Bicuspid anatomy of the aortic valve (BAV) is a common

Transcription

1 Endothelial Nitric Oxide Synthase in Bicuspid Aortic Valve Disease Diana Aicher, MD, Carmen Urbich, MD, Andreas Zeiher, PhD, Stefanie Dimmeler, PhD, and Hans-Joachim Schäfers, PhD Department of Thoracic and Cardiovascular Surgery, University-Hospital, Homburg/Saar; Molecular Cardiology, Department of Internal Medicine III, University of Frankfurt, Frankfurt, Germany Background. The pathogenesis of ascending aortic dilatation in the presence of a bicuspid valve is discussed controversially. Recent experimental evidence suggests that the expression of endothelial nitric oxide synthase (enos) may have an influence on aortic valve anatomy and aneurysmal dilatation of the aorta. We investigated the relationship among enos expression, valve anatomy, and aortic dilatation in the human aortic wall. Methods. Aortic wall specimens from 39 patients with aortic valve disease (bicuspid, n 17; tricuspid, n 22) were studied. The functional aortic valve pathology was regurgitation (n 22), stenosis (n 10), and combined aortic valve disease (n 7). The specimens were obtained intraoperatively from the aortic wall above the noncoronary sinus. The enos protein expression was quantified by western blot analysis after immunoprecipitation from tissue lysates. The enos levels were analyzed for correlation with valve anatomy and ascending aortic diameters. Results. The enos protein expression of aortic endothelial cells was significantly lower in patients with bicuspid as compared with tricuspid aortic valves (4, vs 6, ; p 0.017). In bicuspid aortic valves there was a significant correlation between enos expression and maximum aortic diameter (r 0.530; p 0.029) or sinotubular diameter (r 0.520; p 0.033). In patients with tricuspid aortic valves, no significant correlation between aortic size and enos expression was found. Conclusions. Our results show an association between enos levels and aortic valve anatomy as well as aneurysm formation in patients with bicuspid aortic valves. (Ann Thorac Surg 2007;83:1290 4) 2007 by The Society of Thoracic Surgeons Bicuspid anatomy of the aortic valve (BAV) is a common congenital heart lesion in which dilatation of the proximal aorta occurs in approximately 50% of individuals [1, 2]. The pathogenesis of aortic dilatation in the presence of a bicuspid valve has been discussed controversially. It has been assumed that the abnormal flow in the aortic root caused by a stenotic bicuspid valve leads to increased wall stress and consecutive dilatation as a physiologic phenomenon [3, 4]. Aortic dilatation, however, has also been observed in individuals with BAV with normal hemodynamic function [5 7]. Histopathologic changes of the proximal aortic wall indicating an inherent structural defect have been described in patients with ascending aortic dilatation and bicuspid valve anatomy [8, 9]. On the basis of clinical findings a combined genetic defect has been proposed, possibly leading to the development of a bicuspid valve and structural abnormalities in the aortic wall with variable expression [10, 11]. Endothelium-derived nitric oxide (NO) produced by endothelial NO synthase (enos) has been shown to be involved in the pathogenesis of aneurysms in animal models [12, 13]. Most interestingly, enos knockout mice Accepted for publication Nov 28, Address correspondence to Dr Aicher, Department of Thoracic and Cardiovascular Surgery, University Hospital Homburg, Homburg, Germany; dianaaicher@gmx.de. have been shown to exhibit a high prevalence of congenital BAV [12] and, in combination with deficiency of apolipoprotein E, also spontaneous development of aortic aneurysms [13]. These findings indicate a possible relationship among enos, valve anatomy, and aortic wall structure. It is unclear whether a similar relationship among enos expression, bicuspid valve anatomy, and aortic dilatation exists in humans. We compared enos expression in aortic wall endothelial cells of patients with bicuspid and tricuspid aortic valves (TAV) with and without aortic dilatation. Material and Methods Aortic wall specimens from 39 patients with diseased bicuspid or tricuspid aortic valves (Table 1) undergoing valve surgery were studied for enos expression. The study was approved by the local ethics committee and all patients gave informed consent. The 39 specimens were obtained from consecutive patients undergoing aortic valve replacement/repair. Because the prevalence of bicuspid aortic valve anatomy is approximately 50% in patients undergoing aortic valve surgery we anticipated similar or equal numbers in the two subgroups formed based on the intraoperative appearance of valve morphology. There was no preselection, no further patients were studied, and no patient was excluded by The Society of Thoracic Surgeons /07/$32.00 Published by Elsevier Inc doi: /j.athoracsur

2 Ann Thorac Surg AICHER ET AL 2007;83: enos IN BICUSPID AORTIC VALVE DISEASE 1291 Table 1. Baseline Clinical Characteristics of the Patients Characteristics Bicuspid (n 17) Tricuspid (n 22) p Bonferroni-Holm Age Sex (f/m) 10/7 7/ Aortic valve pathology Stenosis 2/17 (12%) 7/22 (32%) Regurgitation 11/17 (65%) 12/22 (55%) Combined 4/17 (24%) 3/22 (14%) Medical history Hypertension 6/17 (35%) 13/22 (59%) Diabetes mellitus 2/17 (12%) 4/22 (18%) Current smoking 0/17 3/22 (14%) Coronary artery disease 4/17 (24%) 10/22 (46%) Left ventricular ejection fraction (%) Current medication -blocker 9/17 (53%) 13/22 (59%) ACE inhibitor 8/17 (47%) 10/22 (45%) Statin 9/17 (53%) 8/22 (36%) Diuretic 10/17 (59%) 15/22 (68%) Insulin 1/17 (6%) 3/22 (14%) LDL cholesterol mg/dl mg/dl ACE angiotensin-converting enzyme; f female; LDL low density lipoproteins; m male. Aortic wall samples were obtained at the time of aortic valve or ascending aortic replacement. They were excised from the lateral portion of the aorta 5 mm above the sinotubular junction of the noncoronary sinus, placed immediately in liquid nitrogen, and stored at 80 C until processing. Maximum ascending aortic diameters were measured in preoperative computed tomographic scans. Sinotubular and aortoventricular diameters were measured intraoperatively. Aortic wall samples were mortared and lysed in buffer (20 mmol/l Tris [ph 7.4], 150 mmol/l NaCl, 1 mmol/l ethylenediaminetetraacetic acid, 1 mmol/l egtazic acid, 1% Triton-X-100 [Union Carbide, Bound Brook, NJ] 2.5 mmol/l sodium pyrophosphate, 1 mmol/l -glycerolphosphate, 1 mmol/l sodium orthovanadate [Na 3 VO 4 ], 1 g/ml leupeptin, and 1 mmol/l phenylmethylsulfonyl fluoride) for 20 minutes on ice. After centrifugation for 15 minutes at 20,000g (4 C), the protein content of the samples was determined according to Bradford [14]. The protein expression of enos was determined by Western blot analysis of tissue lysates or after immunoprecipitation. To immunoprecipitate enos, protein lysates were incubated with an antibody against enos (1:100; BD Biosciences, Heidelberg, Germany) overnight at 4 C. After incubation with 30 L of protein A/Gagarose beads (Santa Cruz Biotechnology, Heidelberg, Germany)/0.5 mg protein for 1.5 hours at 4 C, the protein-complexes were washed and analyzed for enos expression by Western blotting. The proteins were loaded in SDS-polyacrylamide gels (8%) and blotted onto polyvinylidene difluoride membranes. After blocking with 3% bovine serum albumin (BSA) and 5% milk powder, respectively, at room temperature for two hours, the enos (1: 3000; BD Biosciences), platelet endothelial cell adhesion molecule 1 (PECAM-1; CD31; 1:500; Santa Cruz Biotechnology) and tubulin (1:20; Lab Vision/Neomarkers, Fremont, CA 0) antibodies were incubated in tris buffered saline (50 mm Tris/HCl, ph 8; 150 mm NaCl, 2.5 mm KCl), 0.1% Tween-20, 3% BSA, and 3% milk powder, respectively, overnight at 4 C. After incubation with the secondary antibodies (1:4000; anti-mouse-horseradish peroxidase or anti-goat-horseradish peroxidase; Amersham, Hamburg, Germany) for one hour at room temperature, enhanced chemiluminescence was performed according to the instructions of the manufacturer (Amersham). The autoradiographies were scanned and analyzed semiquantitatively (Fig 1; representative immunoblot). Plateletendothelial cell adhesion molecule 1 (PECAM-1) and tubulin levels were used as endothelial marker and Fig 1. Representative immunoblot of endothelial nitric oxide synthase (enos), tubulin, and platelet endothelial cell adhesion molecule 1 (PECAM-1) in bicuspid and tricuspid aortic valves.

3 1292 AICHER ET AL Ann Thorac Surg enos IN BICUSPID AORTIC VALVE DISEASE 2007;83: loading control, respectively, in order to standardize the examined amounts of aortic wall endothelial cells. The ratio enos/pecam-1/tubulin was calculated and analyzed for possible correlation with aortic valve anatomy and aortic diameters. In order to normalize the measurements of enos for quantitative differences in aortic wall tissue, a ratio enos/pecam-1/tubulin was calculated. This was initially done by Western blot analysis of tissue lysates and, then, compared with data after immunoprecipitation of enos. The resulting data (ratio enos/pecam-1/ tubulin) were similar to the results obtained for enos using immunoprecipitation. In order to amplify the signal of enos expression in a heterogenous lysate of vascular cells (smooth muscle cells, fibroblasts, endothelial cells), we systematically performed the analysis with enos expression after immunoprecipitation. The results were analyzed retrospectively with the patients divided into two groups according to their respective valve anatomy (bicuspid versus tricuspid). Data are expressed as mean standard deviation. Differences between the two groups were tested using the 2 test. A level of less than 0.05 was considered statistically significant. The p values were corrected for multiple testing using the Bonferroni-Holm method. The correlation between variables was estimated by the Spearman rank correlation test. Results The aortic valve was bicuspid in 17 and tricuspid in 22 instances. The groups were comparable in most aspects. There was a trend toward lower age, less coronary artery disease, lower prevalence of hypertension, and higher percentage of treatment with statins in the bicuspid valve patients. None of these characteristics differed significantly between the groups by post hoc testing. Aortic size did not differ between the groups in aortic aortoventricular (BAV mm, TAV mm), sinotubular Fig 3. Bicuspid aortic valve. Correlation between enos expression (determined as ratio enos/pecam-1/tubulin) and maximum aortic diameter in bicuspid aortic valves. There was a significant correlation between enos expression and maximum aortic diameter (r 0.530; p 0.029). ( one aortic wall sample; enos endothelial nitric oxide synthase; PECAM-1 platelet endothelial cell adhesion molecule 1.) diameter (BAV mm, TAV mm), or maximum aortic diameter (BAV mm, TAV mm). Ten of the 39 patients (five with bicuspid; five with tricuspid aortic valves) had aortic dilatation with a maximum diameter of 50 mm or more. The amount of enos in aortic wall endothelial cells was significantly lower in patients with bicuspid aortic valves (4, units) compared with tricuspid aortic valves (6, units; p 0.017) (Fig 2). There was no difference in enos expression among differing valve pathologies (regurgitation; stenosis; mixed lesion) in bicuspid valves (regurgitation 4, ; stenosis 3,489 2,388; mixed lesion 4,993 1,100; p 0.35). Similarly, Fig 2. Expression of enos (determined as ratio enos/pecam-1/ tubulin) in the aortic wall of patients with tricuspid and bicuspid aortic valves. ( one aortic wall sample; enos endothelial nitric oxide synthase; PECAM-1 platelet endothelial cell adhesion molecule 1.) Fig 4. Tricuspid aortic valves. The enos expression and maximum aortic diameter in tricuspid aortic valves. In patients with tricuspid aortic valves no correlation between aortic size and enos expression was found (maximum aortic diameter and enos: r 0.093; p 0.676). ( one aortic wall sample; enos endothelial nitric oxide synthase; PECAM-1 platelet endothelial cell adhesion molecule 1.)

4 Ann Thorac Surg AICHER ET AL 2007;83: enos IN BICUSPID AORTIC VALVE DISEASE 1293 valve pathology was not related to enos level in tricuspid valve anatomy (regurgitation 6, ; stenosis 5, ; mixed lesion 6, ; p 0.49). In patients with bicuspid aortic valves, enos expression was significantly lower in individuals with arterial hypertension compared with patients with normal blood pressure (2, vs 5, ; p 0.007), but there was no correlation between blood pressure and enos expression (r 0.124; p 0.54). In tricuspid aortic valves, enos expression did not differ between patients with hypertension and those with normal blood pressure (5, vs 6, ; p 0.217). There was no correlation between aneurysm and blood pressure (r 0.283; p 0.18). There was no difference in enos expression between patients with or without statin therapy in bicuspid (3, vs 5, , p 0.124) or tricuspid aortic valves (5, vs 6, ; p 0.09). There was a significant correlation between enos expression and maximum aortic diameter (Fig 3; r 0.530; p 0.029) as well as sinotubular diameter (r 0.520; p 0.033) in bicuspid aortic valves. In patients with tricuspid aortic valves no correlation between aortic size and enos expression was found (maximum aortic diameter and enos: r 0.093; p 0.676; Fig 4). Comment Although an increased prevalence of aortic complications in the presence of bicuspid aortic valves has been recognized [15, 16], its etiology is still unclear. The occurrence of proximal aortic dilatation has been related to abnormal flow through a stenosed BAV [3]. This, however, does not explain why dilatation occurs in the presence of aortic regurgitation or even the absence of valve dysfunction in 50% of individuals with bicuspid aortic valves [5 7]. Previous studies have demonstrated histopathologic changes related to cystic medial necrosis, elastic fragmentation, and smooth muscle cell orientation in bicuspid aortic valve disease in comparison with tricuspid aortic valve disease [8]. A recent publication [12] describing a high prevalence of BAV and aneurysm formation in enos knockout mice indicated a potential role of NO in this phenomenon, possibly as a mediator in cardiac development. These findings were also reported by Kuhlencordt and colleagues [13], who found an increased incidence of spontaneous aortic aneurysm formation and dissection in apolipoprotein E/endothelial nitric oxide synthase deficient mice but not in apolipoprotein E knockout mice. Our results from human aortic tissue also show an association between enos expression and bicuspid aortic valve anatomy. Mean enos levels in patients with BAV were significantly lower compared with patients with tricuspid valve anatomy. There was, however, a marked variability in enos levels and relevant overlap of the levels between the two groups. Bicuspid anatomy per se did not predict low levels of enos. The most interesting finding was a significant inverse correlation between aortic diameters and enos expression in BAV, which was not found in tricuspid aortic valves. These indicate that also in humans there is a relationship between enos and the development of aneurysmal dilatation. The present results, however, should be interpreted carefully in view of the limitations of the study. One limitation is the current sample size. A higher number of samples will be required to confirm the current findings. Because only a small segment of ascending aorta was examined, no conclusions can be drawn regarding regional differences in enos levels in the ascending aorta. While it remains uncertain how high enos levels were in other parts of the proximal aorta (such as the aortic root), removal of more specimens is only possible in patients with extensive aneurysms requiring complete replacement of the proximal aorta. This is not possible in patients with normal-sized aortas operated upon for valve replacement only, thus any investigation going into more depth would present a skewed picture. In order to minimize the possible influence of regional differences we analyzed biopsies of similar size, always taken from the same region (ie, slightly above the sinotubular junction on the right lateral circumference of the aorta). The enos may be measured directly on protein basis or indirectly using polymerase chain reaction. Due to limited availability of material (aortic wall) we had to decide on one method to detect enos, which means to prepare either mrna or protein. Because mrna levels do not take into consideration that proteins can be stabilized or destabilized, mrna might be misleading. Therefore, measurement of enos levels by determining the specific protein appears to be the more appropriate method. Finally, comorbidities and respective medication may influence enos expression. Because enos deficiency is associated with hypertension, both by itself [17] and in the presence of apolipoprotein E deficiency [13], it cannot be fully excluded that hypertension contributes to the increased extent of aortic aneurysm development in patients with bicuspid aortic valves. Indeed, in the present study the patients with bicuspid aortic valves with arterial hypertension showed a significantly lower enos level compared with patients with normal blood pressure. It appears unlikely that aneurysm formation is only a consequence of hypertension caused by enos deficiency, because the prevalence of hypertension was lower in that group. This is in agreement with a publication by Chen and colleagues [18], who found that hypertension does not account for the accelerated atherosclerosis and development of aneurysms in male apolipoprotein E/endothelial nitric oxide synthase double knockout mice. They suggest that enos plays important roles in suppressing atherogenesis separate from blood pressure regulation. Nitric oxide expression is also known to be modulated by low density lipoproteins (LDL). Under stimulated conditions (ie, vascular response to endothelium-dependent vasodilators), reduced NO expression has been found in hypercholesterolemic patients [19]. In our patients, a slightly increased LDL level was found in patients with

5 1294 AICHER ET AL Ann Thorac Surg enos IN BICUSPID AORTIC VALVE DISEASE 2007;83: bicuspid aortic valves compared with patients with tricuspid aortic valves, but no correlation was seen between LDL and enos expression (r 0.316). Lubrano and colleagues [19] described that NO-dependent functions can partly be restored by a lowering of serum cholesterol by statins. In our patients, statin therapy did not influence enos expression. It therefore seems unlikely that in our patients altered LDL levels account for reduced enos expression. Although the therapeutic implications of our findings are still unclear, our data provide additional evidence for the structural abnormality of the aortic wall in BAV. We conclude that BAV is associated with reduced enos expression, which inversely correlates with proximal aortic dilatation. We believe that changes in the aortic wall are mainly caused by altered expression of enos and are not a consequence of the hemodynamic changes. Nevertheless, our clinical study obviously cannot provide evidence for any cause-and-effect relationship. Further studies are necessary to clarify the extent of enos deficiency in patients with bicuspid aortic valves and define its role as a risk factor for aortic dilatation and aorta-related cardiovascular events. We thank Dr Thomas Georg from the Department of Mathematics and Statistics for performing the statistical analysis on the data presented in this article. References 1. Ando M, Okita Y, Morota T, Takamoto S. Thoracic aortic aneurysm associated with congenital bicuspid aortic valve. Cardiovasc Surg 1998;6: Nkomo VT, Enriquez-Sarano M, Ammash NM, et al. Bicuspid aortic valve associated with aortic dilatation: a community-based study. Arterioscler Thromb Vasc Biol 2003;23: Holman E. The obscure physiology of poststenotic dilatation: its relation to the development of aneurysms. J Thorac Surg 1954;28: Gale AN, McKusick VA, Hutchins GM, Gott VL. Familial congenital bicuspid aortic valve: secondary calcific aortic stenosis and aortic aneurysm. Chest 1977;72: Keane MG, Wiegers SE, Plappert T, Pochettino A, Bavaria JE, Sutton MG. Bicuspid aortic valves are associated with aortic dilatation out of proportion to coexistent valvular lesions. Circulation 2000;102:III Nistri S, Sorbo MD, Marin M, Palisi M, Scognamiglio R, Thiene G. Aortic root dilatation in young men with normally functioning bicuspid aortic valves. Heart 1999;82: Pachulski RT, Weinberg AL, Chan KL. Aortic aneurysm in patients with functionally normal or minimally stenotic bicuspid aortic valves. Am J Cardiol 1991;67: Bauer M, Pasic M, Meyer R, et al. Morphometric analysis of aortic media in patients with bicuspid and tricuspid aortic valve. Ann Thorac Surg 2002;74: De Sa M, Moshkovitz Y, Butany J, David TE. Histologic abnormalities of the ascending aorta and pulmonary trunk in patients with bicuspid aortic valve disease: clinical relevance to the Ross procedure. J Thorac Cardiovasc Surg 1999; 118: Lindsay J Jr. Coarctation of the aorta, bicuspid aortic valve and abnormal ascending aortic wall. Am J Cardiol 1988;61: Bonderman D, Gharehbaghi-Schnell E, Wollenek G, Maurer G, Baumgartner H, Lang IM. Mechanisms underlying aortic dilatation in congenital aortic valve malformation. Circulation 1999;99: Lee TC, Zhao YD, Courtman DW, Stewart DJ. Abnormal aortic valve development in mice lacking endothelial nitric oxide synthase. Circulation 2000;101: Kuhlencordt PJ, Gyurko R, Han F, et al. Accelerated atherosclerosis, aortic aneurysm formation and ischemic heart disease in apolipoprotein E/endothelial nitric oxide synthase double-knockout mice. Circulation 2001;104: Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976;72: Ward C. Clinical significance of the bicuspid aortic valve. Heart 2000;83: Edwards WD, Leaf DS, Edwards JE. Dissecting aortic aneurysm associated with congenital bicuspid aortic valve. Circulation 1978;57: Huang PL, Huang Z, Mashimo H, et al. Hypertension in mice lacking the gene for endothelial nitric oxide synthase. Nature 1995;377(6546): Chen J, Kuhlencordt PJ, Astern J, Gyurko R, Huang PL. Hypertension does not account for the accelerated atherosclerosis and development of aneurysms in male apolipoprotein E/endothelial nitric oxide synthase double knockout mice. Circulation 2001;104: Lubrano V, Vassalle C, Blandizzi C, et al. The effect of lipoproteins on endothelial nitric oxide synthase is modulated by lipoperoxides. Eur J Clin Invest 2003;33: