452

EXPERIMENTAL STUDIES

Dietary

i e Prevents Atherogenesils in the Coronary Artery of the

er

lester e c

it

BING-YIN WANG, MD, ALAN H. SINGER, MD, PHILIPS. TSAO, PHD,HELMUT DREXLER, MD, JON KOSEK, MD, JOHN P . COOKE, MD, PHD, FACC

Stanford, California

Objectives. This study was designed to t hypothesis thatWg4tm oral supplemental of die L-a e (to provide asustalned eleva of nitric oxide activity) Id inhibit athero-ger s In by

eroletnic rabbits, as

by histomor-measurements.

Background. End

lium-derived nitric oxide Inhibits a num-ber of

that are critical In a

s. Hypercholes-t

red

endothelial nitric oxide activity, and welate that this y

a

reduction in nitricactivity can be revs

acutely by intravenous i

ofL ., the precursor of nitric oxide . We show that dietaryso tatlim of L abrogates the development of cor-onary

hype

rolemic rabbits.M . Male New Zealand White rabbits were fed normal

rabbit chow, 1 % cholesterol chow or 1 % cholesterol chow withd ine or methioniie supplementation to increase theirIntake of these amino acids sixfold. After 1 or 10 weeks of dietaryintervention, the left train and left anterior descending coronaryarteries were harvested for histologic study. Plasma cholesterolmeasurements were elevated to the same degree in all groups of

Endothelium-derived relaxing factor is a potent endogenousvasodilator, released from the endothelium in response to anumber of hemodynamic or humoral stimuli. This substancehas the characteristics of a nitrovasodilator and is nitricoxide or a closely related molecule (1-3) . Nitric oxide isderived from the metabolism of L-arginine to L-citrulline by

From the Division of Cardiovascular Medicine, Stanford UniversitySchool of Medicine, Stanford, California . This work was supported in part bygrants from the National Heart, Lung, and Blood Institute (1PO1HL48638-01), National Institutes of Health, Bethesda, Maryland ; Institute of Biologicaland Clinical Investigation of Stanford University, Stanford, California ; theUniversity of California Tobacco Related Disease Program IRT 215, Oakland,California ; and the Peninsula Community Foundation, San Mateo, California.Dr. Wang is the recipient of a Ho Tim-Stanley Ho-Li Shing Award from theStanford University-Asia Medical Fund . Dr. Singer is the recipient of anAmerican Heart Association-Bugher Fellowship. Dr. Tsao is the recipient ofa National Research Service Award (1F32HL08779) . Dr. Cooke is a recipientof the Vascular Academic Award from the National Heart, Lung, and BloodL..ntitute (1K07HC02660-1), National Institutes of Health .

Manuscript received April 5, 1993 ; revised manuscript received August25, 1993, accepted September 8, 1993 .

Address for corresmgd= Dr. John P. Cooke, Division of Cardiovas-cular Medicine, Cardiovascular Research Center, Stanford University Schoolof Medicine, Stanford, California 94305.

01994 by the American College of Cardiology

rabbits receiving the 1 % cholesterol diet, whereas piss a argininelevels were doubled in the inine-treat group. High density0 rotein ( L) cholesterol values were not ected by argininetreatment.

Results. In rabbits receiving the 1 cholesterol diet, with orwithout methionine supplementation, light and electron micros-copy revealed a marked increase from 1 to 10 weeks in the intinralaccumulation of macrophages, associated with an increase in theint

a

of the left main coronary retry. By contrast, in. de-treated hypercholesterolemic rabbits, there was a near

absence of adherent monocytes and tissue macrophages and noprogression of inti

thickness from 1 to 10 weeks .Conclusions. Dietary supplements of L-arginine prevent ind-thickening in the coronary arteries of by rcholestcrolemic

rabbits . This antiathe enic effect is not due to an alteration inplasma total cholesterol, HDL cholesterol or caloric or nitrogenbalance . The data are consistent with the hypothesis that nitricoxide has antiathe epic properties.

(J Am Loll Cardiol 1994,23 .452-8)

JACC Vol . 23, No . 2February 1994:452-8

nitric oxide synthase, which has recently been cloned andcharacterized (4-6) . In addition to its effects on vasomotion,nitric oxide inhibits monocyte adherence and chemotaxis(7), platelet adherence and aggregation (8-10) and vascularsmooth muscle proliferation (11) . Because these processesare critically involved in atherogenesis, we postulate thatnitric oxide is an antiatherogenic molecule .

Wa and others (12-18) have shown that in animals andhumans, hypercholesterolemia inhibits nitric oxide-dependent vasodilation. We (16-18) recently demonstratedthat this abnormality can be reversed acutely by administer-ing large intravenous doses of L-arginine, the precursor ofnitric oxide . We subsequently found that long-term oraladministration of L-arginine induced a sustained elevation ofplasma arginine and a significant improvement in nitricoxide-dependent vasodilation in hypercholesterolemic rab-bits (19) . This effect occurred in the absence of any alterationin plasma cholesterol. The sustained augmentation of nitricoxide activity was associated with an antiatherogenic effect,as manifested by a 50% to 75% reduction in the thicknessand surface area of intimal lesions in the descending thoracic

0735-1097/941$7 .00

JACC Vol. 23, No . 2February 1994 :452-8

aorta (19) . We now report that supplements of dietaryL-arginine esseptially abolish the development of intimallesions in the left main coronary artery of hypercholester-olemic rabbits .

Methods

Animals. Male New Zealand White rabbits were assignedto one of three treatment groups : I I were fed normal rabbitchow (control group) ; 10 received chow containing 1%cholesterol ; I I received a 1% cholesterol diet supplementedwith 2.25% L-arginine hydrochloride in the drinking waterthroughout the course of the study, and 6 received a 1%cholesterol diet supplemented with 0.9% methionine hydro-chloride in the drinking water . Normal rabbit chow contains0.9% L-arginine and 0 .36% L-methionine by weight. Supple-mentation with 2 .25% L-arginine Cr 0.9% L-Mthionine indrinking water results in a sixfold increase in daily L-arginineor L-Methionine intake, respectively (based on the averagewater and food intake of the animals) . We previouslyshowed that this amount of L-arginine supplementation iswell tolerated and results in a twofold increase in plasma freearginine levels (19) . Supplementation with a 1 .2% L-argininesolution is also well tolerated but only increases plasmaarginine levels by 50% . Lower doses did not have a measur-able effect on plasma arginine . Solutions of L-arginine>2 .25% were not tolerated .

After I or 10 weeks of the dietary intervention, the rabbitswere ligntly sedated with a 3 .0-mg subcutaneous injection ofacepromazine maleate solution (Ayerst Laboratories), andthe central ear artery was cannulated for collection of bloodsamples. Subsequently, the animals were killed with anoverdose of pentobarbital sodium, and the heart and ascend-ing aorta were excised en bloc . Immediately thereafter, theaorta was cannulated and perfused under pressure(80 mm Hg) for I min with cold physiologic saline solution,followed by phosphate-buffered formalin for 5 min . Subse-quently, the left main and left anterior descending coronaryarteries were harvested for histologic study . The thoracicaorta was harvested from a subset of the rabbits (n = 16) forstudies of vascular structure and reactivity (these data havebeen reported separately [19]) . These protocols were ap-proved by the Administrative Panel on Laboratory AnimalCare of Stanford University and were performed in accor-dance with the recommendations of the American Associa-tion for Accreditation of Laboratory Animal Care .

Biochemical studies. Before initiation of the diet, andafter I or 10 weeks of treatment, blood samples werecollected from rabbits in each group for measurement ofplasma free arginine and plasma cholesterol . Blood wasobtained by cannulation of the central ear artery and col-lected into 10-ml glass vials containing potassium ethylene-diaminetetraacetic acid . The sample was then centrifuged at1,800 g for 10 min and the plasma supernatant collected. Formeasurement of plasma arginine, plasma was deproteinized

WANG ET AL .

453ANTIATHEROGENIC EFFECTS OF ARGININE

with 2% sulfosalicylic acid and analyzed for free arginineusing an automated amino acid analyzer (Beckman 6300), aspreviously described (19). Plasma cholesterol measurementswere performed using a modification of the enzymaticmethod of Allain et al . (20) developed by Sigma Diagnostics .This assay utilizes cholesterol oxidase to generate hydrogenperoxide, which is then coupled with a chromogen to yield aquinoneimine dye with an absorbance maximum at 500 nm .

Elistologic studies . Electron microscopy . Morphologicfeatures of the endothelium were examined by transmissionand scanning electron microscopy . For scanning electronmicroscopy, left anterior descending coronary artery seg-ments were bisected longitudinally, dehydrated in gradedethanol, critical point dried, sputter coated with gold andexamined with a IS140 scanning electron microscope . Fortransmission electron microscopy, vascular segments wereplaced in 2% glutaraldehyde, rinsed, postfixed in osmiumtetroxide, dehydrated and embedded in epoxy resin . Thinsections stained with uranyl acetate and lead hydroxide wereexamined with a Phillips 201 or Siemens IA electron micro-scope .

Light microscopy . The entire left main coronary arteryof each rabbit was fixed in 10% buffered formalin andembedded in paraffin . The left main coronary artery wasthen sectioned throughout its length (six sections takenevery 500 tim) for a total of 24 to 36 cross sections/animaland stained with the elastic van Gieson stain . Measurementsof intimal and medial areas of each cross section were madeby a skilled observer using an eyepiece grid with 100counting points (line intersections), each 0 .4 mm apart(21,22) . Total grid area (mm 2 ) was determined by measure-ment of grid dimensions with a stage micrometer at the samemicroscope magnification used for point counts . Total areaof the intima (or media) was defined as the number of gridpoints overlying the intima (or media) . The area of each wascalculated by the formula 0TI(P) x grid area, where P is thenumber of points counted over the intima (or media). Allcross sections from each left main coronary artery (24 to 36cross sections) were analyzed and the values averaged togive a single value for the intimal and medial cross-sectionalarea for each vascular segment . In addition to intimal lesionarea, we determined the circumferential extent of the intimallesion or lesions by using the grid to measure the entirecircumference and that involved by any lesions . The circum-ferential extent of the lesion or lesions is expressed as apercent of the entire circumference .

Data analysis. Data are expressed as mean value ± SEM .Measurements of intimal and medial thickness, intima/mediaratios and circumferential extent were compared using inui-tivariate analysis of variance of independent groups todetermine the overall difference, followed by a post hocFisher test to determine statistical significance betweengroups . Statistical significance was accepted at the 95%significance level (p :s 0 .05) .

454

WANG ET AL .ANTIATHEROGENIC EFFECTS OF ARGININE

TaMe 1. Biochemical Measurements

Tignificamly different from control value at the same time point, p < 0 .05 . Values presented are mean value ± SEM . Arg = rabbits receiving 1% cholesteroldiet supplemented with arginine ; Chol = rabbits receiving chow containing 1% cholesterol; Control = rabbits receiving normal chow ; HDL = high densitylipoprotein; Meth = rabbits receiving 1% cholesterol diet supplemented with methionine .

ResultsBiochemical and hemodynamic measurements. After

I week, plasma cholesterol was significantly elevated inrabbits receiving the 1% cholesterol diet (Table 1) . After 10weeks of cholesterol feeding, the elevation in plasma cho-lesterol was maintained (Table 1) in all groups receiving the1% cholesterol diet .

Plasma arginine levels were doubled at I week in thearginine-treated rabbits compared with levels in the controlgroup (Table 1) . Plasma arginine levels were not altered inthe hypercholesterolemic rabbits receiving L-methionine orvehicle .

In a separate series of studies, we sought to determinewhether arginine treatment altered the level of high densitylipoprotein (HDL) cholesterol values . Rabbits were fednormal chow (n = 6), a 1% cholesterol diet (n = 6) or a 1%cholesterol diet supplemented with 2.25 1-7, arginipt in thedrinking water (n = 6) . After 10 weeks, blood was collectedand processed as previously described for plasma total choles-terol and HDL cholesterol measurements. The total choles-terol and HDL cholesterol in the arg'nine-treated cholesterol-fed rabbits was not different from that in the vehicle-treatedcholesterol-fed rabbits (Table 1) .

%womawvOkommWetwric ants after I week of treat .ment . Electron microscopy. Scanning electron microscopyof the left anterior descending coronary artery and aortafrom rabbits receiving normal chow revealed an intactendothelial lining with fusiform cells all ned in the directionof blood flow, which is typical of normal endothelium . Bycontrast, in both hypercholesterolemic groups, ocr, ~sionalmonocytes were found to be adherent to the endothelium,and occasional deformations of the endothelial surface were

Table 2 . Histomorphometric Measurements of Left Main Coronary Artery

observed. Transmission electron microscopy revealed thatthese deformations were due to small localized collections oflipid-laden macrophages in the subendothelial space .

Light microscopy . In rabbits receiving normal chow, theintimal area was not measurable . In both of the hyper-cholesterolemic groups . rare superficial and localized intimallesions (fatty streaks) were observed in the aorta and leftmain coronary artery . However, the cross-sectional areas ofthe intima and media were not different from those ofanimals receiving normal chow (Table 2) .

Histomorphometric measurements after 10 weeks of treat®ment . Electron microscopy . In rabbits receiving normalchow, there was no difference in the appearance of the leftanterior descending coronary artery endothelium comparedwith that of animals killed at I week . In the left anteriordescending coronary arteries of hypercholesterolemic (cho-lesterol-supplemented) rabbits, small endothelial ulcerationswere observed by scanning electron microscopy . Adherentmonocytes Were abundant (Fig . IA), some of them dome-shaped with typical ruffled surfaces, others with a veryirregular mottled surface . These changes were most notablenear the ostia of branching vessels . Lipid-laden smoothmuscle cells and macrophages were observed within the in-tima and subjacent media by transmission electron micros-copy (Fig. M). In rabbits receiving supplemental arginine,these changes were distinctly less pronounced . Only occa-sional adherent monocytes and small localized collections ofmacrophages in the subendothelial space were observed .

Light microscopy. In rabbits receiving normal chow, theintimal area was not measurable . In the left main coronaryartery of the rabbits receiving supplemental cholesterol,intimal lesions were observed; there was a significant in-

JACC Vol . 23 . No. 2February 1994 :452.-8

'Significantly different from control and arginine-supplemented diet values at the same time period . tPercent of circumference of the cross section involvedby lesions. Values presented are mean value ± SEM . Abbreviations as in Table 1 .

Week I Week 10Control Choi Arg Control Choi Arg Meth

Intimal area (JAM') 0 13 ± 6 6 ± 6 0 193 ± 108* 1 ± I 202 ± 73*Medial area (,um2) 118 ± 19 184 ± 80 194 ± 8 87 ± 4 210 ± 33 127 ± 15 182 ± 23Intimatinedia ratio 0 0.11 ± 0.05 0.01 ± 0 .01 0 0.83 ± 0.35* 0.003 ± 0.003 0 .99 ± 0.29*Circumference (%)t 0 14 ± 6 .6 6 .6 ± 6 .6 0 46 ± le 1 .7 ± 1 .7 55 ± 15*

1 Week 10 Weeks

Control(n = 3) Choi (n = 3) Arg (n = 3)

Control(n = 5-6)

Chol(n = 5-6)

Arg(n = 5-6)

Meth(n = 6)

Plasma cholesterol (mg/dl) 141 t 99 545 t 200* 428 t 247* 55 t 10 643 Q 687 ± %* S'S ± 1VPlasma HDL cholesterol (mg/dl) 8.2 t 1 .3 8.1±1 .8 7 .1 ± 0 .4Plasma arginine (mm) 0.13 t 0.02 0.11 t 0 .02 0 .24 t 0 .05* 0 .15 t 0.02 0.15 0.02 0 .24 ± 0 .04* 0.12 ± 0.02

JACC Vol . 23, No . 2February 1994 :452-8

Figure 1 . A, Scanning electron micrograph of the left anteriordescending coronary artery 10 weeks after initiation of a 1%cholesterol diet reveals monocytes adhering to the endothelium .Monocyte adherence was markedly reduced in hypercholester-olemic rabbits receiving supplemental arginine . B, Transmissionelectron micrograph of the left anterior descending coronary artery10 weeks after initiation of a 1% cholesterol diet reveals a lipid-ladenmacrophage (arrow) arising from the subendothelial space through agap in the endothelial cell layer (asterisks) . Collections of lipid-ladenmacrophages with endothelial deformation and microulcerationwere prominent in hypercholesterolemic rabbits receiving vehiclebut not in those receiving arginine supplementation .

crease in intimal thickness compared with that of left maincoronary arteries harvested at I week in this group, indicat-ing significant progression of disease (Table 2, Fig . 2). Bycontrast, in left main coronary arteries from rabbits receiv-ing supplemental arginine ., only superficial and localized

intimal lesions were observed . There was no increase inintimal thickness or circumferential extent of the lesionscompared with that in left main coronary arteries harvestedfrom rabbits receiving supplemental arginine at I week .Indeed, the intimal area and intimal/medial ratio were notdifferent from those in rabbits receiving normal chow(Fig. 3) .

In the hypercholesterolemic rabbits receiving methioninesupplementation intimal lesions consisting of large collec-tions of lipid-laden macrophages were observed . The intimalarea, intimallmedial ratio and circumference of the coronaryartery lesions were not different from those of the vehicle-treated hypercholesterolemic rabbits (Table 2, Fig . 2) .

DiscussionAndatheregenic effects of L-21'ginille . The salient finding

of this study is that intimal lesions did not develop in thecoronary arteries of hyperc hole sterolemic rabbits receivingL-arginine supplementation . This effect was not due to analteration by L-arginine in plasma total cholesterol or HDLcholesterol. Furthermore, this effect was not mimicked bydietary supplementations of an amino acid that is not metab-olized by nitric oxide synthase . These findings are consistentwith the hypothesis that arginine is antiatherogenic . Wepropose that this property of arginine is due to its conversionto nitric oxide .

L-Arginine is the primary prc cursor of nitric oxide (4) . Inprevious studies we have found that the attenuation byhypercholesterolemia of nitric oxide-dependent vasodila-tion can be acutely reversed by administration of this pre-cursor but not its inactive enantiomer D-aiginine (16-18,23) .In hypercholesterolemic rabbits, endothelium-dependentvasoddation of the thoracic aorta and hind limb resistancevow els is improved by intravenous infusion of 1-arginine(16,17). Endothelium-dependent vasodilation of forearm re-sistance vessels in hypercholesterolemicc humans is alsonormalized by intravenous administration of 1-arginine (18) .The reversibility by arginine of the endothelial dysfunctioninduced by hypercholesterolemia has now been confirmedby other investigators (24-26) .

These findings led us to hypothesize that long-termdietary supplementation of arginine would induce a sus-tained improvement in nitric oxide-dependent vasodilation .We further postulated that this long-term augmentation ofnitric oxide activity would retard atherogenesis . Indeed, wediscovered that a sixfold enrichment of dietary argininedoubled plasma arginine levels, did not affect plasma cho-lesterol values and improved nitric oxide-dependent vaso-dilation in the descending thoracic aorta of hypercholester-olemic rabbits (19) . These effects are associated with asignificant reduction in the extent ant, thickness of intimallesions in the thoracic aorta.

We now report that there is a virtual absence of lesions inthe coronary arteries of arginine-treated hypercholester-olemic rabbits . Arginine is at least as potent an antiathero-

WANG ET AL .

455ANTIATHEROGENIC EFFECTS OF ARGININE

456

WANG ET AL .ANTIATHEROGENIC EFFECTS OF ARGININE

Figure 2. Photomicrographs of representative cross sections from theleft main coronary arteries of rabbits treated with vehicle (A),supplemental methionine (B) and supplemental arginine (C) after 10weeks on a 1% cholesterol diet . There is a striking absence of intimallesions in the coronary arteries of the rabbits given supplementalarginine . All photomicrographs were taken at a magnification of x 100.

Figure 3. Bar graph of intimalmedia ratios of the left main coro-nary artery after I or 10 weeks of treatment in two groups ofhypercholesterolemic rabbits : those given vehicle (CHOL) andthose given supplemental arginine (ARG) . Values shown aremean * SEM (vertical bars) . *Significantly different from all groups,p < 0.05 .

I week

CHOL

10 woo* I weak 10 weak$

ARG

JACC Voi . 23, No. 2February 1994:452-8

genic agent in hypercholesterolemic rabbits as calcium chan-nel antagonists, converting enzyme inhibitors or propranolol(27-29). We now show that this effect of arginine is notmimicked by another L-amino acid, nor is it associated withchanges in total or HDL cholesterol . Previously it has beenhypothesized that increases in dietary arginine may exert anantiatherogenic effect by virtue of favorab!e alterations inthe lipoprotein profile (30), but this explanation seems un-likely in view of the present data .

Role of nitric oxide. We have proposed that arginineenhances the synthesis and release of endothelium-derivednitric oxide from the hypercholesterolemic vessel, overcom-ing an increased degradation (31) or reduced synthesis (32),or both, induced by hypercholesterolemia. An increasedrelease of nitric oxide from the endothelium may inhibitatherogenesis by limiting interactions of circulating bloodelements with the vessel wall . Indeed, monocyte adherenceto endothelial cells in culture, and chemotaxis in response toformyl-methionyl-leucine-phenylalanine, is inhibited byagents that release nitric oxide (7) . This effect on adherenceand chemotaxis may be mediated through effects on theexpression of adhesion molecules. Indeed, nitric oxide ap-pears to modulate the expression of CD11/18 by neutrophils .

JACC Vol. 23, No . 2February 1994 :452-8

In the perfused rat mesenteric venule, N-methyl-arginine (anitric oxide synthase antagonist) increases the adhesion ofneutrophils to the endothelium by an order of magnitude, aneffect that is fully blocked by administration of arginine (33 .The effect of N-methyl-arginine is associated with upregula-tion of CEM 1/18, as assessed by flow cytometry (33) .

The effect of nitric oxide on the expression of thisadhesion molecule may explain the beneficial effect of nitro-vasodilators in reperfusion injury . In reperfusion injury,neutrophil adhesion and infiltration occur after flow is re-stored to the experimentally occluded coronary artery . Thisleukocyte-endothelial cell interaction is inhibited by aninfusion of sodium nitrite (34). Intracoronary infusion ofarginine has the same protective effect, which appears to bedue to its metabolism to nitric oxide because the effect isblocked by nitric oxide synthase antagonists (35) . Thus, it ispossible that inhibition by nitric oxide of leukocyte-endothelial cell interaction is involved in the antiatherogeniceffect of arginine . In addition, the antiplatelet effects ofnitricoxide (8-10) and its ability to suppress vascular smoothmuscle growth (11) may contribute to the salutary effects ofarginine. Finally, there is preliminary evidence that nitricoxide may interfere with the oxidation of low densitylipoprotein cholesterol by macrophages in cell culture (36).

Clinical relevance. Studies in our laboratory are underway to determine the mechanism or mechanisms of theantiatherogenic effect of L-arginine and to determinewhether the findings are relevant to humans . The short-termresponse to arginine supplementation in hypercholester-olemic humans is similar to that of our animal model(16-18,23,26), so it is not unreasonable to anticipate anti-atherogenic effects of dietary arginine supplementation inhumans . Indeed, there is indirect support for this hypothe-sis. A recent epidemiologic study reveals that coronaryartery disease is reduced in heavy consumers of nuts andlegumes (37,38), food items that are rich in arginine (39) .

Conclusions . We find that dietary arginine supplementa-tion inhibits the development of atherosclerotic lesions in theaorta and coronary arteries of the hypercholesterolemicrabbit . This effect is most pronounced in the coronaryarteries, where progression of disease was essentially abol-ished. These effects are not due to changes in caloric ornitrogen balance or plasma cholesterol . The effects of argi-nine may be exerted through its metabolism to nitric oxide .Further studies of the mechanism of this antiatherogeniceffect and its relevance to human disease are warranted,given the potential ramifications that these findings may haveon public health and nutrition .

References1 . Furchgott RF . Studies on relaxation of rabbit aorta by sedium nitrite : the

basis for the proposal that the acid-activatable inhibitory factor fromretractor penis is inorganic nitrite and the enticthelium-derived relaxingfactor is nitric oxide . In : Vanhoutte PM, editor . Mechanisms of Vasodi-lation . New York: Raven, 1988:401-4.

WANG ET AL .

457ANTIATHEROGENIC EFFECTS OF ARGININE

2. lgnarro U, Bums RE, Buga GM, Wood KS . Endothelium-derivedrelaxing factor from pulmonary artery and vein possesses pharmacologicand chemical properties identical to those of nitric oxide radical . Circ Res1987 ;61 :866-79 .

3 . Palmer RMJ, Ferridge AG, Moncada S . Nitric oxide release accounts forthe biological activity of endothelium-derived relaxing factor . Nature1987 ;327 :524-6.

4. Palmer RMJ, Ashton DS, Moncada S . Vascular endothelial cells synthe-size nitric oxide from L-argifline . Nature 1988 ;333:664-6.

5 . Pollock JS, Fdrstermann U, Mitchell JA, et al . Purification and charac-terization of particulate endothelium-derived relaxing factor synthasefrom cultured and native bovine aortic endothelial cells . Proc N411 AcadSci U S A 1991 ;88:10480-4.

6 . Lamas S, Marsden PA, Li GK, Tempst P, Michel T . Endothelial nitricoxide synthase : molecular cloning and characterization of a distinctconstitutive enzyme isoform . Proc Natl Acad Sci U S A 1992 ;89 :6348-52 .

7 . Bath PMW, Hassall DG, Gladwin A-M, Palmer RMJ, Martin JF . Nitricoxide and prostacyclin . Divergence of inhibitory effects on monocytechemotaxis and adhesion to endothelium in vitro. Arterioscler Thromb1991 ;11 :254-60 .

8 . Radomski MW, Palmer RMJ, Moncada S . Comparative pharmacology ofendothelium-derived relaxing factor, nitric oxide, and proslacyclin inplatelets . Br J Pharmacol 1987 ;92 :181-7 .Stantler JS, Mendelsohn ME, Amarante P, et al . N-acetylcystein-e poten-dates platelet inhibition by endothelium-derived relaxing factor . Circ Res1989,65 :789-95 .

10. Radomski MW, Palmer RMJ, Moncada S . An L-arginine/nitric oxidepathway present in human platelets regulates aggregation . Proc Nall AcadSci U S A 1990 ;87 :5193-7 .

11 . Garg UC, Hassid A . Nitric oxide-generating vasodilators and 8-bromo-cyclic guanosine monophosphate inhibit mitogenesis and proliferation ofcultured rat vascular smooth muscle cells . J Clin Invest 1989 ;83 :1774-7 .

12 . Verbeuren TJ, Jordaens FH, Zonnekeyn LL, VanHove CE, Coene MC,Herman AG . Effect of hypercholeserolemia on vascular reactivity in therabbit . Circ Res 1986;58 :552-64 .

13 . Cohen RA, Zitnay KM, Haudenschild CC . Cunningham LD. Loss ofselective endothelial cell vasoactive functions caused by hypercholester-olemia in pig coronary arteries . Circ Res 1988 ;63 :903-!0 .

14 . Kugiyama K, Kerns SA, Morrisett JD, Roberts R, Henry PD . Impairmentof endothelium-dependent arterial relaxation by lysolecithin in modifiedlow-density lipoproteins . Nature 1990 ;344:160-2 .

IS . Ludmer PL, Selwyn AP, Shook TL, et al . Paradoxical vasoconstrictioninduced by acetylcholine in atherosclerotic coronary arteries . N Engl JMed 1986,315 :1046-51 .

16. Girerd XJ, Hirsch AT, Cooke JP, Dzau VJ, Creager MA . L-arginineaugments endothelium-dependent vasodilation in cholesterol-fed rabbits .Circ Res 1990 ;67 :1301-8 .

17. Cooke JP, Andon NA, Girerd XJ, Hirsch AT, Creager MA . Argininerestores cholinergic relaxation of hypercholesterolemic rabbit thoracicaorta . Circulation 1991 ;83 :1057-62 .

18 . Creager MA, Girerd XJ, Gallagher SJ, Coleman S, Dzau VJ . Cooke JP .L-arginine improves endothelium-dependent vasodilation in hypercholes-terolemic humans. J Clin Invest 1992 ;90 :1168-72,

19 . Cooke JP, Singer AH, Tsao P . Zera P, Rowan RA, Billingham ME .Antiatherogenic effects of L-arginine in the hypercholesterolemic rabbit .J Clin Invest 1992 ;901168-72.

20 . Alta., CA, Poon LS, Chan CSG, Richmond W, Fu PC . Enzymaticdetermination of total serum cholesterol [abstract] . Clin Chem 1974 ;20:470-5 .

21 . Aherne WA, Dunnill MS . Morphometry . Sevenoaks, Kent, England :Edwin Arnold, 1982 :33-44 .

22 . Manwaring L, O'Connell DL, Bhagwandeen BS, Zardawi IM, DobsonAL Morphometric analysis of coronary artery stenosis : an accuracy andreliability study . J Pathol 1986 ;156 :111-7 .

23 . Rossitch E Jr, Alexander E 111, Black P, Cooke JP . L-arginine normalizesendothelial function in cerebral vessels from hyuercholesterolemic rab-bits . J Clin Invest 1991 ;87:1295-9.

24 . 'Tanner FC, Noll G, Boulanger CM, Uscher TF . Oxidized low densitylipoproteins inhibit relaxations of porcine coronary arteries . Role ofscavenger receptor and endothelium-derived nitric oxide . Circulation1991 ;83 :2012-20 .

25 . Kuo L, Davis MF, Cannon MS, Chilian WM . Pathophysiological conse-

458

WANG ET AL .ANTIATHEROGENIC EFFECTS OF ARGININE

quences of atherosclerosis extend into the coronary microcirculation :restoration of endothelium-dependent responses by L-arginine . Circ Res1992 ;70 :465-76.

26. Drexler H . Zeiher AM, Meinzer K, Just H . Correction of endothelialdysfunction in the coronary microcirculation of hypercholesterolemicpatients by L-arginine . Lancet 1991 ;338:1546-50 .

27. Henry PD, Bentley KI . Suppression of atherosclerosis in cholesterol-fedrabbits treated with nifedipine . J Clin Invest 1981 ;68 :1366-9.

28 . Chob3ttian AV . 1989. Corcoran Lecture: adaptive and maladaptive re-sponses of the arterial wall to hypertension . Hypertension 1990;15:666-74.

29. Chobanian AV, Brecher P, Chan C. Effect of propranolol on atherogen-esis in the cholesterol-fed rabbit. Cite Res 1985;56 :755-62 .

30 . Minor RL, Myers RP, Guerra R, Bates JN, Harrison DO . Diet-inducedatherosclerosis increases the release of nitrogen oxides from rabbit aorta .J Clin Invest 1990;86 :2109-16.

31 . Kritchevsky D, Tepper SA, Czarnecki SK, Klurfeld DM . Atherogenicityof animal and vegetable protein : influence of the lysine to arginine ratio .Atherosclerosis 1982 ;41 :429-31 .

32. Mitchell JA, Warner TD, Huang Z-J, FOrstermann U, Murad F . NativeLDL inhibits the release of endothelial derived relaxing factor by reducing

the activity of endothelial nitric oxide synthase [abstract] . J Vase Res1992 ;29 :169 .

33 . Kubes P . Suzuki M, Granger DN . Nitric oxide : an endogenous modulatorof leukocyte adhesion . Proc Natl Acad Sci U S A 1991 ;88:4651-5 .

34 . Lefer AM, Tsao P, Johnson G . Role of endothelium-derived relaxingfactor as a cardioprotective agent in myocardial ischemia . In : Rubanyi G,Vanhoutte P . editors. Endothelium-Derived Contracting Factors . Basel:Karger, 1990 :190-7 .

35. Weyrich AS, Ma X-L, Lefer AM . The role of L-arginine in amelioratingreperfusion injury after myocardial ischemia in the cat . Circulation1992 ;86:279- .

36. Mao S . Yates M, Lambert L . et al . Nitric oxide protects the oxidativemodification of low density lipoprotein by macrophages [abstract] .FASEB J 1992 ;6:A1030.

37. Fraser GE, Sabatd J, Beeson WL, Strahan TM . A possible protectiveeffect of nut consumption on risk of coronary heart disease . Arch InternMed 1992 ;152 :1416-24 .

38. Ccoke JP, Tsao P. Singer A, Wang B . Kosek J, Drexler H . Anti-atherogenic effect of nuts : is the answer NO? Arch Intern Med 1993 ;153 :896-9 .

39. Pennington JAT, editor . Bowes and Church's Food Values of PortionsCommonly Used. 15th ed . Philadelphia : Lippincott, 1989 :1-328 .

JACC Vol . 23, No. 2February 1994 :452-8