| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
-308 G/A Polymorphism
From the Institute for Behavioral Sciences, Swiss Federal Institute of Technology, Zürich, Switzerland (P.J., R.v.K., J.E.F.); Department of General Internal Medicine, University Hospital, Bern, Switzerland (R.v.K.) and the Institute of Clinical Chemistry, University Hospital Zürich, Switzerland (F.E.M.).
Address correspondence and reprint requests to Joachim E. Fischer, M.D., M.Sc., Swiss Federal Institute of Technology, Institute for Behavioral Sciences, Turnerstrasse 1, CH-8092 Zürich / Switzerland. E-mail: fischer{at}ifv.gess.etzh.ch
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
gene.
METHODS: We assessed exhaustion in 275 industrial employees (mean age ± SD, 41 ± 9 years, 88% men) using the Maastricht Questionnaire. Subjects were stratified as per exhaustion severity: none (N = 80), moderate (N = 128), and severe (N = 67). The TNF- polymorphism was determined by real-time polymerase chain reaction, and plasma CRP levels were measured by a high-sensitivity immunoassay.
RESULTS: There was a significant interaction between exhaustion and the TNF- polymorphism, explaining 4.5% in the variance of plasma CRP values (F(5,271) = 2.47, p = .033); the result held after controlling for classic cardiovascular risk factors. Adjusted mean CRP levels across exhaustion strata in GA (N = 70) and AA (N = 3) carriers combined were 0.91 mg/l (none), 1.78 mg/l (moderate), and 2.61 mg/l (severe) as compared with 1.24 mg/l, 1.61 mg/l, and 1.36 mg/l for the GG wild-type (N = 202).
CONCLUSION: The findings suggest that the A allele of the TNF- -308 G/A polymorphism may mediate inflammation with exhaustion in a dose-response relationship, while with the GG wild-type exhaustion severity seems unrelated to CRP levels. The finding provides a rationale for gene-environmental interactions by which psychosocial factors may promote atherosclerosis and CAD.
Key Words: exhaustion, • C-reactive protein, • tumor necrosis factor, • polymorphism, • cardiovascular disease.
Abbreviations: BP = blood pressure;; CAD = coronary artery disease;; CRP = C-reactive protein;; ELISA = enzyme-linked immunosorbent assay;; HbA1c = glycosylated hemoglobin A1c;; HDL = high-density lipoprotein cholesterol;; LDL = low-density lipoprotein cholesterol;; PTCA = percutaneous transluminal coronary angioplasty;; TNF- = tumor necrosis factor alpha.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
One recent study performed in patients referred for PTCA because of severe, stable angina proposed that a proinflammatory state might mediate increased coronary risk with exhaustion; exhausted subjects had higher plasma levels of tumor necrosis factor (TNF)- and interleukin-1ß than nonexhausted individuals (5). Such reasoning is much in line with the concept of atherosclerosis as an inflammatory disease (6), and the key role of TNF- and C-reactive protein (CRP) in this concept (7). In a large case–control study, plasma TNF- levels showed a positive association with excess risk for recurrent coronary events after myocardial infarction (8), and numerous studies have shown that CRP is a predictor of future coronary events across different populations (9). Markers of low-grade systemic inflammation such as CRP originate in the liver. Compared with cytokines, in particular TNF-, CRP has a much longer half-life, exceeding 24 hours. Therefore, plasma levels of CRP offer an integrated assessment of inflammatory activity over time. The inflammatory cascade—which ultimately leads to elevated levels of CRP—is triggered by activation of circulating monocytes and of macrophages within atherosclerotic lesions. Autocrine and paracrine activation of monocytes plays a key role in this process. One of the most potent autocrine activators of monocytes is TNF-, which upregulates the synthesis of cytokines (10). Individual differences in the regulation of the TNF- secretion from monocyte/macrophages may, therefore, affect the progression of atherosclerotic lesions.
The last decade has brought much interest in the contribution of genes coding for inflammatory plasma proteins related to CAD (11). For instance, the TNF--308 G/A gene polymorphism is a single base pair polymorphism located at the position –308 in the TNF- gene promoter region mapping to chromosome 6. The TNF- -308 G/A polymorphism has been investigated in 4 populations of CAD patients (12–15). However, none of these studies found a significant relationship between this polymorphism and increased risk for incident CAD and CAD severity (12–15). In the light of such negative results, it has been emphasized that the etiology of a multifactorial disease like CAD is best understood considering interactions of acquired (ie, environmental) and genetic factors (ie, gene polymorphisms) (16). Accordingly, polymorphisms are viewed as amplifiers of environmental effects on the final phenotype rather than direct determinants of plasma protein levels or disease risk (17). With respect to the TNF- -308 G/A polymorphism, it has been shown that cultured blood cells from subjects heterozygous for the A allele of the TNF- -308 G/A polymorphism produce more TNF- than blood cells from subjects homozygous for the G allele on "environmental" stimulation by lipopolysaccharide (18).
There is first evidence to assume that psychosocial characteristics may interact with genotypes much like established cardiovascular risk factors in determining cardiovascular disease (19). For instance, extent of cardiovascular reactivity to mental stress was associated with different polymorphisms in the apolipoprotein E gene (20), the Arg16/Gly polymorphism of the ß2-adrenoreceptor gene (21), and the promoter region of the serotonin transporter gene (5HTTLPR) (22). It has been repeatedly shown that excess cardiovascular reactivity to acute mental stress may promote atherosclerosis (23).
In this study, we investigated the effect of exhaustion on inflammatory activity as measured by high-sensitive CRP in plasma and related to the TNF- -308 G/A polymorphism. We speculated that expression of plasma CRP would be different between exhausted and nonexhausted subjects with respect to the 2 alleles of the promoter region of the TNF- gene.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
For the present study, we excluded all subjects who reported a positive history of cardiovascular disease, immunological disorders, and those who took steroids or other medication affecting immune function. We also excluded participants with an incomplete data set to allow full linear regression approach. These criteria left a study population of 275 subjects (mean age ± SD 41 ± 9 years, 88% male). The population consisted exclusively of white individuals, of whom the majority (>90%) originated from Southern Germany. The ethnic identity of the participants’ parents remained unknown.
Medical Data Acquisition
After a standardized introduction, groups of 12 to 15 participants completed a set of questionnaires during 1-hour sessions. These sessions were held in a room separate from the working place. All questionnaire data were obtained within 1 week. After completing the separate medical questionnaire and after a rest period of 15 minutes, BP was measured twice within 5 minutes by sphygmomanometry, and the average of the two readings was computed.
The medical assessment consisted of a 96-item questionnaire regarding the medical history, risk behavior, physical exercise, sleep, and consumption of alcohol. Questions were derived from the Nurses Health Study (25) and from the MONICA study (26).
Assessment of Vital Exhaustion
We assessed exhaustion by use of the Shortened 9-item Maastricht Exhaustion Questionnaire (27), which was derived from the original 21-item instrument. There is an excellent correlation between scores obtained from the abbreviated and original questionnaire (r = 0.94, p < .001, N = 452) (27). For the purpose of this study, the 9 items were translated into German. The German translation was developed in collaboration with the authors of the original scale. Possible answers to each item were "no" (score = 0), "don’t know" (score = 1), and "yes" (score = 2), resulting in a maximum score of 18. For the analysis, we used both the continuous scale as well as a categorical classification on exhaustion scores. The three categories were defined as follows: "not exhausted" (score 0–3, corresponding to not more than 1 of the 9 questions answered with yes); "mild to moderate exhaustion" (score 4–10) and "severe exhaustion" (score 
11, corresponding to 5 or more questions answered with yes), presumably reflecting clinically relevant levels of exhaustion.
Biochemical Measures
Venous blood for CRP was collected into ice-cooled citrate tubes and immediately centrifuged at 4°C; obtained plasma was snap-frozen at –70°C until further processing. High-sensitive enzyme-linked immunosorbent assays (ELISA) were chosen to measure plasma concentrations of CRP (detection limit 0.1 mg/l; Immulite, DPC Biermann GmbH, Germany) and of TNF- (Quantikine HS, R&D Systems Europe, Abington, United Kingdom). Blood samples for high-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, and for glycosylated hemoglobin A1c (HbA1c) were processed by a commercial laboratory (Synlab, Augsburg, Germany) within 4 hours from blood collection using standard procedures.
Molecular Diagnostics
To determine the TNF- -308 G/A gene polymorphism, we extracted genomic DNA from the leukocyte-containing pellets remaining after centrifugation of coagulated blood using the QIAmp DNA Blood Mini Kit (Qiagen, Hilden, Germany). The TNF- -308 G/A gene polymorphism was assessed by fluorescent real-time polymerase chain reaction with melting curve analysis on a LightCycler (Roche Diagnostics, Rotkreuz, Switzerland) using the TNF- G-308A ToolSet for LightCycler (Genes-4U, Neftenbach, Switzerland) containing specific primers and fluorescent mutation detection oligonucleotide probes, in conjunction with the Roche LightCycler HybProbe Master Mix (Roche Diagnostics, Rotkreuz, Switzerland) according to the manufacturer’s protocols. For statistical analyses, we used the following groups: a) the GG variant, and b) the rarer AA and GA genotype combined. The staff performing the molecular diagnostic assays was strictly blinded to CRP data and to exhaustion scores.
Statistical Analyses
Descriptive data are presented as means ± SD unless in the figure that depicts means ± SEM. To approximate a normal distribution, CRP values were logarithmically transformed. General linear models were employed to elucidate the proportion of variance explained of log-transformed CRP values (dependent variable). Independent variables were exhaustion group, gene variant, and an interaction term between gene variant and exhaustion. Following this crude analysis, we entered possible covariates (smoking status, self-reported physical exercise, self-reported alcohol intake, gender, blood pressure (BP), and lipoproteins). Results were considered statistically significant at the p 
.05 level; all tests were two-tailed. Additional hypothesis-generating analyses included: an extension of the multivariable model by also including plasma levels of TNF- as a covariate, analyses of possible downstream effects regarding the association between plasma levels of CRP and HbA1c, and correlation analyses aiming to elucidate the possibly mediating role of plasma TNF- levels. All calculations were performed using SAS statistical software package (version 8.2, SAS Inc., Cary, NC).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
levels. Moreover, raw exhaustion scores showed a direct association with plasma TNF- levels (r = 0.12, p = .044). Although the crude analysis showed no significant differences in CRP values across strata of exhaustion, raw exhaustion scores were significantly correlated with log-transformed CRP (r = 0.15, p = .009). The adjusted analysis, controlling for gender, age, lipid levels, smoking status, physical exercise, alcohol intake, blood pressure (BP), and HbA1c revealed significant differences in CRP values across exhaustion strata (F(2,261) = 3.54, p = .03). TNF- levels significantly correlated with CRP values (r = 0.26, p < .001).
|
2 p = .47). Therefore, the assumption that expected and observed proportions of genotypes were in equilibrium in our study population was not violated. Table 2 compares individuals with the GG wild-type to participants with the GA or the AA genotype.
|
-308G/A polymorphism alone did not show a significant association with log-transformed CRP levels (p = .54), we found a significant interaction between exhaustion and the polymorphism, explaining 4.5% of the variance in log-transformed CRP (F(3,271) = 2.47, p = .033). GA or AA carriers showed a dose–response relationship between exhaustion and CRP. In contrast, there was no association between exhaustion and CRP in subjects carrying the GG wild-type (Figure 1). This association was retained after controlling for covariates. The full model containing exhaustion, interaction between exhaustion and genotype, and the aforementioned covariates explained 18% of the variance of the log-transformed CRP values (F(19,257) = 2.94, p < .001). Adjusted CRP levels across the 3 strata of exhaustion (none, moderate, severe) in GA and AA individuals were: 0.91 mg/l, 1.78 mg/l, and 2.61 mg/l as compared with 1.24 mg/l, 1.61 mg/l, and 1.36 mg/l for the GG wild-type (Figure 1).
|
plasma levels as an additional covariate in the explanatory model for log-transformed CRP values further increased the explained variance of the adjusted model to 21.5%. In this model, TNF- was a strong and independent predictor (F(1,258) = 15.8, p < .001), whereas the interaction term between exhaustion and genotype retained significance (F(1,258) = 4.3, p = .039).
Association With TNF- and Glycosylated Hemoglobin
To further test the mediational hypothesis, we investigated whether a) the correlation between exhaustion and CRP is higher in the presence of the A allele (r = 0.36, p = .004) as compared with the GG wild-type (r = 0.03, p = .64); b) exhaustion correlates with TNF- in the presence of the A allele (r = 0.22, p = .06) vs. the GG wild-type (r = 0.02, p = .80); c) a correlation exists between TNF- levels and CRP levels (r = 0.11, p = .06); and d) whether introducing TNF- into the multivariable regression models decreases the amount of variance in CRP values explained by exhaustion (change from F = 6.96 to F = 5.86).
Post hoc analyses also revealed that plasma levels of HbA1c were positively associated with log transformed plasma levels of CRP (r = 0.18, p = .0025). This association was more pronounced in subjects carrying at least one A allele (r = 0.24, p = .03) than in participants with the GG wild-type (r = 0.15, p = .03). Estimates from general linear models suggested that an individual at the 75th percentile of CRP values carrying at least 1 A allele had 3.9% higher HbA1c levels than individuals at the 25th percentile of CRP values. The corresponding increase in participants with the GG wild-type amounted to 1.7%.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
levels in exhaustion (5), a finding we replicated in the present study. The primary aim of our study was to test for an association between exhaustion on plasma levels of the acute phase reactant CRP, which is the inflammatory marker most consistently associated with CAD (9). Of note, even slight elevations of CRP predicted cardiovascular events in men and women (31,32). As did these prospective studies, we choose a high-sensitive assay for CRP (33), which allowed us to stratify individuals with CRP levels within the currently accepted normal range (9). We found a weak association between exhaustion scores and CRP and a significant interaction between exhaustion and genotype. Across three strata of exhaustion severity, we found that the group of severely exhausted subjects (score 11–18) carrying the A allele had higher plasma CRP levels than severely exhausted subjects being homozygous for the G allele. In other words, GG homozygous subjects appeared to be protected from inflammatory stimulation that occurs with severe exhaustion in carriers of the A allele. However, consistent with its distribution in the general population, the frequency of the AA genotype in our sample was very low. For the purpose of reasonable statistical analyses, we were forced to group subjects heterozygous and homozygous for the A allele. We thus are unable to state whether there is a dose–response relationship between the A allele dosage and exhaustion in terms of CRP. Studies in much larger populations are clearly needed to resolve this intriguing question.
Do our results suggest that exhaustion might regulate CRP levels via an interaction with the -308 G/A polymorphism in the promoter region of the TNF- gene? Our main finding lends support to the view that the A allele of the TNF- -308 G/A polymorphism mapping to chromosome 6 may mediate effects of exhaustion on plasma CRP levels. TNF- is a potent paracrine activator of monocytes. The association between plasma levels of TNF- and CRP supports the view that monocyte activation may lead to elevated plasma levels of CRP via intermediate steps in the inflammatory cascade (10). We also found an increased association between HbA1c levels and CRP levels in individuals carrying at least 1 A allele. This suggests the possibility that the genotype may affect the interaction between exhaustion and CRP further downstream of the inflammatory cascade, eg, via the glucose metabolism.
Our findings add to the emerging evidence that psychosocial risk factors for CAD may exert some of their adverse impact on vascular health via particular gene polymorphisms (19–22). It is well accepted that individual genetic polymorphisms are best viewed as modulators of cardiovascular risk in the presence of particular environmental factors rather than being unique risk factors on their own (16,17). This is exemplified in terms of research on the role of the TNF- -308 G/A polymorphism in CAD (11). Previous studies have failed to show such an association (12–15). On the other hand, plasma levels of the proinflammatory cytokine TNF- have been positively associated with recurrent myocardial infarction in CAD patients (8). We know of no study that reports a significant association between the TNF- -308 G/A polymorphism and resting plasma TNF- levels. However, when subjected to an "environment" enriched with lipopolysaccharide, blood cells of healthy subjects expressed different amounts of TNF- depending on genotype. In essence, carriers of the A allele had higher TNF- than subjects homozygous for the G allele (18), emphasizing gene–environmental interactions in the expression of a particular phenotype (ie, plasma TNF- levels).
Our study is consistent with such reasoning because the TNF- polymorphisms affected CRP only in the presence of an environment that had resulted in exhaustion. Of course, because our study was associative in nature, we are unable to state whether exhaustion caused high CRP levels via an influence on the promoter site of the TNF- gene (eg, via related neuroendocrine changes), or whether CRP, in presence of the polymorphism, may have caused exhaustion (ie, by causing "sickness behavior"). We favor, however, the former assumption. Moreover, our findings may have important clinical implications because they may help explain why some individuals experience increased cardiovascular risk from prolonged states of psychosocial stress resulting in exhaustion while others remain unaffected.
It should be emphasized, though, that substantial heritable variation is present for most psychosocial risk factors. Thus, the observed interaction could also reflect the interaction between a single genetic locus (TNF- -308 G/A) and heritable psychological predispositions. In this case, the findings are best explained as a gene–gene interaction (epistasis). Following this reasoning, the interaction between the environment and the individual would occur at the level of the psychological heredity. For instance, a longitudinal study showed that individuals with one or two copies of the short allele of the promoter region of the serotonin transporter (5-HTT) gene experienced more depressive symptoms and diagnosable depression in response to a determined number of stressful life events than individuals homozygous for the long allele (34). Given the overlap between exhaustion and depression, a future avenue of investigation might be to examine the gene–gene interaction between the 5-HTT promoter polymorphism and the 308 G/A TNF- polymorphism in relation to life event stress and inflammatory outcome.
Three potential caveats of our study need to be considered. First, we performed a cross-sectional investigation. Therefore, we had to focus on the association between a proxy-indicator for cardiovascular risk (ie, CRP) as the outcome variable. Longitudinal studies following a sufficiently large sample are required to establish the causal role of the observed gene–environmental interaction in modulating cardiovascular risk. Second, we did not include a measure of depressed mood in addition to exhaustion. Because there is some overlap between the two constructs (35), we cannot exclude that the observed gene–environmental interaction applies to depression rather than to exhaustion. Sadness, guilt, and feelings of worthlessness are commonly observed in depressed patients, whereas these affects are not part of the diagnostic criteria for vital exhaustion (36). To be able to differentiate between these 2 entities, future studies need to assess depressed mood or clinical depression related to CRP as mediated by the -308 G/A TNF- polymorphism. Third, we did not measure plasma levels of interleukin-6. Plasma levels of interleukin-6 might be one mediating factor leading from elevated local TNF- levels within an existing atherosclerotic lesion to increased CRP secretion from liver cells.
Taken together, the present study shows that the A allele of the TNF- -308 G/A polymorphism may contribute to elevated plasma levels of CRP in severely exhausted subjects. If confirmed, this finding points to an inflammatory pathway by which exhaustion may promote atherosclerosis. The findings call for further studies on interactions between the genotype and psychosocial factors in terms of cardiovascular biology. Such research might tremendously advance our understanding of how psychosocial risk factors may contribute to CAD.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Received for publication April 27, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. B. Dowd, M. N. Haan, L. Blythe, K. Moore, and A. E. Aiello Socioeconomic Gradients in Immune Response to Latent Infection Am. J. Epidemiol., January 1, 2008; 167(1): 112 - 120. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. McCaffery, H. Snieder, Y. Dong, and E. de Geus Genetics in Psychosomatic Medicine: Research Designs and Statistical Approaches Psychosom Med, February 1, 2007; 69(2): 206 - 216. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C G Parks and G S Cooper Occupational exposures and risk of systemic lupus erythematosus: a review of the evidence and exposure assessment methods in population-and clinic-based studies Lupus, November 1, 2006; 15(11): 728 - 736. [Abstract] [PDF]
|
|
|
|
 |
|
 R. von Kanel, J. E. Dimsdale, P. J. Mills, S. Ancoli-Israel, T. L. Patterson, B. T. Mausbach, and I. Grant Effect of Alzheimer Caregiving Stress and Age on Frailty Markers Interleukin-6, C-Reactive Protein, and D-Dimer J Gerontol A Biol Sci Med Sci, September 1, 2006; 61(9): 963 - 969. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. W. McDade, L. C. Hawkley, and J. T. Cacioppo Psychosocial and Behavioral Predictors of Inflammation in Middle-Aged and Older Adults: The Chicago Health, Aging, and Social Relations Study Psychosom Med, May 1, 2006; 68(3): 376 - 381. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. McCaffery, N. Frasure-Smith, M.-P. Dube, P. Theroux, G. A. Rouleau, Q. Duan, and F. Lesperance Common genetic vulnerability to depressive symptoms and coronary artery disease: a review and development of candidate genes related to inflammation and serotonin. Psychosom Med, March 1, 2006; 68(2): 187 - 200. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Lucini, G. Di Fede, G. Parati, and M. Pagani Impact of Chronic Psychosocial Stress on Autonomic Cardiovascular Regulation in Otherwise Healthy Subjects Hypertension, November 1, 2005; 46(5): 1201 - 1206. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 How's Your Health? Self-Rated Health, Circulating Cytokines, and Your Genes Journal Watch Psychiatry, September 8, 2004; 2004(908): 2 - 2. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
