Sexual dimorphism and testosterone-dependent regulation of somatostatin gene expression in the periventricular nucleus of the rat brain.

Gender differences in hypothalamic somatostatin (SS) secretion may account in part for the sexually dimorphic patterns of GH secretion in rats. Since males have lower baseline serum GH levels than females, and SS inhibits GH secretion, we hypothesized that the SS neurons in the periventricular nucleus (PeN) of the male rat would have greater biosynthetic activity than those of the female. We tested this hypothesis by measuring SS mRNA in cells in the PeN of intact male and proestrous female rats. Using in situ hybridization and a computerized image analysis system, we measured SS mRNA content in individual cells in the PeN and compared signal levels (autoradiographic grains per cell) between male and proestrous female animals. The signal level of SS mRNA in cells of the PeN was significantly greater in males than in proestrous females (males, 210 +/- 7 grains/cell; females, 158 +/- 5 grains/cell; P less than 0.0005), whereas no difference was observed in SS cells of the frontal cortex (males, 100 +/- 0.8 grains/cell; females, 99 +/- 5.9 grains/cell). This difference in SS mRNA levels is likely to be the result of different hormonal environments exerting an influence on neurons of the hypothalamus. To test the hypothesis that testosterone stimulates SS gene expression in neurons of the PeN, adult male rats were castrated and immediately implanted with either empty (sham; n = 3) or testosterone-containing (n = 3) Silastic implants of a size that would deliver physiological levels of testosterone (3.6 +/- 1.5 ng/ml). We observed that castrated animals had significantly lower levels of SS mRNA signal in neurons of the PeN compared with intact animals (intact, 195 +/- 3 grains/cell; castrated, 159 +/- 6 grains/cell; P less than 0.003) and that physiological levels of testosterone prevent this reduction in SS mRNA levels (castrated testosterone-replaced, 182 +/- 4 grains/cell; castrated, 159 +/- 6 grains/cell; P less than 0.003). Furthermore, testosterone-treated castrates had SS mRNA signal levels indistinguishable from those of intact controls (intact, 195 +/- 3 grains/cell; castrated testosterone-replaced, 182 +/- 4 grains/cell). There was no significant difference in SS mRNA levels in neurons of the frontal cortex (intact, 98 +/- 2 grains/cell; castrated, 98 +/- 3 grains/cell; castrated testosterone-replaced, 102 +/- 2 grains/cell).(ABSTRACT TRUNCATED AT 400 WORDS)