Genes expressed in coronary thrombi are associated with ischemic time in patients with acute myocardial infarction

Thrombosis Research 135 (2015) 329–333

Contents lists available at ScienceDirect

Thrombosis Research journal homepage: www.elsevier.com/locate/thromres

Regular Article

Genes expressed in coronary thrombi are associated with ischemic time in patients with acute myocardial infarction R. Helseth a,b,⁎, I. Seljeflot a,b, T. Opstad a,b, S. Solheim a, M. Freynhofer c, H. Arnesen a,b, K. Huber c, T. Weiss c a b c

Center for Clinical Heart Research, Department of Cardiology, Oslo University Hospital Ullevål, Norway Faculty of Medicine, University of Oslo, Norway Department of Cardiology and Intensive Care, Wilhelminenhospital, Vienna, Austria

a r t i c l e

i n f o

Article history: Received 16 September 2014 Received in revised form 29 October 2014 Accepted 30 November 2014 Available online 4 December 2014 Keywords: Myocardial infarction Thrombectomy Gene expression Inflammation Thrombosis

a b s t r a c t Introduction: Reports on the content of aspirated coronary thrombi have until now mainly focused on cellular components. We investigated the genetic expression of selected mediators and proteases actively involved in the pathophysiological process of acute myocardial infarction in aspirated coronary thrombi. Materials and Methods: In this cross-sectional study, RNA from coronary thrombi in 67 subjects with acute myocardial infarction was isolated. Gene expression arrays of selected markers were performed by RT-PCR with relative quantification. Results: Twenty of 22 markers were expressed in N 50% of the samples. The relative quantification of P-selectin correlated negatively to total ischemic time (p = 0.01), while genes related to fibrinolysis (t-PA, u-PA, PAI-1), inflammation (PTX3, CXCL9, MCP-1, IL18, TNFα) and plaque instability (MMP-2 and TIMP-1) correlated positively to total ischemic time (all b 0.05). Long ischemic time (N4.0 hours) associated with a relative reduction in the expression of P-selectin and a relative increase in the expression of t-PA, u-PA, PAI-1, PTX3, CXCL9, MCP-1, IL-18, TNFα, MMP-2 and TIMP-1. The presence of type 2 diabetes associated with 3.2-fold increased PAI-1 expression (adjusted p = 0.033), while the presence of hypertension associated with about 50% reduction of IL-8 and TIMP-1. Smoking and overweight did not affect any markers. Conclusions: The gene expression profile from coronary thrombi differed according to ischemic time, shown by reduced content of platelet markers and increased content of fibrinolytic, inflammatory and plaque instability mediators over time. Patients with type 2 diabetes showed increased expression of PAI-1, indicative of reduced fibrinolysis. © 2014 Elsevier Ltd. All rights reserved.

Introduction Even in times of great medical advances, coronary artery disease (CAD) still remains a major cause of global morbidity [1]. Acute myocardial infarction (AMI) occurs due to rupture or erosion of atherosclerotic coronary plaques, subsequent luminal narrowing and thrombosis, ultimately leading to myocardial ischemia and eventually necrosis. In AMI, with or without ST-elevation in the electrocardiogram (STEMI and NSTEMI), percutaneous coronary intervention (PCI) is the method of choice in order to revascularize the infarct-related artery and to restore coronary blood flow. The importance of greater knowledge about the atherosclerotic process and the subsequent coronary thombus formation is obvious [2]. The introduction of thrombus aspiration before dilatation and stent implantation of the culprit lesion has shown controversial results with ⁎ Corresponding author at: Center for Clinical Heart Research, Department of Cardiology, Oslo University Hospital Ullevål, Kirkeveien 166, PB 4956 Nydalen, N-0424 Oslo, Norway. Tel.: +47 97 14 95 18; fax: +47 2211 9181. E-mail address: [email protected] (R. Helseth).

http://dx.doi.org/10.1016/j.thromres.2014.11.028 0049-3848/© 2014 Elsevier Ltd. All rights reserved.

respect to reducing the risk of re-infarction and death [3–5]. A recent Swedish multicenter randomized controlled trial could not find any benefit of thrombus aspiration on 30-days mortality compared with PCI alone [6]. Thrombus aspiration per se, however, allows more thorough investigation of the occlusive coronary thrombus itself. Existing reports regarding aspirated coronary thrombi have mainly focused on structural and cellular components. Both platelets, fibrin, red blood cells and inflammatory cells seem to be part of the coronary thrombus content [7–11]. It has also been suggested that primitive stem cells are involved [12]. Three different types of coronary thrombi have further been suggested; platelet-rich and low-erythrocyte, mixed platelet and erythrocyte, and erythrocyte-rich, low-platelet thrombi [13]. Whether the cellular composition of the thrombus differs according to the ischemic time or other variables is unclear, although the platelet content is suggested to correlate inversely and the fibrin content positively with time from symptom onset to PCI [7,8,12]. Activated inflammatory cells like neutrophils, macrophages and eosinophils are suggested to contribute to thrombus size [14,15]. The association between coronary thrombus content and clinical outcome is unclear, although there seems to be a link between inflammatory cell density in the aspirated thrombus

330

R. Helseth et al. / Thrombosis Research 135 (2015) 329–333

and impaired myocardial perfusion and lower left ventricular ejection fraction (LVEF) [16]. To the best of our knowledge, there are no reports on gene expression in aspirated coronary thrombi of patients with AMI. Whether haemostatic and inflammatory mediators are genetically expressed at all are unknown. The establishment of a possible genetic profile may theoretically pave the way for more targeted therapy in AMI. The aim of the present study was therefore to investigate coronary thrombi aspirated from patients with AMI with respect to a selection of mediators previously related to plaque rupture, platelet and neutrophil cell activation, coagulation, fibrinolysis and inflammation. We wanted to describe the genetic expression of such selected mediators and explore whether they were associated with total ischemic time and with cardiovascular risk factors like type 2 diabetes mellitus (T2DM), hypertension, smoking and obesity. Materials and Methods

and 280 nm) of the nano drop evaluation was ≈ 2. Gene arrays of the selected markers were performed by use of custom designed TaqMan RT-PCR Arrays and the Viia ™ 7 Real Time PCR System (Applied Biosystems, Foster City, CA, USA). Expression of the selected genes were quantified in relation to a reference sample and normalized to β-2-microglobulin [17]. A relative quantification (RQ) value of 1.0 means that the gene was equally expressed as the reference sample. The selected markers in the array were CD40 ligand (CD40L), protease activated receptor 1 (PAR-1), P-selectin (P-sel), tissue factor (TF), tissue factor pathway inhibitor (TFPI), tissue plasminogen activator (t-PA), urokinase plasminogen activator (u-PA), plasminogen activator inhibitor 1 (PAI-1), myeloperoxidase (MPO), pentraxin 3 (PTX3), C-reactive protein (CRP), chemokine (C-X-C motif) ligand 9 (CXCL9), fractalkine, monocyte chemoattractant protein 1 (MCP-1), interleukin 8 (IL-8), interleukin 1 beta (IL-1β), interleukin 18 (IL-18), tumor necrosis factor alfa (TNFα), interleukin 12 (IL-12), matrix metalloproteinase 2 (MMP-2), matrix metalloproteinase 9 (MMP-9) and tissue inhibitor of metalloproteinases 1 (TIMP-1).

Study Design Statistical Analyses In this descriptive cross-sectional study, subjects with AMI undergoing percutaneous coronary intervention (PCI) were included at Department of Cardiology at Wilhelminenhospital, Vienna, Austria, during the time period November 2007 – October 2010. All patients were participants in the "Wilhelminenspital Stent Registry" (WSP-STENT) and all patients gave their informed consent. Coronary thrombi from 84 patients were aspirated, transferred into kryo-vials and immediately stored at -80 °C. Relevant clinical data was recorded. Ischemic time was defined as time from symptom start to PCI. Hypertension was defined as systolic and/or diastolic blood pressure (BP) N 140/90 mmHg at admission or on basis of a history of hypertension or current use of antihypertensive medication. Diabetes mellitus was defined according to guidelines and if there was a history of diabetes mellitus or the patients used insulin or oral antidiabetics. Smoking habits were defined as current smokers or not, while obesity was defined as body mass index (BMI) above median (27.8 kg/m2).

As most variables, including the RQ values of the genes expressed, were skewly distributed, non-parametric statistical tests were used throughout. Spearman´s rho was used for correlation analyses and Mann-Whitney test for two independent samples when comparing levels of gene expression between groups. A multivariable linear regression model was used for the association between PAI-1 expression and T2DM. Skewed variables were log transformed before entered into the model. P-values of ≤ 0.05 were considered statistically significant. All statistical analyses were performed in IBM SPSS Statistics, Version 21. Results Of the initially aspirated thrombi in 84 patients, the amount and quality were acceptable in samples from 67 patients, thus the presented results are from this cohort.

Laboratory Analyses

Patients Characteristics

RNA from the aspirated coronary thombi were isolated by use of High Pure RNA Tissue Kit (Roche Diagnostics GmbH, Mannheim, Germany), stabilized by lysing buffer and homogenized by use of termomixer (Termomixer Eppendorf, Eppendorf AG, Hamburg, Germany). The quality and quantity were evaluated by use of NanoDrop 1000 (Saveen Werner, Stockholm, Sweden). Quality of a sample was deemed acceptable for further analysis if the quality ratio (the ratio of the absorbance at 260

Baseline characteristics are given in Table 1. The patients had a median age of 56.4 years, 75% were men and median BMI was 27.8 kg/m2. The frequency of hypertension was 77.6 %, about 50 % were current smokers and 21 % had T2DM. The median ischemic time from symptom onset to thrombus aspiration and PCI was 4.0 hours (range 0 – 120 hours) and there were no significant differences in patient characteristics according to low and high ischemic time.

Table 1 Baseline characteristics of the total study cohort (n = 67) and according to total ischemic time. Values are given in number (proportions) or median (25, 75 percentiles) if not otherwise stated.

Female gendera Age (yrs) Family history of CAD Previous AMI Hypertension Current smokers T2DM Heart failure BMI (kg/m2) Total cholesterol (mmol/L) Triglycerides (mmol/L) HDL-cholesterol (mmol/L) LDL-cholesterol (mmol/L) Total ischemic time (h)

Total study polulation

Total ischemic time ≤ 4 h

Total ischemic time N4 h

17 (25 %) 56.4 (46.5, 63.9) 21 (31.3 %) 7 (10.4 %) 52 (77.6 %) 33 (49.3 %) 14 (20.9 %) 3 (4.5 %) 27.8 (24.8, 30.9) 4.9 (3.9, 6.0) 1.5 (1.2, 2.4) 1.0 (0.8, 1.2) 2.8 (2.1, 3.9) 4.0 (0 - 120)b

8 (22.2 %) 55.7 (46.0, 64.0)

8 (29.6 %) 58.7 (51.6, 63.2)

29 (80.6 %) 17 (47.2 %) 6 (16.7 %) 27.2 (23.6, 30.9) 5.1 (3.9, 6.0) 1.5 (1.0, 2.3) 1.1 (0.9, 1.2) 3.0 (2.1, 4.0) 3.0 (2.0, 3.0)

19 (70.4 %) 14 (51.9 %) 7 (25.9 %) 29.4 (26.5, 32.5) 4.7 (3.8, 5.8) 1.6 (2.1, 3.1) 1.0 (0.7, 1.1) 2.5 (2.2, 3.6) 15.0 (5.0, 24.0)

p 0.504 0.739

0.348 0.716 0.369 0.071 0.467 0.380 0.159 0.394 b0.001

CAD: Coronary Artery Disease, AMI: Acute myocardial infarction T2DM: Type 2 diabetes mellitus. BMI: Body mass index. HDL-cholesterol: High density lipoprotein-cholesterol. LDL-cholesterol: Low density lipoprotein cholesterol. P-values refer to differences according to total ischemic time. a Total ischemic time was not available in one female b Median and range.

R. Helseth et al. / Thrombosis Research 135 (2015) 329–333 Table 2 Gene expression in the coronary thrombi (n = 67). Selected markers

Gene expressed

% of samples with gene expression

MMP-2 MMP-9 TIMP-1 CD40L PAR-1 P-selectin TF TFPI t-PA u-PA PAI-1 MPO PTX3 CRP CXCL9 Fractalkine MCP-1 IL-18 IL-1ß IL-8 TNFα IL-12

+ + + + + + + + + + + + + + + + + + + + -

70 96 100 67 88 96 51 99 66 84 91 55 60 0 52 60 85 66 94 100 54 0

For abbreviations: see text.

Gene Expression Twenty of the 22 selected mediators were expressed in more than 50% of the samples, while CRP and IL-12 were not expressed in any sample (Table 2). Gene expression of plaque rupture and plaque instability related mediators (MMP-2, MMP-9 and TIMP-1) were observed in the range of 70 – 100 % of the samples. Likewise, mediators related to platelet activation (CD40L, PAR-1 and P-selectin) were observed in the range of 67 – 96 % of the samples. Gene expression of TF, the activator of the extrinsic pathway of the coagulation cascade, and its inhibitor TFPI were expressed in approximately 50% and 100% of all samples, respectively, while the fibrinolytic mediators t-PA, u-PA and PAI-1 were expressed in the range of 66 – 91 %. Neutrophil cell markers (MPO and PTX3) were expressed in 55-60% of the samples, while the selected inflammatory mediators were expressed in the range of 52 – 100 %. Expressed Genes in Relation to Total Ischemic Time An inverse correlation was observed between total ischemic time and the expression of P-selectin (p = 0.01), while total ischemic time otherwise correlated positively to the expression of t-PA, u-PA, PAI-1, CXCL9, MCP-1, IL-18, TNFα, MMP-2 and TIMP-1 (p b 0.05) (Table 3). When dichotomizing total ischemic time at median level (4 hours), there were significant differences in expression levels according to high and low time frame (Table 3). As illustrated in Fig. 1, compared with thrombi from subjects with low ischemic time, the thrombi with ischemic time N 4 hours showed a 0.7-fold reduction in the expression of P-selectin and an increase in t-PA of 2.2-fold, u-PA 5.8-fold, PAI-1 8.7-fold, PTX3 1.7-fold, CXCL9 3-fold, MCP-1 2.6-fold, IL-18 2.3-fold, TNFα 2-fold, MMP-2 2.8-fold and TIMP-1 3.2-fold.

331

(0 – 26.8), p = 0.027) (Fig. 2a). When adjusting for high vs. low total ischemic time and patient age in a multivariable regression model, this association remained statistically significant (p = 0.033). The presence of hypertension was associated with a 0.5-fold lower expression of IL-8 (median RQ-values (range)): (2.5 (0 – 21.9) vs 5.0 (1.0 – 108.3), p = 0.006) and TIMP-1 (1.6 (0.2 – 71.7) vs 3.5 (0.5 – 16.9), p = 0.007) (Fig. 2b and c). Neither current smoking nor BMI above median (27.8 kg/m 2 ) associated with expression of the studied mediators (data not shown). Discussion The main findings of the study were that 1) aspirated coronary thrombi expressed genes of mediators related to plaque rupture, platelet and neutrophil cell activation, coagulation, as well as fibrinolysis and inflammation 2) the expression levels of most of the genes were associated with total ischemic time 3) PAI-1 was significantly higher expressed in thrombi from patients with T2DM, 4) IL-8 and TIMP-1 were significantly lower expressed in thrombi from patients with hypertension and 5) smoking and overweight did not significantly affect the expression of the studied genes. Although occlusive aspirated coronary thrombi are reported to contain platelets, fibrin, red blood cells, a variety of pro-inflammatory markers and possibly primitive stem cells [7–12], this is the first report on the genetic expression profile of coronary thrombotic material in patients suffering from myocardial infarction. We found that aspirated coronary thrombi of patients with AMI were highly genetically active and that markers of plaque rupture (MMP-2, MMP-9 and TIMP-1), platelet activation (CD40L, PAR-1 and P-selectin), neutrophil cell activation (MPO and PTX3), coagulation and fibrinolysis (TFPI, u-PA, t-PA and Table 3 Spearman correlations between the biomarkers and ischemic time (n = 67), and absolute RQ-values of the biomarkers in relation to median ischemic time. Ischemic time r

RQ-values

MMP-2

0.480

p-value 0.001

MMP-9 TIMP-1

-0.248 0.521

0.054 b0.001

CD40L PAR-1 P-selectin

-0.157 -0.169 -0.330

0.316 0.218 0.010

TF TFPI t-PA

0.186 0.186 0.444

0.301 0.147 0.003

u-PA

0.527

b0.001

PAI-1

0.552

b0.001

MPO PTX3

0.090 0.371

0.612 0.020

CRP CXCL9

-

0.003

Fractalkine MCP-1

0.320 0.489

0.050 b0.001

IL-18

0.509

0.001

Il-1β IL-8 TNFα

0.069 0.225 0.418

0.603 0.076 0.014

IL-12

-

+

1.1 (0.0 – 18.6) 3.1 (0.3 – 32.4)

0.011

+

1.4 (0.3 – 8.7) 4.5 (0.3 – 71.7)

b0.001

+

1.3 (0.1 – 6.1) 0.9 (0.0 – 3.1)

0.087

+ + +

1.3 (0.1 – 7.3) 2.8 (0.1 – 36.8) 6.8 (0.2 – 194.0) 39.1 (1.0 – 232.9) 0.3 (0.0 – 4.1) 2.6 (0.1 – 26.8)

0.016

+

0.6 (0.1 – 2.6) 1.0 (0.1 – 29.0)

0.247

+

0.2 (0.0 – 4.4) 0.6 (0.2 – 8.5)

0.004

+ +

0.9 (0.0 – 66.7) 2.3 (0.0 – 31.2) 4.6 (0.8 – 70.1) 10.7 (0.8 – 89.2)

0.006

+

0.9 (0.1 – 4.4) 1.8 (0.2 – 9.3)

0.187

0.003 b0.001

0.507

Expressed Genes in Relation to Clinical Entities Patient age correlated positively to the expression of MMP-2 (r = 0.364, p = 0.012), TFPI (r = 0.264, p = 0.032), PAI-1 (r = 0.394, p = 0.002) and IL-18 (r = 0.319, p = 0.035), respectively. The presence of T2DM was associated with a 3.2-fold higher expression of PAI-1 (median RQ-values (range)): (1.6 (0.1 – 12.7) vs 0.5

p-value

0.008

-

- refers to ischemic time ≤ median (4.0 hours), + refers to ischemic time N median (4.0 hours). RQ: Relative quantification values.

332

R. Helseth et al. / Thrombosis Research 135 (2015) 329–333

Fig. 1. Fold regulation of the selected mediators as related to total ischemic time ≤ 4 hours as reference (=1).

PAI-1), and a diversity of pro-inflammatory markers were expressed in a large number of samples. CRP was not expressed in any of the samples, which is somewhat surprising as a previous immunohistochemical study of coronary thrombi obtained post-mortem exhibited positive staining for CRP [11]. Probably, cells within a thrombus do not express CRP, as CRP is known to be produced in the liver. Expression of the selected genes differed significantly in relation to the total ischemic time. The inverse correlation between the platelet marker P-selectin and the total ischemic time goes along with previous reports in which the platelet content in aspirated thrombi has been suggested to decrease with increasing ischemic time [7,12]. A recent study could not find any influence of elapsed time of ischemia on Pselectin levels in plasma [18]. They could, however, show reduced expression of CD61 with increased ischemic time, indicative of reduced platelet content. As plasma levels and gene expression levels cannot be directly compared, our findings seem not to be contradictory to their findings. Our finding of increased expression of genes related to fibrinolysis along total ischemic time are also supported by previous reports showing positive correlation between coronary thrombi fibrin content and total ischemic time [7,12]. We furthermore observed increased expression of genes related to neutrophil activation and inflammation with increasing total ischemic time, possibly reflecting increasing involvement of inflammatory cells over time during thrombus organisation. A maintained presence of neutrophils and monocytes in aspirated thrombi regardless of elapsed ischemia time has, however, recently also been discussed [18]. Thus, the association between inflammatory cell content and ischemic time is still not fully understood. Importantly, even though an inverse correlation between P-selectin expression and ischemic time was observed, the RQ absolute values of the mediators at ischemic time of b4.0 hours implied that both t-PA, u-PA, TIMP-1 and IL-18 were equally or more pronounced

expressed than P-selectin, indicating that the thrombi also initially most likely contained fibrin, modulators of plaque rupture and some proinflammatory markers (Table 3). The expression of MMP-2, TFPI, PAI-1 and IL-18 correlated significantly to age which is in line with previous reports [19–21]. Increased levels of these mediators might illustrate the tendency to increased arterial thrombosis formation in patients of higher age [19]. A possible interpretation of the current correlations to patient age may be that increased levels of TFPI in the elderly might be to balance a prothrombotic state with high levels of PAI-1. The presence of T2DM associated with a 3.2-fold increased expression of PAI-1 and the association persisted significantly after adjusting for total ischemic time and age. The other markers were apparently unaffected by the presence of T2DM. PAI-1 is a specific, fast acting inhibitor of both t-PA and u-PA and hence diminishes endogenous fibrinolysis [22]. PAI-1 has previously been linked to the increased tendency of thrombus formation in patients with T2DM [22] and increased levels of PAI-1 are also discussed to contribute substantially to the development of atherosclerosis in T2DM [22]. The observed increased PAI-1 expression in aspirated thrombi from T2DM subjects thereby supports the evidence of the prothrombotic properties of PAI-1 in subjects with T2DM. Expression of IL-8 and TIMP-1 seemed to be less pronounced in patients suffering from arterial hypertension. Inflammation has been discussed to play a role in the pathogenesis of hypertension [23] and high levels of IL-8 have previously been reported to associate with pregnancy-induced hypertension [24]. We observed in contrast that hypertension was associated with lower gene expression of IL-8. As the definition of hypertension in our population also included the use of antihypertensive drugs, any influence of antihypertensive medications on IL-8 regulation can not be excluded. TIMP-1 binds to metalloproteinases and inhibits their proteolytic activity [25]. A recent review reported that hypertensive patients had high circulating levels of both TIMP-1 and MMP-9 [26]. The lower TIMP-1 expression observed in our hypertensive patients might indicate that hypertension per se increases plaque instability through less inhibition of MMP-9. It is now also growing evidence for metalloproteinaseindependent TIMP-mediated actions [26], which might explain why MMP-2 and MMP-9 were unaffected by the hypertensive state. Obviously, antihypertensive therapy might also influence the regulation of TIMP-1. Neither smoking nor BMI seemed to affect any of the expressed genes. This was somewhat surprising, as both smoking and high BMI traditionally are seen as being prothrombotic and pro-inflammatory. However, this might not be true for the micro-environment studied here. Knowledge about the RNA content of coronary thrombi is scarse. By observing that coronary thrombi from subjects with AMI expressed a diversity of genes related to different stages of atherothrombosis and plaque rupture and also that the genetic profile apparently changed along with ischemic time, this report contributes to more basic understanding of the pathophysiological situation in AMI. Increased

Fig. 2. (a): PAI-1 RQ-values according to presence of T2DM, (b) IL-8 RQ-values according to presence of hypertension and (c) TIMP-1 RQ-values according to presence of hypertension. Boxes are given with median values, 25 and 75 percentiles and with 2.5 and 97.5 percentiles. Outliers are marked with *. One outlier in each variable has been omitted to make the figure more readable. P-values refer to differences between groups.

R. Helseth et al. / Thrombosis Research 135 (2015) 329–333

knowledge about coronary thrombi beyond the structural and cellular level may contribute to improve further therapeutic strategies in AMI. The impact of our observations remains to be explored. Limitations and Strenghts The strenght of our study is the experimental model of gene array analyses in aspirated coronary thrombi from patients with AMI. An important limitation is the limited sample size. Due to the novelty of this issue, no power calculation was performed. The selection of genes is based on previous knowledge about the atherothrombotic process, but not all genetic mediators that might be of interest have been investigated. Accordingly, the gene selection might be biased and is open for discussion. The selection of house-keeping gene was based on experience and can be discussed. We could, however, not find any significant correlation between the house keeping gene used and ischemic time (data not shown). Unfortunately, we were not able to dissect the cell type responsible for the expressed genetic mediators. We can furthermore not exclude that the genes expressed are influenced by use of different medications prior to hospital admission, as well as by the acute pharmacological management of AMI, described for instance for statins [27]. Conclusions Gene expression of several pro-thrombotic and pro-inflammatory markers was present in coronary thrombi aspirated from patients with AMI. The genetic expression profile seemed to change according to total ischemic time with a decrease in expression of genes related to platelet activation and an increase in expression of genes related to fibrinolysis, inflammation and plaque instability. Expression of PAI-1 was significantly higher in patients with T2DM, possibly confirming the particular role of impaired fibrinolysis in T2DM. The presence of hypertension seemed to associate with plaque instability through a TIMP-1 mediated mechanism. Addendum R. Helseth contributed to the development of the study protocol, statistics, evaluation of results, preparation and discussion of manuscript. I. Seljeflot contributed in the development of study protocol, laboratory analyses, statistics, and discussion of manuscript. T. Opstad contributed to laboratory analyses, evaluation of results, and discussion of manuscript. S. Solheim, M. Freynhofer, H. Arnesen, K. Huber and T. Weiss all contributed in the development of study protocol, evaluation of results and discussion of manuscript. Disclosures None. Acknowledgements The study was supported by Stein Erik Hagens Foundation for Clinical Heart Research, Oslo, Norway and by grants from the Association for the Promotion of Scientific Research in the Area of Arteriosclerosis, Thrombosis and Vascular Biology (ATVB), Vienna, Austria and “Ada og Hagbart Waages Humanitære og Veldedige stiftelse”. We thank medical laboratory thechnologist Sissel Åkra for her professional and careful handling of the frozen thrombi and subsequent RNA isolation.

333

References [1] Global atlas on cardiovascular disease prevention and control. Geneva: World Health Organization; 2011[Available at: http://www.who.int/mediacentre/factsheets/ fs317/en/index.html. Accessed October 8th, 2013]. [2] Freynhofer MK, Bruno V, Wojta J, Huber K. The role of platelets in athero-thrombotic events. Curr Pharm Des 2012;18:5197–214. [3] Vlaar PJ, Svilaas T, van der Horst IC, Diercks GF, Fokkema ML, de Smet BJ, et al. Cardiac death and reinfarction after 1 year in the Thrombus Aspiration during Percutaneous coronary intervention in Acute myocardial infarction Study (TAPAS): a 1-year follow-up study. Lancet 2008;371:1915–20. [4] Burzotta F, De Vita M, Gu YL, Isshiki T, Lefèvre T, Kaltoft A, et al. Clinical impact of thrombectomy in acute ST-elevation myocardial infarction: an individual patient-data pooled analysis of 11 trials. Eur Heart J 2009;30:2193–203. [5] De Luca G, Dudek D, Sardella G, Marino P, Chevalier B, Zijlstra F. Adjunctive manual thrombectomy improves myocardial perfusion and mortality in patients undergoing primary percutaneous coronary intervention for ST-elevation myocardial infarction: a meta-analysis of randomized trials. Eur Heart J 2008;29:3002–10. [6] Fröbert O, Lagerqvist B, Olivecrona GK, Omerovic E, Gudnason T, Maeng M, et al. Thrombus aspiration during ST-segment elevation myocardial infarction. N Engl J Med 2013;369:1587–97. [7] Silvain J, Collet JP, Nagaswami C, Beygui F, Edmondson KE, Bellemain-Appaix A, et al. Composition of coronary thrombus in acute myocardial infarction. J Am Coll Cardiol 2011;57:1359–67. [8] Nagata Y, Usuda K, Uchiyama A, Uchikoshi M, Sekiguchi Y, Kato H, et al. Characteristics of the pathological images of coronary artery thrombi according to the infarct-related coronary artery in acute myocardial infarction. Circ J 2004;68:308–14. [9] Ohba T, Hata N, Ohaki Y. Angioscopic apparance and histopathology of coronary artery thrombi. Asian Cardiovasc Ann 2003;11:255–7. [10] Kato K, Matsubara T, Iida K, Suzuki O, Sato Y. Elevated levels of pro-inflammatory cytokines in coronary artery thrombi. Int J Cardiol 1999;70:267–73. [11] Sato Y, Hatakeyama K, Yamashita A, Marutsuka K, Sumiyoshi A, Asada Y. Proportion of fibrin and platelets differs in thrombi on ruptured and eroded coronary atherosclerotic plaques in humans. Heart 2005;91:526–30. [12] Iwata H, Sata M, Ando J, Fujita H, Morita T, Sawaki D, et al. Impact of primitive cells in intracoronary thrombi on lesion prognosis: temporal analysis of cellular constituents of thrombotic material obtained from patients with acute coronary syndrome. Heart 2010;96:748–55. [13] Yunoki K, Naruko T, Sugioka K, Inaba M, Itoh A, Haze K, et al. Thrombus aspiration therapy and coronary thrombus components in patients with acute ST-elevation myocardial infarction. J Atheroscler Thromb 2013;20:524–37. [14] Sakai T, Inoue S, Takei M, Ogawa G, Hamazaki Y, Ota H, et al. Activated inflammatory cells participate in thrombus size through tissue factor and plasminogen activator inhibitor-1 in acute coronary syndrome: Immunohistochemical analysis. Thromb Res 2011;127:443–9. [15] Sakai T, Inoue S, Matsuyama TA, Takei M, Ota H, Katagiri T, et al. Eosinophils may be involved in thrombus growth in acute coronary syndrome. Int Heart J 2009;50: 267–77. [16] Arakawa K, Yasuda S, Hao H, Kataoka Y, Morii I, Kasahara Y, et al. Significant association between neutrophil aggregation in aspirated thrombus and myocardial damage in patients with ST-segment elevation acute myocardial infarction. Circ J 2009;73:139–44. [17] Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 2001;25:402–8. [18] Ramaiola I, Padró T, Peña E, Juan-Babot O, Cubedo J, Martin-Yuste V, et al. Changes in thrombus composition and profilin-1 release in acute myocardial infarction. Eur Heart J 2014 [pii: ehu356. [Epub ahead of print]]. [19] Cesari M. Plasminogen activator inhibitor-1 (PAI-1): a key factor linking fibrinolysis and age-related subclinical and clinical conditions. Cardiovasc Ther 2010;28:e72–91. [20] Mitchell CT. Tissue factor pathway inhibitor, vascular risk factors and subclinical atherosclerosis: the Multi-Ethnic Study of Atherosclerosis. Atherosclerosis 2009; 207:277–83. [21] Chung AW. Increased matrix metalloproteinase 2 activity in the human internal mammary artery is associated with ageing, hypertension, diabetes and kidney dysfunction. J Vasc Res 2008;45:357–62. [22] Jankun J, Al-Senaidy A, Skrzypczak-Jankun E. Can inactivators of plasminogen activator inhibitor alleviate the burden of obesity and diabetes? (Review). Int J Mol Med 2012; 29:3–11. [23] Boos CJ, Lip GY. Is hypertension an inflammatory process? Curr Pharm Des 2006;12: 1623–35. [24] Wang T, Wang YY, Zhou R, Song CP, Lin W, Niu XY, et al. Bioactive proteins in healthy pregnancies and preeclampsia: relevance to hypertension and proteinuria. Chin Med J (Engl) 2013;126:2015–20. [25] Brew K, Dinakarpandian D, Nagase H. Tissue inhibitors of metalloproteinases: evolution, structure and function. Biochim Biophys Acta 2000;1477:267–83. [26] Marchesi C, Dentali F, Nicolini E, Maresca AM, Tayebjee MH, Franz M, et al. Plasma levels of matrix metalloproteinases and their inhibitors in hypertension: a systematic review and meta-analysis. J Hypertens 2012;30:3–16. [27] Demyanets S, Kaun C, Maurer G, Huber K, Wojta J. Statins modulate expression of components of the plasminogen activator/plasmin system in human cardiac myocytes in vitro. J Thromb Haemost 2006;4:476–9.