Monocyte toll-like receptor 4 (TLR4) has been implicated in the pathogenesis of atherosclerosis with increased levels in myocardial infarction. The aim of this study was to assess the numbers of TLR4+ monocytes in each monocyte subset in MI, the expression of TLR4 and association with markers of monocyte activation, inflammation, myocardial damage and postmyocardial infarction (MI) cardiac contractility.
Surface expression of TLR4 and numbers of TLR4-expressing monocytes were quantified by flow cytometry of venous blood in 50 patients with ST-elevation MI (STEMI), 48 with non-STEMI (NSTEMI) and 40 with stable coronary artery disease (CAD). These parameters were measured on days 1, 3, 7 and 30 post-MI in STEMI patients. Three monocyte subsets were defined as CD14++ CD16− CCR2+ (Mon1), CD14++ CD16+ CCR2+ (Mon2) and CD14+ CD16++ CCR2− (Mon3). Plasma inflammatory cytokines were assessed using cytometric bead arrays.
There was a significant increase in counts of TLR4+ Mon1 and Mon2 in STEMI patients and TLR4+ Mon2 in NSTEMI patients compared with controls with CAD. Monocyte TLR4+ expression was similar between the groups, and was not changed during follow-up in STEMI patients. Plasma interleukin-6 (IL6) levels correlated positively with TLR4+ Mon2 count (r = 0.54, P <0.001), but negatively with TLR4 expression on Mon2 (r = −0.33, P =0.021).
Following treatment of acute MI, TLR4 expression by individual monocyte subsets is unchanged. An increase in TLR4+ Mon1 and Mon2 count in patients with STEMI and TLR+ Mon2 count in those with NSTEMI is due to an increase in monocyte subset count and not to changes in TLR4 expression. Monocyte counts but not TLR4 expression correlate positively with plasma IL6 levels. We suggest that TLR4 expression may not be a reliable marker of monocyte activation in MI.
Toll-like receptor 4 (TLR4) is a member of the pattern recognition receptor family. It is involved in innate immunity via bacterial endotoxin-induced inflammatory responses, but is also implicated in the pathogenesis of atherosclerosis . The role of TLR4 in cardiovascular disorders can also be attributed to its responsiveness to host ligands, such as heat shock proteins released upon myocardial damage, fibronectin and reactive oxidative species [2, 3]. Although the biological roles of TLR4-mediated pathways represent an antibacterial defence mechanism, chronic stimulation or acute upregulation by these pathways in the setting of coronary artery atherothrombosis might be detrimental, by facilitating excessive inflammation, leucocyte accumulation and release of matrix metalloproteinases . Monocyte TLR4 overexpression in acute myocardial infarction (MI), as demonstrated both in the circulation and ruptured plaque, has been associated with high plasma levels of inflammatory cytokines, such as interleukin (IL)-6 . Moreover, inhibition of TLR4 in mouse models of MI attenuated the inflammatory process and was associated with reduced infarct size .
However, most published studies of monocyte TLR4 overexpression in MI have been conducted in vitro, which may affect the monocyte phenotype. Furthermore, research to date has focused on only two monocyte subsets, whereas three distinct subsets of monocytes have recently been defined, the so-called ‘classical’ CD14++ CD16− CCR2+ (Mon1), CD14++ CD16+ CCR2+ (Mon2) and CD14+ CD16++CCR2− (Mon3) cells .
The objective of this study was to assess the proportion of TLR4+ monocytes within each of the three human monocyte subsets  in patients with ST-elevation MI (STEMI) or non-STEMI (NSTEMI) and matched control subjects with stable coronary artery disease (CAD). In addition, we measured monocyte subset expression of TLR4 and their relation to markers of monocyte proinflammatory activation, myocardial damage, inflammatory cytokines and convalescent post-MI cardiac contractility.
Fifty consecutive patients with STEMI and 48 with NSTEMI admitted to Sandwell and West Birmingham Hospitals NHS Trust were recruited between November 2009 and November 2010. MI was diagnosed and treated according to the European Society of Cardiology guidelines; all STEMI patients underwent primary percutaneous coronary intervention (PCI) . STEMI and NSTEMI patients were compared with an age- and sex-matched group of patients with stable CAD (n =40), confirmed during elective coronary angiography, with no unstable symptoms or hospital admissions for ≥3 months. Exclusion criteria were factors known to affect monocyte count (including infectious and inflammatory disorders and their treatment, cancer, haemodynamically significant valvular heart disease, atrial fibrillation, renal failure and use of hormone-replacement therapy). In addition, none of the patients had a history of MI within the previous 6 months or left ventricular (LV) dysfunction. All study patients received standard treatment according to current guidelines .
In STEMI patients, blood samples were collected at four time-points: day 1 within 24 h of MI (after primary PCI), day 3, day 7 and day 30. A proportion of STEMI patients did not complete follow-up due to withdrawal of consent (15) or death (2). Peripheral venous blood was collected and flow cytometry was performed within 60 min. Plasma was stored at −70°C for analysis in batches. The study was performed in accordance with the Declaration of Helsinki and was approved by the Coventry Research Ethics Committee. All participants provided written informed consent.
Flow cytometric analysis was performed using the BD FACSCalibur flow cytometer (Becton Dickinson, Oxford, UK) as previously described . The technique is robust and highly reproducible. The laboratory coefficient of variation for absolute monocyte count is 1.9%, and for surface and intracellular markers are <5%. CD14, CD16 and TLR4 positivity were defined using appropriate isotype controls and according to current consensus guidelines .
Absolute monocyte subset and monocyte-platelet aggregate counts
Mouse anti-human monoclonal fluorochrome-conjugated antibodies anti-CD16-Alexa Fluor 488 (clone DJ130c, AbDSerotec, Oxford, UK), anti-CD14-PE (clone MфP9; Becton Dickinson) and anti-CCR2-APC (clone 48607, R&D Systems, Abingdon, UK) were mixed with 50 μL fresh EDTA-anticoagulated whole blood in TruCount tubes (Becton Dickinson) containing a defined number of fluorescent count beads. After incubation for 15 min, red blood cells were lysed with 450 μL lysing solution (Becton Dickinson) for 15 min, followed by dilution in 1.5 mL phosphate-buffered saline (PBS) and immediate flow cytometric analysis. Monocyte subsets were defined in accordance with contemporary nomenclature (Fig. 1) [8, 10]. Absolute monocyte subset counts (cells μL−1) were obtained by calculating the number of monocytes relative to the number of count beads according to the manufacturer's instructions.
TLR4 expression on monocyte subsets
For analysis of surface TLR4 expression, 100 μL whole blood was incubated with mouse anti-human monoclonal fluorochrome-conjugated antibodies for 15 min in the dark. Red blood cells were lysed with 2 mL Becton Dickinson lysing solution for 10 min, then washed in PBS followed by immediate flow cytometric analysis. Anti-CD16-Alexa Fluor 488 and anti-CD14-PerCP-Cy5.5 (clone M5E2; Becton Dickinson) were used for classification of monocyte subsets into CD14++ CD16− monocytes (Mon1), CD14++ CD16+ monocytes (Mon2) and CD14+CD16++ monocytes (Mon3). PE-conjugated antibodies against TLR4 (clone 285219, R&D Systems) were used to determine expression of the receptor, quantified as median fluorescence intensity (MFI).
Assessment of intracellular activation of the nuclear factor κB pathway
The intracellular level of IKKβ, a cytoplasmic marker of activation of the nuclear factor κB (NFκB) pathway, was measured from fresh whole blood [8, 11]. Briefly, 100 μL blood was incubated with monoclonal mouse anti-human antibodies against CD16-Alexa Fluor 488 and CD14-PerCP-Cy5.5 for 15 min followed by lysis of red blood cells with 2 mL PharmLyse (Becton Dickinson) for 10 min and washing in staining buffer. The resultant pellet was resuspended in fixation/permeabilization solution (Becton Dickinson) for 20 min and, following centrifugation, in 2 mL Perm/Wash buffer (Becton Dickinson) for 10 min. Following further centrifugation, the pellet was incubated for 30 min with monoclonal mouse anti-human APC-conjugated antibodies (LL-APC-XL conjugation kit; Innova Biosciences, Cambridge, UK) against IKKβ (clone 10A9B6, Abcam, Cambridge, UK), washed and resuspended in 200 μL 2% PBS/2% paraformaldehyde solution (PharmFix; Becton Dickinson) for immediate flow cytometric analysis. Levels of IKKβ were quantified by MFI.
Plasma levels of IL1β, IL6, IL10 and monocyte chemoattractant protein-1 (MCP-1) were measured by cytometric bead array technology as previously described . The FACSCalibur flow cytometer was used for data acquisition, with FCAP Array v2.0.2 software (Burnsville, MN, USA) for data analysis. Commercially available reagent sets, Human IL1β, IL6, IL10 and MCP-1 Flex Sets (all from Becton Dickinson), were used according to the manufacturer's instructions. The inter- and intra-assay coefficients of variation for all assays were <5%. The lower limits of detection were 0.3, 1.0, 0.13 and 1.3 pg mL−1 for IL-1β, IL-6, IL-10 and MCP-1 respectively .
Assessment of left ventricular function
Patients underwent echocardiographic assessment of convalescent LV ejection fraction (EF) using Simpson's method 6 weeks after STEMI by echocardiographers blinded to the study results. The inter- and intraobserver coefficients of variation in our laboratory were <5%.
Following a test of statistical normality, continuous data were expressed as mean [standard deviation] for normally distributed data or median [interquartile range] for nonnormally distributed data. Cross-sectional normally distributed data were analysed using the t-test; the Mann–Whitney test was used for nonnormally distributed data. The Friedman test with Dunn's post hoc test (where appropriate) was used for follow-up analysis of nonnormally distributed data. Correlations between the study parameters were assessed using Spearman's method for nonnormally distributed parameters. Univariate and multivariate linear regression analyses were used to establish predictive value of the study parameters for LVEF at 6 weeks in STEMI patients.
Based on our previous investigation and pilot data , the calculated minimum number of participants required to achieve 80% power to detect a difference of 0.5 SD in mean monocyte count between the study groups was 35 for the cross-sectional part of the study and 25 for the longitudinal analysis.
A P-value of <0.05 was considered statistically significant. SPSS18 (SPSS, Inc, Chicago, IL, USA) and GraphPad 4 (GraphPad Software, Inc., La Jolla, CA, USA) software were used for statistical analyses. Only STEMI and NSTEMI patients who completed follow-up were included in the longitudinal analysis.
A total of 50 patients with STEMI (mean age [standard deviation] 58  years, 86% male), 48 patients with NSTEMI (60  years, 83% male) and 40 patients with stable CAD (60  years, 83% male) were recruited; demographic and clinical details have been reported previously [11, 12]. The study groups were well matched for demographic and clinical parameters, and most concomitant medications. Glycoprotein IIb/IIIa inhibitors were administered to 96% of patients with STEMI but only 10% of those with NSTEMI. All STEMI and NSTEMI patients received clopidogrel or prasugrel, compared with only 75% of patients with stable CAD (P <0.001). Data on TLR4 expression at all four time-points were available for 27 STEMI patients.
As previously reported in STEMI patients , there were significant but numerically smaller increases in Mon1 and Mon2 count and no change in Mon3 count following NSTEMI (Fig. 2). Furthermore, a similar trend was observed in patients with NSTEMI as in those with STEMI, but not stable CAD, with a reduction in the relative proportion of Mon3 but an increase in Mon2 (Fig. 2).
There was a significant increase in TLR4+ Mon1 and Mon2 count in patients with STEMI and in TLR4+ Mon2 count only in those with NSTEMI (Table 1 and Fig. 3). There was no difference in TLR4 expression on any monocyte subset (measured as percentage or MFI) despite the study being powered to detect even small (20%) changes (Table 1). TLR4 expression remained constant during follow-up in patients with STEMI (Table 2).
Table 1. TLR4-expressing monocyte subsets in the study groups
STEMI (n =50)
NSTEMI (n =48)
CAD (n =40)
P-value STEMI vs. CAD
P-value NSTEMI vs. CAD
CAD, coronary artery disease; MFI, median fluorescence intensity; Mon, monocyte; Mon1, CD14++ CD16− CCR2+ monocytes; Mon2, CD14++ CD16+ CCR2+ monocytes; Mon3, CD14+ CD16++ CCR2− monocytes; NSTEMI, non-ST-elevation myocardial infarction; STEMI, ST-elevation myocardial infarction; TLR4, Toll-like receptor 4. Data expressed as median [interquartile range] or mean [standard deviation].
Statistical power (1 − β) to detect a minimum 20% changes in TLR4 expression.
Table 2. Monocyte TLR4 expression during follow-up in patients with ST-elevation myocardial infarction
Data expressed as median [interquartile range]. MFI, median fluorescence intensity; Mon1, CD14++ CD16− CCR2+ monocytes; Mon2, CD14++ CD16+ CCR2+ monocytes; Mon3, CD14+ CD16++ CCR2− monocytes; TLR4, Toll-like receptor 4.
TLR4 (Mon1), MFI
TLR4 (Mon2), MFI
TLR4 (Mon3), MFI
On day 1 following STEMI, troponin level correlated with TLR4+ Mon2 count (r = 0.31, P =0.044) but not with TLR4 expression on Mon2 (or any other subset, P >0.05). Plasma IL6 level correlated positively with TLR4+ Mon2 count (r = 0.54, P < 0.001), but negatively with TLR4 expression on Mon2 (r = −0.33, P =0.021). There was no significant association between TLR4-related monocyte parameters and LVEF 6 weeks post-STEMI. In addition, no significant correlation was observed between monocyte TLR4 expression and IKKβ level (P >0.05).
The results of this study show for the first time that MI is associated with increased numbers of TLR4+ monocytes, but not with higher TLR4 expression by individual monocytes. Moreover the increase in TLR4+ monocytes was attributed to specific human monocyte subsets (Mon1 and Mon2 in STEMI, and Mon2 alone in NSTEMI). Thus, we suggest that TLR4 expression may not be a reliable marker of monocyte activation in MI.
Our findings are in contrast to some previously published data on monocyte TLR4 expression in MI (see Table 3); the reason(s) for this warrant discussion. The results of all studies suggest that TLR4−bearing monocytes are upregulated, however, heterogeneity between the published studies is likely to reflect the different methodological approaches used and populations tested (Table 3). For example, the proportion of monocytes-expressing TLR4 in comparable STEMI cohorts varies from 6% to about 80%; in this study 12% of ‘classical’ monocytes expressed TLR4 in STEMI patients [13-15]. Similar variations are also seen in control samples.
Table 3. Published data on TLR4 expression in myocardial infarction
Cumulative analysis of all published data may provide explanations of these discrepancies. Firstly, the lowest proportions of TLR4-expressing monocytes are seen in studies in which fresh whole blood samples were analysed rather than in those with protocols based on prolonged sample processing (including density centrifugation, adhesion to plastic or immune-magnetic separation and in vitro culture) (Table 3). As there is consistent evidence of monocyte activation following STEMI and high responsiveness of activated monocytes to external stimuli resulting in prompt TLR4 upregulation, it is likely that monocyte TLR4 expression may be increased by sample processing, with monocytes being potentially more sensitive following STEMI due to their activated state .
A second confounding issue is related to variation in the definition of TLR4-expressing monocytes by flow cytometry. Monocytes do not form a separate cluster of TLR4-expressing cells, but rather show a continuous spectrum of cells with varying TLR4 expression and, with the exception of the Mon2 subset, the level of this expression is low. In these circumstances, definition of TLR4 positivity critically depends on the cut-off value. This is reflected in the lower proportion of TLR4+ cells in studies in which isotope controls were used (to account for nonspecific binding) compared with the use of ‘sham’ controls . Given the functionally of TLR4 as a surface coreceptor for CD14, its key role in monocyte-mediated TLR4 expression on most monocytes would be expected. Thus, TLR4 is likely to be expressed on all monocytes (or at least on Mon1 and Mon2), but at such low levels that its expression may be difficult to differentiate from that of isotype controls. Accordingly, monocyte TLR4 expression may be better assessed using MFI rather than percentage of TLR4+ cells. Finally, healthy volunteers were used as controls in several studies; therefore it is difficult to determine whether differences observed were solely due to MI or were confounded by comorbidities, cardiovascular disease risk factors and differences in drug treatment.
Of interest, there appears to be a general trend towards lower monocyte TLR4 expression in more recent studies, which might be a result of more aggressive antithrombotic therapy. This may partly explain the lack of significant differences in monocyte TLR4 expression between the study groups. Almost all STEMI patients and a fraction of NSTEMI patients received platelet glycoprotein IIb/IIIa inhibitors in this study and in addition all received dual antiplatelet therapy (aspirin plus clopidogrel or prasugrel). As monocyte–platelet interactions play a significant role in regulation of monocyte phenotype and activity, the use of antiplatelet therapy could influence monocyte TLR4 expression .
The diversity of monocyte subsets, with regard to phenotype, function and implications in cardiovascular disorders, has been increasingly recognized. Following MI, only the Mon1 and Mon2 but not Mon3 subset numbers are increased, with the most prominent changes seen in Mon2 (e.g. >2.5-fold increase in STEMI patients) . Indeed, the results of the present study show that in patients with MI managed according to contemporary guidelines, and with monocytes classified into three distinct subsets, there is an increase in TLR4+-expressing Mon1 and Mon2 following STEMI and Mon2 alone after NSTEMI due to increases in the numbers of these monocyte subsets but not to a change in TLR4 expression per se.
The relative clinical importance of monocyte count rather than the level of TLR4 expression is supported by the correlation observed between peak troponin and TLR4+ Mon2 count (similarly to the previously reported correlation with total Mon2 count) but not with TLR4 expression on any subset. Of interest, plasma IL6 level correlated positively with TLR4+ Mon2 levels, but negatively with TLR4 expression on Mon2. This observation is in agreement with previous results showing the association between this subset and high IL6 levels. By contrast, the negative association between IL6 level and TLR4 expression by other monocytes may reflect partial loss of TLR4 from activated monocytes, indicating that TLR4 expression may not be a reliable marker of monocyte activation in patients with STEMI. This possibility is also supported by the lack of any observed correlation between monocyte TLR4 expression and a marker of NFκB activation (intracellular IKKβ) as well as the lack of association between monocyte TLR4 expression and LVEF at 6 weeks post-STEMI.
Finally, as shown in Fig. 2, there is an increase in the proportion of Mon2 following STEMI (and to a smaller degree after NSTEMI) with high TLR4 expression and a decrease in the proportion of Mon3 with decreased TLR4 expression. Accordingly, an MI-related increase in TLR4 expression by all CD16+ monocytes (i.e. Mon2 and Mon3 considered as a single population) may be partly due to an increase in the proportion of Mon2 rather than a true increase in TLR4 expression per se .
The present study was mainly descriptive and observational in its nature with monocyte functional activity assessed only by intracellular IKKβ levels in venous blood. Thus, we cannot provide a detailed insight into TLR4-related monocyte pathways nor into their roles within the myocardium. In addition, surface TLR4 level may not be a comprehensive measure of monocyte TLR4 production, as certain amounts of surface TLR4 could be shed into the circulation, and thus intracellular TLR4 mRNA measurement could provide further information. In aiming to determine TLR4 expression as close to the in vivo state as possible (by immediate processing of fresh whole blood samples), we avoided the more prolonged sample processing required for monocyte (subset) isolation and RNA extraction.
Following acute MI treatment, monocyte TLR4 expression by individual subsets is unchanged. The increases in TLR4+ Mon1 and Mon2 count following STEMI and TLR+ Mon2 count alone post-NSTEMI are due to increases in monocyte subset numbers and not to changes in their TLR4 expression. Monocyte counts but not their TLR4 expression correlate positively with plasma IL6 levels. We suggest that TLR4 expression may not be a reliable marker of monocyte activation in MI.
Conflict of interest statement
The authors have no conflicts of interest to declare.