Cardiac troponin I (cTnI), a biomarker for myocardial damage and risk stratification, may be involved in the pathogenesis of cardiovascular diseases, which was ascribed to the effect of cTnI auto-antibodies. Whether or not cTnI itself has a direct impact on acute myocardial injury is unknown. To exclude the influence of cTnI antibody on the cardiac infarct size, we studied the effect of cTnI shortly after myocardial ischaemia–reperfusion (I/R) injury when cTnI antibodies were not elevated. Pretreatment with cTnI augmented the myocardial infarct size caused by I/R, accompanied by an increase in inflammatory markers in the blood and myocardium. Additional experiments using human umbilical vein endothelial cells (HUVECs) showed that the detrimental effect of cTnI was related to cTnI-induced increase in vascular cell adhesion molecule-1 (VCAM-1) expression and VCAM-1 mediated adhesion of human monocytes (THP-1) to HUVECs, which could be neutralized by VCAM-1 antibody. Both toll-like receptor 4 (TLR4) and nuclear factor-κB (NF-κB) were involved in the signalling pathway, because blockade of either TLR4 or NF-κB inhibited the cTnI's effect on VCAM-1 expression and adhesion of monocytes to endothelial cells. Moreover, TLR4 inhibition reduced cTnI-augmented cardiac injury in rats with I/R injury. We conclude that cTnI exacerbates myocardial I/R injury by inducing the adhesion of monocytes to vascular endothelial cells via activation of the TLR4/NF-κB pathway. Inhibition of TLR4 may be an alternative strategy to reduce cTnI-induced myocardial I/R injury.
For CHD, there are many classic risk factors in the development of atherosclerosis and coronary artery occlusion, like dislipidaemia and diabetes, but there is no risk factor for the mal-progression after the acute insult. There are risk factors that hinder an acute recovery. Previous studies have suggested that cTnI may be involved in the pathogenesis of CHD caused by its auto-antibodies. Our study was aimed to investigate whether or not cTnI, in the acute phase of myocardial injury, when auto-antibodies have not been produced, can further damage the myocardium.
Our study demonstrated that cTnI exacerbates myocardial I/R injury by inducing adhesion of monocytes to vascular endothelial cells, via activation of the TLR4/NF-κB pathway.
Our study suggests that cTnI and the TLR4/NF-κB pathway activated by cTnI may be potential targets of intervention in preventing secondary myocardial damage that could translate into improved clinical outcomes.
Troponin, a regulatory complex in the thin filaments of striated muscles, comprises of three subunits, troponin C, T and I [1–3]. Cardiac troponin (cTn) T and I are structurally different from their skeletal muscle counterparts. Circulating cTn T and I isoforms are specific and sensitive markers of cardiac injury, superior to other biomarkers (e.g. creatine kinase) [1–3]. However, whether cTn is merely a result or also the cause of cardiac injury is unclear.
Previous studies have shown that cardiac troponin I (cTnI) may be involved in the pathogenesis of heart disease, especially in coronary heart disease (CHD). A community-based study, consisting of 1203 70-year-old males and followed for up to 10.4 years, showed that cTnI predicted death and first CHD event in those free from cardiovascular diseases at baseline . Moreover, cTnI, by producing auto-antibodies, has been suggested to be involved in the pathogenesis of myocarditis, dilated cadiomyopathy and heart failure [5–7]. Göser and colleagues reported that mice immunized with cTnI, but not cTnT, developed severe myocarditis with increased inflammatory proteins, chemokines and chemokine receptors that was followed by cardiomegaly, myocardial fibrosis, decreased myocardial contractility and increased mortality . Patients, who developed cTnI antibodies after acute myocardial infarction, did not get an increase in left ventricular ejection fraction and stroke volume compared with those who did not develop cTnI antibodies in a follow-up period of 6–9 months . Meanwhile, in patients with acute coronary syndrome, antibodies against cTnI interfere with the immunoassays for cTnI, resulting in apparent negative results, 6–12 h after the myocardial injury . We wondered whether elevated cTnI level per se, rather than the cTnI auto-antibody, worsened the cardiac function. In the present study, we examined the effect of cTnI on cardiac infarct size in the acute phase of a myocardial ischaemia–reperfusion (I/R) rat model, at a time-window before the development of cTnI antibody. The underlying mechanisms were investigated in human umbilical vein endothelial cells (HUVECs) and a human monocyte cell line, THP-1.
MATERIALS AND METHODS
Male Sprague–Dawley (SD) rats that weighed 220–250 g (8 weeks old) were used for the experiments. The rats had free access to tap water and standard rat chow and were kept on a 12-h light/dark cycle. All the experimental procedures were approved by the Animal Care and Use Committee of Third Military Medical University, China.
Human cTnI, mouse anti-human TNNI3 monoclonal antibody, mouse IgG, nuclear factor-κB (NF-κB) inhibitor PDTC, phosphatidyl inositol 3-kinase inhibitor (PI3K) LY294002, lipopolysaccharide (LPS), tumour necrosis factor-α (TNF-α) and mitogen-activated protein kinase (MAPK) inhibitor PD 98059 were obtained from Sigma. Human albumin (HA) was purchased from Shanghai Institute of Biological Products. Toll-like receptor 4 (TLR4) inhibitor TAK-242 was obtained from Merck. The fluorescent dye 2′,7′-bis-(2-carboxyethyl)-5(and -6)-carboxyfluorescein acetoxymethyl ester (BCECF-AM) was obtained from Life Technologies. Protein kinase C (PKC) inhibitor peptide 19–31 was obtained from Calbiochem. Goat anti-cTnI polyclonal antibody, rabbit anti-VCAM-1 polyclonal antibody, TLR4 siRNA (siRNA is targeting the rat TLR4 gene sequence) and control siRNA-A were purchased from Santa Cruz Biotechnology. Rabbit anti-NF-κB P65 polyclonal antibody, mouse anti-phospho-I-κBα polyclonal antibody, rabbit anti-I-κBα polyclonal antibody, anti-histone H3 polyclonal antibody and rabbit anti-β-actin polyclonal antibody were bought from Cell Signaling Technology. All other chemicals were from Sigma Chemicals.
Myocardial ischaemia–reperfusion model and interventions
Anaesthesia was induced by an intraperitoneal (i.p.) injection of sodium pentobarbital (50 mg/kg). Then, the rats were placed on a heating pad (37°C) to maintain their body temperature. After checking for a non-response to pinching of the paw, a thoracotomy was performed at the 4th and 5th left intercostal space to expose the heart. The left anterior descending (LAD) coronary artery, 2–3 mm below the left atrium, was exposed and temporarily ligated by a 6-0 silk suture. Thirty minutes later, the ligation was loosened to allow reperfusion which was confirmed by an epicardial hyperaemic response. Then, the incisions were sutured and the rats were allowed to recover with free access to rat chow and water for an indicated reperfusion period, according to different experimental protocols (Figure 1A).
cTnI pretreatment increases myocardial infarction size in I/R-injured heart
For cTnI intervention, cTnI (10 μg/kg in normal saline) was injected into the left external jugular vein through a small skin incision 10 min before the ligation of the LAD coronary artery. For the intervention using the TLR4 inhibitor, TAK 242 (3 mg/kg in DMSO)  was given intravenously 30 min before the cTnI injection. Sham-operated animals, which served as surgical controls, were subjected to the same surgical procedures except that the LAD coronary artery was not ligated.
After the reperfusion period, the rats were re-anaesthetized by the intraperitoneal injection of sodium pentobarbital (100 mg/kg). As much blood as possible was obtained quickly from the abdominal aorta, right after the abdominal laparotomy. The heart was quickly obtained and rinsed in ice-cold PBS, and then the left ventricle was transected at the ligation site. The apical portion was either preserved at −80°C for subsequent analysis or fixed in 4% (v/w) paraformaldehyde PBS solution for haematoxylin and eosin (H&E) staining.
Evaluation of myocardial infarct size
For evaluation of infarct size, the sutures were loosened at the ligation site after 30 min of ischaemia. At the end of the 6 h reperfusion, rats were re-anesthetized by the intraperitoneal injection of sodium pentobarbital (100 mg/kg) and as much blood as possible was collected from the abdominal aorta, as described above. After re-occlusion of the LAD with the remaining sutures, the coronary arteries were perfused with 2% Evans blue (EB), retrogradely via the aorta cannulated with a PE-50 tubing, in order to differentiate the non-ischaemic area (blue) from the ischaemic area (area at risk, AAR) which had no EB stain. The hearts were excised and rinsed in PBS; the left ventricles were trimmed and frozen at −20°C for 30 min and then sliced into six 2 mm transverse sections from the apex to the base. The slices were then incubated in 1% 2,3,5-triphenyltetrazolium chloride (TTC) in PBS for 20 min at 37°C to differentiate viable tissue (brick red) and infarct (INF), which had no TTC staining. Finally, the slices were fixed with 4% paraformaldehyde for 24 h before both sides were digitally photographed. The AAR and INF on both sides were delineated and calculated using the ImageJ software (National Institutes of Health). The cumulative area for all sections of each heart was used for comparison. The percentage of infarct size was calculated as the ratio of INF to AAR. Individuals conducting the experiments were blinded to the experimental groups .
Serum was separated from blood cells by centrifugation at 1200 g for 10 min and stored at −80°C until use. Serum creatine kinase-MB (CK-MB), lactate dehydrogenase (LDH), TNF-α and interleukin-1β (IL-1β) were quantified using commercially available kits (Jiancheng Bioengineering Institute, Nanjing, China). Serum vascular cell adhesion molecule-1 (VCAM-1), cTnI and cTnI-antibody were measured using enzyme-linked immunosorbent assay kits (VCAM-1 assay kit from R&D Systems, Minneapolis, MN; cTnI and cTnI-antibody assay kits from Westang Bioscience). Myocardial myeloperoxidase (MPO) activity was measured using an assay kit from Jiancheng Bioengineering Institute, then corrected by the protein concentration of myocardial tissue, which was determined using a BCA protein assay kit (Beyotime).
Morphological evaluation of myocardial neutrophil accumulation
Myocardial structure and neutrophil accumulation were examined using H&E staining. After fixation for 48–72 h, the apical portion of the heart, under the ligation site, was dehydrated, paraffin-embedded, transversely sliced into 4 μm sections, de-paraffinized, and then stained with H&E. Polymorphonuclear neutrophils (PMNs) infiltrating the myocardial tissue were counted in 6 high power fields (×400) across a section of three independent sections per sample and five samples per group in a blinded manner. The counts were averaged among three sections per sample for comparison .
Cell culture and treatment
Human monocyte cell line THP-1 [13,14], HUVECs  and rat myocardial cell line H9c2 were obtained from the A.T.C.C. and cultured as described. THP-1 cells in RPMI 1640 medium (Gibco) and HUVECs and H9c2 cells in Dulbecco's modified Eagle's medium (DMEM, Gibco), all supplemented with 10% (v/v) FBS, were cultured at 37°C in humidified 95% air/5% CO2 incubator. When the cells were 70–80% confluent, they were digested with 0.25% trypsin and subcultured in fresh culture medium (CM). All cells were rendered quiescent in serum-free CM for 2 h before the experiments.
H9c2 cells were subjected to anoxia/reoxygenation (A/R) by exposure to an anoxic chamber filled with 5% CO2 and 95% N2 at 37°C for 6 h and followed by reoxygenation for 3 h . Then, the CM was collected for measurement of LDH and cTnI. 70–80% confluent HUVEC monolayers in 12-well plates were treated with CM from A/R H9c2 cells (A/R CM, containing approximately 3.7 ng/ml of cTnI) or human cTnI (5 ng/ml) or other indicated reagent for the indicated times before monocyte adhesion assay. For blockade of cTnI in A/R CM, goat anti-cTnI polyclonal antibodies suitable for rat species (50 ng/ml) were incubated with A/R CM at 4°C overnight before treatment of HUVECs. For neutralization of human cTnI, mouse anti-human cTnI monoclonal antibody (1 μg/ml) or mouse IgG (1 μg/ml) as control was incubated with human cTnI at 4°C overnight before treatment of HUVECs.
Monocyte adhesion assay
THP-1 cells were labelled with 10 μg/ml of fluorescent dye BCECF-AM for 30 min. After the dye-labelling was stopped with RPMI 1640, the cells were suspended in Medium 199 (Gibco). After the indicated treatment, the HUVECs were washed three times with serum-free culture medium. The dye-labelled THP-1 cells (2×105) were added to the HUVECs and incubated at 37°C in a humidified 95% air/5% CO2 incubator for 60 min. Non-adhered cells were removed by gentle washing with RPMI 1640 three times, and the number of adhered green fluorescent THP-1 cells were counted under an inverted fluorescence microscope (excitation at 490 nm) .
Measurement of VCAM-1 in culture medium
Serum-starved HUVECs were incubated with the indicated concentrations of cTnI for 24 h. Soluble VCAM-1 concentration in the culture medium was measured [18,19], using an enzyme-linked immunosorbent assay kit (R&D Systems). Values were expressed as ng per mg of total cell protein, which was determined by BCA assay.
Transfection of HUVECs with TLR4 siRNA
Knockdown of the TLR4 with TLR4 siRNA was accomplished in HUVEC cells (70–80% confluent HUVEC monolayers in 12-well plates) by transfection with 10 nmol/l TLR4 siRNA or scrambled siRNA (scRNA) for 48 h, using Lipofectamine 2000 reagent (Invitrogen) . The efficiency of the TLR4 siRNA transfection was determined by RT-PCR, as previously reported [20,21]. Primer sequences of TLR4 and β-actin used for RT-PCR were: 5′-TGTGGCTCACAATCTTATCCA-3′ (forward) and 5′-CTAAATGTTGCCATCCGAAA-3′ (reverse) for TLR4; 5′-CGTGCGTGACATTAAGGAGA-3′ (forward) and 5′-ATACTCCTGCTTGCTGATCCA-3′ (reverse) for β-actin.
After treatment with the indicated reagent, the HUVECs were washed twice with ice-cold PBS and lysed in HEPES buffer (20 mM, pH 7.4) containing 150 mM NaCl, 100 mM NaF, 10 mM EDTA, 10 mM Na4P2O7, 2 mM sodium orthovanadate, 1 mM PMSF, 2 mg/ml aprotinin, for 15 min. The lysates were collected and their protein concentrations were determined by BCA assay. To obtain the nuclear fraction, commercial nuclear and cytoplasmic protein extraction kit from Beyotime (Shanghai, China) was used according to manufacturer's instructions. Protein concentrations were adjusted to the same final protein concentration using the lysis buffer. Fifty micrograms of protein were electrophoresed in SDS/PAGE and transferred on to nitrocellulose membrane. Non-specific binding sites were blocked with 5% non-fat milk powder in TBS with 0.05% Tween 20 (TBST) for 1 h at room temperature, then incubated overnight at 4°C with the indicated primary antibodies at appropriate dilutions: rabbit anti-VCAM-1 antibody (1:500), rabbit anti-NF-κB P65 antibody (1:1000), mouse anti-phospho-I-κBα antibody(1:1000), rabbit anti-I-κBα antibody(1:1000), rabbit anti-histone H3 antibody(1:1500) and β-actin (1:600). Thereafter, the membranes were washed three times with TBST and incubated with the indicated secondary antibodies [goat anti-rabbit IRDye 800 (1:15000) and donkey anti-mouse IRDye 800 (1:15000) (Li-Cor Biosciences)] for 1 h at room temperature. The protein bands on the nitrocellulose membranes were captured by the Odyssey Infrared Imaging System (Li-Cor Biosciences) and analysed using Quantity One software. Densitometric intensity corresponding to each band was normalized against β-actin or histone H3 .
Translocation of NF-κB p65 from the cytoplasm to the nucleus was evaluated using the Cellular NF-κB Translocation Kit (Beyotime) . Briefly, HUVECs were fixed with 2% paraformaldehyde for 30 min and treated with 0.2% Triton X-100 in TBS for another 30 min. Then, the treated HUVECs were incubated with 1% BSA for 1 h to block non-specific binding and then washed with TBST for 15 min. Thereafter, the HUVECs were incubated with the primary rabbit anti-NF-κB p65 antibody (1:200) for 1 h at room temperature, followed by Cy3-conjugated anti-rabbit IgG (1:1000) (Millipore) for 1 h. Nuclei were stained finally with DAPI for 5 min at room temperature. The slides were viewed under a fluorescence microscope (Eclipse Ti-U, Nikon Corporation) at an excitation wavelength of 405 nm for DAPI and 543 nm for NF-κB p65.
EMSA was performed using the Light-shift Chemiluminescent EMSA Kit (Pierce Chemical) . This non-isotopic method to detect DNA–protein interaction that includes an enhanced luminol substrate for horseradish peroxidase with optimized blocking and washing buffers has a sensitivity equivalent to radioactive phosphorus (32P). The NF-κB consensus DNA double-strand oligonucleotide probe (5′-CTGCCCTGGGTTTCCCCTTGAAGGGATTTCCCTCCGCC-3′), containing the sequence of rat VCAM-1 gene promoter, was labelled with biotin and incubated with the nuclear extracts. After the reaction, the DNA–protein complexes were subjected to a 6% native PAGE and transferred on to a nylon membrane (Millipore). Then, the membrane was immediately cross-linked for 15 min on a UV transilluminator. The chemiluminescence was detected using a luminal/enhancer solution and stable peroxide solution, as described by the manufacturer. The membrane was then exposed to X-ray films from 3 min before development.
The data were analysed by SPSS 13.0 and expressed as mean ± S.E.M. Comparison among groups of more than 2 was made by one-way ANOVA with Bonferroni post-hoc test. P<0.05 was considered statistically significant.
cTnI increases myocardial infarction with an increase in inflammation in I/R injured heart
To determine any effect of cTnI on the myocardial infarction, first we had to distinguish between the contributions of cTnI, itself and cTnI-induced antibody. In the primary immune response, which occurs against an antigen encountered for the first time, several days are required for the clonal expansion and differentiation of naive lymphocytes into effector T-cells and IgG antibody-secreting B-cells . We measured the serum cTnI antibody concentrations before and at different reperfusion times in the rat and found that the serum concentration of cTnI antibody at 3 and 6 h of reperfusion was similar to the control pre I/R and tended to increase after 12 h of reperfusion, reaching significance at 24 h (Supplementary Table S1). By contrast, serum cTnI, as well as LDH and CK-MB, had already increased at 3 h of reperfusion. In order to choose the proper time-point to safely exclude the influence of cTnI antibody on the extent of cardiac infarction and at the same time allow the injured myocardium to fully develop into necrosis in AAR after I/R, in the present study, we chose 6 h of reperfusion time as an observation time-point. Compared with the I/R group, pretreatment with human cTnI (10 μg/kg) worsened the cardiac infarction, determined by infarct size (Figures 1B and 1C) and cardiac injury biomarkers (cTnI, CK-MB and LDH) at 6 h of reperfusion (Figures 1D–1F). HA (10 μg/kg), used instead of human cTnI, as control, had no effect on the above indices.
To investigate the mechanisms underlying the worsening effect of cTnI on myocardial I/R injury, we measured markers of inflammation in the serum and heart. We found that the serum inflammatory factors were increased in cTnI-pretreated I/R rats, as compared with I/R rats. Those inflammatory factors included neutrophil and monocyte activation marker MPO, pro-inflammatory cytokines TNF-α and IL-1β, and adhesion factor VCAM-1 (Figures 2A–2D). PMN in the myocardium were also observed to be increased by H&E staining (Figures 2E and 2F).
cTnI pretreatment increases inflammatory markers in heart and serum of myocardial I/R-injured rat
cTnI increases adhesion of monocytes to HUVECs
Since the number of inflammatory cells was increased in the human cTnI-treated I/R hearts, we wondered if endogenous cTnI could promote the adhesion of monocytes to endothelial cells. Rat myocardial cell line H9c2 cells were subjected to A/R to induce cell injury and release of endogenous cTnI; these cells had impaired cell viability observed under an inverted microscope (Figure 3A). We found that LDH and cTnI concentrations in the culture medium of H9c2 cells were increased after A/R injury (Figures 3B and 3C). Incubation of HUVECs with the culture medium from the A/R H9c2 cells increased the adhesion of THP-1 cells, a monocyte cell line, as indicated above. The effect was abolished by co-incubation with goat anti-cTnI polyclonal antibody (50 ng/ml), neutralizing the endogenous cTnI in the culture medium from A/R H9c2 cells (Figures 3D). Since HUVECs and THP-1 are of human origin, to determine the specificity of the effect of cTnI, we used the commercial cTnI purified from human to treat HUVECs and found that human cTnI increased the adhesion of THP-1 cells to HUVECs in concentration- and time-dependent manners (Figures 3E and 3F). The stimulatory effect began to be evident at the cTnI concentration of 2.5 ng/ml (24 h) (Figure 3E) at 6 h (5 ng/ml), peaked at 12 h and lasted at least for 24 h (Figure 3F). In the concentration response study (Figure 3E), LPS (100 ng/ml) and TNF-α (10 ng/ml) were used as positive controls [26–28]. Mouse anti-human cTnI monoclonal antibody (1 μg/ml for 24 h), by itself, had no effect, but blocked the stimulatory effect of cTnI (5 ng/ml for 24 h) on the adhesion of THP-1 cells to HUVECs. Mouse IgG was used in this experiment to serve as negative control. Results showed that mouse IgG by itself had no effect, and also did not affect cTnI-induced adhesion of monocytes to HUVECs (Figure 3G).
cTnI increases adhesion of THP-1 cells to HUVECs
cTnI increases VCAM-1 protein expression in HUVECs
It is known that adhesion factors are involved in the adhesion of monocytes to endothelial cells. Therefore, we wondered if cTnI increased VCAM-1, a protein that increases the adhesion of white blood cells to endothelial cells. We found that cTnI increased VCAM-1 protein in HUVECs in concentration- and time-dependent manners (Figures 4A and 4B). Soluble VCAM-1 in the culture medium was also increased in a concentration-dependent manner (Figure 4C). Blockade of VCAM-1 with VCAM-1 antibodies (10 μg/ml) abrogated the cTnI (5 ng/ml for 24 h)-mediated adhesion of THP-1 cells to HUVECs (Figure 4D).
cTnI induces VCAM-1 protein expression in HUVECs
Role of TLR4 and NF-κB in cTnI-induced monocyte-endothelial cell adhesion and increase in VCAM-1 expression
To further explore the underlying mechanisms involved in the cell adhesion-promoting effect of cTnI and increase in VCAM-1 protein expression in HUVECs, inflammation-related signalling pathways were screened. Using confocal microscopy, we found that exogenous cTnI mainly localized on the membrane surface of HUVECs, not in the cytosol (Supplementary Figure S1). The cTnI-induced adhesion of THP-1 cells to HUVECs was markedly attenuated in the presence of TLR4 inhibitor (TAK-242, 2 μM) or NF-κB inhibitor (PDTC, 20 μM) (Figures 5A and 5B). By contrast, inhibitors of other inflammation-related signalling molecules, including a PKC inhibitor (PKCI), peptide 19–31 (5 μM), a PI3K inhibitor, LY294002 (30 μM) and a MAPK inhibitor, PD 98059 (20 μM), had no effect (Figure 5B). Moreover, silencing of TLR4 with TLR4 siRNA (Figure 5C) also markedly attenuated the adhesion-promoting effect of cTnI (Figure 5D), accompanied by reduced VCAM-1 expression in HUVECs (Figure 5E). Either TAK-242 or PDTC could attenuate the increase in VCAM-1 protein expression in HUVECs induced by cTnI (Figure 5F), indicating that TLR4 and NF-κB are involved in the cTnI-induced monocyte-endothelial cell adhesion and cTnI-increased VCAM-1 expression in HUVECs.
Role of TLR4 and NF-κB in cTnI-induced monocyte-endothelial cell adhesion and VCAM-1 expression in HUVECs
To further evaluate the role of TLR4 and NF-κB signalling pathway in the cTnI-induced monocyte-endothelial cell adhesion, we studied the effect of cTnI on the protein expression of NF-κB p65, phospho-I-κBα and total I-κBα in HUVECs. We found that cTnI (5 ng/ml) increased the expression of NF-κB p65 in the nucleus (Figure 6A) and phospho-I-κBα in the cytosol (Figure 6B), and decreased the expression of total I-κBα in the cytosol (Figure 6C), in a time-dependent manner. Moreover, the translocation of NF-κB from the cytosol to the nucleus in HUVECs promoted by cTnI was observed by immunofluorescence microscopy (Figure 6D). EMSA showed that cTnI treatment for 30 min increased the binding of NF-κB to VCAM-1 promoter (Figure 6E). Addition of 50-times-molar excess of unlabelled probe eliminated the NF-κB band, confirming the specificity of the binding (results not shown). Both the translocation and binding of NF-κB to VCAM-1 promoter were attenuated either by TAK-242 or PDTC, consistent with the role of TLR4 and NF-κB in the cTnI-induced increase in VCAM-1 expression in HUVECs.
cTnI activates TLR4–NF-κB pathway in HUVECs
Inhibition of TLR4 attenuates cTnI-induced augmentation of myocardial damage in I/R-injured rats
As we found that both TLR4 and NF-κB are involved in the cTnI-induced increase in VCAM-1 expression and adhesion of monocytes with endothelial cells, and it is known that TLR4 is upstream of NF-κB in activating pro-inflammatory molecules, though NF-κB can also be activated by other factors, we wondered if inhibition of TLR4 could reduce the cTnI-augmented myocardial damage in vivo. Therefore, the TLR4 inhibitor TAK-242 (3 mg/kg in DMSO) was injected intravenously 30 min before cTnI pretreatment of myocardial I/R-injured rats (Figure 1C). Compared with the cTnI group, TAK-242 reduced the cTnI-induced increase in infarct size (Figures 7A and 7B) and lowered the increased plasma levels of cardiac injury biomarkers (LDH and CK-MB) 6 h after myocardial I/R injury (Figures 7C and 7D). TAK-242 also markedly reduced the cTnI-induced increase in myocardial MPO and circulating levels of TNF-α, IL-1β and VCAM-1 in myocardial I/R-injured rats, as compared with cTnI group (Supplementary Figure S2).
Inhibition of TLR4 attenuates cTnI-augmented myocardial damage in I/R-injured rats
cTnT and I are the most specific and sensitive laboratory markers of myocardial injury, and therefore have replaced CK-MB as the gold standard for the diagnosis of acute myocardial infarction [1–3]. Moreover, the remarkable clinical value of cardiac troponins lies in its superior prognostic value in patients with unstable [1–3] and stable angina pectoris  and even in healthy elderly males . However, the precise impact of the high levels of circulating troponin, per se, on these subjects is not clear. Previous studies demonstrated that the auto-antibodies of cTnI induced severe autoimmune inflammation in mouse myocardium , resulting in exacerbation of I/R injury , heart failure and death . These studies could be taken together to indicate that antibodies against cTnI may have a detrimental effect on the myocardium. However, it should be noted that not everyone with elevated circulating cTnI has increased cTnI antibody; cTnI auto-antibodies were detected only in 9.2 or 18.4% of patients with ischaemic cardiomyopathy [8,32]. In experimental coxsackievirus B3 myocarditis, auto-antibodies against cardiac troponins, released on day 5, were not detected until 2 weeks after infection . It is not known if cTnI, itself, causes myocardial injury, in acute or chronic heart disease. Although there is an in vitro experiment showing the protective effect of cTnI against hypoxia/reoxygenation-induced myocardial cell injury , the direct effect of cTnI on the heart in vivo is unknown. Our present study found an adverse effect of cTnI on acute cardiac injury; cTnI worsened cardiac I/R injury in rats and the detrimental effect of cTnI has nothing to do with the anti-cTnI antibodies, which we confirmed in both in vivo and in vitro studies. Although the reasons for the difference between our finding and that of others are not clear, the increased inflammatory level, the stimulatory effect of cTnI on VCAM-1 expression in HUVECs and promotion of monocyte adhesion to the I/R injured heart may explain the in vivo and in vitro studies of others .
Adherence factors, including VCAM-1, are critical for the transmigration of leucocytes out of blood vessels and their arrest in tissues [27,28], and play an important role in the pathogenesis of inflammation. VCAM-1 is constitutively present on endothelial cells [35,36]. Soluble VCAM-1 is present in various body fluids [18,19], with elevated levels in patients with atherosclerosis [37,38]. We found that cTnI increased VCAM-1 in HUVECs and also increased the adhesion of monocytes to HUVECs. The cTnI-induced expression of endothelial VCAM-1 could promote more leucocytes to penetrate the artery into the injured heart and augment the cardiac injury from I/R. This could account for the increased number of inflammatory cells in the I/R-injured heart after treatment with cTnI.
TLRs are pattern recognition receptors which are bound and activated by a range of pathogen-associated molecular patterns, including bacterial and viral products [39,40]. Growing body of evidence suggests that TLR signalling is elicited in the absence of infection through endogenous ligands (damage-associated molecular patterns, DAMPs) generated at sites of tissue remodelling and inflammation [41,42], and is already implicated in myocardial damage resulting from myocarditis  and I/R injury . Among the TLRs, TLR4 and TLR2 are most prevalent in the myocardium. Although TLR4 has been shown to play a detrimental role in myocardial I/R injury, the effect elicited by TLR2 activation remains controversial . Endothelial cells are the major sources of TLR4 in arterial thrombosis . We deduced that cTnI, an endogenous component released from myocardiocytes, may play as a DAMP ligand which could be recognized by TLR4, but this hypothesis needs to be confirmed in future research. Recently, Bangert and colleagues reported that high-mobility group box 1 (HMGB1) was increased in TnI-induced experimental autoimmune myocarditis . HMGB1 acts as a DAMP signal recognized by TLRs after passive release into the extracellular milieu during cell death or active secretion by mononuclear and other cell types [47,48]. HMGB1 may also be involved in the pathogenesis of disease caused by cTnI per se.
To further investigate the mechanisms of cTnI-mediated up-regulation of VCAM-1 expression in HUVECs, we observed the localization of cTnI in HUVECs and it resulted, after treatment with cTnI for 6 h, that exogenous cTnI was mainly concentrated on the membrane surface of HUVECs, not in the cytosol. Based on the knowledge that TLR4 is located in cell membrane , and the mechanisms of TLR4-mediated inflammatory responses by deleterious factors, like lipopolysaccharide and heat-shock proteins, exist direct and indirect ways [41,47,48], we speculate that it might be by the activation of TLR4 in the cell membrane that cTnI increased the VCAM-1 expression and monocyte-endothelial cell adhesion, and therefore up-regulated the inflammatory responses. Whether or not cTnI directly or indirectly interacts with TLR4 needs to be determined in the future.
NF-κB is a pivotal molecule downstream of the TLR-MyD88-dependent signalling pathway in activating the transcription of pro-inflammatory cytokines, chemokines and adhesion molecules. The role NF-κB in promoting inflammation is well recognized [49,50]. The prototypic inducible form of NF-κB is the heterodimer of NF-κB1 (P50) and Rel A (P65) combined with the inhibitory protein I-κB in the cytoplasm. I-κB can be phosphorylated by a number of incoming signals from the cell surface, causing self-degradation, releasing NF-κB to translocate to the nucleus and bind to the NF-κB motif of target genes, including cell adhesion molecules and cytokines involved in the inflammatory response. NF-κB serves as a signal upstream of VCAM-1, as many studies have shown that pro-inflammatory cytokines increase VCAM-1 expression, via activation of NF-κB [51–54]. Our present study showed that cTnI activated NF-κB by increasing I-κB phosphorylation, hence increasing the expression of VCAM-1 and adhesion of monocytes to endothelial cells. Furthermore, we confirmed the cTnI-induced NF-κB nuclear translocation by immunofluorescence staining and the binding to VCAM-1 promoter by EMSA.
The involvement of TLR4 and NF-κB induced by endogenous cTnI may provide therapeutic modalities, since treatment with inhibitor of TLR4 could significantly reduce the cTnI-induced increase of inflammation and infarct size in I/R-injured heart, though other methods like cTnI neutralization or NF-κB blockade may also have some prospects.
In summary, we demonstrated that cTnI is not only an outcome and biomarker of myocardial damage, but also an endogenous pathological agent involved in the augmentation of cardiac infarction and inflammation in I/R-injured heart, via a TLR4-NF-κB-dependent pathway. cTnI and the TLR4/NF-κB pathway activated by cTnI may be potential targets of intervention in preventing secondary myocardial damage. In our study, blockade of TLR4 reduced the harmful effect of cTnI in myocardial I/R-injured rats, which may shed light on its translation, if any, into improved clinical outcomes.
Yu Han and Xiang Liao performed experiments, analysed data and drafted the article. Zhao Gao, Sufei Yang, Caiyu Chen and Yukai Liu performed experiments. Yu Han and Chunyu Zeng conceived and designed the study. Wei Eric Wang and Xiongwen Chen made contributions to the experimental design. Pedro Jose, Xiongwen Chen, Ye Zhang and Chunyu Zeng edited and revised the article. All authors contributed to the discussion and revision of the manuscript.
This work was supported by the National Natural Science Foundation of China [grant numbers 31430043 and 31471089]; and the National Basic Research Program of China [grant numbers 2012CB517801 and 2013CB531104].
area at risk
coronary heart disease
cardiac troponin I
damage-associated molecular pattern
haematoxylin and eosin
human umbilical vein endothelial cell
left anterior descending
mitogen-activated protein kinase
phosphatidyl inositol 3-kinase
protein kinase C
toll-like receptor 4
tumour necrosis factor-α
vascular cell adhesion molecule-1
These two authors contributed equally to this work.