Quinapril, an ACE Inhibitor, Reduces Markers of Oxidative Stress in the Metabolic Syndrome

Metabolic Syndrome/Insulin Resistance Syndrome/Pre-Diabetes ORIGINAL ARTICLE

1 3 BOBBY V. KHAN, MD, PHD W. CRAIG HOOPER, PHD ing the link between inflammation, met- 1 1 SRIKANTH SOLA, MD REKHA G. MENON, MD abolic disorders, and cardiovascular 1 1 WRIGHT B. LAUTEN, BS STAMATIOS LERAKIS, MD 2 1 disease (5,6). Chronic inflammation and RAMA NATARAJAN, PHD TAREK HELMY, MD an abnormal pro-oxidant state are both found in the metabolic syndrome and may play a role in its pathogenesis (7,8). The renin-angiotensin system (RAS) plays a central role in the pathogenesis of OBJECTIVE — Patients with the metabolic syndrome often have abnormal levels of proin- atherosclerosis-related diseases. Angio- flammatory and pro-oxidative mechanisms within their vasculature. We sought to determine tensin II, the central molecule in the RAS, whether the ACE inhibitor quinapril regulates markers of oxidative stress in the metabolic syndrome. has multiple effects on inflammation, oxidation, atherosclerotic plaque initiation, RESEARCH DESIGN AND METHODS — Forty patients with the metabolic syndrome and progression (9). In the present study, were randomized in a double-blind manner to either the ACE inhibitor quinapril (20 mg/day) or we determine potential mechanisms by matching placebo for 4 weeks. Serum markers of vascular oxidative stress were measured. which the administration of the ACE inhibitor quinapril regulates mechanisms of RESULTS — After 4 weeks of therapy, serum 8-isoprostane was reduced by 12% in the oxidative stress in subjects with the met- Ϯ Ϯ quinapril group when compared with placebo (quinapril, 46.7 1.0; placebo, 52.7 0.9 abolic syndrome. pg/ml; P ϭ 0.001). Erythrocyte superoxide dismutase activity increased 35% in the quinapril Ϯ Ϯ group when compared with placebo (quinapril, 826.3 17.1; placebo, 612.3 6.9 units/g Hb; RESEARCH DESIGN AND P Ͻ 0.001). In addition, lag time to oxidation of LDL, a marker of oxidative stress, was increased Ϯ METHODS — Men and women aged by 48% in the quinapril group when compared with placebo (quinapril 89.2 9.2 vs. placebo Ն 60.1 Ϯ 12.3 min; P Ͻ 0.001). Therapy with quinapril was well tolerated. 18 years and with the metabolic syndrome were enrolled in the study. The CONCLUSIONS — The addition of the ACE inhibitor quinapril reduces markers of vascular metabolic syndrome was defined using oxidative stress and may attenuate the progression of the pathophysiology seen in the metabolic the National Cholesterol Education Pro- syndrome. gram Adult Treatment Panel III criteria (Table 1), and eligible subjects were re- Diabetes Care 27:1712–1715, 2004 quired to meet at least three of the five criteria (10). Subjects were excluded if they had any of the following: tobacco use he metabolic syndrome is a constel- pared with those without metabolic syn- Ͻ6 months before enrollment, a clinical lation of abnormal glucose and lipid drome (2). history of coronary artery disease or con- T metabolism that has reached epi- Recent studies (3,4) suggest that pro- gestive heart failure, use of an ACE inhib- demic proportions over the last decade oxidative and proinflammatory processes itor or angiotensin receptor blocker Ͻ12 (1). Patients with the metabolic syndrome play a significant role in the progression of months before enrollment, ejection frac- are at considerable risk for developing atherosclerosis. In fact, inflammatory tion Ͻ50% by echocardiography or con- atherosclerosis-related diseases, includ- markers are predictors of cardiovascular trast ventriculogram, systolic blood Ͼ Ͻ ing a two- to fourfold increased risk of events and progression to type 2 diabetes pressure 140 or 100 mmHg, diastolic blood pressure Ͼ90 or Ͻ60 mmHg, stroke and a three- to fourfold increased in healthy individuals as well as those Ͼ Ͼ risk of myocardial infarction when com- with the metabolic syndrome, underscor- HbA1c 7.0%, serum creatinine 2.0 mg/dl, hepatic impairment, or malig- nancy. Written informed consent was ob- ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● tained from all subjects. From the 1Division of Cardiology, Emory University School of Medicine, Atlanta, Georgia; the 2Division of 3 Diabetes and Endocrinology, City of Hope Medical Center, Duarte, California; and the Centers for Disease Study design Control and Prevention, Atlanta, Georgia. Address correspondence and reprint requests to Bobby V. Khan, MD, PhD, Division of Cardiology, Subjects were randomized in a double- Department of Medicine, Emory University School of Medicine, 69 Jesse Hill Dr. SE, #C247, Atlanta, GA blinded fashion to either quinapril 20 mg/ 30303. E-mail: [email protected]. day or matching placebo for 4 weeks. Received for publication 10 February 2004 and accepted in revised form 1 April 2004. Allocation concealment was maintained Abbreviations: E-SOD, erythrocyte superoxide dismutase; RAS, renin-angiotensin system; A table elsewhere in this issue shows conventional and Syste`me International (SI) units and conversion until the end of the study. The dose of the factors for many substances. study drug was chosen based on the re- © 2004 by the American Diabetes Association. sults of prior studies in which we found

1712 DIABETES CARE, VOLUME 27, NUMBER 7, JULY 2004 Khan and Associates

Table 1—Diagnostic criteria for the metabolic syndrome Statistical analysis All data are presented as the mean Ϯ SE. Factor Criteria Comparisons were determined within the ACE inhibitor and placebo groups using 1 Abdominal girth Ͼ40 in (102 cm) in men or Ͼ35 in (88 cm) in women paired-sample t tests. A P value of Ͻ0.05 2 HDL cholesterol Ͻ40 mg/dl in men or Ͻ50 mg/dl in women was considered statistically signiﬁcant, 3 Fasting triglycerides Ͼ150 mg/dl (1.69 mmol/l) and all P values were two sided. Calcula- 4 Blood pressure Ͼ130/85 mmHg tions were performed with SPSS software 5 Fasting glucose Ն110 mg/dl (Ն 6.1 mmol/l) (version 10.0; Statistical Package for the Three out of ﬁve criteria are required for diagnosis of the metabolic syndrome. Social Sciences, Chicago, IL).

that the addition of quinapril 20 mg daily LDL oxidation and lag time to standard therapy in subjects with cor- We have previously observed that the ad- RESULTS — Forty subjects (23 men onary artery disease reduced markers of ministration of an RAS inhibitor to sub- and 17 women) were enrolled in the inflammation and oxidative stress; these jects with coronary artery disease study and followed for 4 weeks. Fol- changes were independent of blood pres- increases the lag time to oxidation of LDL low-up was 100% complete. The two ex- sure reduction (11). Subjects were ad- cholesterol, suggesting increased resis- perimental groups (quinapril and placebo) had similar demographic char- vised to self-administer one-half of the tance to the modification of LDL choles- acteristics (Table 2). The mean age was full dose during the initial 2 days of ther- terol (15). LDL cholesterol was isolated 2 apy, after which they were to take the full 36.2 years; mean BMI 29.3 kg/m ; mean from blood samples. Using CuSO4, oxi- study dose. After 2 weeks, blood pressure LDL cholesterol, HDL cholesterol, and dation of LDL was performed. Spectro- Ϯ was rechecked and blood was drawn to triglyceride levels were 125.9 20.3 photometric analysis was used to Ϯ Ϯ measure serum creatinine and potassium. mg/dl (3.3 0.5 mmol/l), 39.7 5.9 determine the lag time to oxidation of (1.04 Ϯ 0.13), and 229 Ϯ 33 (6 Ϯ 0.8), Fasting blood samples were drawn before LDL (16). We found no evidence of inter- and at the end of therapy at a similar time respectively; mean systolic and diastolic ference of quinapril or its metabolite in Ϯ of day. The study protocol complies with blood pressures were 125.9 8.2 and the LDL oxidation assays. Ϯ the Declaration of Helsinki and was ap- 79.5 10.6 mmHg, respectively; and the Blood glucose was measured using a proved by the institutional review board mean fasting glucose was 112 mg/dl. Nine glucose dehydrogenase method after pre- at the participating institution before its subjects (23%) were on lipid-lowering implementation. cipitation of proteins by trichloroacetic therapy. No statistically significant differ- acid. LDL and HDL fractions were sepa- ences were noted in baseline characteris- Measurement of total serum rated from fresh serum by combined ul- tics between the two treatment groups. 8-isoprostane and erythrocyte tracentrifugation and precipitation. There was a higher incidence of superoxide dismutase Lipoprotein fraction cholesterol and trig- cough in the quinapril group when com- ϭ Plasma samples were centrifuged and lycerides were measured enzymatically. pared with the placebo group (4 vs. 1; P stored at Ϫ80°C. An aliquot was drawn, and enzyme immunoassay (EIA; Cayman Chemical, Ann Arbor, MI) for serum 8-iso- Table 2—Patient demographics and baseline characteristics prostane was performed on each sample in ␮ triplicate. Serum (60 l) was used for anal- Quinapril Placebo ysis, and enzyme immunoassay was performed as previously described (12). The n 20 20 levels of total serum 8-isoprostane were de- Age (years) 35.2 39.6 termined on a plate reader at an optical den- Sex (M/F) 11/9 12/8 sity of 420 nm. Erythrocyte superoxide BMI (kg/m2) 29.8 29.1 dismutase (E-SOD) activity was determined Systolic blood pressure (mmHg) 126 Ϯ 8 130 Ϯ 9 using hemolysates and commercially avail- Diastolic blood pressure (mmHg) 80 Ϯ 10.4 77.8 Ϯ 11.1 able kits (cat. no. SDI 25; Randox Labora- LDL cholesterol (mg/dl) 124.7 Ϯ 21.4 130 Ϯ 16.8 tories, Crumlin, Ireland). Briefly, HDL cholesterol (mg/dl) 40.3 Ϯ 7.0 38.9 Ϯ 6.1 superoxide radicals produced by xanthine Triglycerides (mg/dl) 238 Ϯ 36 219 Ϯ 30 and xanthine oxidase reacts with 2-(4- Fasting glucose (mg/dl) 113 Ϯ 15 111 Ϯ 19 Ϯ Ϯ iodophenyl)-3-(4-nitrophenol)-5- HbA1c (%) 5.8 0.7 5.6 0.8 phenyltetrazolium chloride to form a red Hypertension 5 (25) 6 (30) formazan dye (13). The E-SOD activity is Smoking history 7 (35) 6 (30) then measured by the degree of inhibition of Concomitant medications this reaction. E-SOD activity was expressed Lipid lowering therapy* 5 (25) 6 (30) as units per gram of Hb (14). We found no ␤-Blocker 2 (10) 4 (20) evidence of interference of quinapril or its Diuretic 3 (15) 4 (20) metabolite in the isoprostane or E-SOD Calcium channel blocker 3 (15) 4 (20) assays. Data are means Ϯ SE or n (%). *Includes statin, niacin, resin, or fibrates.

DIABETES CARE, VOLUME 27, NUMBER 7, JULY 2004 1713 Quinapril and oxidative stress

Table 3—Effects of therapy on markers of oxidative stress and metabolic indexes

Baseline After 4 weeks of therapy Quinapril Placebo P Quinapril Placebo P n 20 20 — 20 20 — 8-isoprostane (pg/ml) 54.4 Ϯ 1.4 53.2 Ϯ 1.2 NS 46.7 Ϯ 2 52.7 Ϯ 0.9 0.001 E-SOD activity (units/g Hb) 609.3 Ϯ 9.8 612.7 Ϯ 9 NS 826.3 Ϯ 17.1 612.3 Ϯ 6.9 0.001 Lag time to LDL oxidation (min) 60.5 Ϯ 10.1 61.0 Ϯ 8.7 NS 89.2 Ϯ 9.2 60.1 Ϯ 12.3 0.001 LDL cholesterol (mg/dl) 124.7 Ϯ 21.4 130 Ϯ 16.8 NS 122.9 Ϯ 19.2 133.1 Ϯ 17.8 NS Triglycerides (mg/dl) 238 Ϯ 36 219 Ϯ 30 NS 234 Ϯ 31 227 Ϯ 24 NS Fasting glucose (mg/dl) 113 Ϯ 15 111 Ϯ 19 NS 116 Ϯ 17 115 Ϯ 13 NS Ϯ Ϯ Ϯ Ϯ HbA1c (%) 5.8 0.7 5.6 0.8 NS 5.6 0.6 5.7 0.7 NS Data are means Ϯ SE.

0.05). Furthermore, there was a rise in with quinapril compared with placebo clinical biomarkers of lipid peroxidation serum potassium or creatinine of Ͼ20% (quinapril 89.2 Ϯ 9.2 vs. placebo 60.1 Ϯ (i.e., oxidative stress) (17). Oxidized LDL in 2 of 20 subjects in the quinapril group 12.3 min; P Ͻ 0.001). Furthermore, the is believed to be the most atherogenic (P ϭ NS). There was an average reduction increase in lag time to LDL oxidation was form of LDL; the time required for LDL to of systolic blood pressure by 4 Ϯ 2 mmHg observed in every subject treated with undergo oxidation, or lag time, is another and diastolic blood pressure by 3 Ϯ 2 quinapril (data not shown). We deter- indirect measure of oxidative stress mmHg in the group of subjects treated mined mean values for 8-isoprostane, E- (18,19). Finally, superoxide dismutase with quinapril during the study period SOD, and lag time to LDL oxidation in a catalyzes the reaction of superoxide an- ϭ ϭ Ϫ (P NS). No subject in either group ex- healthy control population (n 19) to be ions (O2 ) to hydrogen peroxide (H2O2), perienced hypotension (i.e., systolic 44.4 Ϯ 5.3 pg/ml, 773 Ϯ 86 units/g Hb, making it a central element in the main- blood pressure Ͻ100 mmHg) during the and 86.0 Ϯ 10.0 min, respectively. No tenance of the vascular redox balance. As study. No changes in glycemic control changes in levels of LDL cholesterol, trig- such, superoxide dismutase is indirectly were observed in either of the two study lycerides, serum glucose, or HbA1c were involved in regulating levels of nitric ox- groups. observed in either the placebo or ide (NO) bioavailability. Inhibition of quinapril study groups (Table 3). bradykinin degradation by ACE inhibi- Treatment with quinapril reduces tors may increase the activity of superox- serum 8-isoprostane in the CONCLUSIONS — This study dem- ide dismutase and modulate the metabolic syndrome onstrates possible mechanisms by which production of NO, leading to the inacti- Activity of 8-isoprostane was significantly the ACE inhibitor quinapril may affect vation of reactive oxygen species, while reduced after 4 weeks of therapy with vascular oxidative processes in subjects also inhibiting various pro-oxidative quinapril compared with the placebo with the metabolic syndrome. Therapy mechanisms within the vasculature (20). (46.7 Ϯ 2 vs. 52.7 Ϯ 0.9 pg/ml; P ϭ with quinapril resulted in increased E- The RAS plays a central role in the 0.001). Furthermore, the decrease in the SOD activity, suggesting that quinapril pathogenesis of atherosclerosis-related expression of 8-isoprostane was observed has antioxidative effects in subjects with diseases. Angiotensin II, the central mol- in every study subject treated with the metabolic syndrome. In addition, lev- ecule in the RAS, has multiple effects on quinapril (data not shown). els of serum 8-isoprostane were de- atherosclerotic plaque initiation and pro- creased, whereas lag time to LDL gression. On a molecular and cellular Treatment with quinapril increases oxidation was increased, findings that are level, blockade of the RAS reduces the ex- E-SOD activity in the metabolic consistent with decreased oxidative stress tent of vascular lesions in atherosclerosis syndrome within the vasculature. Furthermore, our (21), and it appears that these effects may Activity of E-SOD was significantly in- findings suggest that these effects may be be independent of blood pressure reduc- creased after 4 weeks of therapy with at least partly independent of blood pres- tion (22). Although not fully established, quinapril compared with placebo sure reduction. In comparison with pla- these mechanisms include improved en- (826.3 Ϯ 17.1 vs. 612.3 Ϯ 6.9 units/g cebo, no significant changes in systolic dothelial function, plaque stabilization, Hb; P Ͻ 0.001). Furthermore, the in- blood pressure, LDL cholesterol, or met- and regulation of hemodynamic stress; crease in E-SOD activity was observed in abolic control were noted with quinapril NO bioactivity; pulse pressure; and the every study subject treated with quinapril therapy in these patients. ability of macrophages to oxidize LDL (data not shown). Pro-oxidative mechanisms are (23). Our study highlights mechanisms thought to be a hallmark of the athero- by which these agents may be of benefitin Treatment with quinapril increases genic process. We examined several subjects with the metabolic syndrome. lag time to LDL oxidation in markers of oxidation within the vascula- metabolic syndrome ture in this study. The isoprostanes are a Limitations of the study Lag time to oxidation of LDL was signifi- family of free-radical–dependent metab- Our investigation is a short-term study (4 cantly increased after 4 weeks of therapy olites of arachidonic acid that are used as weeks) to determine potential mecha-

1714 DIABETES CARE, VOLUME 27, NUMBER 7, JULY 2004 Khan and Associates nisms by which ACE inhibitors may be 12. Hoffman SW, Roof RL, Stein DG: A reli- References effective in the metabolic syndrome. Due able and sensitive enzyme immunoassay 1. Haffner S, Taegtmeyer H: Epidemic obe- method for measuring 8-isoprostaglandin to its small size (n ϭ 40), we were not able sity and the metabolic syndrome. Circula- to evaluate differences among various F2 alpha: a marker for lipid peroxidation tion 108:1541–1545, 2003 after experimental brain injury. J Neurosci 2. Lakka HM, Laaksonen DE, Lakka TA, Nis- subsets within our study population. In Methods 68:133–136, 1996 kanen LK, Kumpusalo E, Tuomilehto J, addition, it is possible that the response to 13. Erel O, Kocyigit A, Bulut V, Gurel MS: Salonen JT: The metabolic syndrome and therapy with quinapril may have been Reactive nitrogen and oxygen intermedi- total and cardiovascular disease mortality ates in patients with cutaneous leishman- more pronounced if a higher dose of in middle-aged men. JAMA 288:2709– iasis. Mem Inst Oswaldo Cruz 94:179– quinapril had been used (e.g., 40 mg/ 2716, 2002 183, 1999 day). The subjects in our study are at sig- 3. Ross R: Atherosclerosis is an inflamma- 14. McCord JM, Fridowich I: Superoxide dis- nificant risk for the development of tory disease. Am Heart J 138:S419–S420, mutase: an enzymatic function for eryth- 1999 atherosclerosis and its related complica- rocuprein. J Biol Chem 249:6049–6055, 4. Libby P, Ridker PM, Maseri A: Inflamma- tions; therefore, these results may not be 1969 applicable to lower-risk populations. tion and atherosclerosis. Circulation 105: 1135–1143, 2002 15. Khan BV, Navalkar S, Khan QA, Rahman Nevertheless, in comparison with pla- 5. Schmidt MI, Duncan BB, Sharrett AR, ST, Parthasarathy S: Irbesartan, an angio- cebo, therapy with quinapril for 4 weeks Lindberg G, Savage PJ, Offenbacher S, tensin type 1 receptor inhibitor, regulates reduced serum levels of 8-isoprostane, in- Azambuja MI, Tracy RP, Heiss G: Markers the vascular oxidative state in patients creased E-SOD activity, and prolonged of inflammation and prediction of diabe- with coronary artery disease. J Am Coll Cardiol 38:1662–1667, 2001 lag time to LDL oxidation in all subjects. tes mellitus in adults (Atherosclerosis Risk in Communities study): a cohort study. 16. Strawn WB, Chappell MC, Dean RH, Kiv- Lancet 353:1649–1652, 1999 lighn S, Ferrario CM: Inhibition of early atherogenesis by losartan in monkeys Summary 6. Rader DJ: Inflammatory markers of coronary risk. N Engl J Med 343:1179–1182, with diet-induced hypercholesterolemia. The incidence of the metabolic syndrome 2000 Circulation 101:1586–1593, 2000 is increasing throughout the world, with a 7. Ridker PM, Buring JE, Cook NR, Rifai N: 17. Roberts LJ, Morrow JD: Measurement of concomitant increase in the risk of athero- C-reactive protein, the metabolic syn- F2-isoprostanes as an index of oxidative sclerosis-related diseases. Pro-oxidative drome, and risk of incident cardiovascu- stress in vivo. Free Radic Biol Med 28:505– lar events: an 8-year follow-up of 14,719 513, 2000 and proinflammatory mechanisms are 18. Jialal I, Devaraj S: Low-density lipoprotein important in the pathogenesis of athero- initially healthy American women. Circu- lation 107:391–397, 2003 oxidation, antioxidants, and atherosclero- sclerosis, and the present study suggests 8. Festa A, D’Agostino R Jr, Howard G, Myk- sis: a clinical biochemistry perspective. Clin in part the mechanisms by which the ACE kanen L, Tracy RP, Haffner SM: Chronic Chem 42:498–506, 1996 inhibitor quinapril may be beneficial in subclinical inflammation as part of the in- 19. Gaut JP, Heinecke JW: Mechanisms for the prevention of atherosclerosis-related sulin resistance syndrome: the Insulin Re- oxidizing low-density lipoprotein: in- diseases. These findings also reinforce the sistance Atherosclerosis Study (IRAS). sights from patterns of oxidation products growing evidence of oxidative mecha- Circulation 102:42–47, 2000 in the artery wall and from mouse models of atherosclerosis. Trends Cardiovasc Med nisms in the pathogenesis of atherosclero- 9. Rajagopalan S, Harrison DG: Reversing endothelial dysfunction with ACE inhibi- 11:103–112, 2001 sis. Inhibitors of the RAS, such as ACE tors: a new trend. Circulation 94:240– 20. Ehring T, Baumgart D, Krajcar M: Atten- inhibitors, have powerful anti-inflamma- 243, 1996 uation of myocardial stunning by the ACE tory and antioxidant effects within the 10. National Cholesterol Education Program inhibitor ramiprilat through a signal cas- vasculature, and clinical outcome studies Expert Panel on Detection, Evaluation, cade of bradykinin and prostaglandins should be considered to determine the and Treatment of High Blood Cholesterol but not nitric oxide. Circulation 90:1368– utility of these agents in the primary pre- in Adults: Executive Summary of The 1385, 1994 21. Candido R, Jandeleit-Dahm KA, Cao Z: vention of atherosclerosis-related diseases Third Report of The National Cholesterol Education Program (NCEP) Expert Panel Prevention of accelerated atherosclerosis in subjects with the metabolic syndrome. on Detection, Evaluation, and Treatment by angiotensin converting enzyme inhibi- of High Blood Cholesterol in Adults tion I diabetic apolipoprotein E-deficient (Adult Treatment Panel III). JAMA 285: mice. Circulation 106:246–253, 2002 2486–2497, 2001 22. Adams KF: Angiotensin-converting en- Acknowledgments— This study was sup- 11. Lauten WB, Khan QA, Rajagopalan S, Le- zyme inhibitor and vascular remodeling ported by a grant from the Emory Medical rakis S, Rahman ST, Parthasarathy S, in coronary artery disease. Coron Arter Dis Care Foundation and an unrestricted educa- Khan BV: Usefulness of quinapril and 9:675–684, 1998 tional grant from Pfizer Pharmaceuticals. irbesartan to improve the anti-inflamma- 23. Lonn EM, Yosuf S, Jha P: Emerging role of We also thank Reuben Press of Pfizer Phar- tory response of atorvastatin and aspirin angiotensin-converting enzyme inhibi- maceuticals for his assistance in providing the in patients with coronary heart disease. tors in cardiac and vascular protection. quinapril for this study. Am J Cardiol 91:1116–1119, 2003 Circulation 90:2056–2069, 1994

DIABETES CARE, VOLUME 27, NUMBER 7, JULY 2004 1715