Targeting STAT5 or STAT5-Regulated Pathways Suppresses Leukemogenesis of Ph+ Acute Lymphoblastic Leukemia

Abstract

Combining standard cytotoxic chemotherapy with BCR-ABL1 tyrosine kinase inhibitors (TKI) has greatly improved the upfront treatment of patients with Philadelphia chromosome-positive (Ph+) acute lymphoblastic leukemia (ALL). However, due to the development of drug resistance through both BCR-ABL1-dependent and -independent mechanisms, prognosis remains poor. The STAT5 transcription factor is activated by BCR-ABL1 and by JAK2-dependent cytokine signaling; therefore, inhibiting its activity could address both mechanisms of resistance in Ph+ ALL. We show here that genetic and pharmacologic inhibition of STAT5 activity suppresses cell growth, induces apoptosis, and inhibits leukemogenesis of Ph+ cell lines and patient-derived newly diagnosed and relapsed/TKI-resistant Ph+ ALL cells ex vivo and in mouse models. STAT5 silencing decreased expression of the growth-promoting PIM-1 kinase, the apoptosis inhibitors MCL1 and BCL2, and increased expression of proapoptotic BIM protein. The resulting apoptosis of STAT5-silenced Ph+ BV173 cells was rescued by silencing of BIM or restoration of BCL2 expression. Treatment of Ph+ ALL cells, including samples from relapsed/refractory patients, with the PIM kinase inhibitor AZD1208 and/or the BCL2 family antagonist Sabutoclax markedly suppressed cell growth and leukemogenesis ex vivo and in mice. Together, these studies indicate that targeting STAT5 or STAT5-regulated pathways may provide a new approach for therapy development in Ph+ ALL, especially the relapsed/TKI-resistant disease.Significance: Suppression of STAT5 by BCL2 and PIM kinase inhibitors reduces leukemia burden in mice and constitutes a new potential therapeutic approach against Ph+ ALL, especially in tyrosine kinase inhibitor-resistant disease. Cancer Res; 78(20); 5793-807. ©2018 AACR.

Fig. 1.. Effect of STAT5 silencing on cell growth, apoptosis and leukemogenesis of Ph+ leukemia cell lines.

(A-D): Western blot (A), cell growth (B), apoptosis (C), and colony formation (D) of untreated or doxycycline (Doxy)-treated shSTAT5 Ph+ cell lines. Cell growth was determined by MTT assays and data are expressed as optical density (O.D). For Z181 cells, MTT assays were performed until 144 h. Apoptosis was measured as the % of cells expressing activated Caspase 3/7 by flow cytometry analysis. For colony formation assays, cells were seeded in methylcellulose plates (2,500–5,000 cells/dish), with or without Doxy. Colonies were counted 7–10 days after plating. Results are expressed as % inhibition of colony formation in Doxy-treated versus untreated plates. (no asterisk=not significant, *p<0.05, **p<0.01, ***p<0.001). (E) Kaplan-Meier survival plot of untreated or Doxy-treated NSG mice injected with shSTAT5 Ph+ leukemia cell lines. P values indicate statistical significance of the difference in survival between untreated and Doxy-treated mice.

Fig. 2.. Effect of STAT5 silencing on primary Ph+ or Ph-like ALL cells.

A and B) Western blot and colony formation of Ph+ ALL primary cells (newly diagnosed or relapsed/TKI-resistant) (A) or a newly diagnosed Ph-like ALL sample (B) transduced with a scramble or a STAT5 shRNA lentivirus. 100,000 cells/plate were seeded in the presence of puromycin. Colonies were counted 7–10 days after plating. Results are expressed as % inhibition of colony formation of shSTAT5-transduced versus scramble-transduced Ph+ or Ph-like ALL primary cells (***p<0.001).

Fig. 3.. Effect of the STAT5 inhibitor IST5–002 on colony formation and leukemogenesis of Ph+ ALL cells.

Western blot (A) and colony formation (B) of IST5–002-treated primary Ph+ ALL cells. For western blot analysis, cells from Ph+ ALL samples were left untreated (Ctrl) or treated with 5 or 10 μM IST5–002 for 24 h. Whole cell lysates were blotted with anti-p-STAT5, STAT5 and β-actin antibodies. For colony formation assays, cells (100,000/dish) were seeded in methylcellulose plates in presence of medium only or with 5 or 10 μM IST5–002. Colonies were counted after 10 days. Results are expressed as % inhibition of colony formation from IST5–002-treated versus untreated cells (***p<0.001). (C) Kaplan-Meier survival plot of NSG mice injected with SUP-B15 Ph+ ALL cells (10⁶ cells/mouse) and treated with vehicle only or with IST5–002 (50 mg/kg/14 consecutive days) starting 7 days post-cell injection. *p<0.05 denotes statistical significance of the difference in survival between IST5–002-treated versus diluent-treated mice. (D) Peripheral blood leukemia load measured as percentage of CD19+ cells in NSG mice injected with ALL#557 sample before or after treatment with vehicle or IST5–002 (100 mg/kg/14 consecutive days) (right panels); (left panel) fold-change (log₂ % CD19 positivity) in leukemia burden based on data shown in right panels (NS=not significant, *p<0.05, **p<0.01, ***p<0.001).

Fig. 4.. Effect of BIM silencing or BCL-2/MCL-1 expression on the apoptosis of STAT5-silenced Ph+ ALL cells.

(A, B, E) Western blots show expression of BCL-2, MCL-1 and BIM in the indicated STAT5-silenced Ph+ ALL cell lines (A) and in the shSTAT5-BV173 derivative line transduced with the Doxy-regulated shBIM lentivirus (B), or in shSTAT5-BV173 ectopically expressing MCL-1 or BCL-2 (E, left), or in the shSTAT5-shBIM-BV173 line expressing MCL-1 or BCL-2 (E, right); (C, D) % of apoptotic cells detected by Annexin V staining in shSTAT5-BV173 parental and derivative cell lines, untreated or Doxy treated to silence STAT5 expression.

Fig. 5.. Effect of PIM kinase inhibition or STAT5 silencing on signal transduction pathways in Ph+ ALL cells.

Western Blot analysis of PIM-1-regulated proteins in AZD1208-treated Ph+ ALL lines (A) and primary Ph+ ALL samples (B) or in STAT5-silenced Ph+ ALL lines (C).

Fig. 6.. Effect of AZD1208 and Sabutoclax on colony formation of Ph+ leukemia cell lines and primary Ph+ ALL cells.

Methylcellulose colony formation of BV173, SUP-B15 and BV173 (T315I) cell lines (A) or primary Ph+ ALL samples (B), untreated or treated with AZD1208 (3 μM), Sabutoclax (80 nM), or a combination of AZD1208 and Sabutoclax. Colonies were counted 7–10 days after plating; results are expressed as % inhibition of colony formation from drug-treated versus untreated cells (*p<0.05, **p<0.01, ***p<0.001).

Fig. 7.. Effect of AZD1208 and Sabutoclax on Ph+ ALL burden in NSG mice.

(A) Percentage of CD19+ cells in NSG mice injected with ALL#004, ALL#557 or ALL#536 sample before or after treatment with vehicle only, AZD1208, Sabutoclax, or AZD1208 and Sabutoclax; (B) fold-change (log₂ % CD19 positivity) in leukemia burden based on data shown in the A panels (NS=not significant, *p<0.05, **p<0.01, ***p<0.001).