The PIM inhibitor AZD1208 synergizes with ruxolitinib to induce apoptosis of ruxolitinib sensitive and resistant JAK2-V617F-driven cells and inhibit colony formation of primary MPN cells

Abstract

Classical myeloproliferative neoplasms (MPNs) are hematopoietic stem cell disorders that exhibit excess mature myeloid cells, bone marrow fibrosis, and risk of leukemic transformation. Aberrant JAK2 signaling plays an etiological role in MPN formation. Because neoplastic cells in patients are largely insensitive to current anti-JAK2 therapies, effective therapies remain needed. Members of the PIM family of serine/threonine kinases are induced by JAK/STAT signaling, regulate hematopoietic stem cell growth, protect hematopoietic cells from apoptosis, and exhibit hematopoietic cell transforming properties. We hypothesized that PIM kinases may offer a therapeutic target for MPNs. We treated JAK2-V617F-dependent MPN model cells as well as primary MPN patient cells with the PIM kinase inhibitors SGI-1776 and AZD1208 and the JAK2 inhibitor ruxolitinib. While MPN model cells were rather insensitive to PIM inhibitors, combination of PIM inhibitors with ruxolitinib led to a synergistic effect on MPN cell growth due to enhanced apoptosis. Importantly, PIM inhibitor mono-therapy inhibited, and AZD1208/ruxolitinib combination therapy synergistically suppressed, colony formation of primary MPN cells. Enhanced apoptosis by combination therapy was associated with activation of BAD, inhibition of downstream components of the mTOR pathway, including p70S6K and S6 protein, and activation of 4EBP1. Importantly, PIM inhibitors re-sensitized ruxolitinib-resistant MPN cells to ruxolitinib by inducing apoptosis. Finally, exogenous expression of PIM1 induced ruxolitinib resistance in MPN model cells. These data indicate that PIMs may play a role in MPNs and that combining PIM and JAK2 kinase inhibitors may offer a more efficacious therapeutic approach for MPNs over JAK2 inhibitor mono-therapy.

Keywords: JAK2-V617F; MPN; PIM; kinase inhibitor; therapy.

Figure 1. PIM Inhibitors lack significant efficacy against MPN model cells

A. The MPN model/JAK2-V617F-expressing cell lines HEL and SET2 were cultured with the indicated concentrations of the PIM inhibitor SGI-1776. Total viable cells were determined over time using trypan blue exclusion. B. The MPN/JAK2-V617F-expressing cells HEL, Uke1, and BaF3-JAK2-V617F were treated with a range of concentrations of the PIM inhibitor AZD1208 and relative viable cells were determined by MTS assay. Percent growth relative to DMSO control is plotted versus the log of AZD1208 concentration. Fifty percent growth/inhibition is indicated by a dashed line.

Figure 2. PIM inhibitors synergistically enhance the effect of ruxolitinib on the growth of MPN cells

A. HEL and SET2 cells were cultured with DMSO, the JAK2 inhibitor ruxolitinib (Rux) (0.5 μM for HEL, 0.1 μM for SET2), the PIM kinase inhibitor SGI-1776 (3 μM), and the same concentrations of SGI-1776 and Rux in combination. Total viable cells were determined by trypan blue exclusion over time. The data shown represent the total number of HEL cells after three days of treatment and the total number of SET2 cells after ten days of treatment. The dashed line indicates the starting number of cells (2 × 10⁵) and error bars indicate standard deviation. B. HEL and Uke1 cells were treated with the indicated concentrations of SGI-1776 and ruxolitinib and relative viable cell number was determined by MTS assay. The percent inhibition of drugs alone and in combination was determined and the combination index (CI) for each combination was determined by Compusyn (Combosyn, Inc.). A combination index less than 1 indicates the combination therapy demonstrated synergy compared to the same concentrations of drugs used in mono-therapy treatment. C. HEL and BaF3-JAK2-V617F cells were treated with DMSO, AZD1208 (3 μM for HEL and 0.3 μM for BaF3-JAK2-V617F), ruxolitinib (0.25 μM for HEL and 0.1 μM for BaF3-JAK2-V617F), and AZD1208 plus ruxolitinib in combination. Total viable cells were determined over time by trypan blue exclusion. D. HEL and Uke1 cells were treated with the indicated concentrations of AZD1208 and ruxolitinib and relative viable cell numbers were determined by MTS assay. The percent inhibition of drugs alone and in combination was determined and the combination index (CI) for each combination was determined by Compusyn (Combosyn, Inc.).

Figure 3. AZD1208 enhances apoptosis induced by ruxolitinib

A. BaF3-JAK2-V617F and HEL cells were treated with DMSO, AZD1208 (0.3 μM for BaF3-JAK2-V617F and 3 μM for HEL), ruxolitinib (0.1 μM for BaF3-JAK2-V617F and 0.25 μM for HEL), and AZD1208 plus ruxolitinib in combination. Percent viability over time was determined by trypan blue exclusion. Error bars indicate standard deviation. B. The MPN model cell lines Uke1, BaF3-JAK2-V617F, and HEL were treated with DMSO, and the indicated concentrations of AZD1208 and ruxolitinib alone and in combination. Annexin V binding was determined by flow cytometry after 72 hours for Uke1, 48 hours for BaF3-JAK2-V617F, and 48 hours for HEL. Data is represented as the increase in the percent of annexin V positive cells compared to identically treated DMSO-treated cells. Error bars indicate standard deviation of samples treated in triplicate. C. Uke1, BaF3-JAK2-V617F, and HEL cells were treated with DMSO, AZD1208, and/or ruxolitinib, as indicated. Cell lysates were prepared after 24 hours (Uke1 and HEL) or 48 hours (BaF3) of treatment and immunoblots were performed for cleaved PARP, P-BAD (Ser-112), and tubulin (Uke1 and HEL) and/or total BAD (BaF3 and HEL) as controls, as indicated. Note: drug treatment did not alter total BAD expression in Uke1 cells (Supplementary Fig. 1C).

Figure 4. MPN patient erythroid colony formation is inhibited by AZD1208 mono-therapy and synergistically inhibited with AZD1208 and ruxolitinib combination therapy

A. Peripheral blood mononuclear cells (PBMCs) from two MPN patients were plated in methylcellulose, containing cytokines but lacking erythropoietin (Epo), in the presence of DMSO or SGI-1776 (3 μM). Epo-independent erythroid colonies (EECs) were counted 14 days later. Similarly, cells from two healthy controls (HC) were plated in the same medium containing Epo, and erythroid colonies were determined 14 days later. Data are represented as percent of DMSO samples. B. PBMCs from three MPN patients (left) and two healthy controls (right) were plated, as in A., with the indicated doses of AZD1208. Erthyroid colonies were determined 14 days later and are represented as percent of DMSO samples. C. PBMCs from MPN patients were plated as in A. with DMSO, AZD1208, and ruxolitinib alone or in combination. Drug concentrations used: MPN6–10, 0.2 μM AZD1208 and 0.05 μM ruxolitinib; MPN11–15, 0.1 μM AZD1208 and 0.05 μM ruxolitinib; MPN16, 0.1 μM AZD1208, 0.01 μM ruxolitinib; and MPN17, 0.2 μM AZD1208 and 0.1 μM ruxolitinib. Erythyroid colonies were determined 14 days later and are represented as percent of DMSO samples. Error bars indicate standard deviation. D. Summary of data in C. with mean +/− 95% confidence interval indicated. P value was calculated by paired t-test. All samples were from JAK2-V617F-positive MPN patients: samples MPN1, 6, 7, 8, 12, 13, 15, and 16 were from PV patients; MPN3, 5, 9, 10, 14, and 17 were from ET patients; and MPN2, 4, and 11 were from MF patients.

Figure 5. AZD1208 and ruxolitinib suppress downstream signaling of the mTOR pathway

HEL, SET2, Uke1, and BaF3-JAK2-V617F cells were treated with DMSO (−) or the indicated amounts of AZD1208 and ruxolitinib, alone and in combination. Lysates were prepared following 24 hours for HEL cells, 4 hours for SET2 and Uke1 cells, and 72 hours for BaF3-JAK2-V617F cells. Lysates were analyzed by immunoblotting for P-p70S6K (T389), P-S6 (S235/236), P-4EBP1 (T37/46), and as loading controls tubulin (for HEL and Uke1) and GAPDH (for SET2 and BaF3-JAK2-V617F). Note: the loading control blot for HEL cells in this Fig. is the same as shown in Fig. 3, as the same lysates were analyzed in each Fig.

Figure 6. MPN cells that persistently grow in the presence of JAK2 inhibitors are still sensitive to the combination of ruxolitinib and AZD1208

A. Ruxolitinib persistent Uke1-R and BaF3-JAK2-V617F-R cells growing in 1 μM ruxolitinib were plated in 1 μM ruxolitinib and 0.1 or 0.5 μM AZD1208, as indicated. Relative viable cells were determined by MTS assay after 72 hours. B. Ruxolitinib persistent BaF3-JAK2-V617F-R cells were cultured in 1 μM ruxolitinib alone or with 0.25 or 0.5 μM AZD1208 and total viable cells were determined over time by trypan blue exclusion. C. BaF3-JAK2-V617F-R cells growing in 1 μM ruxolitinib were treated with ruxolitinib alone, 0.25 μM AZD1208, or the combination of the two drugs. Cell viability after two and four days was determined by trypan blue exclusion. D. Apoptosis in BaF3-JAK2-V617F-R cells treated with 1 μM ruxolitinib, 0.25 μM AZD1208, or a combination of the two drugs was detected with annexin V staining and flow cytometry after 24, 48, and 72 hours. Error bars indicate standard deviation.

Figure 7. Exogenous expression of PIM1 induces ruxolitinib resistance

A. PIM1L and PIM1S were expressed from a viral promoter in BaF3-JAK2-V617F cells. Cell lysates were immunoblotted for PIM1, actin as a loading control, and MDM2 for a proteasome inhibitor control. The proteasome inhibitor bortezomib was utilized to stabilize PIM1 expression for easier detection of PIM1S. The exogenous PIM1 proteins were FLAG-tagged thus increasing their molecular weight and slowing their mobility in SDS-PAGE compared to endogenous PIM1 proteins. Mobility of endogenous and exogenous PIM1 proteins are indicated with arrows. B. These cells from A., along with vector control, were cultured with DMSO or ruxolitinib (0.5 μM and 1.0 μM) and total viable cells were determined over time by trypan blue exclusion. Similar results were obtained in three independent experiments.