|Home | About | Journals | Submit | Contact Us | Français|
Hypertension occurs with higher prevalence and morbidity in black Americans compared to other groups. Alterations in the signal transduction pathways of seven-transmembrane spanning receptors are found in hypertensive patients. GRKs play an important role in regulating this receptor signaling. The two most abundantly expressed GRKs in the cardiovascular system are GRK2 and GRK5 and each have unique substrates. Understanding changes in expression may give us insight into activated receptors in the pathophysiological progression of hypertension. In heart failure and Caucasian hypertensives, increased GRK2 expression arises because of neurohormal stimulation of particular receptors. GRK2 subsequently desensitizes specific receptors including β-adrenergic receptors. In blood pressure control, β-adrenergic receptor desensitization could lead to increased blood pressure. GRK2 and GRK5 mRNA were evaluated in lymphocytes of black Americans via quantitative real-time PCR. GRK2 mRNA expression directly correlated with systolic blood pressure and norepinephrine levels. GRK2 was elevated >30% among those with systolic blood pressure equal to or greater than 130mmHg. No significant correlation between GRK5 mRNA expression and blood pressure or catecholamines was observed. Diabetic status, age, sex and body mass index (BMI) were also examined compared to GRK2 expression using univariate and multi-variate analyses. GRK2 protein expression was elevated 2-fold in subjects with higher blood pressure and GRK activity was increased >40%. Our data suggest that GRK2, but not GRK5, is correlated with increasing blood pressure in black Americans. Understanding the receptors stimulated by increased neurohormonal activation may give insight into the pathophysiology of hypertension in this at-risk population.
Nearly 45% of black Americans have high blood pressure (BP)1. The prevalence of hypertension and the severity of cardiovascular morbidity are greater for blacks than other racial groups, but the reasons for these differences are not well understood2. The influence of the sympathetic nervous system on BP is mediated predominantly through catecholamines (epinephrine and norepinephrine) binding to adrenergic receptors. β-adrenergic receptors (βARs) are critical for regulating peripheral resistance and disruption in their signaling can lead to an attenuation of vasodilation and subsequent BP elevation3,4. Although the pathophysiology of high BP is undeniably complex, increased vascular resistance5,6, βAR derangement5,7,8, and increased plasma norepinephrine3,9 are likely involved. Further, norepinephrine levels increase early in the progression of hypertension10 and are inversely correlated with lymphocyte βAR density11 suggesting that understanding the impact of neurohormal activation is important to understanding the development and progression of hypertension.
Constant stimulation of βARs by catecholamines in heart failure and hypertension leads to selective βAR downregulation and an overall attenuation of βAR-mediated adenylyl cyclase activity12–14. G protein-coupled receptor kinases (GRKs) are important serine/threonine kinase regulators of agonist-occupied 7 transmembrane-spanning receptors (7TMRs) including βARs15. GRK2 and GRK5 are the most abundant GRKs in the cardiovascular system. In hypertensive rat models, an increase in lymphocyte GRK2 expression was reflected by a parallel increase in VSM GRK2 expression16. In humans, it was found that lymphocyte GRK2 expression directly correlated with GRK2 in biopsies from the right atria of the same patient17. It has been suggested that human lymphocyte GRK2 expression and activity are increased, at least in a small cohort of young Caucasians with borderline hypertension18. Experimentally, we have shown in transgenic mouse models that increasing expression of GRK2 in vascular smooth muscle was sufficient to increase BP19. Our data suggested that attenuation in vasodilation, mediated by βARs, is an important contributor to the high BP in these mice19.
Alterations in adrenergic nervous system activity appears to play a role in development and maintenance of high BP20 and racial variations in the interplay of adrenergic activity and BP have been described21,22. The role of βARs and their desensitization is not well delineated in humans, especially in blacks. The purpose of this study was to quantify lymphocyte GRK2 activity in a human sample of black Americans and determine if there is an association of GRK2 levels with plasma catecholamines and BP. An increase in GRK2 levels may indicate increased neurohormal activity and give insight into the pathogenesis of high BP.
Data were obtained from predominantly black participants who were enrolled in an ongoing cohort study on biomarkers of hypertension and cardiovascular injury. BP measurements were obtained according to well standardized procedures23 and fully described in Supplementary Methods (please see http://hyper.ahajournals.org). Based on the average systolic and diastolic BP, participants were stratified as normal BP (average BP <130/80 mm Hg) or high BP (systolic BP ≥130 mm Hg and/or diastolic BP ≥85 mm Hg. The Adult Treatment Panel III criteria for high BP24 were used to designate high BP in the cohort study because other parameters of metabolic syndrome were under investigation. Individuals who were taking antihypertensive medication for BP control were designated high BP regardless of the BP level. Fasting blood samples were collected into 8ml Vacutainer CPT tube(s) with sodium citrate (Becton Dickinson) which facilitate selection of lymphocytes. Lymphocytes and plasma were isolated and stored at −80°C.
RNA was isolated and cDNA constructed using standard procedures as fully described (Supplemental Methods – please see http://hyper.ahajournals.org). Real-time PCR was performed in duplicate using primers as described in Table S1 (please see http://hyper.ahajournals.org) and product verification was confirmed with a 9-point standard curve, sequencing, melting point curve analysis and agarose gel electrophoresis as fully described in Supplemental Methods (please see http://hyper.ahajournals.org).
Immunoblotting was performed in 35 µg of lymphocyte protein using a GRK2 primary antibody (C15, Santa Cruz Biotechnology) as fully described in Supplemental Methods (please see http://hyper.ahajournals.org).
GRK enzymatic activity was assessed via determination of the extent of light-dependent phosphorylation of rhodopsin according to previously published methods25. Sample phosphorylation results were normalized upon a seven-point standard curve of rhodopsin phosphorylation performed with purified GRK2 (0.03–2000 ng, Invitrogen).
Plasma epinephrine and norepinephrine levels were assessed according to the manufacturer’s protocol (Human Bi-CAT EIA, Alpco Diagnostics) from plasma extracted from the same sample as lymphocyte extraction. Conditions on obtaining the blood samples were uniform in all participants, including overnight fasts, absence of nicotine or caffeine exposure, and 45 minutes of rest23. Samples with significant hemolysis and lipolysis were excluded, as were those that fell outside the linear portion of the standard curve.
The relative quantification of RT-PCR results were performed by using the mathematical model with 2−ΔΔCt method26 as well as being compared to standard amounts of GRK2 and GRK5. GRK2 and GRK5 mRNA expression were calculated and normalized to the levels of 28S mRNA. As described above subjects were stratified into two BP groups (Normal BP: SBP<130 and diastolic BP <85 mm Hg; and High BP: SBP≥130mmHg and/or diastolic BP ≥85 mm Hg. Participants who were taking antihypertensive medications were designated High BP. According to these criteria, there were no individuals who had isolated diastolic hypertension. Therefore the statistical analyses focused on systolic BP as continuous and categorical variable. The most parsimonious multivariable models of GRK2 and GRK5 were utilized to better understand their clinical implications in relation to high BP, diabetes, obesity, and sex. Analyses were completed via the appropriate contrasts from the mixed effects linear regression model, allowing for variance differences among the groups. Contrasts were computed and tested using the appropriate estimates of effects and standard errors as computed from the variance-covariance matrix of parameter estimates. When distributions were not appropriately symmetric even with transformation, then nonparametric regression methods were completed using, for example, median regression. We then completed bi-variate and multi-variate analyses of all parameters. Data are expressed as mean ± S.E.M. Data were analyzed using one-way analysis of variance (ANOVA), two-way ANOVA, or unpaired two-tailed Student’s t-test as indicated.
The study sample included 133 subjects, ages 18 to 67 years (Table S2, S3 – please see http://hyper.ahajournals.org). In this cohort, systolic BP ranged from 90mmHg to 188mmHg. Using a cut point of 130 mmHg as a threshold (systolic BP ≥ 130mmHg), 71% of the sample was below and 29% of the sample had a systolic BP ≥ 130mmHg. The distribution of males (47.5%) and females (52.5%) was equivalent. 132 subjects were black.
Robustness of the quantitative real-time PCR analysis was verified (Figure S1 – please see http://hyper.ahajournals.org). Lymphocytes from the first 99 samples were used to determine GRK2 and GRK5 mRNA expression (Table S2 - please see http://hyper.ahajournals.org). Figure 1 provides data on the relationship of mRNA expression with systolic BP. There is a statistically significant correlation of GRK2 mRNA expression with systolic BP (Figure 1A) and GRK2 mRNA expression was significantly higher in the group of subjects with a systolic BP ≥130mmHg as compared to <130mmHg (Figure 1B). The increase in GRK2 mRNA expression was even more profound in males with systolic BP ≥130mmHg (Figure 1C). There was no significant correlation between GRK5 mRNA expression and systolic BP level for the total sample (Figure 1D), or when the data were analyzed for males and females separately (data not shown).
Univariate analysis was performed for GRK2 and GRK5 mRNA levels with age, sex, systolic BP, body mass index (BMI), diabetic status, and hypertensive treatment (Table 1). Only systolic BP correlated significantly with GRK2 mRNA levels (p=0.004). There was no correlation between GRK5 mRNA and systolic BP in univariate analysis.
Multivariate analysis was performed with GRK2 mRNA as the dependent variable with systolic BP, sex, BMI, diabetic status, and anti-hypertensive treatment status as independent variables (Table 1). In this model, systolic BP is the major determinant of GRK2 mRNA expression with some small additional contribution of sex and BMI.
We also considered GRK2 mRNA expression as compared to diastolic BP using 80mmHg to group the population since this is considered the cutoff value for normal diastolic BP. GRK2 mRNA was increased if diastolic BP ≥80mmHg (p=0.0082, Figure S2 - please see http://hyper.ahajournals.org).
In the next 10 subjects that were recruited, protein expression was determined by immunoblot analysis. There was a significant correlation of GRK2 protein expression with systolic BP, and GRK2 protein was higher in those with systolic BP ≥130mmHg compared to those with systolic BP <130mmHg (Figure 2).
GRK activity in the human lymphocyte extracts was measured using a rhodopsin phosphorylation assay in the next 24 subjects. All GRKs are able to phosphorylate rhodopsin however GRK2 and GRK5 are the most abundant GRKs in the lymphocyte (data not shown). Rhodopsin phosphorylation assay determines total GRK activity in a sample and therefore, we refer to this as GRK activity. Autoradiography of a seven-point standard curve produced with increasing doses of purified GRK2 and rhodopsin substrate in the presence of radio-labeled ATP demonstrates assay sensitivity (Figure 3). Verification of assay robustness is shown in Figure S3 (please see http://hyper.ahajournals.org). GRK activity was increased in subjects with systolic BP ≥130mmHg compared to those with systolic BP <130mmHg (Figure 3). We did not detect GRK5 mRNA changes and therefore predict that the activity changes we observed are due to GRK2 activity because of increased mRNA (Figure 1) and protein expression (Figure 2) although this remains to be fully determined.
In our sample of black adults there was a small but statistically significant correlation of plasma norepinephrine level with systolic BP (Figure 4A). No relationship of plasma epinephrine level with BP was detected (Figure 4B). A direct correlation was detected for plasma norepinephrine levels with GRK2 mRNA (Figure 4C).
In the current study, we found that GRK2, but not GRK5, mRNA, protein and activity directly correlated with systolic BP and plasma norepinephrine levels in black adults, a population at higher risk for hypertension and cardiovascular complications. An increase in 7TMR signaling leads to an increase in GRKs15. Sympathetic nerves of the heart, kidneys and skeletal muscle vasculature in hypertensive patients are activated and sympathetic overactivity in the renal sympathetic outflow is a prominent pathophysiological feature in hypertensive patients27–29. Greater sympathetic nervous system activation in black American adults has been reported29 and our data confirms an increase in plasma norepinephrine levels with increased systolic BP. An increase in sympathetic nervous system activity results in an increase in βAR signaling and subsequent GRK-mediated βAR desensitization. In our black subjects, we found that as BP and norepinephrine levels increase, there is a corresponding increase in lymphocyte GRK2, but not GRK5, expression and activity. The increased norepinephrine and GRK2 we observed may contribute to a decrease in βAR signaling. Hypertensive black Americans also tend to be at higher risk for increased circulating plasma endothelin (ET-1) levels30 and ET-1 receptors are preferentially desensitized by GRK231. Hypertension has been associated with increased plasma levels of other ligands including the components of the renin-angiotensin system32 although hypertension in black Americans tends to be low-renin in nature30. Like endothelin receptors, angiotensin AT1 receptors are regulated by GRK2. Thus, the increase in GRK2 we observe may be due not only to increased plasma norepinephrine, but also to increases in plasma or local levels of endothelin, angiotensin II and/or perhaps other endogenous ligands. Although no change in GRK5 mRNA expression was observed, we did not investigate whether the Q41L polymorphism exists, a nonsynonymous polymorphism of GRK5 that is associated with a cardio-protection phenotype in black Americans33.
Data is accumulating to provide support for a direct relationship between GRKs and BP. We previously found that vascular smooth muscle (VSM) overexpression of either GRK2 or GRK5 alone was sufficient to increase basal BP in mice19,34. Increased GRK2 is associated with increased BP in both lymphocytes and VSM cells of two different high BP rat models, the spontaneously hypertensive rat (SHR) and Dahl salt-sensitive hypertensive rats16. In addition to the animal studies, in a small cohort of young Caucasian hypertensive subjects (n=8), lymphocyte GRK2 protein and activity levels were increased compared to normotensive, age-matched control subjects35. GRK5 is increased in rat VSM during both angiotensin II and norepinephrine-induced hypertension36. The augmented GRK5 level was attenuated by concurrent hydralazine administration suggesting that hemodynamic stress itself was responsible for the GRK5 increase36. Data from our lab suggest that the role of GRK in BP regulation is variable. In previous studies, we found that mice with increased VSM cell GRK2 expression have high BP and also have both VSM and cardiac hypertrophy. In contrast, mice with increased expression of VSM cell GRK5 have high BP without concomitant hypertrophy. Whether increased expression of GRK2 is also a predictor of increased cardiovascular risk and propensity towards increased morbidity remains to be determined.
In the current study, there was a direct correlation of GRK2, but not GRK5, with BP. Although the correlation in this study is low, our sample population was mostly considered to have BP within “normal” limits or “prehypertensive”. It will be particularly important to further test the direct correlation between GRK2 and BP in a greater number of subjects with high BP.
GRK2 upregulation occurs with heart failure37. The increase in GRK2 in heart failure acts to decrease βAR signaling leading to ionotropic and chronotropic decreases. It would seem counterintuitive, at least from a cardiac perspective, that increased GRK2 be associated with an increase in BP. However, the cardiovascular system is a highly complex system with multiple inputs and sites of regulation including heart, vasculature, brain and kidney and a multitude of different receptors and signaling pathways. Previously, we found an increase in cardiac GRK2 accompanies an increase in BP that precedes cardiac dysfunction in the spontaneously hypertensive heart failure rat model38. We predict, and data in this manuscript suggest, that GRK2 levels are at least somewhat reflective of catecholamine activity. It remains to be determined how signaling by other hormones and neurotransmitters affects GRK2 levels but it could be that depending on the etiology of the high BP, the ligand that results in an increase in GRK2 might have minimal cardiac effects.
In American blacks, we have found that norepinephrine levels correlate with systolic BP. In addition, both norepinephrine and systolic BP directly correlate with GRK2 expression and activity. GRK2 is upregulated with stimulation of specific 7TMRs including βARs and its increase can have a profound impact on 7TMR signaling and therefore BP control. Understanding GRK levels may give us insight into the 7TMRs stimulated during hypertension thus potentially broadening our understanding of etiology and progression of the disease and may contribute to improved therapies and more targeted treatment strategies to control hypertension and reduce cardiovascular morbidity in this at-risk population.
Sources of Funding
The studies were funded, in part, by NIH NHLBI. This project was funded, in part, under a grant with the Pennsylvania Department of Health. The Department specifically disclaims responsibility for any analyses, interpretations or conclusions.