Share this article on:

A genome-wide association study-derived candidate gene seeks replication: STK39

Persu, Alexandrea; Vikkula, Miikkab

doi: 10.1097/HJH.0b013e328344b6b3
Editorial commentaries

aDivision of Cardiology, Cliniques Universitaires Saint-Luc, Belgium

bLaboratory of Human Molecular Genetics, de Duve Institute, Université Catholique de Louvain, Brussels, Belgium

Correspondence to Alexandre Persu, Division of Cardiology, Cliniques Universitaires Saint-Luc, Université Catholique de Louvain, 10 Avenue Hippocrate, 1200 Brussels, Belgium Tel: +32 2 764 28 88; fax: +32 2 764 89 80; e-mail:

In 2007–2008, genome-wide association studies (GWASs) have led to the discovery of a host of new loci involved in the pathogenesis of various complex diseases [1]. This ‘genomic gold rush’ [1] has continued relentlessly since then [2]. In contrast with earlier approaches, in many cases, the associations found using GWAS were successfully replicated. An example of such a ‘success story’ was the multireplicated association of the 9p21 locus with coronary artery disease [3,4]. This locus was subsequently associated with other vascular phenotypes, such as sudden and/or arrythmic cardiac death, peripheral artery disease, carotid plaques, abdominal aortic and cerebral aneurysms and – though less consistently – ischemic stroke [5]. Accordingly, in a statement of the American Heart Association, Arnett et al. [6] considered that GWAS might well become the primary research methodology for the discovery of cardiovascular disease genes in the near future.

However, as far as hypertension is concerned, the GWAS approach has not lived up to its expectations [7,8]. Indeed, in a GWAS of 14 000 cases of seven common diseases and 3000 shared controls, one or several association signals were found for all conditions studied apart from hypertension [9]. Potential explanations for these disappointing results include poor coverage of many genes by the Affymetrix chips used in the study and – more importantly – use of controls from the general population that were not assessed for blood pressure (and may, thus, have included up to 25% hypertensive individuals). In addition, hypertension may have fewer common risk alleles of larger effect size than other complex diseases [8,10]. Finally, essential hypertension likely represents a mix of different poorly characterized subsets with specific underlying mechanisms, which may still decrease the statistical power to find an effect.

The first positive result of a GWAS in hypertension, published early in 2009, was the identification of STK39 as a possible hypertension susceptibility gene [11]. Several polymorphisms of this serine/threonine kinase gene were found to be associated with systolic blood pressure (SBP) and diastolic blood pressure (DBP) in a cohort of 542 participants from the Amish Family Diabetes Study. The association was replicated in one Amish and four non-Amish Caucasian samples, reaching a suggestive significance level (P < 10−6) in a meta-analysis including all studies (n = 7125). Interestingly, in the Amish population, the two lead single nucleotide polymorphisms (SNPs) (rs6749447 and rs3754777) were in almost complete linkage disequilibrium with a SNP located in a conserved region of intron 2, namely rs35929607. The latter was considered functional, as the minor G allele was associated with enhanced transcription of the gene in vitro [11].

After phosphorylation by the WNK1 and WNK4 kinases, the product of the STK39, a serine/threonine kinase named SPAK, is able to phosphorylate the furosemide-sensitive Na+/K+/2Cl channel (NKCC2) and the thiazide-sensitive Na+/Cl channel (NCC) [12,13]. The WNK–SPAK–NKCC2/NCC pathway plays, thus, a critical role in the regulation of sodium reabsorption. Furthermore, mutations in the WNK1 and WNK4 genes are at the origin of pseudohypoaldosteronism type II (Gordon syndrome), characterized by hypertension and hyperkalemia, whereas mutations in the genes coding for the NKCC2 and NCC channels are responsible for Bartter type 1 and Gitelman syndrome, respectively [12]. Finally, SPAK knockout mice display a Gitelman-like phenotype and impaired vasoconstriction [14]. The implication of variants from genes belonging to this pathway – including STK39– in salt sensitivity and hypertension is, thus, supported by a strong rationale.

Despite these encouraging elements, further studies did not confirm unequivocally the implication of STK39 in genetic determination of blood pressure. Although in a recent GWAS performed in African–Americans, several SNPs of STK39 were found to be associated with both SBP and DBP – though by far not at the genome-wide significance level [15] – in the two largest GWAS performed in hypertension so far, totalizing more than 60 000 individuals [16,17], no such association was reported. Moreover, Ho et al. [18] attempted replication in a subset of the Women's Health Study. Unfortunately, the STK39 locus was not included in detailed analysis because it did not meet prespecified replication criteria for GWAS (i.e. the β-coefficients and standard errors were not available in the initial publication). Finally, Cunnington et al. [19] found no association between the three SNPs located within STK39– including the functional SNP rs35929607 – and either office or ambulatory blood pressure in a cohort of 1372 members of British Caucasian families.

In this issue of the Journal of Hypertension, Fava et al. [20] tested the hypothesis of an association of the previously identified functional SNP of STK39 (rs35929607) with blood pressure and hypertension in two Swedish cohorts, the Malmö Diet and Cancer (MDC-CVA) (n = 5634) and the Malmö Preventive Project (MPP) (n = 17 894), both at baseline and (for MPP) at follow-up after a mean of 23 years. The minor G allele was significantly associated with higher SBP and DBP in the MDC-CVA cohort, but not in the larger MPP cohort, either at baseline or at follow-up. In both cohorts, the prevalence of hypertension was higher in individuals harboring the G allele, both at baseline and, in the MPP cohort, at follow-up. Nevertheless, after exclusion from the MPP cohort of 2398 individuals who also participated in the MDC-CVA study, these associations were no more significant. Furthermore, higher associations were mainly due to the female subset (and, in post-hoc analysis, were even larger in women taking hormone replacement therapy). Finally, retrospective genotyping of a small cohort of individuals assessed for salt sensitivity disclosed a larger blood pressure decrease in response to salt restriction in participants harboring the less frequent G allele.

The study of Fava et al. [20] is the second study specifically designed to replicate the association of STK39 variants with blood pressure in Caucasians. It is more than 15 times larger than that of Cunnington et al. [19], population rather than family-based and includes longitudinal data. As a whole, it provides patchy evidence in favor of a modest effect of STK39 on hypertension prevalence and possibly incidence. Such an effect might be larger and/or easier to demonstrate in more homogeneous populations, such as Amish [11] or African–Americans [15], sharing an increased proportion of genetic and/or environmental factors (including those related to salt diet and salt sensitivity).

It appears more and more evident that the – otherwise large – heritability of blood pressure is likely to be split between multiple genetic determinants. While most candidates identified in GWAS account for no more than 1–2 mmHg of blood pressure [11,17,18], at the individual level, constructing scores including the best ‘hits’ from GWAS might make more sense than looking for a few selected SNPs [17]. Admittedly, however, the cumulative effect of the top 10 SNPs associated with blood pressure in the Cohorts for Heart and Aging Research in Genomic Epidemiology (CHARGE) Consortium

[17] explains no more than 1% of blood pressure variance, contrasting with the 30–40% heritability of the trait. The apparent paradox of ‘missing heritability’ [21] remains, thus, unsolved.

Another maybe sounder approach – intermediate between somehow naive single-SNP association studies of the 1990s and huge multinational ‘fishing expeditions’ – would be to test in the same cohort the effect of multiple genes belonging to a specific metabolic pathway. Accordingly, the interest of analyzing biological pathways in GWAS, rather than focusing on the analysis of single markers, has been recently highlighted [22,23].

As most genes involved in rare monogenic forms of hypertension [24] and several promising candidates identified in GWAS (including STK39 and, more recently, UMOD) [11,25] code for proteins involved in sodium reabsorption, we anticipate that exhaustive genetic analysis of a comprehensive panel of renal salt-handling genes (including identification of rare mutations by resequencing) [6,26] in a large, well phenotyped cohort might prove a fruitful approach.

Back to Top | Article Outline


1 Topol EJ, Murray SS, Frazer KA. The genomics gold rush. JAMA 2007; 298:218–227.
2 Hindorff LA, Junkins HA, Hall PN, Mehta JP, Manolio TA. A catalog of published genome-wide association studies. Bethesda, Maryland: National Human Genome Research Institute; 2008. [Accessed January 15, 2011]
3 Samani NJ, Erdmann J, Hall AS, Hengstenberg C, Mangino M, Mayer B, et al. Genomewide association analysis of coronary artery disease. N Engl J Med 2007; 357:443–453.
4 Palomaki GE, Melillo S, Bradley LA. Association between 9p21 genomic markers and heart disease. JAMA 2010; 303:648–656.
5 Roberts R, Wells GA, Stewart AFR, Dandona S, Chen L. The genome-wide association study: a new era for common polygenic disorders. J Cardiovasc Transl Res 2010; 3:173–182.
6 Arnett DK, Baird AE, Barkley RA, Basson CT, Boerwinkle E, Ganesh SK, et al. Relevance of genetics and genomics for prevention and treatment of cardiovascular disease. Circulation 2007; 115:2878–2901.
7 Rafiq S, Anand S, Roberts R. Genome-wide association studies of hypertension: have they been fruitful? J Cardiovasc Transl Res 2010; 3:189–196.
8 Zhang K, Weder AB, Eskin E, O'Connor DT. Genome-wide case/control studies in hypertension: only the ‘tip of the iceberg'. J Hypertens 2010; 28:1115–1123.
9 The Wellcome Trust Case Control Consortium. Genome-wide association study of 14 000 cases of seven common diseases and 3000 shared controls. Nature 2007; 447:661–678.
10 Padmanabhan S, Melander O, Hastie C, Menni C, Delles C, Connell JM, Dominiczak AF. Hypertension and genome-wide association studies: combining high fidelity phenotyping and hypercontrols. J Hypertens 2008; 26:1275–1281.
11 Wang Y, O'Connell JR, McArdle PF, Wade JB, Dorff SE, Shah SJ, et al. Whole-genome association study identifies STK39 as a hypertension susceptibility gene. Proc Natl Acad Sci U S A 2009; 106:226–231.
12 Richardson C, Alessi DR. The regulation of salt transport and blood pressure by the WNK-SPAK/OSR1 signalling pathway. J Cell Sci 2008; 121:3293–3304.
13 Welling PA, Chang YPC, Delpire E, Wade JB. Multigene kinase network, kidney transport, and salt in essential hypertension. Kidney Int 2010; 77:1063–1069.
14 Yang SS, Lo YF, Wu CC, Lin SW, Yeh CJ, Chu P, et al. SPAK-knockout mice manifest Gitelman syndrome and impaired vasoconstriction. J Am Soc Nephrol 2010; 21:1868–1877.
15 Adeyemo A, Gerry N, Chen G, Herbert A, Doumatey A, Huang H, et al. A genome-wide association study of hypertension and blood pressure in African Americans. PLoS Genet 2009; 5:1–11.
16 Newton-Cheh C, Johnson T, Gateva V, Tobin MD, Bochud M, Coin L, et al. Genome-wide association study identifies eight loci associated with blood pressure. Nat Genet 2009; 41:666–676.
17 Levy D, Ehret GB, Rice K, Verwoert GC, Launer LJ, Dehghan A, et al. Genome-wide association study of blood pressure and hypertension. Nat Genet 2009; 41:677–687.
18 Ho JE, Levy D, Rose L, Johnson AD, Ridker PM, Chasman DI. Discovery and replication of novel blood pressure genetic loci in the Women's Genome Health Study. J Hypertens 2011; 29:62–69.
19 Cunnington MS, Kay C, Avery PJ, Mayosi BM, Koref MS, Keavney B. STK39 polymorphisms and blood pressure: an association study in British Caucasians and assessment of cis-acting influences on gene expression. BMC Med Genet 2009; 10:135.
20 Fava C, Danese E, Montagnana M, Sjogren M, Almgren P, Engström G, et al. Serine/threonine kinase 39 is a candidate gene for primary hypertension especially in women: results from two cohort studies in Swedes. J Hypertens 2011; 29:484–491.
21 Manolio TA, Collins FS, Cox NJ, Goldstein DB, Hindorff LA, Hunter DJ, et al. Finding the missing heritability of complex diseases. Nature 2009; 461:747–753.
22 Torkamani A, Topol EJ, Schork NJ. Pathway analysis of seven common diseases assessed by genome-wide association. Genomics 2008; 92:265–272.
23 Wang K, Li M, Hakonarson H. Analysing biological pathways in genome-wide association studies. Nature 2010; 11:843–854.
24 Lifton RP, Gharavi AG, Geller DS. Molecular mechanisms of human hypertension. Cell 2001; 104:545–556.
25 Padmanabhan S, Melander O, Johnson T, Di Blasio AM, Lee WK, Gentilini D. Genome-wide association study of blood pressure extremes identifies variant near UMOD associated with hypertension. PLoS Genet 2010; 6:1–11.
26 Ji W, Foo JN, O'Roak BJ, Zhao H, Larson MG, Simon DB, et al. Rare independent mutations in renal salt handling genes contribute to blood pressure variation. Nat Genet 2008; 40:592–599.
© 2011 Lippincott Williams & Wilkins, Inc.