A point mutation in the extracellular domain of KIT promotes tumorigenesis of mast cells via ligand-independent auto-dimerization

Abstract

Mutations in the juxtamembrane and tyrosine kinase domains of the KIT receptor have been implicated in several cancers and are known to promote tumorigenesis. However, the pathophysiological manifestations of mutations in the extracellular domain remain unknown. In this study, we examined the impact of a mutation in the extracellular domain of KIT on mast cell tumorigenesis. A KIT mutant with an Asn508Ile variation (N508I) in the extracellular domain derived from a canine mast cell tumor was introduced into IC-2 cells. The IC-2(N508I) cells proliferated in a cytokine-independent manner and showed KIT auto-phosphorylation. Subcutaneous injection of IC-2(N508I) cells into the dorsal area of immunodeficient BALB/c-nu/nu mice resulted in the formation of solid tumors, but tumor progression was abrogated by treatment with a tyrosine kinase inhibitor (STI571). In addition, the N508I mutant KIT protein dimerized in the absence of the natural ligand, stem cell factor. Structure modeling indicates that the increased hydrophobicity of the mutant led to the stabilization of KIT dimers. These results suggest that this extracellular domain mutation confers a ligand-independent tumorigenic phenotype to mast cells by KIT auto-dimerization that is STI571-sensitive. This is the first report demonstrating the tumorigenic potential of a mutation in the extracellular domain of KIT.

Figure 1. Characterization of IC-2^N508I cells.

(a) sequence of the c-kit gene. The arrow indicates a heterozygous point mutation in codon 508 (1551 A > T) from a clinical sample from a canine diagnosed with MCT. The base number corresponds to GenBank accession no. AF044249. (b) representative flow cytometry analysis data. Cell surface KIT expression in the indicated IC-2 sublines was detected using an anti-KIT-APC antibody (black), and KIT expression in IC-2^vector cells was used as a negative control (gray). (c) growth curves of IC-2 sublines in the presence or absence of 10 ng/mL SCF. Data represent means ± standard deviations (SD) of 3 independent experiments (n = 5 at each time point). ** p > 0.01, compared to SCF (-) cells at each time point.

Figure 2. STI571 sensitivity and PI3K signaling activity in IC-2^N508I cells.

(a, b) representative cell cycle analysis data and distribution of cells in each cell cycle phase as a function of STI571 treatment. Each subline was serum-starved overnight and then cultured with or without 250 nM STI571 for 24 h. After incubation, cells were fixed with 70% ethanol and stained with propidium iodide. Each data point represents the mean ± SD of 3 independent experiments with duplication. ** p > 0.01, relative to untreated control cells. (c) western blot analysis of each IC-2 subline and (d) quantification of the ratio of phosphorylated/total amounts of each protein. Cells were serum-starved overnight, cultured for 4 h in the presence or absence of 10 ng/mL SCF and/or 250 nM STI571, and expression of the indicated proteins was analyzed in western blots. The phosphorylated/total ratios of cells cultured in the presence of SCF were set to 1.

Figure 3. KIT dimerization in IC-2 sublines.

(a) western blot analysis of IC-2^WT cells for KIT. Cells were treated with indicated concentrations of SCF, followed by the BS₃ crosslinker (1 mM). (b) the mean dimer/monomer ratios ± SD of KIT observed in 3 independent experiments are indicated. Lane numbers correspond to those shown in Figure 3a. Arrows indicate monomeric or dimeric forms of KIT. **, p > 0.01 compared to lane 2; , †, ††, p > 0.05, 0.01 compared to lane 3; , ‡, ‡‡, p > 0.05, 0.01 compared to lane 4; and §, p > 0.01 compared to lane 5, respectively. (c, e) western blot analysis of IC-2^N508I cells (c) and IC-2^N814V cells for KIT (e). Cells were treated with the indicated concentrations of SCF and/or 1 mM BS₃. Arrows indicate monomeric or dimeric KIT. (d, f) the mean dimer/monomer ratios ± SD of KIT from 3 independent experiments are shown. Lane numbers correspond to those shown in Figure 3c and e, respectively. ** p > 0.01 compared to lane 1.

Figure 4. In vivo growth and STI571 sensitivity of IC-2^N508I cells.

(a) growth curves of IC-2^N508I cells in vivo. A total number of 5 × 10⁶ cells were injected subcutaneously into the flanks of BALB/c-nu/nu mice, and tumor sizes were measured every 2 or 3 days. STI571 (100 mg/kg) was administered orally daily, starting 10 days after tumor cell transplantations (Day 0 in the graph). Each data point represents the mean ± SD for 6 animals in each group. ** p > 0.01, compared to vehicle-treated mice. (b) immunohistochemical analysis of IC-2^N508I cells. Tissues were collected on Day 11, and phospho-KIT and Ki-67 staining was conducted. Original magnification, × 200. Bar; 100 μm. (c) percentages of phospho-KIT- or Ki-67-positive cells are indicated as means ± SD obtained from 5 randomly selected microscopic fields. ** p > 0.01, compared to vehicle-treated.

Figure 5. Structure model of the extracellular domain of canine KIT and the effect of KIT inhibitors on dimerization.

(a) modeling of the extracellular KIT domain in the SCF-bound condition. Red and blue ribbons indicate main-chains, green wires indicate the hydrophobic residues, and the blue ball & stick models indicates the residue Asn508. Pink and yellow ribbons indicate stem cell factor. (b) comparison of circle values between wild-type and N508I mutant canine KIT. Circle values for residue 508 in wild-type and N508I canine KIT are indicated. (c) western blot analysis of IC-2^N508I cells treated with STI571 or AMN107. Cells were treated with either reagent for 4 h, followed by chemical crosslinking with BS₃. The monomeric and dimeric forms of KIT are indicated with arrows. (d) the relative mean dimer/monomer ratios ± SD from 3 independent experiments are shown. The dimer/monomer ratios of cells cultured in FBS-containing medium were set to 1. *, **, p > 0.05, 0.01, compared to treatment with 10 nM STI571 or AMN107, as indicated. (e) relative mean dimer phosphorylation levels ± SD from 3 independent experiments are indicated. The dimer/monomer ratio observed in cells cultured in FBS-containing medium was set to 1. *, **, p > 0.05, 0.01 compared to cells treated with 10 nM of the indicated reagents. (f) western blot analysis of IC-2^N508I cells treated with STI571 or AMN107. Cells were treated by each agent for 4 h and lysed without BS₃ treatment.

Figure 6. Schematic representation of mutant KIT phenotypes.

This diagrammatic representation describes the correlation of known KIT mutations with activation patterns. While wild-type KIT dimerizes and gets activated only in the presence of SCF, KIT with mutations in either the extracellular or the juxtamembrane domain dimerizes and becomes activated independently of SCF binding. The variant KIT with a mutation in the tyrosine kinase domain does not require SCF stimulation or dimerization for activation. STI571 sensitivities are also indicated. *1, sensitive to STI571, but less sensitive than mutant KIT.