The well-known mitogenic effects of TSH observed in vivo on the thyroid are not always reproducible of human thyroid cells in vitro where conflicting results have been obtained. In order to clarify this issue, we have used primary cultures of human thyroid cells obtained from normal tissue and maintained in serum-free medium for several days. In this in vitro model we have studied the effect of TSH on growth by measuring three different parameters: [3H]-thymidine incorporation, cell counts, and DNA measurement. Monolayer cultures were plated at both low and high cell density (2 x 10(4) and 8 x 10(4) cells/25 mm well, respectively). Although at either cell density cultures were equally able to functionally respond to TSH in terms of cAMP accumulation a significant growth response to TSH was observed only in low density cultures. In high density cultures TSH had an antimitogenic effect. Moreover, TSH potentiated the mitogenic effect of insulin only in low density cultures. In contrast to TSH, FCS induced a similar proliferative response at both high and low cell density. Following TSH stimulation, cAMP content was always increased, paralleling the effect of growth in low density but not in high density cultures. The cAMP analogues dibutyryl-cAMP and 8-bromo-cAMP, as well as cholera toxin and forskolin, did not mimic the mitogenic effect of TSH but had an antiproliferative effect. In addition, these agents blunted the proliferative effect of insulin. These data suggest that in thyroid cells TSH is able to elicit both a mitogenic and an antimitogenic effect depending on the environmental conditions such as cell density.(ABSTRACT TRUNCATED AT 250 WORDS)