1. Ruthenium(III) reacts with nitric oxide (NO) to form stable ruthenium(II) mononitrosyls. Several Ru(III) complexes were synthesized and a study made of their ability to bind NO, in vitro and also in several biological systems following expression of the inducible isoform of nitric oxide synthase (iNOS). Here we report on the properties of two, related polyaminocarboxylate-ruthenium complexes: potassium chloro[hydrogen(ethylenedinitrilo)tetraacetato]ruthenate+ ++ (=JM1226; CAS no.14741-19-6) and aqua[hydrogen(ethylenedinitrilo)tetraacetato]ruthenium (=JM6245; CAS no.15282-93-6). 2. Binding of authentic NO by aqueous solutions of JM1226 yielded a product with an infrared (IR) spectrum characteristic of an Ru(II)-NO adduct. A compound with a similar IR spectrum was obtained after reacting JM1226 with S-nitroso-N-acetylpenicillamine (SNAP). 3. The effect of JM1226 or JM6245 on nitrite (NO2-) accumulation in cultures of macrophages (RAW 264 line) 18 h after stimulating cells with lipolysaccharide (LPS) and interferon-gamma (IFNgamma) was studied. Activation of RAW264 cells increased NO2- levels in the growth medium from (mean+/-1 s.e.mean) 4.9+/-0.5 microM to 20.9+/-0.4 microM. This was blocked by actinomycin D (10 microM) or cycloheximide (5 microM). The addition of JM1226 or JM6245 (both 100 microM) to activated RAW264 cells reduced NO2- levels to 7.6+/-0.2 microM and 8.8+/-0.6 microM, respectively. N(G)-methyl-L-arginine (L-NMMA; 250 microM) similarly reduced NO2- levels, to 6.1+/-0.2 microM. 4. The effect of JM1226 or JM6245 on NO-mediated tumour cell killing by LPS+IFNgamma-activated macrophages (RAW 264) was studied in a co-culture system, using a non-adherent murine mastocytoma (P815) line as the 'target' cell. Addition of JM1226 or JM6245 (both 100 microM) to the culture medium afforded some protection from macrophage-mediated cell killing: target cell viability increased from 54.5+/-3.3% to 93.2+/-7.1% and 80.0+/-4.6%, respectively (n=6). 5. Vasodilator responses of isolated, perfused, pre-contracted rat tail arteries elicited by bolus injections (10 microl) of SNAP were attenuated by the addition of JM1226 or JM6245 (10(-4) M) to the perfusate: the ED50 increased from 6.0 microM (Krebs only) to 1.8 mM (Krebs + JM6245) and from 7 microM (Krebs only) to 132 microM (Krebs + JM1226). Oxyhaemoglobin (5 microM) increased the ED50 value for SNAP from 8 microM to 200 microM. 6. Male Wistar rats were injected with bacterial LPS (4 mg kg(-1); i.p.) to induce endotoxaemia. JM1226 and JM6245 (both 100 microM) fully reversed the hyporesponsiveness to phenylephrine of tail arteries isolated from animals previously (24 h earlier) injected with LPS. Blood pressure recordings were made in conscious LPS-treated rats using a tail cuff apparatus. A single injection of JM1226 (100 mg kg(-1), i.p.) administered 20 h after LPS (4 mg kg(-1), i.p.) reversed the hypotension associated with endotoxaemia. 7. The results show that JM1226 and JM6245 are able to scavenge NO in biological systems and suggest a role for these compounds in novel therapeutic strategies aimed at alleviating NO-mediated disease states.