Replicative stress-induced aging of hematopoietic stem progenitor cells increases oncogenic mutation burden and incidence of myelodysplasia in sickle cell disease mice Free

Introduction: Sickle cell disease (SCD) affects millions globally, results in vascular occlusions, chronic organ damage and early death. An increased risk of myelodysplastic syndrome (MDS) and hematological malignancy (HM) is becoming evident in SCD population-based studies and gene therapy/allogeneic transplants. The specific pathobiology that predisposes to it is unknown, and necessary to address as patients are surviving longer and curative therapies are on the forefront. We hypothesized that increased replicative stress-induced aging in SCD hematopoietic stem and progenitor cells (HSPC) leads to expansion of clones carrying oncogenic mutations (clonal hematopoiesis; CH), MDS and HM.

Methods: HSPC from eight 1-year-old (yo) Townes SCD (SS) mice and 8 age/strain-matched normal controls (AA) were exposed to whole exome sequencing (WES; 400X). CBCs, blood smears and flow cytometry was performed quarterly on peripheral blood (PB) and mice sacrificed when 1 yo. Multipotent progenitor populations (MPP) and long-term (LT) and short-term (ST) hematopoietic stem cells (HSCs) from bone marrow (BM) HSPC were examined. BM cells from 1-yo AA and SS mice (CD45.2) were transplanted into WT (CD45.1) primary and secondary mice for 3 and 6 months. Pre- and post-transplant BM was subject to Targeted Deep DNA Sequencing (TDS; 30,000X) with a custom panel of 450 known oncogenes associated with CH/MDS/HM, to detect low-level somatic variants. Germline single nucleotide variants (SNVs) were excluded using tail DNA TDS and the mouse Single Nucleotide Polymorphism Database.

Results: 1-yo SS HSPCs had significantly increased aging-related enrichment of C>T (P<0.01) and G>A (P<0.05) transition SNVs with 185±22 more mutant genes compared to AA HSPC (P<0.001). Since anemia, leukocytosis, and thrombocytopenia are used to monitor MDS/HM routinely, but they are also features typical of SCD, we developed a flow cytometry-based hematopoietic panel comprising of lymphoid, myeloid and HSPC markers. Myeloid-biased hematopoiesis, with 30% higher myeloid- (P<0.0001) and 20% less lymphoid lineage cells (P<0.0001) were seen in SS PB as early as 3 mo, suggesting aging HSCs. Myeloid bias was also present in BM progenitors, with significantly higher proportions of MPP3 (CD135^-CD150^-CD48⁺LSK) in SS BM. Further, SS BM had decreased HSC-LT (CD135^-CD150⁺CD48^-LSK) and HSC-ST (CD135^-CD150^-CD48^-LSK). Together, increased transition mutations, myeloid-biased hematopoiesis and perhaps, HSC exhaustion, indicated faster aging of SS HSCs than AA HSCs.

Prospectively, 24 AA and 35 SS mice were followed until 1 yo. A high proportion (8-25%) of SS mice developed abnormal hematopoiesis (CD45^dim, CD33⁺CD11b^dim/neg myeloid- and CD19^dim lymphoid populations) in PB, in contrast to 4% AA mice. Furthermore, this abnormality developed early in SS (3 of 35, 7 of 35 and 8 of 32 mice at 3, 6 and 9 mo, respectively) mice in contrast to 9-mo (1 of 24) AA mice. Further, mice with abnormal hematopoiesis had tri-lineage myelodysplasia in PB and BM by morphology and histology.

We tested whether the underlying mechanism was increased HSC replicative stress via serial transplants, and subjected BM cells pre- and post-transplant to TDS. SS BM showed an increase in SNVs from 94–754 vs. 22–29 SNVs in AA BM, and increase in mutant genes from 59–293 in SS BM vs. 20–22 in AA BM, pre- and post-transplant, respectively. Overall, variant allele frequencies (VAF) of oncogenic non-synonymous mutations were higher in SS mice (n=7) and increased from 1.2% pre- to 2.3% post-transplant (n=12). VAF in AA mice (n=6) were 0.6% pre- and increased to 0.8% post-transplant (n=11) (P<0.00001).Conclusion: Herein, using age/strain-matched humanized SS and AA mice, we show accumulation of a significantly higher mutation burden associated with aging in 1-yo SCD mice. Prospective monitoring of cohorts of SS and AA mice show earlier development of myeloid-biased haematopoiesis and myeloid-biased progenitor populations indicative of aging HSCs. Our data suggests accelerated HSC aging from HSC replicative stress as the underlying mechanism. Furthermore, SS mice develop a dysplastic hematopoiesis, consistent with MDS. We confirmed this at a molecular level by serial transplant-induced HSC replicative stress and show increased CH and a high mutational burden in genes associated with HM. Thus curative interventions could be targeted early in life to mitigate the risk of MDS/HM due to increased mutation burden with HSC aging.