overweight and obese humans demonstrate increased vascular

Overweight and Obese Humans Demonstrate IncreasedVascular Endothelial NAD(P)H Oxidase-p47phox Expression

and Evidence of Endothelial Oxidative Stress

Annemarie E. Silver, PhD; Stacy D. Beske, PhD; Demetra D. Christou, PhD; Anthony J. Donato, PhD;Kerrie L. Moreau, PhD; Iratxe Eskurza, MD, PhD; Phillip E. Gates, PhD; Douglas R. Seals, PhD

Background—Obesity may alter vascular endothelial cell protein expression (VECPE) of molecules that influencesusceptibility to atherosclerosis.

Methods and Results—Quantitative immunofluorescence was performed on vascular endothelial cells collected from 108men and women free of clinical disease who varied widely in adiposity (body mass index 18.4 to 36.7 kg/m2; total bodyfat 5.8 to 55.0 kg; waist circumference: 63.0 to 122.9 cm). All 3 expressions of adiposity were positively associated withVECPE of the oxidant enzyme subunit NAD(P)H oxidase-p47phox (part correlation coefficient [rpart] 0.22 to 0.24, allP�0.05) and the antioxidant enzyme catalase (rpart�0.71 to 0.75, all P�0.001). Total body fat was positively associatedwith VECPE of nitrotyrosine (rpart�0.36, P�0.003), a marker of protein oxidation, and, in men, with Ser1177-phos-phorylated endothelial nitric oxide synthase (rpart�0.46, P�0.02), an activated form of endothelial nitric oxide synthase.Overweight/obese subjects (body mass index �25 kg/m2) had 35% to 130% higher VECPE of NAD(P)H oxidase-p47phox, nitrotyrosine, catalase, and the cytosolic antioxidant CuZn superoxide dismutase (all P�0.05), as well as a 56%greater VECPE of the potent local vasoconstrictor endothelin-1 (P�0.05) than normal-weight subjects (body mass index�25 kg/m2). Nuclear factor-�B protein expression was �60% to 100% greater in the most obese adults than in theleanest adults (P�0.01). These relations were independent of sex but were selectively reduced after accounting for theinfluence of plasma C-reactive protein, fasting glucose-insulin metabolism, or serum triglycerides.

Conclusions—Compared with their normal-weight peers, overweight and obese adults demonstrate increased vascularendothelial expression of NAD(P)H oxidase-p47phox and evidence of endothelial oxidative stress, with selectivecompensatory upregulation of antioxidant enzymes and Ser1177-phosphorylated endothelial nitric oxide synthase.Endothelin-1 and nuclear factor-�B protein expression also appear to be elevated in obese compared with lean adults.These findings may provide novel insight into the molecular mechanisms linking obesity to increased risk of clinicalatherosclerotic diseases in humans. (Circulation. 2007;115:&NA;-.)

Key Words: endothelium � obesity � endothelin � free radicals � antioxidants � atherosclerosis

The vascular endothelium plays an essential role in theinitiation and progression of atherosclerosis.1 Vascu-

lar endothelial cells produce proteins that influence sus-ceptibility to the development of atherosclerosis.2 In gen-eral, an “antiatherogenic” vascular endothelium expressesproteins that act to enhance nitric oxide (NO) bioavailabil-ity, inhibit excessive formation of reactive oxygen speciesand oxidative stress, suppress inflammation, and restrictlocal production of vasoconstrictor molecules.3 A“proatherogenic” vascular endothelium typically expressesthe reverse phenotype.

Clinical Perspective p ●●●

Experimental animal obesity and human obesity are asso-ciated with the development of atherosclerosis and increasedprevalence of clinical atherosclerotic diseases.4,5 Increasingtotal body fat and abdominal fat have been linked to impairedNO-dependent endothelial function, oxidative stress, andincreased production of vasoconstrictor proteins such asendothelin-1 (ET-1).6 These changes may in turn be related tochronic low-grade inflammation, insulin resistance, lipidabnormalities, and increases in circulating free fatty acids.7

Received August 9, 2006; accepted November 8, 2006.From the Department of Integrative Physiology (A.E.S., S.D.B., D.D.C., A.J.D., K.L.M., I.E., P.E.G., D.R.S.), University of Colorado, Boulder; and

Department of Medicine (D.R.S., K.L.M.), Divisions of Cardiology and Geriatric Medicine, University of Colorado Health Sciences Center, Denver.The online-only Data Supplement, consisting of expanded methods, a figure, and tables, is available with this article at

http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.106.657486/DC1.Correspondence to Douglas R. Seals, Department of Integrative Physiology, University of Colorado at Boulder, UCB 354, Boulder, CO 80309. E-mail

[email protected]© 2007 American Heart Association, Inc.

Circulation is available at http://www.circulationaha.org DOI: 10.1161/CIRCULATIONAHA.106.657486

1

Vascular Medicine

by guest on February 20, 2018http://circ.ahajournals.org/

Dow

nloaded from

http://circ.ahajournals.org/

Currently, however, there is no information in humansconcerning the effects of adiposity on vascular endothelialexpression of proteins involved in atherogenic processes.

Accordingly, the primary purpose of the present study wasto determine the relations between adiposity and vascularendothelial cell expression of selective proteins involved inNO production (endothelial NO synthase [eNOS]; phosphor-ylated eNOS-Ser1177 [P-eNOS]), oxidant enzyme systems[NAD(P)H oxidase-p47phox; xanthine oxidase], oxidativemodification of proteins (nitrotyrosine), antioxidant enzymes(catalase; cytosolic CuZn superoxide dismutase [CuZnSOD];mitochondrial Mn superoxide dismutase), inflammation (nu-clear factor-�B [NF-�B]; cyclooxygenase-2), and vasocon-striction (ET-1) in adult humans. To do so, vascular endothe-lial cells were obtained from adult men and women varyingwidely in body mass index (BMI), total fat mass, and waistcircumference, and protein expression was measured byquantitative immunofluorescence.8 –10 Relations betweenbody fatness and protein expression were assessed withregression analysis based on the entire cohort and by com-paring subgroups of overweight/obese and normal-weightadults. A secondary purpose was to examine potential inter-mediary mechanisms by which adiposity may be influencingvascular endothelial cell protein expression (VECPE).

MethodsSubjectsA total of 108 adults, 73 men and 35 women, were studied. Allsubjects had resting blood pressure �140/90 mm Hg and were freeof clinical cardiovascular diseases, diabetes mellitus, and other majorchronic diseases as assessed by medical history, physical examina-tion, resting ECG, and blood chemistries. Men older than 40 yearsand women older than 50 years of age were further evaluated withECG and blood pressure responses to incremental treadmill exerciseperformed to volitional exhaustion.11 Subjects were nonsmokers,were not regularly exercising (�90 minutes of aerobic activity perweek), and were not taking medications or dietary supplements (eg,antioxidants) that could influence the results. All procedures wereapproved by the Human Research Committee of the University ofColorado at Boulder. The nature, benefits, and risks of the studywere explained to the volunteers, and their written informed consentwas obtained before participation.

Study ProceduresAll measurements were performed at the University of Colorado atBoulder General Clinical Research Center after an overnight fast.

Body CompositionBody weight, BMI, and waist circumference, an index of abdominalfat,12 were determined as described previously.13 Total body fat wasdetermined with dual-energy x-ray absorptiometry (GE Lunar Corp,Madison, Wis; software version 6.80.002) as described previously.14

Resting Blood PressureResting blood pressure was measured over the brachial artery with a

semiautomated device (Dinamap, Johnson & Johnson, Arlington, Tex).

Vascular Endothelial Cell Protein ExpressionThe procedures used for collection of venous vascular endothelialcells and measurement of VECPE have been described in detail.8–10

Briefly, endothelial cells were collected from an antecubital veinwith 2 sterile J-wires briefly advanced (�4 cm beyond the tip of thecatheter) and retracted through an 18-gauge catheter. The wires weretransferred to a dissociation buffer solution, and cells were recovered

by washing and centrifugation. Cells were fixed with 3.7% formal-dehyde and plated on poly-L-lysine–coated slides (Sigma Chemical,St Louis, Mo).

For immunofluorescence staining, cells were rehydrated with PBSand rendered permeable with 0.1% Triton X-100 (Alfa Aesar, WardHill, Mass). After nonspecific binding sites were blocked with 5%donkey serum (Jackson Immunoresearch, West Grove, Pa), cells wereincubated with monoclonal antibodies for one of the following: eNOS(Transduction Laboratories, San Jose, Calif), P-eNOS (Ser1177,Calbiochem, San Diego, Calif), NAD(P)H oxidase-p47phox (Upstate,Billerica, Mass), xanthine oxidase (United States Biological,Swampscott, Mass), catalase (Abcam, Cambridge, Mass), CuZnSOD(Upstate), Mn superoxide dismutase (Research Diagnostics, Con-cord, Mass), nitrotyrosine (Abcam), ET-1 (Affinity Bioreagents,Golden, Colo); cyclooxygenase-2 (Cayman Chemicals, Ann Arbor,Mich) and NF-�B (Novus, Littleton, Colo). Cells were next incu-bated with CY3-conjugated secondary antibodies (Jackson Immu-noresearch or Research Diagnostics).

For analysis, slides were viewed with a fluorescence microscope(Eclipse 600, Nikon, Melville, NY) and were digitally captured by aPhotometrics CoolSNAPfx digital camera (Roper Scientific, Inc,Tucson, Ariz). Endothelial cells were identified by the presence ofvon Willebrand factor staining, and nuclear integrity was confirmedwith DAPI (4�,6�-diamidino-2-phenylindole hydrochloride) staining.Once endothelial cells with intact nuclei were identified, imageswere captured and analyzed with Metamorph Software (UniversalImaging Corp, Downingtown, Pa) to quantify the intensity of CY3staining (ie, average pixel intensity), hence quantifying VECPE. Thesoftware program allowed for systematic quantification of stainingintensity and eliminated human subjectivity in analysis. Values arereported as ratios of VECPE/human umbilical vein endothelial cell(control cells) VECPE. The reporting of ratios minimizes day-to-dayvariability resulting from differences in intensity of staining betweendifferent staining sessions. Technicians were blinded to subjectidentity during the staining and analysis procedures. Although slideswere not randomly assigned to a staining batch, they were selected sothat each batch included subjects with a wide range of adiposity.

The present technique of quantitative immunofluorescence ofhuman endothelial cells was originally validated against immuno-blotting by Colombo and colleagues,9 with an excellent correlationcoefficient (r�0.99), and more recently, by our laboratory, with asimilarly impressive correlation coefficient (r�0.97, P�0.007; seeonline Data Supplement).

Plasma Markers of Inflammation, Glucose-InsulinMetabolism, and LipidsLipid profiles were determined with the ACE chemistry system (AlfaWassermann Diagnostic Technologies, West Caldwell, NJ). Serum totalcholesterol and triglycerides were analyzed by conventional enzymaticmethods. High-density lipoprotein cholesterol was determined with ahomogenous 2-reagent method, and low-density lipoprotein cholesterolwas calculated with the formula LDL�TC�HDL�(TG/5), in whichLDL indicates low-density lipoprotein; TC, total cholesterol; HDL,high-density lipoprotein; and TG, trigylcerides.15 Serum free fatty acids(Waco Chemicals, Neuss, Germany) and high-sensitivity C-reactiveprotein (Roche Diagnostic Systems, Basel, Switzerland) were measuredby enzymatic methods.

Fasting plasma glucose was measured by enzymatic methods (RocheDiagnostic Systems) and plasma insulin by radioimmunoassay (Diag-nostic Systems Laboratory, Webster, Tex). Insulin resistance wasestimated with the homeostasis model of insulin resistance by theformula [fasting plasma insulin (�U/mL)�fasting plasma glucose(mmol/L)]/25.16 The homeostasis model of insulin resistance has beenvalidated against other measures of insulin sensitivity (eg, intravenousglucose tolerance test) as a reliable estimate of insulin sensitivity.17

Plasma Markers of Oxidative Stress and Antioxidant StatusSerum oxidized low-density lipoprotein was determined by ELISA(ALPCO Diagnostics, Salem, NH), plasma concentrations of thio-barbituric acid reactive substances were determined by a spectro-photometric method (ZeptoMetrix Corp, Buffalo, NY), and 24-hour

2 Circulation February 6, 2007


Dow

nloaded from


urinary isoprostanes were measured by ELISA (Oxford BiomedicalResearch, Oxford, Mich). Plasma total antioxidant status and super-oxide dismutase-1 were measured with the Cobas Mira Plus Chem-istry Analyzer, and glutathione peroxidase was determined with theMira Method (all Randox Laboratories, Crumlin, Colo).

Data AnalysisStatistical analyses were performed with SPSS (version 11.0.3). Todetermine the relations of total and abdominal adiposity with VECPEindependent of sex, multiple linear regression analysis was per-formed. Separate regression models were used for each measure ofVECPE and each measure of adiposity (BMI, total body fat mass,and waist circumference). VECPE was entered as the dependentvariable, whereas sex and an expression of adiposity were entered asindependent variables. Sex was coded as 0 for men and 1 for women.Part (also known as semipartial) correlation coefficients derivedfrom regression analysis were used to determine the association ofadiposity with VECPE, independent of sex. Residual analyses to testthe validity of the regression model assumptions were performed forall regression models, and violations were remedied by data trans-formation. Pearson product-moment correlation coefficients wereused to determine bivariate relations in men and women separately.Similar regression analyses were performed to assess the relationsbetween markers of oxidative stress and antioxidant status and themeasures of adiposity. Pearson product-moment correlation coeffi-cients were used to determine bivariate relations between markers ofoxidative stress and antioxidant status and VECPE.

To further assess associations between adiposity and VECPE, wecompared VECPE in a subsample of overweight/obese (BMI �25kg/m2) and normal-weight (BMI �25 kg/m2) men and women usingt tests for independent group comparisons. The groups were com-posed of subjects from the overall study sample in whom most or allof the proteins were assessed. Further subgroup analyses wereperformed to compare VECPE between the most obese (highestquartile of BMI) and leanest (lowest quartile of BMI) men andwomen, as well as between individuals in the highest and lowestquartiles of waist circumference.

To assess inflammation (C-reactive protein), impaired glucosemetabolism (fasting glucose, fasting insulin, and the homeostasismodel of insulin resistance), resting blood pressure, and blood lipids(total cholesterol, low-density lipoprotein, high-density lipoprotein,triglycerides, and free fatty acids) as potential intermediary mecha-nisms by which adiposity influences VECPE, multiple linear regres-sion analyses were performed as described above. Each factor wasentered as an independent variable in the model, in addition toadiposity and sex. Using this approach, we determined how therelation between adiposity and VECPE was changed by accountingfor these factors in addition to sex. Statistical significance for allanalyses was set at P�0.05.

The authors had full access to and take full responsibility for theintegrity of the data. All authors have read and agree to themanuscript as written.

ResultsRegression Analysis on the Entire CohortMean values and ranges for sex composition, age, anthro-pometry, and body composition of subjects in the entirecohort are presented in Table 1. Subjects varied widely inBMI, total body fat, and waist circumference. Subject infor-mation, mean values, and ranges for each VECPE are shownin the online Data Supplement, Table I.

All 3 measures of adiposity were positively associated withVECPE of NAD(P)H oxidase-p47phox and catalase (Figures 1,

Figure 1. Relations between body mass index and VECPE ofNAD(P)H oxidase-p47phox (A) and catalase (B) adjusted for sexare represented by the regression line. rpart indicates the partcorrelation coefficient derived from multiple linear regressionanalysis; HUVEC, human umbilical vein endothelial cell.

TABLE 1. Subject Characteristics

Mean�SE Minimum–Maximum

Sex, m/f 73/35 � � �

Age, y 40�2 18–71

Height, cm 175�1 156–194

Weight, kg 78.6�1.4 49.7–124.6

Body mass index, kg/m2 25.7�0.4 18.4–36.7

Body fat, % 28.7�1.0 8.4–52.3

Body fat, kg 23.1�1.0 5.8–55.0

Waist circumference, cm 86.2�1.3 63.0–122.9

Total cholesterol, mmol/L 4.76�0.10 3.32–7.69

LDL cholesterol, mmol/L 2.91�0.08 1.50–5.70

HDL cholesterol, mmol/L 1.25�0.03 0.65–2.56

Triglycerides, mmol/L 2.90�0.13 1.22–8.62

CRP, mg/L 1.20�0.23 0.20–8.39

Glucose, mmol/L 5.04�0.07 3.55–6.72

Insulin, pmol/L 49.13�4.06 6.95–145.85

Homeostasis model of insulin resistance 3.67�0.34 0.36–10.73

Free fatty acids, �Eq/L 618�38 237–1179

Systolic blood pressure, mm Hg 115�1 85–139

Diastolic blood pressure, mm Hg 68�1 49–88

Data are mean�SE. m/f indicates male/female. LDL indicates low-densitylipoprotein; HDL, high-density lipoprotein; and CRP, C-reactive protein.

Silver et al Adiposity and Endothelial Cell Oxidative Stress 3


Dow

nloaded from


2A and 2B, and 3; all P�0.05). Total body fat was positivelyassociated with nitrotyrosine VECPE (Figure 2C, P�0.003).The relation between total body fat and P-eNOS VECPEdiffered in men and women (Figure 2D; P�0.04 for interac-

tion): There was a positive association in men (r�0.46,P�0.02) but not in women (P�0.8). There were no othersignificant relations between BMI, total body fat, or waistcircumference and VECPE. Results were similar if totalpercent body fat was used instead of total body fat mass.

Group Comparisons of Overweight/Obese andNormal-Weight AdultsMean values and ranges for sex composition, age, anthro-pometry, and body composition of the subjects in the over-weight/obese and normal-weight groups are presented inTable 2. All measures of adiposity were greater in theoverweight/obese subjects than in the normal-weight subjects(P�0.0001). Subject information, mean values, and rangesfor each VECPE in each group are shown in the online DataSupplement, Table II. Although mean age was higher in theoverweight/obese group, age was not related to expression ofany protein; thus, no correction of the VECPE values for agewas necessary.

Consistent with the results of the regression analysesperformed on the entire subject cohort, VECPE of NAD(P)Hoxidase-p47phox, catalase, nitrotyrosine, and P-eNOS weregreater (35% to 130%) in the overweight/obese subjects thanin the normal-weight subjects (Figure 4A through 4D;P�0.03). In addition, VECPE of CuZnSOD was �40%higher (P�0.04; Figure 4E) and VECPE of ET-1 was 56%higher (P�0.05; Figure 4F) in the overweight/obese groupcompared with normal-weight subjects.

When the most obese subjects were compared to theleanest individuals (ie, the highest and lowest quartiles ofBMI), the group differences in VECPE of catalase, NAD(P)Hoxidase-p47phox, nitrotyrosine, CuZnSOD, ET-1, and P-eNOSgenerally were comparable to the group differences betweenthe overweight/obese and normal-weight individuals indi-cated above, although they were not always statisticallysignificant because of smaller group numbers (online DataSupplement, Table III). The most notable difference was that

Figure 3. Relations between waist circumference and VECPE ofNAD(P)H oxidase-p47phox (A) and catalase (B) adjusted for sexare represented by the regression line. rpart indicates the partcorrelation coefficient derived from multiple linear regressionanalysis; HUVEC, human umbilical vein endothelial cell.

Figure 2. Relations between total bodyfat mass and VECPE of NAD(P)Hoxidase-p47phox, catalase and nitroty-rosine adjusted for sex (A, B, and C) orP-eNOS separately in men and women(D) are presented by the regression lines.D, Black regression line�relation inwomen; gray regression line�relation inmen; r values are Pearson product-moment correlation coefficients. rpart indi-cates the part correlation coefficientderived from multiple linear regressionanalysis; HUVEC, human umbilical veinendothelial cell.



Dow

nloaded from


VECPE of NF-�B was �60% greater in the most obese groupthan in the leanest group (0.45�0.06 [n�10] versus0.28�0.04 [n�9] NF-�B/human umbilical vein endothelialcell intensity, P�0.01).

Similarly, group differences in VECPE of catalase,NAD(P)H oxidase-p47phox, nitrotyrosine, CuZnSOD, ET-1,and P-eNOS between the individuals in the highest andlowest waist circumference quartiles generally were compa-rable to the group differences between the overweight/obeseand normal-weight individuals (online Data Supplement,Table IV). The only notable difference was that VECPE ofNF-�B was �100% greater (0.55�0.09 [n�9] versus0.27�0.03 [n�9] NF-�B/human umbilical vein endothelialcell intensity, P�0.009) in the group with the highest versusthe lowest waist circumference.

Influence of Inflammation, Glucose-InsulinMetabolism, Lipids, and Blood PressureThe relations of BMI, total body fat, and waist circumferencewith VECPE of NAD(P)H oxidase-p47phox were reduced by46% to 65% after we accounted for the influence of serumtriglycerides (Figure 5). There were no other significantinfluences of plasma lipids, blood pressure, markers ofinflammation, or glucose-insulin metabolism on the relationsbetween adiposity and VECPE of NAD(P)H oxidase-p47phox.

The relations of BMI, total body fat, and waist circumfer-ence with VECPE of catalase were reduced by 32% to 44%after we accounted for the influence of C-reactive protein,fasting insulin, or the homeostasis model of insulin resistance(Figure 6). Accounting for plasma lipids, resting bloodpressure, or fasting glucose alone did not significantly influ-

ence the relations between adiposity and VECPE of catalase.The relations between adiposity and VECPE of P-eNOS andnitrotyrosine were not significantly affected after we ac-counted for each of these factors.

Relations Between Systemic Markers of OxidativeStress and Antioxidant Status, Adiposity, and VECPESystemic markers of oxidative stress and antioxidant status werenot consistently related to the measures of adiposity, nor did theydiffer between normal-weight and overweight/obese subgroups(online Data Supplement, Table V). Plasma total antioxidantstatus was positively related to VECPE of CuZnSOD (r�0.48,P�0.01), however, and plasma superoxide dismutase-1 waspositively related to VECPE of catalase (r�0.77, P�0.01).

DiscussionThe primary purpose of the present study was to identifyrelations between total body and/or abdominal fat and theexpression of specific putative atherogenic risk–modulatingproteins in the vascular endothelium of adult humans. Thekey novel findings were that increased adiposity is associatedwith vascular endothelial oxidative stress, increased expres-sion of NAD(P)H oxidase-p47phox, and upregulation of selec-tive antioxidant enzymes. In addition, endothelial ET-1 andNF-�B protein expression appeared be elevated in obesecompared with lean adults. Our findings also suggest thatinflammation, insulin resistance, and plasma triglyceridesmay be among the intermediary mechanisms linking elevatedadiposity to altered VECPE. To the best of our knowledge,these are the first observations concerning the effects of

TABLE 2. Characteristics of a Subsample of Subjects With Lower andHigher Adiposity

Normal Weight Overweight/Obese P

Sex, m/f 23/16 29/13 � � �

Age, y 34�3 42�3 0.04

Height, cm 174�1 174�1 0.8

Weight, kg 67.0�1.5 88.1�1.5 �0.0001

Body mass index, kg/m2 22.0�0.3 29.2�0.4 �0.0001

Body fat, % 23.4�1.6 33.1�1.5 �0.0001

Body fat, kg 15.5�1.1 29.3�1.4 �0.0001

Waist circumference, cm 74.6�1.1 94.9�1.4 �0.0001

Total cholesterol, mmol/L 4.57�0.19 4.84�0.12 0.2

LDL cholesterol, mmol/L 2.75�0.16 3.02�0.11 0.2

HDL cholesterol, mmol/L 1.33�0.05 1.15�0.05 0.02

Triglycerides, mmol/L 2.32�0.14 3.35�0.24 �0.0001

CRP, mg/L 0.99�0.24 1.41�0.45 0.4

Plasma glucose, mmol/L 4.97�0.12 5.15�0.12 0.3

Plasma insulin, mmol/L 37.43�5.67 57.21�7.16 0.04

Homeostasis model of insulin resistance 2.78�0.49 4.30�0.58 0.05

Plasma free fatty acids, �Eq/L 567�51 661�66 0.3

Systolic blood pressure, mm Hg 109�2 118�2 0.003

Diastolic blood pressure, mm Hg 62�2 70�2 0.004

Data are mean�SE. m/f indicates male/female. LDL indicates low-density lipoprotein; HDL,high-density lipoprotein; and CRP, C-reactive protein.



Dow

nloaded from


human obesity on the expression of atherosclerosis-associated proteins in the vascular endothelium.

Adiposity and Vascular Endothelial Expression ofOxidative Stress–Related ProteinsOxidative stress represents increased production and bio-availability of reactive oxygen species such as superoxide(O2

�) relative to antioxidant defenses and is believed to be amajor contributor to the development of atherosclerosis.1,18

Obese individuals are at higher risk of clinical atheroscleroticdiseases,4 and this is postulated to be mediated in part byoxidative stress.19 There also is evidence that oxidative stressimpairs vascular endothelial function in obese humans.20

NAD(P)H oxidase, the major source of O2� in the vasculature,

is a multicomponent protein that is strongly implicated in thedevelopment of atherosclerosis.18,21 NAD(P)H oxidase-p47phox isa key cytosolic accessory protein that associates with membrane-bound NAD(P)H oxidase catalytic subunits (eg, p22phox) tostimulate vascular O2

� production.21 In the present study, BMI,total body fat, and waist circumference all were positivelyassociated with VECPE of NAD(P)H oxidase-p47phox in theoverall cohort. Moreover, NAD(P)H oxidase-p47phox VECPE

was �40% greater in the overweight/obese compared with thenormal-weight subjects. In contrast to NAD(P)H oxidase-p47phox, VECPE of xanthine oxidase, an enzyme that producesequimolar amounts of O2

� and hydrogen peroxide (H2O2),22 wasnot related to body fatness. Together, these results support theconcept that vascular endothelial expression of NAD(P)Hoxidase-p47phox, but not xanthine oxidase, is upregulated inoverweight/obese adult humans.

One consequence of oxidative stress is the modification ofbiomolecules, including proteins, lipids, and DNA.23 Onesuch modification of proteins involves the nitration of ty-rosine residues.24 Thus, elevated nitrotyrosine levels areinterpreted as an indicator of vascular oxidative stress.25

Nitrotyrosine content is increased in the thoracic aorta ofobese rats26; however, it is unknown whether this is observedin human obesity. In the present investigation, VECPE ofnitrotyrosine was positively associated with total body fat inthe entire group. In addition, nitrotyrosine expression was35% higher in the overweight/obese than in the normal-weight subjects. These observations are consistent with theidea that increased total body fat is associated with oxidativestress in the vascular endothelium in humans.

Figure 4. Comparison of VECPE innormal-weight and overweight/obeseadult humans. Values are mean�SE.HUVEC indicates human umbilical veinendothelial cell.



Dow

nloaded from


In the present study, VECPE of catalase, an antioxidantenzyme that catalyzes the conversion of H2O2 to water, waspositively associated with all 3 expressions of adiposity.Indeed, the relations between VECPE of catalase and mea-sures of total and abdominal fat (rpart�0.71 to 75) were by farthe strongest relations observed. Moreover, VECPE of cata-lase was �130% greater in overweight/obese than in normal-weight subjects. Additionally, VECPE of CuZnSOD, theSOD isoform that converts O2

� to H2O2 in the cytosol, was�40% higher in the overweight/obese group, although nodifferences were observed in the mitochondrial isoform (Mnsuperoxide dismutase). Given the increases in NAD(P)Hoxidase-p47phox and nitrotyrosine observed with increasingadiposity, one explanation for the present findings is that the

greater VECPE of catalase and CuZnSOD in the overweight/obese subjects represents a compensatory adaptation to oxi-dative stress. Indeed, reactive oxygen species induce expres-sion of antioxidant enzymes in vascular endothelial cells invitro.27 Furthermore, endothelial expression of antioxidants isincreased in vivo in response to oxidative stress under bothnormal physiological conditions and until the late stages ofatherosclerosis.28 It is not clear why, in the present study, Mnsuperoxide dismutase VECPE would not be similarly ele-vated. One possibility is that elevated body fatness does notcause increased mitochondrial O2

� production, thus negatingthe need for augmented expression of this antioxidantenzyme.

Figure 5. Part correlation coefficients showing the relationsbetween the 3 expressions of adiposity and VECPE of NAD(P)Hoxidase-p47phox adjusted for sex only (left). Additional adjust-ment for serum triglycerides significantly (P�0.005) reduced therelation of adiposity with VECPE of NAD(P)H oxidase-p47phox

(right).

Figure 6. Part correlation coefficients showing the relationsbetween the 3 expressions of adiposity and VECPE of catalaseadjusted for sex only (far left). Additional adjustment forC-reactive protein, plasma fasting insulin concentrations, andvalues for the homeostasis model of insulin resistance (HOMA-IR) significantly (P�0.005) reduced the relation of adiposity withVECPE of catalase (left to right).



Dow

nloaded from


It is noteworthy that increased adiposity was associatedwith increased VECPE of nitrotyrosine, despite selectiveupregulation of enzymatic antioxidants. It is possible that theincreased production of oxidants was greater than the corre-sponding increased expression of these antioxidants, thusallowing the development of oxidative damage. Alterna-tively, increased VECPE of catalase and CuZnSOD may nothave been associated with increased activity of these antiox-idant enzymes.

Finally, we did not observe consistent associations betweenseveral conventional systemic markers of oxidative stress/antioxidant status and adiposity in the present study. Onepossible explanation is that systemic oxidative stress was notpresent and systemic antioxidants were not upregulated in theotherwise healthy overweight and obese subjects in thepresent study, despite clear evidence for the presence ofendothelial oxidative stress and selective increases in endo-thelial antioxidant enzymes. An alternative explanation is thatthe present overweight/obese subjects did have mild tomoderate systemic oxidative stress but that the measures usedsimply were not sensitive enough to depict small differencesamong healthy adults differing in adiposity.

Adiposity and Vascular Endothelial Expression ofProteins Involved in NO ProductionAlthough eNOS protein expression is downregulated inadvanced atherosclerosis,29 eNOS expression is increased inpathophysiological states associated with oxidative stress.30

The latter is thought to be a compensatory attempt to increaseNO production and maintain its bioavailability in the pres-ence of increased vascular O2

� levels.30 Increased eNOSexpression has been observed in thoracic aortic homoge-nates26 and in small coronary arteries31 of obese rats. In thepresent study, however, VECPE of eNOS was not related toadiposity in the entire cohort, and mean values were notdifferent in the overweight/obese compared with normal-weight subjects.

In addition to influencing enzyme concentrations, how-ever, reactive oxygen species can modulate NO production byaffecting eNOS activation.32 For example, H2O2 can activatethe enzyme by phosphorylating eNOS at Ser1177.33 In thepresent investigation, VECPE of P-eNOS was positivelyassociated with total body fat in men and was 54% greater inthe overweight/obese group than in the normal-weight sub-jects. Taken together, the present results indicate that VECPEof eNOS is not increased in humans with elevated total bodyfat but that P-eNOS VECPE is greater in overweight andobese men. It is unclear why elevated adiposity was notassociated with increased P-eNOS VECPE in women. Dif-ferences in sex hormones34 and/or body fat distribution35 maycontribute to sex-related differences in VECPE. Estrogensexert a strong modulatory influence on the vascular endothe-lium,34 and thus, differences in estrogens between the maleand female subjects or within the females may have addedvariability to the VECPE–body fat relations. Females gener-ally demonstrate less abdominal, but greater lower-bodyadiposity than males.35 It is possible that this influencescirculating factors (eg, other hormones and cytokines) thatinteract with the endothelium and modulate gene and protein

expression. Additional investigations will be required tofurther elucidate the influence of sex on VECPE and todetermine the precise mechanisms by which sex modulatesVECPE.

Adiposity and Vascular Endothelial Expression ofET-1 and Inflammatory ProteinsET-1 is a potent vasoconstrictor factor produced by thevascular endothelium and is implicated in processes involvedin the development of endothelial dysfunction and atheroscle-rosis.36 Plasma concentrations of ET-1 are elevated in obesecompared with normal-weight humans37 and are reduced withweight loss.38 Furthermore, tonic ET-1 vasoconstriction isincreased in obese adults.39 However, ET-1 protein content isnot different in aortas of obese compared with lean mice.40 Inthe present investigation, there was no significant relationbetween adiposity and VECPE of ET-1 in the overall cohort,but ET-1 expression was 54% higher in the present sub-sample of overweight/obese compared with normal-weightsubjects. These findings suggest that ET-1 protein expressionmay be increased in the vascular endothelium of obesehumans and that additional work is warranted on this issue.

The important role of inflammation in atherogenesis iswidely accepted,41 and inflammation is considered an impor-tant mechanism by which obesity promotes atherosclerosis.42

Cyclooxygenase-2 is an inducible enzyme that produceseicosanoids believed to promote atherosclerosis,43 whereasNF-�B is a major transcription factor involved in activationof proinflammatory genes.44 In the present study, adipositywas not associated with the expression of cyclooxygenase-2.There was, however, a nonsignificant trend for NF-�BVECPE to be greater in the overweight/obese group than inthe normal-weight group. Moreover, NF-�B VECPE wassignificantly (�60% to 100%) greater in the subjects in thehighest versus the lowest quartiles of BMI and waist circum-ference. As such, the possibility that NF-�B VECPE may beupregulated in human obesity requires further study.

Associations With Total Body and Abdominal FatTotal body fatness and abdominal adiposity are stronglyassociated with endothelial dysfunction and risk of clinicalatherosclerotic diseases.6 In the present study, dual x-rayabsorptiometry–determined total fat mass was more consis-tently related to VECPE than either BMI or waist circumfer-ence. However, more direct measurements of abdominaladiposity (ie, with computed tomography or magnetic reso-nance imaging) will be necessary to determine the relationsbetween endothelial expression of these proteins and total,subcutaneous, and visceral abdominal fat.

Role of Inflammation, Glucose-Insulin Metabolism,Lipids, and Blood Pressure in Adiposity-RelatedDifferences in VECPEObesity likely affects the vascular endothelium via multiplemechanisms, including lipid abnormalities, insulin resistance,hypertension, and increases in inflammation.7,45 The resultsof the present study provide initial insight to support thepossibility that elevated triglycerides, insulin resistance, andinflammation may be among the mechanisms that contribute



Dow

nloaded from


to associations between adiposity and VECPE of NADPHoxidase-p47phox and catalase. Increased plasma triglycerides,46

insulin resistance,47 inflammation,48 and blood pressure49 areassociated with increased atherosclerotic risk. Although in-sight into the mechanisms by which insulin resistance47 andparticularly inflammation48 contribute to atherosclerosis isemerging, the mechanisms by which triglycerides influencethe atherosclerotic process in humans are not well under-stood. Recent in vitro data suggest that triglycerides modifyendothelial cell expression of atherosclerotic genes50; theassociations between plasma triglycerides and endothelialcell expression in humans in vivo have not been examinedpreviously, however. The present data provide preliminaryinsight into a novel mechanism by which triglycerides couldaffect cardiovascular risk, namely, via altered endothelialexpression of a key vascular oxidant enzyme system [ie,NAD(P)H oxidase-p47phox]. None of the factors we examinedhelped explain the relation of obesity with VECPE ofnitrotyrosine and P-eNOS. Resting blood pressure, at leastwithin the normal range of the healthy subjects assessed inthe present investigation, was not associated with any of therelations between adiposity and VECPE. Future studies willneed to determine whether other factors associated withobesity, such as changes in adipokines and renin-angioten-sin-aldosterone activity,7 may be mechanisms linking adipos-ity with altered VECPE.

Study LimitationsWe recognize that we studied a limited number of endothelialproteins and that future studies are needed to establishrelations between adiposity and expression of proteins in-volved in other aspects of atherosclerosis, such as celladhesion, monocyte chemoattraction, thrombosis, and vascu-lar smooth muscle proliferation. Additional studies also willbe required to determine the transcriptional, translational,and/or posttranslational mechanisms underlying adiposity-associated differences in endothelial protein expression, aswell as their functional consequences.

The measurements of VECPE in the present study wereperformed on cells obtained from veins as opposed toarteries; the latter would be optimal given that atherosclerosisis an arterial disease. However, the invasiveness of arterialcell collections is limiting in this regard. We believe that ourmeasurements of VECPE in venous cells are valid for severalreasons. First, although absolute expression of certain pro-teins (eg, eNOS) may differ in arteries and veins, theexpression of other proteins (eg, nitrotyrosine) appears to besimilar.8–10 Second, in our laboratory, we find a highlysignificant positive relation between expression of proteinsmeasured in cells obtained from venous compared witharterial samples collected from subjects on the same day(mean r�0.71, Silver et al, unpublished results). This sug-gests that although absolute VECPE may differ betweenarterial and venous cells, group differences or relationsbetween adiposity and VECPE should be observable withvenous cells. Third, expected differences in VECPE betweenhealthy controls and patients with cardiovascular disease areobserved in cells obtained from veins by the same proceduresas in the present study.9

In the present study, we measured the expression ofNAD(P)H oxidase-p47phox but not its phosphorylation status.Although NAD(P)H oxidase-p47phox phosphorylation is requiredfor NAD(P)H oxidase activation,51 the absolute expression ofNAD(P)H oxidase-p47phox likely influences superoxide produc-tion as well.52 Further studies will be required to identify theinfluence of adiposity on NAD(P)H oxidase-p47phox phosphory-lation status.

Finally, we recognize the semiquantitative nature of theimmunofluorescence technique; however, quantitative immu-nofluorescence has been validated against immunoblotting byour laboratory and others (supplemental materials and Co-lombo et al9). Moreover, all techniques for measuring proteinexpression are semiquantitative. Despite this limitation, wewere able to demonstrate significant relations between mea-sures of adiposity and VECPE. Indeed, it is likely that thestrength of these associations is underestimated and that someassociations may not have been detected owing to thesemiquantitative nature of the technique. Intervention studiesin which total body and abdominal fat are increased withoverfeeding or reduced with caloric restriction may provemore sensitive for establishing the role of adiposity in theregulation of VECPE. Future investigations also are neededto determine the influence of adiposity on gene expression ofendothelium-derived factors.

ConclusionsThe present study provides the first information concerningrelations between adiposity and expression of atherosclero-sis-linked proteins in the vascular endothelium of humans.Our results indicate that elevated body fatness is associatedwith increased vascular endothelial NAD(P)H oxidase pro-tein expression and evidence of endothelial oxidative stress.ET-1 and NF-�B protein expression also appear to beelevated in obese compared with lean adults. Antioxidantenzymes, particularly catalase, are selectively upregulated, asis P-eNOS, possibly as compensatory responses to oxidativestress. The present findings also suggest that elevated triglyc-erides, insulin resistance, and inflammation may be interme-diary mechanisms by which adiposity exerts its influence onVECPE. These observations may provide novel insight intothe molecular mechanisms underlying the association be-tween overweight/obesity and increased risk of clinical ath-erosclerotic diseases in humans.

AcknowledgmentsWe would like to thank Anita Banerjee, Thierry LeJemtel, andAshok Talreja from Albert Einstein School of Medicine for their rolein developing the measurements of VECPE, and Adam Levy, AshleyDepaulis, Brooke Lawson, and Rhea Chiang from the University ofColorado at Boulder for their technical assistance.

Sources of FundingThis work was supported by National Institutes of Health awardsAG006537, AG013038, AG022241, HL07851, and RR00051.

DisclosuresNone.



Dow

nloaded from


References1. Stocker R, Keaney JF Jr. Role of oxidative modifications in atheroscle-

rosis. Physiol Rev. 2004;84:1381–1478.2. Quyyumi AA. Endothelial function in health and disease: new insights

into the genesis of cardiovascular disease. Am J Med. 1998;105:32S–39S.3. Kunsch C, Medford RM. Oxidative stress as a regulator of gene

expression in the vasculature. Circ Res. 1999;85:753–766.4. Hubert HB, Feinleib M, McNamara PM, Castelli WP. Obesity as an

independent risk factor for cardiovascular disease: a 26-year follow-up ofparticipants in the Framingham Heart Study. Circulation. 1983;67:968–977.

5. Jousilahti P, Vartiainen E, Tuomilehto J, Puska P. Sex, age, cardiovas-cular risk factors, and coronary heart disease: a prospective follow-upstudy of 14 786 middle-aged men and women in Finland. Circulation.1999;99:1165–1172.

6. Avogaro A, de Kreutzenberg SV. Mechanisms of endothelial dysfunctionin obesity. Clin Chim Acta. 2005;360:9–26.

7. Williams IL, Wheatcroft SB, Shah AM, Kearney MT. Obesity, athero-sclerosis and the vascular endothelium: mechanisms of reduced nitricoxide bioavailability in obese humans. Int J Obes Relat Metab Disord.2002;26:754–764.

8. Feng L, Stern DM, Pile-Spellman J. Human endothelium: endovascularbiopsy and molecular analysis. Radiology. 1999;212:655–664.

9. Colombo PC, Ashton AW, Celaj S, Talreja A, Banchs JE, Dubois NB,Marinaccio M, Malla S, Lachmann J, Ware JA, Le Jemtel TH. Biopsycoupled to quantitative immunofluorescence: a new method to study thehuman vascular endothelium. J Appl Physiol. 2002;92:1331–1338.

10. Colombo PC, Banchs JE, Celaj S, Talreja A, Lachmann J, Malla S,DuBois NB, Ashton AW, Latif F, Jorde UP, Ware JA, LeJemtel TH.Endothelial cell activation in patients with decompensated heart failure.Circulation. 2005;111:58–62.

11. Gibbons RJ, Balady GJ, Beasley JW, Bricker JT, Duvernoy WF, FroelicherVF, Mark DB, Marwick TH, McCallister BD, Thompson PD Jr, WintersWL, Yanowitz FG, Ritchie JL, Gibbons RJ, Cheitlin MD, Eagle KA, GardnerTJ, Garson A Jr, Lewis RP, O’Rourke RA, Ryan TJ. ACC/AHA guidelines forexercise testing: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee onExercise Testing). J Am Coll Cardiol. 1997;30:260–311.

12. Onat A, Avci GS, Barlan MM, Uyarel H, Uzunlar B, Sansoy V. Measuresof abdominal obesity assessed for visceral adiposity and relation tocoronary risk. Int J Obes Relat Metab Disord. 2004;28:1018–1025.

13. Christou DD, Gentile CL, DeSouza CA, Seals DR. Gates PE. Fatness isa better predictor of cardiovascular disease risk factor profile than aerobicfitness in healthy men. Circulation. 2005;111:1904–1914.

14. Van Pelt RE, Davy KP, Stevenson ET, Wilson TM, Jones PP, DeSouzaCA, Seals DR. Smaller differences in total and regional adiposity withage in women who regularly perform endurance exercise. Am J Physiol.1998;275:E626–E634.

15. Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concen-tration of low-density lipoprotein cholesterol in plasma, without use of thepreparative ultracentrifuge. Clin Chem. 1972;18:499–502.

16. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, TurnerRC. Homeostasis model assessment: insulin resistance and beta-cellfunction from fasting plasma glucose and insulin concentrations in man.Diabetologia. 1985;28:412–419.

17. Bonora E, Targher G, Alberiche M, Bonadonna RC, Saggiani F, ZenereMB, Monauni T, Muggeo M. Homeostasis model assessment closelymirrors the glucose clamp technique in the assessment of insulin sensi-tivity: studies in subjects with various degrees of glucose tolerance andinsulin sensitivity. Diabetes Care. 2000;23:57–63.

18. Madamanchi NR, Vendrov A, Runge MS. Oxidative stress and vasculardisease. Arterioscler Thromb Vasc Biol. 2005;25:29–38.

19. Fenster CP, Weinsier RL, Darley-Usmar VM, Patel RP. Obesity, aerobicexercise, and vascular disease: the role of oxidant stress. Obes Res.2002;10:964–968.

20. Perticone F, Ceravolo R, Candigliota M, Ventura G, Iacopino S, SinopoliF, Mattioli PL. Obesity and body fat distribution induce endothelialdysfunction by oxidative stress: protective effect of vitamin C. Diabetes.2001;50:159–165.

21. Keaney JF Jr. Oxidative stress and the vascular wall: NADPH oxidasestake center stage. Circulation. 2005;112:2585–2588.

22. Berry CE, Hare JM. Xanthine oxidoreductase and cardiovascular disease:molecular mechanisms and pathophysiological implications. J Physiol.2004;555:589–606.

23. Droge W. Free radicals in the physiological control of cell function.Physiol Rev. 2002;82:47–95.

24. Radi R, Peluffo G, Alvarez MN, Naviliat M, Cayota A. Unravelingperoxynitrite formation in biological systems. Free Radic Biol Med.2001;30:463–488.

25. Berlett BS, Stadtman ER. Protein oxidation in aging, disease, and oxi-dative stress. J Biol Chem. 1997;272:20313–20316.

26. Kagota S, Yamaguchi Y, Tanaka N, Kubota Y, Kobayashi K, Nejime N,Nakamura K, Kunitomo M, Shinozuka K. Disturbances in nitric oxide/cyclic guanosine monophosphate system in SHR/NDmcr-cp rats, a modelof metabolic syndrome. Life Sci. 2006;78:1187–1196.

27. Weidig P, McMaster D, Bayraktutan U. High glucose mediates pro-oxidant and antioxidant enzyme activities in coronary endothelial cells.Diabetes Obes Metab. 2004;6:432–441.

28. ’t Hoen PA, Van der Lans CA, Van Eck M, Bijsterbosch MK, Van BerkelTJ, Twisk J. Aorta of ApoE-deficient mice responds to atherogenicstimuli by a prelesional increase and subsequent decrease in theexpression of antioxidant enzymes. Circ Res. 2003;93:262–269.

29. Oemar BS, Tschudi MR, Godoy N, Brovkovich V, Malinski T, LuscherTF. Reduced endothelial nitric oxide synthase expression and productionin human atherosclerosis. Circulation. 1998;97:2494–2498.

30. Li H, Wallerath T, Munzel T, Forstermann U. Regulation of endotheli-al-type NO synthase expression in pathophysiology and in response todrugs. Nitric Oxide. 2002;7:149–164.

31. Katakam PV, Tulbert CD, Snipes JA, Erdos B, Miller AW, Busija DW.Impaired insulin-induced vasodilation in small coronary arteries ofZucker obese rats is mediated by reactive oxygen species. Am J PhysiolHeart Circ Physiol. 2005;288:H854–H860.

32. Cai H, Li Z, Davis ME, Kanner W, Harrison DG, Dudley SC Jr. Akt-dependent phosphorylation of serine 1179 and mitogen-activated proteinkinase kinase/extracellular signal-regulated kinase 1/2 cooperativelymediate activation of the endothelial nitric-oxide synthase by hydrogenperoxide. Mol Pharmacol. 2003;63:325–331.

33. Thomas SR, Chen K, Keaney JF Jr. Hydrogen peroxide activates endo-thelial nitric-oxide synthase through coordinated phosphorylation anddephosphorylation via a phosphoinositide 3-kinase-dependent signalingpathway. J Biol Chem. 2002;277:6017–6024.

34. Orshal JM, Khalil RA. Gender, sex hormones, and vascular tone.Am J Physiol Regul Integr Comp Physiol. 2004;286:R233–R249.

35. Wingard DL. Sex differences and coronary heart disease: a case ofcomparing apples and pears? Circulation. 990;81:1710–1712.

36. Dashwood MR, Tsui JC. Endothelin-1 and atherosclerosis: potential com-plications associated with endothelin-receptor blockade. Atherosclerosis.2002;160:297–304.

37. Parrinello G, Scaglione R, Pinto A, Corrao S, Cecala M, Di Silvestre G,Amato P, Licata A, Licata G. Central obesity and hypertension: the roleof plasma endothelin. Am J Hypertens. 1996;9:1186–1191.

38. Ferri C, Bellini C, Desideri G, Di Francesco L, Baldoncini R, Santucci A,De Mattia G. Plasma endothelin-1 levels in obese hypertensive andnormotensive men. Diabetes. 1995;44:431–436.

39. Mather KJ, Lteif A, Steinberg HO, Baron AD. Interactions betweenendothelin and nitric oxide in the regulation of vascular tone in obesityand diabetes. Diabetes. 2004;53:2060–2066.

40. Barton M, Carmona R, Morawietz H, d’Uscio LV, Goettsch W, Hillen H,Haudenschild CC, Krieger JE, Munter K, Lattmann T, Luscher TF, ShawS. Obesity is associated with tissue-specific activation of renal angioten-sin-converting enzyme in vivo: evidence for a regulatory role of endo-thelin. Hypertension. 2000;35:329–336.

41. Gonzalez MA, Selwyn AP. Endothelial function, inflammation, andprognosis in cardiovascular disease. Am J Med. 2003;115:99S–106S.

42. Lyon CJ, Law RE, Hsueh WA. Minireview: adiposity, inflammation, andatherogenesis. Endocrinology. 2003;144:2195–2200.

43. Linton MF, Fazio S. Cyclooxygenase-2 and inflammation in atheroscle-rosis. Curr Opin Pharmacol. 2004;4:116–123.

44. De Martin R, Hoeth M, Hofer-Warbinek R, Schmid JA. The transcriptionfactor NF-kappa B and the regulation of vascular cell function. Arte-rioscler Thromb Vasc Biol. 2000;20:E83–E88.

45. Rahmouni K, Correia ML, Haynes WG, Mark AL. Obesity-associated hyper-tension: new insights into mechanisms. Hypertension. 2005;45:9–14.

46. Sharrett AR, Ballantyne CM, Coady SA, Heiss G, Sorlie PD, Catellier D,Patsch W; Atherosclerosis Risk in Communities Study Group. Coronaryheart disease prediction from lipoprotein cholesterol levels, triglycerides,lipoprotein(a), apolipoproteins A-I and B, and HDL density subfractions:the Atherosclerosis Risk in Communities (ARIC) Study. Circulation.2001;104:1108–1113.



Dow

nloaded from


47. Nigro J, Osman N, Dart AM, Little PJ. Insulin resistance and atheroscle-rosis. Endocr Rev. 2006;27:242–259.

48. Willerson JT, Ridker PM. Inflammation as a cardiovascular risk factor.Circulation. 2004;109(suppl II):II-2–II-10.

49. Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JLJr, Jones DW, Materson BJ, Oparil S, Wright JT Jr, Roccella EJ; JointNational Committee on Prevention, Detection, Evaluation, and Treatmentof High Blood Pressure; National Heart, Lung, and Blood Institute;National High Blood Pressure Education Program Coordinating Com-mittee. Seventh report of the Joint National Committee on Prevention,Detection, Evaluation, and Treatment of High Blood Pressure. Hyper-tension. 2003;42:1206–1252.

50. Norata GD, Grigore L, Raselli S, Seccomandi PM, Hamsten A, MaggiFM, Eriksson P, Catapano AL. Triglyceride-rich lipoproteins from hy-pertriglyceridemic subjects induce a pro-inflammatory response in theendothelium: molecular mechanisms and gene expression studies. J MolCell Cardiol. 2006;40:484–494.

51. Cai H, Griendling KK, Harrison DG. The vascular NAD(P)H oxidases astherapeutic targets in cardiovascular diseases. Trends Pharmacol Sci.2003;24:471–478.

52. Shi Y, Niculescu R, Wang D, Patel S, Davenpeck KL, Zalewski A.Increased NAD(P)H oxidase and reactive oxygen species in coronaryarteries after balloon injury. Arterioscler Thromb Vasc Biol. 2001;21:739–745.

CLINICAL PERSPECTIVECardiovascular disease risk is increased in overweight and obese humans compared with their normal-weight peers, butlittle is known about the molecular mechanisms involved. Vascular endothelial expression of proteins involved in nitricoxide production, the formation of reactive oxygen species and oxidative stress, inflammation, and local vasoconstrictionare believed to play an important role in determining the risk of vascular disease. In the present study, we sought todetermine whether endothelial expression of such proteins was related to total or abdominal body fatness in adult humanswithout clinical disease. Endothelial cells were collected from antecubital veins, and protein expression was measured withquantitative immunofluorescence. We found that compared with normal-weight adults, overweight and obese individualsdemonstrated increased endothelial expression of (1) NAD(P)H oxidase, an enzyme that produces superoxide anions andis implicated in states associated with vascular oxidative stress; (2) nitrotyrosine, a protein marker of oxidative stress; (3)endothelin-1, an endothelium-derived protein that promotes vasoconstriction; and (4) nuclear factor-�B, a factor thatpromotes oxidant and inflammatory gene transcription. There also were apparently compensatory increases in expressionof an activated form of endothelial nitric oxide synthase and the antioxidant enzymes catalase and CuZn superoxidedismutase. We found that elevated triglycerides, insulin resistance, and inflammation may be among the mechanisms thatcontribute to associations between adiposity and endothelial cell protein expression. These findings indicate thatoverweight and obese adult humans demonstrate altered vascular endothelial expression of proteins that may explain, inpart, their increased risk of cardiovascular disease.



Dow

nloaded from


Moreau, Iratxe Eskurza, Phillip E. Gates and Douglas R. SealsAnnemarie E. Silver, Stacy D. Beske, Demetra D. Christou, Anthony J. Donato, Kerrie L.

Expression and Evidence of Endothelial Oxidative StressphoxOxidase-p47Overweight and Obese Humans Demonstrate Increased Vascular Endothelial NAD(P)H

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 2007 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation published online January 22, 2007;Circulation.

http://circ.ahajournals.org/content/early/2007/01/22/CIRCULATIONAHA.106.657486.citationWorld Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://circ.ahajournals.org/content/suppl/2007/01/30/CIRCULATIONAHA.106.657486.DC1Data Supplement (unedited) at:

http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer this process is available in the

click Request Permissions in the middle column of the Web page under Services. Further information aboutOffice. Once the online version of the published article for which permission is being requested is located,

can be obtained via RightsLink, a service of the Copyright Clearance Center, not the EditorialCirculationin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:


Dow

nloaded from

http://circ.ahajournals.org/content/early/2007/01/22/CIRCULATIONAHA.106.657486.citation

http://circ.ahajournals.org/content/suppl/2007/01/30/CIRCULATIONAHA.106.657486.DC1

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circ.ahajournals.org//subscriptions/


overweight and obese humans demonstrate increased vascular

Documents