Oncogenic Kit controls neoplastic mast cell growth through a Stat5/PI3-kinase signaling cascade

Abstract

The D816V-mutated variant of Kit triggers multiple signaling pathways and is considered essential for malignant transformation in mast cell (MC) neoplasms. We here describe that constitutive activation of the Stat5-PI3K-Akt-cascade controls neoplastic MC development. Retrovirally transduced active Stat5 (cS5(F)) was found to trigger PI3K and Akt activation, and to transform murine bone marrow progenitors into tissue-infiltrating MCs. Primary neoplastic Kit D816V(+) MCs in patients with mastocytosis also displayed activated Stat5, which was found to localize to the cytoplasm and to form a signaling complex with PI3K, with consecutive Akt activation. Finally, the knock-down of either Stat5 or Akt activity resulted in growth inhibition of neoplastic Kit D816V(+) MCs. These data suggest that a downstream Stat5-PI3K-Akt signaling cascade is essential for Kit D816V-mediated growth and survival of neoplastic MCs.

Figure 1

Constitutive activation of Stat5 induces mast cell hyperplasia and leukemic cell infiltrates in the gastrointestinal tract of cS5^F mice. (A) Immunohistochemical analysis of stomach walls of mice that underwent transplantation with cS5^F BM cells versus control mice that received GFPv BM cells (n > 6 mice analyzed in each group and a representative example is shown). To detect neoplastic MCs, sections of the stomach mucosa and submucosa of CS5^F mice and control GFPv-mice were immunostained with an antitryptase antibody (i-vi). The presence of tyrosine^694/695 phosphorylated Stat5 (P-Y-Stat5; vii-xii) and serine⁴⁷³ phosphorylated Akt (P-S-Akt; xiii-xviii) was analyzed with specific antibodies on consecutive tissue sections (original magnification, ×400). An overview of each staining on consecutive sections is shown in the 2 panels labeled “Overview” (iii, iv, viiix, xv, xvi, xxi, xxii) for leukemia cS5^F and control GFPv mice (original magnifications, ×25 and ×100, respectively). H&E staining is shown for organotypic structure comparison (xviiii-xxiv). (B) Quantification of P-Y-Stat5, tryptase, and P-S-Akt stainings was performed on consecutive sections using HistoQuest analysis software (results shown in the scattergram plots are the mean of 4 fields of view for each staining). The cutoff values for background staining were chosen manually using the forward/backward gating tool of the HistoQuest software.

Figure 2

SCF-dependent activation of cS5^F promotes the development of murine mast cells in vitro. (A) Bone marrow (BM) cells from cS5^F mice were cultured in presence of SCF (10 ng/mL) or Flt3L (10 ng/mL) and viable cells were enumerated daily using the trypan blue dye exclusion method. (B) cS5^F-BM cells were deprived of SCF for 3 hours and then restimulated for the indicated times. Cell extracts were prepared and analyzed by Western blot with anti–P-Y-Stat5 and anti–Stat5 antibodies. (C) Two-week-old cS5^F-BM cells grown in culture in presence of SCF were analyzed for expression of FcϵRI and c-Kit by flow cytometry. (D) cS5^F-BM–derived MCs (cS5F-BMMCs) were incubated with anti-IgE alone or/and IgE. Alternatively, cells were treated with DMSO or A23187 and stained with toluidine blue. (E) cS5^F-BMMCs and GFPv-BMMCs were treated or not with LY294002 (1 μM) for 3 days and the percentage of viable cells was determined daily with the trypan blue dye exclusion assay. Results shown are representatives of 3 independent experiments.

Figure 3

Immunohistochemical detection of P-Y-Stat5 and P-S-Akt in neoplastic human mast cells. (A) Adjacent bone marrow (BM) sections from a patient with indolent systemic mastocytosis (ISM) and normal bone marrow (normal BM) sections were stained with an antibody against tryptase for MC detection (brownish staining) and an antibody against P-Y-Stat5 (brownish staining). (B) Similarly, bone marrow sections from a patient with indolent systemic mastocytosis were stained with an antibody against P-Ser⁴⁷³-Akt antibody. Higher magnification of the immunostaining (×400) is also shown.

Figure 4

Detection and localization of P-Y-Stat5 and P-S-Akt in isolated neoplastic mast cells. (A) Expression of P-Y-Stat5 and tryptase was also evaluated by immunocytochemistry in isolated neoplastic MCs (from a patient with mast cell leukemia [MCL]) and normal isolated BM cells. (B) Detection of P-Y-Stat5 and P-S-Akt in neoplastic MCs obtained from patients with MCL in the absence or presence of a blocking phospho-peptide. (C) Immunodetection of P-Y-Stat5, P-S-Akt, and tryptase in the human HMC-1 neoplastic MC line. The P-S-Akt staining was also performed in the presence of a blocking phospho-peptide. (D,E) Cytoplasmic and nuclear extracts (CE and NE) were prepared from HMC-1 cells and analyzed by Western blotting using the indicated antibodies. (F) Detection of cytoplasmic P-Y-Stat5 in HMC-1 cells by FACS. Cells were cultured in control medium (solid line, open graph) or 1 μM PKC412 (solid line, gray graph) at 37°C for 4 hours followed by FACS using an antibody against P-Y-Stat5. The dotted line represents the buffer control. The isotype control exhibited a slight shift compared with the buffer control, but did not change after exposure to PKC412 (not shown). Representatives of 3 independent experiments are shown.

Figure 5

Biologic effects of Stat5 proteins on neoplastic mast cell growth. (A) Schematic representation of TAT-Stat5 proteins. DBD indicates DNA binding domain; SH2, Src-homology domain 2; TAD, transactivation domain. (B) Purity of recombinant TAT-Stat5 proteins was assessed by Coomassie blue staining on SDS-PAGE. L indicates bacterial lysate; E, eluate fraction. (C) HMC-1 cells were transduced with the different TAT-Stat5 proteins (10 nM) during 24 hours. Lysates from transduced cS5^F-BM cells were prepared and analyzed by Western blotting with the indicated antibodies. (D) HMC-1 cells were transduced or not with 10 nM of the different TAT-Stat5 proteins during 9 days and the number of living cells was determined every 3 days using the trypan blue dye assay. Results are the mean of 3 independent experiments. (E) HMC-1 cells were transduced with recombinant lentiviruses expressing a Stat5 shRNA or a luciferase shRNA as control. GFP⁺ cells were sorted by flow cytometry and cultured in normal medium for 9 days. Cell extracts were then prepared and analyzed by Western blot with indicated antibodies. (F) The number of viable GFP⁺ cells expressing Stat5 or luciferase shRNAs was also enumerated every 3 days. Representatives of 2 independent experiments are shown.

Figure 6

Analysis of Stat5 function in human CD34⁺ cells via c-Kit signaling. (A) Purified human CD34⁺ cells from umbilical cord blood were stimulated with recombinant SCF (10 ng/mL) or IL-3 (10 ng/mL) for 30 minutes. Tyrosine phosphorylation of Stat5 was evaluated by Western blot analysis using an anti–P-Y-Stat5 antibody. SCF-mediated activation of Stat5 in CD34⁺ cells was also analyzed by immunocytochemistry with an anti–P-Y-Stat5 (AX1) antibody. (B) Human CD34⁺ cells were transduced or not (PBS) with the different TAT-Stat5 proteins (10 nM) for the indicated times. After extensive washes, the presence of recombinant TAT-Stat5 proteins in CD34⁺ cells were analyzed by Western blotting with the indicated antibodies. (C) CD34⁺ cells (50 × 10³; n = 4) were cultured in the presence of SCF (10 ng/mL) for 20 days. TAT-Stat5 proteins (10 nM) were added or left away (PBS) every 2 days in culture and cells were enumerated every 5 days. (D) Transduced cells grown for 30 days in the presence of SCF were analyzed for expression of FcϵR1 by flow cytometry. Immunocytochemical analysis (IC) was also performed on TAT-cS5^F protein-transduced cells with an antitryptase antibody. (E) The presence of metachromatic granules was detected after staining with toluidine blue. Results shown are representative of 4 independent experiments.

Figure 7

P-Y-Stat5 interacts with PI3-kinase in HMC-1 cells. (A) Stat5 (left panel) or p85 (right panel) was immunoprecipitated from HMC-1 cell extracts with specific or isotype control antibodies. The presence of p85 (left panel) or Stat5 (right panel) in the immunoprecipitates was detected by Western blot. (B) HMC-1 cells were also transduced with 100 nM TAT-wtAkt or TAT-dnAkt for 9 days and the number of living cells was determined every 3 days using the trypan blue dye assay. Results are the mean of 3 independent experiments. (C) Lysates from transduced HMC-1 cells were analyzed by Western blotting using anti–HA and anti–Akt antibodies. (D) HMC-1 cells were transduced or not (NaCl) with the TAT-wtStat5, TAT-dnStat5, TAT-wtAkt, and TAT-dnAkt fusion proteins or a mixture of TAT-wtStat5/TAT-wtAkt or TAT-dnStat5/TAT-dnAkt proteins (ratio 1:1) for 3 days. Cell growth was determined using the trypan blue dye assay. (E) Extracts from HMC-1 cells, either untreated (NaCl) or transduced with TAT-cS5^F and TAT-dnStat5 recombinant proteins for 6 days, were analyzed by Western blot with anti–P-S-Akt and anti–Akt antibodies. Representatives of 3 independent experiments are shown.