Nephrotoxicity of cadmium & lead.

Cadmium and lead are divalent cations with a propensity to settle in the proximal tubule of the nephron, leading to nephrotoxicity. The pathophysiological results, however, tend to diverge. Cadmium in sufficient cumulative dosage leads to the production of the Fanconi syndrome, a generalized proximal tubular reabsorptive defect thought to be related to inhibition of both ATP production and Na-K-ATPase activity. On the other hand, lead accumulation in the proximal tubule leads to hyperuricaemia and gout, presumably by inhibiting uric acid secretion, and diminished glomerular filteration rate (GFR). Fanconi syndrome is seen unusually only in children and experimental animals. Cadmium nephrotoxicity is heralded by increased excretion of beta2-microglobulin, retinol binding protein and alpha1-microglobulin, indicative of decreased proximal tubule function. Beta2-microglobulinuria is not found in lead nephropathy. In lead nephropathy albuminuria is absent or minimal whereas in cadmium nephropathy albuminuria is variable. From the standpoint of pathology, both entities are characterized by tubulointerstitial disease and fibrosis, but only early lead nephropathy is characterized by the presence of proximal tubule nuclear inclusion bodies, due to the combination of lead with a lead binding-protein.

Cardiovascular effects of lead exposure.

Several epidemiological and clinical studies have found a link between chronic lead exposure and elevated blood pressure. In addition, a few population studies have shown possible connection between lead exposure and other cardiovascular disorders including ischaemic coronary heart disease, cerebrovascular accidents, and peripheral vascular disease. The causal link between chronic lead exposure and hypertension (HTN) has been confirmed by several studies in experimental animals. In addition, the effects of lead on the heart and vascular function have been explored in a limited number of in vivo and in vitro studies. The in vivo, ex vivo and in vitro studies conducted in laboratory animal, cultured cells and isolated tissues have helped to elucidate many of the mechanisms by which lead exposure can cause HTN and cardiovascular disease. This review is intended to provide an overview of the epidemiology and the underlying mechanisms of lead-associated HTN and cardiovascular disease.

Is lead exposure the principal cause of essential hypertension?

Lead is a ubiquitous toxin, known to have adverse effects on the body even at low levels of exposure. In this review we explore whether low lead may be the principal or a major contributory cause of essential hypertension, and whether removal of lead from the environment may eventually reduce both the overall incidence of hypertension and the increased incidence with aging.

Influence of lead on rat thoracic aorta contraction and relaxation.

Low levels of lead, but not high levels, produce hypertension. This mystery has not yet been resolved. In this study we compared the in vitro vasoresponsiveness in rat thoracic aorta to low dose (10(-8) mol/L) and high dose (10(-5) mol/L and 10(-4) mol/L) lead acetate. In addition to the direct response to lead, we examined reactivity to norepinephrine, acetylcholine, isoproterenol, phorbol ester, and calcium in the presence and absence of lead. Neither low-dose nor high-dose lead directly affected aortic contractile or relaxant responses. However, lead, only at the highest concentration (10(-4) mol/L), increased the contractions to calcium at all submaximal calcium concentrations. We conclude that low-dose lead must increase blood pressure indirectly through a humoral effect. The reasons for the failure of high-dose lead to influence blood pressure remain to be explored.

Lead-induced hypertension. III. Increased hydroxyl radical production.

Lead-induced hypertension has previously been shown to be closely associated with an increase in reactive oxygen species in low lead (100 ppm)-treated rats. The present study has attempted to define the specific moiety involved by noting the blood pressure (BP), reactive oxygen species (MDA-TBA), hydroxyl radical, and nitrotyrosine responses to infusion of the reactive oxygen species scavenger dimethylthiourea. Dimethylthiourea, a reputed scavenger of hydroxyl radical, normalized BP and MDA-TBA in the lead-treated rats but had no effect in normal control animals. MDA-TBA, hydroxyl radical, and nitrotyrosine, the tissue end product of peroxynitrite, were reduced to or toward normal by dimethylthiourea. The results, therefore, are consistent with the suggestion that either hydroxyl radical or peroxynitrite may be the reactive species affected by lead.

Severe lead-induced peripheral neuropathy in a dialysis patient.

Toxic neuropathy caused by lead (manifested as wrist drop) was a frequent phenomenon before 1925. In modern times, it is a distinct rarity. We report herein a Hispanic woman who developed end-stage renal failure, followed by wrist drop, in whom the maximal total blood lead was 69 microg/dL. Measurements of lead in her tibia and calcaneus by K-x-ray fluorescence, however, showed markedly elevated values. The wrist drop cleared after four treatments with intravenous calcium sodium edetate (Ca EDTA). In vitro studies of (210)Pb uptake by red blood cells (RBC) after incubation with normal or uremic plasma indicated that (210)Pb uptake was inhibited by uremic plasma. These studies suggest the presence of a transport inhibitor in uremia that modifies the distribution of lead between plasma and RBC, leading to lower overall blood values.

Lead promotes hydroxyl radical generation and lipid peroxidation in cultured aortic endothelial cells.

Early studies by our group have shown that lead-induced hypertension (HTN) is closely related to enhanced activity of reactive oxygen species (ROS). In addition, we have found indirect evidence that hydroxyl radical may be the most likely culprit in lead-exposed animals. In the present study, rat aortic endothelial cells were incubated in the presence of 0, 0.01, 0.1, 0.5, and 1.0 ppm lead acetate for 1, 24, and 48 h. At the conclusion of the incubation period cells were harvested and the media were collected. Lipid peroxidation products were measured as malondialdehyde-thiobarbituric acid (MDA-TBA) in the medium and hydroxyl radical was measured as 2,3-dihydroxybenzoic acid (2,3 DHBA) in the cells. After exposure to lead for 48 h, MDA-TBA generation and 2,3 DHBA formation were significantly increased. These data clearly demonstrate that lead exposure promotes hydroxyl radical generation and induces oxidative stress in isolated endothelial cells, mimicking the effects observed in lead-exposed animals. Enhanced inactivation of endothelium-derived nitric oxide by locally produced oxygen free radicals could contribute to endothelial dysfunction and HTN in lead-exposed animals.

Adequate dietary calcium mitigates osteopenia induced by chronic lead exposure in adult rats.

The purpose of the present study was to investigate bone changes in the adult rat exposed to low lead levels during intake of normal dietary calcium and to contrast these findings with data from our earlier studies performed with animals receiving low dietary calcium concurrent with lead exposure. The present study exposed adult rats to 100 ppm lead via drinking water for 12 weeks and assessed bone histology, 1,25-dihydroxyvitamin D, 25(OH)vitamin D and parathyroid hormone levels. No osteopenia was evident by quantitative bone histology, and circulating levels of 1,25-dihydroxyvitamin D, 25(OH) vitamin D and parathyroid hormone were normal. Bone ash findings documented incorporation of significant amounts of lead into bone mineral. These findings document absence of interference with vitamin D metabolism, absence of secondary hyperparathyroidism and absence of osteopenia following 12 weeks of low lead exposure in the adult rat maintained on normal calcium intake. Results stress the importance of adequate calcium intake in our elderly population who may be exposed to cumulative, low-level lead exposure.

Na-K-ATPase inhibitor dissociated from hypertension-associated plasma protein.

It has been demonstrated that human plasma contains a low molecular weight sodium-potassium-stimulated adenosine triphosphatase (Na-K-ATPase) inhibitor, which can be dissociated from a circulating protein with a molecular weight of approximately 12,000 daltons. The dissociated factor was found to have a molecular weight <500 daltons, and shared many characteristics with ouabain. Similar to ouabain, this factor was found to be a potent inhibitor of both the Na-K-ATPase and potassium-stimulated para-nitrophenyl phosphatase (K-pNPPase) enzyme systems, and to bind to both high- and low-affinity binding sites on Na-K-ATPase, but unlike ouabain did not cross-react with digoxin antibody. The factor was further separated by HPLC and electrochemical detection into two active compounds (p-NKAI-1 and p-NKAI-2). P-NKAI-1 was demonstrated on mass spectroscopy to have a molecular weight of 408 daltons. In a vasoconstrictor assay employing rabbit femoral artery segments, this compound was a direct vasoconstrictor and potentiated the vasoconstriction produced by norepinephrine. It behaved similarly to ouabain in counteracting the relaxing effect on rabbit femoral artery of increasing potassium concentrations in the tissue bath.

Summary of a symposium on natriuretic and digitalis-like factors.

An international symposium on natriuretic and digitalis-like factors was convened for the first time since 1992. Topics discussed included structures and biosynthesis of endogenous digitalis-like factors (EDLF), biologic activities, physiology function and role of EDLF in hypertension, and novel natriuretic factors. Progress was reported in determining the exact structure of an isomer of ouabain isolated from bovine hypothalamus. Evidence was presented supporting the existence of a second mammalian EDLF that resembles steroids found in toads (bufodienolides). Support for endogenous synthesis of mammalian EDLF was also presented. Mammalian EDLF were reported to have effects which are different from those possessed by digitalis like steroids derived from plants. New evidence was presented implicating EDLF in various forms of hypertension in humans and animal models. Finally, several unique natriuretic factors that do not inhibit Na, K ATPase and that appear to play a role in mammalian volume regulation were discussed.

Simultaneous measurement of marinobufagenin, ouabain, and hypertension-associated protein in various disease states.

We have previously demonstrated that a 12 kD hypertension-associated protein (HAP) is elevated in essential hypertension and that this protein has the characteristics of natriuresis, inhibition of Na-K-ATPase, displaces 3H-ouabain from binding sites, and is vasoconstrictive in vitro. In the present study, plasma from 101 patients were examined [25 normals (N)age 50, 7 with acute congestive heart failure (CHF), 24 with chronic renal failure (CRF), on dialysis, 5 with idiopathic hyperaldosteronism (PA) and 27 with essential hypertension, untreated (EHT)]. Plasma was extracted with 32% acetonitrile, then analyzed by DELFIA for marinobufagenin and ouabain. In addition, from 32 patients (6 N <50, 6 N >50, 5 CHF, 5 CRF, 6 EHT, and 4 PA) SDS gradient gels were obtained. The 12 kD bands were extracted, analyzed for Na-K-ATPase inhibition, marinobufagenin and ouabain, and compared to 14 kD and 21 kD bands. Marinobufagenin was found to be elevated in CRF, EHT, PA and CHF. Ouabain was increased only in PA. When the relative optical densities of the 12 kD and 21 kD bands were contrasted, CRF, PA, and EHT were found to be increased and CHF to be decreased in the 12 kD band, with no discernible changes in the 21 kD bands. Following extraction of the bands, Na-K-ATPase inhibitory activity measured 38% in 18 pooled 12 kD bands, with essentially no activity found in the 14 kD or 21 kD bands. Thus only the 12 kD HAP band possessed all of the attributes of natriuretic hormone.

Lead-induced hypertension. II. Response to sequential infusions of L-arginine, superoxide dismutase, and nitroprusside.

Administration of 100 ppm lead acetate daily for 3 months caused hypertension in Sprague-Dawley rats, with reversal by treatment with 2,3-dimercaptosuccinic acid (DMSA) (0.5% for 2 weeks). Animals from each group were infused sequentially in 30-min intervals with saline (S), L-arginine (Arg), Arg+ superoxide dismutase (SOD), S, and sodium nitroprusside (SNP). Baseline mean blood pressure (MBP) was elevated in lead-treated animals (Pb) compared to that in controls(C), returning toward normal after DMSA (105 +/- 2 mmHg, C, vs 149 +/- 2, Pb, and 124 +/- 1, DMSA, P < 0.001). Infusion of Arg caused a fall in MBP in all animals, normalizing the MBP in Pb-treated animals. SNP caused a greater fall in MBP in all groups of animals, normalizing the MBP in Pb. Measurement of urinary nitrite + nitrate (NOx) by chemiluminescence revealed at baseline a reduced level in Pb, restored to normal by DMSA (6.6 +/- 1.5 nmol/min/100 g BW, C, vs 3.3 +/- 1.7, Pb, P < 0.05, vs 5.8 +/- 2.6, DMSA, P = NS). Infusion of arginine increased urinary NOx in all groups, but to a lesser degree in Pb and DMSA. Assay of plasma malondialdehyde (MDA) by HPLC, as a measure of reactive oxygen species (ROS), was elevated at baseline in Pb, reduced by DMSA (3.6 +/- 0.4 mumol/L, Pb, vs 1.9 +/- 0.2, C, and 1.9 +/- 0.3, DMSA, P < 0.01). In the Pb group, SOD resulted in a significant fall in MDA (2.0 +/- 0.3 mumol/L, SOD, vs 3.1 +/- 0.1, Arg, P < 0.01), but no further fall in MBP or increase in urinary NOx. Thus, hypertension in lead-exposed animals is related to both diminished NO and increased ROS. The elevation in MBP can be ameliorated by additional NO through infusion of substrate arginine or by treatment with the ROS scavenger, DMSA. Lead-exposed animals show enhanced MBP sensitivity to the NO donors, Arg and SNP, but no further response to SOD, despite a reduction in MDA to normal. We speculate that lead-induced hypertension may be caused by one species of ROS which enhances vascular reactivity, and that provision of additional NO acts to scavenge the ROS and/or acts directly as a vasodilator.

Lead-induced hypertension: interplay of nitric oxide and reactive oxygen species.

An elevation of mean blood pressure was found in rats treated with low lead (0.01% lead acetate) for 3 months, as contrasted to paired Sprague-Dawley control rats. In these rats, measurement of plasma and urine endothelins-1 and -3 revealed that plasma concentration and urinary excretion of endothelin-3 increased significantly after 3 months (plasma: lead group, 31.8+/-2.2, versus controls, 23.0+1.7 pg/mL, P<.001; urinary excretion: lead group, 46.6+11.7, versus controls, 35.6+6.7 pg/24 h, P<.05), whereas endothelin-1 was unaffected. Plasma and urinary nitric oxide (NO) and cyclic GMP concentrations were not significantly changed. However, assay of plasma and kidney cortex malondialdehyde by high-pressure liquid chromatography, as a measure of reactive oxygen species, was elevated in lead-treated rats compared with that in control rats (plasma: lead group, 4.74+1.27, versus controls, 2.14+.49 micromol/L, P<.001; kidney cortex: lead group, 28.75+3.46, versus controls, 16.38+2.37 nmol/g wet weight, P<.001). There was increased NO synthase activity in lead-treated rat brain cortex and cerebellum. In lead-treated rat kidney cortex, the endothelial constitutive NO synthase protein mass was unaffected, whereas the inducible NO synthase protein mass was increased. These data suggest a balance between increased NO synthesis and degradation (by reactive oxygen species) in lead-treated rats, which results in normal levels of NO. Thus, the hypertension may be related to an increase in the pressure substances, endothelin-3 and reactive oxygen species, rather than to an absolute decrease in nitric NO.

Lead-induced hypertension is not associated with altered vascular reactivity in vitro.

In confirmation of a previous study (Am J Hypertens 1993;6:723), mean arterial blood pressure (MBP), as determined by tail cuff plethysmography, was found to be significantly elevated in Sprague-Dawley rats after 3 months of feeding 0.48 mmol/L (100 ppm) lead acetate/day (144 +/- 3.3 [SEM], in lead-treated [L] v 107 +/- 3.3 mm Hg in controls [C], P < .001). Thoracic aorta was excised from L and C animals (n = 6). Segments were suspended in tissue baths with Krebs' bicarbonate solution, then tested sequentially for vasoreactivity to 68 mmol/L K+, followed by graded concentrations of phenylephrine (PE), 0.01 to 0.3 micromol/L, acetylcholine (Ach), 0.001 to 3 micromol/L, nitroprusside (SNP), 0.0001 to 0.1 micromol/L, norepinephrine (NE), 0.001 to 300 micromol/L. There were no differences between L and C animals with respect to either vasoconstrictors (PE and NE) or vasodilators (Ach and SNP). The tissue levels of cGMP measured with and without phosphodiesterase inhibition, and in the absence and presence of either Ach or SNP, were comparable in the two groups. We conclude that the intrinsic vascular responsiveness is unchanged in lead-treated animals. The elevation of MBP is due to the presence of circulating factor(s) and hemodynamic changes.

Altered nitric oxide metabolism and increased oxygen free radical activity in lead-induced hypertension: effect of lazaroid therapy.

Chronic exposure to low levels of lead results in sustained hypertension (HTN) in humans and experimental animals. The mechanism of lead-induced HTN remains unclear. We investigated the possible role of reactive oxygen species (ROS) and their impact on nitric oxide (NO) metabolism in lead-induced HTN. Male Sprague-Dawley rats were treated with lead (100 ppm in drinking water) for twelve weeks. They were then treated with either the potent antioxidant, lazaroid (des-methyl-tirilazad, 5 mg/kg i.p., twice daily) (Pb-Lz group) or placebo (Pb group) for two weeks and monitored for an additional two weeks. A group of normal animals served as controls (N = 6 in each group). Lead administration resulted in marked HTN together with a significant rise in plasma concentration of lipid peroxidation product, malondialdehyde (MDA, reflecting increased ROS generation) and a twofold reduction in urinary excretion of NO metabolites, that is, total nitrates and nitrites (NOx). Lazaroid therapy led to prompt normalization of blood pressure, plasma MDA and urinary NOx. In contrast, blood pressure and plasma MDA remained elevated, and recovery of urinary NOx excretion was slow with placebo therapy. No significant difference was found in creatinine clearance between the study groups during the observation period. Thus, chronic lead exposure resulted in marked HTN coupled with increased ROS production and decreased urinary NOx excretion. Administration of the potent antioxidant, lazaroid, abrogated HTN and reversed the abnormalities of plasma MDA and urinary NOx excretion, thus supporting the role of ROS in lead-induced HTN in this model.

Osteopenia induced by long-term, low- and high-level exposure of the adult rat to lead.

The skeleton, the major site for Pb accumulation, is responsible for the largest fraction of the total body burden, but long-term effects of low-level exposure in adults remain unclear. In this study rats were exposed to low (0.01%; 100 ppm, LoPb) or high (0.5%, 5,000 ppm, HiPb) Pb, low calcium, feeding regimes for 1-12 months. Both LoPb and HiPb animals showed significant 12-month blood Pb levels [LoPb 21 +/- 3 micrograms/dl; HiPb 59 +/- 18; controls 3 +/- 1 (mean +/- SEM), p = 0.001]. Dual energy X-ray densitometry of the femur detected a significant decrease in bone density in HiPb animals by 3 months which remained significantly lowered through 12 months [3 months: HiPb: 0.498 +/- 0.011 (6) vs. control: 0.546 +/- 0.012 (6), p < 0.003]. By 12 months' density was also significantly lowered in LoPb animals (p = 0.001). Mineral analyses of ashed femurs showed a significant lead content after 1, 3, 9 and 12 months' exposure [1 month: LoPb, 0.020 +/- 0.002 (4) (% ash weight) vs control 0.008 +/- 0.0004 (4); HiPb 0.016 +/- 0.001 (8); control 0.007 +/- 0.0004 (6) (p < or = 0.002)]. Ca levels (% ash weight) were significantly lowered at 9 months in HiPb and 12 months in both groups (p < or = 0.04). Quantitative histomorphometry documented significantly elevated osteoid and resorptive trabecular surface features in both Pb groups. The LoPb design produced no overt renal functional abnormalities and resulted in blood Pb values comparable to those in man with modest environmental Pb exposure. The HiPb design resulted in development of lead nephropathy (more severe from months 6-12) and produced blood lead levels comparable to those seen in past industrial exposure. Findings show that Pb is incorporated into bone mineral after only 1 month's exposure to LoPb with significant osteopenia after 12 months' exposure; HiPb caused osteopenia by 3 months. No normal compensatory mechanism was elicited to maintain bone mass. Results stress renewed concern about the effects of cumulative, low-level lead exposure in our elderly population.

Silicon-related syndrome in dialysis patients.

Two dialysis patients with markedly elevated plasma silicon (Si) levels (3,849 and 2,350 micrograms/l, respectively) and a presumed Si-related syndrome are described in this report. One patient presented with transient hypercalcemia in the face of low PTH, vitamin D and plasma A1 levels. Both patients had painful, nodular skin eruptions and aberrant hair growth, characterized as perforating folliculitis on skin biopsy, compatible with known effects of organosilicon compounds in man and animals. Plasma Si was found to be moderately elevated in 30 dialysis patients studied at random (710 +/- 53 micrograms/l, dialysis, vs. 152 +/- 9 micrograms/l, normal control), but there was no significant difference between the arterial values before and after dialysis, implying that the source of Si was ingested foods and fluids rather than dialysate. In these patients, plasma Si was weakly correlated with serum calcium as well as with serum calcium corrected for serum albumen, indicating that Si, like aluminum, may affect calcium metabolism.

Effect of 2,3-dimercaptosuccinic acid on nephrosclerosis in the Dahl rat. I. Role of reactive oxygen species.

2,3-Dimercaptosuccinic acid (DMSA), a sulfhydryl-containing chelator, has previously been shown to reduce mean blood pressure in lead-treated rats. In the present study we have demonstrated that DMSA (0.5% for 5 days every 2 weeks) also reduces mean blood pressure in the Dahl salt-sensitive (SS) rat. Six-week-old Dahl SS and salt resistant (SR) rats were placed on a 0.3% NaCl diet for two weeks, followed by an 8% NaCl diet for four weeks. Eight SS and 8 SR rats remained untreated while 8 SS and 8 SR rats were treated with DMSA. DMSA treatment ameliorated the mean blood pressure rise in the Dahl SS rats (141 +/- 5 vs. 120 +/- 4 mm Hg at 6 weeks, P < 0.001). Nephrosclerosis was severe in untreated SS rats but absent in treated SS rats as well as in both treated and untreated SR rats. Reactive oxygen species formation, as assessed by kidney cortex content of malondialdehyde (MDA) and immunohistochemical demonstration of nitrotyrosine (a byproduct of peroxynitrite) in interlobular arteries, was increased in Dahl SS rats, but abolished by DMSA (MDA 9.65 +/- 0.33 nmol/g wet wt, untreated SS, vs. 6.46 +/- 0.51, treated SS, P < 0.001). The anti-nephrosclerotic action of DMSA was clearly disproportionate to the reduction in blood pressure. We conclude that the effect of DMSA was related instead to the reactive oxygen species scavenging properties of the thiol groups.

Relationship of Na-K-ATPase inhibitors to blood-pressure regulation in continuous ambulatory peritoneal dialysis and hemodialysis.

Inhibitors of sodium-potassium-activated adenosine triphosphatase (Na-K-ATPase) have been implicated in the pathogenesis of hypertension. In the study presented here, an attempt was made to determine whether differences in the plasma levels and the removal rates of high-molecular weight (HMW) and low-molecular weight (LMW) forms of Na-K-ATPase inhibitors might relate to blood-pressure control in hemodialysis (N = six ultrafiltered and N = six non-ultrafiltered) and CAPD (N = six long-term and N = five short-term) patients. The latter group was studied before the initiation of continuous ambulatory peritoneal dialysis (CAPD) and 2 wk after starting the treatment. The mean blood pressure was significantly reduced after dialysis in the nonultrafiltered hemodialysis group and in both CAPD groups. Plasma levels of both HMW and LMW inhibitors were found to be elevated before dialysis in all patients and were modified only slightly after dialysis. Irrespective of whether ultrafiltration was utilized in hemodialysis patients and despite significant losses of both HMW and LMW inhibitors into CAPD effluent. Because CAPD effluent was found to contain vasopressors that were not exclusively Na-K-ATPase inhibitors, losses of these other vasopressors may contribute to improved blood-pressure control in CAPD in contrast to hemodialysis.

Effect of chelation treatment with dimercaptosuccinic acid (DMSA) on lead-related blood pressure changes.

An elevation in mean blood pressure was found in rats treated with low lead (0.01%) for 6 months and then only water for an additional 6 months (discontinuous low lead). No change in blood pressure was found in rats similarly treated with high lead (0.5%) (discontinuous high lead). Administration of DMSA (0.5% in drinking water), for 5 days every 2 months following cessation of lead administration, resulted in a significant lowering of blood pressure in both groups of animals. In the low-lead but not the high-lead group, this was associated with an increase in plasma cyclic GMP (acting as a second messenger for endothelium-derived relaxing factor, EDRF) and a decrease in the plasma concentration of a 12-kDa hypertension-associated protein. Plasma endothelin-3 (ET-3) levels were decreased in discontinuous high-lead rats, increased in discontinuous low-lead rats, but were unaltered by DMSA treatment. We infer that the elevated blood pressure in the discontinuous low-lead rats is related to an increase in the putative vasoconstrictors, ET-3 and the hypertension-associated protein, without a change in the vasodilator, EDRF. With DMSA treatment, plasma cyclic GMP in low-lead rats increased above normal, and the hypertension-associated protein decreased, resulting in lowered blood pressure. DMSA was shown to act as an antioxidant in vitro. Thus the DMSA effect on plasma cGMP (EDRF) may occur via a scavenging effect on EDRF-inactivating reactive oxygen species.