Vasomotor Function of Pig Coronary Arteries after Chronic Coronary Occlusion

ORIGINAL ARTICLE

Vasomotor Function of Pig Coronary Arteries after ChronicCoronary Occlusion

Jinsheng Li, Hector De Leon, Takafumi Ueno, Jianhua Cui, Patrick K. Coussement, Spencer B. King III,Nicolas A. F. Chronos, and Keith A. Robinson

Abstract: Placement of an ameroid constrictor in large-conduitpig coronary arteries causes progressive stenosis and distal myo-cardial ischemia. Blood perfusion in the ischemic region is partlydependent on vasomotor responses to neural and humoral factorsdistal to the occlusion site. To ascertain the degree of impairmentof vascular function in pigs, the authors induced myocardial is-chemia by placing an ameroid constrictor in the left circumflexcoronary artery and examined vascular reactivity and histopa-thology distal to the constriction site. The sensitivity of the distalleft circumflex coronary and nonoccluded control left anteriordescending arteries to PGF2� was similar. After nitric oxideblockade using Nw-nitro-l-arginine methylester (L-NAME), thesensitivity and maximal contraction to PGF2� were significantlyincreased in both the left circumflex coronary (EC50: 5.86 ± 0.74vs. 3.28 ± 0.84 µM; Cmax: 4.63 ± 0.28 vs. 6.25 ± 0.30 g, P < 0.01)and left anterior descending (EC50: 6.57 ± 0.73 vs. 2.78 ± 0.16µM; Cmax: 5.09 ± 0.37 vs. 6.95 ± 0.39 g, P < 0.01) arteries. Sub-stance P-induced relaxation (100 pM) was blocked to a larger de-gree in the distal left circumflex coronary artery when comparedwith the left anterior descending artery (76.9 ± 4.2% vs. 56.4 ±3.1%, P < 0.05). Endothelium-independent relaxation to sodiumnitroprusside was similar in the left circumflex coronary and leftanterior descending arteries before and after nitric oxide block-ade. Histopathologic examination showed no major differencesbetween distal left circumflex coronary artery segments and leftanterior descending artery controls. However, scanning electronmicroscopy showed endothelial hypertrophy and activation inspecimens from the left circumflex coronary arteries. In sum-mary, as a result of the major hemodynamic changes induced bya chronic constriction and eventual occlusion of a large coronaryartery, distal segments underwent adaptive compensatorychanges. Such compensation may be related to an increased nitricoxide production by the hypertrophic endothelium in responseto alterations in coronary hemodynamics.

Key Words: Ameroid constrictor—Myocardial ischemia—Vasoreactivity.

( J Cardiovasc Pharmacol ™ 2003;41: 600–608)

Hypercholesterolemia, hypertension, smoking, and dia-betes have all been associated with impaired endothe-

lial nitric oxide (NO)-mediated vasodilatation (1,2). Theearliest vascular atherosclerotic lesions occur in the endo-thelium lining large blood vessels (3), and coronary arterieslocated distal to an occlusion have been shown to exhibitaltered reactivity to vasoactive agonists (4,5). Headrick etal. found that a brief coronary occlusion and reperfusionresulted in decreased endothelium-dependent relaxationand increased contraction in the segment distal to the oc-clusion (6). Other investigators have reported that chroniccoronary occlusion impairs receptor-dependent cAMP-mediated relaxation (7). Griffin et al. demonstrated that ex-ercise improves endothelium-mediated relaxation ofchronically occluded coronary arteries by enhancing theproduction of NO and endothelium-derived hyperpolariz-ing factor (8).

Gradual narrowing of atherosclerotic coronary arter-ies leads to myocardial ischemia. Impaired responses ofepicardial vessels to vasoactive substances might exacer-bate the compromised perfusion of ischemic myocardium.Therefore, it is important to study the pathologic processand alteration of function in coronary vessels distal to thenarrowing site. Placement of ameroid constrictors on pigcoronary arteries has become an accepted model to mimichuman chronic myocardial ischemia (9,10). Ameroidplacement results in a gradual and progressive reduction ofthe arterial lumen size, which is substantially occluded by 3or 4 weeks. Basal (resting) blood flow to the chronicallyoccluded artery is maintained at levels comparable to non-occluded vessels via collateral circulation (11,12). How-ever, the conduit artery vasomotor function distal to thesite of ameroid constrictors responsible in part for the regu-lation of myocardial perfusion in the affected region hasnot been thoroughly elucidated. We hypothesized thatcoronary arteries distal to a site of chronic occlusion would

The American Cardiovascular Research Institute, Norcross, Georgia,U.S.A.

Received April 2, 2002; accepted August 21, 2002.Address correspondence and reprint requests to Keith A. Robinson, MD,

American Cardiovascular Research Institute, 3155 NorthwoodsPlace, Norcross, GA 30071, U.S.A. E-mail: [email protected]

This study was funded in part by R01 grant HL 60184-01 from the Na-tional Heart, Lung, and Blood Institute (K. A. Robinson) and the RichFoundation (H. De Leon, K. A. Robinson).

600 J Cardiovasc Pharmacol� • Volume 41, Number 4, April 2003

exhibit altered vascular function. To test this hypothesis, anameroid occluder was placed in the left circumflex (LCX)coronary artery of juvenile crossbred farm swine. Alter-ations in vascular function distal to the site of occlusionwere examined by angiographic scoring of collateral flow(using quantitative coronary angiography) and isometrictension responses of isolated vessel rings 8 weeks afterameroid placement. Vascular histopathology and endothe-lial cell morphology were analyzed by bright field andscanning electron microscopy, respectively.

METHODS

Myocardial Ischemia ModelAnimal care and handling conformed to National In-

stitutes of Health and American Heart Association guide-lines and were approved by the Institutional Animal Careand Use Committee of the Atlanta Cardiovascular Re-search Institute. Six juvenile crossbred farm pigs (35–55kg) were sedated with an intramuscular injection of telazol(5 mg/kg), intubated, and anesthetized with inhalant isoflu-rane. A thoracotomy was performed through the fourth leftintercostal space. The pericardium was opened and anameroid constrictor was placed around the proximal LCXjust distal to the main stem of the left coronary arterymatching the size of the vessel (typically 1.75, 2.00, 2.25mm ID). The chest was then closed, and the animals wereallowed to recover and returned to their cages. Animalswere killed 8 weeks later, hearts were harvested, and coro-nary arteries were processed for in vitro organ chamberstudies (n = 6 pigs), and histology and scanning electronmicroscopy (n = 2 pigs).

Coronary AngiographyFormation of collateral vessels was evaluated by

coronary angiography 8 weeks after ameroid placement. A6F JR4 and hockey stick guiding catheter was introducedover a 0.035 “J” wire via femoral artery cutdown. Angiog-raphy was performed on the right and left coronary arteriesin orthogonal LAO and RAO projections. The mean aorticpressure, left ventricular end-diastolic pressure, and meanleft atrial pressure were recorded during the procedure.Quantitative coronary angiography was performedthrough cine review by two blinded experienced angiog-raphers. The stenosis at site of ameroid placement was cal-culated by the following equation: (1 − lumen diameterstenotic site/lumen diameter reference) × 100. The assess-ment of collateral development in the LCX region wasbased on the Rentrop classification (0: absent; 1: filling ofside-branches of a target occluded epicardial coronary ar-tery via collaterals without visualization of the epicardialcoronary artery itself; 2: partial filling of the epicardial seg-ment via collateral arteries; 3: complete filling of the epi-cardial segment).

Histopathologic AnalysisArteries from two animals were used for histologic

analysis. Animals were killed and distal vessels were ex-cised and fixed with 10% neutral buffered formalin over-night. The vessel segments were trimmed of perivascularconnective and fat tissue and prepared for histopathologicanalysis. Vessel segments were embedded in paraffinblocks, sectioned, and stained with hematoxylin and eosin(H&E) and Verheoff-Masson. To study the morphologicfeatures of the vessel at the site where the ameroid constric-tor was placed, sections were dehydrated in a graded seriesof ethanol. Infiltration was performed using Technovit7200 (isobornyl methacrylate) and xylene at 70:30% and100% (twice) dilutions for approximately 4 hours each.Sections were then embedded using a light-polymerizingunit (EXAKT Apparatebau, Wehrheim, Germany). Sec-tions were trimmed of excess plastic and cut into threeblocks, which were mounted on a plexiglas slide usingTechnovit 7210. A second slide was glued to the oppositeside and a ≈150-µm thick section was cut using a precisionband saw (EXAKT Apparatebau). Sections were ground to≈15 µm and polished (EXAKT Apparatebau) using pro-gressively finer grinding papers (P1000, P1200, P2000,P2500) until a smooth surface was obtained, and thenstained with H&E.

Scanning Electron MicroscopyTwo distal LCX and two control normal LAD seg-

ments were processed for scanning electron microscopy tocompare the ultrastructural features of endothelial cells.Formalin-fixed samples were rinsed with cacodylate bufferand postfixed for 1 hour in buffered 1% OsO4. Sampleswere dehydrated in graded ethanol series to 100%, critical-point dried from liquid CO2 with thermoregulation andflow monitoring, and cut transversely and longitudinallyinto four pieces. These were attached with luminal surfacesup to aluminum supports using conductive paste, and mag-netron sputter coated with 20-nm Au/Pd alloy. They wereexamined on the lower stage of a Topcon DS-130 scanningelectron microscope equipped with a LaB6 emitter and 4.8-Mb images were acquired digitally on a Micron Pentium166 PC.

Isometric Tension ResponsesCoronary vessels from six pigs were harvested 8

weeks after ameroid placement and their vasoreactivitywas examined in vitro. After animals were killed, the heartswere harvested and placed in ice-cold Krebs solution. TheLCX artery distal to the ameroid and an identical-sizedLAD coronary artery segment were carefully dissected andcleaned of adherent fat and connective tissue. An LCX seg-ment (1.5–2.0 cm long), ≈1 cm distal from the occlusion

J Cardiovasc Pharmacol� • Volume 41, Number 4, April 2003 Vasomotor Function after Coronary Occlusion

© 2003 Lippincott Williams & Wilkins 601

site, was obtained and cut in four 4-mm rings. Arterial ringswere suspended in individual organ chambers (RadnotiGlass Technology, Inc., Monrovia, CA, U.S.A.). Thechambers were filled with 17 ml freshly made Krebs solu-tion with the following composition (mM): NaCl, 120;MgSO4, 1.17; KH2PO4, 1.18; NaHCO3, 25.0; CaCl2, 2.5;KCl, 4.7; glucose, 5.5; and 10 µM indomethacin at pH 7.4.The solution was continuously infused with 95% O2 and5% CO2 and maintained at 37°C. The vascular rings weregradually stretched to a basal tension of 3 g, which wascontinuously adjusted over approximately 1 hour untilstable. Vessels were subjected to the same passive tensionof 3 g throughout the remainder of the study. Krebs bufferwas changed every 15 min during the equilibration period.

The responsiveness of the vessel rings was tested with40 and 100 mM KCl. Rings were then contracted with arange of concentrations of PGF2� (0.3–30 µM). After wash-ing, rings were preconstricted with a single dose of PGF2�

(10 µM) until they reached a stable plateau (≈7 min). Arte-rial rings were then exposed to a series of endothelium-dependent and endothelium-independent vasodilators.Complete concentration-response curves to substance P(0.01–100 pM) and SNP (1 pM–10 µM), endothelium-dependent and endothelium-independent relaxing agents,respectively, were carried out. After incubation with 100µM Nw-nitro-l-arginine methylester (L-NAME) for 45 min,PGF2� and SNP concentration-response curves were re-peated. Due to the long-lasting effect of endothelin-1 (ET-1), rings were always contracted with ET-1 (0.1 nM–0.1 µM)at the end of each experiment. Vessels were washed for 45min between each concentration-response curve. Cmax wasdefined as the maximal response obtained after comple-tion of a cumulative dose-response curve. Isometric ten-sion was digitized, acquired, and analyzed using aPonemah Tissue Platform System (Gould Instrument Sys-tem, Inc., Valley View, OH, U.S.A.).

Statistical AnalysisData are presented as mean ± standard error of the

mean. Data were analyzed by a two-tailed t test to comparegroup means for LCX and LAD. Dose concentration re-sponses were obtained, plotted, and EC50s were calculatedusing SigmaPlot (SPSS). Significance was established at the95% confidence level (P < 0.05) using SigmaStat (SPSS).

RESULTS

Coronary AngiographyEight weeks after placement of an ameroid constric-

tor, collateral filling of the LCX was demonstrated (Fig.1A). Quantitative coronary angiography analysis showedcomplete stenosis of LCX arteries in the six swine sub-jected to the procedure, with a mean Rentrop score of 2.67± 0.21.

Histopathologic analysisPlastic-embedded sections of segments obtained

from the site of ameroid placement and stained with H&Eshowed that the LCX was completely occluded at that site.Numerous inflammatory infiltrates were found in the oc-cluded region with abundant interspersed collagen (Fig.1B). Extensive inflammatory infiltrates were present out-side the occluded area, within a few millimeters of theameroid constrictor (1C). Histopathologic analysis of distalLCX arterial segments revealed no apparent changeswhen compared with identical size LAD sections. BothH&E (not shown) and Verheoff-Mason stainings (Fig. 2)displayed typical cell morphology of endothelial cells aswell as SMC and fibroblasts present in the medial and ad-ventitial layers, respectively.

Scanning Electron MicroscopyA qualitative scanning electron microscopy assess-

ment revealed a normal cellular surface ultrastructure inLAD segments, with well-defined cell borders and few sur-face indentations or irregularities. The LCX specimens dis-played a more conspicuous surface anatomy with largerendothelial cells, less detectable cell borders, and a sub-stantial number of cells with surface microvilli (Fig. 3). Ap-proximately 40% of the coronary artery luminal surface inthe LCX region distal to the occlusion site showed en-larged endothelial cells with luminal surface microvilli,compared with less than 5% of the LAD luminal surface.

Isometric Tension ResponsesThere were no differences between distal LCX and

control LAD rings in their responses to 40 mM (5.20 ± 0.32vs. 5.09 ± 0.41 g) and 100 mM (5.98 ± 0.30 vs. 6.06 ± 0.28g) KCl. Although maximal contractions of distal LCX toPGF

2�(30 µM) and ET-1 (0.1 µM) were 10% and 23% lower,

respectively, than those observed for LAD, the differenceswere not statistically significant. The sensitivity of bothLCX and LAD to PGF2� was similar (EC50: 5.86 ± 0.74 vs.6.57 ± 0.73 µM). After NO synthase blockade with L-NAME, concentration-response curves to PGF2( wereshifted to the left in LCX (EC50: 5.86 ± 0.74 vs. 3.28 ± 0.84µM, P < 0.01) and LAD (EC50: 6.57 ± 0.73 vs. 2.78 ± 0.16µM, P < 0.01) arteries (Fig. 4). The maximal contractions toPGF2( were also significantly increased after addition of L-NAME in LCX (Cmax: 4.63 ± 0.28 vs. 6.25 ± 0.29 g, P <0.001) and LAD (Cmax: 5.09 ± 0.37 vs. 6.95 ± 0.39 g, P <0.01) arteries (Fig. 4). Endothelium-dependent relaxationto substance P (10–100 pM) was significantly blocked inboth arteries by using L-NAME (P < 0.01) (Fig. 5). Sub-stance P-induced relaxation at the highest concentration(100 pM) was blocked to a larger degree by l-NAME indistal LCX when compared with LAD (76.9 ± 4.2% vs.56.4 ± 3.1%, P < 0.05) (Fig. 5). Concentration-response

J. Li et al. J Cardiovasc Pharmacol� • Volume 41, Number 4, April 2003

602 © 2003 Lippincott Williams & Wilkins

curves to SNP, an endothelium-independent vasodilator,were similar between LCX and LAD at the range of con-centrations used (1 nM–10 µM) (Fig. 6). However, the sen-sitivity of both arteries was significantly enhanced, afterNO blockade with l-NAME, as evidenced by the leftwardshift of both curves (LCX: 355 vs. 47.4 pM, P < 0.01; LAD:401 vs. 46.5 pM, P < 0.01). The ET-1 concentration-dependent responses in LCX and LAD arterial rings werealmost identical (data not shown).

DISCUSSIONPlacing an ameroid constrictor around the LCX re-

sulted in progressive stenosis, chronic myocardial ische-mia, and coronary collateral circulation as previously re-ported in this model (11,13). Despite a total occlusion of theLCX 8 weeks after placement of an ameroid constrictor asrevealed by plastic-embedded sections, histopathologicanalysis of the distal LCX revealed no major changes in

any vascular layer. This finding is remarkable since theconstrictor necessarily evoked dramatic reductions incoronary filling pressure and laminar flow in the LCX.Eight weeks after the occlusion, the resting cardiac hemo-dynamic parameters were normal, and all distal LCXs ex-amined were well filled by contrast media during angiog-raphy (Rentrop score: 2.67 ± 0.21), indicating thatcollateral vessels were gradually opened and/or developedto maintain or reestablish filling of the occluded vessel inan anterograde or retrograde fashion, or a combination ofboth. Interestingly, chronic reduction of myocardial ische-mia by �-adrenergic receptor blockade does not attenuatecollateral development, suggesting that the latter is unre-lated to the extent and duration of myocardial ischemia(10). Collateralization after myocardial ischemia developsfrom intercoronary and extracardiac sources. The sourcesof the latter are the bronchial and internal mammary arter-ies, which might provide up to 30% of resting flow (12).

FIG. 1. A. Coronary angiography 8 weeks after ameroid occlusion. B and C. An LCX plastic-embedded segment at the site ofocclusion 8 weeks after ameroid placement. Collateral filling of the LCX was demonstrated (A), and an extensive fibroproliferativeand inflammatory response was observed surrounding the completely occluded artery (B, arrow). Numerous inflammatoryinfiltrates were present in all areas of the occluded segment (C, arrows) around the ameroid (*) material. Magnification: B = �200,C = �400.

J Cardiovasc Pharmacol� • Volume 41, Number 4, April 2003 Vasomotor Function after Coronary Occlusion

© 2003 Lippincott Williams & Wilkins 603

Endothelial dysfunction is one of the initial changesthat occur in the progression of coronary atherosclerosis(3), which eventually leads to chronic myocardial ische-mia. A number of endogenous and exogenous factors in-cluding heparin (14) and dipyridamole (15) have been im-plicated in the enhanced development of collateral-dependent blood flow in myocardial ischemia.Additionally, direct myocardial injection of an adenoviralvector containing the coding sequence for VEGF and peri-vascular delivery of acidic FGF in the porcine model ofmyocardial ischemia results in greater collateral vessel de-velopment and improved left ventricular function (16,17).Previous studies have suggested that chronic myocardial

ischemia (ameroid occluder model), as well as brief periodsof low-flow ischemia followed by reperfusion, reduces en-dothelium-dependent vasodilation in response to adeno-sine (6,18). In our experiments, distal LCX and nonoc-c luded LAD segments responded s imi lar ly toendothelium-dependent and endothelium-independentvasodilators (substance P and SNP) and vasoconstrictors(PGF2� and ET-1). Only the responses induced by thehighest dose of the endothelium-dependent receptor-mediated agonist, substance P (100 pM), tested after block-ade of NO production, were significantly different betweenoccluded LCX and nonoccluded LAD vessels. Previous re-ports in a similar model have shown that relaxation re-

FIG. 2. Light microscopy of Verheoff-Masson stained sections of control LAD (A, B) and LCX (C, D) distal to the occlusion site.No apparent morphologic differences were detected between LAD and LCX arterial segments. The vessel layers are clearlydelineated in both vessels. No intimal, medial, or adventitial hypertrophy or hyperplasia was present. Magnification: A and C =�40, B and D = �200.

J. Li et al. J Cardiovasc Pharmacol� • Volume 41, Number 4, April 2003

604 © 2003 Lippincott Williams & Wilkins

sponses to bradykinin, another endothelium-dependentreceptor-mediated vasodilator, are impaired in collateral-dependent resistance arteries (19). Pathophysiologic re-sponse differences between the large (≈3 mm) conduit ar-teries used in our study and the small (≈150 µm) resistancearteries in that report may account for the discrepancy.

Long-term exercise has been shown to improve flowreserve in collateral-dependent myocardium (20), and a16-week exercise-training program restored occludedLCX sensitivity to bradykinin to levels observed in arteriesof sedentary animals (19). In agreement with our results,large-conduit collateral-dependent LCX and normal LADcoronary arteries were shown to respond similarly to KCland PGF2� at concentrations similar to those tested in ourstudy (20–100 mM KCl and 0.1–100 µM PGF2�) (5). Al-though the adrenergic innervation of the LCX-perfusedmyocardium is not affected by ameroid constrictor place-ment (21), Rapps et al. found that �-adrenergic receptor-

mediated responses to norepinephrine and phenylephrinein the presence of propranolol were enhanced in LCX oc-cluded arteries compared with normal LAD in a caninemodel of chronic coronary occlusion (5). As in the porcinemodel (18), relaxation to adenosine in the canine modelwas also impaired in large-conduit LCX arteries (5). Sellkeet al. reported that relaxation to VEGF and FGF in vesselsharvested from the collateral-perfused LCX region wasgreater when compared with normally perfused LAD (22).Since relaxations to both VEGF and FGF were inhibited inthe presence of NG-nitro-L-arginine, relaxations werelikely induced through the tyrosine kinase-mediated re-lease of endothelium-derived NO. The authors attributedthe exaggerated responses to an increased expression ofVEGF and FGF receptors.

An increased sensitivity in SNP-induced relaxationwas observed in LCX and LAD coronary arteries after NOblockade with L-NAME. This supersensitivity effect has

FIG. 3. Scanning electron microscopy of control LAD (A, C) and LCX (B, D) distal to the occlusion site. LCX coronary arteriesdisplayed larger endothelial cells when compared with LAD control arteries. Surface microvilli were more often present onendothelial cells in LCX and cell borders were less well defined compared with LAD. Magnification: A and B = �150, C and D =�1500.

J Cardiovasc Pharmacol� • Volume 41, Number 4, April 2003 Vasomotor Function after Coronary Occlusion

© 2003 Lippincott Williams & Wilkins 605

FIG. 4. Concentration-response relationship to PGF2� in arterial segments of LCX distal to the occlusion site and LAD controlvessels in the presence and absence of L-NAME. Both arteries responded similarly and were similarly shifted to the left and upwardsafter NO production blockade.

FIG. 5. Concentration-response relationship to substance P in arterial segments of LCX distal to the occlusion site and LAD controlvessels in the presence and absence of L-NAME. Substance P induced a similar concentration-dependent relaxation in LCX andLAD rings. Substance P-induced relaxation was blocked by L-NAME in both vessels. However, in the presence of L-NAME, LCX ringsdid not relax at any concentration, whereas LAD rings showed a significantly larger substance P-induced relaxation at 100 pM. *P< 0.01, **P < 0.05.

J. Li et al. J Cardiovasc Pharmacol� • Volume 41, Number 4, April 2003

been previously reported and may be modulated at thelevel of the soluble guanylate cyclase (23). Removal of en-dothelium has been shown to potentiate the effects of SNPin porcine coronary arteries (24), and the ability of SNP ornitroglycerin to induce vasorelaxation and increases incGMP is significantly enhanced in arterial rings denudedof endothelium or treated with endothelial NOS inhibitors(23,25). These observations indicate that the endotheliumis a dynamic modulator of the vascular smooth muscle re-laxant effects of nitrovasodilators.

These findings and the results of our present studysuggest that a compensatory adaptive response of the en-dothelium in conduit coronary arteries develops afterchronic stenosis and eventual occlusion, which supportsenhanced vasodilation to endothelium-dependent agonistspromoting smooth muscle cell relaxation. This appears tobe associated with morphologic changes suggestive of en-dothelial cell activation and hypertrophy, and may be cor-related to laminar flow disturbance and reductions in coro-nary filling pressure and flow volume.

CONCLUSIONIn the current study, L-NAME caused greater inhibi-

tion of substance P-induced relaxation in distal segments of

occluded LCX coronary arteries. We hypothesize thatsuch inhibition might be the result of an increased NO pro-duction by an altered endothelium showing both hypertro-phy and surface specializations indicative of cell activation.Although these changes would implicate NO release as thepredominant mechanism for vascular tone regulation inlarge-conduit vessels distal to the occlusion site, the contri-bution of a decreased production of endothelium-derivedhyperpolarizing factor, other NO-independent pathways,or increased expression of substance P receptors cannot beruled out. Except for the changes observed after stimula-tion with substance P, chronic gradual occlusion of theLCX did not substantially modify the vasoreactivity distalto the occlusion site. As the collateral-dependent coronarycirculation developed, adaptive humoral responses com-pensated for the dramatic reduction in blood flow. Themechanisms involved in such adaptive responses may in-clude endothelial reactions to changes in coronary fillingpressure and flow patterns.

ACKNOWLEDGMENTSThe authors thank Connie Micko, Evan Dessasau,

and Daniel Micko for their excellent technical assistance inhistology and animal instrumentation; Robert Apkarian,

FIG. 6. Concentration-response relationship to SNP in arterial segments of LCX distal to the occlusion site and LAD control vesselsin the presence and absence of L-NAME. Similar concentration-dependent relaxation curves were obtained in LCX and LAD rings.Concentration response curves in LCX and LAD arteries were shifted to the left in the presence of L-NAME.

J Cardiovasc Pharmacol� • Volume 41, Number 4, April 2003 Vasomotor Function after Coronary Occlusion

© 2003 Lippincott Williams & Wilkins 607

PhD, and Jeannette Taylor of the IMMF, Department ofChemistry, Emory University for the scanning electron mi-croscopic sample preparation and documentation; andLesley Wood for her editorial input.

REFERENCES1. Landmesser U, Hornig B, Drexler H. Endothelial dysfunction in hy-

percholesterolemia: mechanisms, pathophysiological importance,and therapeutic interventions. Semin Thromb Hemost. 2000;26:529–37.

2. Cai H, Harrison DG. Endothelial dysfunction in cardiovascular dis-eases: the role of oxidant stress. Circ Res. 2000;87:840–4.

3. Kinlay S, Libby P, Ganz P. Endothelial function and coronary arterydisease. Curr Opin Lipidol. 2001;12:383–9.

4. Scheel KW, Galindez TA, Cook B, et al. Changes in coronary andcollateral flows and adequacy of perfusion in the dog following oneand three months of circumflex occlusion. Circ Res. 1976;39:654–8.

5. Rapps JA, Sturek M, Jones AW, et al. Altered reactivity of coronaryarteries located distal to a chronic coronary occlusion. Am J Physiol.1997;273:H1879–87.

6. Headrick JP, Angello DA, Berne RM. Effects of brief coronary oc-clusion and reperfusion on porcine coronary artery reactivity. Circu-lation. 1990;82:2163–9.

7. Heaps CL, Sturek M, Rapps JA, et al. Exercise training restores aden-osine-induced relaxation in coronary arteries distal to chronic occlu-sion. Am J Physiol Heart Circ Physiol. 2000;278:H1984–92.

8. Griffin KL, Laughlin MH, Parker JL. Exercise training improves en-dothelium-mediated vasorelaxation after chronic coronary occlu-sion. J Appl Physiol. 1999;87:1948–56.

9. Asahara T, Masuda H, Takahashi T, et al. Bone marrow origin ofendothelial progenitor cells responsible for postnatal vasculogenesisin physiological and pathologic neovascularization. Circ Res. 1999;85:221–8.

10. Symons JD, Pitsillides KF, Longhurst JC. Chronic reduction of myo-cardial ischemia does not attenuate coronary collateral developmentin miniswine. Circulation. 1992;86:660–71.

11. O’Konski MS, White FC, Longhurst J, et al. Ameroid constriction ofthe proximal left circumflex coronary artery in swine. A model oflimited coronary collateral circulation. Am J Cardiovasc Pathol. 1987;1:69–77.

12. White FC, Carroll SM, Magnet A, et al. Coronary collateral devel-opment in swine after coronary artery occlusion. Circ Res. 1992;71:1490–1500.

13. Roth DM, Maruoka Y, Rogers J, et al. Development of coronary

collateral circulation in left circumflex Ameroid-occluded swinemyocardium. Am J Physiol. 1987;253:H1279–88.

14. Carroll SM, White FC, Roth DM, et al. Heparin accelerates coronarycollateral development in a porcine model of coronary artery occlu-sion. Circulation. 1993;88:198–207.

15. Symons JD, Firoozmand E, Longhurst JC. Repeated dipyridamoleadministration enhances collateral-dependent flow and regionalfunction during exercise. A role for adenosine. Circ Res. 1993;73:503–13.

16. Mack CA, Patel SR, Schwarz EA, et al. Biologic bypass with the useof adenovirus-mediated gene transfer of the complementary deoxy-ribonucleic acid for vascular endothelial growth factor 121 improvesmyocardial perfusion and function in the ischemic porcine heart. JThorac Cardiovasc Surg. 1998;115:168–76.

17. Lopez JJ, Edelman ER, Stamler A, et al. Angiogenic potential of peri-vascularly delivered aFGF in a porcine model of chronic myocardialischemia. Am J Physiol. 1998;274:H930–6.

18. Laham RJ, Simons M, Tofukuji M, et al. Modulation of myocardialperfusion and vascular reactivity by pericardial basic fibroblastgrowth factor: insight into ischemia-induced reduction in endothe-lium-dependent vasodilatation. J Thorac Cardiovasc Surg. 1998;116:1022–8.

19. Griffin KL, Woodman CR, Price EM, et al. Endothelium-mediatedrelaxation of porcine collateral-dependent arterioles is improved byexercise training. Circulation. 2001;104:1393–8.

20. Roth DM, White FC, Nichols ML, et al. Effect of long-term exerciseon regional myocardial function and coronary collateral develop-ment after gradual coronary artery occlusion in pigs. Circulation.1990;82:1778–89.

21. Roth DM, White FC, Mathieu-Costello O, et al. Effects of left circum-flex Ameroid constrictor placement on adrenergic innervation ofmyocardium. Am J Physiol. 1987;253:H1425–34.

22. Sellke FW, Wang SY, Stamler A, et al. Enhanced microvascular re-laxations to VEGF and bFGF in chronically ischemic porcine myo-cardium. Am J Physiol. 1996;271:H713–20.

23. Moncada S, Rees DD, Schulz R, et al. Development and mechanismof a specific supersensitivity to nitrovasodilators after inhibition ofvascular nitric oxide synthesis in vivo. Proc Natl Acad Sci U S A. 1991;88:2166–70.

24. Thorin E. Influence of nitric oxide synthase inhibition and endothe-lin-1 receptor blockade on acetylcholine-induced coronary arterycontraction in vitro in dilated and ischemic cardiomyopathies. J Car-diovasc Pharmacol. 2001;38:90–8.

25. Dinerman JL, Lawson DL, Mehta JL. Interactions between nitroglyc-erin and endothelium in vascular smooth muscle relaxation. Am JPhysiol. 1991;260:H698–701.

J. Li et al. J Cardiovasc Pharmacol� • Volume 41, Number 4, April 2003

608 © 2003 Lippincott Williams & Wilkins