Influence of nitric oxide synthase inhibitors on the ACTH and cytokine responses to peripheral immune signals.

Nitric oxide (NO) is an unstable gas that participates in the response of the hypothalamic-pituitary-adrenal (HPA) axis to a variety of immune signals, including turpentine-induced tissue damage and the systemic injection of the pro-inflammatory cytokine interleukin 1-beta (IL-1beta). Studies that have investigated the role of this gas in the intact rat have relied on blockade of NO formation with the NO synthase (NOS) inhibitor N(omega)nitro-L-arginine-methylester (L-NAME). They have suggested that endogenous NO blunts the ACTH response to intravenous (i.v.) IL-1beta in part by exerting an inhibitory influence on the release of hypothalamic peptides such as corticotropin-releasing factor (CRF) from nerve terminals in the median eminence. It must nevertheless be noted that, at present, evidence for this mode of action remains circumstantial. Significant controversy remains regarding the specificity of the compounds used to block NO formation, the characteristics of their effect in terms of dose and timing of administration, the possibility that their effect is restricted to IL-1beta or can be expanded to other pro-inflammatory cytokines, and the question of whether the possibility that they might also influence ACTH release by altering circulating levels of tumor necrosis factor-alpha (TNF-alpha) and IL-6. The purpose of the present study was to elucidate these points. In the first series of experiments, we determined the i.v. IL-1beta-induced ACTH response to various doses of systematically injected L-NAME (1-100 mg/kg). At 10-100, but not 1 mg/kg, L-NAME significantly (P<0.01) augmented the ACTH response to IL-1beta, with a maximum effect observed at 30 and 100 mg/kg. At the 30 mg/kg dose, L-NAME was equally effective in augmenting the ACTH response when administered between 5 and 240 min prior to the cytokine. The effect of L-NAME was fully mimicked by equivalent doses of other arginine derivatives such as N(omega)-monomethyl-L-arginine (L-NMMA) or N(omega)-nitro-L-arginine (L-NNA), indicating that controversy in the published literature concerning the influence of NO on CRF secretion does not appear to be due to the use of different arginine derivatives. The ability of other cytokines such as TNF-alpha and IL-6 to release ACTH and corticosterone was significantly (P<0.01) augmented by blockade of NO formation in a manner similar to that found with IL-1beta. To test the hypothesis that L-NAME might alter ACTH secretion at least in part by modifying the secretion of pro-inflammatory cytokines, we measured plasma concentrations of TNF-alpha and IL-6 following endotoxin injection in the presence or absence of L-NAME. Blockade of NO formation reduced TNF-alpha but increased IL-6 levels in rats administered the lipopolysaccharide (25 microg/kg i.v.). As L-NAME augments the ACTH response to TNF-alpha as well as IL-6, it is improbable that changes in TNF-alpha and IL-6 secretion during immune stimulation represents an important mechanism mediating the inhibitory influence of endogenous NO on the HPA axis activity. Collectively, these results indicate that the systemic injection of L-NAME very quickly augments the stimulatory effect of pro-inflammatory cytokines on ACTH secretion, and does so for at least 4 h. Other arginine derivatives known to block the activity of constitutive NO syntheses, such as L-NMMA and L-NNA, exert an effect that is virtually identical to that of L-NAME. The ability of L-NAME to increase the ACTH response to i.v. IL-1beta is also observed in rats injected with TNF-alpha and IL-6. Because of the opposite effects of L-NAME on the levels of these two cytokines, the influence of arginine derivatives on ACTH release is probably not due to changes in cytokines produced during immune stimulation such as endotoxemia.