Potential Role of Selected Biomarkers for Predicting the Presence and Extent of Coronary Artery Disease

Key words: Coronary artery disease/Macrophage chemoattractant protein-1/Matrix-metalloproteinase-3/Soluble CD40 ligand/Soluble tumour necrosis factor receptor-2/SYNTAX score Introduction. Since atherosclerosis may in part be an inflammatory disease, circulatory factors related to inflammation may be predictors of coronary artery disease. The aim of this study was to evaluate the association between the level of some circulating biomarkers and the extent of coronary artery disease. Methods. Blood samples were taken from 128 patients with stable forms of coronary heart disease. Macrophage chemoattractant protein-1 (MCP-1), matrix-metalloproteinase-3 (MMP-3), soluble CD40 ligand (sCD40L) and soluble tumour necrosis factor receptor-2 (sTNFR2) were measured by ELISA. Coronary angiography and grading with the SYNTAX score followed. Results. There was no significant interdependence of circulating MCP-1, sCD40L, sTNFR2 levels and SYNTAX score. MMP-3 levels were significantly different in subgroup with coronary artery disease (SYNTAX score > 0): 38.1 μg/l (13.6; 84.1) and subgroup without coronary artery disease (SYNTAX score = 0): 20.4 μg/l (13.1; 82.8), p=0.001. According to the Spearman correlation coefficient there was significant association between MMP-3 level and SYNTAX score (0.358, a=0.05). Conclusions. Our data suggest association between the extent of coronary artery disease and circulating MMP-3. We failed to demonstrate any association with the other investigated biomarkers.


INTRODUCTION
Atherosclerosis is the main cause of myocardial ischemia.In 2008, 17.9% of all deaths in the Czech Republic were due to chronic ischaemic heart disease 1 .These patients risk acute coronary syndrome, progressive heart failure and sudden cardiac death.The most important prognostic value for patients with stable forms of coronary heart disease is left ventricular function, the complexity of the coronary artery disease and serious myocardial ischemia.Coronary angiography remains elementary for the evaluation of coronary disease.However, more recently coronary CT angiography has become more important in that process 2 .For determination of the complexity of coronary artery disease there has been developed an angiographic grading tool -the SYNTAX score.This helps to identify the potential risk for revascularization and aids choosing the optimal revascularisation strategy 3 .
Since atherosclerosis may in part be an inflammatory disease, circulatory factors related to inflammation could be predictors of coronary artery disease.The aim of this study was to evaluate the association between the level of some circulating biomarkers and the extent of coronary artery disease.

MATERIALS AND METHODS
A cohort of 128 subjects was recruited by random selection of patients with stable forms of coronary heart disease (stable angina pectoris, silent ischemia, unclear chest pain) who were sent to our cardiology department.Exclusion criteria, were acute coronary syndrome, renal insufficiency, systemic autoimmune disease, chronic or acute inflammation.The clinical characteristics of pa-ABBREVIATIONS MCP-1 -Monocyte chemoattractant protein-1 MMPs -Matrix-metalloproteinases MMP-3 -Matrix-metalloproteinase-3 TIMPs -Tissue inhibitors of metalloproteinases sCD40L -Soluble CD40 ligand sTNFR2 -Soluble tumour necrosis factor receptor-2 CAD -Coronary artery disease CHD -Coronary heart disease ELISA -Enzyme-linked immunosorbent assay tients are summarized in Table 1, updated treatment in Table 2.
Blood samples were drawn into plastic tubes without anticoagulant and immediately spun at 1,500×g for 10 min (centrifuge MPW 350, Med.Instruments, Spoldzielnia Pracy, Warszawa, Poland) at room temperature.Serum was immediately isolated and stored at −20°C until analysed.Specimens were not repeatedly defrosted before analysis.Serum concentrations of MCP-1, MMP-3, sCD40L and sTNFR2 were measured concurrently by ELISA standard CE method (Bender MedSystems, eBioscience Corporation, San Diego, California, USA) according to the recent manufacturer recommendations.The minimal sensitivity and precision of the method are 2.31 ng/l, variation coefficient (CV) = 6.2% for MCP-1; 0.06 μg/l, CV=5.4% for MMP-3; 0.062 μg/l, CV= 10.7% for sCD40L; 0.053 μg/l, CV=12.9% for sTNFR2 in accordance with the precision data provided.The absorption at 450 nm was used for photometric assay (spectrophotometer Spectra Shell, SLT Labinstrument, Austria).
For HbA1c assessment,blood samples were drawn into plastic tubes containing tripotassium EDTA, blood was processed on the day of withdrawal.For standard biochemical assessments the serum was processed on the day of blood withdrawal.The standard technique of the local laboratory was used.
Coronary angiography followed.The findings were evaluated by two independent experienced interventional cardiologists unfamiliar with the biochemical analysis results.No quantitative coronary analysis tool nor intravascular imaging method were used.SYNTAX score was assessed using original freeware.
Standard descriptive statistics were adopted for the analysis.Continuous variables were described by means of medians and 5-95 th percentile; categorical variables were described using percentages in categories.The statistical significance of differences in continuous variables between groups was analysed using the Mann-Whitney U test; Fisher exact test was applied for analysis of the relationship between binary variables.Statistical dependence between two variables was assessed by Spearman's rank correlation coefficient and means of ML Chi-square test.α=0.05 was adopted as the chosen level of statistical significance in all tests.

RESULTS
Circulating MCP-1 and MMP-3 levels significantly differed between subgroups with and without diabetes mellitus (Table 1).Gender variability of biomarker levels was investigated.MCP-1, MMP-3 and sTNFR2 significantly differed between male and female subgroups (Table 3).
The entire patient group was divided into two subgroups: subjects with coronary artery disease (SYNTAX score > 0) and subjects without significant coronary artery stenosis (SYNTAX score = 0).Biomarker levels in these subgroups were confronted.Only MMP-3 levels significantly differed (Table 4, Table 5).According to the Spearman correlation coefficient, there was a significant association between MMP-3 levels and SYNTAX score for the entire patient group (0.358, a=0.05) as well as for diabetic (0.378, a=0.05) and non-diabetic (0.335, a=0.05) subgroups.There was no significant correlation between SYNTAX score and the other investigated biomarkers.

DISCUSSION
The SYNTAX score is an angiographic grading tool to determine the functional complexity of coronary artery disease.It was derived from preexisting classifications and was published in 2005 (ref. 4).This algorithm was developed for purposes of SYNTAX trial (Synergy Between Percutaneous Coronary Intervention With TAXUS and Cardiac Surgery) and provides the first evidence-based approach to employing optimal revascularization strategies for patients with three-vessel and/or left main coronary artery disease.In principle, the SYNTAX score is the sum of the points assigned to each individual lesion identified in the coronary tree with >50% diameter narrowing in vessels >1.5 mm diameter.The percent diameter stenosis is not a consideration, only the presence of a stenosis from 50-99% diameter, <50% diameter narrowing or total occlusion.
The latest publications are indicative of new applications of this algorithm: a SYNTAX score of >34 also identifies a subgroup with a particularly high risk of cardiac death, independent of age, gender, acute coronary syndrome, ejection fraction, EUROscore and degree of revascularization 5 .Higher SYNTAX scores are predictive of peri-procedural myocardial injury during percutaneous coronary intervention 6 .In this study the SYNTAX score was used for description of coronary artery impairment enabling statistical processing.
Coronary artery disease is widely recognized as an inflammatory disease.Monocyte chemoattractant protein-1 (MCP-1) -alternative name chemokine (C-C motif) ligand 2 (CCL2), is a chemokine involved in the pathogenesis of atherosclerosis.By its chemotactic activity, it causes diapedesis of monocytes from the lumen to the subendothelial space where they become foam cells, initiating fatty streak formation that leads to atherosclerotic plaque formation.Inflammatory macrophages probably play a role in plaque rupture and the resulting ischaemic episodes as well as restenosis after angioplasty.There is strong evidence that MCP-1 plays a major role in myocarditis, ischaemia/reperfusion injury to heart and in transplant rejection 7 .During the early phase of acute myocardial infarction, the plasma levels of MCP-1 were significantly higher in patients with well-developed collateral circulation than patients with absent collateral circulation.These findings suggest that the shear stress-induced overexpression of MCP-1 contributes significantly to the development of coronary collaterals during the early phase of acute myocardial infarction 8 .In middle-aged overweight/ obese individuals MCP-1 was independently associated with CVD mortality 9 .Even though a promising biomar-Potential role of selected biomarkers for predicting the presence and extent of coronary artery disease 1 Continuous variables were described using median and 5-95 th percentile; categorical variables are described using percentage of categories 2 Statistical significance of differences in continuous variables between groups of patients was analysed by means of Mann -Whitney U test; statistical significance of relationship between binary variables was analysed using Fisher exact test; α=0.05 was adopted as a level of significance for all tests and statistically significant results are given in bold and denoted by * ker of atherosclerosis, we have not succeeded in confirming an association between circulating MCP-1 level and SYNTAX score.
It has become clear that the extracellular matrix is not a mere scaffold for cells but that it also has cryptic biological functions that can be revealed on proteolysis.Matrix metalloproteinases (MMPs) are a family of 24 zinc-dependent endopeptidases hydrolyzing components of the extracellular matrix 10 .Tissue inhibitors of metalloproteinases (TIMPs) are their specific inhibitors that participate in controlling the local activities of MMPs in tissues 11 .A loss of activity control may result in diseases such as arthritis, cancer, nephritis, tissue ulcers, and fibrosis 12 .The pathological effects of MMPs and TIMPs in cardiovascular disease involve vascular remodeling, atherosclerotic plaque instability 13 , and left ventricular remodeling after myocardial infarction 14,15 .Some studies with matrix-metalloproteinase-3 (stromelysin-1) show its prospective value.It has been demonstrated that plasma fluctuation in MMP-3 levels can be used as a supplementary independent predictor of cardiovascular events in patients with stable coronary artery disease 16 .MMP-3 is associated with left ventricular dysfunction, adverse left ventricular remodelling and prognosis after AMI 17 .MMP-  1 Categorical variables are described using percentage of categories 2 Statistical significance of relationship between binary variables was analysed using Fisher exact test; α=0.05 was adopted as a level of significance for all tests and statistically significant results are given in bold and denoted by * 3 together with B-type natriuretic peptide are significant independent predictors of cardiac events in patients with dilated cardiomyopathy.Patients with elevations of both were found to have the poorest prognosis 18 .The increased level of MMP-3 may be responsible for coronary artery ectasia formation 19 .Polymorphisms of MMP-3 and TIMP-4 genes affect angiographic coronary plaque progression in non-diabetic and type 2 diabetic patients 20 .
Our findings show that circulating MMP-3 levels correlate with coronary artery disease presence (Table 5) and with SYNTAX score.Assuming that the SYNTAX score has the ability to describe anatomical and functional features of coronary artery disease, we suggest that MMP-3 levels can predict the extent of coronary artery impairment.
CD40 ligand (CD40L) and soluble CD40 ligand (sCD40L) belong to the tumor necrosis factor superfamily, and they are molecules with a dual prothrombotic and proinflammatory role.They are expressed in a variety of tissues such as the immune system (in both B and T cells), the vascular wall, and activated platelets.Soluble CD40L has multiple autocrine, paracrine, and endocrine actions, and it may trigger the key mechanisms of atherothrombosis.CD40/CD40L may participate in the development of coronary atherosclerosis and triggering acute coronary syndromes while sCD40L seems to have a prognostic role not only in people with advanced atherosclerosis but also in the general population 21 .It is hypothesized that sCD40L induces proliferation and migration of vascular smooth muscle cells through activation of matrix metalloproteinases 22 .However the outcomes of studies with CD40L/sCD40L are inconsistent.sCD40L levels are elevated in hypertensive patients and might discriminate hypertensive patients at high risk of cardiovascular events as published by Ferroni et al. 23 .In contrast, another finding is that in newly diagnosed hypertensive men, sCD40L levels are inversely related to insulin sensitivity, with no correlation with BP, other cardiovascular risk factors or degree of subclinical atherosclerosis 24 .In a TACTICS-TIMI 18 trial, sCD40L was not associated with risk in non-ST elevation acute coronary syndrome population treated with a platelet GPIIb/IIIa receptor antagonist 25 .It is known that atorvastatin diminishes sCD40L plasma levels, more markedly in subjects with the metabolic syndrome 26 .
In general, preanalytic conditions are critical in the assessment of sCD40L concentrations.Some reports document the effects of specimen type (serum and Potential role of selected biomarkers for predicting the presence and extent of coronary artery disease MMP-3, matrix-metalloproteinase-3; CAD, coronary artery disease 1 Statistical significance was analysed by means of ML Chi-square test; α=0.05 was adopted as a level of significance for all tests and statistically significant results are given in bold and denoted plasma), processing (time and temperature), and commercial reagent selection on sCD40L ELISA results 27,28 .High-sensitivity ELISA assay suitable for plasma and serum testing (Bender MedSystems) was used.In accordance with the recent manufacturer recommendations,we failed to confirm the association between sCD40L and SYNTAX score.The results can in part be affected by specimen type: in our study blood serum was tested.As for Halldorsdottir et al. the optimum strategy would be to measure sCD40L in platelet-free plasma 28 .Variability in statin administration could play its role as well (Table 2).Tumor necrosis factor-α (TNF-α) is a pleiotropic cytokine involved in a broad spectrum of inflammatory and immune responses including proliferation, differentiation and cell death induction in several cell types.The biological effects of TNF-α are mediated via the cell-surface TNF-α receptors, TNFR1 and TNFR2.Upon activation, these two receptors are proteolytically cleaved from the membrane and shed into the circulation, where they can be measured as soluble forms (sTNFR1, sTNFR2).The level of sTNFR2 in the circulation could therefore be interpreted as the degree of activation of TNF-α signalling in the tissues, making sTNFR2 an interesting parameter for TNF-α activation in the tissues.High levels of sTNFR2 in serum have been documented in multiple inflammatory pathologies 29,30 .Metabolic syndrome and diabetes are also associated with higher plasma concentrations of TNFalpha and its receptors 31 .
Our findings do not show any correlation between sTNFR2 level and coronary artery disease described by SYNTAX score.This result in high probability is not related to laboratory assay.Preanalytic conditions were kept precisely according to the manufacturer recommendations.The explanation can theoretically be in the patient selection -all individuals according to anamnesis, objective examination and laboratory methods had stable atherosclerotis plaques.Presence of unstable plaques could be an important cause of circulating sTNFR2 level elevation.However contrary to this theory is the finding of Mizia-Stec et al. 32 .According to their study in patients with stable coronary artery disease, interventional procedures result in suppression of sTNFR2 32 .Other studies for clarifying of sTNFR2 consequences are needed.
According to our findings the concentrations of circulating MCP-1 and MMP-3 significantly differ between men and women.The difference is also seen in sTNFR2 and sCD40L levels, however the statistical significance is not so strong (Table 3).Several physiological conditions might explain this variability.It is assumed that markers of inflammation strongly correlate with measures of adiposity, and this association is generally stronger in women than in men 33,34 .The second potentially important aspect may involve differences in endogenous sex hormone levels or sex hormone-binding globulin 35,36 .
The main limitation of this study seems to be the coronary lesion significance assessment.The SYNTAX score considers over 50% coronary artery luminal reduction.The generally used estimation of luminal reduction is not exact and can be incorrect in spite of investigator experience.On the other hand, quantitative coronary analysis of the whole coronary bloodstream would have been exceedingly time consuming.Moreover, in the case of diffusely diseased arteries, it can be difficult to find the reference diameter.An intravascular imaging technique could enable precise stenosis quantification.The functional status could be described by fraction flow reserve determination.Using these methods any scoring system can become exact but useless for day-to-day routine.
Another limitation can be spuning of blood samples at room temperature.Using of cooled centrifuge is regarded as the best way of manipulation with sample in preanalytic stage.However this practice was not recommended in the diagnostic set manufacturer recommendations.

CONCLUSIONS
Our results suggest that in patients with stable coronary artery disease, MMP-3 level correlates with the extent of coronary artery disease.No association between the other investigated biomarkers and grade of coronary artery disease was found.

Table 3 .
Gender variability of levels of selected biomarkers.Continuous variables were described using median and 5-95 th percentile 2 Statistical significance of differences in continuous variables between groups of patients was analysed by means of Mann -Whitney U test; α=0.05 was adopted as a level of significance for all tests and statistically significant results are given in bold and denoted by * 1

Table 4 .
Coronary artery disease presence and level of selected biomarkers.

Table 5 .
Coronary artery disease presence and according to level of MMP-3.