Identification and characterization of therapy-related acute myeloid leukemia (tAML) susceptibility factors could help distinguish between high- and low-risk patients and suggest logical targets for prevention and control strategies. Our lab recently identified the anti-apoptotic gene Bcl2 as a candidate tAML susceptibility gene based its proximity to a quantitative trait locus (QTL) peak (Mleu1.1) associated with time-to-leukemia in B6C3F1 × SWR/J F2 intercross mice treated with the potent alkylator N-ethyl-N-nitrosourea (ENU). Recent studies have shown that, in addition to its anti-apoptotic properties, Bcl2 decreases both mismatch repair activity and abasic site repair. Thus, higher Bcl2 expression levels in early hematopoietic cells at baseline or in response to chemotherapy could prevent both apoptosis and DNA repair in the affected cells, leading to the accumulation of potentially leukemogenic mutations in surviving cells. At Mleu1.1, mice homozygous for SWR/J alleles have longer survival compared to mice homozygous for the C57BL/6J haplotype. We hypothesized that Bcl2 could be the underlying QTL gene driving this phenotype, with higher Bcl2 expression predicted in C57BL/6J compared to SWR/J mice. Because early hematopoietic cells are implicated as the transforming population in tAML, we measured Bcl2 RNA expression in Lin−/c-Kit+ (enriched for stem/progenitors) sorted cells using the Illumina BeadChip Platform. Concordant with our hypothesis, Bcl2 expression measured by 3 separate probes was 43.0% (p=0.002), 44.5% (p=0.002), and 33.5% (p=0.06) lower in SWR/J compared to C57BL/6J (n=3 mice per strain). We then measured Bcl2 protein expression by flow cytometry in LSK (Lin−/Sca+/c-Kit+), CMP (Lin−/c-Kit+/Sca−/Fcγ −/CD34+), GMP (Lin−/c-Kit+/Sca−/Fcγ+/CD34+), and MEP (Lin−/c-Kit+/Sca−/Fcγ −/CD34−) cells. In C57BL/6J mice (n=3), 4.49±1.28%, 14.97±2.44%, 5.87±2.53%, and 8.86±1.81% of LSKs, CMPs, GMPs, and MEPs respectively were Bcl2 positive. Concordant with RNA expression data, a lower percentage of LSKs (p=0.02) from SWR/J mice (n=3) were Bcl2+. Specifically, we detected 1.70±0.31%, 13.43±2.87%, 6.09±1.96%, and 6.26±1.64% Bcl2+ LSKs, CMPs, GMPs, and MEPs respectively. To determine the effect of ENU exposure on Bcl2 expression, we treated mice (n=3 per strain) with 100 mg/kg ENU 24 hours prior to flow cytometric analysis. Compared to baseline, treated mice had decreased total bone marrow cellularity (13.0±4.1 × 106 vs. 28.1±6.8 × 106 viable cells recovered per mouse; treated vs. untreated) and increased percent Bcl2+ cells in most compartments of early hematopoiesis in both strains. However, the percent Bcl2+ cells from ENU treated mice was higher in C57BL/6J vs. SWR/J mice (p<0.001 for LSK, CMP, GMP, and MEP). Specifically, 33.37±5.11%, 57.10±7.25%, 32.53±11.78%, and 41.47±4.90% in C57BL/6J mice and 6.89±1.13%, 33.43±5.62% 13.8±4.01%, and 18.63±3.96% in SWR/J mice of LSKs, CMPs, GMPs and MEPs respectively were Bcl2 positive. To confirm the relevance of apoptosis (and of genes underlying the apoptotic response) in our model, we measured apoptosis in primary bone marrow cells following in vitro ENU exposure. Primary bone marrow cells from C57BL/6J mice (n=3) were treated with 100 ug/ml ENU or vehicle for 60 minutes. We determined that ENU causes apoptosis (measured by 7-AAD/Annexin-V staining) in a time and dose dependent manner. At 48 hours after ENU exposure, 47.4±4.3% of treated cells were 7-AAD+/Annexin-V+ compared to 21.7±0.9% of vehicle treated cells. Comparison of apoptosis in whole bone marrow from SWR/J (n=3) and C57BL/6J (n=3) mice following in vitro ENU exposure showed no detectable difference between strains. Our results show that Bcl2 expression in early hematopoietic cells is strain dependent and that these differences are detectable at both the RNA and protein levels. A smaller percentage of hematopoietic stem/progenitor cells express Bcl2 in SWR/J compared to C57BL/6 mice at baseline and C57BL/6J mice exhibit more pronounced increases in percent Bcl2+ cells following ENU exposure. These differences in Bcl2 expression may explain the genetic component of tAML susceptibility we mapped to Mleu1.1. Further investigation of Bcl2 as a candidate tAML susceptibility gene is warranted. Biomarkers of tAML susceptibility could be useful for risk stratification and/or targeted pharmacologic intervention to reduce the incidence of this disease.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author