Sage Journals: Discover world-class research

Abstract

We aimed to test the association between matrix metalloproteinase (MMP) genetic polymorphisms and (a) intima–media thickness in the common carotid (IMTcc) and (b) the presence of plaques in the carotid and femoral bifurcations. Carotid and femoral bifurcations were scanned with ultrasound in 762 Cypriot community dwellers (46% men) over the age of 40 years. IMTcc and the presence of plaques were recorded. The MMP1 1G/2G, MMP3 5A/6A, MMP7 -181A>G, MMP9 R279Q, and MMP12 -82A>G polymorphisms were determined with the TaqMan method. In men, the presence of plaques in any bifurcation was associated with the MMP9 279Q allele (OR_adjusted=4.50; 95% CI=2.0 to 10.1; p<0.001) and the MMP7 -181A allele was associated with the presence of femoral plaques (OR_adjusted=2.61; 95% CI=1.36 to 4.99; p=0.004). In women, the presence of femoral plaques was associated with the MMP12 -82G allele (OR_adjusted=1.9; 95% CI=1.14 to 3.16; p=0.014). Our results suggest that the effect of common MMP genotypes on plaque presence may be site- and sex-dependent.

Keywords

atherosclerosis carotid plaques femoral plaques genetic polymorphisms intima–media thickness (IMT)matrix metalloproteinases (MMP)

Introduction

Carotid intima–media thickness (IMTc)^1–3 and the presence and number of carotid plaques^4–7 as measured with high-resolution ultrasound, are surrogate markers of atherosclerosis and have been shown to be associated with the extent and severity of coronary atherosclerosis⁸ as well as future episodes of myocardial infarction and stroke.^1–4,6,7 Although IMTc is most often measured in studies, doubts have been raised recently as to its clinical utility, whereas the presence of carotid plaques has been shown to have a greater utility in estimating risk for future cardiovascular events, both in observational studies and a meta-analysis.^9,10 In addition to carotid atherosclerosis, atherosclerosis in the femoral arteries (i.e. presence and number of femoral plaques) has also been shown to predict cardiovascular disease (CVD), independent of carotid atherosclerosis,^10–12 and may represent a different pathophysiological process with structural differences.^13–15

Metalloproteinases (MMPs) are a family of endopeptidases with the ability to degrade the extracellular matrix (ECM), which have been shown to be involved in the remodelling of atherosclerotic plaques.¹⁶ Currently, matrix remodelling is thought to contribute to plaque instability, rupture and thrombus formation¹⁷ and is considered a dynamic process. Current concepts of atherosclerosis, which emphasize the remodelling of the arterial wall, provide a framework for understanding how, through weakening and destabilization, angiographically undetectable moderate lesions could become culprits for acute coronary syndromes.¹⁶ Increasing evidence indicates that genetic variability (single nucleotide polymorphisms; SNPs) in the genes coding for MMPs may alter their transcriptional activity and contribute to susceptibility to cardiovascular disease.^16,18

The aim of this study was to test the association between ultrasonically determined measures of carotid and femoral subclinical atherosclerosis (i.e. IMT in the common carotid, carotid and femoral plaque presence) and functional genetic polymorphisms in genes encoding five MMPs (MMP1 1G/2G (rs1799750), MMP3 5A/6A (rs3025058), MMP7 -181A>G (rs11568818), MMP9 R279Q (rs17576), and MMP12 -82A>G (rs2276109)) in community-dwelling middle-aged men and women.

Methods

Subjects

The Cyprus Study is a population-based cohort study of cardiovascular disease and atherosclerosis in 1106 individuals aged 40 years or more from two areas in Cyprus.^19,20 Baseline data have been collected from residents of two randomly selected areas in Cyprus and their relatives living in any major town between 2003 and 2008. All inhabitants were identified through the population list held at the Mayor’s office and all those over the age of 40 years were invited to participate. The overall participation rate of those invited was 95%. The Ethics Committee of the Cyprus Institute of Neurology and Genetics approved the study. All participants provided written informed consent. For this analysis, baseline data of the first 762 subjects recruited, who had complete data on intima–media thickness in the common carotid (IMTcc), plaque and MMP genotypes, were used.

Cardiovascular disease status, risk factors and ultrasonic measurements

Cardiovascular disease status (prevalent cases), risk factors and ultrasonic measurements have been described previously.¹¹ Type 2 diabetes (T2D) was considered present if subjects were taking oral hypoglycemic agents, insulin or had a fasting blood sugar in excess of 7.0 mmol/L. Hypertension was considered present if subjects were on treatment or had a systolic blood pressure (BP) ≥ 140 mmHg or a diastolic BP ≥ 90 mmHg.

All scans were performed using a Philips (ATL) HDI 5000 duplex scanner (Seattle, WA, USA). The IMT complex of the far wall of the common carotid artery was measured at its thickest part (mean of three readings) on both transverse and longitudinal sections 1.5–2.0 cm proximal to the bifurcation which was free from any focal thickening (i.e. atherosclerotic plaques). The mean of the measurements from both carotid arteries was used in the analysis (IMTcc). The extracranial vessels (common carotid artery (CCA), external carotid artery (ECA), internal carotid artery (ICA)) and approximately 1.5–2 cm proximal and distal to the femoral bifurcations were carefully examined in both transverse and longitudinal section for the presence of plaques. An arterial bifurcation (internal carotid or common femoral bifurcation) was classified as being affected by plaque if there was a focal thickening greater than 1.2 mm that did not uniformly involve the whole arterial wall.⁵ The two ultrasonographers who performed the ultrasonic scans were not aware of the clinical, biochemical or genetic risk factors of the subjects. The inter- and intra-observer variabilities of the two ultrasonographers have been reported previously.¹¹

Biochemical measurements

A venous blood sample was obtained from all subjects after a 6–12-hour fast. After at least 15 minutes at room temperature, the samples were centrifuged and serum and plasma were immediately aliquoted and frozen to −80°C until analysis. Serum fasting glucose, total cholesterol (TChol), high-density lipoprotein-cholesterol (HDL-C), low-density lipoprotein-cholesterol (LDL-C) and triglycerides (TG) were quantitatively determined with enzymatic color tests. All tests were performed with the Olympus AU640 automatic analyzer and all reagents were from Olympus Diagnostics (Beckman Coulter International SA, Nyon, Switzerland). Coefficients of variation were less than 5% for all measurements.

DNA was extracted from whole blood in EDTA anticoagulant with the use of commercial kits (Qiagen maxiBlood extraction kit; Qiagen Hilden, Germany) according to the manufacturer’s instructions. The MMP1 1G/2G (rs1799750), MMP3 5A/6A (rs3025058), MMP7 -181A>G (rs11568818), MMP9 R279Q (rs17576), and MMP12 -82A>G (rs2276109) single nucleotide polymorphisms were determined with the TaqMan method. The primers and probes used are shown in the Supplementary Table. Oligonucleotides were synthesized by Applied Biosystems^® (ABI), (Paisly, UK). Amplification was performed in a 5 μl volume with 4.25 ng of genomic DNA, 2.5 μl of TaqMan^® Universal PCR Master Mix, 0.125 μl of 40× Assay Mix and 2.375 μl of DNase-free dH₂O according to the ABI protocol for dry template dispersion. The polymerase chain reaction (PCR) conditions were according to the ABI TaqMan^® protocol and 384-well plates were used. After amplification, the plates were read with the ABI Prism 7900HT Sequence Detection System for allelic discrimination and results were analyzed using the SD 2.1 software. In some cases allelic discrimination was not possible due to poor DNA quality, resulting in slightly different numbers for genotypes. The MMP3 5A/6A polymorphism was additionally analyzed with the Restriction Fragment Length Polymorphism (RFLP) method, thus resulting in no missing genotypes. Missing cases were <5% for all genotypes and therefore no missing data methods were used. A random number generator was used for duplicates and four no template controls (NTCs) were used in every plate to assure quality control. Genotyping was performed at the Centre for Cardiovascular Genetics, UCL, London, UK.

Statistical methods

We tested the Hardy–Weinberg equilibrium on a contingency table of observed-versus-predicted genotypic frequencies with an exact test.

Subclinical atherosclerosis outcomes were classified as binary variables: IMTcc greater or less than 0.07 cm (population mean); presence versus absence of carotid or femoral plaques and presence versus absence of plaques in any of the four bifurcations scanned (0 vs 1–4).

Associations between subclinical atherosclerosis outcomes and MMP genotypes were investigated by computing age- and sex-adjusted basic (Model 1) and multivariable adjusted (Model 2: Model 1 + smoking in pack-years, hypertension, T2D, body mass index (BMI) and TChol) odds ratios using logistic regression. We also conducted sex-specific analyses because of recent evidence of sex-specific associations between the MMP9 (-1562C>T) polymorphism and carotid artery stiffness.²¹ Formal testing for sex interaction was done by computing the interaction term in the multivariable model, as appropriate. To adjust for multiple testing, the Bonferroni correction was used, raising the significance level from 0.05 to 0.01 (0.05/5 SNPs) for each outcome as well as step-wise methods.

All statistical analyses were computed using Stata v.12.0 (StataCorp, College Station, TX, USA).

Results

Of the 762 subjects studied, 113 (14.8%) had clinical evidence of prevalent cardiovascular disease; documented myocardial infarction with hospital admission and/or presence of infarct on electrocardiogram (ECG) in 58, presence of angina or coronary intervention and/or myocardial ischemia on ECG in 43, ischemic hemispheric neurological events in nine (five strokes, four TIAs) and lower limb ischemia in three.

All genotypes were in Hardy–Weinberg equilibrium. Baseline characteristics of the population according to MMP genotype are shown in Table 1 and genotypic frequencies for the Cypriot population studied are shown in Table 2.

Table 1.

Baseline characteristics of the population in all subjects and according to MMP genotypes.

	MMP GENOTYPES
Baseline characteristics	All subjects (762)	MMP1 1G/2G		MMP3 5A/6A		MMP7 -181A>G		MMP9 R279Q		MMP12 -82A>G
		1G1G	1G2G, 2G2G	6A6A	6A5A, 5A5A	GG	AG, AA	RR	RQ, QQ	AA	AG, GG
Age, years^a	60.5 (10.2)	61.2 (9.8)	60.3 (10.3)	60.2 (10.3)	60.7 (10.1)	60.5 (10.1)	60.5 (10.3)	60.5 (10.2)	60.5 (10.3)	60.2 (10.1)	61.1 (10.5)
BMI, kg/m^2,a	28.04 (4.6)	27.9 (4.8)	28.1 (4.5)	28.1 (4.9)	28.0 (4.4)	29.0 (4.6)	27.8 (4.5)	28.1 (4.5)	28.0 (4.7)	28.0 (4.5)	28.1 (4.6)
Ever smokers	38.55%	38.26%	39.32%	36.15%	40.73%	36.84%	39.62%	35.20%	41.93%	39.63%	37.40%
Current smokers	19.21%	20.13%	19.32%	19.26%	19.61%	18.42%	19.72%	18.88%	19.26%	19.92%	19.11%
Pack-years^b	0 (0; 18.5)	0 (0; 20)	0 (0; 15)	0 (0; 15)	0 (0; 20)	0 (0; 11.9)	0.(0; 20)	0 (0; 12.5)	0 (0; 16.25)	0 (0; 20)	0 (0; 12)
Systolic BP, mmHg^a	138.91 (17.5)	138.4 (17.5)	139.1 (17.0)	139.8 (16.6)	138.3 (17.4)	142.0 (16.7)	138.3 (17.2)	139.3 (17.2)	138.4 (16.9)	138.8 (17.1)	139.5 (17.1)
Diastolic BP, mmHg^a	83.72 (9.68)	83.7 (8.9)	83.9 (9.8)	84.6 (9.3)	83.3 (9.8)	85.4 (9.5)	83.6 (9.7)	83.2 (9.8)	84.5 (9.5)	83.5 (9.8)	84.7 (9.5)
Antihypertensive therapy	37.24%	38.93%	36.92%	37.5%	37.07%	39.47%	36.13%	38.52%	35.98%	36.99%	37.40%
Total cholesterol, mmol/L^a	5.84 (1.06)	5.76 (1.02)	5.86 (1.08)	5.85 (1.02)	5.84 (1.10)	6.01 (1.14)	5.80 (1.05)	5.81 (1.09)	5.90 (1.04)	5.83 (1.07)	5.85 (1.02)
HDL-cholesterol, mmol/L^a	1.30 (0.31)	1.25 (0.31)	1.31 (0.32)	1.30 (0.31)	1.30 (0.32)	1.27 (0.29)	1.31 (0.32)	1.30 90.31)	1.30 (0.33)	1.29 (0.32)	1.31 (0.31)
Cholesterol-lowering therapy	18.55%	23.49%	18.12%	19.26%	18.32%	14.47%	20.07%	18.37%	18.98%	19.51%	18.29%
Triglycerides, mmol/L^b	1.5 (1.09; 2.02)	1.48 (1.1; 2.0)	1.49 (1.1; 2.0)	1.53 (1.14; 2.05)	1.48 (1.08; 2.02)	1.7 (1.27; 2.3)	1.44 (1.06; 1.99)	1.48 (1.09; 1.99)	1.51 (1.08; 2.09)	1.48 (1.09; 2.04)	1.52 (1.08; 1.96)
Type 2 diabetes present	13.29%	13.42%	13.33%	12.84%	13.36%	12.5%	13.09%	13.01%	13.03%	11.59%	16.26%
CVD present	14.91%	14.09%	14.38%	15.99%	14.04%	13.82%	14.34%	14.83%	13.32%	14.05%	14.63%
IMTcc, mm^a	0.74 (0.17)	0.75 (0.18)	0.73 (0.17)	0.73 (0.17)	0.74 (0.17)	0.73 (0.15)	0.74 (0.17)	0.74 (0.17)	0.74 (0.17)	0.74 (0.17)	0.74 (0.16)
Carotid plaques	59.21%	58.39%	60.17%	59.46%	60.13%	59.87%	59.69%	55.61%	63.74%	57.93%	63.01%
Femoral plaques	48.42%	44.97%	49.57%	47.30%	50.0%	39.47%	51.48%	46.94%	49.86%	47.97%	50.41%

BMI, body mass index; BP, blood pressure; HDL, high-density lipoprotein; CVD, cardiovascular disease; IMTcc, intima–media thickness in the common carotid.

Mean and standard deviation (±SD) is used for normally distributed values.

Median and interquartile range (IQR) is used for non-normally distributed values.

Table 2.

Genotypic frequencies in the Cyprus Study population.

SNP	Genotypic frequencies
MMP1 (1G/2G) rs1799750	1G1G 20.3% (149/734)	1G2G 50.14% (368/734)	2G2G 29.56% (217/734)
MMP3 (5A/6A) rs3025058	5A5A 15.75% (120/762)	5A6A 45.41% (346/762)	6A6A 38.85% (296/762)
MMP7 (-181A>G) rs11568818	GG 20.97% (152/725)	AG 49.24% (357/725)	AA 29.79% (216/725)
MMP9 (R279Q) rs17576	RR 52.62% (392/745)	RQ 40.40% (301/745)	QQ 6.98% (52/745)
MMP12 (-82A>G) rs2276109	AA 66.67% (492/738)	AG 28.73% (212/738)	GG 4.61% (34/738)

IMTcc and MMP genotypes

In our study population, there was some weak evidence that the MMP12 -82A>G genotype (AA vs AG, GG) was associated with IMT greater than 0.07 cm (population mean) especially before (Model 1: OR=1.51; 95% CI=1.06 to 2.14; p=0.02) adjusting for additional risk factors (Model 2: OR=1.45; 95% CI=0.99 to 2.10; p=0.05). In a step-wise regression, the best model explaining the variation in IMT > 0.07 cm included age, BMI, sex, smoking in pack-years and the MMP12 -82A>G genotype (R² for the model=0.22). Results are shown in detail in Table 3.

Table 3.

Associations between subclinical atherosclerosis outcomes and MMP genotypes in all subjects: (a) Model 1 and (b) Model 2.

	(a) Model 1 (adjusted for age and sex)
	IMTcc <0.07 mm>		Presence of anyplaques		Presence of carotid plaques		Presence of femoral plaques
	OR (95% CI)	p-value	OR (95% CI)	p-value	OR (95% CI)	p-value	OR (95% CI)	p-value
MMP1 (1G1G vs 1G2G, 2G2G) (149/734 vs 585/734)	1.24 (0.82 to 1.88)	0.30	1.42 (0.90 to 2.21)	0.13	1.21 (0.81 to 1.81)	0.34	1.47 (0.97 to 2.23)	0.07
MMP3 (6A6A vs 6A5A, 5A5A) (296/762 vs 466/762)	1.11 (0.79 to 1.55)	0.56	0.87 (0.60 to 1.26)	0.45	0.97 (0.70 to 1.34)	0.84	1.05 (0.75 to 1.47)	0.78
MMP7 (GG vs AG, AA) (152/725 vs 573/725)	1.01 (0.67 to 1.52)	0.97	1.33 (0.86 to 2.05)	0.21	0.97 (0.65 to 1.44)	0.88	1.68 (1.11 to 2.54)	0.014
MMP9 R279Q (RR vs RQ, QQ) (392/745 vs 353/745)	1.10 (0.79 to 1.53)	0.58	1.67 (1.16 to 2.42)	0.006	1.48 (1.07 to 2.04)	0.017	1.07 (0.77 to 1.50)	0.68
MMP12 (AA vs AG, GG) (492/738 vs 246/738)	1.51 (1.06 to 2.14)	0.023	1.38 (0.93 to 2.04)	0.12	1.21 (0.86 to 1.71)	0.27	1.15 (0.81 to 1.65)	0.43
	(b) Model 2 (adjusted for age, sex, smoking in pack-years, hypertension, diabetes, BMI and TChol)
	IMTcc <0.07 mm>		Presence of anyplaques		Presence of carotid plaques		Presence of femoral plaques
	OR (95% CI)	p-value	OR (95% CI)	p-value	OR (95% CI)	p-value	OR (95% CI)	p-value
MMP1 (1G1G vs 1G2G, 2G2G) (149/734 vs 585/734)	1.21 (0.79 to 1.87)	0.39	1.51 (0.95 to 2.42)	0.08	1.22 (0.81 to 1.85)	0.34	1.63 (1.04 to 2.55)	0.032
MMP3 (6A6A vs 6A5A, 5A5A) (296/762 vs 466/762)	1.17 (0.82 to 1.66)	0.39	0.86 (0.58 to 1.26)	0.44	0.97 (0.69 to 1.35)	0.83	1.05 (0.74 to 1.50)	0.78
MMP7 (GG vs AG, AA) (152/725 vs 573/725)	1.11 (0.72 to 1.71)	0.64	1.51 (0.95 to 2.40)	0.08	1.05 (0.70 to 1.59)	0.80	1.88 (1.20 to 2.94)	0.006
MMP9 R279Q (RR vs RQ, QQ) (392/745 vs 353/745)	1.19 (0.84 to 1.68)	0.33	1.67 (1.14 to 2.46)	0.009	1.50 (1.08 to 2.10)	0.016	1.03 (0.73 to 1.47)	0.85
MMP12 (AA vs AG, GG) (492/738 vs 246/738)	1.45 (0.99 to 2.10)	0.052	1.27 (0.84 to 1.92)	0.26	1.14 (0.80 to 1.63)	0.47	1.05 (0.72 to 1.52)	0.81

IMTcc, intima–media thickness in the common carotid; OR, odds ratio; CI, confidence interval; BMI, body mass index; TChol, total cholesterol.

Plaque presence and MMP genotypes

Presence of plaques (0 versus 1–4 bifurcations with plaque) was strongly associated with the MMP9 R279Q polymorphism (RR versus RQ, QQ) both before (Model 1: OR=1.67; 95% CI=1.16 to 2.42; p=0.006) and after controlling for CVD risk factors (Model 2: OR=1.67; 95% CI=1.14 to 2.46; p=0.009) with no change in the effect estimate, indicating that the association was not significantly confounded by the risk factors in the model. Further analysis showed that the association was driven by the presence of plaques in the carotid (Model 1: OR=1.5; 95% CI=1.07 to 2.04; p=0.017) rather than femoral bifurcations (Model 1: OR=1.07; 95% CI=0.77 to 1.50; p=0.68), and the association persisted even after further adjustment (Model 2: OR=1.5; 95% CI=1.08 to 2.10; p=0.016 for carotids). In a step-wise regression, the best model explaining the variance in carotid plaque presence included age, sex, hypertension, the MMP9 R279Q genotype and diabetes (R² for the model=0.14).

There was a trend between presence of femoral plaques and the MMP7 -181A>G polymorphism (GG vs AA, AG) before (Model 1: OR=1.68; 95% CI=1.11 to 2.54; p=0.014) but especially after further adjustment, with carriers of the A allele having nearly twofold increased odds for the presence of femoral plaques (Model 2: OR=1.88; 95% CI=1.20 to 2.94; p=0.006). Results are shown in detail in Table 3.

Given the differences between men and women in subclinical atherosclerosis shown by our team²⁰ and others,^21,22 a sex-specific analysis was also carried out, which showed that both the association between the MMP9 R279Q genotype and plaque presence in the carotids (Model 2: OR_men=2.85; 95% CI=1.67 to 4.86; p<0.001; p for sex interaction=0.003) and the MMP7 -181A>G genotype and plaque presence in the femoral bifurcations were driven by men (Model 2: OR_men=2.61; 95% CI=1.36 to 4.99; p=0.004; p for sex-interaction<0.001). The trend between the MMP12 -82A>G genotype and IMT greater than 0.07 cm was more evident in women (Model 2: OR_women=1.87; 95% CI=1.10 to 3.18; p=0.02; p for sex interaction=0.33) although it did not reach statistical significance. In women, the MMP12 -82AG and GG genotypes showed a weak association with the presence of femoral plaques after adjustment (Model 2: OR_women=1.90; 95% CI=1.14 to 3.16; p=0.014; p for sex interaction<0.001). Results for men and women separately are shown in detail in Table 4.

Table 4.

Associations between ultrasonic features and MMP genotypes in (a) men and (b) women after adjustment for age, smoking in pack-years, hypertension, diabetes, BMI and TChol.

	(a) MEN
	IMTcc <0.07 mm>		Presence of anyplaques		Presence of carotid plaques		Presence of femoral plaques
	OR (95% CI)	p-value	OR (95% CI)	p-value	OR (95% CI)	p-value	OR (95% CI)	p-value
MMP1 (1G1G vs 1G2G, 2G2G) (74/345 vs 271/345)	1.53 (0.84 to 2.78)	0.17	1.26 (0.55 to 2.92)	0.58	1.16 (0.62 to 2.17)	0.63	1.26 (0.68 to 2.35)	0.47
MMP3 (6A6A vs 6A5A, 5A5A) (133/356 vs 223/356)	1.34 (0.82 to 2.19)	0.25	0.76 (0.37 to 1.57)	0.46	0.97 (0.58 to 1.62)	0.91	1.33 (0.79 to 2.24)	0.28
MMP7 (GG vs AG, AA) (66/345 vs 279/345)	1.41 (0.75 to 2.65)	0.75	1.76 (0.76 to 4.09)	0.19	1.03 (0.54 to 1.98)	0.92	2.61 (1.36 to 4.99)	0.004
MMP9 R279Q (RR vs RQ, QQ) (174/346 vs 172/346)	1.19 (0.73 to 1.94)	0.49	4.50 (2.01 to 10.06)	<0.001	2.85 (1.67 to 4.86)	<0.001	1.30 (0.78 to 2.18)	0.32
MMP12 (AA vs AG, GG) (238/347 vs 109/347)	1.12 (0.66 to 1.91)	0.67	0.96 (0.45 to 2.05)	0.91	1.09 (0.63 to 1.88)	0.76	0.54 (0.31 to 0.93)	0.03
	(b) WOMEN
	IMTcc <0.07 mm>		Presence of anyplaques		Presence of carotid plaques		Presence of femoral plaques
	OR (95% CI)	p-value	OR (95% CI)	p-value	OR (95% CI)	p-value	OR (95% CI)	p-value
MMP1 (1G1G vs 1G2G, 2G2G) (75/389 vs 314/389)	0.96 (0.50 to 1.82)	0.89	1.62 (0.92 to 2.87)	0.10	1.31 (0.75 to 2.29)	0.34	2.06 (1.04 to 4.10)	0.04
MMP3 (6A6A vs 6A5A, 5A5A) (161/403 vs 242/403)	1.00 (0.60 to 1.68)	0.99	0.91 (0.57 to 1.45)	0.70	0.96 (0.62 to 1.51)	0.87	0.89 (0.54 to 1.45)	0.63
MMP7 (GG vs AG, AA) (86/380 vs 294/380)	0.86 (0.47 to 1.57)	0.63	1.46 (0.84 to 2.55)	0.18	1.05 (0.62 to 1.80)	0.86	1.54 (0.83 to 2.86)	0.17
MMP9 R279Q (RR vs RQ, QQ) (218/399 vs 181/399)	1.18 (0.71 to 1.96)	0.53	1.18 (0.75 to 1.88)	0.47	0.99 (0.63 to 1.53)	0.95	0.89 (0.54 to 1.46)	0.64
MMP12 (AA vs AG, GG) (254/391 vs 137/391)	1.87 (1.10 to 3.18)	0.02	1.45 (0.89 to 2.37)	0.14	1.17 (0.73 to 1.87)	0.51	1.90 (1.14 to 3.16)	0.014

Discussion

Several matrix metalloproteinases (MMPs) have been reported to be expressed in atherosclerotic lesions; however, the exact role of each one in the formation and progression of atherosclerosis remains to be fully elucidated. We studied functional genetic polymorphisms in genes encoding for five MMPs (MMP1 (1G/2G), MMP3 (5A/6A), MMP7 (-181A>G), MMP9 (R279Q) and MMP12 (-82A>G)) in relation to subclinical atherosclerosis phenotypes as assessed by ultrasound, in a large (n=762), middle-aged, general population cohort.

We report a strong association between the MMP9 279Q allele and a three- to fourfold increase in risk of having any bifurcation affected by plaque or having a plaque in the carotid bifurcation; this association was mainly driven by men (Model 2: OR_men=4.5; p<0.001 for any plaque and OR_men=2.85; p<0.001 for carotid plaques). Men were also more likely to have a plaque in the femoral arteries if they carried the MMP7 -181A allele (Model 2: OR=2.6; p=0.004 and p for sex interaction<0.001). There was some evidence for association between having a femoral plaque and carrying the MMP12 -82G allele in women (Model 2: OR=1.9; p=0.014 and p for sex-interaction=0.01).

A study by Armstrong et al. looking at six polymorphisms in four MMP genes in a similar-sized community cohort, in relation to IMT, reported an association between the MMP9 R279Q genotype and internal carotid artery bulb IMT; however, the authors suggest this may have been a spurious finding, as no association was seen with common carotid or bulb IMT.²³ In the Framingham study, circulating MMP9 was not associated with IMTcc or IMT in the bulb but was, however, associated with internal carotid stenosis <25%²⁴ (i.e. plaque presence). We also show no association between the MMP9 R279Q and IMTcc (>0.07 cm) but report a strong association between the MMP9 279Q allele and presence of plaques in men; especially presence of carotid plaques. These results provide further support to the notion that IMT and atherosclerotic plaque may represent different pathophysiological mechanisms.^25,26 Our results are also in agreement with previously reported sex differences in atherosclerosis phenotypes (carotid plaque and stenosis)²⁷ as well as with a recently published study reporting a sex-specific effect of another MMP9 functional polymorphism and carotid arterial stiffness.²¹

Both MMP9 and MMP12 degrade components of the basement membrane, thus allowing macrophages to penetrate injured tissues during inflammation, as well as promoting vascular smooth muscle cell migration. Jormsjo and colleagues²⁸ reported an association between the MMP12 -82A>G polymorphism and coronary luminal diameter in diabetic patients, while findings by Orbe et al.²⁹ show higher expression of MMP9 in occlusive lesions from carotid but not femoral arteries and further support differential expression of MMPs in different plaque types and arterial beds. A recently published study also did not find any association between the MMP12 -82G allele and carotid plaque susceptibility in Chinese patients with acute cerebral infarctions;³⁰ however, both the differences in the populations studied and the allele frequencies found (3.8% for the G allele in the Chinese population versus 18.9% in the Cypriot population studied) make these two studies difficult to compare. We show an association between the MMP12 -82G allele and femoral plaques in women (p=0.014) but given that this association was in the opposite direction in men (although not reaching statistical significance after adjustment for multiple testing), this may have been a spurious finding.

MMP7 has been found to be expressed in human atherosclerotic lesions but its exact role has not been defined. One recent study reported an association between the MMP7 -181G allele and presence of vulnerable plaques – but not stable plaques – in the carotids in a Chinese population.³¹ We report for the first time an association between the MMP7 -181A>G polymorphism and presence of femoral plaques in men. Femoral plaques have been previously reported to show a different morphology (more fibrocalcific plaques) compared to carotid plaques (more fibrous cap atheroma)¹⁵ and the pattern of disease progression was shown to be sex-dependent.^13,14

In our study population, no association was seen between the MMP1 2G allele and carotid plaques, in contrast to a recent case–control study in a Serbian population.³² However, the association shown was with severe carotid plaques in patients undergoing endarterectomy, which may not be comparable to our general population subjects, most of whom had non-severely stenotic plaques. We show a weak association in women between the MMP1 2G allele and presence of femoral plaques but this result could be due to chance, as it did not reach statistical significance after adjustment for multiple comparisons.

We did not find any association between the MMP3 5A6A polymorphism and the atherosclerotic phenotypes studied, despite some previous reports for an association between the 6A6A genotype and increased IMT.^33–35 These discrepancies could be explained by the much smaller size of those studies as well as different protocols of measuring IMT at different sites. In a recent meta-analysis of the effect of the MMP3 5A6A polymorphism on coronary disease using data from 15 studies, differential effects were suggested, with studies in subjects of European descent showing no association between myocardial infarction and the MMP3 5A allele, and studies in subjects of Asian descent suggesting a positive association.³⁶

It is worth noting that ~19% of our population is on anti-hyperlipidemic treatment and statins have been found to reduce plasma activity of MMP2 and cell death and increase collagen content in human carotid plaques³⁷ with cardioprotective effects. It is possible that those with plaques and/or prevalent CVD have other risk factors as well and are receiving treatment; however, when repeating the analysis, excluding those on anti-hyperlipidemic treatment, results did not change. Contacting an additional sensitivity analysis by excluding those with prevalent CVD from the analysis resulted in no significant difference in the results either (results not shown).

A strength of our study is the community-based sample and the detailed phenotypic characterization of the subjects with regards to ultrasonically determined atherosclerosis, as well as having data from both carotid and femoral bifurcations. A limitation is the use of baseline, cross-sectional data in the analysis, restricting the possibility of causal inferences from our data and allowing for bias. However, we adjusted our findings for common risk factors for atherosclerosis and conducted relevant sensitivity analyses with no changes in our results. Prospective studies measuring progression of IMT and plaque development are still warranted. An additional limitation is that while we assume that the effect of the MMP7, MMP9 and MMP12 variants on plaque is due to their influence on tissue levels of the respective enzymes, we do not have any direct measure of these in order to investigate this.

Our findings are in agreement and provide further support to previous reports,^24,29 suggesting that the proteolytic mechanisms governed by MMPs may differ between carotid and femoral arteries and between sexes³² and suggesting a complex role of MMPs in plaque formation and development. This is in agreement with the theory of disease susceptibility versus disease-modifying genes, where the former may contribute to the initiation of disease while the latter may influence progression and outcome.³⁸

The MMP7 -181A and the MMP9 279Q alleles were associated with a greater than 2.5 times the risk of having an atherosclerotic plaque in the femoral and carotid arteries (respectively) after adjusting for common risk factors in middle-aged men. Our findings are consistent with site-specific matrix remodeling and support a role for the use of common MMP genotypes in studies of subclinical atherosclerosis.

Footnotes

Appendix

Supplementary Table.

Primers and probes used for genotyping.

SNP	Forward primer and VIC-probe	Reverse primer and FAM-probe
MMP1 (1G/2G) rs1799750	F-ACATGTTATGCCACTTAGATGAGGAAA	R-GCGTCAAGACTGATATCTTACTCATAAACAATA
	VIC-TTGAGATAAGTCATATCCTTTCTA	FAM-TTGAGATAAGTCATATCTTTCTA
MMP3 (5A/6A)	F-TCAATGTGGCCAAATATTTTCCCTGTA	R-ATTCTATGGTTCTCCATTCCTTTGATGG
rs3025058	VIC-ACATGGTTTTTTCCCC	FAM-ACATGGTTTTTCCCC
MMP7 (-181A>G)rs11568818	F-GGAGTCAATTTATGCAGCAGACAGA	R-AGTGTTTTCTTTCTTTTTAGAGTCTACAGAACT
	VIC-CTTTGAAAGACAAATACA	FAM-CTTTGAAAGACGAATACA
MMP9 (R279Q)	F-TCCCCCTTTCCCACATCCT	R-CAGGGTTTCCCATCAGCATTG
rs17576	VIC-CTCTACACCCAGGACGG	FAM-TCTACACCCGGGACGG
MMP12 (-82A>G)rs2276109	F-TGCTTTTGTTTGCATGTTTTTGAGATAGA	R-CCGGGTTCTGTGAATATGAATCCT
	VIC-ATGATATCAACTATGAGTCACT	FAM-ATATCAACTGTGAGTCACT

SNP, single nucleotide polymorphism.

Acknowledgements

The authors express their gratitude to all the participants of the Cyprus Study.

Declaration of conflicting interest

The authors declare that there is no conflict of interest.

Funding

This work was supported by a Cyprus Cardiovascular Disease Educational and Research Trust (CCDERT) and Joint Cyprus Research Promotion Foundation, Ministry of Health and Cyprus Heart Foundation grant (No. 41/5PE) and a Research Promotion Foundation grant (PENEK-05/04). SEH is funded by the British Heart Foundation (RG2008/015).

References

Chambless

Heiss

Folsom

. Association of coronary heart disease incidence with carotid arterial wall thickness and major risk factors: the Atherosclerosis Risk in Communities (ARIC) Study, 1987–1993. Am J Epidemiol 1997; 146: 483–494.

Bots

Hoes

Koudstaal

Hofman

Grobbee

. Common carotid intima-media thickness and risk of stroke and myocardial infarction: the Rotterdam Study. Circulation 1997; 96: 1432–1437.

O’Leary

Polak

Kronmal

. Carotid-artery intima and media thickness as a risk factor for myocardial infarction and stroke in older adults. N Engl J Med 1999; 340: 14–22.

Salonen

. Ultrasonographically assessed carotid morphology and the risk of coronary heart disease. Arterioscler Thromb 1991; 11: 1245–1249.

Ebrahim

Papacosta

Whincup

. Carotid plaque, intima media thickness, cardiovascular risk factors, and prevalent cardiovascular disease in men and women: The British Regional Heart Study. Stroke 1999; 30: 841–850.

Hollander

Bots

Iglesias del Sol

. Carotid plaques increase the risk of stroke and subtypes of cerebral infarction in asymptomatic elderly. The Rotterdam Study. Circulation 2002; 105: 2872–2877.

Kitamura

Iso

Imano

. Carotid intima-media thickness and plaque characteristics as a risk factor for stroke in Japanese elderly men. Stroke 2004; 35: 2788–2794.

Kablak-Ziembicka

Tracz

Przewlocki

Pieniazek

Sokolowski

Konieczynska

. Association of increased carotid intima-media thickness with the extent of coronary artery disease. Heart 2004; 90: 1286–1290.

Inaba

Chen

Bergmann

. Carotid plaque, compared with carotid intima-media thickness, more accurately predicts coronary artery disease events: A meta-analysis. Atherosclerosis 2012; 220: 128–133.

10.

Sillesen

Fuster

. Predicting coronary heart disease: From Framingham Risk Score to ultrasound bioimaging. Mt Sinai J Med 2012; 79: 654–663.

11.

Griffin

Nicolaides

Tyllis

. Carotid and femoral arterial wall changes and the prevalence of clinical cardiovascular disease. Vasc Med 2009; 14: 227–232.

12.

Khoury

Schwartz

Gottlieb

Chenzbraun

Stern

Keren

. Relation of coronary artery disease to atherosclerotic disease in the aorta, carotid, and femoral arteries evaluated by ultrasound. Am J Cardiol 1997; 80: 1429–1433.

13.

Dalager

Falk

Kristensen

Paaske

. Plaque in superficial femoral arteries indicates generalized atherosclerosis and vulnerability to coronary death: An autopsy study. J Vasc Surg 2008; 47: 296–302.

14.

Dalager

Paaske

Bayer Kristensen

Marsvin Laurberg

Falk

. Artery-related differences in atherosclerosis expression: implications for atherogenesis and dynamics in intima-media thickness. Stroke 2007; 38: 2698–2705.

15.

Herisson

Heymann

M-F

Chétiveaux

. Carotid and femoral atherosclerotic plaques show different morphology. Atherosclerosis 2011; 216: 348–354.

16.

Galis

Khatri

. Matrix metalloproteinases in vascular remodeling and atherogenesis: the good, the bad, and the ugly. Circ Res 2002; 90: 251–262.

17.

Pollanen

Lehtimaki

Mikkelsson

. Matrix metalloproteinase 3 and 9 gene promoter polymorphisms: joint action of two loci as a risk factor for coronary artery complicated plaques. Atherosclerosis 2005; 180: 73–78.

18.

Humphries

Luong

L-A

Talmud

. The 5A/6A polymorphism in the promoter of the stromelysin-1 (MMP-3) gene predicts progression of angiographically determined coronary artery disease in men in the LOCAT gemfibrozil study. Atherosclerosis 1998; 139: 49–56.

19.

Panayiotou

Nicolaides

Griffin

. Serum total homocysteine, folate, 5,10-methylenetetrahydrofolate reductase (MTHFR) 677C→T genotype and subclinical atherosclerosis. Expert Opin Ther Targets 2009; 13: 1–11.

20.

Panayiotou

Nicolaides

Griffin

. Leukocyte telomere length is associated with measures of subclinical atherosclerosis. Atherosclerosis 2010; 211: 176–181.

21.

Lin

R-T

Chen

C-H

Tsai

P-C

B-L

Hank Juo

S-H

Lin

H-F

. Sex-specific effect of matrix metalloproteinase-9 functional promoter polymorphism on carotid artery stiffness. Atherosclerosis 2012; 223: 416–420.

22.

Halvorsen

Johnsen

Mathiesen

Njolstad

. The association between inflammatory markers and carotid atherosclerosis is sex dependent: the Tromso Study. Cerebrovasc Dis 2009; 27: 392–397.

23.

Armstrong

Abilleira

Sitzer

Markus

Bevan

. Polymorphisms in MMP family and TIMP genes and carotid artery intima-media thickness. Stroke 2007; 38: 2895–2899.

24.

Romero

Vasan

Beiser

. Association of carotid artery atherosclerosis with circulating biomarkers of extracellular matrix remodeling: The Framingham Offspring Study. J Stroke Cerebrovasc Dis 2008; 17: 412–417.

25.

Spence

. Measurement of intima-media thickness vs. carotid plaque: uses in patient care, genetic research and evaluation of new therapies. Int J Stroke 2006; 1: 216–221.

26.

Spence

Hegele

. Noninvasive phenotypes of atherosclerosis: similar windows but different views. Stroke 2004; 35: 649–653.

27.

Iemolo

Martiniuk

Steinman

Spence

. Sex differences in carotid plaque and stenosis. Stroke 2004; 35: 477–481.

28.

Jormsjo

Moritz

. Allele-specific regulation of matrix metalloproteinase-12 gene activity is associated with coronary artery luminal dimensions in diabetic patients with manifest coronary artery disease. Circ Res 2000; 86: 998–1003.

29.

Orbe

Fernandez

Rodrνguez

. Different expression of MMPs/TIMP-1 in human atherosclerotic lesions. Relation to plaque features and vascular bed. Atherosclerosis 2003; 170: 269–276.

30.

Jin

Zhou

. Lack of independent relationship between the MMP-12 gene polymorphism and carotid plaque susceptibility in the Chinese Han population. Vasc Med 2012; 17: 310–316.

31.

X-F

Jin

X-P

P-Y

. Association of a functional polymorphism in the MMP7 gene promoter with susceptibility to vulnerable carotid plaque in a Han Chinese Population. Clin Chem Lab Med 2011; 49: 1735–1741.

32.

Djurić

Stanković

Končar

. Association of MMP-8 promoter gene polymorphisms with carotid atherosclerosis: preliminary study. Atherosclerosis 2011; 219: 673–678.

33.

Gnasso

Motti

Irace

. Genetic variation in human stromelysin gene promoter and common carotid geometry in healthy male subjects. Arterioscler Thromb Vasc Biol 2000; 20: 1600–1605.

34.

Rauramaa

Vaisanen

Luong

L-A

. Stromelysin-1 and interleukin-6 gene promoter polymorphisms are determinants of asymptomatic carotid artery atherosclerosis. Arterioscler Thromb Vasc Biol 2000; 20: 2657–2662.

35.

Rundek

Elkind

Pittman

. 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.

36.

Koch

de Waha

Hoppmann

Schömig

Kastrati

. Haplotypes and 5A/6A polymorphism of the matrix metalloproteinase-3 gene in coronary disease: case–control study and a meta-analysis. Atherosclerosis 2010; 208: 171–176.

37.

Crisby

Nordin-Fredriksson

Shah

Yano

Zhu

Nilsson

. Pravastatin treatment increases collagen content and decreases lipid content, inflammation, metalloproteinases, and cell death in human carotid plaques: implications for plaque stabilization. Circulation 2001; 103: 926–933.

38.

Gale

Martyn

. Variation in the matrix metalloproteinase-1 gene and risk of coronary heart disease. Eur Heart J 2003; 24: 1668–1671.

Association of genotypes at the matrix metalloproteinase (MMP) loci with carotid IMT and presence of carotid and femoral atherosclerotic plaques

Abstract

Keywords

Introduction

Methods

Subjects

Cardiovascular disease status, risk factors and ultrasonic measurements

Biochemical measurements

Statistical methods

Results

IMTcc and MMP genotypes

Plaque presence and MMP genotypes

Discussion

Footnotes

Appendix

Acknowledgements

Declaration of conflicting interest

Funding

References