TGFB1 genetic polymorphisms and coronary heart disease risk: a meta-analysis

  • Yingchang Lu1, 2Email author,

    Affiliated with

    • Jolanda MA Boer2,

      Affiliated with

      • Roza M Barsova3,

        Affiliated with

        • Olga Favorova3,

          Affiliated with

          • Anuj Goel4,

            Affiliated with

            • Michael Müller1,

              Affiliated with

              • Edith JM Feskens1 and

                Affiliated with

                • and on behalf of PROCARDIS CARDIoGRAM Consortium

                  Affiliated with

                  BMC Medical Genetics201213:39

                  DOI: 10.1186/1471-2350-13-39

                  Received: 21 August 2011

                  Accepted: 27 April 2012

                  Published: 18 May 2012

                  Abstract

                  Background

                  Genetic variations in TGFB1 gene have been studied in relation to coronary heart disease (CHD) risk, but the results were inconsistent.

                  Methods

                  We performed a systematic review of published studies on the potential role of TGFB1 genetic variation in CHD risk. Articles that reported the association of TGFB1 genetic variants with CHD as primary outcome were searched via Medline and HuGE Navigator through July 2011. The reference lists from included articles were also reviewed.

                  Results

                  Data were available from 4 studies involving 1777 cases and 7172 controls for rs1800468, 7 studies involving 5935 cases and 10677 controls for rs1800469, 7 studies involving 6634 cases and 9620 controls for rs1982073, 5 studies involving 5452 cases and 9999 controls for rs1800471, and 4 studies involving 5143 cases and 4229 controls for rs1800472. The pooled odds ratios (ORs) for CHD among minor T allele carriers of rs1800469, minor C allele carriers of rs1982073, and minor C allele carriers of rs1800471 versus homozygous major allele carriers was 1.14 (95% confidence interval [CI]: 1.05-1.24), 1.18 (95% CI: 1.04-1.35), and 1.16 (95% CI: 1.02-1.32), respectively. No substantial heterogeneity for ORs was detected among the included Caucasian populations for all SNPs. However, for rs1800471, the statistical significance disappeared after adjusting for potential publication bias. No significant association was found between rs1800468 and rs1800472 variants and CHD risk.

                  Conclusion

                  Minor allele carriers of two genetic variants (rs1800469 and rs1982073) in TGFB1 have a 15% increased risk of CHD.

                  Background

                  Transforming growth factor-β1 (TGFβ1) is a ubiquitously expressed multifunctional cytokine that is involved in many physiological and pathological processes. TGFβ1 has been demonstrated to be of fundamental importance in the development, physiology and pathology of the vascular system. Research into the mechanisms of TGFβ1 signaling over the past two decades has led to the development of a well-accepted canonical signaling cascade involving heterotetrameric complexes of type I and type II serine/threonine-kinase transmembrane receptors together with Smad transcription factors that act as intracellular signaling effectors. However, the exact mechanisms by which TGFβ1 signaling exerts its effects within the vasculature are still incompletely understood [14]. According to the literature [27], TGFβ1 can be secreted by several cell types, including peripheral blood mononuclear cells, macrophages, platelets, endothelial cells, vascular smooth muscle cells (VSMCs), myofibroblasts, and renal cells. Its regulatory function on the vessel wall is directed at endothelial cells, VSMC and extracellular matrix [13, 58]. Although the role of TGFβ1 in the pathogenesis of atherosclerosis is being recognized, the association between plasma TGFβ1 levels and coronary heart disease (CHD) risk is still controversial [6, 911]. There may be several explanations for the controversy: 1), TGFβ1 is a bimodal regulator of both endothelial cells and VSMC proliferation, depending on local TGFβ1 levels, cell density, and/or membrane TGFβ receptors [1, 2, 5, 7, 8, 12]; 2), different pathophysiological stages of CHD may differentially affect the biological effects of TGFβ1 [1, 3, 10]; and 3), circulating TGFβ1 levels may not reflect the real vascular interstitial TGFβ1 levels that are directly involved in the pathogenesis of CHD [35, 13]. Also, animal-model studies of CHD reported inconsistent findings on the role of TGFβ1 in CHD development. This might, however, be due to the dysregulated systemic immune function from different methods used, i.e. injecting TGFβ1 antibodies, infusing a soluble TGFβ receptor, or using transgenic or knockout mice [6, 8].

                  Although the amino acid sequence of the active form of TGFβ1 is highly conserved across mammalian species [7, 14, 15], common TGFB1 genetic variations that could cause variable constitutive or induced expression of TGFB1 or protein structural changes and, as a result, changed TGFβ1 activity, have been identified. They include rs1800468 (−800 G/A) and rs1800469 (−509 C/T) in the promoter region, rs1982073 (868 T/C, Leu10Pro) and rs1800471 (913 G/C, Arg25Pro) in the signal peptide region, and rs1800472 (11929 C/T, Thr263Ile) in the region encoding the precursor part of the protein [4, 1519]. These genetic variants are generally in strong linkage disequilibrium (LD) with each other, and this DNA LD block covers the whole 5′ proximal region of the TGFB1 gene in Caucasian populations [4, 14, 16, 17]. The minor alleles of these genetic variants or the haplotypes where the minor alleles are located, were associated with increased CHD risk in some [11, 16, 20], but not all studies [10, 14, 21, 22], and even an opposite association has been observed [23]. This may partly be explained by a relatively small sample size, different CHD endpoints and/or different study populations in each of the published studies. Demonstrating an association may require a much larger number of subjects, which may be beyond the resource of one single study. Multiple replicated loci have recently been identified from genome-wide association (GWA) studies of CHD. However, they together explain only a small part of its heritability [24, 25]. It has been suggested that the adopted highly stringent statistical criteria and/or the imperfect coverage of genetic variants by current GWA studies might prevent the discovery of potential loci associated with CHD risk [26]. No meta-analysis describing TGFB1 genetic variants in relation to CHD risk exists; therefore, we performed a meta-analysis of published data that tested for TGFB1 genetic variants associated with CHD risk.

                  Methods

                  Selection criteria

                  For inclusion, studies 1) had to be case–control or cohort in design, 2) examined the association between TGFB1 gene polymorphisms and primary outcomes of CHD, coronary artery disease or myocardial infarction (MI), 3) used validated coronary heart disease phenotypes (diagnostic criteria included angiographical confirmation; elevations of cardiac enzymes, changes of electrocardiographic and clinical symptoms according to the World Health Organizations criteria; a documented history of coronary artery bypass graft, percutaneous transluminal coronary angioplasty, or percutaneous coronary intervention), and 4) involved unrelated participants.

                  Search strategy

                  All studies reporting on the association between TGFB1 gene polymorphism and CHD risk published before July 2011 were identified by comprehensive electronic searches of Medline and HuGENet. Terms used for the searches were “TGFB1,” “ischemic heart disease,” “coronary heart disease,” “coronary artery disease,” “acute coronary syndrome,” “myocardial infarction,” and “angina pectoris” combined with “gene,” “genetic,” “variant,” “mutation” or “polymorphism.” Hand searches for related articles among the reference lists of included articles were also performed. If essential information of a study was not presented in the publication, authors were contacted for details. The study was excluded if the information could not be obtained. In addition, the relevant data from the two latest large-scale meta-analyses of CHD GWA studies (PROCARDIS [25] and CARDIoGRAM [24]; Additional file ( 1) were also included for part of the analysis.

                  Data extraction

                  The first author, published year, country, study population, mean age of participants, gender distribution, study design, sample size, outcome, diagnostic criteria, genotyping method, characteristics of the controls, allele frequencies, and genotype distributions were extracted. In PROCARDIS [25] and CARDIoGRAM [24], the studied SNPs were not available in the genome-wide genotyping assay and had to be imputed. The SNPs imputed with high quality (MACH_R2 > 0.3) were included in the analyses.

                  Statistical analysis

                  Deviance from Hardy-Weinberg equilibrium (HWE) was assessed for the controls of each study using Fisher’s exact test. For an Iranian study [11], only data on rs1982073 was utilized in the meta-analysis because other SNPs deviated from HWE. Genotype distributions of controls for studies with case–control design or the entire group for studies with cohort design were used to estimate the frequency of the putative risk allele for each SNP using the inverse variance method (Additional file 2) [27]. Crude ORs with 95% confidence intervals (CI) were used to evaluate the association between genetic polymorphisms and CHD risk. Pooled ORs were calculated for several genetic models, i.e. the co-dominant model, the dominant model, and the recessive model. Since the co-dominant model effects (or additive model effects) cannot be straightforward calculated from the extracted summary data from each study, we presented the effects of two groups of genotype comparison (Additional file 2). Since the using of best-guess genotype from the genotype imputation process can lead to both false positives and loss of power [28], the allele effects from the PROCARDIS [25] and CARDIoGRAM [24] studies were only included in part of the co-dominant model analyses (Additional file 2). Statistical heterogeneity in the ORs across studies was assessed with the Q-test. If there was heterogeneity, ethnicity as source of heterogeneity was explored by pooling the data from Caucasian populations only. If there was no heterogeneity, the fixed-effect model was used to evaluate the overall gene effect; otherwise, the random-effect model was used. Presence of publication bias was explored with Begg’s funnel plot and Egger’s regression test. If potential publication bias existed, the Duval and Tweedie nonparametric “trim and fill” method [29] was used to adjust for it. All reported p values were two-tailed, and statistical significance was defined at the α = 0.05 level. All analyses were performed with the R metafor package [30].

                  Results

                  Study inclusion and characteristics

                  Fifteen citations were identified through the original literature search; none were meta-analyses [10, 11, 14, 16, 2023, 3137]. After full review, two studies were excluded because they were conducted in patient cohorts with a composite end-point that included CHD [31, 32]. Three studies (two European [10, 33] and one Japanese [34]) were conducted without “proper” controls of no CHD history and were also excluded. One additional Chinese study met the inclusion criteria but was excluded due to unavailability of essential information even after contacting the authors [35]. The 9 remaining studies, together with PROCARDIS [25] and CARDIoGRAM [24] studies, were included in the meta-analysis (Table 1) [11, 14, 16, 2023, 36, 37].
                  Table 1

                  Characteristics of studies included in the meta-analysis a

                  Study

                  Country

                  Design

                  Cases

                  Controls

                  Outcome

                  Genotypes

                   

                  Genotypes

                  p_HWEc

                  rs1800468 (−800 G/A)

                    

                  N

                  GG

                  AG

                  AA

                  N

                  GG

                  AG

                  AA

                    

                  Crobu et al, 2008 [22]

                  Italy

                  CC

                  201

                  175

                  25

                  1

                  201

                  168

                  31

                  2

                  0.65

                  MI

                  Sie et al, 2006 [21]

                  Netherlands

                  CO

                  358

                  288

                  66

                  4

                  6098

                  5071

                  984

                  43

                  0.58

                  MI

                  Syrris et al,1998 [14]

                  England

                  CC

                  655

                  541

                  110

                  4

                  244

                  207

                  36

                  1

                  1

                  CAD

                  Cambien et al,1996 [16]

                  FR and NIE

                  CC

                  563

                  472

                  88

                  3

                  629

                  534

                  89

                  6

                  0.28

                  MI

                  rs1800469 (−509 C/T)

                     

                  CC

                  CT

                  TT

                   

                  CC

                  CT

                  TT

                    

                  Sudomoina et al, 2010 [36]b

                  Russia

                  CC

                  264

                  77

                  150

                  37

                  212

                  90

                  103

                  19

                  0.22

                  MI

                  Drenos et al, 2009 [37]

                  England

                  CC

                  240

                  120

                  100

                  20

                  2143

                  1090

                  885

                  168

                  0.56

                  CAD

                  Crobu et al, 2008 [22]

                  Italy

                  CC

                  201

                  67

                  87

                  47

                  201

                  80

                  92

                  29

                  0.76

                  MI

                  Koch et al, 2006 [20]

                  Germany

                  CC

                  3657

                  1581

                  1659

                  417

                  1211

                  564

                  508

                  139

                  0.13

                  MI

                  Sie et al, 2006 [21]

                  Netherlands

                  CO

                  355

                  171

                  156

                  28

                  6037

                  3043

                  2441

                  553

                  0.05

                  MI

                  Syrris et al, 1998 [14]

                  England

                  CC

                  655

                  301

                  284

                  70

                  244

                  124

                  97

                  23

                  0.54

                  CAD

                  Cambien et al, 1996 [16]

                  FR and NIE

                  CC

                  563

                  240

                  257

                  66

                  629

                  263

                  297

                  69

                  0.29

                  MI

                  rs1982073 (868 T/C)

                     

                  TT

                  TC

                  CC

                   

                  TT

                  TC

                  CC

                    

                  Najar et al, 2011 [11]

                  Iran

                  CC

                  900

                  301

                  424

                  175

                  900

                  395

                  403

                  102

                  1

                  MI

                  Crobu et al, 2008 [22]

                  Italy

                  CC

                  201

                  55

                  88

                  58

                  201

                  69

                  101

                  31

                  0.66

                  MI

                  Koch et al, 2006 [20]

                  Germany

                  CC

                  3657

                  1235

                  1802

                  620

                  1211

                  458

                  565

                  188

                  0.55

                  MI

                  Sie et al, 2006 [21]

                  Netherlands

                  CO

                  343

                  135

                  164

                  44

                  5844

                  2322

                  2698

                  824

                  0.37

                  MI

                  Yokota et al, 2000 [23]

                  Japan

                  CC

                  315

                  89

                  185

                  41

                  591

                  149

                  295

                  147

                  1

                  MI

                  Syrris et al, 1998 [14]

                  England

                  CC

                  655

                  242

                  306

                  107

                  244

                  102

                  109

                  33

                  0.68

                  CAD

                  Cambien et al, 1996 [16]

                  FR and NIE

                  CC

                  563

                  181

                  277

                  105

                  629

                  225

                  297

                  107

                  0.62

                  MI

                  rs1800471 (913 G/C)

                     

                  GG

                  GC

                  CC

                   

                  GG

                  GC

                  CC

                    

                  Drenos et al, 2009 [37]

                  England

                  CC

                  234

                  187

                  45

                  2

                  2071

                  1723

                  331

                  17

                  0.78

                  CAD

                  Koch et al, 2006 [20]

                  Germany

                  CC

                  3657

                  3149

                  486

                  22

                  1211

                  1063

                  141

                  7

                  0.33

                  MI

                  Sie et al, 2006 [21]

                  Netherlands

                  CO

                  343

                  297

                  45

                  1

                  5844

                  4992

                  823

                  29

                  0.51

                  MI

                  Syrris et al, 1998 [14]

                  England

                  CC

                  655

                  558

                  95

                  2

                  244

                  214

                  30

                  0

                  0.61

                  CAD

                  Cambien et al, 1996 [16]

                  FR and NIE

                  CC

                  563

                  464

                  92

                  7

                  629

                  546

                  81

                  2

                  1

                  MI

                  rs1800472 (11929 C/T)

                     

                  CC

                  CT

                  TT

                   

                  CC

                  CT

                  TT

                    

                  Drenos et al, 2009 [37]

                  England

                  CC

                  241

                  234

                  7

                  0

                  2145

                  2052

                  89

                  4

                  0.02

                  CAD

                  Koch et al, 2006 [20]

                  Germany

                  CC

                  3657

                  3421

                  231

                  5

                  1211

                  1138

                  72

                  1

                  1

                  MI

                  Syrris et al, 1998 [14]

                  England

                  CC

                  655

                  622

                  33

                  0

                  244

                  237

                  7

                  0

                  1

                  CAD

                  Cambien et al, 1996 [16]

                  FR and NIE

                  CC

                  590

                  563

                  27

                  0

                  629

                  585

                  42

                  2

                  0.20

                  MI

                  FR and NIE, France and Northern Ireland; CC, case–control; CO, cohort; MI, myocardial infarction; CAD, coronary artery disease.

                  a, For rs1800468, rs1982073 and rs1800471, 10090 additional subjects were included for the co-dominant model analysis from the PROCARDIS study [25]; for rs1800469, additional 10090 and 80016 subjects were included for the co-dominant model analysis, respectively from the PROCARDIS study [25] and the CARDIoGRAM study [24] (Additional file 2).

                  b, Additional unpublished data have been included.

                  c, The p values for Hardy-Weinberg equilibrium were derived from Fisher’s exact test.

                  Quantitative synthesis

                  None of the genetic variants were associated with CHD risk when applying a recessive model (data not shown). Rs1800468 and rs1800472 were not associated with CHD in either co-dominant or dominant model (Additional file 2).

                  For rs1800469, both the CT genotype in the co-dominant model and the presence of the minor T allele in the dominant model conferred a risk for CHD when compared to the common CC genotype (OR = 1.14, 95% CI: 1.04-1.25; and OR = 1.14, 95% CI: 1.05-1.24, respectively). The TT genotype conferred a non-significant risk of similar magnitude (Figure 1). For rs1982073, the TC genotype conferred a risk for CHD in the co-dominant model (OR = 1.18, 95% CI: 1.08-1.28), but the CC genotype did not when compared to the common TT genotype (Figure 2). Under a dominant model, the presence of the minor C allele was associated with a 1.18 times increased risk for CHD (Additional file 2).
                  http://static-content.springer.com/image/art%3A10.1186%2F1471-2350-13-39/MediaObjects/12881_2011_987_Fig1_HTML.jpg
                  Figure 1

                  Meta-analysis for coronary heart disease risk depending on the rs1800469 (−509 C/T) polymorphism in the TGFB1 gene. ORs and corresponding 95% confidence intervals (CIs) are shown. Fixed effects were reported because no significant heterogeneity between studies was observed. A. Comparison of the homozygous TT genotype with the wild type CC genotype (p = 0.08); B. Comparison of the heterozygous CT genotype with the wild type CC genotype (p = 0.004); C. Comparison of the TT + CT genotype with the wild type CC genotype (p = 0.003).

                  http://static-content.springer.com/image/art%3A10.1186%2F1471-2350-13-39/MediaObjects/12881_2011_987_Fig2_HTML.jpg
                  Figure 2

                  Meta-analysis for coronary heart disease risk depending on the rs1982073 (868 T/C) polymorphism in the TGFB1 gene. ORs and corresponding 95% confidence intervals (CIs) are shown. A. Comparison of the homozygous CC genotype with the wild type TT genotype (random-effect model, p = 0.26); B. Comparison of the heterozygous TC genotype with the wild type TT genotype (random-effect model, p = 0.0002); C. Comparison of the CC + TC genotype with the wild type TT genotype (random-effect model, p = 0.01).

                  For rs1800471, both the GC genotype in the co-dominant model and the presence of the minor C allele in the dominant model conferred a risk for CHD when compared to the common GG genotype (OR = 1.15, 95% CI: 1.01-1.31; and OR = 1.16, 95% CI: 1.02-1.32, respectively). The CC genotype conferred a 1.25 times increased risk, but this was not statistically significant (Figure 3).
                  http://static-content.springer.com/image/art%3A10.1186%2F1471-2350-13-39/MediaObjects/12881_2011_987_Fig3_HTML.jpg
                  Figure 3

                  Meta-analysis for coronary heart disease risk depending on the rs1800471 (913 G/C) polymorphism in the TGFB1 gene. ORs and corresponding 95% confidence intervals (CIs) are shown. Fixed effects were reported because no significant heterogeneity between studies was observed. A. Comparison of the homozygous CC genotype with the wild type GG genotype (p = 0.49); B. Comparison of the heterozygous GC genotype with the wild type GG genotype (p = 0.03); C. Comparison of the GC + CC genotype with the wild type GG genotype (p = 0.02).

                  After adjusting for multiple testing using Bonferroni correction, all significant associations for rs1800469 and rs1982073 under the co-dominant and dominant models remained. However, for rs1800471, associations were no longer statistically significant (p > 0.017 in Additional file 2). When the relevant allele effects from the PROCARDIS [25] and CARDIoGRAM [24] studies were included in the co-dominant model analyses, the aforementioned associations attenuated; however, the association for rs1982073 persisted (Additional file 2).

                  No substantial heterogeneity for the ORs was detected among the included Caucasian populations (Additional file 2). For rs1982073, however, some heterogeneity existed between Caucasian populations and non-Caucasian populations with regard to both CC and CC + TC vs. TT contrasts (Additional file 2).

                  Sensitivity analysis and publication bias

                  To evaluate the influence of the individual studies on the pooled ORs for rs1800469, rs1982073 and rs1800471, each time, a single study involved in the meta-analysis was deleted. The Rotterdam study [21] and the Japanese study [23] tended to attenuate the pooled ORs in the co-dominant model for rs1800469 and rs1982073, respectively (Additional file 3). No substantial alteration in the observed pooled ORs was observed for rs1800471 (Additional file 3). Begg’s funnel plot and Egger’s regression test were performed to assess potential publication bias for rs1800469, rs1982073 and rs1800471. Although the P values for Egger’s regression tests for all investigated models were > 0.05, Begg’s funnel plot still suggested a certain degree of publication bias, potentially from small studies with significant positive results (Additional file 4). After performing the “trim and fill method” to adjust for potential publication bias, the results for rs1800469 and rs1982073 did not change significantly (data not shown). However, for rs1800471, the statistical significance for both co-dominant and dominant models disappeared (P = 0.07).

                  Discussion

                  Several studies have been carried out to test the hypothesis that genetic polymorphisms in the TGFB1 gene including rs1800468, rs1800469, rs1982073, rs1800471 and rs1800472 might be associated with CHD risk, but data have yielded conflicting results. Possible concerns in genetic association studies are that a positive association might be spurious, while a negative result might be due to a small sample size. In this meta-analysis, we incorporated all eligible studies to date and provided some evidence that rs1800469 and rs1982073 in the TGFB1 gene are associated with CHD risk in Caucasian populations. The inconsistency between the previously reported results for these SNPs might be due to the small sample sizes in most of the studies, especially in combination with a modest effect.

                  Circulating TGFβ1 levels are predominantly under genetic control with a heritability of 0.54 [17]. Both the CHD-associated minor risk alleles of rs1800469 and rs1982073 correlate with an increase in gene expression, TGFβ1 secretion, and plasma TGFβ1 levels [4, 11, 15, 17, 19, 23]. These similar observations might be due to the strong LD between them [4, 14, 16, 17]. Shah et al. [18] demonstrated exclusively in vivo and in vitro recruitment of transcription regulator AP1 to -509 C (the major non-risk allele of rs1800469) leading to transcriptional repression of the TGFB1 gene. However, the exact functional variant in this gene region merits further identification. In support of the aforementioned positive association between CHD risk alleles of TGFB1 and increased TGFβ1 production, increased TGFβ1 levels were observed in different stages of plaque development in some histological studies [7, 3840]. In addition, enhanced TGFβ1 signalling is established to cause cartilaginous metaplasia of vascular media and progressive intima-media thickening after vascular injuries [25, 7, 8, 12]. Interestingly, an increased TGFβ1 regulated gene expression was observed in both atherosclerotic and restenotic lesions [41]. Recently, the TGFβ1 signalling pathway is suggested to be involved in the genetic determining of CHD for the most replicated 9p21.3 locus [4244]. A genetic variant in the SMAD3 gene that encodes one of the downstream activating transcriptional mediators (Smad3) of TGFβ1 signalling [1, 2] was associated with CHD risk in a GWA study [44], which was recently replicated in a large-scale meta-analysis of CHD studies [45].

                  It has previously been shown that abnormal enlargement of human coronary arteries (positive remodelling and aneurysmal coronary lesions) occurs in response to the development of intimal plaque [4649], which is correlated with future acute coronary syndromes and cardiac events [50, 51]. A strong heritable component (h 2  = 0.52) of such abnormal enlargement of the coronary artery in the pathogenesis of coronary artery disease was observed, especially in the proximal coronary artery [52, 53]. Interestingly, the 9p21.3 locus is also associated with increased risk of abdominal aortic aneurysm [5456] and intracranial aneurysm [57, 58]. Recently, high plasma TGFβ1 levels have been implicated in the manifestation of aortic root dilation in Marfan syndrome [6, 13, 59]. Furthermore, genetic variations along the TGFβ1 signalling pathway are associated with coronary artery aneurysm formation and aortic root dilation in Kawasaki diseases [60], whereas mutations in genes of TGFβ1 signalling pathway (TGFBR1, TGFBR2[61, 62], and SMAD3[63]) or TGFβ1 inhibitor genes [64] are implicated in familial or syndromic forms of thoracic aortic aneurysms and dissection. Taken together, this points at altered vascular remodelling from increased TGFβ1 signalling in the pathogenesis of CHD. However, given the fact that TGFβ1 is produced by multiple lineages of resident cells in vascular wall and atherosclerotic lesion and the fact that it acts in an autocrine, paracrine, and endocrine fashion [4], it has been very difficult to pinpoint the exact cellular sources of TGFβ1 that are relevant for the pathogenesis of CHD. More research on this topic is warranted.

                  Some limitations of this meta-analysis should be acknowledged. First, a relatively small number of studies for each SNP was included, and therefore we cannot rule out heterogeneity completely in Caucasian populations although most of the P values for Q-tests were > 0.05. Second, the results in the co-dominant model for rs1800469 and rs1982073 were dominated by the Rotterdam [21] and the Japanese study [23] as shown by the sensitivity analysis. However, this did not affect our main conclusions. Third, the potential publication bias of relatively small sample-sized studies might have affected the results, as there might be eligible studies with negative results that were not published. In the context of these limitations, our current results should be interpreted with caution.

                  Conclusion

                  In conclusion, the present study demonstrates an association between rs1800469 and rs1982073 in the TGFB1 gene and CHD risk in Caucasian populations. Enhanced TGFβ1 signalling may therefore be involved in the pathogenesis of CHD.

                  Abbreviations

                  CHD: 

                  Coronary heart disease

                  CI: 

                  Confidence interval

                  GWA: 

                  Genome-wide association

                  HWE: 

                  Hardy-Weinberg equilibrium

                  LD: 

                  Linkage disequilibrium

                  ORs: 

                  Odds ratios

                  VSMC: 

                  Vascular smooth muscle cells.

                  Declarations

                  Acknowledgments

                  We thank Professor Philippa J. Talmud for sharing data with us. This work was supported by grant 2006B195 of the Netherlands Heart Foundation, and partially (RMB and OF) by grant 8/3-280n-10 of the Moscow Government. See Additional file 1 for the acknowledgments for the PROCARDIS study and the CARDIoGRAM study.

                  Authors’ Affiliations

                  (1)
                  Division of Human Nutrition, Wageningen University and Research Center
                  (2)
                  National Institute for Public Health and the Environment (RIVM)
                  (3)
                  Russian Cardiology Scientific and Production Center, Russian State Medical University
                  (4)
                  Department of Cardiovascular Medicine, University of Oxford

                  References

                  1. Pardali E, Goumans MJ, ten Dijke P: Signaling by members of the TGF-beta family in vascular morphogenesis and disease. Trends Cell Biol 2010,20(9):556–567.PubMedView Article
                  2. Khan R, Agrotis A, Bobik A: Understanding the role of transforming growth factor-beta1 in intimal thickening after vascular injury. Cardiovasc Res 2007,74(2):223–234.PubMedView Article
                  3. Schulick AH, Taylor AJ, Zuo W, Qiu CB, Dong G, Woodward RN, Agah R, Roberts AB, Virmani R, Dichek DA: Overexpression of transforming growth factor beta1 in arterial endothelium causes hyperplasia, apoptosis, and cartilaginous metaplasia. Proc Natl Acad Sci U S A 1998,95(12):6983–6988.PubMedView Article
                  4. Suthanthiran M, Li B, Song JO, Ding R, Sharma VK, Schwartz JE, August P: Transforming growth factor-beta 1 hyperexpression in African-American hypertensives: A novel mediator of hypertension and/or target organ damage. Proc Natl Acad Sci U S A 2000,97(7):3479–3484.PubMed
                  5. Majesky MW, Lindner V, Twardzik DR, Schwartz SM, Reidy MA: Production of transforming growth factor beta 1 during repair of arterial injury. J Clin Invest 1991,88(3):904–910.PubMedView Article
                  6. Frutkin AD, Otsuka G, Stempien-Otero A, Sesti C, Du L, Jaffe M, Dichek HL, Pennington CJ, Edwards DR, Nieves-Cintron M, et al.: TGF-[beta]1 limits plaque growth, stabilizes plaque structure, and prevents aortic dilation in apolipoprotein E-null mice. Arterioscler Thromb Vasc Biol 2009,29(9):1251–1257.PubMedView Article
                  7. Nikol S, Isner JM, Pickering JG, Kearney M, Leclerc G, Weir L: Expression of transforming growth factor-beta 1 is increased in human vascular restenosis lesions. J Clin Invest 1992,90(4):1582–1592.PubMedView Article
                  8. Nabel EG, Shum L, Pompili VJ, Yang ZY, San H, Shu HB, Liptay S, Gold L, Gordon D, Derynck R, et al.: Direct transfer of transforming growth factor beta 1 gene into arteries stimulates fibrocellular hyperplasia. Proc Natl Acad Sci U S A 1993,90(22):10759–10763.PubMedView Article
                  9. Grainger DJ, Kemp PR, Metcalfe JC, Liu AC, Lawn RM, Williams NR, Grace AA, Schofield PM, Chauhan A: The serum concentration of active transforming growth factor-beta is severely depressed in advanced atherosclerosis. Nat Med 1995,1(1):74–79.PubMedView Article
                  10. Wang XL, Sim AS, Wilcken DE: A common polymorphism of the transforming growth factor-beta1 gene and coronary artery disease. Clin Sci (Lond) 1998,95(6):745–746.View Article
                  11. Najar RA, Ghaderian SM, Panah AS: Association of Transforming Growth Factor-beta1 Gene Polymorphisms With Genetic Susceptibility to Acute Myocardial Infarction. Am J Med Sci 2011,342(5):365–370.PubMedView Article
                  12. McCaffrey TA, Consigli S, Du B, Falcone DJ, Sanborn TA, Spokojny AM, Bush HL Jr: Decreased type II/type I TGF-beta receptor ratio in cells derived from human atherosclerotic lesions. Conversion from an antiproliferative to profibrotic response to TGF-beta1. J Clin Invest 1995,96(6):2667–2675.PubMedView Article
                  13. Matt P, Schoenhoff F, Habashi J, Holm T, Van Erp C, Loch D, Carlson OD, Griswold BF, Fu Q, De Backer J, et al.: Circulating transforming growth factor-beta in Marfan syndrome. Circulation 2009,120(6):526–532.PubMedView Article
                  14. Syrris P, Carter ND, Metcalfe JC, Kemp PR, Grainger DJ, Kaski JC, Crossman DC, Francis SE, Gunn J, Jeffery S, et al.: Transforming growth factor-beta1 gene polymorphisms and coronary artery disease. Clin Sci (Lond) 1998,95(6):659–667.View Article
                  15. Yamada Y, Miyauchi A, Goto J, Takagi Y, Okuizumi H, Kanematsu M, Hase M, Takai H, Harada A, Ikeda K: Association of a polymorphism of the transforming growth factor-beta1 gene with genetic susceptibility to osteoporosis in postmenopausal Japanese women. J Bone Miner Res 1998,13(10):1569–1576.PubMedView Article
                  16. Cambien F, Ricard S, Troesch A, Mallet C, Generenaz L, Evans A, Arveiler D, Luc G, Ruidavets JB, Poirier O: Polymorphisms of the transforming growth factor-beta 1 gene in relation to myocardial infarction and blood pressure. The Etude Cas-Temoin de l'Infarctus du Myocarde (ECTIM) Study. Hypertension 1996,28(5):881–887.PubMedView Article
                  17. Grainger DJ, Heathcote K, Chiano M, Snieder H, Kemp PR, Metcalfe JC, Carter ND, Spector TD: Genetic control of the circulating concentration of transforming growth factor type beta1. Hum Mol Genet 1999,8(1):93–97.PubMedView Article
                  18. Shah R, Rahaman B, Hurley CK, Posch PE: Allelic diversity in the TGFB1 regulatory region: characterization of novel functional single nucleotide polymorphisms. Hum Genet 2006,119(1–2):61–74.PubMedView Article
                  19. Dunning AM, Ellis PD, McBride S, Kirschenlohr HL, Healey CS, Kemp PR, Luben RN, Chang-Claude J, Mannermaa A, Kataja V, et al.: A transforming growth factorbeta1 signal peptide variant increases secretion in vitro and is associated with increased incidence of invasive breast cancer. Cancer Res 2003,63(10):2610–2615.PubMed
                  20. Koch W, Hoppmann P, Mueller JC, Schomig A, Kastrati A: Association of transforming growth factor-beta1 gene polymorphisms with myocardial infarction in patients with angiographically proven coronary heart disease. Arterioscler Thromb Vasc Biol 2006,26(5):1114–1119.PubMedView Article
                  21. Sie MP, Uitterlinden AG, Bos MJ, Arp PP, Breteler MM, Koudstaal PJ, Pols HA, Hofman A, van Duijn CM, Witteman JC: TGF-beta 1 polymorphisms and risk of myocardial infarction and stroke: the Rotterdam Study. Stroke 2006,37(11):2667–2671.PubMedView Article
                  22. Crobu F, Palumbo L, Franco E, Bergerone S, Carturan S, Guarrera S, Frea S, Trevi G, Piazza A, Matullo G: Role of TGF-beta1 haplotypes in the occurrence of myocardial infarction in young Italian patients. BMC Med Genet 2008, 9:13.PubMedView Article
                  23. Yokota M, Ichihara S, Lin TL, Nakashima N, Yamada Y: Association of a T29– > C polymorphism of the transforming growth factor-beta1 gene with genetic susceptibility to myocardial infarction in Japanese. Circulation 2000,101(24):2783–2787.PubMedView Article
                  24. Schunkert H, Konig IR, Kathiresan S, Reilly MP, Assimes TL, Holm H, Preuss M, Stewart AF, Barbalic M, Gieger C, et al.: Large-scale association analysis identifies 13 new susceptibility loci for coronary artery disease. Nat Genet 2011,43(4):333–338.PubMedView Article
                  25. A genome-wide association study in Europeans and South Asians identifies five new loci for coronary artery disease Nat Genet 2011,43(4):339–344.
                  26. Sethupathy P, Collins FS: MicroRNA target site polymorphisms and human disease. Trends in genetics: TIG 2008,24(10):489–497.PubMedView Article
                  27. Thakkinstian A, McEvoy M, Minelli C, Gibson P, Hancox B, Duffy D, Thompson J, Hall I, Kaufman J, Leung TF, et al.: Systematic review and meta-analysis of the association between {beta}2-adrenoceptor polymorphisms and asthma: a HuGE review. Am J Epidemiol 2005,162(3):201–211.PubMedView Article
                  28. Marchini J, Howie B: Genotype imputation for genome-wide association studies. Nat Rev Genet 2010,11(7):499–511.PubMedView Article
                  29. Duval S, Tweedie R: Trim and fill: A simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis. Biometrics 2000,56(2):455–463.PubMedView Article
                  30. Viechtbauer W: Conducting meta-analyses in R with the metafor package. Journal of Statistical Software 2010,36(3):1–48.
                  31. Wang Y, Luk AO, Ma RC, So WY, Tam CH, Ng MC, Yang X, Baum L, Lam V, Tong PC, et al.: Independent predictive roles of eotaxin Ala23Thr, paraoxonase 2 Ser311Cys and beta-adrenergic receptor Trp64Arg polymorphisms on cardiac disease in Type 2 Diabetes–an 8-year prospective cohort analysis of 1297 patients. Diabet Med 2010,27(4):376–383.PubMedView Article
                  32. Rao M, Guo D, Jaber BL, Tighiouart H, Pereira BJ, Balakrishnan VS: Transforming growth factor-beta 1 gene polymorphisms and cardiovascular disease in hemodialysis patients. Kidney Int 2004,66(1):419–427.PubMedView Article
                  33. Manginas A, Tsiavou A, Chaidaroglou A, Giamouzis G, Degiannis D, Panagiotakos D, Cokkinos DV: Inflammatory cytokine gene variants in coronary artery disease patients in Greece. Coron Artery Dis 2008,19(8):575–582.PubMedView Article
                  34. Oda K, Tanaka N, Arai T, Araki J, Song Y, Zhang L, Kuchiba A, Hosoi T, Shirasawa T, Muramatsu M, et al.: Polymorphisms in pro- and anti-inflammatory cytokine genes and susceptibility to atherosclerosis: a pathological study of 1503 consecutive autopsy cases. Hum Mol Genet 2007,16(6):592–599.PubMedView Article
                  35. Wang Y, Fu W, Xie F, Chu X, Wang H, Shen M, Sun W, Lei R, Yang L, Wu H, et al.: Common polymorphisms in ITGA2, PON1 and THBS2 are associated with coronary atherosclerosis in a candidate gene association study of the Chinese Han population. J Hum Genet 2010,55(8):490–494.PubMedView Article
                  36. Sudomoina MA, Sukhinina TS, Barsova RM, Favorov AV, Shakhnovich RM, Titov BV, Matveeva NA, Rybalkin IN, Vlasik TN, Ochs MF, et al.: Complex analsis of association of inflammation genes with myocardial infarction. Mol Biol (Mosk) 2010,44(3):463–471.View Article
                  37. Drenos F, Talmud PJ, Casas JP, Smeeth L, Palmen J, Humphries SE, Hingorani AD: Integrated associations of genotypes with multiple blood biomarkers linked to coronary heart disease risk. Hum Mol Genet 2009,18(12):2305–2316.PubMedView Article
                  38. Bobik A, Agrotis A, Kanellakis P, Dilley R, Krushinsky A, Smirnov V, Tararak E, Condron M, Kostolias G: Distinct patterns of transforming growth factor-beta isoform and receptor expression in human atherosclerotic lesions, Colocalization implicates TGF-beta in fibrofatty lesion development. Circulation 1999,99(22):2883–2891.PubMedView Article
                  39. Piao M, Tokunaga O: Significant expression of endoglin (CD105), TGFbeta-1 and TGFbeta R-2 in the atherosclerotic aorta: an immunohistological study. J Atheroscler Thromb 2006,13(2):82–89.PubMedView Article
                  40. Panutsopulos D, Papalambros E, Sigala F, Zafiropoulos A, Arvanitis DL, Spandidos DA: Protein and mRNA expression levels of VEGF-A and TGF-beta1 in different types of human coronary atherosclerotic lesions. Int J Mol Med 2005,15(4):603–610.PubMed
                  41. O'Brien ER, Bennett KL, Garvin MR, Zderic TW, Hinohara T, Simpson JB, Kimura T, Nobuyoshi M, Mizgala H, Purchio A, et al.: Beta ig-h3, a transforming growth factor-beta-inducible gene, is overexpressed in atherosclerotic and restenotic human vascular lesions. Arterioscler Thromb Vasc Biol 1996,16(4):576–584.PubMedView Article
                  42. McPherson R, Pertsemlidis A, Kavaslar N, Stewart A, Roberts R, Cox DR, Hinds DA, Pennacchio LA, Tybjaerg-Hansen A, Folsom AR, et al.: A common allele on chromosome 9 associated with coronary heart disease. Science 2007,316(5830):1488–1491.PubMedView Article
                  43. Helgadottir A, Thorleifsson G, Manolescu A, Gretarsdottir S, Blondal T, Jonasdottir A, Jonasdottir A, Sigurdsson A, Baker A, Palsson A, et al.: A common variant on chromosome 9p21 affects the risk of myocardial infarction. Science 2007,316(5830):1491–1493.PubMedView Article
                  44. Samani NJ, Erdmann J, Hall AS, Hengstenberg C, Mangino M, Mayer B, Dixon RJ, Meitinger T, Braund P, Wichmann HE, et al.: Genomewide association analysis of coronary artery disease. N Engl J Med 2007,357(5):443–453.PubMedView Article
                  45. Angelakopoulou A, Shah T, Sofat R, Shah S, Berry DJ, Cooper J, Palmen J, Tzoulaki I, Wong A, Jefferis BJ, et al.: Comparative analysis of genome-wide-association studies signals for lipids, diabetes, and coronary heart disease: Cardiovascular Biomarker Genetics Collaboration. Eur Heart J 2011, 3:1–2.
                  46. Glagov S, Weisenberg E, Zarins CK, Stankunavicius R, Kolettis GJ: Compensatory enlargement of human atherosclerotic coronary arteries. N Engl J Med 1987,316(22):1371–1375.PubMedView Article
                  47. McPherson DD, Sirna SJ, Hiratzka LF, Thorpe L, Armstrong ML, Marcus ML, Kerber RE: Coronary arterial remodeling studied by high-frequency epicardial echocardiography: an early compensatory mechanism in patients with obstructive coronary atherosclerosis. J Am Coll Cardiol 1991,17(1):79–86.PubMedView Article
                  48. Stiel GM, Stiel LS, Schofer J, Donath K, Mathey DG: Impact of compensatory enlargement of atherosclerotic coronary arteries on angiographic assessment of coronary artery disease. Circulation 1989,80(6):1603–1609.PubMedView Article
                  49. Hermiller JB, Tenaglia AN, Kisslo KB, Phillips HR, Bashore TM, Stack RS, Davidson CJ: In vivo validation of compensatory enlargement of atherosclerotic coronary arteries. Am J Cardiol 1993,71(8):665–668.PubMedView Article
                  50. Schoenhagen P, Ziada KM, Kapadia SR, Crowe TD, Nissen SE, Tuzcu EM: Extent and direction of arterial remodeling in stable versus unstable coronary syndromes: an intravascular ultrasound study. Circulation 2000,101(6):598–603.PubMedView Article
                  51. Okura H, Kobayashi Y, Sumitsuji S, Terashima M, Kataoka T, Masutani M, Ohyanagi M, Shimada K, Taguchi H, Yasuga Y, et al.: Effect of culprit-lesion remodeling versus plaque rupture on three-year outcome in patients with acute coronary syndrome. Am J Cardiol 2009,103(6):791–795.PubMedView Article
                  52. Fischer M, Broeckel U, Holmer S, Baessler A, Hengstenberg C, Mayer B, Erdmann J, Klein G, Riegger G, Jacob HJ, et al.: Distinct heritable patterns of angiographic coronary artery disease in families with myocardial infarction. Circulation 2005,111(7):855–862.PubMedView Article
                  53. Nishioka T, Nagai T, Luo H, Kitamura K, Hakamata N, Akanuma M, Katsushika S, Uehata A, Takase B, Isojima K, et al.: Coronary remodeling of proximal and distal stenotic atherosclerotic plaques within the same artery by intravascular ultrasound study. Am J Cardiol 2001,87(4):387–391.PubMedView Article
                  54. Helgadottir A, Thorleifsson G, Magnusson KP, Gretarsdottir S, Steinthorsdottir V, Manolescu A, Jones GT, Rinkel GJ, Blankensteijn JD, Ronkainen A, et al.: The same sequence variant on 9p21 associates with myocardial infarction, abdominal aortic aneurysm and intracranial aneurysm. Nat Genet 2008,40(2):217–224.PubMedView Article
                  55. Biros E, Cooper M, Palmer LJ, Walker PJ, Norman PE, Golledge J: Association of an allele on chromosome 9 and abdominal aortic aneurysm. Atherosclerosis 2010,212(2):539–542.PubMedView Article
                  56. Thompson AR, Golledge J, Cooper JA, Hafez H, Norman PE, Humphries SE: Sequence variant on 9p21 is associated with the presence of abdominal aortic aneurysm disease but does not have an impact on aneurysmal expansion. Eur J Hum Genet 2009,17(3):391–394.PubMedView Article
                  57. Yasuno K, Bilguvar K, Bijlenga P, Low SK, Krischek B, Auburger G, Simon M, Krex D, Arlier Z, Nayak N, et al.: Genome-wide association study of intracranial aneurysm identifies three new risk loci. Nat Genet 2010,42((5):420–425.PubMedView Article
                  58. Nakaoka H, Takahashi T, Akiyama K, Cui T, Tajima A, Krischek B, Kasuya H, Hata A, Inoue I: Differential effects of chromosome 9p21 variation on subphenotypes of intracranial aneurysm: site distribution. Stroke 2010,4(8):1593–1598.View Article
                  59. Habashi JP, Judge DP, Holm TM, Cohn RD, Loeys BL, Cooper TK, Myers L, Klein EC, Liu G, Calvi C, et al.: Losartan, an AT1 antagonist, prevents aortic aneurysm in a mouse model of Marfan syndrome. Science 2006,312(5770):117–121.PubMedView Article
                  60. Shimizu C, Jain S, Davila S, Hibberd ML, Lin KO, Molkara D, Frazer JR, Sun S, Baker AL, Newburger JW, et al.: Transforming growth factor-beta signaling pathway in patients with Kawasaki disease. Circ Cardiovasc Genet 2011,4(1):16–25.PubMedView Article
                  61. Pannu H, Fadulu VT, Chang J, Lafont A, Hasham SN, Sparks E, Giampietro PF, Zaleski C, Estrera AL, Safi HJ, et al.: Mutations in transforming growth factor-beta receptor type II cause familial thoracic aortic aneurysms and dissections. Circulation 2005,112(4):513–520.PubMedView Article
                  62. Loeys BL, Chen J, Neptune ER, Judge DP, Podowski M, Holm T, Meyers J, Leitch CC, Katsanis N, Sharifi N, et al.: A syndrome of altered cardiovascular, craniofacial, neurocognitive and skeletal development caused by mutations in TGFBR1 or TGFBR2. Nat Genet 2005,37(3):275–281.PubMedView Article
                  63. van de Laar IM, Oldenburg RA, Pals G, Roos-Hesselink JW, de Graaf BM, Verhagen JM, Hoedemaekers YM, Willemsen R, Severijnen LA, Venselaar H, et al.: Mutations in SMAD3 cause a syndromic form of aortic aneurysms and dissections with early-onset osteoarthritis. Nat Genet 2011,43(2):121–126.PubMedView Article
                  64. Prakash SK, LeMaire SA, Guo DC, Russell L, Regalado ES, Golabbakhsh H, Johnson RJ, Safi HJ, Estrera AL, Coselli JS, et al.: Rare copy number variants disrupt genes regulating vascular smooth muscle cell adhesion and contractility in sporadic thoracic aortic aneurysms and dissections. Am J Hum Genet 2010,87(6):743–756.PubMedView Article
                  65. Pre-publication history

                    1. The pre-publication history for this paper can be accessed here:http://​www.​biomedcentral.​com/​1471-2350/​13/​39/​prepub

                  Copyright

                  © Lu et al.; licensee BioMed Central Ltd. 2012

                  This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( http://​creativecommons.​org/​licenses/​by/​2.​0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.