Polymorphisms in MMP Family and TIMP Genes and Carotid Artery Intima-Media Thickness
Background and Purpose— Genetic variation in a number of MMP and TIMP genes have been implicated as risk factors for atherosclerosis, although such studies have been generally small and produced conflicting results. We have therefore sought to address this issue in a large, well-phenotyped community population to assess the effect of a number of polymorphisms in both MMP and TIMP genes on carotid artery intima-media thickness (IMT).
Methods— In a community population (n=1000), IMT was determined using ultrasound in the common carotid artery, carotid bulb, and bifurcation. Eight polymorphisms in 6 MMP genes were genotyped (MMP1 A-519G, MMP2 C-1306T, MMP2 C-735T, MMP3 -1171 5A/6A, MMP9 R279Q, TIMP2 G853A, TIMP3 A-915G, and T-1296C) and assessed for their effect on carotid IMT alone and by interaction with common cardiovascular risk factors.
Results— An association was found between MMP9 R279Q and internal carotid artery bulb IMT (P=0.002), but there was no linear trend between allele number and IMT and no association with common carotid artery or bulb IMT. In addition, 3 interactions were found between polymorphisms and hypertension (MMP1 A-519G, MMP3 5A/6A, TIMP3 T-1296C), the latter 2 of which showed a significant trend test for linearity with increasing copy number and increased internal carotid artery bulb IMT. All remained significant after correction for multiple testing.
Conclusions— Our findings provide little support for genetic variants of MMP as direct risk factors for IMT. However, the interaction findings between MMP variants and hypertension suggest that hypertensive carriers of these alleles may be at greater risk for increased IMT and future cardiovascular disease. These findings need replication in hypertensive populations to assess their effects more fully.
Atherosclerosis, leading to ischemic heart disease and stroke, is a major cause of adult morbidity and mortality in industrialized countries.1 There is significant evidence, from studies of twins and families, that the development of atherosclerosis is controlled at least in part by genetic factors.2 However, few of the responsible genes have been identified to date.3 Remodeling of the extracellular matrix in the arterial wall occurs during all stages of atherosclerosis, including compensatory remodeling, plaque development, and plaque instability.4 There is mounting evidence to indicate that disturbances in arterial extracellular matrix remodeling play a key role in the atherosclerotic process.5
The matrix metalloproteinases (MMPs) are a family of protein-digesting enzymes that are fundamental mediators of matrix turnover,6 and are capable of catalyzing the breakdown of major extracellular matrix components.5 The ability of cells to degrade and rearrange extracellular matrix proteins is crucial for normal growth and development.7 However, aberrant extracellular matrix remodeling is a feature of several major cardiovascular diseases,5,7 and it is thought that deregulation of MMP enzymatic activity may be involved in the underlying etiology of these and other invasive and inflammatory diseases.7
The MMPs are synthesized by a number of cell types in response to various physiological or biological stimuli. Their enzymatic activity is strictly regulated at several levels through control of gene expression, pro-enzyme activation, cellular protein levels, secretory activity, and interactions with endogenous inhibitors, such as the tissue inhibitors of metalloproteinases (TIMP-1, -2, -3, and -4).6,7 Multiple previous studies have shown an association between atherosclerosis and overexpression or underexpression of specific MMPs and TIMPs.4,8,9 However, increased expression of an MMP does not necessarily correlate with increased enzymatic activity, because most are synthesized as inactive zymogens.4
Numerous studies have identified genetic polymorphisms within the MMP genes that alter expression levels of the enzymes,10 implying a possible role in atherosclerotic development. However, studies identifying which of these polymorphisms, if any, are associated with specific cardiovascular diseases have been less successful. Many are based on small sample sizes,11–13 require replication,14,15 or conflict with results of other studies.16,17 In addition, many of the studies are based on late-stage cardiovascular diseases, making meta-analysis impossible because of differing underlying pathology and end points used. Some of the identified polymorphisms have still to be investigated, and no previous studies have yet investigated the association between specific genetic polymorphisms in the TIMP genes and cardiovascular disease.
The use of an intermediate phenotype to focus on a single aspect of the disease process avoids complications caused by the heterogeneity of cardiovascular disease phenotypes and patho-physiological processes. Intima-media thickness (IMT) of the carotid artery, measured by high-resolution ultrasound, correlates closely with pathological measurements of the intima-media complex,18 and is an independent predictor of both myocardial infarction and stroke risk.19 It is increasingly used as an intermediate phenotype for atherosclerosis. Twin and family studies have demonstrated that carotid IMT has significant heritability (30% to 66%).20
In this study we examined the relationship between functional MMP family gene polymorphisms and carotid IMT in 1000 individuals from a community population. We focused on polymorphisms previously reported in genes encoding an MMP from each of the 3 main classes: collagenases, gelatinases, and stromelysins, as well as their endogenous tissue inhibitors, TIMP-2 and TIMP-3.
Materials and Methods
The present investigation used samples from the Carotid Atherosclerosis Progression study (CAPS), details of which have been published previously.21 Briefly, all members of a German primary health care service (n=32 708) who lived within a 50-km radius of 5 study sites in Western Germany were invited to participate. Within a predefined time limit, 6962 individuals agreed to take part. Baseline IMT measurements and blood samples were taken and a full cardiovascular risk profile compiled for each individual as previously described.21 In this study, we included the first 1000 consecutively recruited subjects aged 50 to 65 years. Informed consent on the use of samples for analysis was obtained from all participants before entry and the study was approved by the local research ethics committee.
Ultrasound measurement of IMT was performed using a 7.5- to 10.0-MHz linear array transducer (P700SE; Phillips Medical System). Using antero-oblique ionization, far-wall carotid IMT was visualized within the common carotid artery (CCA), internal carotid artery bulb (ICA), and carotid bifurcation (BIF) on both sides. Digitally captured images were analyzed offline using semi-automated software analysis to produce mean IMT measurements for each carotid region. The instrument parameters and methods used are described in detail in Sitzer et al.22 High levels of inter-observer and intra-observer reproducibility were obtained as previously described.22
The polymorphisms analyzed are shown in Table 1. SNPs were chosen on the basis of demonstration of functionality or because previous studies had associated them with cardiovascular disease. All genotyping was performed blind to patient details. DNA was isolated from 4- to 12-mL whole blood using Nucleon Bacc3 DNA extraction kits (Amersham Biosciences). The MMP-2 C-1306T, MMP-9 R279Q, and TIMP-2 G853A polymorphisms were analyzed by a commercial company (www.kbioscience.co.uk) using a patent-protected process involving automated polymerase chain reaction and single base extension with fluorescent detection. Genotyping of all other polymorphisms was undertaken using restriction fragment-length polymorphism analysis after amplification by polymerase chain reaction (Table 1). The amplified polymerase chain reaction products were subjected to 8-hour digestion with the relevant restriction endonuclease. The resulting fragments were separated on a 2% Micro Agarose gel stained with ethidium bromide, and visualized under ultraviolet light. All ambiguous genotypes were retyped twice then excluded from the analysis if they remained unclear.
Statistical analysis was performed using SPSS (version 12). The distribution of IMT was skewed and therefore IMT measurements were transformed using the reciprocal to obtain a normal distribution. Geometric mean IMT values are given in the text for clarity. There were no differences between right and left CCA, ICA, or BIF IMT values; therefore, mean values were used in the analysis.
Univariate analysis was initially performed to investigate the relationship between each genotype and mean CCA, ICA, and BIF IMT. Multivariable analysis was then conducted using a general linear model (binary logistic regression) to control for the effects of cardiovascular risk factors (age, gender, body mass index, smoking status [pack years], low-density lipoprotein cholesterol, history of arterial hypertension, and diabetes mellitus). In addition, exploratory analyses were undertaken for interactions between genotype and certain risk factors (body mass index, history of arterial hypertension, smoking status, and low-density lipoprotein). The direction of genotype effect for any interaction was determined using ANOVA and the trend test for linearity. The polymorphisms were examined for additive (012), dominant (011), and recessive (001) effects on IMT.
We performed analyses of 6 independent genes, and therefore for our primary analyses we controlled for these multiple comparisons using the Bonferroni correction, raising the significance level from 0.05 to 0.008. This was performed before analysis and represents the significance level that must be met before an association can be considered significant. Because the IMT at different points in the carotid tree (CCA, IMT, and BIF) were highly correlated, applying a Bonferroni correction for the analyses of associations with the different carotid segments would be excessively conservative so we did not apply this. Correlations between CCA and ICA, CCA and BIF, and ICA and BIF had Pearson correlation coefficients of 0.316, 0.399, and 0.420, respectively, all significant at P=<0.001 level.
The population size of 1000 individuals is sufficient to detect a 2% variance explained by a single variant on IMT at α=0.001 and power of 80%. Power was calculated using the Genetic Power Calculator (http://pngu.mgh.harvard.edu/≈purcell/gpc/qtlassoc.html).
Demographics of the study population are shown in Table 2. IMT measurements and DNA for genotyping were available for all 1000 subjects. All genotype distributions were in Hardy Weinberg equilibrium, except the TIMP-3 A-915G polymorphism (χ2=4.177; P=0.041). All results are reported in the supplemental Table I, available online at http://stroke.ahajournals.org.
On univariate analysis, only the associations between MMP9 R279Q and ICA IMT and TIMP3 T-1296C and BIF IMT were found to be significant before correction (P=0.002 and P=0.041, respectively; Table 3). These remained nominally significant after correction for cardiovascular risk factors age, sex, body mass index, smoking status, low-density lipoprotein cholesterol, and history of arterial hypertension (Table 3), although only MMP9 R279Q was under the previously determined threshold of P=0.008 to account for multiple testing. On comparing means between genotype groups, a significant difference was only identified in ICA IMT, with the presence of ≥1 copies of MMP9 279Q allele compared with the presence of no copies (Table 4). Carriers of the MMP9 279Q allele had a lower IMT than noncarriers. No difference in means could be identified with presence or absence of TIMP3 T-1296C. Despite the association with ≥1 copies of MMP9 279Q, there was no significant trend test for linearity with increasing copy number, and presence of 2 copies of the allele did not result in significantly higher ICA IMT than presence of 0 or 1 copies of MMP9 279Q, although 1 copy showed a nonsignificant lower IMT than 0 or 1 copy (data not shown).
An analysis examining gene–environment interactions between MMP genotype and the cardiovascular risk factors body mass index, history of arterial hypertension, smoking status, and low-density lipoprotein cholesterol revealed 3 significant associations with history of arterial hypertension and MMP1 A-519G, MMP3 5A/6A, and TIMP3 T-1296C (Table 5). In all 3 cases the association was with the subpopulation with hypertension, the change was in ICA IMT, and the rare allele led to an increase in IMT. A correction of P=0.0125 was applied because we have sought associations with a single variant but with 4 risk factors. All 3 variants met this corrected threshold. In addition, both MMP3 5A/6A and TIMP3 T-1296C showed a significant trend test for linearity (P=0.023 and P=0.021, respectively) with increasing copy number of the risk allele associated with increased ICA IMT (data not shown).
Considerable experimental data support the role of altered MMP activity in the pathogenesis of various stages of atherosclerosis.6 Previous genetic studies have found associations between MMPs and cardiovascular disease, albeit with some inconsistencies, and replication is needed to confirm or deny these findings. In this large community population, we have examined associations between polymorphisms in a number of MMPs and TIMPs implicated in the pathogenesis of atherosclerosis and carotid IMT, a marker of early atherosclerosis and vessel remodeling.
We examined associations between 8 SNPs and IMT measured at 3 points in the carotid artery: CCA, ICA, and BIF. In a univariate analysis corrected for cardiovascular risk factors, 2 associations were identified, although after correction for multiple testing only MMP9 R279Q remained significant. Despite this positive association, there was no significant linear trend with increasing allele number, and presence of 0 or 2 alleles both gave higher mean IMT than presence of a single allele, suggesting that this association may be spurious. Furthermore, there was no association between this SNP and IMT at other points in the carotid artery. This argues against a biologically relevant associations and the lack of association with other segments suggest it might be a false-positive finding.
On interaction analysis we found interactions between three SNPs and hypertension on ICA IMT, all of which remained significant after correcting for the number of interactions investigated. In addition, there was a significant trend in increasing ICA IMT with increasing copy number of both the MMP3 5A allele and the TIMP3–1296C allele. Unfortunately, the number of individuals with both risk alleles was too small to allow a meaningful joint genotype analysis, although such an investigation in a larger hypertensive population may be warranted.
It is intriguing that the interactions were all with hypertension. Altered MMP activity has been implicated as a mechanism in early vascular remodeling secondary to hypertension.5 IMT thickness appears to represent not only early atherosclerosis but also adaptive remodeling particularly to hypertension.19 Vascular remodeling is limited during the early atherosclerotic changes which changes in IMT represent. As such, the effects of MMP genetics variants may not be pronounced enough to be detected by increases in IMT in normotensive subjects at this early stage, but that as hypertension increases the degree of vascular remodeling the effect of these genetic variants then become manifest. This finding was a secondary analysis and therefore needs replicating in further populations. Future studies should look at not only hypertensive IMT populations but also individuals with more advanced atherosclerotic disease.
A major strength of our study is the large population size. An important factor contributing to previous conflicting results is likely to be the small sample sizes leading to a lack of power, possibly exacerbated by publication bias. Our large sample size provides more robust information and excludes some of the previously identified candidates as having a role in early atherosclerotic changes as measured by carotid IMT. The use of IMT as an intermediate phenotype also offers significant advantages. The use of a continuous variable, rather than a dichotomous variable such as stroke or myocardial infarction, increases statistical power and reduces the problem of incomplete penetrance in which individuals may have subclinical atherosclerosis, which will lead to clinical events in subsequent years. Carotid IMT has been successfully used to show robust replicable associations with other cardiovascular candidate genes.23,24
Our findings provide little support for genetic variants of MMP as direct risk factors for IMT, at least in the early stages of the atherosclerotic process typified by vessel remodeling and thickening. MMPs have been implicated in both the early and later stages of atherosclerosis. IMT provides a measure of early atherosclerosis and remodeling but provides no information on processes involved in later atherosclerosis and plaque rupture. Further studies investigating associations between MMP genotypes and symptomatic disease will be required to determine whether these variants play a role in plaque rupture. In addition, our study only examines the role of genetic variants in MMP genes as risk factors for atherosclerosis, and does not exclude a role of posttranslational changes or other downstream processes altering MMP activity in atherosclerosis.
The interaction findings between MMP variants and hypertension, however, suggests that hypertensive carriers of these alleles may be at greater risk for increased IMT and future cardiovascular disease. These findings need replication in series of hypertensive cases and controls, and populations with later stage atherosclerotic disease, to assess their effects more fully.
Sources of Funding
This work is supported by a grant from St. George’s Hospital Medical Research Committee CA and is supported by a grant from the British Heart Foundation PG/04/080/17377. Sample collection was supported by a grant from the Stiftung Deutsche Schlaganfall-Hilfe (German Stroke Foundation).
- Received April 19, 2007.
- Accepted May 3, 2007.
Chamberlain JC, Galton DJ. Genetic susceptibility to atherosclerosis. Br Med Bull. 1990; 46: 917–940.
Galis ZS, Khatri JJ. Matrix metalloproteinases in vascular remodeling and atherogenesis: the good, the bad, and the ugly. Circ Res. 2002; 90: 251–262.
Jones CB, Sane DC, Herrington DM. Matrix metalloproteinases: a review of their structure and role in acute coronary syndrome. Cardiovasc Res. 2003; 59: 812–823.
Price SJ, Greaves DR, Watkins H. Identification of novel, functional genetic variants in the human matrix metalloproteinase-2 gene: role of Sp1 in allele-specific transcriptional regulation. J Biol Chem. 2001; 276: 7549–7558.
Gnasso A, Motti C, Irace C, Carallo C, Liberatoscioli L, Bernardini S, Massoud R, Mattioli PL, Federici G, Cortese C. Genetic variation in human stromelysin gene promoter and common carotid geometry in healthy male subjects. Arterioscler Thromb Vasc Biol. 2000; 20: 1600–1605.
Rauramaa R, Vaisanen SB, Luong LA, Schmidt-Trucksass A, Penttila IM, Bouchard C, Toyry J, Humphries SE. Stromelysin-1 and interleukin-6 gene promoter polymorphisms are determinants of asymptomatic carotid artery atherosclerosis. Arterioscler Thromb Vasc Biol. 2000; 20: 2657–2662.
Rundek T, Elkind MS, Pittman J, Boden-Albala B, Martin S, Humphries SE, Juo SH, Sacco RL. Carotid intima-media thickness is associated with allelic variants of stromelysin-1, interleukin-6, and hepatic lipase genes: the Northern Manhattan Prospective Cohort Study. Stroke. 2002; 33: 1420–1423.
Pearce E, Tregouet DA, Samnegard A, Morgan AR, Cox C, Hamsten A, Eriksson P, Ye S. Haplotype effect of the matrix metalloproteinase-1 gene on risk of myocardial infarction. Circ Res. 2005; 97: 1070–1076.
Vasku A, Goldbergova M, Izakovicova Holla L, Siskova L, Groch L, Beranek M, Tschoplova S, Znojil V, Vacha J. A haplotype constituted of four MMP-2 promoter polymorphisms (-1575G/A, -1306C/T, -790T/G and -735C/T) is associated with coronary triple-vessel disease. Matrix Biol. 2004; 22: 585–591.
Wang J, Warzecha D, Wilcken D, Wang XL. Polymorphism in the gelatinase B gene and the severity of coronary arterial stenosis. Clin Sci. 2001; 101: 87–92.
Wong M, Edelstein J, Wollman J, Bond MG. Ultrasonic-pathological comparison of the human arterial wall. Verification of intima-media thickness. Arterioscler Thromb. 1993; 13: 482–486.
Bots ML, Hofman A, Grobbee DE. Increased common carotid intima-media thickness: adaptive response or a reflection of atherosclerosis? Findings from the Rotterdam Study. Stroke. 1997; 28: 2442–2447.
Lorenz MW, von Kegler S, Steinmetz H, Markus HS, Sitzer M. Carotid intima-media thickening indicates a higher vascular risk across a wide age range. Prospective data from the Carotid Atherosclerosis Progression study (CAPS). Stroke. 2006; 37: 87–92.
Jerrard-Dunne P, Sitzer M, Risley P, Buehler A, von Kegler S, Markus HS. Inflammatory gene load is associated with enhanced inflammation and early carotid atherosclerosis in smokers. Stroke. 2004; 35: 2438–2443.
Markus HS, Labrum R, Bevan S, Reindl M, Egger G, Wiedermann CJ, Xu O, Kiechl S, Willeit J. Genetic and acquired inflammatory conditions are synergistically associated with early carotid atherosclerosis. Stroke. 2006; 37: 2253–2259.