Under normal physiological conditions, TGF-betas are synthesized predominantly in biologically latent forms. We have purified recombinant latent TGF-beta and radioiodinated it for use in pharmacokinetic studies. We have demonstrated that the latent complex has a much longer plasma half-life in rats than active TGFbetal, and a very different tissue distribution. This suggests that latent TGF-betal may be the form of choice for clinical use. We have further shown that certain members of the steroid hormone superfamily can induce the secretion of active, rather than latent, TGF-betal in specific target tissues, and we propose that this local induction of an endogenous inhibitor of cell growth by steroids could be exploited to develop a new pharmacology of cancer prevention or treatment. In addition, the possibility that active TGF-beta might interact with its cognate receptor intracellularly was investigated by the use of recombinant DNA techniques to add an endoplasmic reticulum retention sequence on to both active and latent TGF-beta. The results indicated that addition of any C-terminal extension to the TGF-betal molecule interferes with biosynthetic processing and destroys biological activity. Thus strategies to develop antagonists or superantagonists of TGF-beta should avoid modification of this region.
