• No results found

Angiotensin II formation in the intact human heart Predominance of the angiotensin converting enzyme pathway

N/A
N/A
Protected

Academic year: 2020

Share "Angiotensin II formation in the intact human heart Predominance of the angiotensin converting enzyme pathway"

Copied!
10
0
0

Loading.... (view fulltext now)

Full text

(1)

Angiotensin II formation in the intact human

heart. Predominance of the

angiotensin-converting enzyme pathway.

L S Zisman, … , B M Groves, M V Raynolds

J Clin Invest.

1995;

96(3)

:1490-1498.

https://doi.org/10.1172/JCI118186

.

It has been proposed that the contribution of myocardial tissue angiotensin converting

enzyme (ACE) to angiotensin II (Ang II) formation in the human heart is low compared with

non-ACE pathways. However, little is known about the actual in vivo contribution of these

pathways to Ang II formation in the human heart. To examine angiotensin II formation in the

intact human heart, we administered intracoronary 123I-labeled angiotensin I (Ang I) with

and without intracoronary enalaprilat to orthotopic heart transplant recipients. The fractional

conversion of Ang I to Ang II, calculated after separation of angiotensin peptides by HPLC,

was 0.415 +/- 0.104 (n = 5, mean +/- SD). Enalaprilat reduced fractional conversion by 89%,

to a value of 0.044 +/- 0.053 (n = 4, P = 0.002). In a separate study of explanted hearts, a

newly developed in vitro Ang II-forming assay was used to examine cardiac tissue ACE

activity independent of circulating components. ACE activity in solubilized left ventricular

membrane preparations from failing hearts was 49.6 +/- 5.3 fmol 125I-Ang II formed per

minute per milligram of protein (n = 8, +/- SE), and 35.9 +/- 4.8 fmol/min/mg from nonfailing

human hearts (n = 7, P = 0.08). In the presence of 1 microM enalaprilat, ACE activity was

reduced by 85%, to 7.3 +/- 1.4 fmol/min/mg in the failing group and […]

Research Article

Find the latest version:

(2)

Angiotensin 11 Formation

in the Intact Human Heart

Predominance of the Angiotensin

i-convi

erting

Enzyme Pathway

LawrenceS. Zisman,* William T. Abraham,* Glenn E. Meixell,* Brian N.Vamvakias,t Robert A. Quaife,* Brain D. Lowes, * Robert L.Roden,* StephanieJ. Peacock,* Bertron M.Groves,*Mary V. Raynolds,*Michael R. Bristow,*

and M.Benjamin Perryman*

*Department of Medicine, Division of Cardiology, B-139, University of Colorado Health Sciences Center, Denver, Colorado 80262, and

'Golden

Pharmaceuticals, Golden, Colorado 80401

Abstract

It has been proposed that the contribution of myocardial tissueangiotensin converting enzyme (ACE) to angiotensin II (AngII) formation in the human heart is low compared with non-ACE pathways. However, little is known about the actualin vivo contribution of these pathways to Ang II formation in the human heart. To examine angiotensin II formation in the intact human heart,weadministered intra-coronary

"2I-labeled

angiotensin I(Ang I) with and without intracoronary enalaprilat to orthotopic heart transplant

re-cipients.Thefractional conversion of AngI toAng II, calcu-latedafter separation of angiotensin peptides by HPLC, was 0.415+0.104 (n = 5,

mean±SD).

Enalaprilat reduced frac-tional conversion by89%,to avalue of0.044+0.053 (n =4, P=0.002).In aseparatestudy ofexplanted hearts,anewly developed in vitroAng II-formingassaywasusedto exam-ine cardiac tissue ACE activity independent ofcirculating

components. ACE activity in solubilized left ventricular membrane preparations from failing hearts was 49.6+5.3 fmol

125I-Ang

II formed per minute per milligram of protein (n = 8, ±SE), and 35.9+4.8fmol/min/mgfrom nonfailing human hearts (n = 7,P = 0.08). In the presence of 1 ,uM

enalaprilat, ACEactivity was reducedby 85%, to7.3±+1.4 fmol/min/mgin thefailinggroup andto

4.6±+1.3

fmol/min/

mg in thenonfailing group (P< 0.001).We conclude that the predominant pathway for angiotensin II formation in the human heart is through ACE. (J. Clin. Invest. 1995.

95:1490-1498.)Key words:angiotensinI *

angiotensin-con-Portions of this workhave appeared in abstract form(Zisman, L.S.,E. Bush,C.S. Taft,M.R. Bristow, M.B. Perryman, and M.V. Raynolds. 1994. Increase inAngiotensin ConvertingEnzyme geneexpressionand activityin thefailinghumanventricle. Circulation. 90:1-578 [Abstr.]. andZisman,L.S.,W.T.Abraham,G.E.Meixell,B.N.Vamvakias,B.D. Lowes,R.L.Roden,R.A.Quaife,B.M.Groves,M.R.Bristow,and M.B. Perryman. 1995. Angiotensin IIformation in the Intact Human Heart: Predominance of the Angiotensin Converting Enzyme pathway. Ac-cepted American College of Cardiology 44th Annual Scientific Session [Abstr.]).

Address correspondencetoM.Benjamin Perryman,Ph.D., Division ofCardiology,B139,Universityof Colorado Health Sciences Center, 4200E.Ninth Ave., Denver, CO 80262. Phone: 303-270-3259; FAX: 303-270-3261.

Receivedfor publication 10 January1995 andacceptedin revised forn 18May1995.

verting enzyme inhibitors - cardiomyopathy - peptidyl

di-peptidase A * renin-angiotensin system

Introduction

The importance of therenin-angiotensin system

(RAS)'

in the development and progression of heart failure has been demon-strated in severallarge, randomized, double-blindclinicaltrials ofangiotensinconverting enzyme inhibitors (ACE-Is). These trials have shown a survival benefit from ACE-I therapy in patients with both symptomaticandasymptomatic left ventricu-lardysfunction (1, 2). This survival benefit correlates with a significant improvement in indices of left ventricular cardiac function, including ejection fraction (3). The VHeft II trial demonstrated that the mortality reduction provided by ACE-I therapy was significantly greater than that of thenon-ACE-I vasodilator combination of hydralazine/isosorbide dinitrate (4).Furthermore, in patients with hypertension and left ventric-ular (LV) hypertrophy, ACE-I therapy may promote regression ofLVhypertrophyto agreaterextentthan other forms of antihy-pertensive therapy (5). These clinical studies suggest that an-giotensinconverting enzyme (ACE) plays animportant role in thepathophysiology of myocardial failure and hypertrophy, and that,atthe cardiac tissuelevel, thisresponse may be indepen-dent of peripheral hemodynamics. However, little is known about ACE substrate metabolism in the intact human heart.

The twomajor peptide substrates for ACEareangiotensin I(AngI) andbradykinin. ACEremovesthetwoterminal amino acids from thedecapeptide AngI toyieldtheactive octapeptide hormone angiotensin II (Ang II). Ang II regulates systemic blood pressurebyseveralmechanisms, including receptor-me-diatedvasoconstriction, promotion of sodium resorption inthe proximal tubule of the kidney, and stimulation of aldosterone secretionfrom theadrenal medulla. Bradykininhasvasodilatory andantiproliferative properties (6); ACE, by hydrolyzing bra-dykinin, renders it inactive. Thus, ACE promotes systemic vaso-constrictionbyproducing AngII andconcurrently catabolizing bradykinin.

Recentstudiesinanimalmodels,culturedcells, andhuman cardiac tissue have demonstrated thatcomponents of the RAS are functionally expressed in the myocardium (7, 8). These data suggestthatlocallyproduced AngIImaymodulatecardiac function.Theautocrine/paracrine effectsof intracardiacAngII

1.Abbreviations usedinthis appear: ACE,angiotensinconverting

en-zyme;ACE-I,angiotensinconverting enzymeinhibitor;AngI, angioten-sinI; Cs,coronarysinus;CSBF,coronary sinus bloodflow; IC, intracor-onary;IDC,idiopathicdilatedcardiomyopathy; ISC,ischemic cardiomy-opathy;LV, leftventricular; RAS, renin-angiotensin system.

J.Clin. Invest.

©DTheAmerican Society for ClinicalInvestigation,Inc. 0021-9738/95/09/1490/09 $2.00

(3)

may bedistinct from theeffects of systemically produced Ang II;potential effects of locally produced AngHinclude facilita-tion of sympatheticneurotransmission(9-11 ), positiveor neg-ativeinotropiceffects( 12, 13),and directmitogeniceffects on cardiomyocytes (14, 15).

Currently, there is controversy regardingthemajor pathway for Ang II formation in the human heart. Although ACE has beenshowntomediate 70%ofAngI toAngHconversion in isolated rat hearts, in vitro studies of the human heart have suggested that ACE mediates only 11% ofAng II formation (16, 17). Based on these in vitro studies, it has been proposed that the enzyme human heart chymase is primarily responsible for cardiac AngII formation(18). However,the actual in vivo contribution of ACE and non-ACE mediatedpathways to Ang Imetabolism in thehumanheart has notbeen studied. Accord-ingly,wedevelopedanovelprotocoltoquantifythe conversion ofAng Ito AngH across themyocardial circulation in human subjects andto determine the fraction of conversion mediated by ACE. Thisprotocolincorporates theuseofanintracoronary (IC) infusion of 231I-labeledAng I with a sequential concomi-tantIC infusion of the ACE inhibitor enalaprilat. Comparison of Ang I to Ang H conversion with and without enalaprilat allowsacalculationof the fractional conversion that is mediated in vivo by ACE. Additionally, we developed animproved in vitrotissue ACE assay that permits quantitative measurement of human heart ACE activity independent ofcirculating RAS components and applied this assay, in a set of experiments separate from the in vivoAngHformation protocol,to measure ACEactivity inLVmyocardium fromeightfailing and seven nonfailing explanted humanhearts.

Methods

Reagents. Sodiumiodide(123I),sodiumthiosulfatefree, specific activity 32.3-61.1 mCi/ml at the time ofcalibration,wasobtained fromNordion International,Inc.(Vancouver,B.C., Canada). AngiotensinIfor product synthesiswas obtained from Sigma Chemical Co. (St. Louis, MO). Chloramine T was obtained from J. T. Baker Chemical Co. (Phil-lipsburg, NJ).Potassiumhydroxide, sodiumchloride,sodiumphosphate (monobasic), glacial aceticacid, sodium hydroxide,sodium thiosulfate USP,HPLCgrade methanol,and acetonitrilewereobtained fromFisher Scientific (Pittsburgh, PA). Ang I, Ang II,andAng HI forstandards, 85%orthophosphoric acid, triflouroaceticacid,bovine serum albumin (fractionV, RIAgrade),Triton-X100,Hepes,potassium chloride,zinc sulfate, EDTA, 1,10-phenanthroline, trichloroacetic acid, deoxycholic acid,andsodiumazide were obtained fromSigmaChemicalCo. Enala-prilatwaskindly provided by MerckResearch Laboratories(WestPoint, PA). The renin inhibitor Ro 42-5892 waskindly providedby Hoffman-LaRoche, Inc.(Nutley, NJ).Medium mesh P4 andAG1X8resins were obtained from Bio-Rad Laboratories (Hercules, CA). 25% HSA and sterile pyrogen free water were obtainedfromBaxterHealthcare Corpo-ration(Deerfield,IL).'"I-labeledangiotensinI, "5I-labeledangiotensin H, and '25I-labeled angiotensin IV were obtained from New England Nuclear(Bedford, MA). 3H-Hipglyglywas obtained from Hycor Bio-medical, Inc. (Irvine CA).

Subjects, invivoAng llfornationprotocol. Fiveorthotopic heart transplantrecipientswith normal left ventricular function and coronary anatomywererecruitedfrom the group of patients undergoing routine annualsurveillance cardiac catheterization at the University of Colorado Health Sciences Center. All patients were on a standard regimen of cyclosporin (doserange 100-125 mg b.i.d.), azathioprine (Imuran; dose range 50-150 mgq.d.), and diltiazem (dose range 120-360 mg q.d.). Twopatients were onaspirin 81 mg q.d., and three patients were on furosemide(Lasix;dose range 20 mgq.d.to40 mgb.i.d.).Nopatients

were on ACE-I therapy for at least 12 mo before enrollment. Four patients were male, one was female. Ages ranged from 34 to 59 yr. Lugol's solution was administeredstartingtheday beforethestudyand was given for a total of 3 d to block thyroid uptake of123I. Thisstudy was approvedby the Colorado Multiple Institution Review Board, and written informedconsent was obtained from all subjectsbefore study.

'23Iodination ofAngl. Iodination ofAngI wasperformed usingthe chloramine T method (19) andpurified asdescribed by Admiraal et al.(20) with slight modifications. All reagents wereof USPgradeor equivalent andprepared with sterile,pyrogen-freewater.Glassware was depyrogenized bybaking at 200C for 12 h. Plasticware andcolumns weredepyrogenized with 1% KOH and then autoclaved at 120C for 45 min. 1 mg Ang I was reconstituted with 4 ml 0.25 M NaH2PO4, pH 7.5. 20 1l reconstituted Ang I was added to a 1.5-ml polypropylene microcentrifugetube containing 40 pl 0.25 M NaH2PO4, pH 7.2, 10

Al

KI(1 ng/1Al in 0.25 M NaH2PO4, pH 7.4), and 10 LI 1231. 20

ILI

14 mM chloramineT was added, and thereaction vessel vortexed for20 s. To stopthereaction, 20 l 22 mM sodiumthiosulfate wasadded. lodination wasverified by thin layer chromatography. This reaction was repeated until a total of 10 mCi 123j activitywas used. Thereactions were pooled andloaded via a flowadaptor (Bio-RadLaboratories) onto a2.5 X 10 cmAGlX8column. Elution wascarried outwith 0.1 M acetic acidcontaining 0.2% HSA to removefree 1231I.Labeledproduct wasdiverted to a sterile 30 ml vial and loadedbyaperistaltic pump through a flow adaptor onto a 2.5 x 50 cm P4 column. The P4resin wasprepared in0.05 M aceticacid/0.I MNaCl,pH6.0, forautoclaving and equilibratedovernight with elutionbuffer(0.05 Macetic acid/0.1 MNaCI/0.2%HSA). Product eluted from thecolumnwascollected in four fractions of 15, 30, 30, and 15 ml volume. The detection system consisted of a NaIscintillationdetector,aphotomultiplier tube,asingle channel analyzer, anamplifier,and a quad counter/timer(EG&G Ortec, OakRidge,TN).Calibrationwasperformed usinga137CSsource.The second fractioncontainedpredominantlymono-iodinated Ang I(90%) as verified by HPLC. This fraction was counted in an ion chamber (Capintec Inc., Ramsey, NJ) to determine total activity. The pH was adjusted to 7.4 with 5%NaOH,andthefraction filteredthrough a 0.2-tmAcrodisc(Gelman Sciences, Inc., Ann ArborMI).Before adminis-tration tothepatients,analiquot ofthisproduct was subjected to pyrogen testing (Limulus assay; CapeCod Associates, WoodsHole, MA). In addition, sterility testing was performed (Infinity Laboratories, Inc., Littleton, CO).

Cardiac catheterization. Standard procedures for routine left and rightheart catheterization wereemployed. Coronary sinus (CS) cathe-terization wasperformedvia therightinternaljugular veinusing a 7F Webster thermodilution catheterforblood samplingfromthe CSand formonitoringof coronarysinusbloodflow (CSBF). Catheter position in the left coronary artery wasconfirmed by contrast injection before theinitiationof the IC infusion of any study medication. The CS catheter positionwas confirmed bycontrast injection and blood sampling for percentage ofoxygen saturation. Delivery of the IC medications was controlledbyanautomatic infusionpump (Harvard Apparatus, North Reading, MA). A 5F JL4 catheter was used for infusion of the IC medications to allow simultaneous sampling from the femoral artery sheath and the CS.CSBF wasmeasured using the standard technique ofthermodilution aspreviously described (21). A timeline for the in vivo protocolis presentedinFig. 1.

After routine catheterization was completed, anICinfusion of Ang

Iwasgiven for 5 min (0.01

Ag/min)

to saturatenonspecific binding in theplastictubing and catheters.AnICinfusion of AngIlabeled with theradioisotope 123Ito aspecific activity of 440-1,890

ACi/ug

atthe timeof administration was infused via the Judkins catheter into the left main coronary arteryat1 ml/min, 0.01-0.05

Ag/min.

The infusionwas

given for a period of6 min, during which time blood samplingwas

(4)

ICEnalaprilat Figure 1. Timeline for

IC AngI IC1231-AngI the in vivo AngII

forma-. - tionprotocol.Aftera

5-o 5 8 11 16 min infusion of

unla-Time(minutes)

beled Ang

I,

an

infusion

of '23I-labeledAng I wasadministered into the left main coronary artery for 6min; a concom-itant IC infusion of enalaprilat (0.01 mg/min) was then added for 5 min. Bloodwassampled from the coronarysinusandfemoral artery, and coronary sinus bloodflowwasmeasured at the indicated timepoints.

ofenalaprilat at 0.01 mg/min was administered after 6 min of 123i-labeled Ang I infusion, and blood sampling was performed after 5 min of the combined infusion. CSBF measurements were performed at baselineandduringthe 231I-labeledAng Iandsimultaneous'23I-labeled AngI/enalaprilatinfusion.The 0.01 mg/min doseofIC enalaprilat was chosenbecausehigher doses (i.e., 0.05 mg/min), which were previously reported to have no systemic hemodynamic effects (22), were found in some cases to lowersystemic blood pressure (unpublished observa-tions).

Dosimetryof'231-labeledAng 1.24-h urine collections were obtained in 8-h aliquots toquantify urinary excretion of123Iactivity. Thoracic and abdominalimages wereacquired witha scannerequippedwith a general purpose,parallel-hole collimator(Body scan,Siemens Medical, Inc.,Iselin, NJ). A 10%window centered on the159KeVpeakfor 1231 wasused,and theacquisition timewas10min.Scansandthyroid uptake studies wereperformedwithin 6 h ofcompletionof thestudy protocol; thyroid uptake studies were repeated 24-48 h later to quantify 123I uptake. The rate of urinary excretion of radioactivity was fit to an

exponential decayfunctiontoyieldanexponentialelimination constant (INPLOT; GraphPad Software,Inc., SanDiego, CA). This constant, thetotal administeredactivity, the fraction ofactivityeliminated in the urine, and the measured fraction ofactivity taken upby the thyroid

wereusedtocalculatetime-activitycurves(23).Theresidence times calculatedforthe total body, heart, andthyroidwereusedtocalculate thedosetothese compartments(MIRDOSE 2;OakRidgeInstitutefor Science, OakRidge, TN).

Sample analysis. Thecombined HPLC/RIA method of Danseret

al. (24)with slight modifications was used to separate and quantify unlabeledAng I, Ang II,and Ang III,aswellasiodinatedAngIand Ang II. Briefly, blood samples were immediately placed in inhibitor solution containing0.01 mM renin inhibitor(Ro 42-5892), 6.25 mM EDTA, 1.25 mM 1,10-phenanthroline, and 0.02 mMenalaprilat(final concentration in blood). Plasmawas extracted on C18 Sep-pak

car-tridges (Waters Chromatography, Bedford, MA)eluted withmethanol, lyophilized, and reconstituted in mobilephase. HPLC wasperformed withaC18Nucleosil column(AlltechAssociates, Inc., Deerfield, IL). MobilephaseA consisted of 0.085%orthophosphoric acid/0.02%

so-diumazide; mobilephase B consisted of methanol. Thegradientwas

isocratic withA65%-B35% from 0-9 min followedby agradientof mobile phaseB to55%over9 minat45°C.Fractions werecollected, neutralized to pH7.4,and counted inagammacounter.Aftercounting, thesampleswerelyophilized,reconstituted in RIAbuffer,andfractions assayed for AngIand Ang II. Ang IIwasmeasuredusingthe Nichols Institute RIA Kit (San Juan Capistrano, CA); Ang I was measured usingthe RIAfrom Peninsula Laboratories (Belmont, CA).Retention times of Ang I, AngII, and Ang III were determined using known standards andwere15.5, 7.5,and5.8 min,respectively.Retention times of'23I-labeledAngI, '23I-labeledAngII,and'23I-labeledAng IVwere

determinedusing peptideslabeledwith1251 andwere19, 12.5,and 15.5 min. Theactivityof thesamples resulting from the1231Iwasallowedto

decaytobackgroundlevels beforecompletionof the RIA.

Mathematical modeling. The venous equilibrium model used by

Danser et al. (24) to study angiotensin metabolism across different vascular beds inpigstreatsthe intravascular space and the tissue

com-partment as one well-stirred chamber. The mathematical arguments of this model have been previously described (24, 25). The fractional conversion of arterially delivered Ang I is defined as

f,= fractional conversion ofAngItoAngII (1)

= E,x {k2t/(klt + k2t)},

whereElis the extraction ratio of Ang I, k2 is the first-order rate constant for conversion of Ang I to Ang II, andk,is the first-order rate constant for degradation of Ang I and Ang II to other metabolites including Ang III and Ang IV.

If the venous equilibrium model is correct, then the fractional con-version of 1231I-labeledAng I to 1231I-labeledAng H actually measured in samples taken from the CS should be the same as that predicted by Eq. 1. The fractional conversion of 123I-labeled Ang I to '23I-labeled Ang IImeasured in the CS is given by the following formula:

fm = (123IAIk(cpm))/(1231AIIcs(cpm) +

1231AIcs(cpm)),

(2)

where123IMIk (cpm)=counts perminute of'23I-labeledAng II in the CS, and

1231AIL

(cpm) = countsper minute of 1231-labeledAng I in theCS.

Under the condition fv > fin, there is net uptake of Ang II. In a simplified two-compartment model, this condition is accounted for by an effective uptake constant for Ang H:

Keff= {f4 fm

}/t{fm(l

-fv)} (3) This relationisderived fromtheassumptionthat ifoneadded backthe

countsderived from '231-labeled AngIIthataretaken upbythetissue compartmentduringpassagethroughthemyocardialvascularbed,then f,wouldequal

fm* = (

1231AIIcs(cpm)

+ Keff{

123IAIk,(cpm)})/(Total,,(cpm)

+ KffI123IAIIks(cpm)}). (4) Because this protocol requiresdetermination of arteriovenous

differ-ences acrossthemyocardial vascular bed,itisnecessarytoknow the arterial input concentration ofunlabeled Ang I andAng H distal to

the infusion site. The measuredinputconcentration ofAngIinfusion concentrationisgiven bythefollowing formula:

(AI)Im

=

J (AI)cath

X

Qcath

}

/Qcs

+ (AI)fa

where(AI)tm = AngIconcentration in the left main coronary artery distal to Ang I infusion site; (AI)ch = Ang I concentration in the infusion catheter;(AI)fa = AngIconcentration in the femoral artery; Qcath=infusion catheter flow rate; andQc, =coronary sinus blood flow. Statistical analysis was performed with Statview 512+ (BrainPower, Inc., Calabasas, CA). Unlessotherwisenoted, valuesareexpressedas

the mean±SE. For within-group comparisons, statistical significance

wastestedusingthepairedttest; forbetween-group comparisons, the unpairedt test wasused.

Invitrocardiac tissue ACE assay: characteristicsofassayedhearts. 16 human hearts obtained atthe time ofexplantationwereevaluated. Thetime fromexplanationtomembranepreparationwasless than30 min, as routinely practiced in ourlaboratory. Eightof the explanted hearts werefrom patients withend-stage heart failure resulting from idiopathicdilatedcardiomyopathy (IDC),andseven werefrom donors withnormally functioningheartsthat couldnotbeplacedfor transplanta-tion. One heartassayedatmultiplesubstrate concentrationswasfrom

apatientwithend-stageheart failuresecondarytoischemic cardiomyop-athy (ISC).Themeanage of theIDChearts and thenonfailinghearts

was43±5.5 and 48±5.7 years,respectively (P= NS).ThemeanLV

ejectionfraction fortheIDCheartswas0.18±2.5,thepulmonary capil-lary wedgepressurewas21.7±2.4mmHg,and themeancardiac index

(5)

capil-lary wedgepressure greater than 15 mmHg or cardiac index less than 2.0liter/min/m2.Forthenonfailing hearts,inallcases, the LVejection fraction was greater than 50%. More detailed hemodynamic data for the donor hearts was unavailable; however, the 61-adrenergicreceptor density in thefailing hearts was 34.7±4.1 fmol/mg, whereas the /3k-adrenergic receptor density in thenonfailinghearts was63.5±5.4 fmol/ mg (P=0.002). The use of61-adrenergic receptor density as a quantita-tive criterion for the classification of humanmyocardialtissueasfailing versus nonfailing has been previously established by our laboratory (26). Such quantitation supplements standard hemodynamic variables for whichonlyqualitative information is typically availablefor donor hearts.Of the eight IDC heartsassayed,five had beenexposedto ACE-Isduring thepreceding 6 mo and two had not received ACE-I therapy for aperiod of at least 3 mo before transplantation. The one ISC heart examined had a left ventricularejectionfraction of0.11 and ahistory ofprior long-term ACE-I exposure. All of theIDC patients wereon

regimensof digoxin, Lasix, andcoumadinfor aperiodof time exceeding 3 mobeforetransplantation. None were onaspirin. Therewasnoprior history of ACE-I exposure in the nonfailing group, andnopatients in this group weretaking digoxin, diuretics, oraspirin.

TissueACE assay:sample analysis.Ahigh-yieldcrudemembrane preparation of ventricularmyocardiumwasprepared as previously de-scribed (27). Themembranepreparationwassolubilized at40Cin an

equalvolume ofresuspensionbuffer(0.01MHepes, 0.3 MKCl,0.6% Triton-XIOO,pH 7.5) for 4 h and centrifuged at 25,000 g for 1 h. The supernatant was dialyzedextensively against 0.01 M Hepes, pH 8.1, 0.3 MKCl,0.02%Triton-X100, 100pmZnSO4to removethe EGTA in themembrane preparation suspension buffer and ACE-I if present. This combination of solubilization and dialysis was derived from the method of Bull et al. (28). 100IL sample was diluted in 400 tlACE

buffer (0.05MHepes, pH8.0,0.1%NaN3).50 pl ofthisdilutionwas

incubated with 50 fmol of the ACE substrate lnI-labeledAng I for 20 min at 370C. The 125I-labeled Ang I was reconstituted in 2 ml ACE buffer 1 (0.05 M Hepes, 0.1 MNaCl,0.01%Triton-X1OO,

50jiM

ZnSO4, pH8.0) and diluted to achieve aconcentrationof 1fmol/jlin0.05 M Hepes, 0.1% NaN3, pH 8.0. Whenperformingassays togenerate data for Lineweaver-Burk plots, four different concentrations of substrate were used: 25 fmol, 50 fmol, 75 fmol, and 150 fmol per reaction. Appropriate dilutions of concentrated l"I-labeledAng Iweremadeto

achieve this range of values. To inhibit ACE in these preparations, parallelsamplesweremade with theadditionof 1

j1M

enalaprilat. All sampleswereassayedinduplicate. The reactionwas stoppedwith 300 kd0.2% trifluoroacetic acid. 133

01

acetonitrilewasadded and 200

ILI

of thesamplewaschromatographedon anHPLC systemusingaVydac 201HS54column(Hesperia, CA). Mobile phaseAwas0.2% trifluoro-acetic acid inH20; mobilephaseB wasacetonitrile. Thegradientwas isocratic from 0 to 3 min with mobilephase B25%;from 3 to 12 min

agradientincreasedmobile phase Bto40%. The flow rate was 1.2 ml/ min, and the temperature25°C.Under theseconditions, l25I-labeledAng

Ieluted at9.5 min, whereasl"I-labeledAng Heluted at 7 min. Elution timesweredeterminedusing known standards of "nI-labeledAng I and 125I-labeled Ang II. 0.5-ml fractions were collected and counted in a gammacounter.ACEactivitywascalculatedaccordingtothefollowing formula:

Units(fmol/min/mg) = Totalactivity* (AII)/{(AI)*s.a.* (sam-ple volume)(time) (mg/ml)IwhereAll = '"I-labeledAng II activity,

Al = 125I-labeled Ang Iactivity, and s.a. = specific activity of 1I-labeled AngI atthe time ofsample analysis, which included correction fordecayandcounting efficiency;mg/ml indicates theprotein concen-tration.

When the ACE specific substrate 3H-Hipglygly was used in this assay, ACE activity was onthe order of 5-10 nmol/min/mg versus 20-30nmol/min/mg in nonfailing versus failing hearts and was found

to parallel ACE activity determined using '25-Ilabeled Ang I as the substrate.

Protein assay. BSA was reconstituted in dialysis buffer to a concen-trationof 1mg/ml.Totalproteinfrom 0to0.125 mg wasadded toa totalvolume of 1 ml H20 in 0.025-mg increments. 50 and 75

plI

of

dialyzed sampleswerebroughtto atotalvolume of 1 mlwithH20.To the BSA standards and the samples, 100 ILI0.15% deoxycholic acid wasadded followed by 100IL 72%TCA.Aftervortexing, thestandards

andsampleswerecentrifugedat2000g for 15 minutes. The supernatant wasdiscarded, and a Lowry assay wasperformedtoquantifythe precipi-tatedproteins (29). Absorbanceat750 nm wasmeasured with a mi-croplatereader(Thermomax; MolecularDevices, Sunnyvale, CA).

Lineweaver-Burkplots.ACEactivityinsolubilizedmembrane prep-arations from threefailinghearts and threenonfailingheartswas

mea-suredatfour different substrate concentrations. Thereciprocalof

sub-strateconcentrationwasplotted againstthereciprocalof ACEactivity. Linearregressionwasperformedtodetermine the best fit of the data for each heartassayed (INPLOT; GraphPad Software, Inc., San Diego, CA). The Km andV., werecalculated for each heartasthenegative reciprocalof thex-interceptand the inverse of they-intercept, respec-tively.

Results

Hemodynamics ofpatients enrolledinthein vivoAng I1forma-tionprotocol.Thebaseline heartrateof the fivepatientsinthe in vivostudywas74.8±6.9 beats per minute. Themeansystolic blood pressurewas 131.2±13.4mmHg, and the mean diastolic blood pressure was 83.2±9.4 mmHg. Themean LV end dia-stolic pressure for the five patients was 9.2±3.4mmHg. The mean pulmonary capillary wedgepressure, cardiac index, and LV ejection fraction were 7.4+2.3 mmHg, 2.92+0.38 liters/

min/m2,

and0.69±0.08, respectively.ThemeanCSBFat base-linebefore the infusion of any study medicationwas93.1±14.0 ml/min. The mean CSBF after the 6-min IC infusion of1231_

labeled AngIwas 106±8.0ml/min, and themeanCSBF after the5-min combined IC

231I-labeled

AngI/enalaprtinfusionwas

106.1±7.43

ml/min. In no case was a decrease in CSBF ob-served during the IC

1231I-labeled

Ang I infusion, nor did a

significant changein CSBFoccurduringthe ICenalaprilat infu-sion. Thus,appropriatetothedesign of thisstudy,thedoses of infused medications werein thesubpharmacologic range.

Dosimetryof

123"-labeled

AngI. Thetotalradioactivity ad-ministered toeach patientwas 274+41.7

sCi.

The percentage ofurinary excretion of total administered activity at24 h was

66.1±5.9%, and the fraction of total activity taken up by the

thyroid was 0.037±0.008. The dose to the thyroid was 0.373±0.071 rad, totheheart, 0.987±0.145 rad, and the total body dose was

0.022±0.003

rad. The dose tothe heart repre-sents a maximum estimate because it was assumed that the residence time in the heartwasthesame astotalbody residence time. Fig. 2 demonstrates the distribution of

1231-labeled

Ang Infused activity in images obtained2 h after the study in one

patientwho didnotreceiveintracoronary enalaprilat.A signifi-cant proportion of activity is retained in the heart. A small amountofactivitycanbeappreciatedin theregionof the thyroid aswellasthekidneys.

ACE-mediated AngI conversion in theintacthuman heart. A measure of the fractional conversion of Ang I to Ang H acrossthemyocardialvascular bed wasobtainedby takingthe ratio of activity of the labeled Ang H to the sum ofactivity

(6)

AP

+=thyroid

=heart

-4= kidney

60

E { 50

,E

040

0 30

E 20

-0

c 10

'-T

*Baseline oEnalaprilat

-p<0.10vs. NF

**p<0.001vs.baseline

I*

i1-Failing nwS

PA

Figure 2. Biodistributionof'23I-labeledAng I 2 hafterthe in vivo study.Asignificant portion of activity is retainedin themyocardium subtendedby the left coronary artery system. A small amount of activity canbeappreciatedin theregion ofthethyroid and thekidneys. The toppanelrepresents theanteroposterior(AP)view, and the bottom panel theposteroanterior(PA) view.

During the steady-state IC infusion of

1231-labeled

Ang I, fractional conversion of AngI toAngII was0.415+0.104, and

0.430±0. 158(mean±SD) asdetermined, respectively, bythe

activityratio and RIA methods(n=5).Intracoronary

enalapri-lat(0.01 mg/min for 5 min) reduced fractional conversion by

89% to 0.044±0.053 (activity ratio method) or 0.024±0.018 (RIA method) (n =

4,

mean±SD,P= 0.002)(Fig. 3).

After 3 min of the'23I-labeledAngIinfusion,the calculated effectiveuptakeconstantfor

1231-labeled

AngII was0.79±0.38, whereas the effectiveuptakeconstant for

'231I-labeled

AngII at 6 min 0.232±0.361. The first-orderrateconstantfordegradation

of labeled and unlabeledAngIandAngII toothermetabolites

0.6 Figure 3.The fractional conversionof'23I-labeled = 0.5

AngIto

'23I-labeled

Ang

L CD

0.4 IIacrossthe intact

hu-man heartbefore and

8-

0.3 duringconcomitant

ad-~ _ ministrationof IC enala-t . 0.2 prilat. During the steady-"0.1 p.002 stateIC infusionof 1231

labeledAngI,fractional

0 conversion of'23I-labeled

IC Ang +Enalapnilat Ang I to

123I-labeled

Ang

IH

was

0.415±0.104

(mean±SD,n= 5).ICenalaprilatreduced fractional conversionto

0.044±0.053 (n= 4,mean±SD,P=0.002),indicatingthat89% of

'23I-labeledAngIconversionto '23I-labeledAngHwasmediatedby ACE.

Nonfailing

n=7

Tissue ACEActivity

Figure 4. ACE activity in solubilized LV membranes prepared from explanted failing(n=8, all IDC)andnonfailing(n=7) human hearts. Sampleswereincubated with'25I-labeledAng I in the presence and absence of the ACE-I enalaprilat, and the formation of'"I-labeledAng IIwas measured. 85% ofl2I-labeledAng IIformation in vitro was mediatedby ACE, and there was a trend toward higher ACE activity in thefailingversus thenonfailing hearts.

was 0.137±0.034

min',

and the first-order rate constant for conversion of labeled and unlabeled Ang I to Ang II was 0.163±0.048

min'

at 6 min. These data indicate mean net uptake of AngHandarelative balance between AngII forma-tion anddegradation at this 6-min time point. The extraction ratio of combined labeled and unlabeled Ang I was 0.794+0.048, and the extraction ratio of combined labeled and unlabeled AngHwas0.522±0.09.

ACE enzymatic activityinhumanmyocardium. ACE

activ-ityinsolubilized LV membrane preparations from failing hearts was 49.6±5.3 fmol '251-labeledAng IIformed/per minute per milligram of protein (n =8, all IDC), and 35.9±4.8 fmol/min/

mig in LVpreparations from nonfailing human hearts (n = 7, P = 0.08 failing vs. nonfailing). In the presence of 1 ,tM

enalaprilat, ACE activity in the failing group was reduced to 7.3±1.4fmol/min/mg and in the nonfailing group to 4.6±1.3 fmol/min/mg,areduction of 85% (P<0.001)(Fig.4). Tissue ACE activity in thetwo IDC hearts that had no prior history of ACE-I exposureranged from 41.3to62.1fmol/min/mg with a meanof51.7±6.9fmol/min/mg.

Lineweaver-Burk plots wereused to calculate the Km and

VM,,M

for ACE inasmallset offailing (n = 3, 2 IDC, 1 ISC) and nonfailing (n = 3) human hearts. For the failing hearts,

the Kmwas 223±57 f M and for thenonfailinghearts, the Km was 255±65 fM (P = NS). The

V,,,

for the failing hearts was391±65 fmol/min/mg, whereastheV,,, for thenonfailing heartswas 298±66fmol/min/mg (P = NS) (Fig. 5).

Discussion

Themajorfindingsof thisstudymay be summarizedasfollows:

(1)conversion ofAngI toAngH canbemeasured inthe intact humanheart; and (2)In agroup oforthotopichearttransplant

(7)

0.05- Figure 5. Lineweaver-Burk plots of human heart tissue ACE activity. ACE 0.04- activity was assayed at

fourdifferent concentra-tions of'25I-labeledAng I: § 0.03- 25, 50, 100, and 150 fmol/ E incubation. Representative

0 plots for onefailingheart

0.02- (o) and onenonfailing heart(-) are shown (mean±SD). The failing 0.01- heart was from a patient withIDC and no prior his-toryof ACE-Iexposure. 0- , | | , | Linear regressionwasused -0.01 0 0.01 0.02 0.03 0.04 0.05 to determine the

x-inter-1/S(fmol) cept(-1/Km) and the

y-intercept(1/V,,,,,)for each heartassayed.The lowery-interceptfor thefailingheart indicates that theV,,,, of ACEforthisheartwashigherthan that for thenonfailing heart. For the failing hearts(2 IDC, 1ISC),theK. was223±57fM; for thenonfailinghearts(n= 3),theKm was255±65fM (P=NS). TheVm,,for the failinghearts,391±65 fmol/min/mg,was30%higher than theV,,,, for thenonfailing hearts,whichwas298±66fmol/min/ mg(P= NS).

The primary purpose of this study was to measure ACE-andnon-ACE-mediated Ang H formation in the intact human heart. Inpreviously reported work in isolatedrathearts, ACE mediated 70% of the conversion of Ang ItoAng II (16). In pigs, in vivo ACE-mediated conversion of Ang I to Ang H across the heartwasapproximately40%(24). However, based onthe results ofanin vitro tissue assay, itwasshown that ACE mediated only 11% of Ang II formation in the human heart, and that the balance ofAng I-to-Ang H conversionwaspotentially

mediatedby human heart chymase (17, 30-32).

Studies of

'251-labeled

Ang I metabolism have been per-formedacross therenal, hepatic, and skeletal muscle vascular beds in humans (20, 33) as well as in pigs (24, 25). Prior studies of the acuteeffects of IC ACE-I administration in the intact human heart have also been undertaken and have demon-strated aneffect onindices of cardiac function (22, 34). We have adapted these previously described methods specifically

tostudyAngIIformationacrossthe humanmyocardial circula-tion. Theprotocolthatwehavedeveloped incorporatestheuse ofan IC infusion of

123I-labeled

AngI with asequential, con-comitant IC infusion of the ACE-Ienalaprilat.Thisprotocol is

uniquein three ways:(1)becauseofconcernsregarding dosim-etry, wechose touse

123I-labeled

Ang I rather that

"2I-labeled

Ang I; (2)to avoidpotential systemic andpulmonary effects, we administered the labeled Ang I via the IC rather than the intravenousroute(35); and (3) by usingadoseof ICenalaprilat that was 10-50-fold lowerthan thatused in otherstudies, we avoided potential confounding pharmacologic effects of this medication.

Usingtwodifferent methodsof sample analysis,theactivity ratiomethodandthecombined HPLC/RIAmethod, we calcu-lated the fractional conversion ofboth labeled and unlabeled Ang I to Ang H across the heart. The combined HPLC/RIA

method incorporatesthefractional conversion term definedby the venousequilibrium model. Although both methods of analy-sisdemonstrated thatAng H formationwasprimarilymediated

by ACE, a simplified two-compartment model revealed evi-dence fornet uptakeof '23I-labeledAng II.That the effective

uptake constant for

'23-labeled

AngH was higher after 3 min compared with 6 min of the

123I-labeled

AngIinfusion indicates an earlydistribution phase into the extravascular/tissue com-partment before equilibrium. We chose to apply the venous

equilibrium model to study Ang H formation in the heart be-causeit had beenappliedwith apparentsuccess to thestudyof Ang I metabolism in other organs in human subjects (20).

However, it isprobablynot anideal modeltodescribe the tissue distribution ofangiotensin peptidesin the human heart.Indeed,

Dell'Italiaetal. (36) reportedthat tissueAng

HI

levels indog

hearts may behigher than those measured inplasma,and Danser et al.(37) have presenteddata that suggest that tissueAngII levels inpigheartsmaybehigherthan would bepredicted by

the venousequilibriummodel.

Because ourin vivoprotocolincludes contributions of both membrane-bound andcirculating ACE, wereexamined the im-portance of tissue ACE activity ex vivo. To develop an

im-proved in vitro tissue ACE assay,we useddialysisto remove

ACE-I, when present, from a solubilized human heart mem-branepreparation before incubation with substrate. This novel tissue ACE assay revealed that invitro,85% of

'25I-labeled

Ang

IIformationwasmediatedbyACE and therefore corroborated our in vivo data. In addition, there was atrend toward

higher

ACEactivity in thefailinghearts.

To explore further the regulation of cardiac tissue ACE

activity in the failing heart, we determined the Km and

V,,,.,

for ACE from Lineweaver-Burk plots. Measurementof these variablescould reveal differences betweennon-failingand

fail-inghearts that would be missed if the assaywere

performed

at

only onesubstrate concentration. Suchdifferences, if present,

might provide important clues about the

regulation

of tissue ACE in heart failure. For example, a lower K. in a

failing

heart would suggest that the affinityof ACE for substrate was increasedcomparedwith anonfailing heart.In contrast, a

sig-nificant increase in the

Vat,,.,

of ACE in a failing heart would suggest either an increased ACE density in the membrane or anincreasedcatalytic rateper molecule of enzyme. However, althoughthe V,, for ACE in the

failing

humanheartswas30%

higherthan that in the nonfailinghumanhearts studiedovera range of substrateconcentrations,this difference didnotachieve statistical

significance.

Thus, additional

experiments

are

re-quired to determine if the secondary activation of the RAS

precipitated byheartfailure affects these variables in the human heart.

Bothourin vivo and in vitro data indicate that ACE is the

major Ang

HI-forming

enzyme in the human heart. Several factors may contribute to the marked difference between our data and those Urata et al. (17). First, the on-off kinetics of

captopril,

the ACE-Iused

by

Urataetal. (17) toinhibit ACE

activity,may differ fromthoseofenalaprilat; duringthe incuba-tionofthe membrane

preparation

withsubstrate,

captopril

may not be an effective in vitro ACE-I. The need to address this

possibility has been raised by Kinoshita et al.

(38). Second,

the presence of nonsolubilized membrane may preventaccess ofACEtosubstrateorinsomeother wayinhibit the

enzyme's

invitroactivity.Thereisaprecedentfortheuseofasolubilized membranepreparationto measurecardiac tissue ACE

activity

(8)

have been optimized for ACE. Fourth, the membrane prepara-tions used by this group were made from previously frozen tissue, whereas our preparations were made from freshly ex-planted hearts.

The crude membrane preparations used in our assay are enriched two-tothreefoldin

fi-adrenergic

receptor binding rela-tive to a washedpellet homogenate, indicating that this prepara-tion is an enriched sarcolemmal fraction (26). However, the percentage of endothelial plasmalemma in this preparation is notknown.Preparation of sheep cardiac sarcolemma according to the method of Jones et al. (40), which includes a contractile protein precipitation step as doesourown, has been shown to enrich ACE activity compared with that of either adisrupted tissueorwashed homogenatepreparation. However, ACE activ-ity determined from a sarcolemmal preparation of myocytes previously separated from endothelial cells on a sucrose gradi-ent was loweredby a factor of 15 compared with whole heart (41). Based on these data, it was argued that the observed enrichment in ACE activity inthe crude sarcolemmal prepara-tionwassecondarytothe presenceof endothelial plasmalemma. Whether or not the ACE activity we observed resulted from ACE solubilized from endothelial cell membranes and/or the sarcolemma is notknown.

We found a trend toward increased cardiac tissue ACE activ-ity in the failing IDC hearts compared with nonfailing hearts assayed in vitro. These data are consistent with both animal and humanstudies that have demonstratedanincrease in ACE gene expression infailing hearts ( 16, 42-44).Infailing human heart, ACEmRNAabundance was foundto be threefoldhigher than in nonfailing heart (43). Chymase mRNA abundance in the same study was not different in failing versus nonfailing left ventricles, and, inaseparate study, chymaseactivities in tissue extracts from nonfailing normal hearts orfailing hearts (IDC andISC)were notdifferent (32).

Limitations.In ourstudy, the majority of patients with heart failure in whom in vitro cardiac ACEactivity was measured had at some time received ACE-I therapy. It is possible that theincreaseweobserved incardiac ACEactivity in the failing heart was a secondary effect of this treatment rather than the result of regulatory mechanisms associated with myocardial failure or hypertrophy. The potential mechanism underlying

such aneffect could be induction of ACE expression through an as yet uncharacterized feedback loop triggered by chronic ACEinhibition. When ACE-Isareremoved,asinourassayby dialysis, the functional effect of such increased expression

would be revealed. Evidence for such a feedback mechanism has been found in lung ACE ofrats treated with the ACE-I

quinapril, and King and Oparil have clearly shown that ACE inhibitionupregulates ACEmRNAandactivityin both endothe-lial and neuronal cell cultures (45-47). However, Studer et al. (43) found nosignificant difference in cardiac ACE gene expression in patients treated with ACE-Is and patients not treatedwith ACE-Is. Inaddition, Wollertetal. (48) have dem-onstratedthat, while LV ACEactivity and ACE mRNa abun-dancewereincreased inratswith myocardial infarctions, long-term,high-doseACE-Itherapydidnotalter cardiac tissue ACE

activity.Furthermore,therelatively high cardiac ACE activities intheIDC heartsassayed inourstudy that hadnoprior history of ACE-I exposure suggest a primary role for chronic heart failureintheupregulation of cardiac tissue ACE. Nevertheless,

given theevidence for ACEupregulation triggered by ACE-I treatment in cell culture, further studies of cardiac tissue not

previously exposedtoACE-Is areessential toisolate theeffect of heart failure.

Additional drug interactions may have affected both our measurements of in vivo Ang H formation and our in vitro tissue Ang H forming assay. Regarding ourin vivo protocol, thepatients for whom initial experience has been obtainedwere allorthotopic heart transplant recipients on astandardregimen ofcyclosporin. Cyclosporin has been showntoincreaseanAng II-mediated rise in intracellular calcium in rat vascular smooth muscleand mesangial cells (49, 50). Furthermore, cyclosporin has been showntoinduceatwofold increase in gene expression of the angiotensin ATlA receptor in aortic endothelial cells (51). Thus, because cyclosporin affects the signal transduction pathway through which AngIIeffectsare atleastin part medi-ated,we cannotexcludethepossibilitythatthis immunosuppres-sive agent altersAngImetabolism. However, all of the patients studied in the in vivoprotocolwerealsotakingcalcium channel blockers and werenormotensive; it could therefore be argued thatpotential cyclosporin effectsonthe RASmediated through changes in intracellular calciumwerealso interrupted.

Two ofthe five patients in whom in vivo cardiac Ang II formation was studied were on low-dose aspirin. Aspirin, a prostaglandin synthetase inhibitor, has been showntoattenuate angiotensin-mediated contraction of aortic muscle and Ang II sensitivity inasubsetof pregnant patients (52, 53). Conversely, pretreatment withaspirin has been showntoreduce the vasodila-tory effect of intraarterial enalaprilat in the forearm vascular bed(54). However,aspirinwhen combined withACE-Itherapy had no discernible effect on hemodynamic status in patients with heart failure(55). Although it isatenablehypothesis that

aspirinalters Ang II formation in the vascular beds of diverse organs, there was no observed difference in the relationship

between ACEand non-ACE mediated AngIIformation in the intact hearts of thepatientswho weretaking aspirin compared with those who were not (data not shown). Furthermore it should bekeptinmind that ourstudy of in vivo cardiac Ang IIformationwas notsimplyanexamination of AngI metabo-lismin themyocardial vascular bed. Conversion ofI toAngII was examinedfor the entiremyocardial tissue compartmentin

equilibrium with themicrovascular bedsubtendingthe heart. Regarding our in vitro assay, all of the IDCpatients from whom cardiac tissue was obtained were on chronic diuretic therapy with furosemide. This diuretic has been showntoresult in an increase in systemic Ang II levels (56). In addition,

sodium balance has been showntoalter the hemodynamic re-sponse to the Ang H receptor antagonist saralasin (57). It is

possiblethat altered volumestatusand sodium balanceresulting

from chronic diuretictherapymay contributetotheupregulation

of ACE-mediated Ang II formation observed in the failing heartsstudied.

Conclusion. Thepresentstudy represents the first report of conversion of Ang I toAngII in the intacthuman heart. This conversionmaybe mediatedbybothcirculating and

"tissue"-specificormembrane-boundACE. Therelativecontribution of

circulatingversustissue-bound ACEtoAngIIformationacross themyocardialvascular bedcannotbedetermined fromourin vivoprotocol. However, ithas beenshownpreviouslythat the half-life of

251I-labeled

AngIinplasmaranges from 1.7 to2.6 minutes (33). This half-life is an order ofmagnitude greater than the transit time across the human myocardial circulation (58) and therefore suggeststhatmostof the conversion ofAng

(9)

ACE. Furthermore, theinvitrotissue ACE assay performedin thepresent study corroboratesour findings in the intact human heart. We therefore conclude thatthepredominant pathway for Ang II formation in the human heart is through ACE. These resultsreinforce the clinicalimportance of ACE inhibitor ther-apy for thetreatment of heartfailure and provide support for thehypothesis thatcardiac-specific ACE inhibitors will provide additionaltherapeutic benefit.

Acknowledgments

We are grateful for the invaluableassistance of the entire Cardiac Cathe-terizationLaboratory nursing staff atUniversity Hospital: Cathy Clark, R.N., ClareConrardy, R.N., KevinMcShane, R.N., Karen Moore, R.N., AnneOliva,R.N., RhondaOlmsted,R.N., LanaOverturf,R.C.P.T.,and Kim Trace, R.N., aswell as NinaLeitman, C.N.M.T., and Michael R. Whitney, C.N.M.T., in the department of Nuclear Medicine. Tremen-dousgratitude is expressed to Frank Stewart and JillJones, who assisted inthe preparation of the manuscript.

L.Zisman was supported by an American College of Cardiology/ MerckFellowship. W. T. Abraham wassupported byaClinical Associa-tion Physician Award from the National Institutes of Health General Clinical Research Centers program.

References

1.Pfeffer,M.A., E.Braunwald, and the SOLVDInvestigators. 1992. Effect ofcaptoprilonmortality andmorbidity inpatientswithleft ventriculardysfunction

aftermyocardialinfarction. N. Engl. J. Med. 327:669-677.

2.SOLVDInvestigators. 1991. Effect ofenalaprilonsurvival inpatientswith reduced left ventricularejectionfractions andcongestiveheartfailure. N. Engl. J. Med.325:293-302.

3. Konstam, M. A., M. F. Rousseau, M. W. Kronenberg, J. E.Udelson,J.

Melin,D.Stewart,N.Dolan, T. R.Edens,S.Ahn, D. Kinanetal.1992. Effects of theangiotensin converting enzyme inhibitorenalaprilonthelong-term progres-sion of left ventriculardysfunction inpatients with heart failure. Circulation. 86:431-438.

4.Cohn,J. N.,G.Johnson,S.Ziesche,F.Cobb,G.Francis,F.Tristani,R.

Smith,B.Dunkman, H.Loeb,M.Wongetal. 1991. Acomparisonofenalapril

withhydralazine-isosorbide dinitrate in thetreatmentof chroniccongestive heart failure. N. Engl. J. Med. 325:303-310.

5.Dahlof,B., K. Pennert, and L. Hansson. 1992. Reversal of left ventricular hypertrophy in hypertensive patients: ametaanalysis of 109treatmentstudies. Am. J.Hypertens. 5:95-110.

6. Stewart, J. 1989. Roles for kinins in inflammation. In Handbook of Inflam-mation, Vol. 6: Mediators of theInflammatoryProcess. P.Henson,and R.Murphy,

editors. ElsevierSciencePublishing Co., New York. 189-217.

7.Dzau, V. J., and A. Hirsch. 1990.Emergingrole of tissuerenin-angiotensin

systems incongestiveheartfailure. Eur. Heart J. 11:65-71.

8. Dzau, V. J. 1988. Circulatingversus local renin-angiotensin system in cardiovascular homeostasis.Circulation. 77:4-13.

9.Hilgers, K.,R.Veelken,G.Rupprecht,P.Reeh,F.Luft,and J. Mann. 1993.

AngiotensinII facilitates sympathetictransmission inrathind limb circulation. Hypertension. 21:322-328.

10.Gilbert,E.M.,A.Sandoval,P.Larrabee,D. G.Renlund,J. B.O'Connell,

and M. R. Bristow. 1993.Lisinoprillowers cardiacadrenergicdrive andincreases

,/-receptordensityin thefailinghuman heart.Circulation. 88:472-480.

1. Abraham,W.T., B. D.Lowes,C. P. Rose, P. Larrabee, and M. R. Bristow. 1994.AngiotensinIIselectivelyincreases cardiacadrenergic activityinpatients

with heartfailure.J. Am. Coll.Cardiol. 23:215A.

12. Capasso, J. M., P.Li,X.Zhang, L. G. Meggs, andP. Anversa. 1993. Alterations in Ang Hresponsiveness in left andright myocardium after infarction-induced heartfailure inrats.Am.J.Physiol.264:H2056-H2067.

13.Li, P., E. H. Sonnenblick, P. Anversa, and J. M.Capasso. 1994.

Length-dependentmodulationof Ang IIinotropisminratmyocardium:effects of myocar-dial infarction. Am. J. Physiol.266:H779-H786.

14.Dzau, V., and R. Re. 1994. Tissueangiotensinsystem in cardiovascular medicine:aparadigm shift? Circulation. 89:493-498.

15. Sadoshima J.-I., Y. Xu, H.Slayter,and S. Izumo. 1993. Autocrine release ofangiotensin II mediates stretch-induced hypertrophy of cardiac myocytes in vitro.Cell. 75:977-984.

16.Schunkert,H.B.,S. S.Jackson,F. J.Tang,J. F. M.Schoen,C. S.Smits, B. H. Apstein, and B. H. Lorell. 1993. Distribution and functional significance

ofcardiac angiotensin converting enzyme in hypertrophied rat hearts. Circulation. 87:1328-1339.

17.Urata, H., B.Healy, R. W. Stewart, F. M. Bumpus, and A. Hussain. 1990. Angiotensin II-forming pathways in normal and failing human hearts. Circ. Res. 66:883-890.

18.Urata, H., A.Kinoshita, K. S. Misono, F. M. Bumpus, and A. Husain. 1990. Identification of a highly specific chymase as the major angiotensin II-forming enzyme in the human heart. J. Biol. Chem. 265:22348-22357.

19.Greenwood, F. C., W. M. Hunter, and J. S. Glover. 1963. The preparation of'31Ilabeled human growth hormone ofhigh specific radioactivity.Biochem. J. 89:114-123.

20.Admiraal, P. J. J., F. H. M.Derkx,A. H.JanDanser, H.Pieterman,and A. D. H. Schalekamp. 1990. Metabolism and production of angiotensin I in different vascular beds insubjectswithhypertension.Hypertension. 15:44-55.

21.Ganz,W., K.Tamura, and H. S. Marcus. 1971. Measurement of coronary sinusblood flowby continuous thermodilution inman.Circulation. 44:181-195. 22.Foult, J.-M., 0. Tavolaro, I. Antony, and A. Nitenberg. 1988. Direct

myocardialand coronary effects ofenalaprilatinpatientswith dilated cardiomyop-athy: assessment by abilateral intracoronary infusion technique. Circulation. 77:337-344.

23.Smith, T. 1988. Asimplified recycling model for thedosimetryof radioio-dide.Phys.Med. Biol. 33:1141-1157.

24.Danser, A. H.J., M. M. G. Koning, P. J. J.Admiraal,F. H. M.Derkx,

P. D.Verdouw, and M. A. D. H.Schalekamp. 1992. Metabolism ofangiotensin

I by different tissues in the intact animal. Am. J.Physiol.263:H418-H428. 25.Danser, A. H.J., M. M. G. Koning, P. J. J.Admiraal,L. M. A.Sassen,

F. H. M.Derkx, P. D.Verdouw,and M. A. D. H.Schalekamp. 1992.Production ofangiotensinsIand IIattissuesites in intactpigs.Am.J.Physiol. 263:H429-H437.

26. Bristow, M. R., R. Ginsburg, V.Umans, M. Fowler, W. Minobe, R. Rasmussen, P. Zera, R.Menlove, P. Shah, S. Jamieson,etal.1986.,/1-and

,/2-adrenergic receptorsubpopulations innonfailingand failinghuman ventricular

myocardium: couplingof both receptorsubtypestomuscle contraction and selec-tive/31-receptor down-regulationinheartfailure. Circ. Res. 59:297-309.

27.White,M., R.Roden,W.Minobe,M. F.Khan,P.Larrabee,M.

Wollmer-ing,J. D.Port, F.Anderson,D.Campbell,A. M.Feldman,etal. 1994.Age-related

changes in 3-adrenergicneuroeffector systems in the human heart. Circulation. 90:1225-1238.

28.Bull, H. G., N. A.Thomberry,and E. H. Cordes. 1985.Purification of

angiotensin-convertingenzyme from rabbitlungand humanplasma byaffinity chromatography. J.Biol. Chem. 260:2963-2972.

29. Lowry,0. H., N. J.Rosebrough, A. L.Farr, and R. J. Randall. 1951. Proteinmeasurementwith the Folin phenol reagent. J. Biol. Chem. 193:265-275. 30.Kinoshita, A.,H.Urata,F. M. Bumpus,and A. Husain. 1991. Multiple

determinants for thehighsubstratespecificityofanangiotensin II-forming

chy-masefrom the human heart. J. Biol. Chem 266:19192-19197.

31. Urata, H., A.Kinoshita,D. M.Perez,K.S.Misono,F. M.Bumpus,R. M.

Graham,and A.Husain. 1991.Cloningof the gene and cDNA for human heart

chymase.J.Biol. Chem. 266:17173-17179.

32.Urata, H.,K.Boehm,A.Philip,A.Kinoshita,J.Gabrovsek,F.Bumpus,

and A.Husain. 1993. Cellular localization andregionaldistributionofan

angioten-sinII-forming chymasein the heart. J. Clin. Invest.91:1269-1281.

33.Admiraal,P. J.J.,A. H. J.Danser,M. S.Jong,H.Pieterman,F. H. M.

Derkx,and M. A. D. H.Schalekamp. 1993.RegionalangiotensinHproduction

in essentialhypertensionand renal artery stenosis.Hypertension.21:173-184. 34.Haber, H.,E.Powers,L.Gimple,C.Wu,K.Subbiah,W.Johnson,and A. M. Feldman. 1994. Intracoronaryangiotensin-converting enzyme inhibition

improvesdiastolicfunction inpatientswithhypertensiveleft ventricular hypertro-phy.Circulation. 89:2616-2625.

35.Magrini, F.,P.Reggiani,M.Ciulla, R.Meazza,andG. Fratianni. 1992.

Coronary haemodynamiceffects ofangiotensinIIin mild essentialhypertension

inman.Clin. Sci.(Lond.). 82:133-137.

36.Dell'Italia, L.J., Q. C. Meng, I. Straeter-Knowlen, G. H. Hankes, S.

Bishop, S.Oparil,and T. Elton. 1994. Increased angiotensin peptidelevels in cardiac tissueofdogswith chronic mitralregurgitation. Circulation. 90:I-451a.

(Abstr.)

37.Danser,A. H.J.,J. P.vanKats,P.J. J.Admiraal,F. H. M.Derkx,J. M. J. Lamers, P. D. Verdouw, P. R. Saxena, and M. A. D. H. Schalekamp. 1994. Cardiac renin and angiotensins: uptake fromplasma versus in situ synthesis.

Hypertension. 24;37-48.

38.Kinoshita, A.,H.Urata,F. M.Bumpus,andA.Husain. 1993. Measurement ofangiotensinIconvertingenzyme inhibition in the heart. Circ. Res. 73:51-60. 39.Grima, M.,C.Welsch,E.Giesen-Crouse,C. Coquard,M.Barthelmebs,

and J. Imbs. 1990.Age-relatedvariations intissueangiotensin convertingenzyme activities:comparisonbetweenspontaneouslyhypertensiveandWistar-Kyotorats.

J.Hypertens. 8:697-702.

40.Jones,L.R.,H. R.Besch, Jr.,J. W.Fleming,M. M.McConnaughey,and A. M. Watanabe.1979.Separationof vesicles of cardiac sarcolemma from vesicles of cardiacsarcoplasmicreticulum: comparativebiochemical analysis of

(10)

41.Tomlins,B., S. E. Harding, M. S. Kirby, P. A. Poole-Wilson, and A. J. Williams. 1986. Contamination of a cardiac sarcolemmal preparation with endo-thelialplasma membrane. Biochim. Biophys. Acta. 856:137-143.

42.Hirsch, A., C. Talsness, H. Schunkert, M. Paul, and V. Dzau. 1991. Tissue-specific activation of cardiac angiotensin converting enzyme in experimental heart failure. Circ. Res.69:475-482.

43. Studer, R., H. Reinecke, B. Muller, J. Holtz, H. Just, and H. Drexler. 1994.Increasedangiotensin-Iconverting enzyme geneexpressionin the failing humanheart. J. Clin. Invest. 94:301-310.

44.Zisman, L. S., E. Bush, C. S. Taft, M. R. Bristow, M. B. Perryman, and M. V.Raynolds. 1994. Increase in angiotensin converting enzyme gene expression andactivity in the failing human ventricle. Circulation. 90:1-578. (Abstr.)

45.Schunkert,H., J. R.Ingelfinger,A. T.Hirsch,Y.Pinto, W. J. Remme, H.

Jacob,and V. J. Dzau. 1993. Feedbackregulation ofangiotensin converting en-zymeactivity and mRNA levels by angiotensin H. Circ. Res. 72:312-318.

46.King, S. J., and S. Oparil. 1992. Converting-enzyme inhibitors increase converting-enzyme mRNA and activity in endothelial cells. Am. J. Physiol. 263:C743-C749.

47.King, S. J., S. Oparil, and K. H. Berecek. 1991. Neuronal

angiotensin-converting enzyme (ACE) gene expression is increased by converting enzyme inhibitors (CEI). Mol. Cell. Neurosci. 2:13-20.

48.Wollert,K.C.,R.Studer, B. vonBulow,and H. Drexler. 1994. Survival aftermyocardialinfarction in the rat: role of tissueangiotensin-convertingenzyme inhibition. Circulation. 90:2457-2467.

49. Pfelischifter, J., and U. T. Ruegg. 1987. Cyclosporin A augments angioten-sin11-stimulatedrise in intracellular free calcium in vascular smooth muscle cells. Biochem. J. 248:883-887.

50.Meyear-Lehnert, H., D. Bokemeyer, U. Friedrichs, S. Drechsler, and H. J.

Kramer. 1993. CellularsignalingbycyclosporineA incontractile cells: interac-tions with atrial natriureticpeptide.Clin. Invest. 71:153-60.

51. Iwai, J., Y. Kanayama, N. N. Inoue, M. Okamura, and T. Takeda. 1993. Increased endothelial gene expression of angiotensin ATlA receptor in cyclosporine inducedhypertensiverats.Eur. J.Pharmacol. 248:341-344.

52. Altura, B. M., and B. T. Altura. 1976. Vascular smooth muscle and

prostaglandins.Fed. Proc. 35:2360-2366.

53. Caruso, A., S.Ferrazzani,S. DeCarolis,A.Lanzone, and S. Mancuso. 1991.Low-dose aspirin and vascular response in pregnant patients sensitive to angiotensinII.Int. J. Gynaecol. Obstet. 35:123-128.

54.Nakamura, M., T. Funakoshi, N. Arakawa, H. Yoshida, S. Makita, and K.Hiramori. 1994. Effect ofangiotensin-convertingenzyme inhibition on endo-theliumdependentperipheralvasodilation inpatientswith chronic heart failure. J. Am. Coll.Cardiol. 24: 1321-1327.

55. van Wijngaarden, J., A. J. Smit, P. A. de Graeff, W. H. van Gilst, S. A. van derBroek, D. J. van Veldhuisen, K.I.Lie, and H.Wesseling. 1994. Effects ofacetylsalicylicacid onperipheralhemodynamics in patients with chronic heart failure treated with angiotensin-converting enzyme inhibitors. J. Cardiovasc. Pharmacol.23:240-245.

56.Semple,P. F. 1980. The effects ofhemorrhageand sodiumdepletionon plasmaconcentrations of angiotensin II and [des-Asp 1 ] angiotensin II in the rat. Endocrinology. 107:771-773.

57.Cody,R.J., A. B.Covit,G. L.Schaer,and J. H. Laragh. 1984. Estimation ofangiotensinIIreceptoractivityin chroniccongestive heart failure. Am. Heart J. 108:81-89.

References

Related documents

Deigning to permit the Plaintiffs to publish the three papers filed under seal with the complaint, and dismissing Plaintiffs’ future work as “hypothetical,” the private Defendants ask

Keywords: World Cities; Nation-States; Global Competition; Global Companies; Market-Driven Management.. Flow Economy, Global Corporation

Keywords: Early warning system, Financial crisis, Emerging markets, Vietnam‟s economy, Signal approach model, Crisis prediction.. JEL Classification: E17,

The conclusions emerged from the quantitative and qualitative research for the semester course: « Contemporary Horizons of Folklore Studies, (Gender, Internet and

Firstly, At P15, in Arg1 KO mice, the density of myelinated axons was significantly reduced in subcortical white matter of the motor cortex, pyramidal tract in medulla, and CST in

In vitro studies in neonatal rat ventricular myocytes demonstrated the feasibility of mitigating aldosterone-induced mitochondrial superoxide levels, blunting noncanonical

But as estimated by IMF China’s growth will slow down in coming years and India will become the fastest growing economy of the world by 2019.In such scenario

The main purpose of this study was performed to investigate prognostic factors status at smaller tumors in patients with stage IB1 cervical cancer (CC) who underwent modified