Article Text

Acute coronary syndromes
Increased serum levels and expression of S100A8/A9 complex in infiltrated neutrophils in atherosclerotic plaque of unstable angina
  1. S Miyamoto1,
  2. M Ueda2,
  3. M Ikemoto3,
  4. T Naruko4,
  5. A Itoh4,
  6. S Tamaki5,
  7. R Nohara1,
  8. F Terasaki6,
  9. S Sasayama7,
  10. M Fujita3
  1. 1
    Division of Cardiology, Kitano Hospital, Tadukekofukai Medical Research Institute, Osaka, Japan
  2. 2
    Department of Pathology, Osaka City University Graduate School of Medicine, Osaka, Japan
  3. 3
    Human Health Sciences, Kyoto University Graduate School of Medicine, Kyoto, Japan
  4. 4
    Department of Cardiology, Osaka City General Hospital, Osaka, Japan
  5. 5
    Division of Cardiology, Takeda Hospital, Kyoto, Japan
  6. 6
    Third Department of Internal Medicine, Osaka Medical College, Takatsuki, Japan
  7. 7
    Heart Bio-Mechanics Centre, Doshisha University, Kyoto, Japan
  1. Professor M Fujita, Human Health Sciences, Kyoto University Graduate School of Medicine, 53 Kawaharacho, Shogoin, Sakyo-ku, Kyoto 606–8507, Japan; mfujita{at}kuhp.kyoto-u.ac.jp

Abstract

Background: The S100A8/A9 complex is expressed in a subset of activated neutrophils and macrophages in acute inflammatory lesions associated with various diseases.

Objective: To investigate (a) whether serum S100A8/A9 levels are increased in patients with unstable angina (UA); and (b) whether S100A8/A9 expression is upregulated in coronary atherosclerotic plaques of patients with UA.

Design: Serum S100A8/A9 levels in 39 patients with stable angina (SA) and 53 patients with UA were measured. In addition, the presence of the S100A8/A9 complex in directional coronary atherectomy specimens was studied immunohistochemically. Cell types which stain positive for S100A8/A9 were identified by immunodouble staining with neutrophils and macrophages.

Results: Mean (SD) serum S100A8/A9 levels were significantly higher in patients with UA than in those with SA (3.25 (3.08) μg/ml vs 0.77 (0.31) μg/ml, p<0.05). In patients with UA, immunodouble staining clearly showed that the S100A8/A9 complex was expressed in infiltrated neutrophils and occasional macrophages. The S100A8/A9-positive area was significantly higher in UA than in SA (mean (SD) 18.3 (14.2)% vs 1.3 (2.4)%, respectively, p<0.001).

Conclusions: The S100A8/A9 complex may be involved in the inflammatory process of coronary atherosclerotic plaques in patients with UA.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Accurate diagnosis of unstable angina (UA) at the earliest stage would improve prognosis through appropriate treatment without delay. Diagnostic tests for myocardial injury include echocardiography, radioisotope scintigraphy and measurement of circulating levels of the MB isoform of creatine kinase, and troponin T to identify myocardial necrosis. However, none of these markers directly detect plaque instability or rupture before myocardial damage appears. The availability of a sensitive and specific early biomarker of plaque instability, whose levels become raised before, or in the absence of a rise of, other markers of myocardial cell injury, might improve the diagnosis and therapeutic decision making of UA at the earliest stage.

At present, there are many suggestions about a link between local inflammation and UA.1 2 We hypothesised that the S100A8/A9 complex,35 which is primarily expressed by circulating activated neutrophils and macrophages, might be such a biomarker and might identify patients with unstable atherosclerotic plaques. The S100A8/A9 complex is a non-covalent heterodimeric molecule, which is composed of S100A8 (10.8 kDa) and S100A9 (13.2 kDa). Each subunit has two calcium-binding sites.68 The secretion of the S100A8/A9 complex toward vascular walls may be involved in the migration of leucocytes into inflammatory tissues such as proatherosclerotic lesion.9

In this study, we measured circulating S100A8/A9 complex levels in patients with stable angina (SA) and UA, using monoclonal antibodies against the S100A8/A9 complex.10 Additionally, we compared the diagnostic accuracy of serum S100A8/A9 complex levels with serum C-reactive protein (CRP) levels in the differentiation between SA and UA. We also immunohistochemically examined the expression of the S100A8/A9 complex in specimens obtained from patients with SA and UA undergoing directional coronary atherectomy (DCA).

METHODS

This study complies with the Declaration of Helsinki. The study protocol was approved by the institutional review committees of Takeda Hospital and Osaka City General Hospital, and the subjects gave informed consent.

Study I

The cohort consisted of 92 patients with either SA or UA. SA was diagnosed in 39 patients and defined as chest pain typical of myocardial ischaemia on exertion. UA was diagnosed in 53 patients and defined as angina at rest. The patients with UA were further divided into class II (n = 29) and class III (n = 24) according to the Braunwald’s criteria.11 We excluded patients with non-atherosclerotic inflammatory disorders such as pneumonia and vasculitis in order to avoid potentially confounding results with respect to the S100A8/A9 complex. A total of 38 age- and sex-matched healthy volunteers served as controls (26 men and 12 women; mean (SD) age 63 (12) years). None of the control subjects had hypercholesterolaemia or diabetes mellitus, four had a history of hypertension and five were current smokers. All four hypertensive subjects were in stage I according to the criteria established by the Joint National Committee VII12; none used antihypertensive drugs.

The following data were obtained: age, sex and risk factors (hypertension as defined by the Joint National Committee VII,12 diabetes mellitus as defined by the WHO Study Group,13 hypercholesterolaemia defined as the serum total cholesterol levels ⩾5.7 mmol/l and cigarette smoking).

Measurement of serum levels of S100A8/A9 complex, CRP, d-dimer and fibrinogen degradation product (FDP)

Venous blood samples from all patients were obtained on admission to hospital. Measurement of serum S100A8/A9 complex levels was performed using a sandwich ELISA that had been newly developed in our laboratory.10 Serum levels of CRP, d-dimer and FDP were measured by the conventional method.

Study II

Coronary atherectomy specimens were obtained by DCA from the culprit lesion in 51 patients with either SA (n = 26) or UA (n = 25). The patients with UA comprised 15 patients in Braunwald’s class I, 3 patients in class II, and 7 in class III11 undergoing percutaneous coronary intervention for a single primary lesion at Osaka City General Hospital, Osaka. The culprit lesion was identified on the basis of clinical, ECG and coronary angiographic data. The patients in whom the culprit lesion was not identified were excluded from this study. In all patients, the procedure was performed against a native de novo atherosclerotic lesion. Immediately after atherectomy, the tissue specimens were carefully oriented along their longest axis, snap-frozen and stored at −80°C. The snap-frozen samples obtained by DCA were subsequently serially sectioned to produce sections of 6 μm in thickness, and then fixed in acetone. All first sections were stained with H&E; the other sections were used for immunohistochemical staining.

Immunohistochemistry

To identify the S100A8/A9 complex, a rabbit polyclonal antibody was used. The methods of antibody production and specificity testing have been reported previously.10 Immunohistochemical identification of cells was achieved using antibodies directed against vascular smooth muscle cells (1A4; DAKO, Glostrup, Denmark), macrophages (EBM11; DAKO), neutrophils (CD66b; Beckman Coulter, Fullerton, California, USA) and platelets (glycoprotein IIb/IIIa, DAKO). Mac387 (Abcam, Cambridge, Massachusetts, USA), a marker of neutrophils and activated macrophages, was also used.

Single staining

The sections were subjected to a three-step staining procedure, with the use of a streptavidin–biotin complex with horseradish peroxidase. Horseradish peroxidase activity was visualised with 3-amino-9-ethylcarbazole, and the sections were faintly counterstained with haematoxylin. The specificity and results obtained with anti-S100A8/A9 polyclonal antibody were checked by omission of the primary antibodies and use of a non-immune mouse IgG antibody (DAKO) as a negative control.

Immunodouble staining

For the simultaneous identification of smooth muscle cells and macrophages, immunodouble staining was performed based on two primary antibodies of a different IgG subclass (1A4/EBM11), as described previously.14 In this immunodouble staining, we visualised the enzymatic activity of β-galactosidase for 1A4 in turquoise (BioGenex Kit; BioGenex, San Ramon, California, USA) and the activity of alkaline phosphatase for EBM11 in red (New Fuchsin Kit; DAKO). To identify cell types that stained positive for the S100A8/A9 complex, we also performed immunodouble staining for neutrophils/S100A8/A9, macrophages/S100A8/A9 and Mac387/S100A8/A9 according to the method previously reported with minor procedural modifications.14 In this staining procedure, alkaline phosphatase was visualised with fast blue BB (blue, neutrophils or macrophages, or Mac387) and peroxidase was visualised with 3-amino-9-ethylcarbazole development (red, S100A8/A9).

Quantitative analysis

The surface area occupied by S100A8/A9-positive cells was quantified using computer-aided planimetry and expressed as a percentage of the total tissue area of the atherectomy specimen. The morphometric analysis was performed by a single investigator who was unaware of the patient’s clinical characteristics.

Statistical analysis

Data are presented as the mean (SD). Statistical comparisons were performed using unpaired two-tailed Student’s t tests or one-way analysis of variance with Scheffe’s test where appropriate, with a p value <0.05 taken to indicate significance. Spearman correlation coefficients were used to evaluate the relations between S100A8/A9 complex levels and variables related to each cardiovascular risk factor, inflammation and coagulation. Receiver operating characteristic (ROC) analysis was also carried out on the serum levels of the S100A8/A9 complex and CRP for the differentiation between SA and UA. This analysis plots the true-positive fraction (sensitivity) against the false-positive fraction (1−specificity) by changing the cut-off value for the test. Areas under the ROC curves indicate the relative accuracy of diagnostic tests for identification of UA among patients with angina.15

RESULTS

Serum S100A8/A9 complex levels

Table 1 shows the characteristics of patients included in study I. There were no differences between patients with SA and UA for age, sex and risk factors except for body mass index. The mean (SD) body mass index was significantly higher in patients with SA than in patients with UA (25 (4) kg/m2 vs 23 (4) kg/m2, p = 0.013). Control subjects had no risk factors or coronary artery disease (blood pressure 124 (10)/74 (8) mm Hg, fasting blood glucose 5.5 (0.5) mmol/l, low-density lipoprotein cholesterol 2.59 (0.36) mmol/l, high-density lipoprotein cholesterol 1.27 (0.26) mmol/l, triglycerides 1.19 (0.28) mmol/l). Serum S100A8/A9 complex levels in patients with UA were significantly higher than in those with SA and in control subjects (UA 3.25 (3.08) μg/ml; SA 0.77 (0.31) μg/ml; controls 0.27 (0.11) μg/ml). There were no statistically significant differences in serum S100A8/A9 complex levels between the classes II and III of UA (class II 2.6 (1.9) μg/ml; class III 4.1 (4.0) μg/ml). Also, there were no correlations between the serum S100A8/A9 complex levels and each cardiovascular risk factor.

Table 1 Patients’ characteristics with measurements of serum levels of the S100A8/A9 complex and C-reactive protein (CRP)

Serum levels of CRP, d-dimer and FDP

In all subjects of the control group, serum CRP levels were <1 mg/l. Serum CRP levels were significantly higher in patients with UA than in those with SA (UA 31 (61) mg/l; SA 2 (2) mg/l). There were no statistically significant differences in serum CRP levels between the classes II and III of UA (class II 35 (59) mg/l; class III 26 (65) mg/l). Serum d-dimer and FDP levels tended to be higher in patients (n = 10) with UA than in those (n = 8) with SA (d-dimer: UA 3.56 (4.91) μg/ml; SA 0.53 (0.24) μg/ml, p = 0.10, FDP: UA 5.5 (4.0) μg/ml; SA 2.5 (0.0) μg/ml, p = 0.06). There were significant correlations between serum S100A8/A9 levels and serum levels of d-dimer and FDP (n = 18, rs = 0.772, p<0.001 for the d-dimer; n = 18, rs = 0.872, p<0.001 for FDP).

Serum S100A8/A9 complex levels as a diagnostic biomarker of UA

Figure 1 shows ROC curves for the serum levels of the S100A8/A9 complex and CRP in 53 patients with UA and 39 patients with SA. The areas below the curves were 0.95 (95% CI 0.91 to 0.99) for the S100A8/A9 complex and 0.78 (95% CI 0.69 to 0.87) for CRP. This difference between the S100A8/A9 complex and CRP is significant (p<0.05). Given a cut-off value of 1.2 μg/ml for the S100A8/A9 complex, serum S100A8/A9 complex levels can significantly (p<0.05) discriminate between patients with UA and patients with SA among these patients and showed 84.9% sensitivity and 89.7% specificity for the diagnosis of UA. In contrast, a CRP cut-off value of 5 mg/l, which had comparable specificity (92.3%), showed lower sensitivity (47.2%) for the diagnosis of SA. In this study there was no correlation between the serum S100A8/A9 complex levels and serum CRP levels.

Figure 1 Receiver operating characteristic curves of serum levels of the S100A8/A9 complex and C-reactive protein (CRP) for diagnosis of angina in 53 patients with unstable angina and 39 patients with stable angina. The true-positive fraction (sensitivity) was plotted against the false-positive fraction (1−specificity) by changing cut-off values for the test.

Immunohistochemical quantification in atherectomy specimens

Table 2 shows the characteristics of patients included in study II. Age, sex, and risk factors did not differ between patients with SA and UA. In the atherectomy specimens of patients with SA, S100A8/A9 positivity was found in only occasional macrophages (fig 2). In contrast, in coronary lesions of patients with UA, 22/25 lesions (88%) showed the presence of the S100A8/A9 complex. Double immunostaining for neutrophils (or macrophages) and the S100A8/A9 complex demonstrated that the vast majority of S100A8/A9-positive cells were neutrophils (fig 3), and S100A8/A9 positivity was also found in occasional macrophages. Neutrophil infiltration was found in unstable plaques and at sites of platelet thrombus overlying the site of plaque injury (fig 3). Double immunostaining for Mac387 and S100A8/A9 also showed staining positivity for S100A8/A9 in neutrophils and occasional macrophages (fig 3B). In these experiments, sections treated with a non-immune IgG antibody gave a negative result. Figure 4 shows the S100A8/A9-positive area for each individual lesion in the two groups. The S100A8/A9-positive area was significantly higher (p<0.001) in UA (18.3 (14.2)%) than in SA (1.3 (2.4)%).

Figure 2 Micrographs of a culprit lesion in a patient with stable angina (SA). (A) Immunodouble staining (smooth muscle cell (SMC), turquoise; macrophage (MΦ, red) shows no macrophages in the lesion dominated by SMCs. (B) Section adjacent to (A), stained with antineutrophil CD66b antibody, shows no positivity in the macrophage-poor area. (C) Adjacent section, stained with anti-S100A8/A9 antibody, shows no positivity as with neutrophils. Bars indicate 200 µm.
Figure 3 (A–E) Micrographs of an atherectomy specimen obtained from a culprit lesion in a patient with unstable angina (UA). (A) Immunodouble staining (smooth muscle cell (SMC), turquoise; macrophage (MΦ, red) shows an area with macrophages. (B) Section stained with anti-glycoprotein (GP) IIb/IIIa antibody shows the presence of platelet thrombus at the site of plaque injury. (C) Section stained with antineutrophil CD66b antibody shows the presence of neutrophils. (D) Section stained with anti-S100A8/A9 antibody shows distinct staining positivity. (E) Double immunostaining for neutrophils (blue) and S100A8/A9 (red) discloses that neutrophils show double staining (purple), indicating that S100A8/A9 is expressed in neutrophils. Bar: panels A, B, C, D 200 µm; panel E 50 µm. (F–J) Micrographs of an atherectomy specimen obtained from a culprit lesion in a patient with UA. (F) Immunodouble staining (SMC, turquoise; MΦ, red) shows large numbers of macrophages. (G) Section stained with anti-GP IIb/IIIa antibody shows that only a limited area shows staining positivity for platelet GP IIb/IIIa. (H) Section stained with antineutrophil CD66b antibody shows neutrophil infiltration in a plaque tissue. (I) Double immunostaining for neutrophil (blue) and S100A8/A9 (red) discloses that neutrophils show double staining (purple), indicating that S100A8/A9 is expressed in neutrophils. (J) Double immunostaining for Mac387 and S100A8/A9 shows that S100A8/A9 is expressed in Mac387-positive cells (neutrophils and occasional macrophages). Bar: panels F, G, H 100 µm; panels I, J 50 µm.
Figure 4 The percentage of S100A8/A9-positive area in atherectomy specimens obtained from culprit lesions in patients with stable angina (SA) and unstable angina (UA). Error bars indicate the mean (SE).
Table 2 Patients’ characteristics with directional coronary atherectomy

DISCUSSION

We found an association between UA and high serum S100A8/A9 complex levels. Both S100A8 and S100A9 are members of S100 family of calcium-binding proteins. This molecule constitutes 45% of neutrophil cytosolic protein, suggesting that it has a key role in the activities of cells.5 Circulating S100A8/A9 complex levels are used as an inflammatory marker in the diagnosis of juvenile rheumatoid arthritis16 and acute renal allograft rejection.17 Recently, Ye et al reported that the other neutrophil-derived S100A12 has a potentially important role in the development of Kawasaki disease.18

Increased levels of the S100A8/A9 complex and CRP in patients with UA are consistent with growing evidence suggesting that inflammation has a pivotal role in these syndromes. The accumulation of neutrophils and macrophages in unstable atherosclerotic lesions stimulates the production of metalloproteinases19 and may be associated with increases in the serum CRP levels.20

Recently, Healy et al suggested that the S100A8/A9 complex not only induces inflammatory response but also evokes thrombogenic response by performing transcriptome analyses of megakaryocyte-derived mRNAs in patients with acute ST-segment elevation myocardial infarction and stable coronary artery disease.21 Thus, it is likely that the S100A8/A9 complex is involved in the pathophysiological process of UA. However, our study cannot answer questions such as whether the S100A8/A9 complex is a primary or secondary component of UA. Further studies are required to characterise the importance of the S100A8/A9 complex within unstable vulnerable plaques and to elucidate the diagnostic and prognostic significance of raised S100A8/A9 complex levels in patients with UA.

Our immunohistochemical study using atherectomy specimens clearly demonstrated that the percentage of S100A8/A9-positive area in the culprit lesions of patients with UA is significantly higher than in those of patients with SA. It is presently well accepted that intraplaque inflammation plays a key part in plaque destabilisation and, hence, in the pathophysiology of UA.22 23 Our present findings not only support this concept but also suggest a pivotal role of the S100A8/A9 complex in the genesis of coronary plaque instability and the development of UA. Naruko et al investigated this possibility in patients with UA; neutrophils within the culprit lesions were detected in 44%, whereas this was the case in only 6% in patients with SA. These observations suggest that neutrophils are heavily involved in the pathogenesis of acute coronary syndromes.24

Several biomarkers have already been shown to be associated with the presence of vulnerable coronary plaques. Neopterin, a by-product of the guanosine triphosphate pathway, is produced by activated macrophages after stimulation by interferon-γ released by T lymphocytes, and serves as an activation marker for monocytes/macrophages.25 Recently, we demonstrated that neopterin-positive macrophage scores showed a significant positive correlation with the number of neutrophils in human coronary atherosclerotic lesions (supplementary fig 1).26 These findings suggest that S100A8/A9 localisation may be associated with neopterin expression in human coronary unstable plaques.

There is an important limitation inherent to our study. We did not use an ultrasensitive immunonephelometric latex-enhanced assay but a traditional immunonephelometric assay. In the data we obtained, the serum CRP levels were fairly high, particularly in patients with UA. Therefore, we did not report CRP levels obtained by high-sensitivity CRP measurements. The extremely high CRP levels in five patients with UA may have been due to the coexistence of high-grade inflammatory conditions.

In conclusion, the distinct presence of neutrophils and macrophages expressing the S100A8/A9 complex in atherosclerotic plaques underlying UA strongly suggests that the S100A8/A9 complex has an important role in mediating destabilisation of atherosclerotic plaques.

REFERENCES

Footnotes

  • Competing interests: None declared.

  • Ethics approval: Study protocol approved by the institutional review committees of Takeda Hospital and Osaka City General Hospital.

  • ▸ An additional figure is published online only at http://heart.bmj.com/content/vol94/issue8