The neu/erbB-2 protooncogene encodes a transmembrane tyrosine kinase homologous to receptors for polypeptide growth factors. The oncogenic potential of the presumed receptor is released through multiple genetic mechanisms including a point mutation, truncation of non-catalytic sequences and overexpression. The latter mechanism appears to be relevant to human cancers as elevated expression of the neu/erbB-2 gene is frequently observed in solid tumors of various adenocarcinomas. It is therefore conceivable that strategies aimed at the biochemical mechanism of action of the neu/erbB-2 tyrosine kinase may contribute to the treatment of certain human cancers. To this aim we undertook a multiple research approach consisting of the following directions: (i) The neu/erbB-2 ligand--a systematic screening of potential biological sources of the hypothetical hormone molecule, that presumably binds to the neu/erbB-2 protein, resulted in detection of a candidate activity in the medium of certain cultured transformed cells. Partial purification indicated that the factor is a 30-35 kDa glycoprotein. Further studies revealed several biochemical characteristics of the factor that may be helpful for complete purification and structural analysis of this novel hormone. (ii) Signal transduction by neu/erbB-2--using a chimeric receptor approach and various mutants we found that all the oncogenic forms of the neu/erbB-2 are constitutively coupled, both physically and functionally, to a multi-protein complex of signaling molecules. The latter includes the phosphatidylinositol-specific phospholipase C gamma and a phosphatidylinositol kinase. Thus, the metabolism of inositol lipids is probably a major biochemical pathway utilized by the neu/erbB-2 tyrosine kinase. (iii) Tumor inhibitory antibodies--we generated a panel of monoclonal antibodies to the presumed receptor. Surprisingly, some antibodies almost completely inhibited the growth of tumor cells in athymic mice, whereas one antibody significantly accelerated the rate of tumor growth in animals. Interestingly, the inhibitory antibodies conferred a mature phenotype to cultured breast cancer cells, implicating terminal differentiation in tumor retardation.