Eplerenone Prevents Salt-Induced Vascular Remodeling and Cardiac Fibrosis in Stroke-Prone Spontaneously Hypertensive Rats
We examined the effect of different levels of salt intake on the role of aldosterone on cardiac and vascular changes in salt-loaded stroke-prone spontaneously hypertensive rats (SHRSP). Eleven-week-old SHRSP were fed high-salt (4.2% NaCl), normal-salt (0.28%), or low-salt (0.03%) diets with or without eplerenone (100 mg/kg per day, in food) for 5 weeks. A group of high-salt SHRSP was also treated with hydralazine (25 mg/kg per day). Blood pressure increased more in high-salt rats than in other groups (P<0.001). Eplerenone prevented further blood pressure rise in salt-loaded rats, with little effect on control and low-salt SHRSP. Increased media-to-lumen ratio of mesenteric resistance arteries induced by salt (P<0.01) was prevented by eplerenone (P<0.01). Maximal acetylcholine-induced vasodilation was impaired under salt loading (P<0.01), but improved under eplerenone (P<0.01). Eplerenone prevented (P<0.01) increased heart weight and left and right ventricular collagen deposition induced by high salt. Blood pressure lowering by hydralazine in high-salt SHRSP did not influence endothelial function or left ventricular collagen. Our study demonstrates salt-dependency of aldosterone effects on severity of hypertension, endothelial dysfunction, and cardiac and vascular remodeling in SHRSP. These effects were attenuated by eplerenone, particularly in the salt-loaded state, underlining the pathophysiological role of aldosterone in salt-sensitive hypertension.
Aldosterone is recognized as a target for therapeutic intervention in cardiac failure1 and hypertension.2,3 Beneficial effects of aldosterone antagonism on end-organ damage have also been studied in animal models. In aldosterone-infused rats, spironolactone or the more specific antagonist eplerenone, prevented left ventricular inflammation and fibrosis,4–6 renal inflammation and proteinuria,7 stiffening of the carotid artery,8 resistance artery remodeling, endothelial dysfunction, and activation of NADPH oxidase.9,10 In stroke-prone spontaneously hypertensive rats (SHRSP), a model of genetic hypertension, spironolactone reduced renal damage and proteinuria.11,12
Most studies using mineralocorticoid receptor antagonism in aldosterone-infused rats4–8 or in SHRSP were performed after salt loading.11,12 We questioned whether salt induces end-organ damage in SHRSP through mechanisms that involve mineralocorticoids. Aldosterone suppression may be impaired in salt-loaded SHRSP both with regard to plasma aldosterone concentrations13 and tissue specific aldosterone synthase activity,14,15 suggesting that aldosterone could play a role in the aggravation by salt of end-organ damage in SHRSP.
In the present study, we tested the hypothesis that salt-induced vascular and cardiac remodeling in SHRSP depends, at least in part, on aldosterone-mediated actions by examining the effects of mineralocorticoid antagonism in SHRSP receiving different levels of salt in their diet.
The study was approved by the Animal Care Committee of the Clinical Research Institute of Montreal and was conducted in accordance with the recommendations of the Canadian Council of Animal Care. Male SHRSP, aged 9 to 11 weeks, were divided into 6 groups, n=7/group plus 4 additional rats per group that were only used for blood pressure (BP) monitoring by radiotelemetry. They received either low-salt (0.03% NaCl), normal-salt (0.28%), or high-salt (4.2%) diets with or without eplerenone (100 mg/kg per day, in food; Pharmacia/Pfizer) for 5 weeks. In a second set of experiments, rats received a high-salt diet with or without hydralazine (25 mg/kg per day, in food, Sigma; n=5 per group plus 3 additional for telemetry). A third set of rats compared SHRSP and normotensive WKY, both on a normal salt diet, to assess whether plasma aldosterone levels and markers of oxidative excess are altered in SHRSP (n=6 per group). 24-hour sodium excretion was measured in urine collected in metabolic cages at the end of the experiment. Radiotransmitters (Data Sciences) were implanted 12 days before the end of the experiment, and BP was recorded for the last 7 days every 5 minutes for 10 seconds. At the end of the experiment, the mesenteric vasculature was dissected, and one segment was used for preparation of small arteries. The rest of the vascular bed and a sample of the apex of the heart were used for measurement of NADPH oxidase activity. A mid cross-section of the heart was fixed in 4% paraformaldehyde.
Preparation and Study of Small Arteries
A third-order branch of the mesenteric artery was isolated and mounted on a pressurized myopraph as described previously.16 After precontraction with 10−5 mol/L norepinephrine, endothelium-dependent and independent relaxation was assessed with acetylcholine (10−9–10−4 mol/L) and sodium nitroprusside (10−9–10−3 mol/L), respectively. To study oxidative excess, vessels were pretreated with tempol (10−3 mol/L; 4-hydroxy-2,2,6,6-tetramethyl piperidinoxyl), superoxide dismutase (80 U/mL), and catalase (100 U/mL). Vascular morphology and mechanics were studied as previously described.16 At the end of the experiment, vessels were fixed for histological analysis.
Tissue sections of mesenteric arteries and hearts were stained with Sirius red F3BA (0.5% in saturated aqueous picric acid, Aldrich Chemical Company) for assessment of collagen, as previously described.4
Measurement of NADPH Oxidase Activity
Activity of NADPH oxidase was measured in a lucigenin luminescence assay as previously described.17
Measurement of Plasma Thiobarbituric Acid-Reacting Substances and Plasma Aldosterone Concentration
Plasma thiobarbituric acid-reacting substances (TBARS) were measured colorimetrically.18 Plasma aldosterone concentration was measured using a radioimmunoassay kit (ICN Diagnostics) according to the manufacturer’s instructions.
For telemetric BP, the means of all readings every hour were calculated. For 7 days the means of the same hours of the day were pooled. Data are presented as mean±SEM. Groups were compared using Student t test, 1-way ANOVA, 2-way ANOVA, or ANOVA for repeated measurements as appropriate. Post-hoc testing was performed using Bonferroni (2-way ANOVA) or Newman-Keuls (1-way and repeated measures) test. P<0.05 was significant.
Blood Pressure, Heart Weight, and Plasma Aldosterone Concentration
After 4 weeks of salt-loading, BP increased significantly in SHRSP (P<0.001; Figure 1). Treatment with eplerenone blunted (P<0.001) the rise in systolic and diastolic BP in salt-treated SHRSP, with greater effect on systolic BP. There were no important differences between low- and normal-salt diets with and without eplerenone. The typical day-night BP rhythm was present in all groups, with highest values at approximately 5:00 am, at which time the animals are most active. Hydralazine reduced systolic and diastolic blood pressure in salt-loaded SHRSP (Figure 1).
Heart:body weight ratio was significantly increased in salt-loaded SHRSP (Table 1). Eplerenone prevented (P<0.05) its increase and was most effective in animals on a high-salt diet. Hydralazine significantly reduced relative heart weight to a similar extent as eplerenone in salt-loaded SHRSP (Table 2).
|Parameter||Low Salt||Low Salt+Eplerenone||Normal Salt||Normal Salt+ Eplerenone||High Salt||High Salt+ Eplerenone||Salt P||Eplerenone P||Interaction P|
|Values are mean±SEM. BW indicates body weight; HW/BW, relative heart weight; PAC, plasma aldosterone concentration; CSA, cross-sectional area; Lv, left ventricle; Mes. Art., mesenteric arteries; NADPH Ox. Act., NADPH oxidase activity. TBARS are expressed as malonyl dialdehyde equivalents. NADPH oxidase activity is expressed as counts/min per mg dry tissue weight.|
|Bonferroni post-hoc testing: *P<0.001, †P<0.01, ‡P<0.05 vs same salt loading without eplerenone.|
|Collagen media, % of area||9.0±0.7||7.3±0.5||9.1±0.6||7.7±0.6||10.1±0,9||8.2±0.4||NS||<0.01||NS|
|Interstitial collagen, % of area|
|NADPH Ox. Act.|
|Parameter||High Salt||High Salt+Hydralazine||P|
|Values are mean±SEM. BW indicates body weight; HW/BW, relative heart weight; CSA, cross-sectional area.|
|Interstitial cardiac collagen content, collagen area/total area|
Twenty-four-hour sodium excretion increased according to salt intake (data not shown), with no differences between eplerenone-treated and untreated animals. Plasma aldosterone concentration decreased (P<0.0001) with increased salt intake (Table 1). Eplerenone treatment resulted in significantly (P<0.0001) increased plasma aldosterone. SHRSP have significantly higher plasma aldosterone levels compared with normotensive WKY rats (484±51 versus 272±66 pg/mL; P<0.05), indicating upregulation of the aldosterone system in SHRSP.
Morphology, Mechanical Properties, and Endothelial Function of Mesenteric Arteries
Media thickness of mesenteric resistance arteries increased significantly (P<0.05) with salt-loading in SHRSP. Eplerenone prevented the increase of media thickness (P<0.05), with the most pronounced effects in animals on high-salt diet (Table 1). Lumen diameter tended to be smaller in salt-loaded animals and higher under eplerenone treatment. Media-to-lumen ratio was increased (P<0.01) with greater salt intake. This was prevented (P<0.01) by eplerenone (Figure 2). Treatment with eplerenone was most effective in animals on high-salt diet. These results paralleled changes in media thickness and relative heart weight. Hydralazine significantly reduced media-to-lumen ratio (Table 2).
Strain was significantly reduced with increased salt (P<0.001 between the different salt treatments without eplerenone, Figure 3A). Eplerenone treatment prevented decrease of strain significantly (P<0.001) in salt-loaded SHRSP, but not in rats receiving normal or low-salt diet. There were minor changes in the stress-strain relationship with slight displacement to the left with greater salt intake (Figure 3B) and increase in the slope of the elastic modulus-stress relationship (Figure 3C), both prevented by eplerenone, but which, however, did not reach statistical significance. Thus the significant changes in vessel mechanics may have been mostly because of vessel geometry.
There were no significant differences between groups in endothelium-independent vasorelaxation curves after sodium nitroprusside stimulation (data not shown). Endothelium-dependent vasorelaxation was similar in all groups (full curves not shown). However, maximal vasodilation to acetylcholine (10−4 mol/L) was significantly less (P<0.01) in rats on high-salt diet without eplerenone compared with other groups (Figure 4A). Pretreatment of vessels with antioxidants partially restored maximal vasodilatory responses (Figure 4B). Hydralazine did not improve endothelial dysfunction in salt-loaded SHRSP (Table 2).
Collagen Content of Mesenteric Arteries and the Left Ventricle
3Collagen content in the media of mesenteric resistance arteries tended to increase with salt loading (Table 1). Eplerenone treatment significantly reduced (P<0.01) media collagen content.
In the left ventricle, subepicardial (P<0.01), midmyocardial (P<0.001), and subendocardial (P<0.0001), as well as right ventricular (P<0.0001) interstitial collagen, increased significantly with salt loading (Table 1). These effects were prevented by eplerenone (P<0.001). Changes in collagen content paralleled changes in relative heart weight. Hydralazine had a small but significant inhibitory effect on collagen content in the right ventricle (Table 2).
TBARS and NADPH Oxidase Activity
Plasma TBARS, an indirect marker of global oxidative excess, were not different in the groups with or without eplerenone (Table 1). When comparing SHRSP and WKY on normal salt diet, the mean plasma TBAR levels tended to be higher in SHRSP (3.29±0.64 versus 2.20±0.32 nmol/mL malonyl dialdehyde equivalents, respectively), but statistical significance was not achieved. NADPH oxidase activity in mesenteric arteries and heart was unaltered (Table 1).
In the present study, we show for the first time that in SHRSP receiving different levels of salt intake, mineralocorticoid antagonism with eplerenone prevented vascular remodeling and cardiac fibrosis, particularly in the rats exposed to the high-salt diet. Studies from our laboratory showed a beneficial effect of spironolactone on vascular hypertrophic remodeling and vascular function of mesenteric resistance arteries in aldosterone-infused rats.9,10 Eplerenone reduced carotid artery stiffness in aldosterone-infused salt-loaded rats,8 associated with differences in fibronectin, not in collagen or elastin density. To our knowledge, this is the first time that the effect of the more selective mineralocorticoid antagonist eplerenone has been investigated in resistance arteries. In high-salt conditions, the increased stiffness of SHRSP vessels was blunted, at least in part because of changes in vessel geometry, as shown by the lack of statistical significance in stress-strain and elastic modulus-stress relationships. However, these curves result from complex mathematical transformations with significant loss of statistical power. Because these findings were paralleled by the results of vascular collagen content, we cannot rule out some effect of salt on vascular collagen content and geometry-independent stiffness.
Endothelial dysfunction in salt-loaded SHRSP was partly restored by antioxidant pretreatment, suggesting that reactive oxygen species may contribute to the impaired endothelial function. Previous studies demonstrated oxidative excess in SHRSP.19 Plasma TBARS, an indirect marker of global oxidative excess tended to be higher in salt-loaded SHRSP. Vascular and cardiac NADPH oxidase activity, a major source of reactive oxygen species in cardiovascular tissues, was however not significantly altered in salt-loaded SHRSP. Oxidative excess thus probably plays only a minor role in salt-induced changes observed in our study. Sources other than NADPH oxidase may contribute to reactive oxygen species in this experimental paradigm.
The protective effect of eplerenone on cardiac fibrosis is in agreement with other studies. In humans, eplerenone reduced death in heart failure1 and left ventricular hypertrophy in hypertension.20 In dogs with left ventricular dysfunction, eplerenone attenuated left ventricular remodeling.21 In contrast, a recent study showed severe heart failure and cardiac fibrosis in a knock-down model of the mineralocorticoid receptor, suggesting that mineralocorticoid receptors may be cardioprotective.22 However, in that study, cardiomyocyte-specific mineralocorticoid receptors were downregulated, without consideration of the noncardiac aldosterone system. Accordingly, it is difficult to reconcile those findings with data obtained in the present study.
There are few studies simultaneously addressing the effect of different levels of salt intake on aldosterone-induced end-organ damage in hypertension. We demonstrate significant effects of salt on target organ damage, prevented partly by eplerenone. Spironolactone prevented high salt-induced cardiac hypertrophy and fibrosis in normotensive Wistar rats23 and media hypertrophy in salt-fed normotensive Sprague-Dawley rats.24 Although in the present study blood pressure lowering with hydralazine improved media to lumen ratio and cardiac hypertrophy, it did not influence endothelial function or left ventricular fibrosis. Thus, it is likely that eplerenone had additional blood pressure-independent beneficial actions on target organs. This is supported by the reduction by eplerenone of cardiac collagen content in the right ventricle similar to that in the left ventricle. Studies on human25 and rat26 cardiac myofibroblasts support a direct and blood pressure-independent effect of aldosterone on cardiac fibrosis. A small reduction in right ventricular collagen content by hydralazine may be related to the profound blood pressure-lowering effect of this drug.
One mechanism of salt-induced hypertension is impaired pressure natriuresis, in which the renin-angiotensin-aldosterone system has been implicated.27 Blockade of these effects could contribute to beneficial actions of mineralocorticoid antagonism. However, 24-hour sodium excretion was unchanged by eplerenone, although there may have been initial differences in natriuresis under treatment. Other studies have shown impaired suppression or a rise of plasma aldosterone in SHRSP under high-salt, stroke permissive diets. In the present study, plasma aldosterone levels decreased with salt loading. Enhanced local cardiac aldosterone production has been reported in SHRSP compared with WKY rats15 and under salt treatment in SHRSP14 and WKY rats.28 Thus, aldosterone associated end-organ damage may be attributed to direct effects of aldosterone on cardiovascular tissues. Under low-salt conditions aldosterone levels are much higher with less end-organ damage and less effect of aldosterone antagonism. This underscores the degree to which high salt sensitizes tissues to aldosterone-induced cardiovascular injury.
Findings from the present study indicate a pivotal role of the level of salt intake on the effect of aldosterone on cardiac and vascular changes in hypertension. Our data provide important insights into the pathophysiological significance of salt in the role of mineralocorticoids such as aldosterone, and the therapeutic potential for mineralocorticoid receptor blockade for protection of the vasculature and the heart in salt-sensitive hypertension.
This study was supported by grants 13570, 44108, and a Group Grant to the Multidisciplinary Research Group on Hypertension, all from the Canadian Institutes of Health Research, and a research grant from Pharmacia/Pfizer. D.H.E. was supported by a grant from the Deutsche Forschungsgemeinschaft. M.I. was supported by a grant from the Québec Hypertension Society. C.S. was supported by a grant from the Italian Society of Hypertension. The authors are grateful to Suzanne Diébold, Manon Laprise, and André Turgeon for their excellent technical assistance.
- 1 Pitt B, Remme W, Zannad F, Neaton J, Martinez F, Roniker B, Bittman R, Hurley S, Kleinman J, Gatlin M. Eplerenone Post-Acute Myocardial Infarction Heart Failure Efficacy and Survival Study Investigators. Eplerenone, a selective aldosterone blocker, in patients with left ventricular dysfunction after myocardial infarction. N Engl J Med. 2003; 348: 1309–1321.CrossrefMedlineGoogle Scholar
- 2 White WB, Duprez D, St Hillaire R, Krause S, Roniker B, Kuse-Hamilton J, Weber MA. Effects of the selective aldosterone blocker eplerenone versus the calcium antagonist amlodipine in systolic hypertension. Hypertension. 2003; 41: 1021–1026.LinkGoogle Scholar
- 3 Flack JM, Oparil S, Pratt JH, Roniker B, Garthwaite S, Kleiman JH, Yang Y, Krause SL, Workman D, Saunders E. Efficacy and tolerability of eplerenone and losartan in hypertensive black and white patients. J Am Coll Cardiol. 2003; 41: 1148–1155.CrossrefMedlineGoogle Scholar
- 4 Park JB, Schiffrin EL. Cardiac and vascular fibrosis and hypertrophy in aldosterone-infused rats: role of endothelin-1. Am J Hypertens. 2002; 15: 164–169.CrossrefMedlineGoogle Scholar
- 5 Sun Y, Zhang J, Lu L, Chen SS, Quinn MT, Weber KT. Aldosterone-induced inflammation in the rat heart: role of oxidative stress. Am J Pathol. 2002; 161: 1773–1781.CrossrefMedlineGoogle Scholar
- 6 Rocha R, Rudolph AE, Frierdich GE, Nachowiak DA, Kekec BK, Blomme EA, McMahon E, Delyani J. Aldosterone induces a vascular inflammatory phenotype in the rat heart. Am J Physiol Heart Circ Physiol. 2002; 283: H1802–H1810.CrossrefMedlineGoogle Scholar
- 7 Blasi ER, Rocha R, Rudolph AE, Blomme EA, Polly ML, McMahon EG. Aldosterone/salt induces renal inflammation and fibrosis in hypertensive rats. Kidney Int. 2003; 63: 1791–1800.CrossrefMedlineGoogle Scholar
- 8 Lacolley P, Labat C, Pujol A, Delcayre C, Benetos A, Safar M. Increased carotid wall elastic modulus and fibronectin in aldosterone-salt-treated rats: effects of eplerenone. Circulation. 2002; 106: 2848–2853.LinkGoogle Scholar
- 9 Virdis A, Neves MF, Amiri F, Viel E, Touyz RM, Schiffrin EL. Spironolactone improves angiotensin-induced vascular changes and oxidative stress. Hypertension. 2002; 40: 504–510.LinkGoogle Scholar
- 10 Pu Q, Neves MF, Virdis A, Touyz RM, Schiffrin EL. Endothelin antagonism on aldosterone-induced oxidative stress and vascular remodeling. Hypertension. 2003; 42: 49–55.LinkGoogle Scholar
- 11 Griffin KA, Abu-Amarah I, Picken M, Bidani AK. Renoprotection by ACE inhibition or aldosterone blockade is blood pressure-dependent. Hypertension. 2003; 41: 201–206.LinkGoogle Scholar
- 12 Rocha R, Chander PN, Khanna K, Zuckerman A, Stier CT, Jr. Mineralocorticoid blockade reduces vascular injury in stroke-prone hypertensive rats. Hypertension. 1998; 31: 451–458.CrossrefMedlineGoogle Scholar
- 13 Volpe M, Rubattu S, Ganten D, Enea I, Russo R, Lembo G, Mirante A, Condorelli G, Trimarco B. Dietary salt excess unmasks blunted aldosterone suppression and sodium retention in the stroke-prone phenotype of the spontaneously hypertensive rat. J Hypertens. 1993; 11: 793–798.CrossrefMedlineGoogle Scholar
- 14 Takeda Y, Yoneda T, Demura M, Furukawa K, Miyamori I, Mabuchi H. Effects of high sodium intake on cardiovascular aldosterone synthesis in stroke-prone spontaneously hypertensive rats. J Hypertens. 2001; 19: 635–639.CrossrefMedlineGoogle Scholar
- 15 Takeda Y, Yoneda T, Demura M, Miyamori I, Mabuchi H. Cardiac aldosterone production in genetically hypertensive rats. Hypertension. 2000; 36: 495–500.CrossrefMedlineGoogle Scholar
- 16 Iglarz M, Touyz RM, Amiri F, Lavoie MF, Diep QN, Schiffrin EL. Effect of peroxisome proliferator-activated receptor-alpha and -gamma activators on vascular remodeling in endothelin-dependent hypertension. Arterioscler Thromb Vasc Biol. 2003; 23: 45–51.LinkGoogle Scholar
- 17 Chen X, Touyz RM, Park JB, Schiffrin EL. Antioxidant effects of vitamins C and E are associated with altered activation of vascular NADPH oxidase and superoxide dismutase in stroke-prone SHR. Hypertension. 2001; 38: 606–611.CrossrefMedlineGoogle Scholar
- 18 Schistermann EF, Faraggi D, Browne R, Freudenheim J, Dorn J, Muti P, Armstrong D, Reiser B, Trevisan M. TBARS and cardiovascular disease in a population-based sample. J Cardiovasc Risk. 2001; 8: 219–225.MedlineGoogle Scholar
- 19 Park JB, Touyz RM, Chen X, Schiffrin EL. Chronic treatment with a superoxide dismutase mimetic prevents vascular remodeling and progression of hypertension in salt-loaded stroke-prone spontaneously hypertensive rats. Am J Hypertens. 2002; 15: 78–84.CrossrefMedlineGoogle Scholar
- 20 Pitt B, Reichek N, Willenbrock R, Zannad F, Phillips RA, Roniker B, Kleiman J, Krause S, Burns D, Williams GH. Effects of eplerenone, enalapril, and eplerenone/enalapril in patients with essential hypertension and left ventricular hypertrophy. Circulation. 2003; 108: 1831–1838.LinkGoogle Scholar
- 21 Suzuki G, Morita H, Mishima T, Sharov VG, Todor A, Tanhehco EJ, Rudolph AE, McMahon EG, Goldstein S, Sabbah HN. Effects of long-term monotherapy with eplerenone, a novel aldosterone blocker, on progression of left ventricular dysfunction and remodeling in dogs with heart failure. Circulation. 2002; 106: 2967–2972.LinkGoogle Scholar
- 22 Beggah AT, Escoubet B, Puttini S, Cailmail S, Delage V, Ouvrard-Pascaud A, Bocchi B, Peuchmaur M, Delcayre C, Farman N, Jaisser F. Reversible cardiac fibrosis and heart failure induced by conditional expression of an antisense mRNA of the mineralocorticoid receptor in cardiomyocytes. PNAS. 2002; 99: 7160–7165.CrossrefMedlineGoogle Scholar
- 23 Lal A, Veinot JP, Leenen FHH. Prevention of high salt diet-induced cardiac hypertrophy and fibrosis by spironolactone. Am J Hypertens. 2003; 16: 319–323.CrossrefMedlineGoogle Scholar
- 24 Simon G, Jaeckel M, Illyes G. Development of structural vascular changes in salt-fed rats. Am J Hypertens. 2003; 16: 488–493.CrossrefMedlineGoogle Scholar
- 25 Neumann S, Huse K, Semrau R, Diegeler A, Gebhardt R, Buniatian GH, Scholz GH. Aldosterone and D-glucose stimulate the proliferation of human cardiac myofibroblasts in vitro. Hypertension. 2002; 39: 756–760.CrossrefMedlineGoogle Scholar
- 26 Stockand JD, Meszaros JG. Aldosterone stimulates proliferation of cardiac fibroblasts by activating Ki-RasA and MAPL1/2 signaling. Am J Physiol Heart Circ Physiol. 2003; 284: H176–184.CrossrefMedlineGoogle Scholar
- 27 Hall JE. The kidney, hypertension, and obesity. Hypertension. 2002; 41: 625–633.Google Scholar
- 28 Takeda Y, Yoneda T, Demura M, Miyamori I, Mabuchi H. Sodium-induced cardiac aldosterone synthesis causes cardiac hypertrophy. Endocrinology. 2000; 141: 1901–1904.CrossrefMedlineGoogle Scholar