Thrombopoietin/MPL participates in initiating and maintaining RUNX1-ETO acute myeloid leukemia via PI3K/AKT signaling

Abstract

Oncogenic mutations in components of cytokine signaling pathways elicit ligand-independent activation of downstream signaling, enhancing proliferation and survival in acute myeloid leukemia (AML). The myeloproliferative leukemia virus oncogene, MPL, a homodimeric receptor activated by thrombopoietin (THPO), is mutated in myeloproliferative disorders but rarely in AML. Here we show that wild-type MPL expression is increased in a fraction of human AML samples expressing RUNX1-ETO, a fusion protein created by chromosome translocation t(8;21), and that up-regulation of Mpl expression in mice induces AML when coexpressed with RUNX1-ETO. The leukemic cells are sensitive to THPO, activating survival and proliferative responses. Mpl expression is not regulated by RUNX1-ETO in mouse hematopoietic progenitors or leukemic cells. Moreover, we find that activation of PI3K/AKT but not ERK/MEK pathway is a critical mediator of the MPL-directed antiapoptotic function in leukemic cells. Hence, this study provides evidence that up-regulation of wild-type MPL levels promotes leukemia development and maintenance through activation of the PI3K/AKT axis, and suggests that inhibitors of this axis could be effective for treatment of MPL-positive AML.

Figure 1

t(8;21)–positive AML cells express MPL and respond to THPO induction. (A) Expression Correlation View (2856 probe sets) of 285 human AML samples, adapted from Valk et al. The 16 clusters identified on the basis of the Correlation View are shown by different colors alongside the Correlation View (1-16). Clusters 9 and 13, aggregating all AML cells with inv(16) and t(8;21), respectively, are indicated by horizontal lines. Expression of MPL (detected by 3 independent probe sets 207550_at, 211903_s_at, 216825_s_at) is plotted in the latter 3 columns in which the bars are proportional to the MPL expression level. (B) Relative proliferation levels of 86 human AML cells in response to THPO as single cytokine. Samples were incubated with THPO or PBS for 3 days, and proliferation was tested by ³H-thymidine incorporation. The fold increase in proliferation with THPO was normalized to that without cytokine and ordered by increasing response. Samples carrying t(8;21) translocation (black bar) and inv(16) rearrangement (“v” mark) are shown. (C) MPL transcript levels determined by quantitative RT-PCR in human AML samples that were subjected to proliferation assays. The MPL levels are the average of duplicate values and relative to the average value of 6 healthy CD34⁺ BM samples (REL). Correlation between MPL expression and THPO response was performed by R statistics (R = 0.572, P < .001).

Figure 2

MPL cooperates with RUNX1-ETO in leukemia development in mice. (A) Experimental design. Top: diagram of retroviral constructs used in the transplantation experiment. Bottom: the transduction-transplantation assay. BM cells (circles) harvested from wild-type mice are cotransduced with MIG-R1E, MIG-R1E9a, or MIG, and MID or MID-Mpl retroviruses were GFP-sorted and transplanted into irradiated recipient mice. (B) Kaplan-Meier plot showing the survival of mice transplanted with BM cells transduced with MIG-R1E/MID-MPL (dashed line/circle, n = 5), MIG-R1E9a/MID-MPL (solid line, circle; n = 12), MIG-R1E9a/MID (solid line, open circle; n = 11), MIG-R1E/MID (dashed line, open circle; n = 8), MID-MPL (dotted line, circle; n = 4), and MID (dotted line; n = 4); mice were followed for 210 days (experimental endpoint). (C) Kaplan-Meier survival curve of secondary transplants from MIG-R1E/MID-MPL (dashed line; n = 5), MIG-R1E9a/MID (solid line, circle; n = 3) leukemic cells, and MID-MPL BM (dotted line; n = 4). (D) Kaplan-Meier survival curve of secondary transplantations with MIG-R1E9a/MID (left; n = 6) and MIG-R1E/MID-MPL (right line; n = 6) leukemic cells, treated with vehicle (dashed line) or INCB18242 (solid line), from day 14 after transplantation (arrow).

Figure 3

Pathology of RUNX1-ETO/MPL myeloid leukemia. (A) Quantification of peripheral blood WBCs from recipient mice expressing MID-Mpl, MIG-R1E9a/MID-MIG, MIG-R1E9a/MID-Mpl, and MIG-R1E/MID-Mpl. Individual values (triangle) and mean (line) are represented. (B) Representative immature cells detected in peripheral blood of MIG-R1E/MID-Mpl (left) and MIG-R1E-MID (right) leukemic mice; original magnification 100×. (C) Representative flow cytometric analysis of GFP(+)/hCD4(+) gated peripheral blood from MIG-R1E9a/MID (top) and MIG-R1E9a/MID-Mpl (bottom) leukemic mice, for lineage (CD41, CD3 B220, Mac1, Gr1, and Ter119) and leukemic cell marker (c-kit). (D) Spleen cross sections showing the architecture of wild-type (wt; 16 weeks old) spleen, and MIG-R1E9a/MID-Mpl (14 weeks old), and MIG-R1E/MID (26 weeks old) spleens from leukemic mice (original magnification ×50). (E) Time course analysis of erythroid progenitors in peripheral blood of recipients transduced with MID-Mpl (n = 8). The percentage of donor cells hCD4(+) cells expressing Mpl (left) and Ter119(+) (right) cells.

Figure 4

Endogenous Mpl activated by PL2 induces AML in cooperation with R1E in mice. (A) Experimental design. Top: schematic representation of retroviral constructs used in the transplantation assay. Bottom: the transduction-transplantation assay. BM cells (circles) harvested from wild-type mice are cotransduced with either MIG or MIG-R1E and MID or MID-PL2 retroviruses, and transplanted into 4- to 8-week-old irradiated recipient mice. (B) Kaplan-Meier plot showing survival curve of mice transplanted with MIG-R1E/MID-PL2 (dotted line; n = 11), MIG-R1E/MID (short dashed line; n = 12), MIG/MID-PL2 (gray line; n = 8), and MIG (solid line, n = 8) transduced BM cells; secondary transplantations of MIG-R1E/MID-PL2 leukemic cells (long dashed line; n = 12); experimental end point: 24 weeks. (C) Flow cytometric analysis of hCD4 (cells expressing PL2) and Mpl receptor expression in GFP-gated peripheral blood leukemic cells from MIG-R1E/MID-PL2 (top) and MIG-R1E/MID (bottom) mice compared with untransduced cells (gray shaded). (D) Flow cytometric analysis of the expression of lineage and c-kit markers in (GFP(+)hCD4(+)–gated cells from MIG-R1E/MID-PL2 mice.

Figure 5

The expression of Mpl is not regulated by R1E. (A) Relative expression levels (REL) of Mpl transcript in cDNA from GFP(+)-sorted BM progenitors transduced with MIG, MID-PL2, or MIG-R1E, using qRT-PCR and normalized to the average of MIG sample; experiments were performed in triplicate. (B) Relative expression levels (REL) of Mpl transcript in cDNA from GFP(+)-sorted MIG-R1E/MID-PL2, MIG-R1E/MID-MPL, and MIG-R1E/MID (n = 3) leukemic cells, using qRT-PCR and normalized to MIG-R1E/MID-Pl2; experiments were performed in triplicate. (C) Epression of Mpl in cell surface of hCD4(+)- or GFP(+)-gated Lin(−) BM cells pretransduced with MID-MPL (left), MIG (middle left), and MIG-R1E (middle right) retroviruses was determined by flow cytometry, and compared with untransduced BM cells (gray shaded). Mpl ecpression in transduced cells is shown in left panel. (D) Expression of MPL in cell surface of GFP(+) MIG-R1E9a/MID (n = 2, left panels) and MIG-R1E/MID-MPL (n = 2, middle panels) leukemic cells was determined by flow cytometry, and compared with untransduced BM cells (gray shaded). Mpl expression in transduced cells is shown in left panel.

Figure 6

Leukemic cells expressing R1E and Mpl are sensitive to THPO signaling through Jak2/Pi3k/Akt. (A) Western blot analysis of signaling proteins activated by Mpl. MIG-R1E/MID (not expressing Mpl) and MIG-R1E/MID-MPL leukemic cells were stimulated with 0, 1, 2.5, 5, and 10 ng/mL THPO after serum starvation. Expression of phospho-Jak2, Jak2, phospho-Stat5, Stat5, phospho-Akt1, Akt1, phospo-Erk1/2, and Erk1/2 was tested by immunoblot analysis. (B) Apoptosis analysis (annexin V(+), 7-AAD-) of MIG-R1E/MID-MPL and MIG-R1E/MID leukemic cells estimated after 48-hour treatment with PBS (black), Thpo (gray), or Thpo and rapamycin (white); P < .01 (*), Student t test). (C) Apoptosis analysis (annexin V(+), 7-AAD-) of MIG-R1E/MID-MPL leukemic cells estimated after 48-hour treatment with PBS (white) or Thpo (black) with pretreatment of no inhibitor (none) or inhibitors for mTor (rapamycin), PI3K (wortmanin), Jak2 (TG101348), and MEK (PD98059). Experiments were performed in quadruplicate; P < .001 (*), Student t test. (D) Cell-cycle analysis (propidium-iodine staining) of MIG-R1E9a/MID-MPL treated as in panel C; subG1 (gray), G0/1 cells (dark gray), S phase cells (white), and G₂/M cells (black); P < .001 Student t test), (*). (E) Proliferation assays of MIG-R1E/MID-MPL leukemic cells after 48-hour culture with IL-3 (6 ng/mL), IL-6 (1 ng/mL), SCF (10 μg/mL), Thpo (20 ng/mL), or the combination of cytokines; each in triplicate. P < .001 (*), Student t test. (F) Kaplan-Meier survival curve of transplantations of MIG-R1E9a/MID-PL2 (left) and MIG-R1E9a/MID (right) leukemic cells in mice injected with mTOR inhibitor rapamycin (solid line, n = 8) or vehicle (dashed line, n = 8). P < .0001 log-rank test.