The paracrine mode of action for IGFs has been hypothesized on the basis of the finding that these peptides are synthesized ubiquitously. As IGFs circulate in very high concentrations (nearly 1 mg/l) they should have an endocrine function; moreover since the liver seems to be responsible for most of the circulating IGF-I it should be recognized that the basis of the "hormonal" IGF-I residues in the liver, whereas the extrahepatic synthesis provides support for local paracrine or autocrine actions. According to the endocrine hypothesis, the liver, which lacks IGF-I receptors, produces both IGF-I and IGF binding protein (IGFBP) and, possibly, the acid-labile subunit. All these components combine in the circulation, and only free IGF-I or small molecular weight complexes cross the capillary wall and reach the target cells. The paracrine hypothesis is based on the fact that fibroblasts are able to synthesize both IGF peptides and IGFBP, whereas keratinocytes are not, although they possess IGF receptors and are sensitive to IGF action. Since affinity of BP3 (which seems to be produced by peripheral fibroblasts) for IGF-I is higher than that of fibroblasts and, to a lesser degree, of keratinocytes, it is likely that IGF-I released from fibroblasts preferentially binds to BP3; the BP3-IGF-I complex reduces the autocrine effects on fibroblasts and preferentially directs the peptide binding to keratinocytes. The secretion of both IGFs and of IGF carrier proteins has been reported in several neoplastic epithelial cell lines.(ABSTRACT TRUNCATED AT 250 WORDS)