1. Experiments were performed to examine the effects of putative non-endothelial nitric oxide on the soluble guanylate cyclase activity of severe atherosclerotic aortae from hypercholesterolaemic rabbits fed a cholesterol rich diet for 45 weeks. 2. The guanosine 3'5'-cyclic monophosphate (cyclic GMP) content of aortae from rabbits fed either a control diet or a diet containing 0.3% cholesterol for 45 weeks was quantified in saline extracts or in trichloracetic acid/either extracts by use of a competitive immunoenzymatic assay. Rabbit anti-cyclic GMP immunoglobulin G was covalently linked to the solid phase, in order to avoid false positive results due to high rabbit immunoglobulin G concentrations in the atherosclerotic saline extracts. 3. Saline extracts of atherosclerotic aortae which were harvested immediately after death (intact aortae) contained about 6 fold more cyclic GMP than control aortae when expressed in pmol cyclic GMP mg-1 protein. The cyclic GMP concentrations in trichloracetic acid/ether extracts of atherosclerotic and control aortae expressed in pmol mg-1 fresh tissue were not significantly different. 4. Neointimal-medial explants from atherosclerotic and control aortae were placed in a physiological saline solution and incubated at 37 degrees C for six hours in an incubator gassed with 5% CO2. Before the incubation, the cyclic GMP concentrations in saline extracts of atherosclerotic explants (0.74 +/- 0.27 pmol mg-1) were found to be 17 fold higher than those of control explants (0.043 +/- 0.008 pmol mg-1). The cyclic GMP content of control explants decreased significantly after 6 h of incubation, while that of atherosclerotic explants remained elevated. 5. Chronic administration of NG-nitro-L-arginine methyl ester, a non selective inhibitor of nitric oxide synthases, at 12 mg kg-1 day-1 subcutaneously for one month did not reduce the cyclic GMP concentration of intact atherosclerotic aortae, while that of intact aortae from control rabbits decreased by 63.4 +/- 7.6%. 6. These data show that atherosclerotic aortae harvested immediately after death from hypercholesterolaemic rabbits contain higher concentrations of cyclic GMP than control aortae when measured in saline extracts. In vitro, the persistence of the cyclic GMP production in atherosclerotic neointimal medial explants suggests that the guanylate cyclase is activated by an endogenous mediator. This mediator could be NO, synthesized by non endothelial nitric oxide synthases. The results confirm our previous findings on atherosclerotic blood vessel reactivity, but further studies are needed to elucidate why treatment with NG-nitro-L-arginine methyl ester did not decrease the cyclic GMP content of atherosclerotic rabbit aortae.