BMI1 Collaborates with RUNX1/AML1 Mutants in the Development of Human Myelodysplastic Syndrome (MDS) / Acute Myeloid Leukemia (AML)

Abstract 275

RUNX1/AML1 gene has been investigated in the pathogenesis of hematopoietic diseases and point mutations of RUNX1 have been frequently detected in patients with MDS or AML. We have found RUNX1 mutations in patients with MDS and MDS-related AML, including therapy-related cases. The mutations were distributed throughout the full length of the RUNX1 protein, and replacement of D171 amino acid in runt homology domain was the most frequent target of mutations in the RUNX1 gene. The D171N mutant showed a loss of normal RUNX1 trans-activation potential and dominant-negative trans-activation suppression, suggesting that this mutant may have some oncogenic potential in addition to the loss of function. In mouse transplantation systems D171N-transduced mice exhibited hyper-proliferative AML with multilineage dysplasia in collaboration with Evi1 overexpression. This impressive result indicated that RUNX1 mutations may be a cause of MDS with a leukemogenic potential. However in human, most of D171-mutated patients were MDS refractory anemia with excess blasts, and EVI1 overexpression was observed in a patient with MDS rapidly progressed to AML. Instead, most of patients with RUNX1 mutations displayed a high expression level of BMI1, suggesting that mouse phenotypes were not always meets clinical features of the patients with the mutations, partly because gene circumstances in mouse are different from those in human. Thus, biological analysis using human hematopoietic stem cells was considered to be necessary to clarify the molecular mechanisms of the RUNX1 mutations in the pathogenesis of MDS/AML.

To examine the effect of RUNX1 mutants, we transduced the D171N mutant into human CD34⁺ cells from cord blood cells. In colony forming cell (CFC) assay, the number of burst forming unit-erythroid colonies was significantly decreased, while the number of granulocyte-macrophage colony-forming unit colonies was increased. Most of the D171N-transduced cells expressed myeloid lineage markers. The D171N cells showed replating capacity for 3 replatings, suggesting that they have self-renewal advantage. The presence of progenitors with long-term self-renewal capabilities was confirmed by long-term culture-initiating cell assay. The D171N cells showed a drastic increase in the number of colonies. However, long-term culture in complete cytokine liquid medium showed that the D171N cells grew a little, exhibiting lower proliferation ability than control. The percentage of CD34⁺ cells increased slightly, but gradually decreased with a maximum around day 35. At this point, although the percentage of CD34⁺/CD38⁺ cells did not increase in comparison to control cells, the percentage of CD34⁺/CD38⁻ cells increased to 4 %. On day 35, a vast majority of the control cells terminally differentiated into mature myeloid cells and monocytes, whereas the D171N cells contained a large number of immature cells and displayed morphological abnormalities in all 3 hematopoietic lineages. Cell cycle analysis revealed that most of the D171N cells accumulated in G1 phase on day 53 when the cells stopped proliferating. These results indicated that the D171N cells had no proliferation ability although the mutant probably gives rise to the multi-lineage dysplasia of hematopoietic cells with increase in the number of blasts that is the main characteristic of MDS.

Because the D171N cells showed low expression level of BMI1, we next performed stepwise transduction of BMI1 following D171N. In CFC assay, stepwise vector-transduced D171N cells seemed to no longer have colony forming ability, whereas stepwise BMI1-transduced D171N cells displayed an increase in both colony forming ability and replating capacity. Moreover, CD34⁺ cell population remained in the stepwise BMI1-transduced D171N cells. Furthermore, the cells showed long-term proliferation with a retained CD34⁺ cell fraction. Morphological findings showed myeloid cell dysplasia with increased blast-like cells.

Taken together, the results of the D171N forced expression demonstrate that the mutation requires collaborating genes for proliferation. EVI1 and BMI1 may add distinct proliferating forces to the D171N cells, reflected in clinical features of MDS patients with their overexpression. Our results support the concept that RUNX1 mutations may become one of the genetic classification categories of myeloid neoplasms.