The nucleotide sequence of a rainbow trout transforming growth factor beta (TGF-beta) peptide is presented, which translates into a 382 amino acid precursor molecule containing a 20 amino acid leader and a mature peptide of 112 amino acids. The mature peptide has nine conserved cysteines and a conserved proline (position 36) and glycine (position 46), all characteristics of TGF-beta superfamily molecules. Within the precursor region are three glycosylation sites, two in common with known TGF-beta s, an integrin binding site (RGD) and the tetrabasic peptide cleavage site (RKKR). The full 3' untranslated region (UTR) consists of 542 nucleotides with a polyadenylation signal 16 nucleotides upstream of the poly(A) tail. The 5' UTR contains an open reading frame with the potential to encode an eleven amino acid peptide, which may have significance for regulation of TGF-beta translation. A wide tissue distribution of TGF-beta message was detected by RT-PCR; in blood leukocytes, kidney macrophages, brain, gill, and spleen tissue but not liver. A phylogenetic tree reveals the trout TGF-beta sequence is most related to xenopus TGF-beta 5, with these sequences and that of chicken TGF-beta 4 grouping with mammalian TGF-beta 1 s. The impact of the trout sequence on current theories of TGF-beta isotype evolution is discussed.