Fig 2: Combined targeting of JAK2 and c-Myc is effective against CRLF2/JAK2 mutant B-ALLs. (A) MHH-CALL4-pLMS-shSCR, MHH-CALL4-pLMS-shJAK2.209, and MHH-CALL4-pLMS-shJAK2.2826 cells were treated with vehicle (Veh; DMSO) or 500 nM iBET151 (BETi) for 24 h at day 12 after transduction, and mRNA levels for MYC were determined by qPCR. (B) Cells in A were analyzed for the percentage of shRNA-expressing cells (GFP+) by flow cytometry (days 0, 2, and 4 after drug treatment), and data were normalized to day 0. (C) MHH-CALL4 cells were exposed to vehicle (Veh; DMSO), 500 nM JQ1, 1000 nM ruxolitinib (Rux), or the combination (Combo) for 72 h, and the percentage of Annexin V-positive (AnnV+) cells was assessed. (D) Immunoblotting for the indicated targets was performed on lysates from MHH-CALL4 cells treated as in C for 24 h. (E) MHH-CALL4 cells were transduced with constitutive (pLMS-eBFP2) retroviral vectors harboring shMYC.1891 or shRenilla.713 (shRen.713). At day 5 after transduction, transduced and nontransduced populations were passaged in vitro for 12 d with vehicle (Veh; DMSO) or 1000 nM ruxolitinib (Rux). Drugs were replenished to the indicated concentrations, and the percentage of eBFP2+ cells was assessed by flow cytometry every 3 d. (F) Immunoblotting for the indicated targets was performed against MHH-CALL4-pLMS-shRenilla.713 (shRen.713) and MHH-CALL4-pLMS-shMYC.1891 (shMYC.1891) cells treated with vehicle (Veh; DMSO) or 1000 nM ruxolitinib (Rux) for 24 h. (G, H) Cohorts of C57BL/6 wild-type CD45.1+ mice were transplanted with Eµ-Crlf2/Jak2R683G B-ALL. After 12 d, mice were treated with vehicle (Veh), 50 mg/kg JQ1 once per day, 90 mg/kg ruxolitinib (Rux) twice per day, or the combination (Combo; JQ1 and ruxolitinib). Spleen weight, total peripheral WBC count, and percentage of GFP+ cells in the spleens of individual mice in each treatment group were assessed after 3 d of therapy. A photograph of spleens from mice in each treatment arm is shown. n = 3 per group. The ruler scale is in centimeters. (I) Immunoblotting for the indicated targets was performed on lysates from Eμ-Crlf2/Jak2R683G B-ALLs treated with vehicle (Veh; DMSO), 500 nM JQ1, 1000 nM ruxolitinib (Rux), or the combination ex vivo for 8 h. (J) Kaplan-Meier survival curve of recipient mice transplanted with Eµ-Crlf2/Jak2R683G B-ALLs treated as in G from 3 d after transplantation (the gray area indicates the period of drug treatment). β-Actin served as a loading control for all immunoblots. Error bars in A–C represent SEM (n = 2); error bars in E represent SEM (n = 3). (*) P < 0.05; (**) P < 0.01; (***) P < 0.001; (****) P < 0.0001; (ns) nonsignificant.

Fig 3: Crlf2 and mutant Jak2 cooperate to induce murine B-ALL development in vivo. (A) Schematic of the Eµ-Crlf2 transgene. (B) Schematic overview of a fetal liver transplantation mouse model of leukemia. Fetal liver cells isolated at embryonic day 13.5 (E13.5) from Eµ-Crlf2 and wild-type C57BL/6 CD45.2+ mice were transduced with pMSCV-IRES-GFP (MIG), MIG-Jak2WT, MIG-Jak2R683G, or MIG-Jak2P933R and transplanted into sublethally irradiated (9.0 Gy) wild-type C57BL/6 CD45.1+ recipients. The resulting primary leukemias were collected and retransplanted into secondary wild-type C57BL/6 CD45.1+ recipients to derive a transplantation model for subsequent functional in vivo experiments. (C) Kapler-Meier survival curves of cohorts of wild-type C57BL/6 CD45.1+ mice transplanted with primary Eµ-Crlf2/Jak2R683G (M#9; n = 8) and Eµ-Crlf2/Jak2P933R (M#8; n = 6) cells. (D) At terminal disease or the conclusion of the experiment, mice in C were autopsied, and splenic tumor burden was assessed by weight. (E) Hematoxylin–eosin (H&E)-stained spleen and liver sections from moribund wild-type recipient mice transplanted with primary Eµ-Crlf2/Jak2R683G and Eµ-Crlf2/Jak2P933R cells. Bars, 100 µm. (F) Light microscopy of May-Grunwald-Giemsa-stained peripheral blood smears from moribund secondary recipient mice of Eµ-Crlf2/Jak2P933R cells show marked numbers of medium to large blast cells with high nuclear to cytoplasmic ratio. (G) Flow cytometry analysis of Crlf2 and GFP expression on splenocytes isolated from C57BL/6 wild-type mice and moribund secondary recipient mice transplanted with Eµ-Crlf2/Jak2R683G cells. A representative flow cytometry plot is shown. (H) Immunoblotting was performed against the indicated targets using lysates isolated from nontransformed C57BL/6 wild-type splenocytes, GFP+ Eµ-Crlf2/Jak2R683G, and Eµ-Crlf2/Jak2P933R cells. Hsp90 served as loading control. (I) Nontransformed C57BL/6 wild-type splenic B-cells, GFP+ Eµ-Crlf2/Jak2R683G, and Eµ-Crlf2/Jak2P933R cells were assayed by PCR of genomic DNA (gDNA) for VDJ rearrangements using the indicated V gene family-specific VH primers. (J) Flow cytometric analysis of the GFP+ Eµ-Crlf2/Jak2R683G cell population reveals a pre-B-cell immunophenotype (B220lowIgM-IgD-CD43-CD24+ckit+CD11b-CD3-).

Fig 4: RNAi-mediated Jak2 depletion induces an anti-proliferative response independent of cell death in Eµ-Crlf2/Jak2R683G B-ALLs in vivo. (A) Schematic illustration of the experimental design. Splenocytes from moribund bound secondary recipients (mouse 7, mouse 8, and mouse 9 from recipients in Fig. 3F, referred to here as Jak2 knockdown-persistent clone 7 [C#7], clone 8 [C#8], and clone 9 [C#8]) of Eµ-Crlf2/Jak2R683G-pREBIR-shJak2.3323 (Dox on) leukemia cells were harvested and engrafted into tertiary recipient mice. n = 6 for each clone. For each clone, mice were left untreated (Dox off; n = 3) or treated with Dox (Dox on; n = 3) from day 0 after transplantation. (B) Kaplan-Meier survival curve of Dox-treated (Jak2 off; n = 3 per clone) and untreated (Jak2 on; n = 3 per clone) recipient mice of Jak2 knockdown-persistent Eµ-Crlf2/Jak2R683G-pREBIR-shJak2.3323 leukemia cells. Gray shading indicates the period of Dox administration. (C) Representative flow cytometric plot showing eBFP2/dsRed proportions of Jak2 knockdown-persistent Eµ-Crlf2/Jak2R683G-pREBIR-shJak2.3323 cells (GFP+) in the spleens of Dox-treated (Dox-on) and untreated (Dox-off) recipients in B at terminal disease. (D) Immunoblotting against the indicated targets was performed using lysates isolated from splenocytes of individual moribund bound Dox-treated (Dox on; eBFP2+/dsRed+ cells; M5–M8) and untreated (Dox off; eBFP2+/dsRed- cells; M1–M4) recipients in B. (E) Representative flow cytometric plot of BrdU incorporation and 7-AAD staining of eBFP2+/dsRed+ splenocytes from recipients of Eµ-Crlf2/Jak2R683G-pREBIR-shRenilla.713 (shRen.713) and Eµ-Crlf2/Jak2R683G-pREBIR-shJak2.3323 B-ALLs (shJak2.3323) at 48 h after in vivo shRNA induction. (F) Quantification of S-, G0/G1-, and G2M-phase cell percentages of cell cycle analysis from Figure 6E. Error bars represent SEM (n = 2) performed in triplicate (three independent mice). (*) P < 0.05; (ns) nonsignificant (P > 0.05).

Fig 5: Eµ-Crlf2/Jak2 mutant B-ALLs persist despite sustained Jak2 inhibition or depletion in vivo. (A) Schematic representation of the in vivo application of Dox-inducible (Tet-on) pREBIR system. Established Eµ-Crlf2/Jak2 mutant B-ALLs were freshly isolated, transduced with pREBIR harboring shRNAs targeting Jak2 (shJak2.3323 or shJak2.1028) or shRenilla.713, FACS-sorted for transduced populations (eBFP2+), and retransplanted into NSG recipient mice. shRNAs were transiently induced by Dox administration in vivo. (B) Cohorts of mice were transplanted with Eµ-Crlf2/Jak2R683G-pREBIR-shRenilla.713 (mouse 1 [M1] and mouse 2 [M2]; n = 2), Eµ-Crlf2/Jak2R683G- pREBIR-shJak2.3323 (M3–M5; n = 3), or Eµ-Crlf2/Jak2R683G-pREBIR-shJak2.1028 (M6–M8; n = 3) cells and treated with Dox for 48 h at 10 d after engraftment prior to being sacrificed. Immunoblotting was performed for the indicated targets using whole-cell lysates from dsRed+/eBFP2+ splenocytes. Tubulin served as a loading control. (C) GSEA of JAK2 knockdown human MHH-CALL4 cells at day 5 after transduction (MHH-CALL4-pLMS-shSCR [shSCR] relative to MHH-CALL4-pLMS-shJAK2.209 cells [shJAK2.209]) demonstrates significant enrichment of gene sets induced (“Eµ-Crlf2/Jak2R683G_Jak2KD_Genes Down”) and repressed (“Eµ-Crlf2/Jak2R683G_Jak2KD_Genes Up”) in Jak2 knockdown mouse Eµ-Crlf2/Jak2R683G B-ALL cells (Eµ-Crlf2/Jak2R683G-pREBIR-shJak2.3323 relative to Eµ-Crlf2/Jak2R683G-pREBIR-shRenilla.713; 48 h after shRNA induction in vivo). (D) Cohorts of recipient mice were transplanted with Eµ-Crlf2/Jak2R683G-pREBIR-shRenilla.713 (n = 2), Eµ-Crlf2/Jak2R683G-pREBIR-shJak2.3323 (n = 3), or Eµ-Crlf2/Jak2R683G-pREBIR-shJak2.1028 (n = 3) cells for 10 d and treated with Dox for 48 h prior to being sacrificed and analyzed for spleen weight. Error bars represent SEM. n = 2. A photograph of spleens from mice in each treatment arm is shown. The ruler scale is in millimeters. (E) Peripheral blood from recipient mice of Eµ-Crlf2/Jak2R683G-pREBIR-shRenilla.713 (n = 9), Eµ-Crlf2/Jak2R683G–pREBIR-shJak2.3323 (n = 10), or Eµ-Crlf2/Jak2R683G–pREBIR-shJak2.1028 (n = 10) cells at week 3 after injection was analyzed for total peripheral WBC count. All mice were treated with Dox from day 10 after tumor transplantation. (F) Kaplan-Meier survival curve of NSG recipient mice treated as in D. Gray shading indicates the period of Dox administration. (G) Kaplan-Meier survival curve of recipient mice of Eµ-Crlf2/Jak2R683G B-ALL cells treated with vehicle (0.5% methyl cellulose; n = 6) or 90 mg/kg ruxolitinib twice per day (n = 6) from day 3 after injection by oral gavage. Gray shading indicates the period of drug treatment. (H) At terminal disease, mice from G were autopsied, and splenic burden was assessed by spleen weight. (I) Quantification of dsRed+/eBFP2+ proportions in GFP+ splenocytes collected from moribund bound Dox-treated recipient mice from F at terminal disease. (J) Western blot analysis was performed for total Jak2 using lysates isolated from dsRed+/eBFP2+ splenocytes collected from individual Dox-treated moribund recipients of Eµ-Crlf2/Jak2R683G-pREBIR-shRen.713 (M1–M3; n = 3) and Eµ-Crlf2/Jak2R683G-pREBIR-shJak2.3323 (M4–M7; n = 4) cells from F at terminal disease. Tubulin served as a loading control. (K) Immunoblotting was performed against the indicated targets using lysates isolated from splenocytes collected from individual vehicle (M1–M2; n = 2) or ruxolitinib-treated (M3–M6; n = 4) moribund recipients of Eµ-Crlf2/Jak2R683G B-ALLs at terminal disease. β-Actin served as a loading control. (*) P < 0.05; (**) P < 0.01; (****) P < 0.0001.