Abstract

Although numerous studies have implicated signal transducers and activators of transcription (STAT)5 in Bcr/Abl-dependent transformation and survival, the role of STAT5 in chronic myelogenous leukemia (CML) has not been fully elucidated. In mice, the CML-like myeloproliferative disorder can occur in the absence of STAT5A/5B, but the repopulating and colony-formation defects of STAT5A/5B double knockout (−/−) bone marrow (BM) cells have made it difficult to separate Bcr/Abl-specific and Bcr/Abl-non-specific effects. To clarify the role of STAT5 signaling in P210BCR/ABL leukemogenesis, we have expressed Bcr/Abl in primary murine STAT5A-deficient (−/−) BM cells, which have no demonstrable intrinsic hematopoietic defects. Interestingly, the absence of STAT5A profoundly decreased the incidence of the pure CML-like disease. Only 24% of wild-type mice reconstituted with P210-transduced STAT5A(−/−) BM cells developed CML, compared to 93% of P210-STAT5A(+/+)- and 83% of P210(+/−)-reconstituted mice. The remainder of P210-STAT5A(−/−)-reconstituted mice developed an acute lymphoblastic leukemia (ALL)-like disease (20%) or a CML/ALL mix (56%), underscoring the relative importance of STAT5 signaling in P210-induced myeloid leukemia compared to P210-induced lymphoid leukemia. There was also a trend for increased latency of the CML-like disease in P210-STAT5A(−/−)-reconstituted mice, with P210-STAT5A(−/−)-reconstituted mice having a median survival of 38 days, compared to only 22 days for P210-STAT5A(+/+) animals. Compared to P210-control cells, primary leukemia cells from P210-STAT5A(−/−) mice had decreased constitutive expression of the STAT5 target genes Bcl-XL and oncostatin M, but preserved expression of pim-1 and cis. Of mice that developed CML, P210-STAT5A(−/−)-reconstituted animals had greater splenomegaly and tended to have higher Bcr/Abl expression than P210-control animals, suggesting that enhanced expression of Bcr/Abl might compensate for the absence of STAT5A in these mice. Importantly, there was no difference in the number of unique P210-STAT5A(−/−) or P210-STAT5A(+/+) leukemic clones by Southern blot, indicating that the reduced incidence of CML in STAT5A (−/−) mice was not due to the decreased availability of CML progenitors. Altogether, these findings indicate that STAT5, particularly STAT5A, plays an important role in the efficient induction of the CML-like disease in mice. A greater understanding of the STAT5 target genes important in CML will provide important insights into the pathophysiology of CML, and may provide additional molecular targets for CML therapy.

Author notes

Corresponding author