- Email: [email protected]
The impact of endothelin-1 genetic analysis and job strain on ambulatory blood pressure
Journal of Psychosomatic Research 71 (2011) 97–101
Contents lists available at ScienceDirect
Journal of Psychosomatic Research
The impact of endothelin-1 genetic analysis and job strain on ambulatory blood pressure☆,☆☆ Sheldon W. Tobe a,⁎, Brian Baker b, Katie Hunter a, Alex Kiss c, Nancy Perkins a, Lissette Gomez d, Yu Feng d, Karen Wigg d, Cathy L. Barr d,e a
Division of Nephrology, Sunnybrook Health Sciences Centre, Toronto, Canada Department of Psychiatry, University Health Network, Toronto, Canada Institute for Clinical Evaluative Sciences, Toronto, Canada d Genetics and Development Division, Toronto Western Research Institute, University Health Network, Toronto, Canada e Program in Neurosciences and Mental Health, The Hospital for Sick Children, Toronto, Canada b c
a r t i c l e
i n f o
Article history: Received 16 May 2010 Received in revised form 1 January 2011 Accepted 4 January 2011 Keywords: Ambulatory blood pressure Endothelin-1 Job strain
a b s t r a c t Objective: An interaction between the endothelin-1 gene (EDN1), blood pressure (BP) and social determinants has been previously found. Using a well-characterized cohort of participants, the impact of associations between genetic factors and job strain on BP was evaluated. Methods: A cross-sectional analysis of five polymorphisms covering the EDN1, of which 2 were previously reported to be associated with BP, was performed. Study subjects had previously completed a baseline evaluation including 24-h ambulatory BP monitoring and an assessment of job strain. This report presents the findings for 184 subjects who gave DNA samples for genetic analysis. One-way analysis of variance (ANOVA) was performed between each genetic marker and 24-h systolic blood pressure (SBP) and diastolic blood pressure (DBP), as well as two-way ANOVAs to test the interaction effect with job strain. Results: Trends for relationships were observed between SBP and two polymorphisms: rs10478694 and rs5369. An interaction between job strain and those heterozygous for two polymorphisms showed higher SBP (P=.029 and .008) and a tendency for higher DBP. All findings were more significant when analyses were confined to the 139 Caucasian subjects. Conclusion: This is the first study to report an interaction between the EDN1 gene, job strain and BP, supporting previous evidence of a role of this gene in the interaction between environmental stress and ambulatory BP. Given the limited sample size, the results should be considered preliminary, and further studies are required. © 2011 Elsevier Inc. All rights reserved.
Introduction More than one in five adults develops hypertension, the most significant modifiable risk factor for heart disease and stroke. The onset of hypertension is influenced by both environmental and genetic factors; however, the interplay of these factors and of their underlying mechanisms remains unclear. Understanding these factors may help to identify and manage individuals at risk. Amongst
Abbreviations: ABPM, ambulatory blood pressure monitoring; BMI, body mass index; DBP, diastolic blood pressure; ET-1, endothelin-1; EDN1, endothelin-1 gene; JCQ, job content questionnaire; JS, job strain; SBP, systolic blood pressure. ☆ Division of Nephrology and Genetics and Development Division, University of Toronto. ☆☆ Funds for this study were provided by a grant from the Hospital for Sick Children Psychiatric Endowment Fund and the Heart and Stroke Foundation of Ontario. ⁎ Corresponding author. Division of Nephrology, Sunnybrook Health Sciences Centre A240 2075 Bayview Avenue, Toronto, ON, Canada M4N 3M5. Tel.: +1 416 480 6901; fax: +1 416 480 6940. E-mail address: [email protected] (S.W. Tobe). 0022-3999/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.jpsychores.2011.01.003
vasopressor factors that might be implicated in the development of essential hypertension, endothelin-1 (ET-1) expression has been shown to be linked to resistance artery remodeling and so may confer a vulnerability to the development of hypertension [1]. Endothelin-1 is best known as a potent regulator of vasoconstriction, and there is now evidence that indicates that it has a number of diverse functions [2], including a role in the central nervous system [2,3]. Endothelin-1 has been clearly associated with hypertension in animal and pharmaceutical models [4,5]; however, clinical evidence demonstrating a direct association with hypertension and plasma levels of ET-1 has been less straightforward [4,6]. A number of known confounding factors may come into play in these studies, including gonadal hormones [7–9] and ethnicity [7]; thus, the composition of the samples with regard to gender and ethnic background may influence the results. Furthermore, as 80% of ET-1 secretion is abluminal (towards the vessel wall), plasma levels may not reflect activation. Mental [10,11] and physiological stressors have been shown to influence ET-1 plasma levels [12], and the impact of these factors can be difficult to control. Genetic background may also be a factor, with
98
S.W. Tobe et al. / Journal of Psychosomatic Research 71 (2011) 97–101
one study finding increases in plasma ET-1 after a mental stressor only in offspring of hypertensive parents [12]. Based on the evidence implicating ET-1 in hypertension, the ET-1 gene (EDN1) has been the focus of a number of genetic association studies, with association findings for resting blood pressure (BP), mostly limited to specific populations such as obese subjects [13–15]. The specific genetic changes in EDN1 that have been associated with BP include the rs5370 polymorphism located in exon 5 that changes the amino acid at codon 198 from lysine to asparagine (Lys198Asn) and the A nucleotide insertion/deletion polymorphism located at position 138 (rs10478694). The Lys198Asn polymorphism has been the focus of the majority of studies to date, with several studies reporting association of the T allele coding for asparagine with higher BP [13–16]. Treiber et al. [16] demonstrated that while carrier status of the T allele (Asn) was not related to resting BP, there was a significant interaction between this allele and BP reactivity when exposed to a stressor, suggesting an interaction of environmental factors with genetic variation in this gene contributing to higher BP. Among the study cohort, the greatest change in BP was seen in those individuals with the T allele (Asn) status from lower socioeconomic backgrounds. More recently, an interaction was demonstrated between anger expression, T(Asn) allele carrier status and total peripheral resistance index while exposed to a stressor such that those T allele carriers with poor anger management skills demonstrated the greatest total peripheral resistance index reactivity consistent with an endothelin effect [17]. Job strain, a construct of excessive work demands and limited control in the work setting, is an environmental stressor that we and others have shown to be associated with elevations of ambulatory BP in men and women [18–23]. The recently completed Double Exposure study examined the interaction over time of both job strain and marital cohesion in normotensive and unmedicated hypertensives [22]. Job strain was strongly associated with higher systolic blood pressure (SBP) even controlling for related covariates, while martial cohesion alone was not [24]. The Double Exposure cohort presented an opportunity to determine whether there was an interaction between the role of genetic factors and the environmental stressor job strain on BP. To look for interactions, we chose the two polymorphisms of EDN1 previously associated with BP and three others to cover the major haplotypes.
Methods Subjects The Double Exposure study examined the effects of job strain and marital cohesion on ambulatory BP in unmedicated normotensive or hypertensive (BPb160/100 mmHg, by the BPTru device [VSM Medical, Vancouver, BC, Canada]) male and female subjects aged 40–65 who were employed and cohabiting with a significant other [22]. At the baseline study visit, questionnaires were completed including the Job Content Questionnaire (JCQ) (described below), and ambulatory BP was measured over 24 h (monitor 90207l Spacelabs Medical Inc., Redmond, WA, USA). Blood pressure was recorded every 15 min during the day and every hour during the night (11 PM–7 AM), and readings were averaged; more detailed methodology has been reported [22,24]. Ambulatory subjects were recruited by advertisements posted around an academic health sciences centre in Toronto, enrolled into the Double Exposure study from July 2001 until July 2003 and followed for 1 year. Three years following completion of the study, subjects were contacted, requesting their participation in the genetic study, and sample collection was completed by July 2008. At the follow-up visit, 184 subjects provided a buccal swab for DNA extraction and genetic analysis.
All research was approved by the Research Ethics Boards for the Hospital for Sick Children and Sunnybrook Health Sciences Centre and was performed in accordance with the Declaration of Helsinki. All participants provided written consent following oral and written explanation of the research. Job Content Questionnaire The JCQ [25,26] is a well-validated, reliable, self-administered instrument designed to measure social and psychological characteristics of work including job strain, which refers to those individuals with high occupational demands and low job control [26,27] and which has been extensively used in relation to cardiovascular parameters including ambulatory BP [18–23]. Job strain (JS) was defined as those individuals in the 20th percentile for high job demands and low job latitude from the JCQ [18–20,27]. Genotyping All DNA samples were collected by extraction from cheek cells (buccal samples) using a standard protocol. Five markers were identified a priori to be genotyped. These included two polymorphisms which have previously been demonstrated to be associated with BP: the Lys198Asn polymorphism (rs5370) [13–16] and the 138ins/delA polymorphism (rs10478694) [28–29]. Either of these two DNA changes may be contributing to BP; alternatively, these polymorphisms may be in linkage disequilibrium with the as yet unknown functional change(s) that contributes to increased BP. Because the functional DNA change(s) is not clearly identified, we genotyped three additional, markers (rs5369, rs1476046, rs2071943) chosen to detect the major haplotypes of the gene (http://www. hapmap.org/cgi-perl/gbrowse/gbrowse/hapmap/) to ensure adequate coverage of the genetic information. Polymorphisms were genotyped using the 5′ nuclease assay (TaqMan assay) [30]. Primers were designed to amplify the region on either side of the polymorphism, and specific probes were designed for each allele and labeled with a florescent tag that was measured on the ABI 7900 (Applied Biosystems, Foster City, CA, USA). Statistical analyses Descriptive statistics were calculated for all variables of interest. Continuous measures such as age were summarized using means and standard deviations, whereas categorical measures were summarized using counts and percentages. Baseline 24-h SBP and diastolic blood pressure (DBP) values were used for all analyses. Job strain was again scored as a binary variable for those in the 20th percentile for high job demands and low job latitude from the JCQ. One-way analysis of variance (ANOVA) was performed for each genetic marker separately and BP (both systolic and diastolic). Each polymorphism was analyzed categorically for each of the three genotype categories for the oneway ANOVA. Two-way ANOVA was also performed to test the interaction effect of the genetic marker and job strain on both SBP and DBP. Homozygous individuals for the less common allele were not compared using the two-way ANOVA models because of low sample numbers, which result in insufficient statistical power to fit such a model. The small sample numbers similarly do not allow the model to control for possible confounders such as age, gender etc. All analyses were performed using SAS statistical software, version 9.1 (SAS Institute Inc., Cary, NC, USA). Results Of the original Double Exposure participants at baseline, 184 (74%) participants provided DNA samples. Of those participating in this study, 39 (21.7%) were classified as having job strain. Mean 24-h ambulatory BP for the subjects participating in the
S.W. Tobe et al. / Journal of Psychosomatic Research 71 (2011) 97–101
Female (%) Age Caucasian (%) BMI Smoking (%) Alcohol (N10 drinks/week) (%) ABPM, mmHg Job strain (%)
Subjects in study (n=184)
Subjects not participating (n=68)
100 (54.3) 50.9±6.0 139 (75.5) 26.7±4.5 15 (8.2) 21 (11.8) 121.3±10.1/ 75.6±7.8 39 (21.7)
37 (54.7) 50.6±7.4 52 (81.2) 26.6±4.0 5 (7.8) 7 (10.9) 120.5±11.5/ 74.1±7.1 13 (20.6)
All data are presented as mean±S.D. unless otherwise indicated. No statistically significant differences between those participating and those not participating. genetics study was 121.1±10.4 mmHg/75.2±7.6 mmHg. Of the participants, 100 (54.3%) of the subjects were female, 139 (75.5%) were Caucasian and overall there were low rates of smoking and alcohol consumption (Table 1). Baseline characteristics were not significantly different for those not participating in the genetics study. Genotype distributions were in accordance with Hardy–Weinberg equilibrium. One-way ANOVA for genotype on SBP approached statistical significance for two of the five markers: rs5369 (P=.065) and rs10478694 (P=.052), one of the markers previously associated with elevated BP. For the rs5369 marker, SBP was lowest in the individuals homozygous for the A allele (111.7±7.6 mmHg). However, there were only three individuals with this genotype; thus, it is not possible to make conclusions concerning this genotype. The SBP for the other two genotypes was similar with the slightly higher SBP for those subjects heterozygous (GA) for the rs5369 marker as compared to those homozygous for the G allele (123.8±1.6 mmHg vs. 120.8±0.8 mmHg, respectively). For the rs10478694 polymorphism, SBP was lowest in those homozygous for the minor allele, the insertion allele (118.7±7.3 mmHg); however, the number of individuals in this class was low (n=11). Systolic BP was slightly higher for those subjects heterozygous for the insertion allele as compared to those subjects homozygous for the deletion (123.3±0.96 mmHg vs. 120.5±1.3 mmHg, respectively). The three remaining markers were not significantly associated with BP. For the one-way ANOVA for DBP and genotype, the rs10478694 polymorphism also showed a trend for an association (P=.089), with individuals homozygous for the A nucleotide insertion allele showing lower DBP than the other two genotypes (Table 2). The other markers were not significant. To determine whether there was an interaction with job strain, a two-way ANOVA was performed (Table 3). Because of low numbers of participants homozygous for the A allele for the rs5369 polymorphism (three subjects) and for the insertion allele of the rs10478694 polymorphism (11 subjects), these were not included. This analysis revealed an interaction between job strain and genotype on SBP for both rs5369 (P=.008) and rs10478694 (P=.029; Figs. 1 and 2, respectively) such that, within the low job strain group, the SBP was very similar between heterozygotes and those homozygous for the common allele. However, within the job strain group, there was a significant difference between the two genotype classes, with those heterozygous showing higher BP and the homozygotes for the common allele showing lower BP than their counterparts with the same genotypes in the low job strain group (Table 3). The same pattern emerged for the two-way ANOVA with DBP, revealing a significant interaction between job strain and genotype for rs5369 (P=.016) and a trend for rs10478694 (P=.124; Figs. 3 and 4, respectively, and Table 4). For both markers in the group with low job strain, the heterozygotes and homozygotes had very similar DBP, whereas in those with job strain, the heterozygotes showed higher DBP than their counterparts in the low job strain category and the homozygotes for the common allele that had lower DBP. A further analysis was performed for the 139 Caucasian participants as there have been reports of genetic heterogeneity amongst ethnic groups, and the results were more significant (data not shown). The non-
Table 3 Mean SBP (mmHg) by job strain and genotype.
rs5369 AG GG
No job strain
Job strain
121.9 [118.31–125.49] (n=31) 121.4 [119.61–123.19] (n=108)
131.3 [124.30–138.30] (n=8) 118.9 [115.32–122.48] (n=30) F=7.29, P=.008
rs10478694 A/del 122.4 [119.39–125.41] (n=46) del/del 121.1 [118.90–123.30] (n=84)
127.4 [122.00–132.80] (n=14) 117.7 [113.70–121.70] (n=25) F=4.86, P=.029
Data presented as mean and confidence intervals.
Caucasians were of mixed ethnicities (Black, Asians, East Indian), and there were insufficient participants to perform the analysis in the non-Caucasian groups.
Discussion While a wealth of information from animal models and pharmacological interventions supports the role of ET-1 as a factor in BP, the direct relationship of ET-1 plasma levels to hypertension is still controversial [4]. Genetic studies of the relationship of EDN1 with resting BP have also been mixed, with associations found most consistently in restricted populations such as the obese. The results from this study build on previous evidence for EDN1 as an important genetic factor in the risk for the development of high BP, particularly in combination with job strain, an environmental stressor which has a documented impact on BP. Furthermore, these results indicate that genetic studies that do not take such stressors into consideration may underestimate the contribution of this gene to BP and may partly account for the conflicting findings from these studies. Interesting information is also provided on the EDN1 polymorphisms associated with BP. Previous research has predominantly focused on the Lys198Asn polymorphism as a possible risk factor, with the T allele associated with increased BP in several studies of European, Australian, Japanese and American subjects [13–15]. 132 130
SBP (mmHg)
Table 1 Baseline demographics for subjects who completed the genetics follow-up and those that did not.
99
128 126 124 Homozygous GG
122
Heterozygous AG
120
Interaction P=.008
118 116 No JS
JS
rs5369 Fig. 1. Systolic blood pressure by genotype (rs5369) and job strain.
128 Table 2 Mean SBP and DBP (mmHg) by genotype.
rs10478694 A/A (n=11) A/del (n=61) del/del (n=112)
Systolic
Diastolic
111.66 [102.88–120.44] 123.82 [120.59–127.06] 120.85 [119.18–122.53] F=2.77, P=.065
71.00 [65.92–76.08] 76.26 [73.67–78.85] 75.46 [74.17–76.75] F=0.68, P=.509
118.7 [114.30–123.10] 123.85 [121.11–126.59] 120.21 [118.36–122.06] F=3.01, P=.052
72.18 [66.75–77.61] 77.08 [75.11–79.05] 75.09 [73.65–76.53] F=2.46, P=.089
Data presented as mean and 95% confidence intervals.
SBP (mmHg)
rs5369 AA (n=3) AG (n=39) GG (n=142)
126 124 122
Homozygous for deletion
120
Heterozygous
118
Interaction P=.029
116 No JS
JS
rs10478694 Fig. 2. Systolic blood pressure by genotype (rs10478694) and job strain.
100
S.W. Tobe et al. / Journal of Psychosomatic Research 71 (2011) 97–101
82
DBP (mmHg)
80 78 Homozygous GG
76
Heterozygous AG
74
Interaction P=.016 72 No JS
JS
rs5369 Fig. 3. Diastolic blood pressure by genotype (rs5369) and job strain.
80
DBP (mmHg)
79 78 77 76
Homozygous for deletion
75
Heterozygous
74
Interaction P=.124
73 72 No JS
JS
rs10478694 Fig. 4. Diastolic blood pressure by genotype (rs10478694) and job strain.
Currently, there is no known change in function related to the alternative alleles of the Lys198Asn polymorphism. The Lys198Asn polymorphism is located in preproendothelin but is cleaved from the mature protein, though it could still play a role in the function of ET-1 by influencing messenger RNA (mRNA) or protein stability. However, a study in hypertensive patients did not find a significant difference in plasma levels of ET-1 in individuals with the different alleles [31], and no differences in protein expression could be found in cell supernatants from cells transfected with the two variants [31]. We found no evidence for an association of BP with the Lys198Asn marker possibly due to the low numbers in this study. Results were only significant for the rs10478694 polymorphism and a marker in linkage disequilibrium with rs10478694, rs5369. The ins/del polymorphism is located at a position 138 bp downstream of the transcription initiation site (in some reports described as 134 bp) in the first untranslated exon [28]. The insertion allele has been associated with increased levels of ET-1 in plasma [31]. Specifically, individuals heterozygous for the insertion allele showed higher plasma ET-1 levels than those homozygous for the deletion allele [31]. Endothelin-1 levels of individuals who were homozygous for the insertion allele were not reported in that paper [31]. The insertion allele has been associated with increased expression in luciferase gene expression assays that may be due to increased stability of the mRNA
Table 4 Mean DBP (mmHg) by job strain and genotype. No job strain
Job strain
rs5369 AG GG
74.9 [71.99–77.81] (n=31) 75.9 [74.44–77.36] (n=108)
81.4 [77.23–85.57] (n=8) 73.9 [70.94–76.86] (n=30) F=5.95, P=.016
rs10478694 A/del del/del
76.4 [74.07–78.73] (n=46) 75.6 [74.01–77.19] (n=84)
78.7 [74.96–82.44] (n=14) 73.4 [70.08–76.71] (n=25) F=2.39, P=.124
Data presented as mean and 95% confidence intervals.
[32], consistent with the increased levels of ET-1 in plasma of individuals heterozygous for the insertion allele, but not with findings of lower BP in individuals with this allele [29]. Increased stability of the mRNA may result in higher levels of the protein, thus predicting higher BP; however, RNA levels do not necessarily predict protein levels, and high protein levels may not directly correlate with plasma levels or BP. Indeed, studies of the mouse knockout model observed a paradoxical increase in BP in mice having one copy of the gene and reduced plasma ET-1 [2]. Our data in the one-way analyses show lower SBP and DBP in the 11 subjects who were homozygous for the insertion allele and marginally higher BP in individuals heterozygous for the insertion allele compared to those homozygous for the deletion allele. Thus, our results are not consistent with a simple role of increased expression of the gene predicted by the insertion allele from the in vitro expression studies and higher ambulatory BP. However, when job strain was included in the analyses, we then observed that individuals heterozygous for this allele showed higher BP than their low job strain counterparts and that those homozygous for the deletion allele showed lower BP than both genotype classes in the low job strain category. One possibility is that the insertion allele, or an allele in linkage disequilibrium with the allele, is responsible for the induction of the gene by psychological stressors. Unfortunately, all of the 11 individuals homozygous for the insertion allele were in the low job strain category, and we were not able to analyze this category in the two-way analyses. A similar picture on the relationship of this gene and BP and an interaction effect with job strain was observed with the alleles of the marker rs5369, a synonymous DNA change in exon 3 (Glu106Glu). With job strain, heterozygous individuals showed increased BP, while those homozygous for the common allele showed decreased BP, compared to their counterparts with the same genotype in the low job strain categories. To the best of our knowledge, this variant has not been studied in any functional assay; thus, currently, we have no information to implicate these alleles in the function of the gene. This marker is in linkage disequilibrium (D'=0.77, r2=0.285) with the rs10478694 marker, and the results may reflect the association with the rs10478694. Alternatively, the results from both markers may reflect association with an as yet unidentified functional allele elsewhere in the gene. If these findings can be replicated, they might provide a method of identifying individuals at risk of developing hypertension from job strain and raise the possibility of psychogenomic interventions. A limitation of our study was the sample size that did not allow us to investigate the homozygous classes for the minor alleles in the twoway analyses; this also resulted in the loss of power from these 14 subjects. A much larger study will be required to have a sufficient sample to test the homozygous classes and would also provide more power for the three markers that were not significant. It will be interesting to see whether people with job strain and homozygous for the polymorphisms rs5369 (AA) and rs10478694 (A/A) have higher BP as suggested from this study. Due to sample size, we were not able to analyze the data by other confounders like age, social economic status or gender. Sex hormones modulate ET-1 levels [7–9], and increases in ET-1 plasma levels were previously reported to be greater in males than females in response to video game and cold stressors [7]; thus, the males in the sample may contribute more to the genetic findings than females, but a larger sample would be required to determine if this is the case. Blood pressure and job strain were measured close in time. The interaction between job strain, BP and genetic polymorphisms is an important finding; however, the significance for clinical practice has not yet been elucidated. In summary, our study, while based on a limited sample size, strengthens the evidence supporting the EDN1 gene as an important risk factor in the development of hypertension in conjunction with environmental stressors. While preliminary, our study supports an
S.W. Tobe et al. / Journal of Psychosomatic Research 71 (2011) 97–101
interaction effect between this gene and an important, documented stress factor, job strain, as contributing to BP. The differences observed between the two genotype classes when job strain is included in the analyses support previous data indicating that genetic variation in EDN1 contributes to a differential response to stress, altering either differences in the induction of the gene, stability of the RNA or change of function of the protein. The challenge now is to determine the molecular mechanism by which this gene responds to environmental stressors and how genetic variation influences the response to these stressors. References [1] Dhaun N, Goddard J, Kohan D, Pollock D, Schiffrin E, Webb D. Role of endothelin-1 in clinical hypertension: 20 years on. Hypertension 2008;52:452–9. [2] Kuwaki T, Kurihara H, Cao W, Kurihara Y, Unekawa M, Yazaki Y, et al. Physiological role of brain endothelin in the central autonomic control: from neuron to knockout mouse. Prog Neurobiol 1997;51:545–79. [3] Delarue C, Conlon J, Remy-Jouet I, Fournier A, Vaudry H. Endothelins as local activators of adrenocortical cells. J Mol Endocrinol 2004;32:1–7. [4] Schiffrin E. Endothelin: role in experiemental hypertension. J Cardiovasc Pharmacol 2000;35(4 Suppl 2):S33–5. [5] Barton M, Yanagisawa M. Endothelin: 20 years from discovery to therapy. Can J Physiol Pharmacol 2008;86:485–98. [6] Schiffrin E. Vascular endothelin in hypertension. Vascul Pharmacol 2005;43: 19–29. [7] Treiber F, Kapuku G, Davis H, Pollock J, Pollock D. Plasma endithelin-1 release during acute stress: role of ethnicity and sex. Psychosom Med 2002;64:707–13. [8] Polderman K, Stehouwer C, van Kamp G, Dekker G, Verheugt F, Gooren L. Influence of sex hormones on plasma endothelin levels. Ann Intern Med 1993;118:429–32. [9] Bilsel A, Moini H, Tetik E, Aksungar F, Kaynak B, Ozer A. 17Beta-estradiol modulates endothelin-1 expression and release in human endothelial cells. Cardiovasc Res 2000;46:579–84. [10] Wagner O, Christ F, Wojta J, Vierhapper H, Parzer S, Nowotny P, et al. Polar secretion of endothelin-1 by cultured endothelial cells. J Biol Chem 1992;267: 16066–8. [11] Morganiti A, Marana I, Airoldi F, Alberti C, Nador B, Palatresi S. Plasma endothelin levels: a meaningless number? J Cardiovasc Pharmacol 2000;35(4 Suppl 2):S21–3. [12] Noll G, Wenzel R, Schneider M, Oesch V, Binggeli C, Shaw S, et al. Increased activation of sympathetic nervous system and endothelin by mental stress in normotensive offspring of hypertensive parents. Circulation 1996;93:866–9. [13] Tiret L, Poirier O, Hallet V, McDonagh T, Morrison C, McMurray J, et al. The Lys198Asn poylmorphism in the endothelin-1 gene is associated with blood pressure in overweight people. Hypertension 1999;33:1169–74. [14] Asai T, Ohkubo T, Katsuya T, Higaki J, Fu Y, Fukuda M, et al. Endothelin-1 gene variant associates with blood pressure in obese Japanese subjects: the Ohasama study. Hypertension 2001;38:1321–4.
101
[15] Jin J, Nakura J, Wu Z, Yamamoto M, Abe M, Tabara Y, et al. Association of endothelin-1 gene variant with hypertension. Hypertension 2003;41:163–7. [16] Treiber F, Barbeau P, Harshfield G, Kang H, Pollock D, Pollock J, et al. Endothelin-1 gene Lys198Asn polymorphsim and blood pressure reactivity. Hypertension 2003;42:494–9. [17] Rabineau K, Treiber F, Poole J, Ludwig D. Interactive effects of anger expression and ET-1 Lys198Asn polymorphism on vasoconstriction reactivity to behavioral stress. Ann Behav Med 2005;30:85–9. [18] Schnall P, Schwartz J, Landsbergis P, Warren K, Pickering T. Relation between job strain, alcohol and ambulatory blood pressure. Hypertension 1992;19:488–94. [19] Light K, Turner J, Hinderliter A. Job strain and ambulatory work blood pressure in healthy young men and women. Hypertension 1992;20:214–8. [20] Fauvel J, Quelin P, Ducher M, Rakotomalala H, Laville M. Perceived job stress but not individual cardiovascular reactivity to stress is related to higher blood pressure at work. Hypertension 2001;38:71–5. [21] Clays E, Leynen F, De Bacquer D, Kornitzer M, Kittel F, Karasek R, et al. High job strain and ambulatory blood pressure in middle-aged men and women from the Belgian job stress study. J Occup Environ Med 2007;49:260–7. [22] Tobe S, Kiss A, Szalai J, Perkins N, Tsigoulis M, Baker B. Impact of job and marital strain on ambulatory blood pressure results from the double exposure study. Am J Hypertens 2005;18:1046–51. [23] Brisson C, Laflamme N, Moisan K, Milot A, Masse B, Vezina M. Effect of family responsibilities and job strain on ambulatory blood pressure among white-collar women. Psychosom Med 1999;61:205–13. [24] Tobe S, Kiss A, Sainsbury S, Jesin M, Geerts R, Baker B. The impact of job strain and marital cohesion on ambulatory blood pressure during 1 year: the double exposure study. Am J Hypertens 2007;20:148–53. [25] Karasek R, Pieper C, Schwartz J, Fry L, Schrier D. Job Instrument Questionnaire and user's guide. Revision 1.1 ed. New York: Columbia University; 1985. [26] Karasek R, Baker D, Marxer F, Ahlbom A, Theorell T. Job decision latitude, job demands, and cardiovascular disease: a prospective study of Swedish men. Am J Public Health 1981;7:694–705. [27] Karasek R, Theorell T, Schwartz J, Schnall P, Pieper C, Michale J. Job characteristics in relation to the prevalence of myocardial infarction in the US Health Examination Survey (HES) and the Health and Nutrition Survey (HANES). Am J Public Health 1988;78:910–8. [28] Stevens P, Brown M. Genetic variability of the ET-1 and the ETA receptor genes in essential hypertension. J Cardiovasc Pharmacol 1995;26(Suppl 3):S9–S12. [29] Dong Y, Wang X, Zhu H, Treiber F, Snieder H. Endothelin-1 gene and progression of blood pressure and left ventricular mass: longitudinal findings in youth. Hypertension 2004;44:884–90. [30] Livak K. Allelic discrimination using fluorogenic probes and the 5′ nuclease assay. Genet Anal 1999;14:143–9. [31] Tanaka C, Kamide K, Takiuchi S, Kawano Y, Miyata T. Evaluation of the Lys198Asn and -134delA genetic polymorphisms of the endothelin-1 gene. Hypertens Res 2004;27:367–71. [32] Popowski K, Sperker B, Kroemer H, John U, Laule M, Stangl K, et al. Functional significance of a hereditary adenine insertion variant in the 5′-UTR of the endothelin-1 gene. Pharmacogenetics 2003;13:443–4.
Contents lists available at ScienceDirect
Journal of Psychosomatic Research
The impact of endothelin-1 genetic analysis and job strain on ambulatory blood pressure☆,☆☆ Sheldon W. Tobe a,⁎, Brian Baker b, Katie Hunter a, Alex Kiss c, Nancy Perkins a, Lissette Gomez d, Yu Feng d, Karen Wigg d, Cathy L. Barr d,e a
Division of Nephrology, Sunnybrook Health Sciences Centre, Toronto, Canada Department of Psychiatry, University Health Network, Toronto, Canada Institute for Clinical Evaluative Sciences, Toronto, Canada d Genetics and Development Division, Toronto Western Research Institute, University Health Network, Toronto, Canada e Program in Neurosciences and Mental Health, The Hospital for Sick Children, Toronto, Canada b c
a r t i c l e
i n f o
Article history: Received 16 May 2010 Received in revised form 1 January 2011 Accepted 4 January 2011 Keywords: Ambulatory blood pressure Endothelin-1 Job strain
a b s t r a c t Objective: An interaction between the endothelin-1 gene (EDN1), blood pressure (BP) and social determinants has been previously found. Using a well-characterized cohort of participants, the impact of associations between genetic factors and job strain on BP was evaluated. Methods: A cross-sectional analysis of five polymorphisms covering the EDN1, of which 2 were previously reported to be associated with BP, was performed. Study subjects had previously completed a baseline evaluation including 24-h ambulatory BP monitoring and an assessment of job strain. This report presents the findings for 184 subjects who gave DNA samples for genetic analysis. One-way analysis of variance (ANOVA) was performed between each genetic marker and 24-h systolic blood pressure (SBP) and diastolic blood pressure (DBP), as well as two-way ANOVAs to test the interaction effect with job strain. Results: Trends for relationships were observed between SBP and two polymorphisms: rs10478694 and rs5369. An interaction between job strain and those heterozygous for two polymorphisms showed higher SBP (P=.029 and .008) and a tendency for higher DBP. All findings were more significant when analyses were confined to the 139 Caucasian subjects. Conclusion: This is the first study to report an interaction between the EDN1 gene, job strain and BP, supporting previous evidence of a role of this gene in the interaction between environmental stress and ambulatory BP. Given the limited sample size, the results should be considered preliminary, and further studies are required. © 2011 Elsevier Inc. All rights reserved.
Introduction More than one in five adults develops hypertension, the most significant modifiable risk factor for heart disease and stroke. The onset of hypertension is influenced by both environmental and genetic factors; however, the interplay of these factors and of their underlying mechanisms remains unclear. Understanding these factors may help to identify and manage individuals at risk. Amongst
Abbreviations: ABPM, ambulatory blood pressure monitoring; BMI, body mass index; DBP, diastolic blood pressure; ET-1, endothelin-1; EDN1, endothelin-1 gene; JCQ, job content questionnaire; JS, job strain; SBP, systolic blood pressure. ☆ Division of Nephrology and Genetics and Development Division, University of Toronto. ☆☆ Funds for this study were provided by a grant from the Hospital for Sick Children Psychiatric Endowment Fund and the Heart and Stroke Foundation of Ontario. ⁎ Corresponding author. Division of Nephrology, Sunnybrook Health Sciences Centre A240 2075 Bayview Avenue, Toronto, ON, Canada M4N 3M5. Tel.: +1 416 480 6901; fax: +1 416 480 6940. E-mail address: [email protected] (S.W. Tobe). 0022-3999/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.jpsychores.2011.01.003
vasopressor factors that might be implicated in the development of essential hypertension, endothelin-1 (ET-1) expression has been shown to be linked to resistance artery remodeling and so may confer a vulnerability to the development of hypertension [1]. Endothelin-1 is best known as a potent regulator of vasoconstriction, and there is now evidence that indicates that it has a number of diverse functions [2], including a role in the central nervous system [2,3]. Endothelin-1 has been clearly associated with hypertension in animal and pharmaceutical models [4,5]; however, clinical evidence demonstrating a direct association with hypertension and plasma levels of ET-1 has been less straightforward [4,6]. A number of known confounding factors may come into play in these studies, including gonadal hormones [7–9] and ethnicity [7]; thus, the composition of the samples with regard to gender and ethnic background may influence the results. Furthermore, as 80% of ET-1 secretion is abluminal (towards the vessel wall), plasma levels may not reflect activation. Mental [10,11] and physiological stressors have been shown to influence ET-1 plasma levels [12], and the impact of these factors can be difficult to control. Genetic background may also be a factor, with
98
S.W. Tobe et al. / Journal of Psychosomatic Research 71 (2011) 97–101
one study finding increases in plasma ET-1 after a mental stressor only in offspring of hypertensive parents [12]. Based on the evidence implicating ET-1 in hypertension, the ET-1 gene (EDN1) has been the focus of a number of genetic association studies, with association findings for resting blood pressure (BP), mostly limited to specific populations such as obese subjects [13–15]. The specific genetic changes in EDN1 that have been associated with BP include the rs5370 polymorphism located in exon 5 that changes the amino acid at codon 198 from lysine to asparagine (Lys198Asn) and the A nucleotide insertion/deletion polymorphism located at position 138 (rs10478694). The Lys198Asn polymorphism has been the focus of the majority of studies to date, with several studies reporting association of the T allele coding for asparagine with higher BP [13–16]. Treiber et al. [16] demonstrated that while carrier status of the T allele (Asn) was not related to resting BP, there was a significant interaction between this allele and BP reactivity when exposed to a stressor, suggesting an interaction of environmental factors with genetic variation in this gene contributing to higher BP. Among the study cohort, the greatest change in BP was seen in those individuals with the T allele (Asn) status from lower socioeconomic backgrounds. More recently, an interaction was demonstrated between anger expression, T(Asn) allele carrier status and total peripheral resistance index while exposed to a stressor such that those T allele carriers with poor anger management skills demonstrated the greatest total peripheral resistance index reactivity consistent with an endothelin effect [17]. Job strain, a construct of excessive work demands and limited control in the work setting, is an environmental stressor that we and others have shown to be associated with elevations of ambulatory BP in men and women [18–23]. The recently completed Double Exposure study examined the interaction over time of both job strain and marital cohesion in normotensive and unmedicated hypertensives [22]. Job strain was strongly associated with higher systolic blood pressure (SBP) even controlling for related covariates, while martial cohesion alone was not [24]. The Double Exposure cohort presented an opportunity to determine whether there was an interaction between the role of genetic factors and the environmental stressor job strain on BP. To look for interactions, we chose the two polymorphisms of EDN1 previously associated with BP and three others to cover the major haplotypes.
Methods Subjects The Double Exposure study examined the effects of job strain and marital cohesion on ambulatory BP in unmedicated normotensive or hypertensive (BPb160/100 mmHg, by the BPTru device [VSM Medical, Vancouver, BC, Canada]) male and female subjects aged 40–65 who were employed and cohabiting with a significant other [22]. At the baseline study visit, questionnaires were completed including the Job Content Questionnaire (JCQ) (described below), and ambulatory BP was measured over 24 h (monitor 90207l Spacelabs Medical Inc., Redmond, WA, USA). Blood pressure was recorded every 15 min during the day and every hour during the night (11 PM–7 AM), and readings were averaged; more detailed methodology has been reported [22,24]. Ambulatory subjects were recruited by advertisements posted around an academic health sciences centre in Toronto, enrolled into the Double Exposure study from July 2001 until July 2003 and followed for 1 year. Three years following completion of the study, subjects were contacted, requesting their participation in the genetic study, and sample collection was completed by July 2008. At the follow-up visit, 184 subjects provided a buccal swab for DNA extraction and genetic analysis.
All research was approved by the Research Ethics Boards for the Hospital for Sick Children and Sunnybrook Health Sciences Centre and was performed in accordance with the Declaration of Helsinki. All participants provided written consent following oral and written explanation of the research. Job Content Questionnaire The JCQ [25,26] is a well-validated, reliable, self-administered instrument designed to measure social and psychological characteristics of work including job strain, which refers to those individuals with high occupational demands and low job control [26,27] and which has been extensively used in relation to cardiovascular parameters including ambulatory BP [18–23]. Job strain (JS) was defined as those individuals in the 20th percentile for high job demands and low job latitude from the JCQ [18–20,27]. Genotyping All DNA samples were collected by extraction from cheek cells (buccal samples) using a standard protocol. Five markers were identified a priori to be genotyped. These included two polymorphisms which have previously been demonstrated to be associated with BP: the Lys198Asn polymorphism (rs5370) [13–16] and the 138ins/delA polymorphism (rs10478694) [28–29]. Either of these two DNA changes may be contributing to BP; alternatively, these polymorphisms may be in linkage disequilibrium with the as yet unknown functional change(s) that contributes to increased BP. Because the functional DNA change(s) is not clearly identified, we genotyped three additional, markers (rs5369, rs1476046, rs2071943) chosen to detect the major haplotypes of the gene (http://www. hapmap.org/cgi-perl/gbrowse/gbrowse/hapmap/) to ensure adequate coverage of the genetic information. Polymorphisms were genotyped using the 5′ nuclease assay (TaqMan assay) [30]. Primers were designed to amplify the region on either side of the polymorphism, and specific probes were designed for each allele and labeled with a florescent tag that was measured on the ABI 7900 (Applied Biosystems, Foster City, CA, USA). Statistical analyses Descriptive statistics were calculated for all variables of interest. Continuous measures such as age were summarized using means and standard deviations, whereas categorical measures were summarized using counts and percentages. Baseline 24-h SBP and diastolic blood pressure (DBP) values were used for all analyses. Job strain was again scored as a binary variable for those in the 20th percentile for high job demands and low job latitude from the JCQ. One-way analysis of variance (ANOVA) was performed for each genetic marker separately and BP (both systolic and diastolic). Each polymorphism was analyzed categorically for each of the three genotype categories for the oneway ANOVA. Two-way ANOVA was also performed to test the interaction effect of the genetic marker and job strain on both SBP and DBP. Homozygous individuals for the less common allele were not compared using the two-way ANOVA models because of low sample numbers, which result in insufficient statistical power to fit such a model. The small sample numbers similarly do not allow the model to control for possible confounders such as age, gender etc. All analyses were performed using SAS statistical software, version 9.1 (SAS Institute Inc., Cary, NC, USA). Results Of the original Double Exposure participants at baseline, 184 (74%) participants provided DNA samples. Of those participating in this study, 39 (21.7%) were classified as having job strain. Mean 24-h ambulatory BP for the subjects participating in the
S.W. Tobe et al. / Journal of Psychosomatic Research 71 (2011) 97–101
Female (%) Age Caucasian (%) BMI Smoking (%) Alcohol (N10 drinks/week) (%) ABPM, mmHg Job strain (%)
Subjects in study (n=184)
Subjects not participating (n=68)
100 (54.3) 50.9±6.0 139 (75.5) 26.7±4.5 15 (8.2) 21 (11.8) 121.3±10.1/ 75.6±7.8 39 (21.7)
37 (54.7) 50.6±7.4 52 (81.2) 26.6±4.0 5 (7.8) 7 (10.9) 120.5±11.5/ 74.1±7.1 13 (20.6)
All data are presented as mean±S.D. unless otherwise indicated. No statistically significant differences between those participating and those not participating. genetics study was 121.1±10.4 mmHg/75.2±7.6 mmHg. Of the participants, 100 (54.3%) of the subjects were female, 139 (75.5%) were Caucasian and overall there were low rates of smoking and alcohol consumption (Table 1). Baseline characteristics were not significantly different for those not participating in the genetics study. Genotype distributions were in accordance with Hardy–Weinberg equilibrium. One-way ANOVA for genotype on SBP approached statistical significance for two of the five markers: rs5369 (P=.065) and rs10478694 (P=.052), one of the markers previously associated with elevated BP. For the rs5369 marker, SBP was lowest in the individuals homozygous for the A allele (111.7±7.6 mmHg). However, there were only three individuals with this genotype; thus, it is not possible to make conclusions concerning this genotype. The SBP for the other two genotypes was similar with the slightly higher SBP for those subjects heterozygous (GA) for the rs5369 marker as compared to those homozygous for the G allele (123.8±1.6 mmHg vs. 120.8±0.8 mmHg, respectively). For the rs10478694 polymorphism, SBP was lowest in those homozygous for the minor allele, the insertion allele (118.7±7.3 mmHg); however, the number of individuals in this class was low (n=11). Systolic BP was slightly higher for those subjects heterozygous for the insertion allele as compared to those subjects homozygous for the deletion (123.3±0.96 mmHg vs. 120.5±1.3 mmHg, respectively). The three remaining markers were not significantly associated with BP. For the one-way ANOVA for DBP and genotype, the rs10478694 polymorphism also showed a trend for an association (P=.089), with individuals homozygous for the A nucleotide insertion allele showing lower DBP than the other two genotypes (Table 2). The other markers were not significant. To determine whether there was an interaction with job strain, a two-way ANOVA was performed (Table 3). Because of low numbers of participants homozygous for the A allele for the rs5369 polymorphism (three subjects) and for the insertion allele of the rs10478694 polymorphism (11 subjects), these were not included. This analysis revealed an interaction between job strain and genotype on SBP for both rs5369 (P=.008) and rs10478694 (P=.029; Figs. 1 and 2, respectively) such that, within the low job strain group, the SBP was very similar between heterozygotes and those homozygous for the common allele. However, within the job strain group, there was a significant difference between the two genotype classes, with those heterozygous showing higher BP and the homozygotes for the common allele showing lower BP than their counterparts with the same genotypes in the low job strain group (Table 3). The same pattern emerged for the two-way ANOVA with DBP, revealing a significant interaction between job strain and genotype for rs5369 (P=.016) and a trend for rs10478694 (P=.124; Figs. 3 and 4, respectively, and Table 4). For both markers in the group with low job strain, the heterozygotes and homozygotes had very similar DBP, whereas in those with job strain, the heterozygotes showed higher DBP than their counterparts in the low job strain category and the homozygotes for the common allele that had lower DBP. A further analysis was performed for the 139 Caucasian participants as there have been reports of genetic heterogeneity amongst ethnic groups, and the results were more significant (data not shown). The non-
Table 3 Mean SBP (mmHg) by job strain and genotype.
rs5369 AG GG
No job strain
Job strain
121.9 [118.31–125.49] (n=31) 121.4 [119.61–123.19] (n=108)
131.3 [124.30–138.30] (n=8) 118.9 [115.32–122.48] (n=30) F=7.29, P=.008
rs10478694 A/del 122.4 [119.39–125.41] (n=46) del/del 121.1 [118.90–123.30] (n=84)
127.4 [122.00–132.80] (n=14) 117.7 [113.70–121.70] (n=25) F=4.86, P=.029
Data presented as mean and confidence intervals.
Caucasians were of mixed ethnicities (Black, Asians, East Indian), and there were insufficient participants to perform the analysis in the non-Caucasian groups.
Discussion While a wealth of information from animal models and pharmacological interventions supports the role of ET-1 as a factor in BP, the direct relationship of ET-1 plasma levels to hypertension is still controversial [4]. Genetic studies of the relationship of EDN1 with resting BP have also been mixed, with associations found most consistently in restricted populations such as the obese. The results from this study build on previous evidence for EDN1 as an important genetic factor in the risk for the development of high BP, particularly in combination with job strain, an environmental stressor which has a documented impact on BP. Furthermore, these results indicate that genetic studies that do not take such stressors into consideration may underestimate the contribution of this gene to BP and may partly account for the conflicting findings from these studies. Interesting information is also provided on the EDN1 polymorphisms associated with BP. Previous research has predominantly focused on the Lys198Asn polymorphism as a possible risk factor, with the T allele associated with increased BP in several studies of European, Australian, Japanese and American subjects [13–15]. 132 130
SBP (mmHg)
Table 1 Baseline demographics for subjects who completed the genetics follow-up and those that did not.
99
128 126 124 Homozygous GG
122
Heterozygous AG
120
Interaction P=.008
118 116 No JS
JS
rs5369 Fig. 1. Systolic blood pressure by genotype (rs5369) and job strain.
128 Table 2 Mean SBP and DBP (mmHg) by genotype.
rs10478694 A/A (n=11) A/del (n=61) del/del (n=112)
Systolic
Diastolic
111.66 [102.88–120.44] 123.82 [120.59–127.06] 120.85 [119.18–122.53] F=2.77, P=.065
71.00 [65.92–76.08] 76.26 [73.67–78.85] 75.46 [74.17–76.75] F=0.68, P=.509
118.7 [114.30–123.10] 123.85 [121.11–126.59] 120.21 [118.36–122.06] F=3.01, P=.052
72.18 [66.75–77.61] 77.08 [75.11–79.05] 75.09 [73.65–76.53] F=2.46, P=.089
Data presented as mean and 95% confidence intervals.
SBP (mmHg)
rs5369 AA (n=3) AG (n=39) GG (n=142)
126 124 122
Homozygous for deletion
120
Heterozygous
118
Interaction P=.029
116 No JS
JS
rs10478694 Fig. 2. Systolic blood pressure by genotype (rs10478694) and job strain.
100
S.W. Tobe et al. / Journal of Psychosomatic Research 71 (2011) 97–101
82
DBP (mmHg)
80 78 Homozygous GG
76
Heterozygous AG
74
Interaction P=.016 72 No JS
JS
rs5369 Fig. 3. Diastolic blood pressure by genotype (rs5369) and job strain.
80
DBP (mmHg)
79 78 77 76
Homozygous for deletion
75
Heterozygous
74
Interaction P=.124
73 72 No JS
JS
rs10478694 Fig. 4. Diastolic blood pressure by genotype (rs10478694) and job strain.
Currently, there is no known change in function related to the alternative alleles of the Lys198Asn polymorphism. The Lys198Asn polymorphism is located in preproendothelin but is cleaved from the mature protein, though it could still play a role in the function of ET-1 by influencing messenger RNA (mRNA) or protein stability. However, a study in hypertensive patients did not find a significant difference in plasma levels of ET-1 in individuals with the different alleles [31], and no differences in protein expression could be found in cell supernatants from cells transfected with the two variants [31]. We found no evidence for an association of BP with the Lys198Asn marker possibly due to the low numbers in this study. Results were only significant for the rs10478694 polymorphism and a marker in linkage disequilibrium with rs10478694, rs5369. The ins/del polymorphism is located at a position 138 bp downstream of the transcription initiation site (in some reports described as 134 bp) in the first untranslated exon [28]. The insertion allele has been associated with increased levels of ET-1 in plasma [31]. Specifically, individuals heterozygous for the insertion allele showed higher plasma ET-1 levels than those homozygous for the deletion allele [31]. Endothelin-1 levels of individuals who were homozygous for the insertion allele were not reported in that paper [31]. The insertion allele has been associated with increased expression in luciferase gene expression assays that may be due to increased stability of the mRNA
Table 4 Mean DBP (mmHg) by job strain and genotype. No job strain
Job strain
rs5369 AG GG
74.9 [71.99–77.81] (n=31) 75.9 [74.44–77.36] (n=108)
81.4 [77.23–85.57] (n=8) 73.9 [70.94–76.86] (n=30) F=5.95, P=.016
rs10478694 A/del del/del
76.4 [74.07–78.73] (n=46) 75.6 [74.01–77.19] (n=84)
78.7 [74.96–82.44] (n=14) 73.4 [70.08–76.71] (n=25) F=2.39, P=.124
Data presented as mean and 95% confidence intervals.
[32], consistent with the increased levels of ET-1 in plasma of individuals heterozygous for the insertion allele, but not with findings of lower BP in individuals with this allele [29]. Increased stability of the mRNA may result in higher levels of the protein, thus predicting higher BP; however, RNA levels do not necessarily predict protein levels, and high protein levels may not directly correlate with plasma levels or BP. Indeed, studies of the mouse knockout model observed a paradoxical increase in BP in mice having one copy of the gene and reduced plasma ET-1 [2]. Our data in the one-way analyses show lower SBP and DBP in the 11 subjects who were homozygous for the insertion allele and marginally higher BP in individuals heterozygous for the insertion allele compared to those homozygous for the deletion allele. Thus, our results are not consistent with a simple role of increased expression of the gene predicted by the insertion allele from the in vitro expression studies and higher ambulatory BP. However, when job strain was included in the analyses, we then observed that individuals heterozygous for this allele showed higher BP than their low job strain counterparts and that those homozygous for the deletion allele showed lower BP than both genotype classes in the low job strain category. One possibility is that the insertion allele, or an allele in linkage disequilibrium with the allele, is responsible for the induction of the gene by psychological stressors. Unfortunately, all of the 11 individuals homozygous for the insertion allele were in the low job strain category, and we were not able to analyze this category in the two-way analyses. A similar picture on the relationship of this gene and BP and an interaction effect with job strain was observed with the alleles of the marker rs5369, a synonymous DNA change in exon 3 (Glu106Glu). With job strain, heterozygous individuals showed increased BP, while those homozygous for the common allele showed decreased BP, compared to their counterparts with the same genotype in the low job strain categories. To the best of our knowledge, this variant has not been studied in any functional assay; thus, currently, we have no information to implicate these alleles in the function of the gene. This marker is in linkage disequilibrium (D'=0.77, r2=0.285) with the rs10478694 marker, and the results may reflect the association with the rs10478694. Alternatively, the results from both markers may reflect association with an as yet unidentified functional allele elsewhere in the gene. If these findings can be replicated, they might provide a method of identifying individuals at risk of developing hypertension from job strain and raise the possibility of psychogenomic interventions. A limitation of our study was the sample size that did not allow us to investigate the homozygous classes for the minor alleles in the twoway analyses; this also resulted in the loss of power from these 14 subjects. A much larger study will be required to have a sufficient sample to test the homozygous classes and would also provide more power for the three markers that were not significant. It will be interesting to see whether people with job strain and homozygous for the polymorphisms rs5369 (AA) and rs10478694 (A/A) have higher BP as suggested from this study. Due to sample size, we were not able to analyze the data by other confounders like age, social economic status or gender. Sex hormones modulate ET-1 levels [7–9], and increases in ET-1 plasma levels were previously reported to be greater in males than females in response to video game and cold stressors [7]; thus, the males in the sample may contribute more to the genetic findings than females, but a larger sample would be required to determine if this is the case. Blood pressure and job strain were measured close in time. The interaction between job strain, BP and genetic polymorphisms is an important finding; however, the significance for clinical practice has not yet been elucidated. In summary, our study, while based on a limited sample size, strengthens the evidence supporting the EDN1 gene as an important risk factor in the development of hypertension in conjunction with environmental stressors. While preliminary, our study supports an
S.W. Tobe et al. / Journal of Psychosomatic Research 71 (2011) 97–101
interaction effect between this gene and an important, documented stress factor, job strain, as contributing to BP. The differences observed between the two genotype classes when job strain is included in the analyses support previous data indicating that genetic variation in EDN1 contributes to a differential response to stress, altering either differences in the induction of the gene, stability of the RNA or change of function of the protein. The challenge now is to determine the molecular mechanism by which this gene responds to environmental stressors and how genetic variation influences the response to these stressors. References [1] Dhaun N, Goddard J, Kohan D, Pollock D, Schiffrin E, Webb D. Role of endothelin-1 in clinical hypertension: 20 years on. Hypertension 2008;52:452–9. [2] Kuwaki T, Kurihara H, Cao W, Kurihara Y, Unekawa M, Yazaki Y, et al. Physiological role of brain endothelin in the central autonomic control: from neuron to knockout mouse. Prog Neurobiol 1997;51:545–79. [3] Delarue C, Conlon J, Remy-Jouet I, Fournier A, Vaudry H. Endothelins as local activators of adrenocortical cells. J Mol Endocrinol 2004;32:1–7. [4] Schiffrin E. Endothelin: role in experiemental hypertension. J Cardiovasc Pharmacol 2000;35(4 Suppl 2):S33–5. [5] Barton M, Yanagisawa M. Endothelin: 20 years from discovery to therapy. Can J Physiol Pharmacol 2008;86:485–98. [6] Schiffrin E. Vascular endothelin in hypertension. Vascul Pharmacol 2005;43: 19–29. [7] Treiber F, Kapuku G, Davis H, Pollock J, Pollock D. Plasma endithelin-1 release during acute stress: role of ethnicity and sex. Psychosom Med 2002;64:707–13. [8] Polderman K, Stehouwer C, van Kamp G, Dekker G, Verheugt F, Gooren L. Influence of sex hormones on plasma endothelin levels. Ann Intern Med 1993;118:429–32. [9] Bilsel A, Moini H, Tetik E, Aksungar F, Kaynak B, Ozer A. 17Beta-estradiol modulates endothelin-1 expression and release in human endothelial cells. Cardiovasc Res 2000;46:579–84. [10] Wagner O, Christ F, Wojta J, Vierhapper H, Parzer S, Nowotny P, et al. Polar secretion of endothelin-1 by cultured endothelial cells. J Biol Chem 1992;267: 16066–8. [11] Morganiti A, Marana I, Airoldi F, Alberti C, Nador B, Palatresi S. Plasma endothelin levels: a meaningless number? J Cardiovasc Pharmacol 2000;35(4 Suppl 2):S21–3. [12] Noll G, Wenzel R, Schneider M, Oesch V, Binggeli C, Shaw S, et al. Increased activation of sympathetic nervous system and endothelin by mental stress in normotensive offspring of hypertensive parents. Circulation 1996;93:866–9. [13] Tiret L, Poirier O, Hallet V, McDonagh T, Morrison C, McMurray J, et al. The Lys198Asn poylmorphism in the endothelin-1 gene is associated with blood pressure in overweight people. Hypertension 1999;33:1169–74. [14] Asai T, Ohkubo T, Katsuya T, Higaki J, Fu Y, Fukuda M, et al. Endothelin-1 gene variant associates with blood pressure in obese Japanese subjects: the Ohasama study. Hypertension 2001;38:1321–4.
101
[15] Jin J, Nakura J, Wu Z, Yamamoto M, Abe M, Tabara Y, et al. Association of endothelin-1 gene variant with hypertension. Hypertension 2003;41:163–7. [16] Treiber F, Barbeau P, Harshfield G, Kang H, Pollock D, Pollock J, et al. Endothelin-1 gene Lys198Asn polymorphsim and blood pressure reactivity. Hypertension 2003;42:494–9. [17] Rabineau K, Treiber F, Poole J, Ludwig D. Interactive effects of anger expression and ET-1 Lys198Asn polymorphism on vasoconstriction reactivity to behavioral stress. Ann Behav Med 2005;30:85–9. [18] Schnall P, Schwartz J, Landsbergis P, Warren K, Pickering T. Relation between job strain, alcohol and ambulatory blood pressure. Hypertension 1992;19:488–94. [19] Light K, Turner J, Hinderliter A. Job strain and ambulatory work blood pressure in healthy young men and women. Hypertension 1992;20:214–8. [20] Fauvel J, Quelin P, Ducher M, Rakotomalala H, Laville M. Perceived job stress but not individual cardiovascular reactivity to stress is related to higher blood pressure at work. Hypertension 2001;38:71–5. [21] Clays E, Leynen F, De Bacquer D, Kornitzer M, Kittel F, Karasek R, et al. High job strain and ambulatory blood pressure in middle-aged men and women from the Belgian job stress study. J Occup Environ Med 2007;49:260–7. [22] Tobe S, Kiss A, Szalai J, Perkins N, Tsigoulis M, Baker B. Impact of job and marital strain on ambulatory blood pressure results from the double exposure study. Am J Hypertens 2005;18:1046–51. [23] Brisson C, Laflamme N, Moisan K, Milot A, Masse B, Vezina M. Effect of family responsibilities and job strain on ambulatory blood pressure among white-collar women. Psychosom Med 1999;61:205–13. [24] Tobe S, Kiss A, Sainsbury S, Jesin M, Geerts R, Baker B. The impact of job strain and marital cohesion on ambulatory blood pressure during 1 year: the double exposure study. Am J Hypertens 2007;20:148–53. [25] Karasek R, Pieper C, Schwartz J, Fry L, Schrier D. Job Instrument Questionnaire and user's guide. Revision 1.1 ed. New York: Columbia University; 1985. [26] Karasek R, Baker D, Marxer F, Ahlbom A, Theorell T. Job decision latitude, job demands, and cardiovascular disease: a prospective study of Swedish men. Am J Public Health 1981;7:694–705. [27] Karasek R, Theorell T, Schwartz J, Schnall P, Pieper C, Michale J. Job characteristics in relation to the prevalence of myocardial infarction in the US Health Examination Survey (HES) and the Health and Nutrition Survey (HANES). Am J Public Health 1988;78:910–8. [28] Stevens P, Brown M. Genetic variability of the ET-1 and the ETA receptor genes in essential hypertension. J Cardiovasc Pharmacol 1995;26(Suppl 3):S9–S12. [29] Dong Y, Wang X, Zhu H, Treiber F, Snieder H. Endothelin-1 gene and progression of blood pressure and left ventricular mass: longitudinal findings in youth. Hypertension 2004;44:884–90. [30] Livak K. Allelic discrimination using fluorogenic probes and the 5′ nuclease assay. Genet Anal 1999;14:143–9. [31] Tanaka C, Kamide K, Takiuchi S, Kawano Y, Miyata T. Evaluation of the Lys198Asn and -134delA genetic polymorphisms of the endothelin-1 gene. Hypertens Res 2004;27:367–71. [32] Popowski K, Sperker B, Kroemer H, John U, Laule M, Stangl K, et al. Functional significance of a hereditary adenine insertion variant in the 5′-UTR of the endothelin-1 gene. Pharmacogenetics 2003;13:443–4.