Endothelial Dysfunction and Cardiorenal Injury in Experimental Salt-Sensitive Hypertension
Effects of Antihypertensive Therapy
Background Pharmacological control of hypertension has contributed to a significant decrease in cardiovascular morbidity and mortality, although the beneficial effect on cardiac and renal diseases has been far more modest than the reduction in stroke. The endothelium plays a crucial homeostatic role in the regulation of vascular tone thrombogenesis and vascular remodeling. We studied the relationship between endothelial dysfunction and cardiorenal injury in hypertensive rats and evaluated the effects of two classes of antihypertensive agents commonly used in the clinical setting, a diuretic (DIU) and an ACE inhibitor (CEI).
Methods and Results Dahl salt-sensitive rats (DS) given high dietary salt (4% NaCl) developed hypertension (systolic blood pressure [SBP], 218±9 versus 147±3 mm Hg in DS given 0.5% NaCl; P<.001), which was associated with impaired endothelium-dependent relaxations (EDRs) in aortic rings (ED50, 5.44±.18 versus 7.51±.10; P<.05) and mesenteric vessels (area under the curve, 299±11 versus 217±11 arbitrary units; P<.05). Hypertensive DS also demonstrated depressed nitric oxide synthase activity in the aorta (0.76±.15 versus 2.83±.17 nmol · min−1 · g protein−1; P<.05), left ventricular hypertrophy (0.43±.02 versus 0.29±.02 g ventricular weight/100 g body weight; P<.05), glomerular injury (histological injury score: 151±8 versus 11±2; P<.05), and increased urinary protein excretion (95±21 versus 25±5 mg/24 hours; P<.05). Treatment of DS rats with the CEI perindopril (4.56 mg · kg−1 · d−1) did not affect SBP (225±6 mm Hg) but modestly improved EDR (ED50: 6.07±.37; P<.05 versus hypertensive DS) as well as proteinuria and glomerular histology. Addition of the DIU indapamide (1.44 mg · kg−1 · d−1) normalized SBP (151±2 mm Hg; P<.05), EDR (ED50, 7.33±.08; P<.05), left ventricular hypertrophy (0.27±.01 g/100 g body weight; P<.05), and proteinuria (31±4 mg/24 hours; P<.05) and prevented glomerular injury (injury score: 30±2; P<.05). Monotherapy with DIU reduced SBP (175±3 mm Hg; P<.05) and normalized EDR and left ventricular hypertrophy but did not provide effective renal protection. In hypertensive DS, impaired EDR and left ventricular hypertrophy were positively correlated with SBP. In addition, we found a significant correlation between cardiac hypertrophy and endothelial dysfunction. Indeed, a hierarchical regression analysis revealed that impaired aortic EDR, and therefore decreased aortic compliance, positively contributed to left ventricular hypertrophy in addition to but independently of SBP [F(2, 37)=6.29; P=.004].
Conclusions These studies suggest a dissociation of the endothelial, cardiac, and renal effects of antihypertensive therapy in hypertension and may explain the variable success of antihypertensive regimens in treating hypertension while preventing cardiac and renal damage.
During the last two decades, pharmacological control of hypertension has contributed to a significant decrease in cardiovascular morbidity and mortality, although the beneficial effect on cardiac and renal diseases has been far more modest than the reduction in stroke.1 2 3 Several explanations for these discrepancies in target-organ protection have been offered, including differences in target-organ sensitivity to the salutary effects of antihypertensive therapy.4 In the case of the heart, failure to effectively reduce coronary artery disease in hypertensive patients has been attributed, at least in part, to inadequate treatment of concomitant risk factors such as dyslipidemia and LVH.5 6 In the case of the kidney, clinical and experimental evidence suggests that to prevent or arrest renal failure, control of systemic hypertension must be accompanied by control of glomerular hypertension.4 7 Thus, an ideal antihypertensive regimen, in addition to normalizing systemic blood pressure, must provide specific cardiac and renal protection.
Clinical studies of populations of patients defined as “salt sensitive” revealed that these patients manifest LVH, microalbuminuria, and paradoxical suppression of atrial natriuretic peptide in response to salt loading.8 9 10 Studies from our laboratory, as well as others, have demonstrated that hypertensive DS rats have many features similar to the aforementioned patients and also develop LVH, renal injury, and endothelial dysfunction.11 12 13 14 Thus, DS rats may be a useful model in which to study target-organ damage in the setting of hypertension.
In the current study, we studied in DS rats the individual and combined effects of two classes of antihypertensive drugs commonly used in the clinical setting: DIUs and CEIs. The effects of these agents on blood pressure, EDR in aorta and mesenteric artery, aortic NOS activity, LVH, and renal injury were evaluated. In addition, we investigated the relationship between blood pressure, impaired endothelial function, and LVH.
The studies reported herein suggest that (1) there is a correlation between impaired aortic endothelial function and LVH that is independent of blood pressure, and (2) levels of systemic blood pressure achieved with antihypertensive therapy may not correlate with the degree of protection provided by these agents in different target organs. These findings may explain the heretofore disparate success achieved clinically in arresting or preventing cardiac and renal hypertensive injuries.
Male, 6-week-old (190 to 210 g body weight [BW]) DS rats from the Brookhaven strain were purchased from Harlan Sprague Dawley (Indianapolis, Ind). Indapamide and perindopril were gifts from Servier Laboratories (Paris, France). [14C]l-arginine was purchased from Amersham International. Other chemicals used were purchased from Sigma Chemical Co. Dowex resin was purchased from Bio-Rad Laboratories.
Five groups of DS rats were used in this study. All rats had free access to water and were housed five per cage in facilities accredited by the American Association for Accreditation of Laboratory Animal Care. The experimental protocols were approved by the Institutional Animal Care and Use Committee.
All drugs were given for 8 weeks by daily gavage in the morning. One percent hydroxyethylcellulose in distilled water was used as a vehicle for all drugs. The normotensive control group was fed a normal salt diet (0.5% NaCl) and given vehicle (VEH-LS; n=8) by gavage, whereas the hypertensive control group was fed a high-salt diet (4% NaCl) and was also given vehicle by gavage (VEH-HS; n=8). In pilot studies, we administered perindopril, a CEI, in variable doses to DS rats receiving high dietary salt and failed to reduce blood pressure even with a high dose (4.5 mg · kg BW−1 · d−1). Therefore, similar to the clinical situation, we added a DIU, indapamide (1.44 mg · kg BW−1 · d−1), and succeeded in normalizing blood pressure in the DS rats on a high-salt diet. The remaining three groups were all fed the 4% NaCl diet. The DIU group (DIU-HS; n=18) was given indapamide in a dose of 1.44 mg · kg BW−1 · d−1. The CEI group (CEI-HS; n=8) was given perindopril 4.5 mg · kg BW−1 · d−1, and the DIU+CEI group (DIU+CEI-HS; n=14) was given indapamide 1.44 mg · kg BW−1 · d−1 and perindopril 4.5 mg · kg BW−1 · d−1. The experimental diets had protein, mineral, and fat contents similar to the standard rat diet.
SBP was measured every 2 weeks in conscious rats by the tail-cuff method (Physiograph MK IV, Narco Biosystems).15
At the end of 8 weeks, the animals were weighed, anesthetized with sodium pentobarbital (30 mg/kg BW IP), and killed by exsanguination. Aortas and mesenteric vessels were obtained for organ-chamber experiments and perfusion studies, respectively (see below). Hearts from all rats were obtained for determining HW/BW ratios, an index of LVH. Kidneys were removed for morphological examination.15 Thoracic aortas were also obtained for the determination of NOS activity.
Measurement of Urinary Excretion of Protein and Endothelin-1
After 8 weeks of treatment, 24-hour urine excretion was obtained in metabolic cages as previously described.15 Urine creatinine was measured with a creatinine analyzer (Beckman Instruments). Urine protein concentration was determined with the use of a BioRad protein assay kit16 and expressed in milligrams per 24 hours (UproV). Urinary ET-1 was measured by radioimmunoassay (Amersham International)17 and expressed in nanograms per milligram of creatinine.
Kidney tissue was fixed in buffered formalin and stained with hematoxylin and eosin and periodic acid-Schiff. Histological evaluation of the kidney was performed by one of the authors (L.R.) without knowledge of the experimental group from which the section came. The glomerular lesions were evaluated by use of a semiquantitative scoring method described previously.15 Briefly, the severity of the glomerular lesion was graded from 0 to 4+ according to the percentage of glomerular involvement. The GIS was calculated by summing the products of the severity score and the percentage of glomeruli displaying the same degree of severity.
Vascular responses in aortas were tested in organ chambers according to techniques previously described.18 Briefly, aortic rings were mounted horizontally between two stirrups in organ chambers filled with 25 mL oxygenated modified Krebs-Ringer bicarbonate solution (in mmol/L: NaCl 118.6, KCl 4.8, CaCl2 2.5, MgSO4 1.2, K2PO4 1.2, NaHCO3 25.1, glucose 10.1, and EDTA 0.026) at 37°C. One stirrup was connected to an anchor and the other to a force transducer (Gould Statham UTC2) for recording of isometric tension. Relaxations to ACh (10−9 to 10−4 mol/L) and SNP (10−9 to 10−4 mol/L) were studied in rings precontracted to 70% of the maximal contraction to l-norepinephrine obtained in each individual ring. Contraction responses to ET-1 (10−10 to 10−7 mol/L) were also studied in aortic rings from all groups of rats. Contraction to ET-1 was expressed as percentage of the response to 100 mmol/L KCl in the same ring.
Vascular Responses in Mesenteric Arteries
Rats were anesthetized by injection of sodium pentobarbital (30 mg/kg BW IP), and the abdominal wall was opened. Mesenteric arteries were prepared for perfusion according to the method described by McGregor.19 The superior mesenteric artery was cannulated with an 18- or 19-gauge needle via the abdominal aorta and perfused without ischemia. The entire preparation was placed into a water-jacketed container maintained at 37°C and perfused with modified Krebs-Ringer bicarbonate solution. This solution was maintained at 37°C and gassed with 5% CO2 and 95% O2 to obtain a pH of 7.4. Phenylephrine (10−6 mol/L) was added to the Krebs-Ringer solution to maintain vascular tone and perfusion pressure. The preparation was perfused at a constant flow rate of 5 mL/min with a roller pump, and the perfusate was allowed to drain freely and discarded. Perfusion pressure was recorded with a pressure transducer connected to a polygraph. Baseline perfusion pressure was maintained at 90 to 100 mm Hg. The preparation was allowed to equilibrate for 40 minutes, and then responses to ACh (10−9 to 10−4 mol/L) were studied. Responses to ACh were expressed as percent decrease of perfusion pressure.
Determination of NOS Activity in Aortas
Thoracic aortas were excised from rats, frozen in liquid nitrogen, and stored at −80°C until use. Aortas were homogenized in 3.5 vol of ice-cold buffer solution containing 50 mmol/L Tris-HCl (pH 7.4), 0.1% mercaptoethanol, 0.1 mmol/L EDTA, 0.1 mmol/L EGTA, 2 μmol/L leupeptin, 1 μmol/L pepstatin A, 1 mmol/L PMSF, and 1% Triton-X100 by use of an Omni TH homogenizer (Omni International). The homogenates were centrifuged at 20 000g for 45 minutes, and the supernatants were used for measuring NOS activity and protein concentration.
The oxidation of l-arginine by NOS was monitored by the conversion of [14C]l-arginine to [14C]l-citrulline at 37°C, which is l-nitro arginine inhibitible.20 Briefly, 60 μL of the supernatant was added to 200 μL of assay buffer containing 50 mmol/L KH2PO4, 1 mmol/L MgCl2, 1 mmol/L CaCl2, 50 mmol/L valine, 1 mmol/L l-citrulline, 20 μmol/L l-arginine, 1 mmol/L dithiothreitol, 2 mmol/L NADPH, 3 μmol/L BH4, 3 μmol/L FAD, 3 μmol/L FMN, and 0.5 μCi/mL [14C]l-arginine HCl. The mixture of supernatant and assay buffer solution was incubated for 30 minutes at 37°C in the presence or absence of either EGTA (1 mmol/L) or EGTA (1 mmol/L) plus l-nitro arginine (1 mmol/L) to determine the level of the Ca2+-dependent and -independent activities. After incubation, the reaction was stopped by adding 500 μL of ice-cold solution containing 20 mmol/L HEPES, 2 mmol/L EDTA, and 2 mmol/L EGTA. Incubated mixture was loaded onto 1-mL columns of Dowex resin (Na form). Columns were then eluted with 500 μL of distilled water. The amounts of [14C]l-citrulline were determined by a liquid scintillation counter. NOS activity was expressed as nanomoles of [14C]l-citrulline formed per gram of protein per minute.
Calculations and Statistical Analysis
Relaxations in aortic rings were expressed as a percent decrease in tension.18 Contractions to ET-1 in aortic rings were expressed as a percentage of the contractile response to 100 mmol/L KCl of each individual aortic ring, performed at the beginning of each experiment. For statistical analysis, the concentration of the agonist (expressed as −log [mol/L]) evoking 50% relaxation (ED50) and maximal relaxation and contraction were calculated.18 A hierarchical regression analysis was used to assess the independent contribution of blood pressure and aortic endothelial function (ED50 to ACh) to LVH. In mesenteric vessels, the AUC of the response to ACh was also calculated, and the results were expressed in arbitrary units (higher values of AUC correlate with an impaired response to ACh). Results of the experiments are given as mean±SE. Statistical analysis was performed by one-way ANOVA between groups (StatView 512+, BrainPower Inc). Significance was assumed at P<.05.
Blood Pressure, BW, and HW
SBP, BW, and HW/BW ratio are shown in Table 1⇓. The time course of SBP progression in each experimental group is presented in Fig 1⇓. The SBP in the VEH-HS group was significantly higher than that in the VEH-LS group (P<.001). Treatment with the CEI did not affect SBP. Treatment with the DIU significantly reduced but did not normalize SBP. However, combination therapy with CEI+DIU normalized SBP.
HW/BW ratio in the hypertensive VEH-HS group was higher than that in the normotensive VEH-LS group (Table 1⇑). Monotherapy with CEI failed to normalize either SBP or HW in DS rats fed a high-salt diet. Although the DIU did not normalize SBP, HW was normalized by this therapy. Combination therapy with CEI and DIU normalized both SBP and HW.
Endothelium-Dependent and -Independent Vascular Relaxations in Aortas
Responses to ACh (10−9 to 10−4 mol/L) were studied in aortic rings from all groups of rats. Relaxation in response to ACh was significantly reduced in aortic rings from VEH-HS rats compared with that from VEH-LS rats (P<.01; Fig 2A⇓). Treatment with DIU or DIU+CEI normalized the response to ACh, whereas monotherapy with CEI did not normalize but partially improved EDRs to ACh. The ED50 of the response to ACh in aortic rings was calculated (Table 2⇓). Treatment with DIU or DIU+CEI significantly improved the ED50 from 5.58±0.28 (VEH-HS) to 7.10±0.19 (DIU-HS) and 7.22±0.12 (DIU+CEI-HS), respectively (P<.05). Addition of indomethacin (10−5 mol/L) to the organ bath did not improve the response to ACh of aortic rings of hypertensive VEH-HS rats or CEI-HS rats (data not shown). This suggests that endothelium-contracting factors derived from cyclooxygenase were not responsible for the impaired vascular relaxation.12
Correlations Among Blood Pressure, LVH, and Endothelial Function
We found a significant positive correlation between SBP and HW/BW ratio (Fig 3A⇓) and between SBP and ED50 of aortic rings to ACh (Fig 3B⇓). Furthermore, we also found a significant correlation between ED50 to ACh and HW/BW ratio (Fig 3C⇓). Therefore, a hierarchical regression analysis was used to assess the independent contribution of aortic endothelial function (ED50) to the presence of LVH. The aortic ED50 to ACh variable in combination with the SBP variable accounted for significantly more HW/BW ratio variance (multiple R2=0.60) than did the SBP predictor alone [R2=0.47; F(2,37)=6.29; P=.004]. This suggests that independent of the actual levels of blood pressure, impaired aortic endothelial function fostered LVH.
Vascular Responses in Mesenteric Arteries
Responses to ACh (10−9 to 10−5 mol/L) were investigated in the isolated perfused mesenteric arteries from all groups (Fig 4⇓; Table 2⇑). Relaxation responses to ACh in the VEH-HS group were significantly attenuated compared with the VEH-LS group (P<.01). Similar to what we observed in aortas, in mesenteric arteries DIU and DIU+CEI significantly improved EDRs to ACh and decreased the AUC (P<.01), whereas CEI did not. However, in contrast to aortas, CEI failed to improve relaxation to ACh in mesenteric arteries. Moreover, as shown in Fig 4⇓, the maximal relaxation to ACh in mesenteric vessels was normalized by DIU and by DIU+CEI but not by CEI.
Responses to ET-1 in Aortas
Aortic contraction to ET-1 (10−10 to 10−7 mol/L) was evaluated in all groups of rats. The responses to ET-1 in aortic rings from hypertensive VEH-HS rats were attenuated compared with normotensive VEH-LS rats (Fig 5⇓). Maximal contraction in VEH-HS was significantly lower than that in VEH-LS (P<.01; Table 2⇑). CEI did not affect the response of aortic rings to ET-1; however, DIU as well as DIU+CEI normalized the response to ET-1 (P<.01).
As shown in Fig 6⇓, the VEH-HS group had a more than threefold higher UproV than the VEH-LS group (95±21 versus 26±6 mg/24 hours, respectively; P<.05). CEI reduced UproV despite its lack of effect on blood pressure. DIU also decreased UproV significantly. However, UproV in the DIU-HS group and the CEI-HS group was not reduced to values similar to those obtained in the normotensive VEH-LS group, whereas treatment with DIU+CEI was effective in normalizing UproV.
GIS was measured in kidneys from all experimental groups (Fig 7⇓). VEH-HS rats demonstrated a markedly increased GIS compared with VEH-LS rats (151±8 versus 11±2, respectively; P<.05). However, whereas treatment with DIU or CEI reduced GIS by ≈50%, the combination of DIU and CEI decreased GIS by 80% (to 30±2).
Urinary ET-1 Excretion
Fig 8A⇓ shows ET-1 excretion in urine from all groups of rats. Compared with the VEH-LS group, the hypertensive VEH-HS rats showed an increased excretion of ET-1 (2.08±0.14 versus 1.49±0.08 ng/mg of creatinine, respectively; P<.05). Treatment with DIU or DIU+CEI reduced ET-1 excretion significantly, whereas the CEI did not decrease ET-1 excretion in hypertensive DS rats. Moreover, urinary excretion of ET-1 was significantly correlated with UproV (r=.69, P<.01; Fig 8B⇓) and with the glomerular injury observed histologically.
NOS Activity in Aortas
Ca2+-dependent NOS activity was determined in aortas of all groups of rats (Fig 9⇓). NOS was significantly lower in hypertensive VEH-HS rats than in normotensive VEH-LS rats (0.755±.150 versus 2.829±.169 nmol · min−1 · g protein−1; P<.05). DIU or DIU+CEI normalized NOS activity; however, CEI did not affect NOS activity. No significant Ca2+-independent NOS was found in the aortas studied.
Hypertensive DS rats developed impaired EDRs in aortas and mesenteric vessels, and this finding was accompanied by LVH and glomerular injury. Vascular relaxation to SNP, an endothelium-independent vasorelaxant, remained normal. The combination of the CEI (perindopril) and the DIU (indapamide) effectively controlled blood pressure as well as prevented endothelial dysfunction, LVH, and glomerular injury.
CEI monotherapy failed to normalize blood pressure. This finding was not unexpected because clinically and experimentally, high dietary salt promotes resistance to the antihypertensive effect of CEI.21 22 23 In these hypertensive DS rats, perindopril did not reduce LVH. CEI provided partial protection to the aortic but not the mesenteric endothelium.
Monotherapy with indapamide resulted in striking normalization of LVH as well as EDR in aortas and mesenteric vessels while only partially reducing SBP. The mechanisms involved in the beneficial cardiovascular effects of indapamide-independent blood pressure reduction are unclear at the present and may be due to a unique pharmacological effect of this agent.24 Furthermore, recent experimental and clinical studies have made similar observations.25 26
Our studies also revealed that in hypertensive rats with impaired EDR, vascular contraction in response to endothelin is also attenuated. Moreover, normalization of EDR with antihypertensive therapy consisting of either DIU or DIU+CEI was accompanied by a restoration of ET-1–mediated vascular contraction. Studies in other models of experimental hypertension have shown increased ET-1 synthesis in the vascular wall,27 impaired vascular contraction in response to ET-1,28 29 and partial improvement of blood pressure with endothelin receptor antagonists.30 Because NO has been reported to downregulate ET-1 synthesis,31 the increased vascular ET-1 in these hypertensive models may be linked to decreased NO bioactivity, whereas impaired contraction in response to exogenous ET-1 may be due to downregulation of endothelin receptors. In the present study, the fact that hypertensive therapy improved both relaxation to ACh and constriction to ET-1 supports this hypothesis.
The current study confirms that there is a positive correlation between blood pressure and LVH (Fig 3A⇑) and a negative correlation between blood pressure and EDR (Fig 3B⇑). Most importantly, we found that independent of blood pressure, there is a positive correlation between impaired aortic EDR and LVH (Fig 3C⇑). The endothelium plays an important homeostatic role in the maintenance of normal vascular compliance.32 33 Previous clinical and experimental studies34 35 36 have demonstrated that in hypertension, although pressure and volume overload are the major determinants of LVH, other contributing factors are important. On the basis of our findings, we suggest that aortic endothelial function may also be a contributing factor to the development of LVH. This hypothesis is supported by observations made in hypertensive patients by use of invasive intravascular recordings, which have demonstrated that the NO donor SNP completely normalizes vascular compliance independently of blood pressure effects.37
We found that NOS activity in the aortas of hypertensive DS rats was lower than in normotensive DS rats. Hayakawa et al previously reported that renal NO release is reduced in hypertensive DS rats.11 The decreased NOS activity shown in aortas of hypertensive DS rats is compatible with the studies reported previously and can explain the impaired EDR in this hypertensive model.11 12 13 14 Moreover, SHR develop significantly less LVH than DS rats in response to hypertension of similar severity. Because SHR but not DS rats upregulate cardiac NOS activity, the reduced LVH observed in SHR has been attributed, at least in part, to the inhibitory effects of cardiac NO on vascular growth.14 In the present studies, treatment with DIU or DIU+CEI improved aortic cNOS activity; however, CEI did not. We speculate that normalization of NOS activity can explain, at least in part, the improvement in EDR by these antihypertensive treatments. These observations suggests that the beneficial effects of CEI on vascular endothelium38 39 are less evident in salt-sensitive hypertensive animals receiving high dietary salt.
It has been reported that hypertensive Dahl rats, due to deficient autoregulation of preglomerular vascular resistances, develop high glomerular capillary pressure and subsequent glomerular injury early in the course of hypertension.7 15 In the present studies, neither monotherapy with CEI nor the DIU indapamide was accompanied by effective renal protection. Only combination therapy with CEI and DIU arrested glomerular injury. The modest reduction in glomerular injury and proteinuria observed with the CEI may be due to the known effects of these agents in reducing glomerular capillary pressure independently of their effects on SBP.7
In summary, we have studied vascular function in conduit as well as resistance vessels and concomitant structural changes in the heart and the kidney in a genetic model of salt-dependent hypertension. We have demonstrated a close link between aortic endothelial function and LVH that is independent of blood pressure levels. Furthermore, these findings may shed light on the pathophysiological mechanisms by which the beneficial effect of antihypertensive agents may vary in different target organs (eg, prevention of LVH without concomitant renal protection). In addition, our studies emphasize that the salutary effects of these agents depend on hemodynamic as well as nonhemodynamic factors. These studies may provide insights for the treatment of hypertension in patients, with the goal of preventing end-organ injury.
Selected Abbreviations and Acronyms
|AUC||=||area under the curve (in arbitrary units)|
|CEI-HS||=||converting-enzyme inhibitor plus high salt|
|DIU+CEI-HS||=||diuretic, converting-enzyme inhibitor plus high salt|
|DIU-HS||=||diuretic plus high salt|
|ED50||=||concentration of agonist (−log[mol/L]) that evokes 50% relaxation|
|GIS||=||glomerular injury score|
|LVH||=||left ventricular hypertrophy|
|NOS||=||nitric oxide synthase|
|SBP||=||systolic blood pressure|
|SHR||=||spontaneously hypertensive rats|
|UproV||=||urinary protein excretion|
|VEH-HS||=||vehicle plus high salt|
|VEH-LS||=||vehicle plus low salt|
This study was supported by a grant from the Department of Veterans Affairs and Servier. The authors gratefully acknowledge Janeth Guerra and Dawn Holms for their technical assistance, Dr Jonathan Tolins for reviewing the manuscript, and Barb Devereaux for secretarial support.
- Received January 13, 1997.
- Revision received May 12, 1997.
- Accepted May 20, 1997.
- Copyright © 1997 by American Heart Association
MacMahon S, Peto R, Cutler J, Collins R, Sorlie P, Neaton J, Abbott R, Godwin J, Dyer A, Stamler J. Blood pressure, stroke, and coronary heart disease, I: prolonged differences in blood pressure—prospective observational studies corrected for the regression dilution bias. Lancet. 1990;335:765-774.
Collins R, Peto R, MacMahon S, Hebert P, Fiebach NH, Eberlein KA, Godwin J, Qizilbash N, Taylor JO, Hennekens CH. Blood pressure, stroke, and coronary heart disease, II: short-term reductions in blood pressure—overview of randomized drug trials in their epidemiological context. Lancet. 1990;335:827-838.
Levy D, Garrison RJ, Savage DD, Kannel WB, Castelli WP. Left ventricular mass and incidence of coronary heart disease in an elderly cohort: the Framingham Heart Study. Ann Intern Med. 1989;110:101-107.
Campese VM, Tawadrous M, Bigazzi R, Bianchi S, Mann AS, Oparil S, Raij L. Salt intake and plasma atrial natriuretic peptide and nitric oxide in hypertension. Hypertension. 1996;28:335-340.
Schmieder RE, Messerli FH, Garavaglia GE, Nunez BD. Dietary salt intake: a determinant of cardiac involvement in essential hypertension. Circulation. 1988;78:951-956.
Bigazzi R, Bianchi S, Baldari D, Sgherri G, Baldari G, Campese VM. Microalbuminuria in salt-sensitive patients: a marker for renal and cardiovascular risk factors. Hypertension. 1994;23:195-199.
Hayakawa H, Hirata Y, Suzuki E, Sugimoto T, Matsuoka H, Kikuchi K, Nagano T, Hirobe M, Sugimoto T. Mechanisms for altered endothelium-dependent vasorelaxation in isolated kidneys from experimental hypertensive rats. Am J Physiol. 1993;264:H1535-H1541.
Luscher TF, Raij L, Vanhoutte PM. Endothelium-dependent vascular responses in normotensive and hypertensive Dahl rats. Hypertension. 1987;9:157-163.
Raij L, Luscher TF, Vanhoutte PM. High potassium diet augments endothelium-dependent relaxations in the Dahl rat. Hypertension. 1988;12:562-567.
Hayakawa H, Raij L. The link among nitric oxide synthase activity, endothelial function and aortic and ventricular hypertrophy in hypertension. Hypertension. 1997;29(pt 2):235-241.
Raij L, Nagy J, Coffee K, DeMaster EG. Hypercholesterolemia promotes endothelial dysfunction in vitamin E- and selenium-deficient rats. Hypertension. 1993;22:56-61.
McGregor D. Effect of sympathetic nerve stimulation on the vasoconstrictor responses in perfused mesenteric blood vessels of the rat. J Physiol. 1965;177:21-30.
Singer DR, Markandu ND, Cappuccio FP, Miller MA, Sagnella GA, MacGregor GA. Reduction of salt intake during converting enzyme inhibitor treatment compared with addition of a thiazide. Hypertension. 1995;25:1042-1044.
Peeler TC, Baker KM, Esmurdoc CF, Chernin MI. Angiotensin converting enzyme inhibition in Dahl salt-sensitive rats. Mol Cell Biol. 1991;104:45-50.
Uehara Y, Shirahase H, Nagata T, Ishimitsu T, Morishita S, Osumi S, Matsuoka H, Sugimoto T. Radical scavengers of indapamide in prostacyclin synthesis in rat smooth muscle cell. Hypertension. 1990;15:216-224.
Lariviere R, Thibault G, Schiffrin EL. Increased endothelin-1 content in blood vessels of deoxycorticosterone acetate–salt hypertensive but not in spontaneously hypertensive rats. Hypertension. 1993;21:294-300.
Dohi Y, Luscher TF. Endothelin in hypertensive resistance arteries: intraluminal and extraluminal dysfunction. Hypertension. 1991;18:543-549.
Deng LY, Schiffrin EL. Effects of endothelin on resistance arteries of DOCA-salt hypertensive rats. Am J Physiol. 1992;262:H1782-H1787.
Li JS, Lariviere R, Schiffrin EL. Effect of a nonselective endothelin antagonist on vascular remodeling in deoxycorticosterone acetate–salt hypertensive rats: evidence for a role of endothelin in vascular hypertrophy. Hypertension. 1994;24:183-188.
Ryan US, Zhong R, Hayes BA, Visner G, Sauther ML. Regulation of endothelin-1 expression in normal and transfected endothelial cells. J Cardiovasc Pharmacol. 1993;22:S38-S41.
Stefas L, Levy BI. Effects of saralasin on arterial compliance in normotensive and hypertensive rats: role of endothelium. Hypertension. 1991;18:II-30-II-36.
Benetos A, Bouaziz H, Albaladejo P, Guez D, Safar ME. Carotid artery mechanical properties of Dahl salt-sensitive rats. Hypertension. 1995;25:272-277.
Boutouyrie P, Laurent S, Girerd X, Benetos A, Lacolley P, Abergel E, Safar M. Common carotid artery stiffness and patterns of left ventricular hypertrophy in hypertensive patients. Hypertension. 1995;25:651-659.
Frohlich ED. Overview of hemodynamic and non-hemodynamic factors associated with left ventricular hypertrophy. J Mol Cell Cardiol. 1989;21:3-10.
Ting CT, Brin KP, Lin SJ, Wang SP, Chang MS, Chiang BN, Yin FC. Arterial hemodynamics in human hypertension. J Clin Invest. 1986;78:1462-1471.
Clozel M, Kuhn H, Hefti F. Effects of angiotensin converting enzyme inhibitors and of hydralazine on endothelial function in hypertensive rats. Hypertension. 1990;16:532-540.
Shultz PJ, Raij L. Effects of antihypertensive agents on endothelium-dependent and endothelium-independent relaxations. Br J Clin Pharmacol. 1989;28:151S-157S.