CYT387, a novel JAK2 inhibitor, induces hematologic responses and normalizes inflammatory cytokines in murine myeloproliferative neoplasms

Abstract

Activating alleles of Janus kinase 2 (JAK2) such as JAK2(V617F) are central to the pathogenesis of myeloproliferative neoplasms (MPN), suggesting that small molecule inhibitors targeting JAK2 may be therapeutically useful. We have identified an aminopyrimidine derivative (CYT387), which inhibits JAK1, JAK2, and tyrosine kinase 2 (TYK2) at low nanomolar concentrations, with few additional targets. Between 0.5 and 1.5muM CYT387 caused growth suppression and apoptosis in JAK2-dependent hematopoietic cell lines, while nonhematopoietic cell lines were unaffected. In a murine MPN model, CYT387 normalized white cell counts, hematocrit, spleen size, and restored physiologic levels of inflammatory cytokines. Despite the hematologic responses and reduction of the JAK2(V617F) allele burden, JAK2(V617F) cells persisted and MPN recurred upon cessation of treatment, suggesting that JAK2 inhibitors may be unable to eliminate JAK2(V617F) cells, consistent with preliminary results from clinical trials of JAK2 inhibitors in myelofibrosis. While the clinical benefit of JAK2 inhibitors may be substantial, not the least due to reduction of inflammatory cytokines and symptomatic improvement, our data add to increasing evidence that kinase inhibitor monotherapy of malignant disease is not curative, suggesting a need for drug combinations to optimally target the malignant cells.

Figure 1

Efficacy of CYT387 against JAK2-dependent cell lines in vitro. (A) Chemical structure of CYT387. (B) Ba/F3-EpoR parental (cultured in media supplemented with 3 U/mL erythropoietin) or Ba/F3-EpoR-JAK2^V617F cells (cultured in the absence of exogenous cytokines) were plated in 96-well plates over a dose gradient of CYT387 for 3 days at which point cell viability was measured with an MTS tetrazolium salt assay. Values represent mean ± SEM. n = 3; *P < .05. (C) Baf3-EpoR-JAK2^V617F cells were plated in graded concentrations of CYT387 for 1 day (immunoblot) or 2 days (trypan blue exclusion) at which point cell death was measured by trypan blue exclusion and apoptosis was determined by immunoblot with antibodies specific for cleaved caspase 3 and tubulin. Values represent mean ± SEM. n = 3; *P < .05. (D) Ba/F3-EpoR-JAK2^WT or Ba/F3-EpoR-JAK2^V617F cells were serum-starved then plated in 6-well plates over a dose gradient of CYT387 for 16 hours at which point cells were lysed and whole-cell extracts were subjected to immunoblot analysis using antibodies specific for total or phospho-JAK2, total or phospho-ERK1/2, total or phospho-STAT5, or β-actin.

Figure 2

Efficacy of CYT387 against JAK2-dependent malignancy in vivo. (A) Balb/c mice were administered CYT387 by oral gavage and pharmacokinetics were measured by analysis of plasma concentrations at indicated time points. Values represent mean ± SEM. (n = 3) (B) Balb/c mice were subjected to bone marrow transplantation with bone marrow donor cells retrovirally transduced to express JAK2^V617F. Thirty-four days after transplantation, mice exhibited symptoms of MPN as measured by elevated white blood cell counts and hematocrit. Mice were divided into 3 groups and initiated on twice daily oral gavage administration of vehicle control, 25 mg/kg CYT387, or 50 mg/kg CYT387 (n = 12 per group). White blood cell (WBC) count of mice was monitored weekly for 83 days after bone marrow transplantation. (C) Mice were treated as in panel B and hematocrit levels (HCT) of mice were monitored weekly for 83 days after bone marrow transplantation. (D) Mice were treated as in panel B and the percentage of granulocytes in the peripheral blood of mice was monitored weekly for 83 days after bone marrow transplantation. (E) Mice were treated as in panel B and percentage of lymphocytes in the peripheral blood of mice was monitored weekly for 83 days after bone marrow transplantation. (F) Mice were treated as in panel B and the spleen weight of 3 mice was measured at the start of treatment and the spleen weight of all mice was measured at the end of treatment. For comparison purposes, spleen weight from normal mice is included. For panels B through F, values represent mean ± SEM, *P < .05 in a t test comparing 25 mg/kg or 50 mg/kg treatment groups with the 0 mg/kg vehicle control. The gray box indicates the normal range in mice. The arrow indicates initiation of CYT387 treatment.

Figure 3

Histopathology after CYT387 treatment of JAK2-dependent malignancy in vivo. Balb/c mice were subjected to bone marrow transplantation with bone marrow donor cells retrovirally transduced to express JAK2^V617F. Thirty-four days after transplantation, mice exhibited symptoms of MPN as measured by elevated white blood cell counts and hematocrit. Mice were divided into 3 groups and initiated on twice daily oral gavage administration of vehicle control, 25 mg/kg CYT387, or 50 mg/kg CYT387 (n = 12 per group). At day 83 after bone marrow transplantation, all mice were killed and representative histologic sections are shown from: (A) peripheral blood stained for reticulocytes, (B) spleen stained with H&E, (C) liver (H&E), (D) bone marrow (H&E), and (E) bone marrow (reticulin stain). The left column represents 0 mg/kg CYT387, the middle column represents 25 mg/kg CYT387, and the right column represents 50 mg/kg CYT387. Magnification is ×10 or ×40 for insets.

Figure 4

Effect of CYT387 on hematopoietic cell lineages in MPN in vivo. Balb/c mice were subjected to bone marrow transplantation with bone marrow donor cells retrovirally transduced to express JAK2^V617F. Thirty-four days after transplantation, mice exhibited symptoms of MPN as measured by elevated white blood cell counts and hematocrit. Mice were divided into 3 groups and initiated on twice daily oral gavage administration of vehicle control, 25 mg/kg CYT387, or 50 mg/kg CYT387 (n = 12 per group). At day 83 after bone marrow transplantation, all mice were sacrificed and spleen and bone marrow cells were stained with fluorescently conjugated antibodies specific for: (A) Gr1 (Granulocytes), (B) CD19/B220 (B cells) and CD3 cells (T cells), (C) CD71 (early erythroid progenitors) and Ter119 (late erythroid progenitors), (D) cKit⁺CD34⁻ (stem cells/early progenitors) and cKit⁺CD34⁺ (late progenitors), (E) CD41 (megakaryocytes). Values represent mean ± SEM and *P < .05 in a t test comparing 25 mg/kg or 50 mg/kg treatment groups with the 0 mg/kg vehicle control. A complete list of numerical values and statistical analyses is found in supplemental Table 2. Representative dot plots are found in supplemental Figure 4.

Figure 5

Effect of CYT387 on JAK2 V617F⁺ allelic burden in MPN in vivo. Balb/c mice were subjected to bone marrow transplantation with bone marrow donor cells retrovirally transduced to express JAK2^V617F. Thirty-four days after transplantation, mice exhibited symptoms of MPN as measured by elevated white blood cell counts and hematocrit. Mice were divided into 3 groups and initiated on twice daily oral gavage administration of vehicle control, 25 mg/kg CYT387, or 50 mg/kg CYT387 (n = 12 per group). At day 83 after bone marrow transplantation, all mice were sacrificed and genomic DNA was isolated from splenocytes. A qPCR assay was developed to assess the relative level of GFP, normalized to the genomic GAPDH locus. (A) Parental Ba/F3 (GFP-negative) were mixed at varying ratios with Ba/F3-JAK2^V617F cells (GFP-positive). Genomic DNA was isolated and subjected to qPCR using primers specific for GFP or the GAPDH genomic locus. All values were normalized to GAPDH and then to the highest expressing well of genomic DNA from 100% GFP-positive cells. This value was set at 100%. Values represent mean ± SEM. (B) Genomic DNA from splenocytes of mice treated with vehicle control, 25 mg/kg CYT387, or 50 mg/kg CYT387 were subjected to qPCR using primers specific for GFP or the GAPDH genomic locus. The standard curve of Ba/F3 cells shown in panel A was amplified simultaneously, and all values were normalized to GAPDH and then to the highest expressing well of genomic DNA from 100% GFP-positive Ba/F3 cells. Each point represents the genomic GFP level in an individual mouse. (C) The values from panel B were averaged and presented in a bar graph. Values represent mean ± SEM and *P < .05 in a t test comparing 25 mg/kg or 50 mg/kg treatment groups with the 0 mg/kg vehicle control.

Figure 6

Effect of CYT387 on cytokine concentrations during MPN in vivo. Balb/c mice were subjected to bone marrow transplantation with bone marrow donor cells retrovirally transduced to express JAK2^V617F. Thirty-four days after transplantation, mice exhibited symptoms of MPN as measured by elevated white blood cell counts and hematocrit. Mice were divided into 3 groups and initiated on twice daily oral gavage administration of vehicle control, 25 mg/kg CYT387, or 50 mg/kg CYT387 (n = 12 per group). At day 83 after bone marrow transplantation, all mice were sacrificed and serum was harvested from peripheral blood for cytokine analysis with a multiplexed 96-well ELISA-based assay. Bar graphs represent mean cytokine levels for: (A) VEGF (vascular endothelial growth factor), IFN-γ (interferon-γ), IL-3 and IL-17, (B) IL-9, LIF (leukocyte inhibitory factor) and IL-1β, (C) TNF-α (tumor necrosis factor-α), (D) IL-6, (E) IL-10, (F) G-CSF (granulocyte colony stimulation factor). Values represent mean ± SEM and *P ≤ .1 in a t test comparing normal mouse, 25 mg/kg, or 50 mg/kg treatment groups with the 0 mg/kg vehicle control. A complete list of numerical values and statistical analyses is found in supplemental Table 4.

Figure 7

Resistant subclone evasion of CYT387 in vitro. (A) Baf3-EpoR cells expressing JAK2^V617F were subjected to ENU mutagenesis in the presence of CYT387. Cells were plated at 2 × 10⁵ cells per well of a 96 well plate in media with 4μM CYT387. Wells were examined every 5 days for colony outgrowth over a 38-day period. Colonies that grew out were expanded and subsequently serum-starved overnight in the presence of 4μM CYT387 and subjected to immunoblot analysis using antibodies specific for total or phospho-JAK2 or β-actin. (B) Densitometric analysis of immunoblots in panel A. Expression levels for total JAK2 () and phospho JAK2 (■) were normalized to the levels observed in JAK2^WT cells that are sensitive to CYT387. (C) Genomic DNA was isolated from CYT387-resistant Ba/F3 cells and levels of genomic GFP or GAPDH were assessed by quantitative PCR. A standard curve of varying ratios of GFP-positive/GFP-negative Ba/F3 cell mixtures was included as in Figure 5A. Levels of genomic GFP were normalized as in Figure 5 (normalized first to GAPDH, then to the highest value well on the standard curve) and are presented as percent GFP-positive. Values represent mean ± SEM (n = 3).