Glutamate peripherally administered exerts somatostatin-releasing action in the conscious rat

Abstract This study was designed to determine whether glutamate is able to stimulate somatostatin release from in vivo conscious animals when somatostatin release is monitored in unanaesthetized rats stereotaxically implanted with a push-pull cannula in the median eminence. One week after implantation, the median eminence was perfused with artificial cerebrospinal fluid alone or with the addition of either CGS 19755, an N-methyl-D-aspartate (NMDA) receptor antagonist (1(10(-6) M), or glutamate (10(-5) M). The latter (which is able to cross the brain-blood barrier at large doses) was also peripherally administered (1 g/kg ip). Median eminence perfusate samples were collected every 15 min and somatostatin was measured by a sensitive radioimmunoassay. In rats receiving ip glutamate injection, somatostatin release from the median eminence was significantly increased (73.3+/-10.4 versus 24.8 +/- 6.2; P < 0.01; n = 5) when compared to baseline levels measured in the same animals, but no effect was observed when local perfusion of the median eminence with glutamate (10(-4) to 10(-5) M; n = 6) was performed. Glutamate-induced somatostatin release was completely blunted (n = 5) by prior local administration of CGS 19755 (10(-5) M), a potent NMDA-type receptor antagonist able to cross the blood-brain barrier. In contrast, administration of glutamic acid diethylester, a competitive antagonist of non-NMDA receptors, at doses of 10(-4) M (n = 4), was not able to alter this response. Our results are the first in vivo evidence in favour of a neuroendocrine role for glutamate on somatostatin release whose site of action seems to exclude the median eminence.