Effect of cyclic GMP-dependent vasodilators on the expression of inducible nitric oxide synthase in vascular smooth muscle cells: role of cyclic AMP.

1. In the present study we examined whether interleukin-1 beta (IL-1 beta) increases the activity of adenylyl cyclase in vascular smooth muscle cells and determined its role in the cytokine-induced expression of the inducible nitric oxide synthase (iNOS) and activation of nuclear transcription factor-kappa B (NF-kappa B). In addition the interaction between cyclic AMP- and cyclic GMP-elevating agonists on the IL-1 beta-stimulated expression of iNOS was examined. 2. Exposure of vascular smooth muscle cells to IL-1 beta stimulated the formation of cyclic AMP but not of cyclic GMP. The intracellular level of cyclic AMP reached a maximum within 1 h and then gradually declined over the next 5 h. This IL-1 beta (60 u ml-1)-stimulated formation of cyclic AMP was modest (about 3 fold at 60 u ml-1 for 1 h) compared to that evoked by isoprenaline (about 9 fold at 3 x 10(-6) M for 2 min). 3. The IL-1 beta (60 u ml-1 for 24 h)-stimulated accumulation of nitrite, which was taken as an index of NO production, was concentration-dependently increased by preferential inhibitors of cyclic AMP-dependent phosphodiesterases (rolipram and trequinsin). This effect was reproduced by a specific activator of the cyclic AMP-dependent protein kinase(s) A, Sp-8-CPT-cAMPS (10(-4) M) but was prevented by a specific inhibitor of cyclic AMP-dependent protein kinase(s) A, Rp-8-CPT-cAMPS (10(-4) M). These compounds alone [rolipram (10(-6) M), trequinsin (3 x 10(-6) M) and Sp-8-CPT-cAMPS (10(-4) M)] slightly but significantly increased the release of nitric oxide while Rp-8-CPT-cAMPS elicited no such effect. 4. Inducible NOS protein was expressed in IL-1 beta (30 u ml-1, 24 h)-stimulated smooth muscle cells as assessed by Western blot analysis. The level of iNOS protein was markedly increased in smooth muscle cells which had been exposed to IL-1 beta in combination with either rolipram (3 x 10(-6) M) or Sp-8-CPT-cAMPS (10(-4) M) but was reduced in those exposed to IL-1 beta and Rp-8-CPT-cAMPS (10(-4) M). A weak expression of iNOS protein was found in smooth muscle cells which had been exposed to either Sp-8-CPT-cAMPS or rolipram alone for 24 h while Rp-8-CPT-cAMPS elicited no such effect. 5. Exposure of smooth muscle cells to IL-1 beta (30 u ml-1) for 30 min increased the level of NF-kappa B-DNA complexes in nuclear extracts as detected by electrophoretic mobility shift assay. Similar levels of NF-kappa B-DNA complexes were found in cells which had been exposed to IL-1 beta in combination with either Sp-8-CPT-cAMPS (10(-4) M), trequinsin (10(-6) M) or rolipram (10(-6) M). None of the modulators alone affected the basal level of NF-kappa B binding activity. 6. NO-donors [sodium nitroprusside (SNP) 10(-4) M; dinitrosyl-iron-di-L-cysteine-complex (DNIC), 10(-4) M; 3-morpholino-sydnonimine (SIN-1), 10(-4) M] and atrial natriuretic factor (10(-6) M) significantly increased the IL-1 beta (30 or 60 u ml-1, 24 h)-stimulated expression of iNOS protein and activity as assessed indirectly by the conversion of oxyhaemoglobin to methaemoglobin. In the absence of IL-1 beta, SNP (10(-4) M, 24 h) but not the other cyclic GMP-dependent vasodilators caused a modest expression of iNOS protein. No such effect was found in smooth muscle cells exposed to SNP in combination with Rp-8-CPT-cAMPS (10(-4) M) while an increased level of iNOS protein was found in those exposed to SNP in combination with either Sp-8-CPT-cAMPS (10(-4) M) or rolipram (3 x 10(-6) M). 7. Exposure of vascular smooth muscle cells to either S-nitroso-L-cysteine (Cys-SNO, 10(-4) M), SNP (10(-4) M) or SIN-1 (10(-4) M) for 35 min affected minimally the basal activation of NF-kappa B but abolished that evoked by IL-1 beta (30 u ml-1 added during the last 30 min). However, addition of Cys-SNO following the stimulation with IL-1 beta (during the last 5 min of the 30 min exposure period) reduced the level of NF-kappa B-DNA complexes only slightly. 8. These data indicate that the cyclic AMP-dependent pathway plays a decisi