Objectives C-reactive protein (CRP), an inflammation marker, is a strong independent risk factor for cardiovascular disease. Vessels are able to express CRP; however, the molecular mechanism behind this expression is not clear.
Methods Reverse transcription PCR and ELISA were used to detect messenger RNA and proteins of CRP and nuclear factor κB (NF-κB) activity in vessel rings stretched with different mechanical strains.
Results Interleukin (IL)-6 treatment did not induce CRP expression in vessel rings of white rabbits in the absence of mechanical strain. In contrast, IL-6 augmented CRP expression in vessel rings stretched with mechanical strains of 3 and 5 g (CRP mRNA, IL-6: 11.367±1.68 and 12.78±0.76 vs vehicle: 7.27±0.88 and 8.3±0.91 folds, respectively; CRP, IL-6: 12.79±1.62 and 14.05±2.1 vs vehicle: 7.72±1.04 and 8.16±1.52 folds, respectively; p<0.05 vs 0 g group and vehicle control group; n=5), and this effect was completely blocked by treatment with gadolinium III chloride hexahydrate (GdCl3). Moreover, IL-6 treatment increased NF-κB activity in vessels stretched with a mechanical strain of 3 g, and this effect was blocked by stretch-activated channel inhibitors (streptomycin or GdCl3) and the NF-κB peptide inhibitor SN50, but not by the inactive SN50 analogue SN50M. We also performed similar experiments on human internal mammary arteries and obtained similar results.
Conclusions These results indicate that the inflammatory cytokine IL-6 alone does not induce CRP synthesis in vessels in the absence of mechanical strain; however, IL-6 augments mechanical strain-induced CRP synthesis in vessels via the stretch-activated channel–NF-κB pathway.
Statistics from Altmetric.com
C-reactive protein (CRP), an inflammation marker, is generally recognised as having a proatherogenic effect.1 ,2 It is a marker of plaque instability and in patients with acute myocardial infarction a marker of left ventricular remodelling and long-term prognosis.3 ,4 Inflammation (including non-infectious inflammation) stimulates the release of CRP from the liver. Vessels are also able to express CRP, although the molecular mechanism is still not clear.5 The relationship among CRP, cardiovascular diseases and inflammation remains controversial.6 ,7 Some investigators have proposed that atherosclerosis or some other cardiovascular diseases are inflammatory diseases and CRP is a marker of inflammation in such diseases; others consider CRP to be a causal mediator of cardiovascular disease. CRP synthesis in vessels is considered an inflammatory process.5 ,8
Inflammation (including non-infectious inflammation) increases the secretion of the cytokines interleukin (IL)-6 and IL-1β, which in turn encourage CRP expression through different signalling pathways. IL-6 plays a central role in all cells, and IL-1β cannot induce CRP expression alone.9 IL-6 mediates the binding of C/EBPβ and C/EBPγ to two sites on the CRP promoter at −53 and −219, and it mediates the binding of STAT3 to a site at −108, which consequently enhances CRP gene transcription.10
Mechanical strain on vessels of patients with cardiovascular diseases, such as hypertension and heart failure etc., is usually substantially increased. We have previously shown that CRP expression in vessels is induced under mechanical strain, which in turn activates stretch-activated channels (SAC) in the absence of inflammation.9 Vessel damage alone (non-infectious inflammation), which is usually considered the initial basis of restenosis after percutaneous coronary intervention (PCI) and atherosclerosis etc., did not increase CRP expression.11 In the current study, we investigated the relationship between CRP induction, inflammatory cytokines and mechanical strain in vessels.
Materials and methods
Reagents and solutions
Acetylcholine, prostaglandin F2α, gadolinium III chloride hexahydrate (GdCl3), and sulphate streptomycin were purchased from Sigma-Aldrich (St Louis, Missouri, USA). SN50 (a cell-permeable peptide inhibitor of NF-κB) and its derivative SN50M (a cell-permeable inactive control peptide) were purchased from Merck (Darmstadt, Germany). Recombinant human IL-6 and IL-1β were purchased from Hyclone (Gaithersburg, Maryland, USA), and diltiazem was purchased from Tanabe (Tianjing, China). Krebs–Henseleit (K–H) solution contained 118 mM NaCl, 4.76 mM KCl, 1.2 mM NaH2PO4, 1.2 mM MgSO4, 25 mM NaHCO3, 1.25 mM CaCl2 and 11 mM glucose.
All experimental protocols complied with the guide for the care and use of laboratory animals published by the US National Institutes of Health and the animal care and use committees of the hospitals and universities. Male New Zealand white rabbits (Experimental Animal Center, Guangdong, China) weighing 2.5−3.0 kg were used in the current study. All animals were housed in an alternating light/dark (12 h/12 h cycle) environment, and normal rabbit chow and water were freely available throughout the experiment. Rabbits were anaesthetised with ketamine (35 mg/kg, intramuscularly), xylazine (5 mg/kg, intramuscularly) and acepromazine (0.75 mg/kg, intramuscularly). Anaesthesia was maintained during the procedure of removing aortas for all experiments with isoflurane inhalation via a mask.
Human sample collection
Our study conformed to the ethical principals of the Declaration of Helsinki. The study protocols were approved by the ethics committee of the hospital and the university, and written informed consent was obtained from all patients. Internal mammary arteries (IMA) were collected from five patients with unstable angina pectoris (coronary stenosis ≥70% on angiography) and myocardial infarction who underwent coronary bypass surgery in our hospital. Patients with any other diseases associated with increased CRP levels, including inflammatory disorders, malignancy or infection, were excluded from the current study.
Measurement of CRP expression levels
ELISA experiments (Bioscience, Guangzhou, China) were performed to determine the protein expression levels of high sensitivity C-reactive protein (hsCRP) from vessel ring samples, as described previously.11 Cytoplasmic hsCRP levels were detected in cytoplasmic protein, which was extracted from vessel samples using a protein extraction kit (Viagene Biotech Company, Ninbo, China) according to the manufacturer's instructions, and the protein samples were stored at −80°C before analysis. All experiments were performed in triplicate.
RT–PCR analysis of CRP expression
Total RNA was isolated from vessels with trizol reagent (Invitrogen, Carlsbad, California, USA). Reverse transcription (RT)–PCR experiments were carried out using the PrimeScript RT–PCR Kit (TaKaRa Biotechnology Company, Dalian, China) as described previously.11 Total RNA was reverse-transcribed into complementary DNA using an oligo(dT) primer with PrimeScript RTase. Rabbit CRP primers (forward: 5′–GGGTGTTTGCTTGCTTAT–3′; reverse: 5′–CCCAGGAAGTCCAGGTAT–3′) were designed to amplify a 456-bp fragment of rabbit CRP. Rabbit β-actin primers (forward: 5′–GTCCTTCCTGGGCATGGAG–3′; reverse: 5′–GGCGTACAGGTCCTTGCG–3′) were designed to amplify a 98-bp fragment of rabbit β-actin. PCR reactions were performed using the following conditions: denaturation at 94°C (5 min), 24 cycles for β-actin and another more eight cycles (with new 1U Taq polymerase, total 32 cycles) for CRP consisting of 94°C (30 s), 57°C (30 s) and 72°C (1 min), and then elongation at 72°C (5 min). Amplified PCR products were analysed by 2% agarose gel electrophoresis with ethidium bromide staining and confirmed by sequencing. The resulting DNA bands were imaged using a charge-coupled device camera (UVItec Limited, Cambridge, UK) and analysed by using the ImageJ 1.37 system from Wayne Rasband (National Institutes of Health, Bethesda, Maryland, USA). CRP messenger RNA expression was normalised to β-actin mRNA expression in each sample. The primers and experimental conditions for human CRP and β-actin mRNA expression were used as previously described.12
Vascular ring preparation
The rabbit vascular rings were prepared as previously described.11 ,12 The rabbit aortas were dissected and endothelial cells were removed by gently rotating the aortic section with the tip of a forceps. Each vessel was cut into 4−6 mm ring segments, which were divided into different groups according to the study protocol. Ring segments were then suspended vertically on stainless steel hooks in a tissue chamber containing K–H solution at 37°C in an atmosphere containing 95% oxygen and 5% carbon dioxide. Mechanical strain generated by aortic smooth muscle was measured using a force transducer (JH-2, Beijing, China) and recorded using the BL-420 experimental system of biological function (Chengdu TME Technology Co. Ltd., China). Resting tension was set to 0.2 g. After 90 min of equilibration, the rings were activated with KCl (50 mM) to determine their integrity. The vascular rings with removed endothelial cells were assessed at the beginning of each experiment to show that 0.5 μM acetylcholine did not produce relaxation of rings that were precontracted with 1.5 μM prostaglandin F2α (Sigma-Aldrich). Thereafter, the vascular rings were treated with diltiazem (10 µM) for 1 h before stretching using a mechanical strain of 0 g (0 g group), 1 g (1 g group), 3 g (3 g group), or 5 g (5 g group) for 20 min. GdCl3, streptomycin, SN50 (a cell-permeable peptide inhibitor of NF-κB), and its derivative SN50M (a cell-permeable inactive control peptide) were applied to evaluate the effects of the mechanical strain-induced SAC/NF-κB pathway on aortic hsCRP expression. GdCl3, streptomycin, SN50 and SN50M were each dissolved in phosphate-buffered saline as stock solutions and diluted in K–H solution containing diltiazem (10 µM) to prepare working solutions for pretreating the vascular rings. Vascular rings in the control group were pretreated with the same volumes and concentrations of vehicle solutions. After pretreatment for 1 h, the vascular rings were stretched with a mechanical strain of 3 g and 20 pg/ml IL-6 for 20 min. Rings were then frozen in liquid nitrogen and stored at −80°C until analysis.
Measurements of NF-κB activation
NF-κB subunit (p50 and p65) expression was detected in the nuclear protein extract from the vascular rings using the NF-κB p50/p65 EZ-TFA transcription factor assay kit (Millipore Corporation, Billerica, New York, USA) as described.11 ,12 This assay combines the principles of the electrophoretic mobility shift assay with a 96-well plate-based ELISA. During the assay, the capture probe, a double-stranded biotinylated oligonucleotide containing the DNA binding consensus sequence for NF-κB (5′–GGGACTTTCC–3′), was combined with nuclear protein extracts. In this assay, the active form of NF-κB contained in the nuclear protein extract bound to its consensus sequence, and the extract/probe/buffer mixture was then directly transferred to a streptavidin-coated plate. The active NF-κB protein was immobilised on the capture probe bound to the streptavidin plate well, and inactive unbound material was washed away. The bound NF-κB transcription factor subunits (p50 and p65) were detected using specific primary antibodies. A highly sensitive horseradish peroxidase-conjugated secondary antibody was then used for sensitive colorimetric detection, which can be read in a spectrophotometric plate reader at 450 nm. The wild-type consensus oligonucleotide (non-biotinylated control) was used as a specific competitor for NF-κB binding to monitor the specificity of the assay. A mutated consensus oligonucleotide that had no effect on NF-κB binding served as the internal negative control.
All of the data were expressed as mean±SE. Differences between means were evaluated by analysis of variance and Student's t test. Coefficient of product–moment correlation was calculated to assess the correlation between aorta diameter changes and serum hsCRP expression levels. A p value of less than 0.05 was considered statistically significant. Statistical analyses were performed using Stata V.6.0 software.
The effect of IL-6 treatment and mechanical strain on CRP expression in vessels
First, we excluded the potential interference from calcium channels on stretch-induced CRP expression by pretreating vessel rings with diltiazem. As shown in figure 1A–C, diltiazem pretreatment did not affect CRP expression in vessel rings stretched by a mechanical strain of 1, 3, or 5 g. Second, we tested the effects of IL-6 on CRP expression in mechanically stretched vessel rings. Treatment with IL-6 (from 3 pg/ml to 1 ng/ml) did not affect CRP expression in vessel rings in the absence of mechanical strain (figure 2A,B). However, IL-6 (3 pg/ml) augmented CRP mRNA and protein expression levels in vessel rings stretched with 3 and 5 g (*#p<0.05; n=5, figure 3A–C). The effects of IL-6 treatment were completely blocked by treatment with 25 µM GdCl3 (*#p<0.05; n=5, figure 4A–C). Collectively, the results of these experiments showed that IL-6 treatment only induced CRP expression in vessels stretched by mechanical strain and that SAC activation was required for IL-6-mediated CRP induction.
Regulation of IL-6-induced CRP expression in stretched vessels by SAC–NF-κB pathway
As shown in figure 5A,B, IL-6 (3 pg/ml) induced CRP expression in vessels stretched with 3 g, and this expression was blocked by the NF-κB peptide inhibitor SN50 (18 μM) but not by the inactive SN50 derivative SN50M. IL-6 (3 pg/ml) treatment significantly increased the activity of the NF-κB p50 and p65 subunits in vessels stretched with 3 g. This effect was blocked by SAC inhibitors (streptomycin (200 μM) and GdCl3 (25 μM)) and by SN50 (18 μM), but not by SN50M (18 μM) (figure 5C,D).
The effect of IL-6 treatment and mechanical strain on CRP expression in human IMA
We also observed the relationship of IL-6 treatment and mechanical strain on CRP expression in human IMA. IMA were collected from five patients with a mean age of 68.7±12.8 years (table 1) who had undergone PCI because of unstable angina. Similar results were obtained. As shown in figure 6A–C, IL-6 (20 pg/ml) augmented CRP mRNA and protein expression levels in vessel rings stretched with 3 and 5 g (CRP mRNA: 9.48±0.59 and 11.09±0.74 fold, respectively; CRP protein: 10.72±1.04 and 11.86±1.32 fold, respectively; p<0.05 vs the 0 g group and the vehicle control group; n=5). The effects of IL-6 treatment were completely blocked by treatment with 25 µM GdCl3 (CRP mRNA expression in 3 and 5 g: 0.95±0.29 and 1.10±0.47 fold, respectively; CRP protein expression in 3 and 5 g: 0.97±0.27 and 1.14±0.43 fold, respectively; p<0.05 vs the 0 g group and vehicle control group; n=5) (figure 6D–F). Collectively, the results of these experiments showed that IL-6 treatment only induced CRP expression in IMA stretched by mechanical strain, and that SAC activation was required for IL-6-mediated CRP induction.
In the current study, we demonstrated that treatment with IL-6 at concentrations much higher than pathophysiological concentrations (from 3 pg/ml to 1 ng/ml) did not affect CRP expression in vessel rings in the absence of mechanical strain. The inflammatory cytokines, IL-6 and IL-1β, were reported to induce CRP expression in vessel cells,5 but the concentration used in those studies was also much higher than that detected under pathophysiological conditions.13 ,14 However, we found that a physiological concentration of IL-6 augmented mechanical strain-induced CRP expression in rabbit vessel rings, and this effect was completely blocked by SAC blockers and the NF-κB peptide inhibitor SN50, but not by the inactive SN50 derivative SN50M. In addition, we repearted some of these experiments in human IMA and obtained similar results.
SAC are one of the major classes of molecules involved in mechanosensitive signal transduction.15–17 SAC include three main types of channels: non-selective cation channels, K+-selective channels and Cl−-selective channels.16–19 SAC activation is rapid and occurs within several seconds.19 Mechanical agitation activates SAC, and they then transduce external mechanical stimuli into physiological signals. Meanwhile, calcium channels, which are abundant in vessels, affect vascular tone and contraction, and they are closely associated with mechanical strain.19 As expected, when the calcium antagonist diltiazem was applied in our system, it did not affect stretch-induced CRP synthesis.
In our studies, sustained mechanical strain on vessels for 20 min sufficiently induced CRP expression, and these CRP expression levels were of similar levels to those detected in vessels stretched for 1 or 2 h (data not shown). Vessel rings stretched with a mechanical strain of 1.5−2 g as the rest tension are usually considered to have the best physiological status for most pharmacological and physiological experiments.20 Wendling et al21 reported that 6.5 g of mechanical strain equals 80 mm Hg pressure in IMA according to the law of Laplace. The mechanical strain (<5 g) used to stretch the vessel rings in the current study would be equivalent to approximately 40–50 mm Hg of pressure on the vessels, which was much lower than the physiological pressure. However, Wendling et al21 reported that the maximal active tension in the IMA occurred at a length of 3.2±0.5 mm, which approximately corresponded to a 40% increase from the basal length of 2.3 mm.22 In our previous studies, vessel rings stretched with 3 g of mechanical strain already exhibited a 40% increase in their length.9 Furthermore, the law of Laplace for a cylinder (circumferential wall tension=pressure×radius) is not appropriate for applications related to stretched vessel rings because vessel rings are flat instead of cylindrical when they are stretched. Flat vessels under mechanical strain only exhibit diameter extension, whereas vessels in vivo exhibit circumferential extension. Our previous studies showed that CRP expression in 3 mm vessel ring segments stretched with 3 g reached a peak and then decreased with 5 g;11 however, in the current study, CRP expression did not decrease in 4–6 mm vessel ring segments stretched with 5 g. Therefore, the applied mechanical strain and width of the ring segments appear to contribute significantly to the regulation of CRP expression.
IL-6 also augmented mechanical strain-induced NF-κB P50 and P65 subunits, and this effect was also blocked by SAC inhibitors and SN50 but not by SN50M. Activated NF-κB has previously been shown to upregulate IL-6 transcription;22 however, IL-6 has also been shown to induce the phosphorylation of an inhibitor of κB (IκB) and result in NF-κB activation within several minutes.23 ,24 Therefore, we asked whether mechanical strain-induced CRP expression in vessels is dependent on the SAC-(NF-κB)–IL-6 pathway or is independent of IL-6 but dependent on the SAC–NF-κB pathway. The results of the current study clearly demonstrated that exogenous treatment with IL-6 in a physiological range had no effect on increasing NF-κB and CRP synthesis in vessels without mechanical strain-induced SAC activation. Instead, IL-6 synergised with mechanical strain to induce NF-κB and CRP expression. Taken together, the results of the current study demonstrated that IL-6-induced CRP expression occurs via the SAC–NF-κB pathway, and that mechanical strain-induced CRP expression is dependent on a synergistic interaction between IL-6 and the SAC–NF-κB pathway.
Previous research indicated that the serum hsCRP peak concentration usually occurred within 38 h after PCI, and increased hsCRP was related to cardiovascular events after PCI.25 Our work indicated that acute increased serum hsCRP after PCI might only come from livers responding to surgery-induced inflammatory reaction, but persistently increased serum hsCRP after PCI might come from complicated and multiple stenosis of vessels. The major stenosis was resolved during PCI but the remains, which might not reach the standard (<70% stenosis) for PCI but increase mechanical strain on vessels, existed.
In summary, these results indicated that IL-6 alone cannot stimulate CRP synthesis in vessels in the absence of mechanical strain, but augment mechanical strain-induced CRP synthesis in damaging vessels via the SAC–NF-κB pathway.
The authors would like to thank Dai Gang for his technical assistance with the animal models.
HZ and YL contributed equally to this manuscript.
Contributors Individual contributions include design, analysis and interpretation of data, drafting and revision of manuscript: LT; analysis and interpretation of data, drafting and revision of manuscript: YL; experiments: HZ, YL and GH; analysis of data and collection of samples: GH, XG, YL, CL and ZW; interpretation of data, revision of manuscript: BO and YZ. YL and HZ contributed equally to this work.
Funding This study was supported by funding provided by the National Natural Science Foundation of China (nos 30770897/C03030201 and 81170241 to LT).
Competing interests None.
Patient consent Obtained.
Ethics approval This study conformed to the ethical principals of the Declaration of Helsinki. The study protocols were approved by the ethics committee of the hospital and the university.
Provenance and peer review Not commissioned; externally peer reviewed.
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.