A TRUNCATED K-SAM PRODUCT LACKING THE DISTAL CARBOXYL-TERMINAL PORTION PROVIDES A REDUCED LEVEL OF AUTOPHOSPHORYLATION AND GREATER RESISTANCE AGAINST INDUCTION OF DIFFERENTIATION

The K-sam gene was originally cloned from KATO-III human gastric cance r cells and Is identical to the bek or keratinocyte growth factor (KGF ) receptor (KGFR) or fibroblast growth factor receptor 2 gene. K-sam g enerates several variant transcripts by alternative splicing, and the most abundant K-sam transcript in KATO-III cells was cloned as the K-s am-IIC3 cDNA, which has the KGF-binding motif and a short carboxyl ter minus lacking a putative phospholipase C-gamma 1 association site, Tyr -769. The K-sam-IIC3 cDNA was distinct from the K-sam-IIC1 cDNA, which was the same as the previously reported KGFR cDNA. The K-sam-IIC1 pro duct contains a long carboxyl terminus with Tyr-769. K-sam-IIC3 showed greater transforming activity in NIH 3T3 cells than did K-sam-IIC1, a nd in gastric cancer cell lines in general, the level of K-sam-IIC3 mR NA was greater than that of K-sam-IIC1 mRNA. Here we report that the K -sam-IIC3 product was less autophosphorylated than the K-sam-IIC1 prod uct in NIH 3T3 transfectants. K-sam-IIC1 transfected keratinocytes sho wed a stronger mitogenic response to KGF than did K-sam-IIC1 transfect ants. Moreover, K-sam-IIC3-transfected L6 myoblast cells hardly differ entiated when cultured in differentiation inducing medium and growth w as not significantly affected, while K-sam-IIC1 transfectants showed a differentiated phenotype,vith a reduced growth rate. These data indic ate the difference in the signal transduction mediated by two KGFR typ e K-sam variants generated by alternative splicing which might he invo lved in certain differentiation and carcinogenesis scenarios.