Inhibition of mTORC1/C2 signaling improves anti-leukemia efficacy of JAK/STAT blockade in CRLF2 rearranged and/or JAK driven Philadelphia chromosome-like acute B-cell lymphoblastic leukemia

Abstract

Patients with cytokine receptor-like factor 2 rearranged (CRLF2-re) subgroup Philadelphia chromosome-like B-cell acute lymphoblastic leukemia (Ph-like B-ALL) have a high relapse rate and poor clinical outcomes. CRFL2-re Ph-like B-ALL is characterized by heightened activation of multiple signaling pathways, including the JAK/STAT and PI3K/AKT/mTOR pathways. We hypothesized that the combined inhibition by JAK2 and mTOR inhibitors would induce an additive antileukemia effect in CRLF2-re Ph-like B-ALL. In this study, we tested the antileukemia efficacy of the type I JAK inhibitor ruxolitinib and type II JAK inhibitor NVP-BBT594 (hereafter abbreviated BBT594) [1] alone and combined with allosteric mTOR inhibitor rapamycin and a second generation ATP-competitive mTOR kinase inhibitor AZD2014. We found that BBT594/AZD2014 combination produced robust anti-leukemic effects in Ph-like cell lines in vitro and in patient-derived xenograft (PDX) cells cultured ex vivo. JAK2/mTOR inhibition arrested the cell cycle and reduced cell survival to a greater extent in Ph-like B-ALL cells with CRLF2-re and JAK2 mutation. Synergistic cell killing was associated with the greater inhibition of JAK2 phosphorylation by BBT594 than by ruxolitinib and the greater inhibition of AKT and 4E-BP1 phosphorylation by AZD2014 than by rapamycin. In vivo, BBT594/AZD2014 co-treatment was most efficacious in reducing spleen size in three Ph-like PDX models, and markedly depleted bone marrow and spleen ALL cells in an ATF7IP-JAK2 fusion PDX. In summary, combined inhibition of JAK/STAT and mTOR pathways by next-generation inhibitors had promising antileukemia efficacy in preclinical models of CRFL2-re Ph-like B-ALL.

Keywords: JAK; Ph-like ALL; mTOR.

Figure 1. In vitro antileukemia efficacy of dual JAK2 and mTOR inhibition in Ph-like B-ALL cell lines

MHH-CALL4 and MUTZ-5 cells were treated with 0.25-0.8μM BBT594 (BBT), AZD2014 (AZD), or combinations for 72 h, then the numbers of viable cells were determined by CTG assay. The cell inhibition curves were plotted with the live cell number normalized to those of DMSO-treated controls: (A) MHH-CALL4 cells, (B) MUTZ-5 cells. Treated cells were fixed in 90% methanol and then stained with propidium iodine to determine the effects of the treatments on the cell cycle by flow cytometry: (C) MHH-CALL4 cells, (D) MUTZ-5 cells. The cells were stained with annexin V/DAPI to quantify cell apoptosis by flow cytometry: (E) MHH-CALL4 cells, (F) MUTZ-5 cells. ^* p<0.05, ^** p<0.005, ^*** p<0.0005 as determined by unpaired Student t-test. (G, H) BaF/3 cells expressing the Ph-like B-ALL-associated JAK2 R683G parental (G) and ruxolitinib-“persistent” cells (H) were treated with 0.25-1.0μM BBT594, AZD2014 or the combination for 72 h, and cell viability was analyzed by CTG assay. The viability of cells treated was selected inhibitors was to those of DMSO-treated controls, and expressed as % viable cells.

Figure 2. Inhibitory effects of dual JAK2 and mTOR inhibition on the PI3K/mTOR pathway

(A) Phospho-flow cytometry analysis demonstrated increased phosphorylation of JAK-2(Tyr1008), STAT5(Ty694)), ERK(T202/Y204) and AKT/pS6 [AKT(Ser473)],4E-BP1(T37/46), and rS6(S240/244) after stimulation with TSLP(25ng/mL) for 30 min. MHH-CALL4 and MUTZ-5 cells were treated with indicated compounds for 1 h, then the phosphorylation signals of PI3K/mTOR pathway proteins were determined by phospho-flow cytometry. Heat map data depict the changes in phosphoprotein levels by median fluorescence intensity. Color scale represents normalization (Z-score) of each condition for each signal. (B) MHH-CALL4 cells were stimulated with TSLP for 30min followed by treatment with ruxolitinib (Ruxo), BBT594 (BBT), rapamycin (Rap), or AZD2014 (AZD) or one of their combinations at the indicated doses for 1 h, after which the expression of PI3K/mTOR pathway proteins were probed by Western blot. (C) MHH-CALL4 cells were infected with empty vector (EV), with a wild-type (WT) or with mutant (5A) 4E-BP1 constructs, then treated with indicated concentrations of doxycycline (Doxy) for 72 h to induce p-4E-BP1 expression. Levels of p-4E-BP1 were determined by Western blot. (D) MHH-CALL4 cells transfected with 4E-BP1 constructs were treated with doxycycline (1μg/mL) for 72 h to induce 4E-BP1 and then with ruxolitinib (Ruxo) or BBT594 (BBT) for 72 h. Cell viability was analyzed by CTG assay. ^* p<0.05, ^** p<0.005, ^*** p<0.0005, ^**** p<0.0001 as determined by unpaired Student t-test.

Figure 3. Inhibitory effects of dual JAK2 and mTOR inhibition by Reverse Phase Protein Array (RPPA) analysis

MHH-CALL4 and MUTZ-5 cells were stimulated with TSLP for 30 min, treated with indicated compounds for 1 h, after which the cell lysates were prepared for RPPA analysis. The heatmap showed the differentially expressed proteins in at least on treatment group with adjust ANOVA p (FDR) < 0.1. Pearson distance metric and Ward's minimum variance was used to cluster the samples and proteins.

Figure 4. Antileukemia efficacy of dual JAK2 and mTOR inhibition ex vivo in Ph-like B-ALL patient-derived xenograft (PDX) models

Cells collected from Ph-like B-ALL PDX models were treated with DMSO, ruxolitinib (Ruxo; 1 μM), BBT594 (BBT; 1 μM), rapamycin (Rap; 0.2 μM), AZD2014 (AZD; 0.2 μM), or one of their combinations for 48 h. The cells were then analyzed for (A-C) apoptosis by annexin V/DAPI or (D) the mTOR and JAK2/STAT phosphoproteins gated on CD19+ cells. (D) Heat map data depict the changes in phosphoprotein levels by fluorescence median intensity. Color scale represents normalization (Z-score) of each condition for each signal.

Figure 5. Anti-leukemia efficacy of dual JAK2 and mTOR inhibition in vivo in the P2RY8-CRLF2/JAK2 mutation PDX

(A) Treatment schema for in vivo experiment with PDX 8 (P2RY8-CRLF2/JAK2 T875N) and PDX 13 (R2RY8-CRLF2/JAK2 R683S). Mice were treated with vehicle, ruxolitinib (Ruxo, 2gm/mouse/day) BBT594 (BBT; 100mg/kg/d), AZD2014 (AZD; 30 mg/kg/d), or the combinations when higher than 30% leukemia cells engraftment was detected in peripheral blood (PB) for a total of 7 days (3 days on, 2 days off, 4 days on). 4 h after the last dosing was completed, mice were sacrificed for tumor burden assessment. Human leukemia cell percentages in the peripheral blood, bone marrow (BM), and spleen were measured by quantitative flow cytometry. Tumor burden was represented by spleen weight and morphology (B, F, C and G) and fraction of engraftment (D, H). Heat map data depict the changes in phosphoprotein levels by fluorescence median intensity (E, I).

Figure 6. Anti-leukemia efficacy of dual JAK2 and mTOR inhibition in vivo in the ATF7IP-JAK2 fusion PDX

Mice were injected with PDX 16 (ATF7IP-JAK2) and treated as indicated in Figure 5A. Tumor burden was represented by spleen weight and morphology (A, B), engraftment (C) and signaling (D).