• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Am J Hypertens. Author manuscript; available in PMC Jul 26, 2009.
Published in final edited form as:
PMCID: PMC2715837
NIHMSID: NIHMS124211

G-Protein-Coupled Receptor Kinase 4 Polymorphisms and Blood Pressure Response to Metoprolol Among African Americans: Sex-Specificity and Interactions

Abstract

BACKGROUND

African Americans have a disproportionate burden of hypertension and comorbid disease. Pharmacogenetic markers of blood pressure response have yet to be defined clearly. This study explores the association between G-protein-coupled receptor kinase type 4 (GRK4) variants and blood pressure response to metoprolol among African Americans with early hypertensive nephrosclerosis.

METHODS

Participants from the African American Study of Kidney Disease and Hypertension (AASK) trial were genotyped at three GRK4 polymorphisms: R65L, A142V, and A486V. A Cox proportional hazards model, stratified by gender, was used to determine the relationship between GRK4 variants and time to reach a mean arterial pressure (MAP) of 107 mm Hg, adjusted for other predictors of blood pressure response. Potential interactions between the three polymorphisms were explored by analyzing the effects of gene haplotypes and by stratifying the analysis by neighboring sites.

RESULTS

The hazard ratio with 95% confidence interval by A142V among men randomized to a usual MAP (102–107 mm Hg) was 1.54 (1.11–2.44; P = 0.0009). The hazard ratio by A142V with R65/L65 or L65/L65 was 2.14 (1.35–3.39; P = 0.001). Haplotype analyses were consistent but inconclusive. There was no association between A142V and blood pressure response among women.

CONCLUSIONS

Results suggest a sex-specific relationship between GRK4 A142V and blood pressure response among African-American men with early hypertensive nephrosclerosis. Men with a GRK4 A142 were less responsive to metoprolol if they had a GRK4 L65 variant. The effect of GRK4 variants and blood pressure response to metoprolol should be studied in larger clinical trials.

Although African Americans have a disproportionate burden of hypertension and associated comorbid diseases, blood pressure management is inadequate in the majority of patients despite numerous treatment alternatives. Genetic variation is thought to contribute to blood pressure response. Examples of candidate genes involved in the physiological pathway of β-adrenoreceptor blockers such as metoprolol are G-protein-coupled receptor kinases (GRKs). GRK4 is capable of phosphorylating and/or desensitizing many G-protein-coupled receptors, including activated forms of the dopamine receptor and β-adrenoceptors (Figure 1).1

Figure 1
Schematic of GRK4 pathway. GRK4 phosphorylates G-protein-coupled receptor proteins (GPCRs), such as β-adrenoreceptors, resulting in subsequent binding of β-arrestin and uncoupling of GPCRs mediated intracellular signaling.

Single-nucleotide polymorphisms (SNPs) on the GRK4 gene also have been associated with hypertension in human studies (Table 1).27 GRK4 is thought to play an important role in the regulation of β-adrenoreceptor density, and GRK4 activity has been shown to decrease with β-adrenoreceptor blockers such as atenolol.8 In this study, we looked at the relationship between GRK4 genotypes and haplotypes and blood pressure response among African Americans with early hypertensive nephrosclerosis randomized to treatment with metoprolol from the African American Study of Kidney Disease and Hypertension Study (AASK). As described in our previous study,9 the analysis focused on the time to reach a mean arterial pressure (MAP) of ≤107 mm Hg, a clinically reasonable blood pressure treatment goal of ~140/90 mm Hg, and one of the target MAP end points defined in the original trial.10,11

Table 1
GRK4 polymorphisms

METHODS

Details of the AASK protocol and this ancillary AASK Genetics Study have been previously published.912 Of the original 1,094 African-American men and women with a clinical diagnosis of hypertensive nephrosclerosis, 994 were eligible to participate in the AASK Genetics Study. Of these, 850 consented to participate; 839 participants had adequate DNA samples for genotyping, including 328 participants randomized to treatment with metoprolol. In addition to randomization to the study drugs (ramipril, metoprolol, and amlodipine), participants were also randomized to either a usual MAP (102–107 mm Hg) or a low MAP (≤92 mm Hg) group. Unlike the primary double-blinded drug, MAP goal categorization was known to both investigators and patients.

Genetics

Genotyping was done with Roche Molecular Systems (Pleasanton, CA), using a platform that included 38 polymorphisms that have been associated with hypertension and cardiovascular disease. Three nonsynonymous GRK4 polymorphisms were analyzed: R65L (rs2960306), A142V (rs1024323), and A486V (rs1801058) (Table 1; Figure 2).

Figure 2
Schematic of GRK4 gene and polymorphisms. The GRK4 gene, located on 4p16.3, consists of ~2,225 base pairs and 16 exons. Three GRK4 polymorphisms were analyzed in this study: (i) R65L (rs2960306) (ii) A142V (rs1024323), and (iii) A486V (rs1801058). These ...

Genomic DNA was extracted from whole blood using the PureGene blood DNA kit (Gentra Biosystems, MN). Genotype assays for SNPs were developed based on flanking genomic DNA sequence (http://www.ncbi.nlm.nih.gov/SNP/), and each subject was genotyped using an immobilized probe approach. Each DNA sample was amplified in two multiplex PCRs using biotinylated primers, hybridized to two linear arrays of immobilized, sequence-specific oligonucleotide probes, and detected colorimetrically. Genotype assignments were made by capturing images with a flatbed scanner and using proprietary software developed by Roche Molecular Systems to resolve probe signals into genotypes for all polymorphisms. Discordant or ambiguous results were resolved by repeat PCR or hybridization. Deviation from Hardy–Weinberg equilibrium was tested using the Pearson goodness of fit (χ2) test statistic. Linkage disequilibrium coefficients (D′) between polymorphism were determined using SAS software expectation maximization algorithm (Version 9.1, SAS Institute, Cary, NC).

Statistical analysis

Since this cohort had early renal insufficiency, analysis focused on the first year of randomization because progressive renal disease could obscure any relationship between genotype and drug response. Baseline characteristics and preliminary outcomes between GRK4 variants (or genotypes) were first explored. A Cox proportional hazards model was used to explore the relationship between the time (days) to reach an MAP of 107 mm Hg and GRK4 variant (or genotype) and gene haplotype. Participants had to have two consecutive MAPs at or below 107 mm Hg, and have the average of all remaining MAPs in the first year at or below 107 mm Hg. Participants who did not reach an MAP of 107 mm Hg in the first year of randomization were considered treatment failures. This MAP of 107 mm Hg was chosen because it was a clinically reasonable goal (corresponding to a blood pressure of ~140/90 mm Hg) and a target MAP for those randomized to usual MAP in the original AASK trial. This MAP was also the outcome analyzed for the low MAP randomization group because few of these participants reached the low MAP of 92 mm Hg, resulting in limited power to analyze this group based on the lower MAP goal.

MAP goal randomization (low or usual) violated the Cox proportional hazards assumption and these groups were analyzed separately. A Cox model was run for each variant (or genotype) assuming an additive (or quantitative) relationship between the variant and blood pressure response. Based on these initial results, significant relationships (P ≤ 0.05) were further explored. The “hazard” ratio of reaching an MAP of 107 mm Hg was then estimated controlling for other potential predictors of blood pressure response such as renal function (serum creatinine or glomerular filtration rate), age, and body mass index. Other predictors of blood pressure were then included in the model if the parameter appeared to be contributing to the model (i.e., the hazard ratio for the covariate was significant as determined by the confidence interval or P value, or the parameter resulted in an increased maximum likelihood or an overall increase in model significance).

Because previous studies have suggested that the activity for several polymorphisms related to hypertension may be sex-specific,13,14 the interaction between gender and GRK4 polymorphisms was first explored by entering a sex by polymorphism interaction term in the Cox model. Cox models were also stratified by gender. Interactions between GRK4 polymorphisms were similarly explored. GRK4 haplotypes were also used to explore the effect of different combinations of polymorphisms; haplotypes were entered into a Cox model to determine whether there was a significant difference in blood pressure response between participants with zero, one, or two gene haplotypes.

Finally, it should be noted that other antihypertensive drugs were used to manage blood pressure based on a standardized protocol: diuretic → α-adrenergic antagonist rarr; central alpha adrenergic agonist.10 The average number of medications per day in the first year of randomization was used as a proxy to control for the use of other antihypertensive medications.

False discovery and power

False discovery resulting from multiple comparison testing was corrected using a more stringent significance criterion. Because these polymorphisms were in moderate linkage disequilibrium, SNPSpD was used to calculate a revised P value required to keep the type 1 error rate below 0.05 (http://gump.qimr.edu.au/general/daleN/SNPSpD/). Based on linkage disequilibrium structure of the three sites, there were an effective 2.8 comparisons (as opposed to 3 under the assumption of independence), resulting in a revised P value of 0.02 (ref. 15). Using a two-sided test of proportions, this study (N = 150, 1/2 each in usual vs. low MAP goal groups) had at least 85% power to detect ~30% response differences between groups, assuming a minor allele frequency of at least 25%. The power of this study dropped to 70% with a minor allele frequency of 15%. Statistical analyses (including power analyses) were done using SAS 9.1 (SAS Institute, Cary, NC) and STATA 9.2 (StataCorp, College Station, TX).

Specificity control and population stratification

Because GRK4 is not in the pharmacological pathway of amlodipine (a calcium channel blocker) and ramipril (an angiotensin-converting enzyme inhibitor), participants randomized to these drugs (N = 497) were used as a specificity control for the effect of genotype on response. In addition, genetic admixture among African Americans could have resulted in false positive or spurious associations between genotype (or haplotype) and phenotype.16 To address this, a generalized analysis of molecular variance17 was used to test the relationship between the genetic background and blood pressure response (day to reach an MAP of 107 mm Hg) using an identity-by-state distance matrix based on additional genotypes of 83 biallelic markers.

RESULTS

Genetics

Of the 328 AASK study participants randomized to metoprolol, 317 (97%) were successfully genotyped at R65L, 315 (96%) at A142V, and 317 (97%) at A486V. Although all three polymorphisms were in Hardy–Weinberg equilibrium, there were only six participants with V486/V486. All three sites were in linkage disequilibrium (Figure 2). GRK4 haplotypes were inferred for 319 participants and the majority had at least one copy of L65/V142/A486, R65/A142/A486, or R65/V142/A486.

Ancillary AASK study participant characteristics

There were no significant differences in gender, age, baseline MAP, body mass index, serum creatinine, and glomerular filtration rate between our cohort (N = 328) and the original AASK cohort (N = 1,094).11 There were 197 men (mean age 55 ± 10 years) and 131 women (mean age 54 ± 10 years). There was complete follow-up in the first year of randomization, 10,787 days at risk with seven treatment failures among those randomized to a low MAP (based on reaching an MAP of 107 mm Hg) and 23,374 days at risk with 24 treatment failures among those randomized to a usual MAP. There were no significant baseline differences between those who were randomized to low and usual MAP, or between those who did and did not reach an MAP of 107 mm Hg (data not shown).

Characteristics by GRK4 genotypes

Although there was an approximate 50:50 randomization to the low and usual MAP groups by R65L and A142V, there were more participants randomized to a low MAP with A486/V486. There were no other differences in baseline characteristics by GRK4 variants (Table 2).

Table 2
Baseline characteristics and preliminary outcomes, stratified by GRK4 genotypes

Time to reach map of 107 mm Hg: Cox proportional hazards model

Usual MAP (102–107 mm Hg)

With men and women combined, the hazard of reaching a target MAP of 107 mm Hg by R65L was not significant (adjusted P = 0.11, results not shown). There was a marginally significant relationship between A486V and blood pressure response (adjusted P = 0.04, results not shown). Although the interaction term between gender and A486V was not significant, the association seemed to be only among women based on a stratified analysis (adjusted P = 0.05, Table 3). It should be noted that the few homozygous V486/V486 participants (N = 4) were excluded from the analysis.

Table 3
Adjusteda hazard ratios of reaching an MAP of ≤ 107mm Hg by GRK4 variants and GRK4 haplotypes for participants randomized to usual MAP (102 to 107 mm Hg)

Among men and women, there was also a marginal association between A142V and blood pressure response (adjusted P = 0.05, results not shown), as well as a marginally significant interaction term for A142V and gender (P = 0.08, results not shown). Stratification by gender suggested that this association was only significant among men (unadjusted P = 0.05, data not shown). Although baseline MAP and average number of medications were significant predictors of blood pressure response, MAP was more significant among men and was retained in the model. Other potential predictors of blood pressure (age, renal function, body mass index, and socio demographic variables) were not significant. All variables retained in the final model met the proportional hazards assumption. Among men, the adjusted (for baseline MAP) hazard ratio based on an additive relationship with V142 and blood pressure response was 1.54, with a 95% confidence interval of 1.11–2.44 (P = 0.009; Table 3 and Figure 3a).

Figure 3
Adjusted Cox proportional hazards regression curves. Days to reach an MAP of 107 mm Hg adjusted for baseline MAP for men randomized to usual MAP (102–107 mm Hg) is shown by A142V. (a) Men with A142 responded slower in comparison to men with V142/V142 ...

Potential interactions between A142V and neighboring sites were then explored using imputed haplotypes as the unit of analysis in the Cox model. There were too few women with two copies of L65/V142/A486 and R65/V142/A486 for analysis. Among men randomized to a usual MAP, the R65/V142/A486 haplotype was marginally significant (P = 0.03; Table 3). There were no significant haplotype models among women.

The possibility of an interaction between A142V and R65L among men randomized to a usual MAP was then further analyzed. A multiplicative interaction term between A142V and R65L was significant (P = 0.04, results not shown). Although there was limited power to detect an association (N = 22), there did not appear to be any differences in blood pressure response by A142V if men also had R65/R65 (P = 0.85; Table 4 and Figure 3b). However, men with A142 responded more slowly if they had L65 (Figure 3c). An additive A142V Cox model with R65/L65 was significant (P = 0.01; Table 4). Although not significant (N = 22), an additive A142V model with L65/L65 was of a similar magnitude and direction as with R65/L65. An additive A142V model with either R65/L65 or L65/L65 was strongly significant: 2.14 (1.35–3.39; P = 0.001; Table 4). Moreover, this hazard ratio for A142V with an L65 variant (2.14) was higher (and more significant) than the hazard ratio without stratification (1.54, as noted previously), supporting an interaction between A142V and R65L. There were few individuals with a V486/V486, limiting the ability to explore interactions between A142V and this third site using a stratified analysis.

Table 4
Adjusteda hazard ratios of reaching an MAP of ≤107mm Hg for A142V stratified by R65L among participants randomized to usual MAP (102–107 mm Hg)

Low MAP (≤92 mm Hg)

Although there was a strong relationship between A486V and blood pressure response among men (P < 0.001; results not shown), the interpretation of this was limited because men with V486/V486 were excluded (N = 2). There were no other significant relationships between GRK4 variants and blood pressure response among men and women randomized to a low MAP.

Specificity control and population stratification

There were no significant associations between R65L or A142V and blood pressure response among men or women randomized to ramipril and amlodipine in either the low or usual MAP group (results not shown). There was a marginally significant association at A486V among men in the usual MAP group (P = 0.03; results not shown). Generalized analysis of molecular variance showed no significant relation between overall genetic profile (at 83 single nucleotide variants) and the blood pressure response trait (days to MAP of 107 mm Hg), indicating that population stratification did not contribute to blood pressure response in this data set (P = 0.17).

DISCUSSION

The AASK genetics data presented a unique opportunity to explore the relationship between GRK4 variants and blood pressure response to metoprolol within the first year of randomization among African Americans with early hypertensive nephrosclerosis. With each additional A142, men randomized to a usual MAP (102–107 mm Hg) were 50% less likely to reach a target MAP of 107 mm Hg (P = 0.009). Further exploration for an interaction with neighboring R65L suggested that those with A142 were less responsive only if they had L65 (P = 0.04 for interaction term). With each additional A142, men were twice as less likely to reach the MAP goal if they had L65 (P = 0.001), as discussed further below.

There was evidence that A486V was a predictor of blood pressure response among men randomized to low MAP P (< 0.001) and among women randomized to usual MAP P ( = 0.05). A full analysis of this site is, however, limited because there were few diploid V486/V486 individuals.

A142V and blood pressure

As a result of regulating dopaminergic receptor density, GRK4 is thought to have an important role in the physiology of hypertension.1,18 Increased GRK4 activity results in enhanced receptor phosphorylation, endocytosis, and subsequent uncoupling of the G-protein complex.5,19 In comparison to the GRK4 A142 wild type, transgenic mice with the V142 variant are hypertensive as a result of increased serine phosphorylation of the dopaminergic receptor, subsequent uncoupling of the G-protein complex, decreased intracellular cyclic AMP levels and impaired natriuresis.4,6,19 Although this variant was not associated with differences in baseline MAP in this data set, it should be remembered that the AASK trial data only included patients already diagnosed with essential hypertension (and early nephrosclerosis). However, V142 was associated with faster blood pressure response to metoprolol (P = 0.009), perhaps as a result of lower β-adrenoreceptor activity.

Gender differences

Autosomal and sex genes are often differentially expressed in male and female tissues14 and have been shown to contribute differentially to multifactorial traits/diseases.13 Given this growing body of evidence of a differential role of gender on genetic expression of complex traits such as hypertension, we systematically investigated the interaction of gender with GRK4 variants in this study. Based on a stratified analysis, the association between A142V and blood pressure response was specific for men, suggesting a gene by gender interaction. While the interaction term was marginally significant (P = 0.08), we had limited power to do a full Cox analysis controlling for the interaction between A142V and gender. Although we had fewer women and less power to detect an association among women, the hazard ratio for women was 1.02 (0.65–1.61), supporting the null hypothesis of no association among women.

Haplotype analyses and exploring interactions between polymorphisms

In this data set there were few diploid L65/V142/A486 and R65/V142/A486 individuals resulting in limited power to explore interactions using haplotypes as the unit of analysis. In addition, the three sites studied were only in moderate linkage disequilibrium, increasing the likelihood of error resulting from imputing haplotypes. However, haplotype analyses were consistent as men randomized to a usual MAP with an A142 in the haplotype (i.e., R65/A142/A486) responded slower (P = 0.09) and those with a V142 in the haplotype (i.e., R65/V142/A486) responded faster (P = 0.03). Further exploration using a stratified analysis suggested that men with A142 seemed to respond much slower only if they had an L65 variant (P = 0.001). Potential biological mechanisms explaining this observation have yet to be explored with in vitro molecular studies.

Study limitations, drug specificity, and population stratification

Limitations of this study design have been previously detailed9 and are also briefly summarized here. Although the AASK Genetics Study is a representative subset of the original AASK cohort,10,11 how loss to follow-up or death might bias these results cannot be determined. Results may also be specific to African Americans with hypertensive nephrosclerosis. As a result of limited genetic diversity, this study was also not adequately powered to explore fully the relationship between A486V or GRK4 haplotypes and blood pressure. Baseline MAP and other antihypertensive medications were strong predictors of blood pressure response in this data set (the likelihood ratio test χ2 for the A142V model with and without adjustment for baseline MAP went from 5.52 (P = 0.02) to 11.81 (P = 0.003)). Although several classes of antihypertensive medications were used, the average number of medications was a proxy measure for specific antihypertensive medication classes administered according to a specified protocol across study sites.

Because GRK4 is not in the pharmacological pathways of angiotensin-converting enzyme inhibitors and calcium channel blockers, participants randomized to treatment with amlodipine and ramipril were used as a specificity control. Although we did not find any significant associations between R65L and A142V genotypes and blood pressure response, there was a marginal association between A486V among men randomized to a usual MAP. As noted above, A486V should ideally be fully studied in a larger data set. Using generalized analysis of molecular variance17 at 83 biallelic loci, we did not find any significant relationship between population stratification and blood pressure response (P = 0.14).

In conclusion, GRK4 A142V may be a predictor of blood pressure response to metoprolol among African-American men with early hypertensive nephrosclerosis. Men with an A142 variant were less responsive to metoprolol if they also had an L65. This effect was not observed among women. Given the high burden of hypertension and end organ disease among African Americans, the relationship between GRK4 variants and blood pressure response warrants further study in larger clinical trials.

Acknowledgments

We appreciate the support of the NIH/NCMHD-sponsored (MD000220) EXPORT/CRCHD minority health center, as well as the NIH/NCRR-sponsored (RR00827) General Clinical Research Center. Satellite Research, NIH (K23 RR020822-01A1, DK048689, RR000071, DK057867, DK60702, RR11145) and the Department of Veterans Affairs.

Footnotes

Disclosure: V.H.B. works for Roche Molecular Systems. M.S.L. receives funding from King Pharmaceuticals. Other authors declared no conflict of interest.

References

1. Neve KA. Novel features of G protein-coupled receptor kinase 4. Mol Pharmacol. 2006;69:673–676. [PubMed]
2. Bengra C, Mifflin TE, Khripin Y, Manunta P, Williams SM, Jose PA, Felder RA. Genotyping of essential hypertension single-nucleotide polymorphisms by a homogeneous PCR method with universal energy transfer primers. Clin Chem. 2002;48:2131–2140. [PubMed]
3. Gu D, Su S, Ge D, Chen S, Huang J, Li B, Chen R, Qiang B. Association study with 33 single-nucleotide polymorphisms in 11 candidate genes for hypertension in Chinese. Hypertension. 2006;47:1147–1154. [PubMed]
4. Wang Y, Li B, Zhao W, Liu P, Zhao Q, Chen S, Li H, Gu D. Association study of G protein-coupled receptor kinase 4 gene variants with essential hypertension in northern Han Chinese. Ann Hum Genet. 2006;70:778–783. [PubMed]
5. Speirs HJ, Katyk K, Kumar NN, Benjafield AV, Wang WY, Morris BJ. Association of G-protein-coupled receptor kinase 4 haplotypes, but not HSD3B1 or PTP1B polymorphisms, with essential hypertension. J Hypertens. 2004;22:931–936. [PubMed]
6. Wang Z, Armando I, Asico LD, Escano C, Wang X, Lu Q, Felder RA, Schnackenberg CG, Sibley DR, Eisner GM, Jose PA. The elevated blood pressure of human GRK4gamma A142V transgenic mice is not associated with increased ROS production. Am J Physiol Heart Circ Physiol. 2007;292:H2083–H2092. [PubMed]
7. Zhu H, Lu Y, Wang X, Snieder H, Treiber FA, Harshfield GA, Dong Y. The G protein-coupled receptor kinase 4 gene modulates stress-induced sodium excretion in black normotensive adolescents. Pediatr Res. 2006;60:440–442. [PubMed]
8. Leineweber K, Rohe P, Beilfuss A, Wolf C, Sporkmann H, Bruck H, Jakob HG, Heusch G, Philipp T, Brodde OE. G-protein-coupled receptor kinase activity in human heart failure: effects of β-adrenoceptor blockade. Cardiovasc Res. 2005;66:512–519. [PubMed]
9. Bhatnagar V, O’Connor DT, Schork NJ, Salem RM, Nievergelt CM, Rana BK, Smith DW, Bakris GL, Middleton JP, Norris KC, Wright JT, Cheek D, Hiremath L, Contreras G, Appel LJ, Lipkowitz MS. Angiotensin-converting enzyme gene polymorphism predicts the time-course of blood pressure response to angiotensin converting enzyme inhibition in the AASK trial. J Hypertens. 2007;25:2082–2092. [PMC free article] [PubMed]
10. Wright JT, Jr, Kusek JW, Toto RD, Lee JY, Agodoa LY, Kirk KA, Randall OS, Glassock R. Design and baseline characteristics of participants in the African American Study of Kidney Disease and Hypertension (AASK) Pilot Study. Control Clin Trials. 1996;17:3S–16S. [PubMed]
11. Wright JT, Jr, Bakris G, Greene T, Agodoa LY, Appel LJ, Charleston J, Cheek D, Douglas-Baltimore JG, Gassman J, Glassock R, Hebert L, Jamerson K, Lewis J, Phillips RA, Toto RD, Middleton JP, Rostand SG. African American Study of Kidney Disease and Hypertension Study Group. Effect of blood pressure lowering and antihypertensive drug class on progression of hypertensive kidney disease: results from the AASK trial. JAMA. 2002;288:2421–2431. [PubMed]
12. Wright JT, Jr, Agodoa L, Contreras G, Greene T, Douglas JG, Lash J, Randall O, Rogers N, Smith MC, Massry S. African American Study of Kidney Disease and Hypertension Study Group. Successful blood pressure control in the African American Study of Kidney Disease and Hypertension. Arch Intern Med. 2002;162:1636–1643. [PubMed]
13. Wang X, Zhu H, Dong Y, Treiber FA, Snieder H. Effects of angiotensinogen and angiotensin II type I receptor genes on blood pressure and left ventricular mass trajectories in multiethnic youth. Twin Res Hum Genet. 2006;9:393–402. [PubMed]
14. Yang X, Schadt EE, Wang S, Wang H, Arnold AP, Ingram-Drake L, Drake TA, Lusis AJ. Tissue-specific expression and regulation of sexually dimorphic genes in mice. Genome Res. 2006;16:995–1004. [PMC free article] [PubMed]
15. Nyholt DR. A simple correction for multiple testing for single-nucleotide polymorphisms in linkage disequilibrium with each other. Am J Hum Genet. 2004;74:765–769. [PMC free article] [PubMed]
16. Suarez-Kurtz G, Pena SD. Pharmacogenomics in the Americas: the impact of genetic admixture. Curr Drug Targets. 2006;7:1649–1658. [PubMed]
17. Nievergelt CM, Libiger O, Schork NJ. Generalized analysis of molecular variance. PLOS Genet. 2007;3:e51. [PMC free article] [PubMed]
18. Watanabe H, Xu J, Bengra C, Jose PA, Felder RA. Desensitization of human renal D1 dopamine receptors by G protein-coupled receptor kinase 4. Kidney Int. 2002;62:790–798. [PubMed]
19. Felder RA, Sanada H, Xu J, Yu PY, Wang Z, Watanabe H, Asico LD, Wang W, Zheng S, Yamaguchi I, Williams SM, Gainer J, Brown NJ, Hazen-Martin D, Wong LJ, Robillard JE, Carey RM, Eisner GM, Jose PA. G protein-coupled receptor kinase 4 gene variants in human essential hypertension. Proc Natl Acad Sci USA. 2002;99:3872–3877. [PMC free article] [PubMed]
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...