrow failure (BMF), but for which HSCT is also associated with a high incidence of delayed donor engraftment, graft failure and morbidity due to chemotherapy and radiation sensitivity.

In previous studies of the MPL^-/- model of congenital amegakaryocytic thrombocytopenia, we discovered that decreased numbers of megakaryocytes and/or long-term (LT)-HSC caused by absent thrombopoietin/MPL signaling resulted in diminished capacity of the MPL^-/- BM niche microenvironment to efficiently engraft wildtype (WT) donor HSC after HSCT. This observation led us to hypothesize that pre-existing BMF, regardless of cause, may generally alter HSC niche elements critical for engraftment after HSCT. We now present data testing this hypothesis in mouse models of Fanconi Anemia (FA) and Shwachman-Diamond Syndrome (SDS).

To model FA, we used double mutant FANCC^-/-;FANCG^-/- mice, which have previously been demonstrated to develop manifestations of BMF. In our facility, FANCC^-/-;FANCG^-/- mice did not develop evidence of BMF, as measured by blood counts, BM cellularity, and HSC quantitation out to 1 year of age even with induction of HSC cycling by polyinosinic-polycytidilic acid (pIpC). Using SCT assays in which GFP⁺ WT donor BM was transplanted into FA recipients (FANCC^-/-, FANCG^-/-, FANCC^-/-;FANCG^-/-) or WT controls after 800-1100 cGy total body irradiation (TBI), efficiency and durability of donor engraftment was assessed at 3 weeks and 5 months, respectively, after primary SCT by competitive secondary HSCT. Unexpectedly, all FA recipients demonstrated more rapid BM reconstitution and enhanced early hematopoietic progenitor engraftment of WT donor BM compared with WT controls. However, there was no significant difference in LT-HSC engraftment efficiency or durability in the FA recipient groups versus WT controls.

To model SDS, we first attempted to generate a model of osteoblast-specific SBDS deletion by crossing Col1A1^Cre mice with mice carrying floxed SBDS alleles (SBDS^fl/fl). However, this Col1A1^CreSBDS^Exc model proved to be embryonic lethal at E12-E15 due to profound developmental abnormalities. We thus generated an inducible model by crossing SBDS^fl/fl mice with Mx1^Cre+ mice, and inducing SBDS deletion in Mx1-inducible BM hematopoietic and osteolineage niche cells by pIpC administration. SBDS deletion induced by pIpC was confirmed with PCR methods.

Within 4 weeks of pIpC induction, Mx1^CreSBDS^Exc mice show evidence of BMF, including a 40% reduction in platelet counts, inverted myeloid/lymphoid cell ratio in blood, and up to 80% reduction in immunophenotypic BM LT-HSC. When we attempted HSCT with GFP⁺ WT donor BM in Mx1^CreSBDS^Exc and littermate control recipients following 1100 cGy TBI, 90% of Mx1^CreSBDS^Exc recipients died by day 9 after HSCT, whereas all littermate controls survived. BM sections from Mx1^CreSBDS^Exc recipients demonstrated persistent aplasia out to 1 week post-HSCT, suggesting that Mx1^CreSBDS^Exc died from primary graft failure. Competitive secondary HSCT performed from BM harvested 1 week after primary HSCT demonstrated severe deficits in GFP⁺ WT donor HSC engraftment in Mx1^CreSBDS^Exc recipients. Our initial studies as to why donor engraftment fails in Mx1^CreSBDS^Exc recipients have identified that compared to WT, Mx1^CreSBDS^Exc mice show significantly reduced expansion of osteolineage niche cells after TBI, a response that we have previously shown to be critical for efficient donor engraftment after HSCT.

Taken together, our data demonstrate that pre-existing BMF and SBDS deficiency in BM osteolineage cells, but not FANCC or FANCG deficiency without co-existent BMF, significantly reduce the capacity of the BM niche to engraft donor HSC. Future studies defining specific niche deficits in our inducible SBDS deficiency model will provide critical insights into cellular and molecular pathways required for efficient engraftment after HSCT.