This study demonstrates that c-kit deletion plus insertion mutations occur in approximately one third of cases of AML with inv(16) karyotype. This cytogenetic variant occurs in approximately 5% of all AML cases, giving an overall incidence for this specific c-kit mutation subtype in AML of around 2%. Although c-kit mutations have been reported previously in a single case of acute myeloid leukaemia, a number of observations suggest that the receptor may be more frequently involved in the pathogenesis of this disease (Beghini et al, 1998). Firstly, the c-kit product is detectable in blast cells from most cases of AML; secondly, the addition of stem cell factor results in the proliferation of some human leukaemic cell lines, such as M07E, and AML blasts and, lastly, c-kit has been reported to be activated, i.e. phosphorylated, in the absence of exogenous SCF in a proportion of AML cases (Wang et al, 1989; Ikeda et al, 1991; Kanakura et al, 1993).
The functional significance of the exon 8 deletion plus insertion mutations at present is unclear. c-kit contains five immunoglobulin-like repeats in the extracellular domain, a single transmembrane domain and a cytoplasmic domain that is split by a kinase insert sequence into the adenosine triphosphate (ATP) binding and phosphotransferase regions (Yarden & Ulrich, 1988). Monoclonal antibody inhibition studies and the analysis of chimaeric mouse–human and deletion mutant kit proteins has enabled the identification of important functional elements in the extracellular region of the receptor (Blechman & Yarden, 1995). The three N-terminal immunoglobulin-like units are involved in ligand binding, the fourth immunoglobulin-like unit is functionally important for receptor dimerization, whereas the function of the fifth domain remains unclear. It is of interest, therefore, that the deletion–insertion mutations described in this study are located in the fifth immunoglobulin-like domain, providing the first evidence in support of an important role for this region of the protein. The structural importance of the deleted region is suggested by the involved amino acids, especially Asp 419, being highly conserved in the c-kit proteins of man (Yarden et al, 1987), rat (Tsujimura et al, 1991), mouse (Qui et al, 1988), cattle (Kubota et al, 1994), chicken (Sasaki et al, 1993) and goat (Tanaka et al, 1997). c-kit receptor activation due to small deletions, albeit in the juxta-membrane domain, has been previously reported. Tsujimura et al (1996) described a seven amino acid deletion (Thr573-His579) in a murine mast cell line (FMA3), and Hirota et al (1998) reported c-kit activation in human gastrointestinal stromal tumours resulting from deletions located within an 11-amino acid stretch (Lys550-Val560). The role of the fourth immunoglobulin-like domain in activating the transforming potential of the related receptor, c-fms, has been highlighted (van Daalen Wetters et al, 1992). Mutations of amino acids 301 and 374 were necessary, but not sufficient, for c-fms dimerization and subsequent cell transformation, in the absence of the ligand (Carlberg & Rohrschneider, 1994). The fifth immunoglobulin-like domain therefore may have some, as yet undefined, function, or with domain 4 may co-operate in receptor dimerization, and Asp 419 changes may therefore lead to possible constitutive activation of c-kit. It is interesting that AML patients with inv(16) karyotype have a complete response rate to treatment of 70–80% and median duration of remission that is significantly prolonged compared to most other subtypes of acute myeloid leukaemia (Grimwade et al, 1998). The influence of the described c-kit mutations in terms of prognosis remains at present unclear.
In conclusion, we report the occurrence of c-kit deletion–insertion mutations in acute myeloid leukaemia and demonstrate their strong association with inv(16) karyotype. By inference from other studies of c-kit and related receptors, the changes reported here would suggest a gain-of-function mutation. However, the biological significance and transforming capacity of these deletion–insertion mutations are at present unclear and are currently being addressed in our laboratory by receptor tyrosine phosphorylation studies and by transfection of mutant-type c-kit genes into human cell lines. Clear and decisive information from such experiments should lead to a greater understanding of kit structure–function relationships as well as the mechanisms underlying regulation of normal and malignant myelopoiesis.