Abstract
Hepatocyte growth factor/scatter factor (HGF/SF) is a potent mitogen, motogen, and morphogen for epithelial cells expressing its tyrosine kinase receptor, the c-met proto-oncogene product, and is required for normal development in the mouse. Inappropriate stimulation of Met signal transduction induces aberrant morphogenesis and oncogenesis in mice and has been implicated in human cancer. NK1 is a naturally occurring HGF/SF splice variant composed of only the amino terminus and first kringle domain. While the biological activities of NK1 have been controversial, in vitro data suggest that it may have therapeutic value as an HGF/SF antagonist. Here, we directly test this hypothesis in vivo by expressing mouse NK1 in transgenic mice and comparing the consequent effects with those observed for mice carrying an HGF/SF transgene. Despite robust expression, NK1 did not behave as an HGF/SF antagonist in vivo. Instead, NK1-transgenic mice displayed most of the phenotypic characteristics associated with HGF/SF-transgenic mice, including enlarged livers, ectopic skeletal-muscle formation, progressive renal disease, aberrant pigment cell localization, precocious mammary lobuloalveolar development, and the appearance of mammary, hepatocellular, and melanocytic tumors. And like HGF/SF-transgenic livers, NK1 livers had higher levels of tyrosine-phosphorylated complexes associated with Met, suggesting that the mechanistic basis for the effects of NK1 overexpression in vivo was autocrine activation of Met. We conclude that NK1 acts in vivo as a partial agonist. As such, the efficacy of NK1 as a therapeutic HGF/SF antagonist must be seriously questioned.