Associations of the Serum/Glucocorticoid Regulated Kinase Genes With BP Changes and Hypertension Incidence: The Gensalt Study

Abstract

BACKGROUND

Single-marker and novel gene-based methods were employed to examine the associations of the serum/glucocorticoid regulated kinases (SGK) gene family with longitudinal blood pressure (BP) changes and hypertension incidence in a family-based cohort study.

METHODS

Totally, 1,768 Chinese participants from the Genetic Epidemiology Network of Salt Sensitivity (GenSalt) follow-up study were included in the current analyses. Nine BP measures were obtained at each of 3 visits during the GenSalt follow-up study. Mixed-model and Gene-based analyses were used to examine the associations of the SGK gene family with longitudinal BP phenotypes. Bonferroni correction was applied to account for multiple testing.

RESULTS

After an average 7.2-year follow-up, 32.2% (513) of participants free of hypertension at baseline developed hypertension. Four novel SNPs in the SGK1 gene were predictive of the longitudinal BP phenotypes. The major alleles of SGK1 rs1763498 and rs114414980 conferred 2.9- and 2.5-fold increased risks of hypertension development, respectively (P = 1.0×10⁻⁴ and 6.0×10^–4, respectively). In addition, the major allele of SGK1 rs229133 was significantly associated with 0.4mm Hg larger annual increases in systolic BP (P = 4.2×10⁻⁴), while the major allele of rs6924468 was significantly associated with 0.2mm Hg smaller annual increases in diastolic BP (P = 4.2×10⁻⁴). Gene-based analyses revealed an association of the SGK1 gene with risk of hypertension development (P = 7.4×10⁻³). No evidence for the SGK2 and SGK3 genes was found.

CONCLUSIONS

The findings of the current study suggest that the SGK1 gene may play a role in long-term BP regulation and hypertension incidence.

China’s aging population, increased urbanization, and Westernization have contributed to the emergence of hypertension and related cardiovascular diseases as major public health challenges in this country.^1–5 An estimated 33.6%, or 335.8 million Chinese adults, had hypertension in 2010,⁴ and a total of 2.33 million cardiovascular deaths were attributable to increased blood pressure (BP) in 2005.⁵ As a complex trait, BP is influenced by environmental and genetic factors and their interactions. Previous heritability studies have shown that roughly 31–68% of interindividual variation in BP is determined by genetic background.⁶ Although numerous genetic studies have identified loci influencing BP regulation, the heritability of this complex trait remains largely unexplained.

The serum/glucocorticoid regulated kinases (SGK) family plays a pivotal role in determining sodium balance, blood volume, and BP level through its regulation of Na+ transport in the kidney.^7,8 The SGK family consists of 3 isoforms encoded by the SGK1, SGK2, and SGK3 genes, all of which are highly expressed in the kidney.^9–11 Previous studies have shown that SGK1 can regulate renal Na+ reabsorption through a variety of receptors including the epithelial Na+ channel, Na+/H+ exchanger 3, Na+ K+ ATP, and Na+−Cl− cotransporter.^8,12–14 In support of these findings, physiology studies have reported that SGK family genes are correlated with renal Na+ retention and BP changes in mice. Regardless of NaCl intake, SGK1 knock-out mice showed significantly higher urine Na (+) excretion than wild-type mice.^15,16 Furthermore, compared to SGK1 knock-out mice, wild-type mice had higher BP increases in response to high-salt diet and smaller BP decreases in response to low-salt diet.^17,18 Recent in vitro and in vivo studies showed that SGK1 can promote tissue inflammation and cell senescence, providing a new insight into BP regulation.^19,20 Similar to SGK1, SGK2, and SGK3 can also influence Na+/H+ exchanger 3 activity and potentially regulate Na+ transport.⁸ SGK’s biological relevance along with evidence from in vitro and in vivo studies make the SGK genes important candidates for genetic study of BP phenotypes. However, only 3 studies have tested the association between the SGK gene family and BP, with inconsistent findings.^21–23 Further, none have examined whether variants in the SGK gene family can predict longitudinal BP changes and hypertension incidence in the Chinese population. The sparse evidence from human study highlights the importance of examining the association between the SGK gene family and longitudinal BP phenotypes.

In the present study, both single-marker and gene-based analyses were employed to examine the associations of common variants in the SGK gene family with longitudinal BP changes and hypertension incidence among 1,768 Han Chinese participants of the Genetic Epidemiology Network of Salt Sensitivity (GenSalt) follow-up study.

METHODS

Study population

The GenSalt study is a family-based dietary feeding study conducted in 6 rural villages located in north China and designed to examine the interaction of genes and sodium intake on BP. Detailed information on the GenSalt study design and methods has been published previously.²⁴ In brief, a community-based BP screenings was carried out among persons aged 18–60 years in the study villages to identify potential probands. Those with mean systolic BP (SBP) of 130–160mm Hg and/or diastolic BP (DBP) of 85–100mm Hg and no use of antihypertensive agents were recruited, as well as their parents, siblings, spouse, and offspring. Individuals who had stage 2 hypertension, secondary hypertension, and a history of clinical cardiovascular disease or diabetes or were pregnant, heavy alcohol drinkers, or currently on a low-sodium diet were excluded from the study.

Written informed consents were obtained from all the GenSalt participants after detailed explanation of the study. The study has been approved by the Institutional Review Boards at all the participating institutions.

Baseline data collection

The first GenSalt study visit consisted of a 3-day baseline examination and subsequent 21-day dietary intervention that was carried out in 2003–2005. During the 3-day baseline examination, information about family pedigree, demographic characteristics, personal and family medical history, and lifestyle risk factors was obtained using a standard questionnaire which was administered by trained staff. Three morning BP measurements were obtained according to a standard protocol on each of the 3 days of the baseline examination. All BP readings were measured by trained and certified technicians using a random-zero sphygmomanometer.²⁵ BP was measured with the participant in the sitting position after 5 minutes of rest. Additionally, participants were advised to avoid alcohol, cigarette smoking, coffee/tea, and exercise for at least 30 minutes before their BP measurements. Mean BP at baseline was calculated as the mean of 9 measurements from the 3-day baseline examination. Body weight and height were measured twice in light indoor clothing without shoes. Body mass index (BMI) was calculated as kilograms per square meter (kg/m²).

Follow-up data collection

As part of the GenSalt follow-up study, 2 additional study visits were carried out in 2008–2009 and 2011–2012. Each follow-up visit consisted of a 3-day clinical examination which employed the same standardized protocol as that of the baseline examination. Again, mean BP was calculated as the average of 9 BP measurements obtained during each of the two follow-up examinations. Hypertension was defined as having a SBP of 140mm Hg and/or DBP of 90mm Hg and/or use of antihypertensive agents.

Among 1,906 eligible participants from 633 families who participated in the GenSalt baseline examination, 117 participants were missing BP data at both follow-up studies and another 21 participants were missing genotype data. The remaining 1,768 participants (92%) were included in the current analysis.

Genotype, imputation, and quality control

Three genes in the SGK gene family (SGK1, SGK2, and SGK3) were selected based on their potential biological effect on BP regulation. Within the 3 candidate genes (including ±5,000 bps), 82 SNPs were genotyped on the Affymetrix 6.0 platform (Affymetrix, Santa Clara, CA). To provide better coverage of the 3 candidate gene regions, imputation was conducted based on the phase I version 3 1000 G reference panel using minimac software.²⁶ Prior to imputation, standard quality control including correction of Mendelian inheritance errors and filtering of SNPs based on deviations from Hardy–Weinberg equilibrium, minor allele frequency, and call rate were conducted. Postimputation quality control included the further removal of SNPs with R ² values <0.50 and minor allele frequency <0.01. Tag-SNPs were selected from the imputed dataset using pairwise r ² thresholds of less than 0.8 and HaploView software.²⁷ After linkage disequilibrium pruning, a total of 101 SNPs in 3 genes remained for the current analysis. Characteristics of the SNPs, including information on location, genotype, and quality are presented in Supplementary Table 1. The linkage disequilibrium structure for SNPs in each of the 3 genes are shown in Supplementary Figures 1–3.

Statistical analysis

Characteristics of the GenSalt study participants at baseline and during follow-up were summarized as means ± SD or percentages, as appropriate. Due to the longitudinal, family-based GenSalt design, mixed-effect regression models were used to test the associations between SNPs and longitudinal BP changes and hypertension incidence.^28,29 Autoregressive and compound symmetry covariance matrices were used to accommodate the correlations of repeated measurements within individuals and of individuals within families, respectively. For the examination of additive associations between SNPs and longitudinal BP changes, a genotype by follow-up time interaction term and the main effects of these variables were included in the models as fixed effects. Additionally, models were adjusted for the fixed effects of age, gender, and BMI using the PROC MIXED procedure in SAS (version 9.3; SAS Institute, Cary, NC). The following mixed effects model was employed:

$$\begin{array}{ll}{\gamma }_{ijk}=& {\beta }_0+{\beta }_1\times {\text{ age}}_{ij}+{\beta }_2\times {\text{gender}}_{ij}+{\beta }_3\\ & \times {\text{ BMI}}_{ij}+{\beta }_4\times {\text{ SNP}}_{ij}+{\beta }_5\times {\text{ time}}_{ijk}+{\beta }_6\\ & \left({\text{SNP}}_{ij}\times {\text{time}}_{ijk}\right)+a_i+ b_{ij}+ e_{ijk}\end{array}$$

where, γ _ijk represents the BP value for the ith individual in the jth family at the kth visit; β ₀ is the mean BP after accounting for covariates, genetic effects, and the interaction term; the terms age _ij, gender _ij, and BMI _ij represent baseline age, gender, and BMI of the ith individual in the jth family, respectively; SNP _ij models the genetic main effect, where the genotype is coded under an additive model; time _ijk is the follow-up time from baseline for the ith individual in the jth family at the kth visit; and the SNP _ij × time _ijk term is the linear interaction between follow-up time and genetic effects. This term was modeled as a fixed effect parameter and used to test the association of each SNP with longitudinal BP change. The random effects terms a _i and b _ij account for the relatedness among individuals in the same family as well as the correlations of repeated measurements among individuals nested within families. The last term represents the residual.

For examination of the additive associations between SNPs and hypertension incidence, a multilevel logistic regression model was employed after excluding 173 participants with hypertension at baseline using the PROC GLIMMIX procedure in SAS.³⁰ Hypertension incidence was modeled as:

$$\begin{array}{ll}{\gamma }_{ijk}=& {\beta }_0+{\beta }_1\times {\text{ age}}_{ij}+{\beta }_2\times {\text{ gender}}_{ij}+{\beta }_3\\ & \times {\text{ BMI}}_{ij}+{\beta }_4\times {\text{ SNP}}_{ij}+{\beta }_5\\ & \times {\text{ time}}_{ijk}+ a_i+ b_{ij}+ e_{ijk}\end{array}$$

where γ _ijk represents hypertension status of the ith individual in the jth family at the kth visit; β ₀ is the log odds of hypertension after accounting for covariates, genetic effects; the terms age _ij, gender _ij, and BMI _ij respresent baseline age, gender, and BMI of the ith individual in the jth family, respectively; SNP _ij models the genetic effect, where the genotype is coded under an additive model; and time _ijk is the follow-up time from baseline for the ith individual in the jth family at the kth visit. P values for β ₄ were used to assess the significance of the association of each SNP with hypertension incidence.

The overall association of each candidate gene with longitudinal BP changes and hypertension incidence was evaluated using the truncated product method (TPM), which combines P values from single-marker association analyses.^31,32 Due to its ability to accommodate correlations among SNPs and increased power to detect gene-based associations compared to other meta-analysis methods, the TPM is preferred for combining P values across genes or genomic loci. To evaluate associations between genes and longitudinal BP changes, the P value for the genotype by follow-up time interaction term was used. For hypertension incidence, the P value for the genotype term was used. The truncation point was set as τ = 0.05, and the P value for the TPM was estimated by 100,000 simulations. Sensitivity analyses were carried out using the TPM after excluding significant SNPs within a gene (identified by single-marker analyses) to examine their influence on the gene-based analysis. Gene-based analysis was conducted using R software (version 3.0.1; http://www.r-project.org. Accessed on 9/17/2016). Additionally, in a sensitivity analysis to evaluate the robustness of findings from the TPM, the Gene-based Association Test that uses Extended Simes procedure (GATES)³³ was also employed.

Bonferroni correction was used to adjust multiple testing in single-marker and gene-based analyses.³⁴ The α-thresholds for single-marker and gene-based analyses were 4.95×10⁻⁴ (0.05/101) and 0.017 (0.05/3), respectively.

RESULTS

General characteristics of the 1,768 participants at each of the 3 visits are presented in Table 1. In summary, the participants were 39 years of age and had a BMI of 23.4kg/m² and an SBP and DBP of 116.9 and 73.8mm Hg at baseline, respectively. Among them, 924 were male and 173 had hypertension. During a mean of 7.2-year follow-up, the average SBP and DBP increased approximately 12mm Hg and 8mm Hg, respectively, and 32.2% of participants free of hypertension at baseline developed this condition.

Table 1.

General characteristics of 1,768 GenSalt participants during the follow-up study

	Baseline (N = 1768)	Visit 1 (N = 1687)	Visit 2 (N = 1626)
Age, mean ± SD, years	39±9.2	43.9±9.2	46.7±9.2
Male, n (%)	924 (52.3)	878 (52.1)	845 (52.0)
BMI, mean ± SD, kg/m2	23.4±3.2	24.1±3.4	24.8±3.5
SBP, mean ± SD, mm Hg	116.9±14.1	122.4±16.8	129.1±17.3
DBP, mean ± SD, mm Hg	73.8±10.2	78.9±11	82.2±11.1
Hypertension, n (%)	173 (9.8)	431 (25.6)	650 (40.0)
Hypertension incidence, n (%)a	—	265 (16.6)	513 (32.2)
Follow-up time, mean ± SD, years	—	4.4±1.2	7.2±0.9

Abbreviations: BMI, body mass index; DBP, diastolic blood pressure; SBP, systolic blood pressure.

^aExcludes participants with hypertension at baseline (N = 173).

The associations of 101 tag-SNPs in 3 SGK genes with longitudinal BP changes and hypertension incidence are shown in Figure 1 and Supplementary Table 1. Table 2 presents significant SNPs associated with longitudinal BP phenotypes. SKG1 SNPs rs1763498 and rs114414980 were significantly and marginally associated with hypertension incidence after Bonferroni correction for multiple comparisons (P = 1.0×10⁻⁴ and 6.0×10⁻⁴, respectively). Each copy of the major alleles of rs1763498 and rs114414980 were correlated with 2.86- and 2.46-folds increased risk of hypertension development. Although there were not significant associations between the 2 SNPs with longitudinal BP changes, a similar trend was observed for average BP change per year. Additionally, SGK1 rs229133 was significantly associated with longitudinal changes in SBP after Bonferroni correction for multiple comparisons (P = 4.2×10⁻⁴). The major allele was associated with 0.43mm Hg greater increases in SBP. Furthermore, the association between SGK1 rs6924468 and longitudinal changes in DBP was also significant after Bonferroni correction for multiple comparisons (P = 4.2×10⁻⁴). The major allele was associated with 0.23mm Hg lesser annual increases in DBP. Although not statistically significant, similar trends in the associations of rs229133 and rs6924468 were observed for the other BP phenotypes.

Figure 1.

Log₁₀ P values for the association of 101 single nucleotide polymorphisms (SNPs) in SGK1, SGK2, and SGK3 with longitudinal changes in systolic and diastolic blood pressure, as well as hypertension incidence. The gray and white circles show P values for the testing of genotype by follow-up time interactions for systolic blood pressure and diastolic blood pressure, respectively, while the black circles show P values for the testing of the effect of SNPs on hypertension incidence. Four labeled SNPs were significantly or marginally associated with longitudinal BP phenotypes after Bonferroni correction for multiple testing.

Table 2.

Association of top SNPs with BP changes and hypertension incidence among GenSalt study participants

							SBP		DBP		HTN
Gene	CHR	SNP	Position	CA	OA	CAF	β (SE)	P	β (SE)	P	OR (95% CI)	P
SGK1	6	rs1763498	134546106	C	T	0.96	0.32 (0.18)	0.08	0.21 (0.14)	0.14	2.86 (1.67, 4.89)	1.0×10−4a
	6	rs114414980	134567989	C	T	0.97	0.23 (0.19)	0.22	0.17 (0.16)	0.28	2.46 (1.47, 4.11)	6.0×10−4b
	6	rs6924468	134580214	G	A	0.86	−0.17 (0.10)	0.09	−0.23 (0.06)	4.2×10−4a	0.81 (0.65, 1.01)	0.07
	6	rs229133	134642370	T	C	0.93	0.43 (0.12)	4.2×10−4a	0.22 (0.09)	0.01	0.91 (0.67, 1.26)	0.58

Abbreviations: CA, code allele; CAF, coded allele frequency; CHR, chromosome; DBP, diastolic blood pressure; HTN, hypertension incidence; OA, other allele; SBP, systolic blood pressure; SNP, single nucleotide polymorphism.

^aSignificant after Bonferroni adjustment.

^bMarginally significant after Bonferroni adjustment.

Additional gene-based analyses found that SGK1 was significantly associated with risk of hypertension development (P = 7.4×10⁻³) after Bonferroni correction for multiple testing (Table 3). SGK1 also showed a nominally significant association with longitudinal DBP changes (P = 0.02). Sensitivity analyses showed that the associations between SGK1 and hypertension incidence and longitudinal DBP changes were attenuated (P = 0.25 and 0.11, respectively) after removing corresponding SNPs identified in single-marker analyses. Neither of SGK2 and SGK3 was associated with any longitudinal BP phenotypes. None of the 3 genes was associated with longitudinal SBP changes. Similar results were obtained using the GATES analyses.

Table 3.

Gene-based associations of 3 SGK genes with longitudinal BP phenotypes in the GenSalt study

Gene	SBP TPM/GATES	DBP TPM/GATES	HTN TPM/GATES
SGK1	0.56/0.03	0.02/0.03	7.4×10−3a/7.0×10−3a
SGK3	0.15/0.92	0.16/0.75	0.20/0.35
SGK2	0.63/0.82	0.41/0.98	0.77/1.00

Abbreviations: DBP, diastolic blood pressure; GATES, Gene-based Association Test that uses Extended Simes procedure; HTN, hypertension incidence; SBP, systolic blood pressure; TPM, truncated product method.

^aSignificant after Bonferroni adjustment.

DISCUSSION

To our knowledge, the current study is the first to examine the association between the SGK gene family and longitudinal BP changes and hypertension incidence in the Han Chinese population. We identified 3 novel SGK1 SNPs significantly associated with longitudinal BP phenotypes and another one which was marginally significant after stringent correction for multiple testing. In the current analysis, the major alleles of SNPs rs1763498 and rs114414980 predicted greater risk of developing hypertension. In addition, the major allele of rs229133 was significantly associated with larger longitudinal increases in SBP, while the major allele of rs6924468 was significantly associated with smaller longitudinal increases in DBP. Similar to the single-marker findings, gene-based analyses documented a significant association of SKG1 with risk of hypertension development. There was no evidence of associations between the SGK2 and SGK3 genes and variants with any of the longitudinal BP phenotypes. In summary, these findings highlight the potential importance of SGK1 as a regulator of BP and contribute to the accumulating evidence of a genomic underpinning for BP regulation and hypertension.

Our study provides the first evidence of an influence of the SGK1 gene on longitudinal BP phenotypes in the Chinese population. Similar to the findings of the current analysis, a previous study conducted in a population of European ancestry also suggested a role for SGK1 in longitudinal BP change.²¹ Wowern et al. identified 2 SNPs (rs1743966 and rs1057293) that were associated with percent change in BP during an 11.2-year follow-up, findings which had been identified previously for the usual BP phenotype by Busjahn et al. ²³ These 2 variants were tested but not replicated by the GenSalt study. However, rs1743966 and rs1057293 are not likely to be functional, representing intronic and synonymous exonic variations, respectively. Given the lack of obvious functional relevance, these 2 variants may not causally associate with BP and may instead be in linkage disequilibrium with the true causal variants. If the true causal variants are on different haplotypes in East Asians and Europeans, replication would be unlikely in East Asians. Furthermore, these variants could interact with environmental factors or other genetic variants that differ between populations, which would be another potential explanation for the lack of replication. The consistent identification of variants in SGK1 in distinct populations implicate this gene in BP regulation. Furthermore, physiological studies have demonstrated the involvement of SGK1 in regulating tissue inflammation with known contribution to hypertension pathophysiology.^19,20 In aggregate, these data provide strong evidence for a role of SGK1 in hypertension development.

In concordance with single-marker analyses, gene-based analyses revealed a significant association of the SGK1 gene with risk of hypertension development. However, sensitivity analysis showed that the association was mainly driven by variants identified in the single-marker analyses. Since the current analysis provided excellent coverage of common variation in the SGK1 gene (with imputation based on 1000 G data), these data do support modest effects of additional common variants in this gene on the longitudinal BP phenotypes. Therefore, sequencing of the linkage disequilibrium regions of identified SNPs may be helpful to identify the causal variants underlying the observed association signals.

Neither single-marker nor gene-based analyses identified associations of the SGK2 and SGK3 genes with longitudinal BP phenotypes in our study. Although the evidence from association studies of these genes in humans is sparse,^7,35 their role in regulating renal Na+ transport has been documented by a variety of in vivo and in vitro studies.^8,36,37 While the current study provides no evidence for association of common variants in these genes, we cannot rule out the possibility of rare variant or structural variant associations not examined here.

The current study has several important strengths. It is the first study to examine the association between the SGK gene family and “longitudinal BP phenotypes” in a Han Chinese cohort. The homogenous genetic background of GenSalt participants makes the analyses robust to population stratification. In addition, stringent quality control in genotyping and BP measurements, the use of 9 repeated BP measures at each study visit, and the high-hypertension event rate, which is in line with recent epidemiologic data from the same region of China,^38,39 may have enhanced the power to detect associations. Furthermore, genotype imputation based 1000 G provided outstanding coverage of common and low-frequency variants in the SGK gene regions. However, some potential limitations should be noted. After a mean of 7.2 years, 8% of participants were lost to follow-up. Compared to those who completed the study, those lost were similar in gender and BMI but slightly younger (36.5 vs. 39 years) with slightly lower SBP (114 vs. 116mm Hg) and DBP (71.3 vs. 73.6mm Hg). Given these small differences and the high-response rates (92%), we believe losses-to-follow-up should have very minimal influence on our findings. Caution should also be taken when interpreting the effect size of identified SNPs due to the possibility of the winner’s curse. As such, replication of the identified SNPs in an East Asian population is needed to confirm our findings and obtain more accurate estimates of SNP-BP effect sizes. In addition, because only common and low-frequency variation were assessed, any rare and structural variants influencing BP would be missed by the current study. Therefore, we cannot exclude the possibility of SGK2 or SGK3 associations with longitudinal BP phenotypes.

In conclusion, our study identified novel variants in the SGK1 gene that significantly associated with longitudinal BP phenotypes. While gene-based findings also supported an association between SGK1 and longitudinal BP phenotypes, these findings were primarily driven by loci identified in the single-marker analyses. We observed no associations of SGK2 and SGK3 with longitudinal BP phenotypes. The findings contribute important information toward better understanding the genomic mechanisms and biological pathways underlying BP regulation. Such information may prove useful for development of novel pharmaceutical options in the treatment and control of BP, and for target preventive intervention to people with high risk of hypertension. Future studies are still needed to replicate our findings. Furthermore, functional studies will be important for identifying the causal variants driving the observed SGK1 gene signals.

SUPPLEMENTARY MATERIAL

Supplementary materials are available at American Journal of Hypertension (http://ajh.oxfordjournals.org).

DISCLOSURE

The authors declared no conflict of interest.