Regulation of tumour necrosis factor production by adrenal hormones in vivo: insights into the antiinflammatory activity of rolipram.

Abstract

1. The role of adrenal hormones in the regulation of the systemic and local production of tumour necrosis factor (TNF alpha) was examined in male Balb/c mice. 2. Intraperitoneal injection of 0.3 mg E. coli lipopolysaccharide (LPS, 0111:B4) led to high levels of circulating TNF alpha without stimulating TNF alpha production in the peritoneal cavity. Systemic production of TNF alpha in response to LPS was increased in adrenalectomized animals and in normal animals treated with the beta-adrenoceptor antagonist, propranolol. The glucocorticoid antagonist, RU 486, did not modify systemic TNF alpha production. These results indicate that systemic TNF alpha production is regulated by adrenaline but not by corticosterone. 3. When mice were primed with thioglycollate, TNF alpha was produced in the peritoneal cavity in response to low dose LPS (1 micrograms). The levels of TNF alpha in the peritoneal cavity were not enhanced by adrenalectomy or by treatment with either propranolol or RU 486, indicating local production of TNF alpha in the peritoneal cavity is not regulated by adrenaline or corticosterone. 4. The phosphodiesterase type IV (PDE-IV) inhibitor, rolipram, inhibited both the systemic production of TNF alpha in response to high dose endotoxin (ED50 = 1.3 mg kg-1) and the local production of TNF alpha in the peritoneal cavity in response to low dose endotoxin (ED50 = 9.1 mg kg-1). In adrenalectomized mice there was a slight reduction in the ability of rolipram to inhibit the systemic production of TNF alpha (ED50 = 3.3 mg kg-1) while the ability of rolipram to inhibit the local production of TNF alpha in the peritoneal cavity was virtually abolished (24% inhibition at 30 mg kg-1). The glucocorticoid antagonist, RU 486, also reduced the ability of rolipram to inhibit local TNF alpha production while propranolol was without effect. 5. Systemic treatment with rolipram increased the plasma concentrations of corticosterone in normal mice but not in adrenalectomized mice indicating that rolipram can cause adrenal stimulation in vivo. 6. In summary, these data indicate that systemic production of TNF alpha in response to high dose endotoxin is controlled differently from the local production of TNF alpha in response to low dose endotoxin. The systemic production of TNF alpha is regulated by catecholamines, but not by corticosterone, while the local production of TNF alpha in the peritoneal cavity is not regulated by basal levels of either catecholamines or corticosterone. 7. These data also show that the ability of rolipram to inhibit the local production of TNF alpha is dependent on the release of corticosterone from the adrenal glands.