DIABETES, VOL. 49, MAY 2000 857 Nitric Oxide Synthesis and Isoprostane Production in Subjects With Type 1 Diabetes and Normal Urinary Albumin Excretion Sharon O’Byrne, Pablo Forte, L. Jackson Roberts II, Jason D. Morrow, Atholl Johnston, Eric Änggård, R.D.G. Leslie, and Nigel Benjamin The role of nitric oxide (NO) and free radicals in the development of microvascular disease in type 1 dia- betes remains unclear. We have measured NO and iso- prostane (a stable marker of in vivo lipid peroxidation) production in 13 type 1 diabetic subjects with normal urinary albumin excretion and 13 healthy volunteers. Whole-body NO synthesis was quantified by measuring the urinary excretion of 15 N-nitrate after the intra- venous administration of L-[ 15 N] 2 -arginine. The urinary excretion of the major urinary metabolite of 15-F 2t -iso- prostane (8-iso-prostaglandin-F 2 ), 2,3-dinor-5,6-dihydro- F 2t -IsoP, was quantified as a marker of in vivo lipid per- oxidation. Whole-body NO synthesis was significantly higher in diabetic subjects compared with control sub- jects (342 vs. 216 nmol 15 N-nitrate/mmol creatinine [95% CI of the difference 45–207], P = 0.005). This increase was not explained by a difference in renal function between the 2 groups. There was no difference in 2,3-dinor-5,6-dihydro-F 2t -IsoP excretion between dia- betic subjects and control subjects (44.8 ± 7.8 vs. 41.4 ± 10.0 ng/mmol creatinine, mean ± 95% CI). However, there was an inverse correlation between NO synthesis and free radical activity in subjects with diabetes (r = –0.62, P = 0.012) that was not observed in control sub- jects (r = 0.37, P = 0.107). We conclude that whole-body NO synthesis is higher in type 1 diabetic subjects with normal urinary albumin excretion than in control sub- jects. The inverse correlation between isoprostane pro- duction and NO synthesis in diabetic subjects is consis- tent with the hypothesis that NO is being inactivated by reactive oxygen species. Diabetes 49:857–862, 2000 T he Diabetes Control and Complications Trial has confirmed that tight blood glucose control signifi- cantly delays the onset and slows the progression of microvascular complications in type 1 diabetes (1). However, despite optimum blood glucose control, a sub- stantial proportion of diabetic subjects in this study developed serious microvascular complications. In contrast, in the Wis- consin Epidemiological Study, a significant proportion (30%) of subjects with type 1 diabetes did not manifest severe microvascular complications after 10 years of follow-up, although they had chronic poorly controlled diabetes (2). It would be useful if we could understand this discrepancy underlying the development of microvascular disease so that those at risk could be identified. The benefit/risk ratio of maintaining tight blood glucose control could be increased if targeted toward individuals at greatest risk of complications due to hyperglycemia. The complex issues involved in both the pathogenesis of and individual susceptibility to the angiopathy associated with type 1 diabetes have been well documented (3–5). The endothelium and, in particular, the production of nitric oxide (NO) have been the subject of considerable research in recent years. NO maintains basal vasodilator tone, is a potent platelet anti-aggregant, reduces the ability of monocytes to adhere to endothelial cells and to oxidize LDL cholesterol, and inhibits the proliferation of vascular smooth muscle cells (6,7), thus serving an important physiological role in the nor- mal microvasculature. Functional studies of the action of NO in type 1 diabetes have produced conflicting results, especially when the effects of muscarinic agonist–mediated release of NO (e.g., acetylcholine, carbachol, and methacholine) in the vascular bed in the forearm are measured (8–14). N G -monomethyl-L- arginine (L-NMMA), a competitive inhibitor of endothelial nitric oxide synthase, provides a more specific measurement of basal vasodilatation due to NO. Two studies have shown impaired forearm vasoconstriction to L-NMMA, implying a reduced contribution of NO to vascular tone (9,10). How- ever, this defect may only be present in type 1 diabetic sub- jects with microalbuminuria (urinary albumin excretion rate [UAER] 20–200 μg/min) (10). Until recently, the methodology to accurately quantify free radical production in vivo has suffered from limitations that have made interpretation difficult. The discovery of F 2 -iso- prostanes, a group of biologically active compounds pro- From the Department of Clinical Pharmacology (S.O., P.F., A.J., N.B.), St Bartholomew’s and the Royal London School of Medicine and Dentistry, Charterhouse Square; the Department of Diabetes and Metabolism (R.D.G.L.), St Bartholomew’s and the Royal London School of Medicine and Dentistry, West Smithfield; the William Harvey Research Institute (E.Ä.), London, U.K.; and the Department of Pharmacology (L.J.R., J.D.M.), Van- derbilt University, Nashville, Tennessee. Address correspondence and reprint requests to S. O’Byrne, MD, Department of Clinical Pharmacology, Charterhouse Square, London EC1M 6BQ, U.K. E-mail: s.r.o’byrne@mds.qmw.ac.uk. Received for publication 27 August 1999 and accepted in revised form 19 January 2000. L-NMMA, N G -monomethyl-L-arginine; MANOVA, multivariate analysis of variance; NO, nitric oxide; O 2 – , superoxide; ONOO – , peroxynitrite; UAER, urinary albumin excretion rate.