Functional Variant of CYP4A11 20-Hydroxyeicosatetraenoic Acid Synthase Is Associated With Essential Hypertension
http://www.100md.com
循环学杂志 2005年第1期
the Department of Medicine, Divisions of Clinical Pharmacology (J.V.G., N.J.B.) and Nephrology (Y.W., J.H.C.)
the Department of Biochemistry (A.B., M.R.W.), Vanderbilt University Medical School, Nashville, Tenn
BioVentures Inc (E.P.D., K.E.W., S.W.G.), Murfreesboro, Tenn
Department of Biostatistics (L.A.C., C.-Y.G., S.D.), Boston University School of Public Health, Boston, Mass
National Heart, Lung, and Blood Institute/Framingham Heart Study (C.J.O.), Framingham, Mass.
Abstract
Background— The CYP4A11 arachidonic acid monooxygenase oxidizes endogenous arachidonic acid (AA) to 20-hydroxyeicosatetraenoic acid (20-HETE), a metabolite with renovascular and tubular functions. Mice with targeted disruption of Cyp4a14, a murine homologue of CYP4A11, have severe hypertension. We combined molecular and biochemical approaches to identify a functional variant of the CYP4A11 20-HETE synthase and determine its association with hypertensive status in 2 independent human populations.
Methods and Results— A thymidine-to-cytosine polymorphism at nucleotide 8590 resulted in a phenylalanine-to-serine substitution at amino acid 434. Expression of cDNA with serine 434 resulted in a protein with a significantly reduced AA and lauric acid metabolizing activity. In a population of 512 whites from Tennessee, the age, body mass index, and gender-adjusted OR of having hypertension attributable to the 8590C variant was 2.31 (95% CI 1.41 to 3.78) compared with the reference 8590TT genotype. In subjects from the Framingham Heart Study, the adjusted ORs of hypertension associated with the 8590C variant were 1.23 (CI 0.94 to 1.59; n=1538) in all subjects and 1.33 (CI 1.01 to 1.77; n=1331) when subjects with diabetes were excluded. No association of the variant with hypertension was detected in a population of 120 blacks.
Conclusions— We identified a variant of the human CYP4A11 (T8590C) that encodes for a monooxygenase with reduced 20-HETE synthase activity. The association of the T8590C variant with hypertension supports its role as a polygenic determinant of blood pressure control in humans, and results obtained from the large population database suggest that the relevance of the variant may vary according to hypertension comorbidity.
Key Words: genetics ; hypertension, renal ; lipids ; metabolism
Introduction
Hypertension is a significant independent risk for cardiac, renal, and cerebrovascular morbidity and mortality.1 Buffered by evidence of a genetic basis for hypertension,2 much effort has been directed toward the identification of relevant genes, characterization of their variants, and determination of genetic and environmental interacting factors.3–5 The targeting of genes involved in renal sodium and fluid regulation has met with success in identifying monogenic variants, which has greatly enhanced our mechanistic understanding of hypertension,3 and the knockout of specific genes that have resulted in a hypertensive phenotype has expanded the list of potential hypertension gene candidates. Common gene variants shown to contribute to hypertension6–10 appear to vary according to gender,8 ethnicity,10–12 or associated comorbidities.13 Thus, the identification of polygenic determinants has not been straightforward.
See p 5
The CYP4A arachidonic acid (AA) monooxygenases oxidize endogenous AA to 20-hydroxyeicosatetraenoic acid (20-HETE). The 20-HETE metabolite has the capacity to act in either a prohypertensive or antihypertensive manner depending on its expression at renovascular or tubular sites, respectively.14–17 We recently reported the characterization of Cyp4a14 (–/–) mice as a model of gender-specific severe hypertension18 and now investigate the CYP4A human homologues as candidate genetic determinants in hypertension. In addition to CYP4A11, which encodes a renal 20-HETE synthase in humans,19,20 we and others recently identified a second CYP4A gene, CYP4A22.
The primary goal of the present study was to define and characterize genetic variants of 20-HETE synthase and to define their role in human hypertension. As a prerequisite, we addressed the question of whether CYP4A22 had enzymatic activity or a role in renal fatty acid metabolism. We present data that CYP4A22 lacks functional activity in the kidney. Therefore, we combined molecular and biochemical approaches to identify functional variants in CYP4A11 and test their relevance to hypertension in 2 independent human populations.
Methods
Subjects
Tennessee Cohort
Normotensive subjects were volunteers recruited through health screening examinations performed primarily at Vanderbilt University. Subjects were defined as normotensive if they had seated systolic and diastolic pressures of less than 140 and 90 mm Hg, respectively, and no history of identification or treatment for hypertension. Hypertensive subjects were ascertained from the Vanderbilt Hypertension Center and had either established hypertension, defined as antihypertensive treatment for >6 months, or a seated diastolic pressure >90 mm Hg on at least 3 occasions. Diastolic blood pressure was defined by the absence of Korotkoff sounds (phase V). Secondary hypertension was excluded by history and physical examination. Ethnicity was self-reported, and inclusion in a particular ethnic group generally required that a proband’s parents and grandparents recognized themselves by the same ethnic designation as that of the proband. All subjects gave written consent, and the Vanderbilt University Institutional Review Board approved the protocol. All genomic samples were coded to protect donor identity.
Framingham Offspring Cohort
The design of the Framingham Heart Study has been detailed previously.21 The offspring cohort comprises offspring (and their spouses) of the original Framingham participants (recruited in 1948). The Framingham Heart Study cohort is almost entirely white. Those in the offspring cohort were invited to have follow-up laboratory and physical and laboratory examinations approximately every 4 years starting in 1971. DNA samples were obtained for 1920 individuals from June 1987 to February 1991.21 Blood pressure measurements were the mean of 2 physician-obtained readings taken >5 minutes apart in the left arm of the seated subject with a mercury column sphygmomanometer, in which appearance of Korotkoff sounds (phase I) denoted systolic blood pressure and disappearance (phase V) determined diastolic blood pressure.8 Hypertension was defined as systolic blood pressure 140 mm Hg or diastolic blood pressure 90 mm Hg or current use of antihypertensive medication.8 Diabetes (type 2) was defined as use of hypoglycemic medication or fasting plasma glucose levels >7 mmol/L (>125 mg/dL) at any of 6 examinations. Participants considered in the present study (n=1880) are a subset of unrelated individuals who provided blood samples for DNA extraction at the sixth examination cycle (n=2933). Selection criteria were established to provide DNA samples of biologically unrelated subjects with approximately equal numbers of men and women who have characteristics similar to the remainder of the participants. Included in this report were 1538 (82%) of the 1880 subjects genotyped for T8590C. All subjects gave written consent, and the protocol was approved by the Boston University Medical Center Institutional Review Board. Further details of the Framingham Heart Study are available online (http://www.nhlbi.nih.gov/about/framingham).
Polymorphism Discovery and Genotyping
In addition to single base differences found throughout the gene sequences, CYP4A11 and CYP4A22 are distinguished by the presence in CYP4A11 of short nucleotide insertions (10 bp) in introns 1, 3, 4, and 11 and deletions (18 bp) in introns 4, 6, and 11.22 Polymerase chain reaction (PCR) amplification was performed with intron-specific primers designed to maximize the difference between the 2 genes and to generate overlapping CYP4A11 specific amplicons covering all 12 exons and intervening introns.22 The selectivity of this approach was corroborated by both sequence analysis and restriction enzyme digestion with sites unique to the CYP4A11 gene.22 Primer sequences, reaction conditions, DNA extraction, and genotyping methods can be found at www.bioventures.com. To define CYP4A11 variants, both forward and reverse sequencing of PCR products were performed in a screening subset (n=34) of predominantly hypertensive subjects divided equally by gender and ethnicity. Variants were numbered by nucleotide position in reference to the translation start site.
Biochemical Characterization
By reverse transcription-PCR of a mixture of total kidney RNAs isolated from 15 nonsexed individuals (Clontech, Palo Alto, Calif), we obtained a single cDNA encoding CYP4A22 (GenBank accession: AY280371). CYP4A11 cDNA was expressed and protein purified as described except the PD-10 column was replaced by hydroxyapatite and 1% Emulgen 913 was replaced by 0.02% sodium cholate.19 Membrane fractions were isolated by differential centrifugation.23 Partially purified proteins were dialyzed against Tris-Cl buffer (pH 7.5) and kept at –80° until use. Enzyme incubations were performed at 35°C in 0.05 mol/L Tris-Cl buffer (pH 7.5) containing 10 mmol/L MgCl2, 1 mmol/L GSH, 1 mmol/L NADPH, 50 μg/mL dilauroylphosphatidylcholine, 8 mmol/L isocitric acid, 0.5 IU/mL isocitric dehydrogenase, and 70 to 100 μmol/L [1-14C]labeled lauric or AA (20 μCi/mol).23 Purified recombinant P450s were mixed with a 10-molar excess of purified liver P450 reductase and an equimolar amount of purified cytochrome b5 and incubated for 15 minutes at room temperature before the addition of reaction components.23 Reaction products were resolved and quantified.19
Statistical Methods
To determine whether the CYP4A11 polymorphisms were in Hardy-Weinberg equilibrium, actual and predicted genotype counts were compared by a 2 test statistic with 2 degrees of freedom. Logistic regression models were used to compute the ORs for the dichotomous trait of hypertension (the presence or absence of hypertension) with the genotype indicator variable(s) as the primary independent variable. Three analytical models were used to consider the crude relationship between these variables and adjustment for age, gender, and body mass index (BMI). Analysis was performed for men and women separately given that we previously observed a significant effect of gender on hypertension in mice with targeted disruption of Cyp4a14, a murine homologue of CYP4A11. The analysis of the effect of diabetes on the CYP4A11 variant in the Framingham cohort was prespecified on the basis of animal24 and human25 data that suggest that insulin downregulates CYP4A expression, potentially impacting the effect of the variant. Diabetes prevalence information was not available for the Tennessee cohort. Descriptive data are presented as mean±SD for continuous variables and as proportions for categorical variables. ORs and 95% CIs are presented from logistic regression analyses. A 2-tailed probability value of 0.05 was considered significant. Analyses were done with SPSS for Windows (version 11.0, SPSS; Tennessee data) and SAS (release 8.2, SAS Institute Inc; Framingham data).
Results
CYP4A11 Is the Functional CYP4A 20-HETE Synthase in Humans
CYP4A11 has been cloned, expressed, and characterized as a renal 20-HETE synthase.19,20 However, the enzymatic activity of the other human CYP4A gene product, CYP4A22, and its role in renal fatty acid metabolism are unknown, and its mRNA is detected in the kidney only after reverse transcription-PCR amplification.22 Western blot analysis of the membrane preparations from cells expressing CYPs 4A11 and 4A22 cDNAs showed the presence of CYP4A11 with expected electrophoretic mobility in CYP4A11-transfected cells (Figure 1, lanes C and D). Similar but less intense immunoreactive bands were detected in membranes from CYP4A22-transfected bacteria (Figure 1, lane B). The absorption spectra of purified recombinant CYPs 4A11 and 4A22 showed the presence of a typical P450 Soret band at 409 nm (not shown). On reduction with Na2S2O4 and CO addition, only CYP4A11 generated a typical reduced P450 CO-bound complex (not shown), which suggests that the CYP4A22 cDNA encodes a nonfunctional protein. This was confirmed by incubation of the partially purified P450s with [1-14C]-AA. As shown by reverse-phase HPLC radiochromatograms (Figure 2), CYP4A11 but not CYP4A22 catalyzed the metabolism of AA to 20- and 19-HETE (90% and 10% of total products, respectively). Incubation of these proteins with lauric acid, a prototype CYP4A substrate,19,20,23 demonstrated that only CYP4A11 catalyzed the formation of 11- and 12-hydroxydocosanoic acids (not shown), which confirms that CYP4A22 lacks hydroxylase activity. Sequence comparisons showed that the glycine at position 130, conserved among all CYP4A isoforms, is replaced by serine in CYP4A22. Site-specific mutation of the CYP4A11 glycine residue 130 to serine resulted in a protein that lacked a functional heme group and was catalytically inactive (not shown).
CYP4A11 Polymorphism Discovery
Screening of CYP4A11 revealed 9 variants with allelic frequencies >1% (Table 1) and in Hardy-Weinberg equilibrium within ethnic groups (not shown). Additional variants with allele frequencies <1% and ethnic comparisons of variants are posted at www.bioventures.com. Of the 2 variants in the open reading frame, the cytosine-for-thymidine transition at nucleotide 8590 (exon 11) resulted in a nonsynonymous phenylalanine (F)-to-serine (S) substitution at residue 434 of CYP4A11, which was chosen for functional characterization.
Functional Effects of the CYP4A11 Gene Polymorphism
To determine the functional consequences of an F-to-S replacement at amino acid 434, we generated the corresponding cDNA by site-specific mutation, expressed it in Escherichia coli, partially purified it, and compared the fatty acid hydroxylating activities of the wild-type (F434) and mutated (S434) proteins. As shown in Table 2, compared with F434 (wild type), the S434 replacement reduced by more than half the 20-HETE synthase activity of CYP4A11, with only a small effect on its apparent Km values and thus, its affinity for AA (Table 2). Similarly, the apparent Vm of the lauric acid conversion to 12-hydroxylaurate was reduced by more than half, but for this fatty acid, there was also a significant reduction in apparent Km values (Table 2). These results suggest that the T8590C variant, which corresponds to the F434S protein polymorphism, affects the catalytic activity of the 20-HETE synthase through a loss-of-function mechanism.
The CYP4A11 8590 C Allele is Associated With Hypertension
Tennessee Cohort
To determine whether an association exists between the CYP4A11 variant and hypertension, each of 512 subjects were genotyped at the T8590C (also referred to as F434S) locus (blacks: hypertensive n=60, normotensive, n=60; and whites: hypertensive n=197, normotensive n=195). As displayed in Table 3, age was similar by blood pressure status and between gender in each ethnic group. BMI and mean arterial pressure were greater in hypertensive than in normotensive subjects but were not different between genders. All subjects with hypertension were receiving antihypertensive medications. Compared with the reference 8590TT genotype, there was no association of the 8590C allele with hypertension in blacks (Table 4). In whites, the 8590C allele was associated with an increased prevalence of hypertension after adjustment for age, gender, and BMI in both men and women (OR 2.31, 95% CI 1.41 to 3.78). No significant interaction between gender and genotype was detected, nor were there significant differences in age and BMI across genotype groups.
Framingham Offspring Cohort
To test the hypothesis raised in the initial study findings in an independent cohort and to further characterize the association of the CYP4A11 8590C (434S) variant with hypertension, we characterized this variant in the large population-based Framingham cohort (n=1538). Table 5 shows baseline characteristics of male and female subjects.21 Approximately 50% of the men and 40% of the women were hypertensive, and 60% to 75% of them were taking antihypertensive medications. Table 6 shows that the OR of having hypertension attributable to the 8590C allele was 1.21 (95% CI 0.93 to 1.56) after adjustment for age, BMI, and gender and was 1.23 (95% CI 0.94 to 1.59) with a model that additionally adjusted for cigarette and alcohol use. After exclusion of diabetics, the comparable ORs were 1.31 (95% CI 0.99 to 1.74) and 1.33 (95% CI 1.01 to 1.77), respectively. ORs tended to be greater in men than in women, but no significant interaction between gender and CYP4A11 genotypes was detected.
Discussion
In the present study, we have provided evidence that although CYP4A11 functions as a 20-HETE synthase, CYP4A22 does not contribute to renal 20-HETE biosynthesis. Furthermore, to determine the relevance of genetic variation in CYP4A11 to hypertension, we identified a coding polymorphism (T8590C), biochemically characterized 8590C as a loss-of-function variant, and demonstrated its association with hypertension in whites. Initial findings of associations between biologically plausible functional variants are often not replicated in follow-up studies, and even when associations are replicated, they are often less strong than in the initial validation cohort.27 For this reason, we conducted a study of replication in a population-based cohort of whites. Our results suggest the existence of a modest association finding in the Framingham cohort. Clearly, our findings strongly support further study of the relationship of variants in this gene with hypertension in other populations.
Both animal14,16 and human28,29 data support CYP4A11 as a candidate gene in hypertension. Numerous studies in the spontaneously hypertensive and Dahl salt-sensitive rats have shown an association between CYP4A 20-HETE synthase expression levels and/or activity and (1) altered nephron-specific transport of sodium chloride and (2) increased blood pressure.14,16 These studies support the contention that 20-HETE exerts prohypertensive and antihypertensive actions that are critically dependent on the site of biosynthesis and action along the nephron.14–17 In the renal tubule, 20-HETE blocks sodium transport and functions primarily as a natriuretic, antihypertensive substance.16,17 In contrast, 20-HETE acting in the renal vasculature has vasoconstricting, prohypertensive properties.16 Analyses of CYP4A expression and of 20-HETE synthase activity demonstrated abundant CYP4A expression and enzyme activity in the proximal and distal tubule, 2 segments of the rat nephron identified as targets for the natriuretic, antihypertensive actions of 20-HETE.14–17 The relevance of these observations to the pathophysiology of human hypertension has been highlighted by measurements of urinary sodium and 20-HETE excretion that show the ability of dietary salt to regulate 20-HETE excretion28 and impairment of 20-HETE-dependent natriuretic mechanisms in salt-sensitive compared with salt-resistant hypertensive patients.28 Moreover, recent studies of the natriuretic response to furosemide in humans suggested that 20-HETE modulates its natriuretic effects and that altered 20-HETE natriuresis is linked to human salt-sensitive hypertension.29
The use of the Framingham cohort allowed us to address the effects of gender and phenotypic context. Significant gender differences have long been recognized in both the prevalence of hypertension and target-organ damage attributable to increased blood pressure, the mechanisms of which are largely unknown.30 The rationale to investigate the effect of gender was prompted by the finding that hypertension in Cyp4a14 (–/–) mice, a human homologue of CYP4A11, is sexually dimorphic and androgen sensitive.18 Although Cyp4a14 does not metabolize arachidonic acid, Cyp4a14 (–/–) mice had an upregulation of Cyp4a12 that resulted in increased 20-HETE synthase activity. In contrast, such an overt effect of gender was not detected in the present study, and although CYP4A11 accounts for significant renal 20-HETE synthase in humans, the role that androgens play in regulating its renal expression is yet to be defined. Nevertheless, the possibility exists that additional, non-CYP4A-dependent -hydroxylases may contribute to 20-HETE biosynthesis in the human kidney.17
The well-characterized nature of the Framingham cohort allowed us to analyze the effect of the CYP4A11 variant without a potential confounding influence of diabetes. Both animal24 and human studies25 have suggested that diabetes induces and insulin suppresses CYP4A expression. We hypothesized that the effects of a loss-of-function CYP4A variant may be muted in the context of altered glucose/insulin regulation (eg, diabetes). That contention is supported by the Framingham data, which show a significant association of the CYP4A11 variant with hypertension in the subpopulation without diabetes. These data serve to focus future validation studies and demonstrate the utility of a large, well-characterized genetic resource in the characterization of polygenic effects underlying hypertension. The apparent differences in the relationship of the variant and hypertension between the Tennessee and Framingham cohorts are not known but may relate to population-specific factors. The prevalence of diabetes in the Tennessee cohort was not known, and hypertensive subjects tended to be younger, heavier, and to have higher mean blood pressures than those in the Framingham cohort.
The lack of a significant association of the CYP4A11 variant and hypertension in blacks is somewhat surprising, despite an increased allelic frequency of the CYP4A11 variant in blacks compared with whites, given the potential link of 20-HETE to salt-sensitive hypertension28,29 and the reported greater prevalence of salt-sensitive hypertension in blacks.31 This fact could be due to a limited sample size of blacks or possibly to population stratification, a limitation inherent to the population case-control study design.32 Given the data in the present study that establish a significant loss of CYP4A11 metabolizing activity with the 8590C variant, there exists the possibility that other genetic or physiological factors modify the activity of the CYP4A system in this ethnic group. Further studies are required to confirm the lack of association of the CYP4A11 variant in blacks and to address CYP4A regulation.
In summary, we have identified a coding variant of the human CYP4A11 gene that results in a CYP4A11 protein with reduced 20-HETE synthase activity and that has a greater prevalence in hypertensive compared with normotensive whites. These results, in conjunction with published studies showing abundant renal CYP4A11 expression23 and antihypertensive, natriuretic effects of 20-HETE,16,17 provide compelling evidence that a genetically controlled alteration in CYP4A11 expression or activity plays a role as a determinant of polygenic blood pressure control in humans.
Acknowledgments
This work was supported by DK38226 (Dr Capdevila), DK28350 (Dr Waterman), HL04221, RR00095 (Dr Gainer), HL65193, HL67308, HL60906 (Dr Brown), and N01-HC-25195 (Dr O’Donnell). cDNAs were sequenced using facilities supported by the National Cancer Institute (CA68485).
Footnotes
Drs Gainer and Bellamine contributed equally to this article.
References
Stokes J III, Kannel WB, Wolf PA, D’Agostino RB, Cupples LA. Blood pressure as a risk factor for cardiovascular disease: the Framingham Study: 30 years of follow-up. Hypertension. 1989; 13: I-13–I-18.
Ward R. Familial aggregation and genetic epidemiology of blood pressure. In: Laragh JH, Brenner BM, eds. Hypertension Pathophysiology, Diagnosis, and Management. New York, NY: Raven Press; 1990: 81–100.
Lifton RP, Gharavi AG, Geller DS. Molecular mechanisms of human hypertension. Cell. 2001; 104: 545–556.
Pratt RE, Dzau VJ. Genomics and hypertension: concepts, potentials, and opportunities. Hypertension. 1999; 33: 238–247.
Halushka MK, Fan JB, Bentley K, Hsie L, Shen N, Weder A, Cooper R, Lipshutz R, Chakravarti A. Patterns of single-nucleotide polymorphisms in candidate genes for blood-pressure homeostasis. Nat Genet. 1999; 22: 239–247.
Jeunemaitre X, Soubrier F, Kotelevtsev YV, Lifton RP, Williams CS, Charru A, Hunt SC, Hopkins PN, Williams RR, Lalouel JM. Molecular basis of human hypertension: role of angiotensinogen. Cell. 1992; 71: 169–180.
Caulfield M, Lavender P, Farrall M, Munroe P, Lawson M, Turner P, Clark AJ. Linkage of the angiotensinogen gene to essential hypertension. N Engl J Med. 1994; 330: 1629–1633.
O’Donnell CJ, Lindpaintner K, Larson MG, Rao VS, Ordovas JM, Schaefer EJ, Myers RH, Levy D. Evidence for association and genetic linkage of the angiotensin-converting enzyme locus with hypertension and blood pressure in men but not women in the Framingham Heart Study. Circulation. 1998; 97: 1766–1772.
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 U S A. 2002;%19; 99: 3872–3877.
Gainer JV, Brown NJ, Bachvarova M, Bastien L, Maltais I, Marceau F, Bachvarov DR. Altered frequency of a promoter polymorphism of the kinin B 2 receptor gene in hypertensive African-Americans. Am J Hypertens. 2000; 13: 1268–1273.
Atwood LD, Kammerer CM, Samollow PB, Hixson JE, Shade RE, MacCluer JW. Linkage of essential hypertension to the angiotensinogen locus in Mexican Americans. Hypertension. 1997; 30: 326–330.
Jain S, Tang X, Narayanan CS, Agarwal Y, Peterson SM, Brown CD, Ott J, Kumar A. Angiotensinogen gene polymorphism at-217 affects basal promoter activity and is associated with hypertension in African-Americans. J Biol Chem. 2002; 277: 36889–36896.
Tiago AD, Samani NJ, Candy GP, Brooksbank R, Libhaber EN, Sareli P, Woodiwiss AJ, Norton GR. Angiotensinogen gene promoter region variant modifies body size-ambulatory blood pressure relations in hypertension. Circulation. 2002; 106: 1483–1487.
Carroll MA, McGiff JC. A new class of lipid mediators: cytochrome P450 arachidonate metabolites. Thorax. 2000; 55 (suppl 2): S13–S16.
Capdevila JH, Falck JR. The CYP P450 arachidonic acid monooxygenases: from cell signaling to blood pressure regulation. Biochem Biophys Res Commun. 2001; 285: 571–576.
McGiff JC, Quilley J. 20-Hydroxyeicosatetraenoic acid and epoxyeicosatrienoic acids and blood pressure. Curr Opin Nephrol Hypertens. 2001; 10: 231–237.
Roman RJ. P-450 metabolites of arachidonic acid in the control of cardiovascular function. Physiol Rev. 2002; 82: 131–185.
Holla VR, Adas F, Imig JD, Zhao X, Price E Jr, Olsen N, Kovacs WJ, Magnuson MA, Keeney DS, Breyer MD, Falck JR, Waterman MR, Capdevila JH. Alterations in the regulation of androgen-sensitive Cyp 4a monooxygenases cause hypertension. Proc Natl Acad Sci U S A. 2001; 98: 5211–5216.
Hoch U, Zhang Z, Kroetz DL, Ortiz de Montellano PR. Structural determination of the substrate specificities and regioselectivities of the rat and human fatty acid omega-hydroxylases. Arch Biochem Biophys. 2000; 373: 63–71.
Imaoka S, Ogawa H, Kimura S, Gonzalez FJ. Complete cDNA sequence and cDNA-directed expression of CYP 4A11, a fatty acid omega-hydroxylase expressed in human kidney. DNA Cell Biol. 1993; 12: 893–899.
Kannel WB, Feinleib M, Mcnamara PM, Garrison RJ, Castelli WP. Investigation of coronary heart disease in families: Framingham Offspring Study. Am J Epidemiol. 1979; 110: 281–290.
Bellamine A, Wang Y, Waterman MR, Gainer JV III, Dawson EP, Brown NJ, Capdevila JH. Characterization of the CYP 4A11 gene, a second CYP4A gene in humans. Arch Biochem Biophys. 2003; 409: 221–227.
Helvig C, Dishman E, Capdevila JH. Molecular, enzymatic, and regulatory characterization of rat kidney cytochromes P450 4A2 and 4A3. Biochemistry. 1998; 37: 12546–12558.
Shimojo N, Ishizaki T, Imaoka S, Funae Y, Fujii S, Okuda K. Changes in amounts of cytochrome-P450 isozymes and levels of catalytic activities in hepatic and renal microsomes of rats with streptozocin-induced diabetes. Biochem Pharmacol. 1993; 46: 621–627.
Laffer CL, Laniado-Schwartzman M, Nasjletti A, Elijovich F. 20-HETE and circulating insulin in essential hypertension with obesity. Hypertension. 2004; 43: 388–392.
Capdevila JH, Falck JR, Dishman E, Karara A. Cytochrome-P-450 arachidonate oxygenase. Methods Enzymol. 1990; 187: 385–394.
Lohmueller KE, Pearce CL, Pike M, Lander ES, Hirschhorn JN. Meta-analysis of genetic association studies supports a contribution of common variants to susceptibility to common disease. Nat Genet. 2003; 33: 177–182.
Laffer CL, Laniado-Schwartzman M, Wang MH, Nasjletti A, Elijovich F. Differential regulation of natriuresis by 20-hydroxyeicosatetraenoic acid in human salt-sensitive versus salt-resistant hypertension. Circulation. 2003; 107: 574–578.
Laffer CL, Laniado-Schwartzman M, Wang MH, Nasjletti A, Elijovich F. 20-HETE and furosemide-induced natriuresis in salt-sensitive essential hypertension. Hypertension. 2003; 41: 703–708.
Reckelhoff JF. Gender differences in the regulation of blood pressure. Hypertension. 2001; 37: 1199–1208.
Grim CE, Robinson M. Blood pressure variation in blacks: genetic factors. Semin Nephrol. 1996; 16: 83–93.
Ziv E, Burchard EG. Human population structure and genetic association studies. Pharmacogenomics. 2003; 4: 431–441.(James V. Gainer, MD; Aoua)
the Department of Biochemistry (A.B., M.R.W.), Vanderbilt University Medical School, Nashville, Tenn
BioVentures Inc (E.P.D., K.E.W., S.W.G.), Murfreesboro, Tenn
Department of Biostatistics (L.A.C., C.-Y.G., S.D.), Boston University School of Public Health, Boston, Mass
National Heart, Lung, and Blood Institute/Framingham Heart Study (C.J.O.), Framingham, Mass.
Abstract
Background— The CYP4A11 arachidonic acid monooxygenase oxidizes endogenous arachidonic acid (AA) to 20-hydroxyeicosatetraenoic acid (20-HETE), a metabolite with renovascular and tubular functions. Mice with targeted disruption of Cyp4a14, a murine homologue of CYP4A11, have severe hypertension. We combined molecular and biochemical approaches to identify a functional variant of the CYP4A11 20-HETE synthase and determine its association with hypertensive status in 2 independent human populations.
Methods and Results— A thymidine-to-cytosine polymorphism at nucleotide 8590 resulted in a phenylalanine-to-serine substitution at amino acid 434. Expression of cDNA with serine 434 resulted in a protein with a significantly reduced AA and lauric acid metabolizing activity. In a population of 512 whites from Tennessee, the age, body mass index, and gender-adjusted OR of having hypertension attributable to the 8590C variant was 2.31 (95% CI 1.41 to 3.78) compared with the reference 8590TT genotype. In subjects from the Framingham Heart Study, the adjusted ORs of hypertension associated with the 8590C variant were 1.23 (CI 0.94 to 1.59; n=1538) in all subjects and 1.33 (CI 1.01 to 1.77; n=1331) when subjects with diabetes were excluded. No association of the variant with hypertension was detected in a population of 120 blacks.
Conclusions— We identified a variant of the human CYP4A11 (T8590C) that encodes for a monooxygenase with reduced 20-HETE synthase activity. The association of the T8590C variant with hypertension supports its role as a polygenic determinant of blood pressure control in humans, and results obtained from the large population database suggest that the relevance of the variant may vary according to hypertension comorbidity.
Key Words: genetics ; hypertension, renal ; lipids ; metabolism
Introduction
Hypertension is a significant independent risk for cardiac, renal, and cerebrovascular morbidity and mortality.1 Buffered by evidence of a genetic basis for hypertension,2 much effort has been directed toward the identification of relevant genes, characterization of their variants, and determination of genetic and environmental interacting factors.3–5 The targeting of genes involved in renal sodium and fluid regulation has met with success in identifying monogenic variants, which has greatly enhanced our mechanistic understanding of hypertension,3 and the knockout of specific genes that have resulted in a hypertensive phenotype has expanded the list of potential hypertension gene candidates. Common gene variants shown to contribute to hypertension6–10 appear to vary according to gender,8 ethnicity,10–12 or associated comorbidities.13 Thus, the identification of polygenic determinants has not been straightforward.
See p 5
The CYP4A arachidonic acid (AA) monooxygenases oxidize endogenous AA to 20-hydroxyeicosatetraenoic acid (20-HETE). The 20-HETE metabolite has the capacity to act in either a prohypertensive or antihypertensive manner depending on its expression at renovascular or tubular sites, respectively.14–17 We recently reported the characterization of Cyp4a14 (–/–) mice as a model of gender-specific severe hypertension18 and now investigate the CYP4A human homologues as candidate genetic determinants in hypertension. In addition to CYP4A11, which encodes a renal 20-HETE synthase in humans,19,20 we and others recently identified a second CYP4A gene, CYP4A22.
The primary goal of the present study was to define and characterize genetic variants of 20-HETE synthase and to define their role in human hypertension. As a prerequisite, we addressed the question of whether CYP4A22 had enzymatic activity or a role in renal fatty acid metabolism. We present data that CYP4A22 lacks functional activity in the kidney. Therefore, we combined molecular and biochemical approaches to identify functional variants in CYP4A11 and test their relevance to hypertension in 2 independent human populations.
Methods
Subjects
Tennessee Cohort
Normotensive subjects were volunteers recruited through health screening examinations performed primarily at Vanderbilt University. Subjects were defined as normotensive if they had seated systolic and diastolic pressures of less than 140 and 90 mm Hg, respectively, and no history of identification or treatment for hypertension. Hypertensive subjects were ascertained from the Vanderbilt Hypertension Center and had either established hypertension, defined as antihypertensive treatment for >6 months, or a seated diastolic pressure >90 mm Hg on at least 3 occasions. Diastolic blood pressure was defined by the absence of Korotkoff sounds (phase V). Secondary hypertension was excluded by history and physical examination. Ethnicity was self-reported, and inclusion in a particular ethnic group generally required that a proband’s parents and grandparents recognized themselves by the same ethnic designation as that of the proband. All subjects gave written consent, and the Vanderbilt University Institutional Review Board approved the protocol. All genomic samples were coded to protect donor identity.
Framingham Offspring Cohort
The design of the Framingham Heart Study has been detailed previously.21 The offspring cohort comprises offspring (and their spouses) of the original Framingham participants (recruited in 1948). The Framingham Heart Study cohort is almost entirely white. Those in the offspring cohort were invited to have follow-up laboratory and physical and laboratory examinations approximately every 4 years starting in 1971. DNA samples were obtained for 1920 individuals from June 1987 to February 1991.21 Blood pressure measurements were the mean of 2 physician-obtained readings taken >5 minutes apart in the left arm of the seated subject with a mercury column sphygmomanometer, in which appearance of Korotkoff sounds (phase I) denoted systolic blood pressure and disappearance (phase V) determined diastolic blood pressure.8 Hypertension was defined as systolic blood pressure 140 mm Hg or diastolic blood pressure 90 mm Hg or current use of antihypertensive medication.8 Diabetes (type 2) was defined as use of hypoglycemic medication or fasting plasma glucose levels >7 mmol/L (>125 mg/dL) at any of 6 examinations. Participants considered in the present study (n=1880) are a subset of unrelated individuals who provided blood samples for DNA extraction at the sixth examination cycle (n=2933). Selection criteria were established to provide DNA samples of biologically unrelated subjects with approximately equal numbers of men and women who have characteristics similar to the remainder of the participants. Included in this report were 1538 (82%) of the 1880 subjects genotyped for T8590C. All subjects gave written consent, and the protocol was approved by the Boston University Medical Center Institutional Review Board. Further details of the Framingham Heart Study are available online (http://www.nhlbi.nih.gov/about/framingham).
Polymorphism Discovery and Genotyping
In addition to single base differences found throughout the gene sequences, CYP4A11 and CYP4A22 are distinguished by the presence in CYP4A11 of short nucleotide insertions (10 bp) in introns 1, 3, 4, and 11 and deletions (18 bp) in introns 4, 6, and 11.22 Polymerase chain reaction (PCR) amplification was performed with intron-specific primers designed to maximize the difference between the 2 genes and to generate overlapping CYP4A11 specific amplicons covering all 12 exons and intervening introns.22 The selectivity of this approach was corroborated by both sequence analysis and restriction enzyme digestion with sites unique to the CYP4A11 gene.22 Primer sequences, reaction conditions, DNA extraction, and genotyping methods can be found at www.bioventures.com. To define CYP4A11 variants, both forward and reverse sequencing of PCR products were performed in a screening subset (n=34) of predominantly hypertensive subjects divided equally by gender and ethnicity. Variants were numbered by nucleotide position in reference to the translation start site.
Biochemical Characterization
By reverse transcription-PCR of a mixture of total kidney RNAs isolated from 15 nonsexed individuals (Clontech, Palo Alto, Calif), we obtained a single cDNA encoding CYP4A22 (GenBank accession: AY280371). CYP4A11 cDNA was expressed and protein purified as described except the PD-10 column was replaced by hydroxyapatite and 1% Emulgen 913 was replaced by 0.02% sodium cholate.19 Membrane fractions were isolated by differential centrifugation.23 Partially purified proteins were dialyzed against Tris-Cl buffer (pH 7.5) and kept at –80° until use. Enzyme incubations were performed at 35°C in 0.05 mol/L Tris-Cl buffer (pH 7.5) containing 10 mmol/L MgCl2, 1 mmol/L GSH, 1 mmol/L NADPH, 50 μg/mL dilauroylphosphatidylcholine, 8 mmol/L isocitric acid, 0.5 IU/mL isocitric dehydrogenase, and 70 to 100 μmol/L [1-14C]labeled lauric or AA (20 μCi/mol).23 Purified recombinant P450s were mixed with a 10-molar excess of purified liver P450 reductase and an equimolar amount of purified cytochrome b5 and incubated for 15 minutes at room temperature before the addition of reaction components.23 Reaction products were resolved and quantified.19
Statistical Methods
To determine whether the CYP4A11 polymorphisms were in Hardy-Weinberg equilibrium, actual and predicted genotype counts were compared by a 2 test statistic with 2 degrees of freedom. Logistic regression models were used to compute the ORs for the dichotomous trait of hypertension (the presence or absence of hypertension) with the genotype indicator variable(s) as the primary independent variable. Three analytical models were used to consider the crude relationship between these variables and adjustment for age, gender, and body mass index (BMI). Analysis was performed for men and women separately given that we previously observed a significant effect of gender on hypertension in mice with targeted disruption of Cyp4a14, a murine homologue of CYP4A11. The analysis of the effect of diabetes on the CYP4A11 variant in the Framingham cohort was prespecified on the basis of animal24 and human25 data that suggest that insulin downregulates CYP4A expression, potentially impacting the effect of the variant. Diabetes prevalence information was not available for the Tennessee cohort. Descriptive data are presented as mean±SD for continuous variables and as proportions for categorical variables. ORs and 95% CIs are presented from logistic regression analyses. A 2-tailed probability value of 0.05 was considered significant. Analyses were done with SPSS for Windows (version 11.0, SPSS; Tennessee data) and SAS (release 8.2, SAS Institute Inc; Framingham data).
Results
CYP4A11 Is the Functional CYP4A 20-HETE Synthase in Humans
CYP4A11 has been cloned, expressed, and characterized as a renal 20-HETE synthase.19,20 However, the enzymatic activity of the other human CYP4A gene product, CYP4A22, and its role in renal fatty acid metabolism are unknown, and its mRNA is detected in the kidney only after reverse transcription-PCR amplification.22 Western blot analysis of the membrane preparations from cells expressing CYPs 4A11 and 4A22 cDNAs showed the presence of CYP4A11 with expected electrophoretic mobility in CYP4A11-transfected cells (Figure 1, lanes C and D). Similar but less intense immunoreactive bands were detected in membranes from CYP4A22-transfected bacteria (Figure 1, lane B). The absorption spectra of purified recombinant CYPs 4A11 and 4A22 showed the presence of a typical P450 Soret band at 409 nm (not shown). On reduction with Na2S2O4 and CO addition, only CYP4A11 generated a typical reduced P450 CO-bound complex (not shown), which suggests that the CYP4A22 cDNA encodes a nonfunctional protein. This was confirmed by incubation of the partially purified P450s with [1-14C]-AA. As shown by reverse-phase HPLC radiochromatograms (Figure 2), CYP4A11 but not CYP4A22 catalyzed the metabolism of AA to 20- and 19-HETE (90% and 10% of total products, respectively). Incubation of these proteins with lauric acid, a prototype CYP4A substrate,19,20,23 demonstrated that only CYP4A11 catalyzed the formation of 11- and 12-hydroxydocosanoic acids (not shown), which confirms that CYP4A22 lacks hydroxylase activity. Sequence comparisons showed that the glycine at position 130, conserved among all CYP4A isoforms, is replaced by serine in CYP4A22. Site-specific mutation of the CYP4A11 glycine residue 130 to serine resulted in a protein that lacked a functional heme group and was catalytically inactive (not shown).
CYP4A11 Polymorphism Discovery
Screening of CYP4A11 revealed 9 variants with allelic frequencies >1% (Table 1) and in Hardy-Weinberg equilibrium within ethnic groups (not shown). Additional variants with allele frequencies <1% and ethnic comparisons of variants are posted at www.bioventures.com. Of the 2 variants in the open reading frame, the cytosine-for-thymidine transition at nucleotide 8590 (exon 11) resulted in a nonsynonymous phenylalanine (F)-to-serine (S) substitution at residue 434 of CYP4A11, which was chosen for functional characterization.
Functional Effects of the CYP4A11 Gene Polymorphism
To determine the functional consequences of an F-to-S replacement at amino acid 434, we generated the corresponding cDNA by site-specific mutation, expressed it in Escherichia coli, partially purified it, and compared the fatty acid hydroxylating activities of the wild-type (F434) and mutated (S434) proteins. As shown in Table 2, compared with F434 (wild type), the S434 replacement reduced by more than half the 20-HETE synthase activity of CYP4A11, with only a small effect on its apparent Km values and thus, its affinity for AA (Table 2). Similarly, the apparent Vm of the lauric acid conversion to 12-hydroxylaurate was reduced by more than half, but for this fatty acid, there was also a significant reduction in apparent Km values (Table 2). These results suggest that the T8590C variant, which corresponds to the F434S protein polymorphism, affects the catalytic activity of the 20-HETE synthase through a loss-of-function mechanism.
The CYP4A11 8590 C Allele is Associated With Hypertension
Tennessee Cohort
To determine whether an association exists between the CYP4A11 variant and hypertension, each of 512 subjects were genotyped at the T8590C (also referred to as F434S) locus (blacks: hypertensive n=60, normotensive, n=60; and whites: hypertensive n=197, normotensive n=195). As displayed in Table 3, age was similar by blood pressure status and between gender in each ethnic group. BMI and mean arterial pressure were greater in hypertensive than in normotensive subjects but were not different between genders. All subjects with hypertension were receiving antihypertensive medications. Compared with the reference 8590TT genotype, there was no association of the 8590C allele with hypertension in blacks (Table 4). In whites, the 8590C allele was associated with an increased prevalence of hypertension after adjustment for age, gender, and BMI in both men and women (OR 2.31, 95% CI 1.41 to 3.78). No significant interaction between gender and genotype was detected, nor were there significant differences in age and BMI across genotype groups.
Framingham Offspring Cohort
To test the hypothesis raised in the initial study findings in an independent cohort and to further characterize the association of the CYP4A11 8590C (434S) variant with hypertension, we characterized this variant in the large population-based Framingham cohort (n=1538). Table 5 shows baseline characteristics of male and female subjects.21 Approximately 50% of the men and 40% of the women were hypertensive, and 60% to 75% of them were taking antihypertensive medications. Table 6 shows that the OR of having hypertension attributable to the 8590C allele was 1.21 (95% CI 0.93 to 1.56) after adjustment for age, BMI, and gender and was 1.23 (95% CI 0.94 to 1.59) with a model that additionally adjusted for cigarette and alcohol use. After exclusion of diabetics, the comparable ORs were 1.31 (95% CI 0.99 to 1.74) and 1.33 (95% CI 1.01 to 1.77), respectively. ORs tended to be greater in men than in women, but no significant interaction between gender and CYP4A11 genotypes was detected.
Discussion
In the present study, we have provided evidence that although CYP4A11 functions as a 20-HETE synthase, CYP4A22 does not contribute to renal 20-HETE biosynthesis. Furthermore, to determine the relevance of genetic variation in CYP4A11 to hypertension, we identified a coding polymorphism (T8590C), biochemically characterized 8590C as a loss-of-function variant, and demonstrated its association with hypertension in whites. Initial findings of associations between biologically plausible functional variants are often not replicated in follow-up studies, and even when associations are replicated, they are often less strong than in the initial validation cohort.27 For this reason, we conducted a study of replication in a population-based cohort of whites. Our results suggest the existence of a modest association finding in the Framingham cohort. Clearly, our findings strongly support further study of the relationship of variants in this gene with hypertension in other populations.
Both animal14,16 and human28,29 data support CYP4A11 as a candidate gene in hypertension. Numerous studies in the spontaneously hypertensive and Dahl salt-sensitive rats have shown an association between CYP4A 20-HETE synthase expression levels and/or activity and (1) altered nephron-specific transport of sodium chloride and (2) increased blood pressure.14,16 These studies support the contention that 20-HETE exerts prohypertensive and antihypertensive actions that are critically dependent on the site of biosynthesis and action along the nephron.14–17 In the renal tubule, 20-HETE blocks sodium transport and functions primarily as a natriuretic, antihypertensive substance.16,17 In contrast, 20-HETE acting in the renal vasculature has vasoconstricting, prohypertensive properties.16 Analyses of CYP4A expression and of 20-HETE synthase activity demonstrated abundant CYP4A expression and enzyme activity in the proximal and distal tubule, 2 segments of the rat nephron identified as targets for the natriuretic, antihypertensive actions of 20-HETE.14–17 The relevance of these observations to the pathophysiology of human hypertension has been highlighted by measurements of urinary sodium and 20-HETE excretion that show the ability of dietary salt to regulate 20-HETE excretion28 and impairment of 20-HETE-dependent natriuretic mechanisms in salt-sensitive compared with salt-resistant hypertensive patients.28 Moreover, recent studies of the natriuretic response to furosemide in humans suggested that 20-HETE modulates its natriuretic effects and that altered 20-HETE natriuresis is linked to human salt-sensitive hypertension.29
The use of the Framingham cohort allowed us to address the effects of gender and phenotypic context. Significant gender differences have long been recognized in both the prevalence of hypertension and target-organ damage attributable to increased blood pressure, the mechanisms of which are largely unknown.30 The rationale to investigate the effect of gender was prompted by the finding that hypertension in Cyp4a14 (–/–) mice, a human homologue of CYP4A11, is sexually dimorphic and androgen sensitive.18 Although Cyp4a14 does not metabolize arachidonic acid, Cyp4a14 (–/–) mice had an upregulation of Cyp4a12 that resulted in increased 20-HETE synthase activity. In contrast, such an overt effect of gender was not detected in the present study, and although CYP4A11 accounts for significant renal 20-HETE synthase in humans, the role that androgens play in regulating its renal expression is yet to be defined. Nevertheless, the possibility exists that additional, non-CYP4A-dependent -hydroxylases may contribute to 20-HETE biosynthesis in the human kidney.17
The well-characterized nature of the Framingham cohort allowed us to analyze the effect of the CYP4A11 variant without a potential confounding influence of diabetes. Both animal24 and human studies25 have suggested that diabetes induces and insulin suppresses CYP4A expression. We hypothesized that the effects of a loss-of-function CYP4A variant may be muted in the context of altered glucose/insulin regulation (eg, diabetes). That contention is supported by the Framingham data, which show a significant association of the CYP4A11 variant with hypertension in the subpopulation without diabetes. These data serve to focus future validation studies and demonstrate the utility of a large, well-characterized genetic resource in the characterization of polygenic effects underlying hypertension. The apparent differences in the relationship of the variant and hypertension between the Tennessee and Framingham cohorts are not known but may relate to population-specific factors. The prevalence of diabetes in the Tennessee cohort was not known, and hypertensive subjects tended to be younger, heavier, and to have higher mean blood pressures than those in the Framingham cohort.
The lack of a significant association of the CYP4A11 variant and hypertension in blacks is somewhat surprising, despite an increased allelic frequency of the CYP4A11 variant in blacks compared with whites, given the potential link of 20-HETE to salt-sensitive hypertension28,29 and the reported greater prevalence of salt-sensitive hypertension in blacks.31 This fact could be due to a limited sample size of blacks or possibly to population stratification, a limitation inherent to the population case-control study design.32 Given the data in the present study that establish a significant loss of CYP4A11 metabolizing activity with the 8590C variant, there exists the possibility that other genetic or physiological factors modify the activity of the CYP4A system in this ethnic group. Further studies are required to confirm the lack of association of the CYP4A11 variant in blacks and to address CYP4A regulation.
In summary, we have identified a coding variant of the human CYP4A11 gene that results in a CYP4A11 protein with reduced 20-HETE synthase activity and that has a greater prevalence in hypertensive compared with normotensive whites. These results, in conjunction with published studies showing abundant renal CYP4A11 expression23 and antihypertensive, natriuretic effects of 20-HETE,16,17 provide compelling evidence that a genetically controlled alteration in CYP4A11 expression or activity plays a role as a determinant of polygenic blood pressure control in humans.
Acknowledgments
This work was supported by DK38226 (Dr Capdevila), DK28350 (Dr Waterman), HL04221, RR00095 (Dr Gainer), HL65193, HL67308, HL60906 (Dr Brown), and N01-HC-25195 (Dr O’Donnell). cDNAs were sequenced using facilities supported by the National Cancer Institute (CA68485).
Footnotes
Drs Gainer and Bellamine contributed equally to this article.
References
Stokes J III, Kannel WB, Wolf PA, D’Agostino RB, Cupples LA. Blood pressure as a risk factor for cardiovascular disease: the Framingham Study: 30 years of follow-up. Hypertension. 1989; 13: I-13–I-18.
Ward R. Familial aggregation and genetic epidemiology of blood pressure. In: Laragh JH, Brenner BM, eds. Hypertension Pathophysiology, Diagnosis, and Management. New York, NY: Raven Press; 1990: 81–100.
Lifton RP, Gharavi AG, Geller DS. Molecular mechanisms of human hypertension. Cell. 2001; 104: 545–556.
Pratt RE, Dzau VJ. Genomics and hypertension: concepts, potentials, and opportunities. Hypertension. 1999; 33: 238–247.
Halushka MK, Fan JB, Bentley K, Hsie L, Shen N, Weder A, Cooper R, Lipshutz R, Chakravarti A. Patterns of single-nucleotide polymorphisms in candidate genes for blood-pressure homeostasis. Nat Genet. 1999; 22: 239–247.
Jeunemaitre X, Soubrier F, Kotelevtsev YV, Lifton RP, Williams CS, Charru A, Hunt SC, Hopkins PN, Williams RR, Lalouel JM. Molecular basis of human hypertension: role of angiotensinogen. Cell. 1992; 71: 169–180.
Caulfield M, Lavender P, Farrall M, Munroe P, Lawson M, Turner P, Clark AJ. Linkage of the angiotensinogen gene to essential hypertension. N Engl J Med. 1994; 330: 1629–1633.
O’Donnell CJ, Lindpaintner K, Larson MG, Rao VS, Ordovas JM, Schaefer EJ, Myers RH, Levy D. Evidence for association and genetic linkage of the angiotensin-converting enzyme locus with hypertension and blood pressure in men but not women in the Framingham Heart Study. Circulation. 1998; 97: 1766–1772.
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 U S A. 2002;%19; 99: 3872–3877.
Gainer JV, Brown NJ, Bachvarova M, Bastien L, Maltais I, Marceau F, Bachvarov DR. Altered frequency of a promoter polymorphism of the kinin B 2 receptor gene in hypertensive African-Americans. Am J Hypertens. 2000; 13: 1268–1273.
Atwood LD, Kammerer CM, Samollow PB, Hixson JE, Shade RE, MacCluer JW. Linkage of essential hypertension to the angiotensinogen locus in Mexican Americans. Hypertension. 1997; 30: 326–330.
Jain S, Tang X, Narayanan CS, Agarwal Y, Peterson SM, Brown CD, Ott J, Kumar A. Angiotensinogen gene polymorphism at-217 affects basal promoter activity and is associated with hypertension in African-Americans. J Biol Chem. 2002; 277: 36889–36896.
Tiago AD, Samani NJ, Candy GP, Brooksbank R, Libhaber EN, Sareli P, Woodiwiss AJ, Norton GR. Angiotensinogen gene promoter region variant modifies body size-ambulatory blood pressure relations in hypertension. Circulation. 2002; 106: 1483–1487.
Carroll MA, McGiff JC. A new class of lipid mediators: cytochrome P450 arachidonate metabolites. Thorax. 2000; 55 (suppl 2): S13–S16.
Capdevila JH, Falck JR. The CYP P450 arachidonic acid monooxygenases: from cell signaling to blood pressure regulation. Biochem Biophys Res Commun. 2001; 285: 571–576.
McGiff JC, Quilley J. 20-Hydroxyeicosatetraenoic acid and epoxyeicosatrienoic acids and blood pressure. Curr Opin Nephrol Hypertens. 2001; 10: 231–237.
Roman RJ. P-450 metabolites of arachidonic acid in the control of cardiovascular function. Physiol Rev. 2002; 82: 131–185.
Holla VR, Adas F, Imig JD, Zhao X, Price E Jr, Olsen N, Kovacs WJ, Magnuson MA, Keeney DS, Breyer MD, Falck JR, Waterman MR, Capdevila JH. Alterations in the regulation of androgen-sensitive Cyp 4a monooxygenases cause hypertension. Proc Natl Acad Sci U S A. 2001; 98: 5211–5216.
Hoch U, Zhang Z, Kroetz DL, Ortiz de Montellano PR. Structural determination of the substrate specificities and regioselectivities of the rat and human fatty acid omega-hydroxylases. Arch Biochem Biophys. 2000; 373: 63–71.
Imaoka S, Ogawa H, Kimura S, Gonzalez FJ. Complete cDNA sequence and cDNA-directed expression of CYP 4A11, a fatty acid omega-hydroxylase expressed in human kidney. DNA Cell Biol. 1993; 12: 893–899.
Kannel WB, Feinleib M, Mcnamara PM, Garrison RJ, Castelli WP. Investigation of coronary heart disease in families: Framingham Offspring Study. Am J Epidemiol. 1979; 110: 281–290.
Bellamine A, Wang Y, Waterman MR, Gainer JV III, Dawson EP, Brown NJ, Capdevila JH. Characterization of the CYP 4A11 gene, a second CYP4A gene in humans. Arch Biochem Biophys. 2003; 409: 221–227.
Helvig C, Dishman E, Capdevila JH. Molecular, enzymatic, and regulatory characterization of rat kidney cytochromes P450 4A2 and 4A3. Biochemistry. 1998; 37: 12546–12558.
Shimojo N, Ishizaki T, Imaoka S, Funae Y, Fujii S, Okuda K. Changes in amounts of cytochrome-P450 isozymes and levels of catalytic activities in hepatic and renal microsomes of rats with streptozocin-induced diabetes. Biochem Pharmacol. 1993; 46: 621–627.
Laffer CL, Laniado-Schwartzman M, Nasjletti A, Elijovich F. 20-HETE and circulating insulin in essential hypertension with obesity. Hypertension. 2004; 43: 388–392.
Capdevila JH, Falck JR, Dishman E, Karara A. Cytochrome-P-450 arachidonate oxygenase. Methods Enzymol. 1990; 187: 385–394.
Lohmueller KE, Pearce CL, Pike M, Lander ES, Hirschhorn JN. Meta-analysis of genetic association studies supports a contribution of common variants to susceptibility to common disease. Nat Genet. 2003; 33: 177–182.
Laffer CL, Laniado-Schwartzman M, Wang MH, Nasjletti A, Elijovich F. Differential regulation of natriuresis by 20-hydroxyeicosatetraenoic acid in human salt-sensitive versus salt-resistant hypertension. Circulation. 2003; 107: 574–578.
Laffer CL, Laniado-Schwartzman M, Wang MH, Nasjletti A, Elijovich F. 20-HETE and furosemide-induced natriuresis in salt-sensitive essential hypertension. Hypertension. 2003; 41: 703–708.
Reckelhoff JF. Gender differences in the regulation of blood pressure. Hypertension. 2001; 37: 1199–1208.
Grim CE, Robinson M. Blood pressure variation in blacks: genetic factors. Semin Nephrol. 1996; 16: 83–93.
Ziv E, Burchard EG. Human population structure and genetic association studies. Pharmacogenomics. 2003; 4: 431–441.(James V. Gainer, MD; Aoua)