Investigation of Impact Induced By AML1-ETO on Self-Renewal and Differentiation of Hematopoietic Stem and Progenitor Cells in Pre-Leukemic Stage

AML1-ETO fusion gene is the most common cytogenetic subtype of AML, which occurs in approximately 15% of acute myeloid leukemia (AML) patients. More than 85% of patients could achieve complete remission after induction chemotherapy. However, it was less effective on the patients over the age of 60 who cannot tolerate standard chemotherapy. And nearly 30% of patients relapsed eventually with a dismal outcome. Therefore, identification of the initiating abnormal cells to inhibit the early occurrence induced by AML1-ETO is an urgent problem.

In this study, an inducible conditional AML1-ETO knock-in murine model (Aml1^Eto/+; Mx1-Cre) was successfully established, hereafter called AE^KI. The stage of malignant transformation in hematopoietic stem and progenitor cells (HSPC) was identified. We found that the absolute number of LT-HSCs (1.43×10⁶ vs 4.85×10³, p<0.0001), ST-HSCs (7.90×10⁶ vs 1.88×10⁴, p<0.0001) and MPPs (6.07×10⁵ vs 1.82×10⁴, p=0.0358) in bone marrow (BM) increased significantly in AE^KI group compared with that of control group, while CMPs (8.13×10³ vs 1.83×10⁵, p=0.0029), GMPs (8.48×10³ vs 3.12×10⁵, p=0.0026) and MEPs (1.25×10⁴ vs 2.03×10⁵, p=0.001) decreased. It suggested that the activation of AML1-ETO in vivo resulted in excessive accumulation of HSCs and differentiation blockage from HSCs to HPCs. However, AML1-ETO alone was insufficient to induce the onset of overt leukemia after 300-day observation of survival.

Furthermore, the characteristics of AE^KI HSPCs in the pre-leukemic stage were investigated. Colony forming assay (CFC) was performed to evaluate the stemness of HSCs in vitro. The equivalent of 2×10³ LT-HSCs were isolated from BM of each group and plated for 14 days. AE^KI LT-HSCs gave rise to dysplastic colonies and lost replating capability compared to that of control group (colony counts: 19.33 vs 30.17, p=0.016). Moreover, a progressive decrease in PB chimerism was observed in AE^KI recipients via competitive transplantation assay in vivo (15.8% vs 62.5%, p<0.0001), for which BM cellularity exhibited a significant decrease (7.24×10⁶ vs 4.81×10⁷, p<0.0001). But it led to an notably expansion in HSC number (LT-HSC:1.18×10⁵ v 8.59×10³, p=0.0035) which had a same tendency as the phenotype of primary mice. Based on Ki67 and DAPI staining, proportion of HSCs in G₁ phase increased (59.85% vs 38.63%, p=0.0007), while that in G₀ (33.58% vs 44.2%, p=0.045) and S/G₂/M (4.86% vs 13.41%, p=0.0002) phases decreased. These results implied that AML1-ETO induction could reduce the quiescence and self-renewal ability of HSCs obviously in comparison with normal HSCs.

Bulk RNA sequencing was performed in LT-HSCs and GMPs sorted from both groups. Genesets related to cell proliferation, cell cycle regulation and Myc targets were enriched in LT-HSCs from control group. It was also found that genesets of tricarboxylic acid (TCA) cycle, glycolysis and fatty acid metabolism were enriched in LT-HSCs from control group. To validate it, Seahorse extracellular flux assay was performed to evaluate oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) on HSPC population (c-kitenriched cells). We found that OCR and ECAR both decreased significantly in AE^KI group that supported the results from RNA-seq. RNA-seq data also showed that fatty acid metabolic pathway was enriched in AE^KI group during the process from LT-HSCs to GMPs, which suggested that AE^KI HSPCs maintained abnormal function through excessive fatty acid supply while it was not necessary during normal hematopoiesis. To confirm the above speculation, the mice were fed with fat-free or normal food after the AML-ETO induction in vivo to evaluate the impact of fatty acid on hematopoiesis. Interestingly, BM chimerism increased on the fat-free group (31.56% vs 15.68%, p=0.013), with the absolute number of LT-HSCs (5.16×10⁵ vs 1.80×10⁶, p=0.068), ST-HSCs (1.28×10⁶ vs 2.72×10⁶, p=0.026) and MPPs (4.70×10⁴ vs 2.53×10⁵, p=0.013) reduced obviously compared with in a normal diet. It suggested the excessive accumulation of HSCs was partly recovered.

In summary, our study suggested that AML1-ETO induction could cause HSCs to lose self-renewal potential and lead to differentiation disruption during the pre-leukemic stage of AML. Limiting exogenous lipid uptake was able to rescue the phenotype of differentiation disrupt of HSPCs partly and it could be a promising therapeutic strategy.

Wang:AstraZeneca: Consultancy; Abbvie: Consultancy, Membership on an entity's Board of Directors or advisory committees.