Plasma C-Terminal Pro–Endothelin-1 is Associated with Left Ventricular Mass Index and Aortic Root Diameter in African American Adults with Hypertension

Abstract

Background

Endothelin-1 (ET-1), a circulating vasoactive peptide with a potent vasoconstricting and mitogenic properties, may contribute to target-organ damage in hypertension. A novel assay for the measurement of C-terminal pro-endothelin-1 (CT-proET-1), a stable fragment of ET-1 precursor, has been recently validated as a reliable measure of ET-1 activity in the plasma. We tested the hypothesis that in African American adults with hypertension, plasma CT-proET-1 is associated with left ventricular (LV) mass and aortic root diameter.

Methods

Participants included 1041 African Americans (65±9 y, 72% women) with hypertension. Plasma CT-proET-1 was measured by an immunoluminometric assay. LV Mass and aortic root diameter were measured according to the American Society of Echocardiography guidelines, and LV mass was indexed by height to the power 2.7 (LVMi). Multivariable regressions analyses (using generalized estimating equations) were employed to assess whether plasma CT-proET-1 was associated with LVMi and aortic root diameter independent of potential confounding variables.

Results

Plasma CT-proET-1 was significantly correlated with LVMi (r =0.21, P <0.0001) and aortic root diameter (r =0.09, P= 0.004). In separate multivariable regression models that adjusted for age, sex, body mass index, total and high density lipoprotein cholesterol, smoking history, diabetes, previous history of myocardial infarction or stroke, and blood pressure-lowering medication and statin use, log CT-proET-1 was significantly associated with greater LVMi (P =0.001) and greater aortic root diameter (P =0.006).

Conclusion

CT-proET-1 is independently associated with LVMi and aortic root diameter and may be a marker of target-organ damage in African Americans adults with hypertension.

Background

Endothelin (ET)-1 is a 21-amino acid vasoactive peptide, derived mainly from vascular endothelial cells ¹ and also produced by various tissues including vascular smooth muscle cells ¹, cardiac myocytes,² and fibroblasts.³ The initial product of human ET-1 gene is a 212 amino-acid peptide, pre-proET-1, which is converted to pro-ET-1 after removal of a short amino acid sequence.² Pro-ET-1 is cleaved to a 38-amino acid precursor, big ET-1, which is finally modified by the ET-1-converting enzyme to produce ET-1.² ET-1 acts by binding to two distinct receptors (ET_A and ET_B), found on vascular smooth muscle cells and the myocardium.² ET-1 mediates vasoconstriction by binding to ET_A receptors on the underlying smooth muscle cells, augments sympathetic activity, increases sodium retention, activates the renin-angiotensin-aldosterone system, and induces myocardial hypertrophy and hypertrophic remodeling of small arteries in hypertension.² ET-1 acts in an autocrine and/or paracrine manner and most of the ET-1 produced by the endothelium is secreted towards the adjacent vascular smooth muscle cells,⁴ which makes plasma ET-1 levels a reflection of the spillover to the circulation rather than actual ET-1 production.

Reliable quantification of plasma ET-1 has been impeded by its short plasma half-life (1–2 min),⁵ its non-specific binding to plasma proteins,⁶ and its intermediate clearance by binding to receptors during pulmonary passage.⁷ Recently, an immunoluminometric assay for the measurement of C-terminal fragment of prepro-ET-1 (CT-pro-ET-1), a stable nonfunctional peptide that is produced in equimolar amount to ET-1, has been validated for the measurement of ET-1 activity in the plasma and may provide a better assessment of the actual release of ET-1 gene products.⁸

Early detection of target-organ damage in patients with hypertension is of utmost interest in clinical practice. Although clinical variables such as age, diabetes, and hyperlipidemia are established risk factors for acceleration of arteriosclerosis and consequent target-organ damage in patients with hypertension, much of the interindividual variation in measures of target-organ damage cannot be accounted for based solely on known clinical risk factors. New biomarkers of etiologic pathway of target-organ damage may explain some of the remaining interindividual variability and may facilitate early detection of target-organ damage in hypertension. African Americans have a disproportionate burden of hypertension, and African American hypertensives are more prone to adverse cardiovascular events and hypertension-related target-organ damage compared to hypertensives from other ethnic groups.⁹ Left ventricular (LV) hypertrophy and aortic root dilation are well-recognized manifestations of target-organ damage in patients with hypertension and are associated with increased cardiovascular morbidity and mortality.^10-12 Multiple studies have implicated ET-1 in the development and maintenance of hypertension, particularly in African Americans.^13-15 However, little is known about its role in mediating target-organ damage in hypertensives. We therefore investigated whether plasma levels of CT-proET-1 were associated with echocardiographic measures of target organ damage: LV mass and aortic root diameter, in African Americans with hypertension.

Methods

This study was part of the Proteomic Markers of Arteriosclerosis Study which is investigating the association of multiple markers in various etiologic pathways of vascular diseases with several phenotypes of arteriosclerosis.¹⁶ Participants were members of sibships that were initially enrolled in Jackson MS and Rochester MN between June 1996 and August 2000 in the Genetic Epidemiology Network of Arteriopathy of the Family Blood Pressure Program Study, a multicenter, community-based study that aims to identify genetic variants influencing blood pressure (BP) levels and the development of target-organ damage due to hypertension.¹⁷ The Jackson MS cohort of Atherosclerosis Risk in Communities study that had originally been a probability sample of persons with driver's licenses, was used to ascertain African- American sibships. The sampling frame of Rochester cohort was the Mayo Clinic diagnostic index and medical record linkage system of the Rochester Epidemiology Project. It was used to identify non-Hispanic white residents of Olmsted County, MN with a diagnosis of essential hypertension made before age 60. Participants were enrolled if two or more members of a sibship had hypertension. The only exclusionary criterion at enrollment was the presence of a secondary cause of hypertension (such as documented renal artery stenosis or advanced renal insufficiency) in the index sibs. Echocardiographic evaluation was only performed on the African American cohort and because of that, the Rochester, MN cohort was not included in our analyses. Study protocols were approved by the human studies review board at the University of Mississippi Medical Center and Mayo Clinic. Written informed consent was obtained from each participant. The present study included 1041 African Americans who had hypertension. Normotensive subjects were excluded.

Height was measured by stadiometer, weight was measured by electronic balance, and body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters. Resting systolic BP and diastolic BP were measured by random zero sphygmomanometer (Hawskley and Sons, London, UK) after participants had rested for at least 10 min in the supine position. Three measures at least 2 min apart were taken and the average of the second and third measurements was used. The diagnosis of hypertension was established based on BP levels measured at the study visit (≥140/90 mm Hg) or a prior diagnosis of hypertension and current treatment with antihypertensive medications. Diabetes mellitus was defined as a fasting serum glucose ≥126 mg/dL, random glucose ≥200 mg/dL, history of physician diagnosis of diabetes mellitus, or use of insulin or oral hypoglycemic agent. ‘Ever’ smoking was defined as having smoked >100 cigarettes. Information about the use of medications was obtained from the participants at the time of the study visit. Each prescription drug that was recorded at the study visit was assigned a code number corresponding to the first six digits of the Medi-Span Generic Product Identifier (Medi-Span Inc., Indianapolis, IN). This number identifies pharmacologically equivalent drug products and was used to categorize agents with a similar therapeutic action. BP-lowering medications were classified as diuretics, beta-blockers, calcium-channel blockers, or renin-angiotensin-aldosterone system inhibitors.

Blood was drawn after an overnight fast. Serum creatinine, glucose, total cholesterol, high-density lipoprotein (HDL) cholesterol, and triglyceride concentrations were measured by standard methods on a Hitachi 911 Chemistry analyzer (Roche Diagnostics, Indianapolis, IN). Estimated glomerular filtration rate (eGFR) was calculated using the abbreviated equation from the Modification of Diet in Renal Disease Study using standardized serum creatinine value.¹⁸

Plasma levels of CT-proET-1

Plasma was collected at the time of blood sampling in plastic vials containing ethylenediaminetetraacetic acid. Samples were placed on ice and then centrifuged at 3000 × g and frozen at -80°C until assayed. Plasma CT-proET-1 was measured by a novel commercial assay in the chemiluminescence/coated tube-format (BRAHMS AG, Hennigsdorf/Berlin, Germany) as previously described.⁸ Briefly, tubes were coated with a purified sheep polyclonal antibody raised against a peptide representing amino acids 168-181 of pre–proET-1. A purified sheep polyclonal antibody raised against a peptide representing amino acids 200-212 of pre–proET-1 was labeled with MACN acridinium–NHS ester (InVent GmbH, Braunschweig, Germany) and used as tracer. Dilutions of a peptide representing amino acids 169-212 of pre–proET-1 in normal horse serum served as standards. The immunoassay was performed by incubating 50 μL of samples/standards and 200 μL of tracer in coated tubes for 2 hours at room temperature. Tubes were washed 4 times with 1 mL of LIA wash solution (BRAHMS AG), and bound chemiluminescence was measured using a LB952T luminometer (Berthold, Bad Wildbad, Germany). The assay (mean reference range, 44.3 ± 10.6 pmol/L) has an analytical detection limit of 0.4 pmol/L. Since plasma CT-proET-1 values <10 pmol/L are likely to reflect degradation, 59 participants with levels <10 pmol/L were excluded from our analyses.

Echocardiographic Measures

Participants were examined with 2-dimensional and Doppler echocardiography on an Acuson 128XP/10c (Acuson, Malvern, PA) using 2.5-, 3.5-, and 5.0-MHz transducers while in the lateral decubitus position. A standardized protocol was followed under which the parasternal acoustic long and short-axis views were used to record on videotape 10 or more consecutive beats of 2-dimensional and M-mode recordings. LV internal dimension, LV septal wall thickness, and LV posterior wall thickness were measured at end-diastole and end-diastole according to American Society of Echocardiography recommendations, averaging 3 or more cardiac cycles. Correct orientations of planes for imaging and Doppler recordings were verified using standardized protocols. LV mass was derived from the simplified cubed equation formula ¹⁹ and indexed by height to the power 2.7 (LVMi) to normalize heart size to body size. LV hypertrophy was defined as an LVMi of ≥51 g/m^2.7 in men and women. Relative wall thickness was calculated as the sum of LV septal and posterior wall thickness at the end-diastole divided by LV internal dimension at the end-diastole. The LV ejection fraction was derived semi quantitatively by the primary cardiologist using a modified Quinones technique and visual assessment of the LV apex. aortic root diameter was measure in parasternal 2-dimentional long-axis view, using color flow mapping to help delineate tissue-blood interfaces when necessary, at sinuses of Valsalva level by M-mode tracings as the maximal distance between the two leading edges of the anterior and posterior aortic root walls at the end-diastole.²⁰

Statistical Methods

Statistical analyses were carried out using SAS v 9.1 (SAS Institute, Cary, NC). Because of sibships in the sample, we used generalized estimating equations to account for intrafamilial correlations. Continuous variables were expressed as mean±SD. Categorical variables were expressed as number (percentage). Values for plasma CT-proET-1, LVMi, and eGFR were log transformed to minimize skewness. Spearman correlation coefficients were used to assess the correlation between CT-proET-1 levels with LVMi and aortic root diameter. We used multivariable regression analyses to identify variables independently associated with plasma CT-proET-1 from among age, sex, BMI, systolic BP, diastolic BP, total and HDL cholesterol, smoking history, diabetes, previous history of myocardial infarction or stroke, and medication (BP-lowering and statin) use. Initially, we constructed separate univariable linear regression models to assess whether plasma CT-proET-1 was associated with LVMi, LV posterior wall thickness, LV septal wall thickness, LV relative wall thickness, and aortic root diameter. Next, we constructed separate multivariable linear regression models that adjusted for age and sex, and finally in fully adjusted models that included conventional cardiovascular risk factors (age, sex, BMI, systolic BP, diastolic BP, total and HDL cholesterol, smoking history, diabetes, previous history of myocardial infarction or stroke, and medication (BP-lowering and statin) use. Statistical significance was determined at 2-sided P-value of <0.05.

Results

The characteristics of the study participants are shown in Tables 1 and 2. Of the 1041 participants, 729 (72.5%) were women and 346 (33.5%) had diabetes. The mean plasma CT-proET-1 was 50.68 pmol/L and the median was 49.05 pmol/L. Plasma levels of CT-proET-1 increased with greater age, higher systolic BP, lower eGFR, lower HDL, and were higher in subjects with history of myocardial infarction and/or stroke (Table 2). A significant proportion (19%) of the interindividual variation in plasma levels of CT-proET-1 could be explained by these clinical variables.

Table 1.

Characteristics of participants.

Characteristics	(n = 1041)
	Mean±SD or n (%)
Mean age, years	64.8±8.6
Sex, women, n (%)	754 (72.4%)
BMI, kg/m2	32.0±6.7
Mean systolic BP, mm Hg	142±21
Mean diastolic BP, mm Hg	80±11
Mean pulse pressure, mm Hg	62±18
Heart rate, per minute	67±11
Mean total cholesterol, mmol/L	5.2±1.1
Mean HDL cholesterol, mmol/L	1.5±0.5
Mean eGFR, mL/s	1.63±0.52
Ever-smoker, n (%)	422 (41%)
Diabetes mellitus, n (%)	346 (33.2%)
Previous history of MI or stroke, n (%)	140(13.5%)
Beta-blocker, n (%)	216 (20.8%)
Calcium-channel blocker, n (%)	368 (35.4%)
Diuretic, n (%)	589 (56.6%)
RAAS inhibitor, n (%)	512 (49.2%)
Statin, n (%)	226 (21.7%)
Aspirin, n (%)	374 (35.9%)
Mean CT-proET-1, pmol/L	50.7±20.5

^*Continuous variables are presented as means ± standard deviation, whereas categorical variables are presented as counts and percentages.

^‡BMI, body mass index; BP, blood pressure; CT-proET-1, C-terminal proendothelin-1; eGFR, estimated glomerular filtration rate; HDL, high- density lipoprotein; LVMi, left ventricular mass index; RAAS, renin-angiotensin-aldosterone system.

Table 2.

Echocardiography Characteristics.

Characteristics	(n = 1013)
	Mean±SD or n (%)
Mean LV mass, g	166.8±48.7
Mean LVMi, g/m2.7	41.2±11.4
LV hypertrophy, n (%)	157 (15.5%)
Mean LV dimension-diastole, mm	142±21
Mean LV septum wall thickness, mm	9.1±13
Mean LV posterior wall thickness, mm	8.4±12
Mean LV relative wall thickness	0.32±0.05
Mean left atrial diameter, mm	36.3±5.3
Mean LV ejection fraction, %	60±9
Mean aortic root diameter, mm	33.6±4.1

^*Continuous variables are presented as means ± standard deviation, whereas categorical variables are presented as counts and percentages.

LV, left ventricle; LVMi, left ventricular mass index.

Plasma CT-proET-1 was significantly correlated with LVMi (r =0.21, P <0.0001) (Figure 1). After adjustment for age and sex, log CT-proET-1 was significantly associated with log LVMi (P <0.001), LV posterior wall thickness (P =0.002), LV septal wall thickness (P =0.003), and the presence of LV hypertrophy (P =0.002) but not LV relative wall thickness, left atrial diameter, or LV ejection fraction (analyses not shown). After additional adjustment for cardiovascular risk factors and medication use, log CT-proET-1 remained significantly associated with greater log LVMi (P =0.002) (Table 4). Other variables associated with greater log LVMi included age, higher BMI, higher systolic BP, history of MI and/or stroke, diabetes, use of beta-blockers and/or calcium channel blockers (Table 4). These variables and log CT-proET-1 explained 24% of interindividual variation of LVMi.

Figure 1.

Correlation of plasma CT-proET-1 levels with aortic root diameter and left ventricular mass index.

Table 4.

Multivariable models for the associations of plasma CT-proET-1 with log LVMi and aortic root diameter.

	Log LVMi		Aortic root diameter
	β±SE	P value	β±SE	P value
Model 1 (unadjusted)	0.11±0.02	<0.001	0.09±0.02	<0.001
Model 2 (age and sex adjusted)	0.11±0.03	<0.001	0.07±0.02	0.005
Model 3 (fully adjusted)‡
Age, year	0.05±0.01	<0.0001	0.06±0.01	<0.0001
Male sex	0.03±0.02	NS	0.43±0.03	<0.0001
BMI, kg/m2	0.01±0.001	<0.0001	0.005±0.002	0.009
Systolic BP, mm Hg	0.05±0.01	<0.0001	-0.002±0.001	0.011
Diastolic BP, mm Hg	-	-	0.007±0.001	<0.0001
Diabetes	0.04±0.02	0.020	-	-
History of MI and/or stroke	0.05±0.02	0.024	-	-
Beta-blocker use	0.06± 0.02	0.002	-	-
Calcium channel blocker use	0.03± 0.01	0.024	-	-
Diuretic use	-	-	0.044±0.02	0.061
Log CT-proET-1	0.05± 0.02	0.002	0.06±0.02	0.008

R² of LVMi model is 0.24 and R2 of aortic root diameter model is 0.29.

^*BMI, body mass index; BP, blood pressure; eGFR, estimated glomerular filtration rate; HDL, high-density lipoprotein; LVMi, left ventricular mass index; MI, myocardial infarction; CT-proET-1, C-terminal proendothelin-1.

^‡Model 3 is adjusted for age, sex, BMI, systolic BP, diastolic BP, eGFR, total cholesterol and HDL cholesterol, smoking history, diabetes, previous history of myocardial infarction or stroke, and the use of medications (BP-lowering and statin).

Plasma CT-proET-1 was significantly correlated with aortic root diameter (r =0.09, P= 0.004) (Figure 1). After adjustment for age and sex, log CT-proET-1 was significantly associated with ARD (P <0.007), and after additional adjustment for cardiovascular risk factors and medication use (P =0.001). Variables independently associated with larger aortic root diameter were age, male sex, higher BMI, higher systolic BP, lower diastolic BP and log CT-proET-1 (Table 3). These variables and log CT-proET-1 explained 29% of interindividual variation of aortic root diameter.

Table 3.

Determinants of plasma CT-proET-1: multivariable regression.

	β±SE	P value
Age, years	0.01±0.002	<.0001
BMI, kg/m2	0.004±0.002	0.079
Systolic BP, per 10 mmHg increase	0.013±0.005	0.0160
Heart rate, per minute	0.0020±0.0011	0.046
HDL cholesterol, mmol/L	-0.002±0.0010	0.010
Log eGFR	-0.44±0.04	<0.0001
History of MI or stroke	0.10±0.04	<0.007

R² of the model is 0.19

BMI, body mass index; BP, blood pressure; eGFR, estimated glomerular filtration rate; HDL, high-density lipoprotein; MI, myocardial infarction.

Discussion

The concept of measuring a nonfunctional, stable peptide derived from a common precursor peptide instead of the problematic measurement of a bioactive, rapidly cleared peptide has been applied with great success for the measurement of natriuretic peptides.^21,22 In this study, we measured plasma levels of CT-proET-1, a nonfunctional, stable fragment of ET-1 precursor in a community-based sample of African Americans with hypertension, and we observed that higher plasma CT-proET-1 levels were associated with greater LVMi and greater aortic root diameter. These associations were independent of age, sex, conventional cardiovascular risk factors, eGFR, and medication use suggesting a potential use of plasma CT-proET-1 as marker of target-organ damage in the setting of hypertension.

The mean/median levels of plasma CT-proET-1 in our study were higher than the previously reported mean/median levels in apparently healthy individuals (44.3 pmol/L/∼45 pmol/L), likely due to the fact that all participants in our study were hypertensive. Consistent with previous reports,^8,23 we found plasma levels of CT-proET-1 to be higher in older subjects and to be similar in men and women. Higher systolic BP, higher heart rate, lower eGFR, lower HDL cholesterol, and history of MI and stroke were additional independent predictors of higher plasma CT-proET-1 levels in African American adults with hypertension (Table 2). Renal filtration is an important determinant of plasma CT-proET-1 levels; consideration of a confounding effect of decreased renal function on any observed association between plasma CT-proET-1 levels and target-organ damage manifestations in hypertensives is pertinent. In our study, however, the observed association between plasma CT-proET-1 and LVMi and aortic root diameter were independent of eGFR.

Plasma ET-1 was reported to be higher in hypertensives with LVH than in hypertensives without LVH.^24-26 The results of studies that evaluated the associations of ET-1 with LVMi are not consistent. While a few small studies (n <60) reported a significant association of plasma CT-proET-1 with LVMi,^24,25,27 other studies (n =50-120) found no significant association ^28-30. In our study, plasma CT-proET-1 was associated with LVMi after adjustment for adjustment for conventional cardiovascular risk factors and medication use (P =0.002). Additionally, after further adjustment for aortic root diameter and LV ejection fraction, plasma CT-proET-1 remained significantly associated with LVMi (P =0.008). Moreover, using LVMi ≥51 gm/m^2.7 to define LV hypertrophy in both sexes, higher plasma CT-proET-1 was associated with LV hypertrophy after multivariable adjustment (P =0.025) (analyses not shown). Consistent with a previous report,²⁷ we found plasma CT-proET-1 to be associated with LV relative wall thickness in a univariate model only. We did not find a significant association with left atrial diameter and LV ejection fraction. Left atrial volume is a more accurate measurement of left atrial size than left atrial diameter but the required images to calculate left atrial volume were not available for this study.

LV hypertrophy is more prevalent in African Americans ⁹ and is associated with increased cardiovascular morbidity and mortality.^10,11 Previous studies suggest that the sympathetic nervous system and the renin-angiotensin aldosterone system contribute to the development and maintenance of LV hypertrophy by inducing myocardial hypertrophy and fibrosis. Multiple animal studies suggest a role of ET-1 in the development of myocardial hypertrophy.^31-33 ET-1 is secreted by cardiac myocytes and fibroblasts in response to mechanical stretching and pressure overload,³³ in turn promoting cardiac myocyte hypertrophy by inducing muscle-specific gene transcripts and variety of immediate early genes,³⁴ and enhancing norepinephrine ³⁵ and angiotensin II-induced cardiac hypertrophy.²

Previous cross-sectional studies demonstrated that aortic root dilation (defined as aortic root diameter >97.5^th percentile predicted by subject's body surface area) is more prevalent in patients with hypertension,³⁶ associated with LV hypertrophy,³⁷ other target-organ damage,³⁷ and adverse cardiovascular events.¹² aortic root diameter has been reported to be mainly determined by age and body size and the impact of BP on aortic root diameter, independent of age and body size, remains controversial.³⁸ We report for the first time a significant association between plasma CT-proET-1 and aortic root diameter. Other variables that were independently associated with larger aortic root diameter were age, male sex, higher BMI, higher systolic BP, lower diastolic BP (Table 3). ET-1 may directly contribute to aortic root dilation by maintaining high BP and indirectly by increasing the expression of matrix metalloproteinase.^39,40

Our findings suggest that plasma level of CT-proET-1 may be a marker of hypertension-related changes in LV mass and aortic root diameter in African American adults with essential hypertension. Although plasma levels of CT-proET-1 were significantly associated with LVMi, its ability to discriminate between the presence and absence of LV hypertrophy was modest. Using only plasma levels of CT-proET-1 to predict the presence of LV hypertrophy, the area under the receiver operating characteristics curve to predict LV hypertrophy was 0.62 (analyses not shown). Strengths of our study include the relatively large sample of African American adults with hypertension included in our study, using uniform protocols including questionnaires, anthropometric and laboratory measurements, and the novel assay used to measure plasma CT-proET-1. Several limitations need to be recognized in our study including that the generalizability of our findings to other racial/ethnic groups is not established. The cross-sectional design of our study precludes interpretations about causality. In addition, plasma CT-proET-1 and echocardiographic measures were measured only once, and variation with time could not be taken into account. However, this would bias the results towards the null and underestimate the true associations. Most participants were on antihypertensive medications and although we corrected for medication use in our analyses, the effect of antihypertensive medications on the observed associations cannot be completely ruled out. However, we did not find a particular class of antihypertensive medications to be associated with plasma CT-proET-1 levels. We adjusted for multiple covariates that may be associated with plasma CT-proET-1 levels or target-organ damage in hypertension. However, the possibility of residual confounding remains.