Twelve female rats weighing approximately 150 g received in the submaxillary gland a pellet capable of releasing 3.5 microg GHRH/h for 60 days. Another eight sex- and weight-matched animals received placebo pellets in the same place. After two months the animals were killed, heart blood was collected and pituitary and submaxillary glands were carefully dissected. Pituitary GH content in both placebo- and GHRH-treated animals showed similar values, but plasma GH and IGF-I levels were significantly lower in the animals carrying GHRH pellets (P<0.03); these animals also had a significantly higher GH content in the submaxillary gland (19.2+/-8 ng/mg protein) compared with the placebo-treated group (1.1+/-0.3 ng/mg protein). GH mRNA was present only in the submaxillary gland of GHRH-treated rats as determined by PCR-Southern blot and by in situ hybridization methods. It is concluded that high local GHRH levels are capable of inducing transdifferentiation in submaxillary gland cells to synthesize GH.

To test whether salivary tissue can secrete pituitary hormones, female Sprague-Dawley rats were hypophysectomized (hypox) and the following were transplanted to the sella turcica: parotid gland (group 3, n=33), adrenal gland (group 4, n=30), muscle (group 5, n=24). Group 2 (n=21) had the sella turcica filled with dentist's cement. In addition a group of rats (group 1, n=22) remained intact as controls. All groups were followed for 8 months. Daily vaginal smears showed normal cyclicity in controls and constant dioestrus in all hypox groups. Blood samples, taken once every 30 days before and after LHRH stimulation, showed significantly lower (P<0.001) plasma LH values in all hypox groups compared with controls. In group 3, a gradual and significant increase (P<0.05) was observed in the LH response to LHRH in parallel with a partial recovery of oestrous smears. No LH modification was observed in the other hypox groups. Plasma prolactin (PRL) levels were also very low in all hypox groups and were unaltered throughout the study. At the end of the experiments, half the animals were killed by decapitation and the hypothalamic-pituitary areas carefully dissected, homogenized and analysed for LH and PRL content. The remaining animals were perfused with 4% paraformaldehyde to obtain fixing of the whole body tissues. Hypothalamic and transplant areas were carefully dissected, frozen, cut and submitted to immunochemical procedures. LH content in the graft of group 3 animals was markedly (P<0.001) lower than in the control pituitary, but significantly higher (P<0.05) than in the other hypox groups. Immunochemistry showed LH and PRL positive cells in the graft of group 3 animals, whereas neither positive cells, nor LH content were observed in the parotid gland in situ. Experiments were completed with in vitro cultures of parotid glands in the presence or absence (controls) of synthetic hypothalamic hormones or rat hypothalamic extracts. After 1.5 weeks of culture, a significantly higher LH concentration (P<0.05) was observed in the wells treated with synthetic hypothalamic hormones (216+/-46 pg/ml vs 41+/-6 pg/ml in controls). When hypothalamic extracts were used, the LH levels increased more markedly (1834+/-190 pg/ml vs 36+/-6 pg/ml in controls) and those values were maintained during 3 weeks of culture. Immunostaining of these cultures showed a positive LH reaction in the epithelial cells found in the hypothalamic extract-treated wells. Both in vivo and in vitro studies confirm the transdifferentiation of parotid gland tissue to pituitary hormone-producing cells under hypothalamic influence.