Heterozygous Germ Line Deletion of a 2Mb Interval in Mice That Models Loss of 7q22 in Human Myeloid Malignancies Results in Defective Hematopoietic Stem Cell Function Reminiscent of Premature Aging

Abstract 2340

Monosomy 7 and deletion 7q [del(7q)] are among the most common cytogenetic alterations found in myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML). However, little is known regarding how chromosome 7 deletions contribute to the pathogenesis of myeloid malignancies. We harnessed chromosome engineering in mice to investigate the in vivo consequences of deleting a contiguous 2 Mb interval on chromosome 5A3 that is syntenic to a segment of chromosome band 7q22 that is commonly deleted in human myeloid malignancies (Blood 88:1930–5, 1996). This region contains 14 known genes, and homozygous deletion is embryonic lethal. Heterozygous 5A3^+/del mice are grossly normal and have normal peripheral blood counts, but they have a 20% reduction in spleen weight and a 25% reduction in bone marrow (BM) cellularity per femur. We did not observe a significant difference in the frequency of phenotypic long-term hematopoietic stem cells (LT-HSC) (c-kit⁺, Sca-1⁺, Lin⁻, CD150⁺, CD41⁻, CD48⁻) between 5A3^+/del mice and their wild-type (WT) littermates; however, there is a ∼30% reduction in the frequency of multipotent progenitors (MPP) (c-kit⁺, Sca-1⁺, Lin⁻, CD150⁻, CD41⁻, CD48⁻), suggesting a defect in the transition from LT-HSC to MPP. Since 5A3^+/del mice did not spontaneously develop evidence of hematologic malignancy in >2 years of observation, and transplantation enforces enhanced HSC cycling which may reveal subtle HSC phenotypes not obvious in the steady state, we performed competitive repopulation assays to directly assess the function of 5A3^+/del HSC in vivo. Unfractionated test BM from WT or 5A3^+/del animals (CD45.2) were mixed with WT competitor BM (CD45.1) at a 1:1 or 1:2 ratio, followed by transplantation into lethally irradiated WT recipients (CD45.1+2). Although 5A3^+/del BM supported reconstitution of all hematopoietic lineages 6 months after adoptive transfer, reductions of 46% (p=0.0003) and 45% (p=0.0010) in comparison to WT were observed in the repopulation of B and T cell lineages, respectively. Reconstitution of the myeloid lineage was reduced to a lesser extent (25% reduction; p=0.0921), and analysis of the lineage distribution of myeloid, B, and T cell lineages within 5A3^+/del derived cells revealed a myeloid lineage skewing reminiscent of the pattern of repopulation exhibited by aged HSC. Interestingly, a progressive decline in the repopulating ability of 5A3^+/del BM upon cell differentiation was observed: when 5A3^+/del test BM were transplanted in a more stringent 1:2 test:competitor ratio, we observed a 29% reduction (p=0.1657) in contribution to the K⁺L⁻S⁺ (c-kit⁺, Lin⁻, Sca-1⁺) stem cell compartment in comparison to WT, but a 39% reduction (p=0.0498) in the more mature progenitor K⁺L⁻S⁻ (c-kit⁺, Lin^-, Sca-1⁻) population. To investigate if the marked deficit in repopulation is intrinsic to the LT-HSC population and to assess if haploinsufficiency of the 5A3 region in the microenvironment modulates the phenotype, we purified and transplanted 15 WT or 5A3^+/del LT-HSC together with 250,000 unfractionated WT BM competitors into WT or 5A3^+/del recipients respectively. Transplantation of 5A3^+/del LT-HSC demonstrated a similar trend of global repopulating deficit and overall pattern of altered lineage distribution regardless of the genotype of the recipient, which infers a dominant cell intrinsic mechanism of action of the 5A3 deletion. Meanwhile, transplantation of WT LT-HSC into 5A3^+/del recipients did not significantly modulate repopulating potential. To determine if 5A3^+/del LT-HSC is defective in self-renewal, we analyzed reconstitution in the LT-HSC compartment. 5A3^+/del LT-HSC repopulated the LT-HSC compartment as efficiently as WT controls regardless of the recipient genotype, and subfractionation of the 5A3^+/del derived KLS population to LT-HSC and MPP subsets revealed a significantly biased distribution towards LT-HSC by 1.39-fold (p=0.0146), suggesting increased self-renewal. Together, these data demonstrate that haploinsufficiency of the 5A3 interval leads to defective HSC function including a perturbed stem cell compartment, diminished overall repopulating potential and a myeloid-biased differentiation pattern. The phenotypic resemblance to aged HSC is intriguing given the increased incidence of MDS with monosomy 7 in elderly individuals.