Author + information
- Received October 18, 2000
- Revision received July 10, 2001
- Accepted July 23, 2001
- Published online November 1, 2001.
- ↵*Reprint requests and correspondence: Dr. Jean-François Tanguay, Research Center, Montreal Heart Institute, 5000 Bélanger Street East, Montreal, Quebec, Canada, H1T 1C8
The goal of this research was to study the effect of locally delivered 17β-estradiol (17β-E) during angioplasty on endothelial function after percutaneous transluminal coronary angioplasty (PTCA) at four weeks.
The endothelium plays a major role in the structural and functional integrity of coronary arteries and is damaged by PTCA.
Juvenile swine were subjected to PTCA, after which each artery was randomly-assigned to 600-μg 17β-E delivered locally, an equal volume of vehicle (V) or PTCA alone. After four weeks, the improvement in endothelial function was assessed by angiography using intracoronary acetylcholine (Ach) infusion and by immunohistochemistry.
At 10−5mol/l and 10−4mol/l Ach, significant vasoconstriction was noted in arteries treated with PTCA alone (p < 0.01 and p < 0.0001, respectively) and with PTCA plus V (p < 0.02 and p < 0.001, respectively). No significant vasoconstrictive response to Ach was observed in arteries treated with PTCA plus 17β-E. Immunohistochemistry of vessels four weeks after PTCA revealed enhanced re-endothelialization (p < 0.0005) and endothelial nitric-oxide synthase (eNOS) expression (p < 0.0005) in PTCA plus 17β-E-treated arteries compared with the other two treatment groups. Arteries treated with 17β-E showed significantly lower neointima formation, which correlated inversely with the extent of re-endothelialization and eNOS expression.
Locally delivered 17β-E significantly enhances re-endothelialization and endothelial function after PTCA, possibly by improving the expression of eNOS. Since endothelial dysfunction can promote both restenosis and coronary spasm, local 17β-E administration is a promising new approach to improve long-term results after PTCA.
The vital role of the endothelium in the regulation of vascular tone is well recognized (1). The endothelium also has important inhibitory effects on platelet aggregation, monocyte adhesion and vascular smooth muscle cell (SMC) proliferation (2). Endothelial injury, with consequent endothelial dysfunction, is caused by percutaneous transluminal coronary angioplasty (PTCA) (3)and may play an important role in subsequent restenosis (4). Endothelial function worsens progressively after balloon injury of porcine coronary arteries (5). Prolonged systemically administered 17β-estradiol (17β-E) accelerates endothelial recovery in vivo after arterial injury (6). Since endothelial injury due to PTCA is a local event, we hypothesized that local delivery of a single dose of 17β-E after PTCA may enhance endothelial recovery.
Animal preparation and handling
The committee approved the study protocol for ethical animal research of the Montreal Heart Institute. Eleven juvenile farm pigs (three immature females with intact ovaries and eight castrated males) weighing 20 to 25 kg were utilized as previously described (7).
Using a standard PTCA balloon-catheter (balloon/artery ratio 1.1 to 1.2:1), three successive 30-s inflations at 10 atm were made with 30-s intervals between inflations. The location of each PTCA was the proximal to midportion of the left anterior descending coronary artery (LAD), left circumflex coronary artery and right coronary artery (RCA). For local delivery, the InfusaSleeve catheter (LocalMed Inc.) was used. Each coronary artery was randomized to receive 600 μg 17β-E (in 5 ml), vehicle (V) alone (5 ml) or PTCA alone. The vessel location was LAD (3, 4, 4), left circumflex coronary artery (4, 2, 5), RCA (4, 5, 2) in the 17β-E, PTCA plus V or PTCA alone arm, respectively. In the last two animals, only two arteries were randomized to treatment with either PTCA plus 17β-E or PTCA alone; each uninjured (naive) coronary artery was utilized to assess response to acetylcholine (Ach) and immunohistochemistry of normal porcine coronary arteries.
Intracoronary Ach infusion
All 11 animals underwent cardiac catheterization four weeks after PTCA. After a baseline coronary angiogram, selective cannulation of the proximal portion of a coronary artery was performed with a single-lumen balloon catheter (TotalCross, Schneider, Minneapolis, Minnesota) for administration of vasoactive agents. Acetylcholine (10−7mol/l, 10−6mol/l, 10−5mol/l and 10−4mol/l) was infused for 3 min each at 1 ml/min (to achieve estimated coronary artery blood concentrations of 10−9mol/l, 10−8mol/l, 10−7mol/l and 10−6mol/l, respectively ) through the lumen-port of the catheter. Coronary angiography was performed after each dose. Immediately after infusion of the highest concentration of Ach and angiography, 100 μg of nitroglycerin was administered via the lumen-port of the catheter, and a coronary angiogram was performed. The same protocol was repeated for the other two coronary arteries.
Quantitative coronary angiography
Coronary angiography was performed as previously reported (7)with a single-plane imaging system (Electromed, St-Eustache, Québec). An independent observer blinded to the treatment-assignment of vessels performed each segment measurement in three consecutive end-diastolic frames, and the results were averaged.
The animals were euthanized at four weeks as previously described (7). From the injured segments and the two naive arteries, serial sections of 3 to 5 mm were made. The sections were then treated with incremental concentrations of alcohol, followed by xylene and paraffin. Slices of 6 μm thickness were prepared and stained with Verhoeff’s stain for morphometry. A minimum of three slices for each artery was analyzed by morphometry, and the results were averaged. The areas of: 1) external elastic lamina; 2) internal elastic lamina; and 3) lumen were measured by digital planimetry, and the neointimal area ( − ) and media area ( − ) were obtained. The morphologic percentage of stenosis was calculated as 100 (1 − /).
Immunohistochemical analyses were performed on cross sections of arteries. For each artery, two sections obtained from the region demonstrating the maximal neointimal response (as observed on morphometry) were analyzed and the results averaged. The lumen circumference and the sum-total of the luminal border staining positively were measured for each section. The percentage of re-endothelialization and the percentage of endothelial nitric oxide synthase (eNOS) expression were calculated as percentages of the lumen circumference of the cross section staining positively by analysis with the lectin Dolichos biflorusagglutinin and eNOS stains, respectively. Immunohistochemical measurements were made under high-power (×1,000) by an examiner blinded to the treatment groups with appropriate positive and negative controls. For lectin immunohistochemistry, the 6 μm slices were first treated with hydrogen peroxide and methanol to block endogenous peroxide, incubated with the Dolichos biflorusagglutinin (Sigma, Oakville, Ontario), followed by treatment with 3,3′-diaminobenzidine (Vector Laboratories, Burlingame, California) and counterstaining with hematoxylin. For eNOS analysis, after blocking of endogenous peroxide and nonspecific antibodies, the slices were treated serially with the primary mouse anti-eNOS antibody (Bio/Can Scientific, Mississauga, Ontario), the secondary goat anti-mouse antibody (Vector Laboratories), incubated with avidin-biotin (Vector Laboratories), treated with 3,3′-diaminobenzidine (Vector Laboratories) and, finally, counterstained with hematoxylin.
The primary analysis end point was the endothelial function assessed by quantitative coronary angiography (QCA) after Ach infusion. Values are expressed as mean ± SEM. Basal and post-nitroglycerin coronary-artery diameters were compared among the three treatment groups with one-way analysis of variance. For each treatment group, Bonferroni-corrected two-tailed ttests were used for comparisons between the basal coronary artery diameter and diameter at the end of each Ach infusion and after nitroglycerin. The Kruskal-Wallis test was used for comparison of morphometry, lectin and eNOS expression among the three treatment groups. The linearity of relationships between variables was analyzed with Pearson correlation coefficients. Values were considered significant if p <0.05.
There were no significant differences in basal coronary-artery diameter (2.45 ± 0.18 mm for PTCA plus 17β-E; 2.75 ± 0.10 mm for PTCA alone and 2.77 ± 0.15 mm for PTCA plus V, p = 0.4) among the three treatment groups. No significant differences in vessel distribution between each treatment arm was observed.
No changes in heart rate, electrocardiogram or blood pressure were noted during local delivery or during intracoronary infusion of vasoactive agents except during the infusion of 10−4mol/l Ach in three arteries (two in the PTCA alone group and one in the PTCA plus V group) with a strong vasoconstrictive response associated with ST-segment elevation. Although the ST-segment elevation and vasospasm were immediately relieved after administration of nitroglycerin, these three measurements were excluded from the QCA analysis.
Response of the PTCA alone group to Ach
Compared with the basal coronary artery diameter (2.75 ± 0.1 mm), there were no significant changes in coronary artery diameter after intracoronary infusion of 10−7mol/l (2.63 ± 0.1 mm, p = 0.4) and 10−6mol/l (2.63 ± 0.1 mm, p = 0.4) Ach. At 10−5mol/l and 10−4mol/l, significant vasoconstrictive responses were noted (2.3 ± 0.12 mm, p = 0.01 and 1.79 ± 0.16 mm, p < 0.0001, Figs. 1 and 2). ⇓⇓After nitroglycerin, coronary artery diameter increased from 1.79 ± 0.16 mm after 10−4mol/l Ach to 2.45 ± 0.15 mm (p < 0.01 vs. 10−4mol/l Ach; p = 0.1 vs. basal diameter).
Response of the PTCA plus V group to Ach
Compared with the basal coronary artery diameter (2.77 ± 0.15 mm), 10−7mol/l Ach did not change coronary artery diameter (2.60 ± 0.13 mm, p = 0.4). A trend towards significant vasoconstriction was noted with 10−6mol/l Ach (2.33 ± 0.16 mm, p = 0.06). Significant vasoconstriction was produced by 10−5mol/l (2.24 ± 0.15 mm, p < 0.02) and at 10−4mol/l (1.89 ± 0.16 mm, p < 0.001) Ach infusion, respectively. Nitroglycerin reversed the vasoconstriction (from 1.89 ± 0.16 mm after 10−4mol/l Ach to 2.69 ± 0.16 mm after nitroglycerin [p < 0.004 vs. 10−4mol/l Ach; p = 0.7 vs. basal diameter]).
Response of PTCA plus 17β-E group to Ach
In the vessels treated with local delivery of 17β-E, no significant vasoconstrictive response to Ach (2.45 ± 0.18 mm under basal conditions, 2.39 ± 0.17 mm, 2.33 ± 0.17 mm, 2.30 ± 0.17 mm, 2.24 ± 0.17 mm after 10−7mol/l, 10−6mol/l, 10−5mol/l and 10−4mol/l Ach infusion, respectively [with corresponding p values of 0.8, 0.6, 0.6 and 0.4, respectively]) occurred at any concentration. A mild vasodilation was observed after administration of nitroglycerin: from 2.24 ± 0.17 mm after 10−4mol/l Ach to 2.50 ± 0.16 mm after nitroglycerin (p = 0.4; p = 0.9 for after nitroglycerin vs. basal diameter).
There were no significant differences in the final coronary-artery diameter achieved after administration of nitroglycerin among the three groups.
Immunohistochemical analyses were performed on all 11 animals. Three arterial segments were lost/damaged during harvesting of samples (two from the PTCA alone group, one from the PTCA plus V group). Re-endothelialization, assessed by the lectin Dolichos biflorusagglutinin stain (Fig. 3A)was greater in vessels treated with PTCA plus 17β-E compared with the other two groups (89.8 ± 1.6% for PTCA plus 17β-E, 69.3 ± 2.3% for PTCA alone and 72.8 ± 1.7% for PTCA plus V, p < 0.0005). Endothelial nitric oxide synthase expression (Fig. 3B)was higher in vessels treated with 17β-E (36.5 ± 3.2% for PTCA plus 17β-E, 10.1 ± 1.7% for PTCA alone and 9.2 ± 1.4% for PTCA plus V, p < 0.0005).
Significant inverse correlations between the extent of re-endothelialization (r = −0.45, p < 0.02) and between eNOS expression (r = −0.65, p < 0.0005) and the response to Ach were observed (date not shown).
Response of naive coronary arteries
To confirm that the abnormalities in Ach response, re-endothelialization and eNOS expression in post-PTCA vessels were due to PTCA and not simply a consequence of tissue handling or other technical factors, we studied these variables in two naive coronary arteries. No vasoconstrictive response was seen (basal diameter 2.92 ± 0.05 mm) in response to Ach at 10−6, 10−5and 10−4mol/l (2.99 ± 0.02 mm, 3.02 ± 0.02 mm, 3.04 ± 0.04 mm, respectively, p = NS). Immunohistochemical analysis revealed 100% lectin expression and 57.6 ± 5.2% eNOS expression.
The extent of morphologic tissue injury (9)induced by the initial angioplasty was similar among the three groups. Arteries treated with PTCA plus 17β-E demonstrated a significantly smaller neointimal response to injury compared with the other two treatment groups. Neointimal area (0.49 ± 0.1 mm2for PTCA plus 17β-E vs. 1.04 ± 0.22 mm2for PTCA alone vs. 1.31 ± 0.38 mm2for PTCA plus V, p < 0.05), neointimal/media area (0.75 ± 0.16 vs. 2.17 ± 0.49 vs. 1.97 ± 0.47, p < 0.01) and percentage of morphologic stenosis (18.7 ± 4% vs. 37.3 ± 7.6% vs. 32.2 ± 6.3%, p = 0.05) were lower in PTCA plus 17β-E arteries.
This study demonstrates for the first time that local delivery of 17β-E immediately after PTCA enhances subsequent re-endothelialization, eNOS expression and endothelial function at the site of injury. Besides its critical role in the regulation of vascular tone, the normal endothelium functions as a barrier between blood elements and underlying vascular SMC. Endothelium-derived NO, a potent vasodilator, also inhibits monocyte adherence and platelet aggregation and adhesion (10)and inhibits vascular SMC migration (11)and proliferation (12).
Endothelium regeneration and dysfunction after PTCA
After experimental arterial injury, re-endothelialization rates of 81% (13)and <50% (14)have been observed. In a study of specimens of restenotic lesions obtained by atherectomy in humans, no endothelial cells could be demonstrated (15). Coronary artery spasm due to endothelial dysfunction has been demonstrated at the site of PTCA (3). Patients with unstable angina due to restenosis experience significantly less recurrent angina upon treatment with intravenous nitroglycerin and no benefit with intravenous heparin, suggesting abnormal vasoconstriction at the site of PTCA (16). This phenomenon of abnormal vasoconstriction was also observed in our experiments (upon provocation with Ach) in arteries treated with PTCA alone or PTCA plus V; these changes resolved immediately upon administration of the endothelium-independent vasodilator nitroglycerin.
Effects of 17β-E on the endothelium
Therapy with subcutaneously implanted 17β-E pellets has been reported to enhance re-endothelialization after arterial injury (6). In this study, local treatment with a single dose of 17β-E at the time of PTCA was followed by nearly complete re-endothelialization (89.8 ± 1.6%), significantly greater than that observed in the groups not treated with 17β-E. The ability of 17β-E to increase vascular endothelial growth factor synthesis (17)and its effect on basic fibroblast growth factor may be responsible for the enhanced re-endothelialization (18). In cell culture assay, 17β-E treatment resulted in a 50% decrease in apoptosis of human umbilical-vein endothelial cells exposed to tumor necrosis factor-α (TNF-α) (19). Of note, increased TNF-α expression is known to occur after balloon injury (20). The effect of 17β-E to upregulate eNOS expression that we observed may be responsible for beneficial effects on endothelial function, as the vascular response to Ach is closely related to eNOS expression (21). In support of this notion, a strong inverse relationship was seen in our study between the vasoconstrictive response to Ach and eNOS expression. Estrogen has been observed to induce NOS (22,23).
The observations from the immunohistochemical analysis of the naive and the injured coronary arteries indicate that impaired re-endothelialization and downregulation of eNOS expression are evident four weeks after PTCA and are significantly improved by local delivery of 17β-E at the time of the initial procedure.
A unifying hypothesis for the responses we observed is that PTCA is associated with endothelial damage and eNOS downregulation, which prevents the vasodilatory response to Ach mediated by endothelial NO production. By improving eNOS expression, 17β-E allows the endothelium-dependent vasodilatory response of Ach to counteract its direct vasoconstricting action, thus preventing Ach-induced vasoconstriction at the site of local injury. The vasodilation by nitroglycerin in Ach-constricted arteries that had been subjected to PTCA alone is consistent with this concept, since exogenous nitroglycerin (a NO donor) simply provides a local NO-related dilation that the eNOS-deficient angioplastied segment cannot provide for itself.
Endothelial dysfunction and restenosis
In a small observational study, the majority of patients with documented spasm at the site of PTCA subsequently developed restenosis (24). Recently, endothelial dysfunction at the site of balloon angioplasty has been demonstrated as an independent factor for the development of restenosis (25). In this study, we demonstrated that enhanced endothelial recovery after a single dose of locally-delivered 17β-E is associated with significantly reduced neointima formation. Neointima formation correlated inversely with the extent of re-endothelialization and eNOS expression (data not shown), supporting the notion that estrogen-induced endothelial protection suppressed neointima formation.
Both rapid nongenomic and genomic effects have been postulated in the influence of 17β-E on coronary vasculature (26,27). Although protein synthesis was not quantified in this study, the enhanced eNOS expression and the response to Ach observed as late as 28 days after a single dose of 17β-E are consistent with a genomic effect. This is the first study to suggest a genomic effect after local 17β-E therapy of coronary arteries in vivo.
Spillover of 17β-E into systemic circulation could have occurred during local delivery. However, for systemic 17β-E to be effective, therapy for at least two weeks (with treatment beginning one week before arterial injury) is required (6), as the 17β-E that enters the circulation is rapidly metabolized (28). Even assuming that estrogen spillover could exert a systemic effect, all three treatment arms were applied in each animal so that 17β-E spillover should have occurred equally for all vessels subjected to PTCA. Systemic effects of 17β-E would, if anything, provide equal protection to all vessels. The differences noted between the arteries treated with PTCA plus 17β-E compared with PTCA alone and PTCA plus V cannot, therefore, be attributed to such systemic effects. The relation between the infusion rate of 1 ml/min of Ach and final estimated concentration in the coronary vascular bed was derived for the LAD artery (8)and may vary for the other two coronary arteries. Interarterial variations in coronary flow are unlikely to have had a significant bearing on the results of the study, as the three treatment groups were randomly assigned to include an equal representation of the different coronary arteries. In addition, immunohistochemical evaluation clearly established a significant beneficial effect on re-endothelialization and eNOS expression in 17β-E-treated arteries compared with the other two treatment groups and was not subject to artifacts related to assumptions about coronary flow. As the female pigs were prepubertal and the male pigs were castrated, it may not be appropriate to make generalizations about a gender effect from the data obtained. The effects of 17β-E may be reduced with higher levels of testosterone or in subjects with aging-related endothelial dysfunction. Finally, stents have become widely used, but evaluating endothelial function in a stented vessel remains limited and cannot be extrapolated from our results.
We conclude that a single dose of 17β-E delivered locally after balloon injury can significantly improve re-endothelialization and enhance endothelial function at the injured site as late as one month after PTCA. Besides the beneficial vascular effects of improved endothelial function, this observation may be of additional importance because endothelial dysfunction is an independent predictor of restenosis (25). Improved endothelial function correlated significantly with decreased neointima formation in estrogen-treated vessels. The approach of local 17β-E administration during PTCA merits further study, with a view to potential clinical value in the prevention of vascular dysfunction and restenosis.
The authors are indebted to Pascale Geoffroy, MSc, and Julie Lebel for technical support and to Dr. Martin G. Sirois, PhD, Dominique Lauzier and Veronique Philibert for their assistance with immunohistochemistry and morphometry.
☆ Dr. Tanguay is supported by the Montreal Heart Institute.
- endothelial nitric oxide synthase
- left anterior descending artery
- percutaneous transluminal coronary angioplasty
- quantitative coronary angiography
- right coronary artery
- smooth muscle cell
- tumor necrosis factor-α
- Received October 18, 2000.
- Revision received July 10, 2001.
- Accepted July 23, 2001.
- American College of Cardiology
- ↵Rubanyi GM. The role of endothelium in cardiovascular homeostasis and diseases. J Cardiovasc Pharmacol 1993;22 Suppl 4:S1–14.
- Shimokawa H.,
- Flavahan N.A.,
- Vanhoutte P.M.
- Krasinski K.,
- Spyridopoulos I.,
- Asahara T.,
- van der Zee R.,
- Isner J.M.,
- Losordo D.W.
- Chandrasekar B.,
- Tanguay J.F.
- Lerman A.,
- Webster M.W.,
- Chesebro J.H.,
- et al.
- Karas S.P.,
- Gravanis M.B.,
- Santoian E.C.,
- Robinson K.A.,
- Anderberg K.A.,
- King S.B. III.
- Cooke J.P.,
- Tsao P.S.
- Sarkar R.,
- Meinberg E.G.,
- Stanley J.C.,
- Gordon D.,
- Webb R.C.
- ↵Cornwell TL, Arnold E, Boerth NJ, Lincoln TM. Inhibition of smooth muscle cell growth by nitric oxide and activation of cAMP-dependent protein kinase by cGMP. Am J Physiol 1994;267:C1405–13.
- Hayashi Y.,
- Tomoike H.,
- Nagasawa K.,
- et al.
- Bauriedel G.,
- Windstetter U.,
- DeMario S.J. Jr.,
- Kandolf R.,
- Hofling B.
- Doucet S.,
- Malekianpour M.,
- Theroux P.,
- et al.
- Hyder S.M.,
- Stancel G.M.,
- Chiappetta C.,
- et al.
- Spyridopoulos I.,
- Sullivan A.B.,
- Kearney M.,
- Isner J.M.,
- Losordo D.W.
- Tanaka H.,
- Sukhova G.,
- Schwartz D.,
- Libby P.
- Weiner C.P.,
- Lizasoain I.,
- Baylis S.A.,
- Knowles R.G.,
- Charles I.G.,
- Moncada S.
- Hollman J.,
- Austin G.,
- Gruentzig A.R.,
- Douglas J.S. Jr.,
- King S.B. III.
- Lafont A.,
- Durand E.,
- Samuel J.L.,
- et al.
- Williams J.K.,
- Adams M.R.,
- Herrington D.M.,
- Clarkson T.B.
- Wellman G.C.,
- Bonev A.D.,
- Nelson M.T.,
- Brayden J.E.
- ↵Murad F, Kuret JA. Estrogens and progestins. In: Gilman AG, Rall TW, Nies AS, Taylor P, editors. The Pharmacological Basis of Therapeutics. New York, NY: Pergamon Press, 1990:1384–412.