OUP user menu

Downregulation of miR-133 and miR-590 contributes to nicotine-induced atrial remodelling in canines

Hongli Shan, Yong Zhang, Yanjie Lu, Ying Zhang, Zhenwei Pan, Benzhi Cai, Ning Wang, Xuelian Li, Tieming Feng, Yuan Hong, Baofeng Yang
DOI: http://dx.doi.org/10.1093/cvr/cvp130 465-472 First published online: 27 April 2009

Abstract

Aims The present study was designed to decipher molecular mechanisms underlying nicotine's promoting atrial fibrillation (AF) by inducing atrial structural remodelling.

Methods and results The canine model of AF was successfully established by nicotine administration and rapid pacing. The atrial fibroblasts isolated from healthy dogs were treated with nicotine. The role of microRNAs (miRNAs) on the expression and regulation of transforming growth factor-β1 (TGF-β1), TGF-β receptor type II (TGF-βRII), and collagen production was evaluated in vivo and in vitro. Administration of nicotine for 30 days increased AF vulnerability by ∼eight- to 15-fold in dogs. Nicotine stimulated remarkable collagen production and atrial fibrosis both in vitro in cultured canine atrial fibroblasts and in vivo in atrial tissues. Nicotine produced significant upregulation of expression of TGF-β1 and TGF-βRII at the protein level, and a 60–70% decrease in the levels of miRNAs miR-133 and miR-590. This downregulation of miR-133 and miR-590 partly accounts for the upregulation of TGF-β1 and TGF-βRII, because our data established TGF-β1 and TGF-βRII as targets for miR-133 and miR-590 repression. Transfection of miR-133 or miR-590 into cultured atrial fibroblasts decreased TGF-β1 and TGF-βRII levels and collagen content. These effects were abolished by the antisense oligonucleotides against miR-133 or miR-590. The effects of nicotine were prevented by an α7 nicotinic acetylcholine receptor antagonist.

Conclusion We conclude that the profibrotic response to nicotine in canine atrium is critically dependent upon downregulation of miR-133 and miR-590.

KEYWORDS
  • Nicotine
  • Atrial fibrosis
  • MiR-133
  • MiR-590

1. Introduction

A large number of people smoke cigarettes and/or use over-the-counter nicotine products (patches and gums) to satisfy nicotine addiction. Nicotine, one of the main constituents of tobacco and cigarette smoking, is a tertiary amine composed of a pyridine and a pyrrolidine ring. Serious, sometimes fatal, cases of atrial fibrillation (AF) have been reported in patients with and without detectable structural and/or history of heart disease from using a nicotine product, particularly when patients have smoked while using nicotine patches.1,2 In an American Multicenter Study, designed to determine the safety of nicotine, many participating patients were withdrawn from the study because they developed cardiac arrhythmias with nicotine use.27

The most important characteristic of AF is the changes in atrial structure and electrophysiology, which promotes persistent AF even in the absence of progressive underlying heart disease, so-called atrial remodelling. Two principal interrelated forms of atrial remodelling have been identified in animal models of AF: atrial structural remodelling (ASR) and atrial electrical remodelling.810 ASR is characterized by severe local conduction slowing and heterogeneous electrical activities associated with structural abnormalities with prominent fibrosis between and within atrial muscle bundles. Cardiac collagen plays an important role in maintaining the structural integrity and overall geometry of heart, whereas extracellular matrix restricts and locates individual myocytes and myofibrils, enabling optimal conduction and coordination of force generated during cardiac contraction. It is known that the atrium is more susceptible to atrial fibrosis than ventricle. An increase in atrial fibrosis can impair cell-to-cell coupling, providing a fixed morphological substrate that favours the occurrence of AF.1115 Indeed, recent studies have demonstrated that nicotine causes a significant and highly heterogeneous increase in atrial interstitial fibrosis in human16 and canine atria.17,18 Nicotine-induced atrial fibrosis may contribute to ASR thereby increased AF vulnerability in cigarette smokers. However, how nicotine induces atrial fibrosis remained unclear.

Transforming growth factor-β1 (TGF-β1) regulates the production and deposition of collagens and induces gene expression related to the development of myocardial fibrosis, for instance, connective tissue growth factor (CTGF).1922 TGF-β1 increases the abundance of messenger RNA (mRNA) for collagen types I and III and enhances collagen type I synthesis in cardiac fibroblasts in culture.23,24 Whether TGF-β1 is involved in nicotine-induced ASR remained unknown. The present study was designed to experimentally decipher molecular mechanisms underlying nicotine's ability to promote ASR and AF. Our results suggest an important role of upregulation of TGF-β1 and TGF-β receptor type II (TGF-βRII) by nicotine via downregulating miR-133 and miR-590 in the detrimental ASR leading to AF.

2. Methods

2.1. Drug administration

Experimental studies were carried out in mongrel dogs of either sex weighing from 21.3 to 28.6 kg. The animals were randomly divided into three groups: control, nicotine (5 nmol/L kg−1, Sigma Chemicals), nicotine (50 nmol/L kg−1) groups. Venous catheters were implanted into the vena jugularis externa for long-term drug infusion. Dogs of nicotine group were infused with nicotine (5 nmol/L kg−1 or 50 nmol/L kg−1) for consecutively 30 days, and control dogs were injected with equal volume of saline. All animal procedures were approved by the Ethics Committee for Animal Experiments, Harbin Medical University and confirmed with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85–23, revised 1996).

The concentrations of nicotine used were based on previous studies as well as our preliminary experiments. For example, 5 and 50 nmol/L kg−1 (i.v.) nicotine yielded blood levels of ∼50 and 500 nmol/L, respectively, at 1 h after administration, as determined by gas liquid chromatography. It has been shown that the average blood concentration of nicotine in regular smokers is 220 nmol/L and that the level can reach 440 nmol/L after consumption of a single cigarette.25,26

2.2. Whole dog studies

On the 31st day, the dogs were anesthetized and medium sternotomy was performed. A 15-stimulus basic train at a basic cycle length (2 ms, twice-threshold current pulses) of 150 ms was followed by a premature extrastimulus (S2s, 2 ms, twice-threshold current) at right atrial appendage (RAA) to induce AF.15,27 AF vulnerability at each site was defined by the ability of single S2s to induce in a reproducible fashion AF that lasted > 1 s. Overall vulnerability in each dog was defined as the percentage of sites at which AF was inducible. Because in some dogs AF was inducible by single extrastimuli, AF was also induced by stimulating the RAA with 10 Hz, 2 ms stimuli at four times threshold current for 2–10 s. To calculate mean AF duration, AF was induced with burst pacing ten times for AF duration < 10 min and twice for AF duration > 10 min. AF that lasted > 20 min was terminated by direct current electrical cardioversion, and 20 min was allowed before repeating AF induction.

2.3. Histochemistry

After the above in vivo study, the heart was quickly dissected and four atrial tissue samples were dissected from both atria (including appendage and free wall). The samples were immersed in 10% neutral buffered formalin for 24 h and stained with Masson-trichrome. Fibrous tissue was quantified with SigmaScan 4.0 (Jandel Scientific) as previously described.2830 Twenty areas of 0.3 × 0.4 mm2 and 2 mm apart were analysed in each slide, and each area was divided into 100 squares with the help of a grid. The number of collagen points (blue stain) at the 100 intersection points in the grid was scored as 1 (present) or 0 (absent). Results are expressed as the percentage of area occupied by fibrosis to the total atrial area examined.31

2.4. Human preparations

We obtained the human tissues from the Second Affiliated Hospital of Harbin Medical University under the procedures approved by the Ethnic Committee for Use of Human Samples of the Harbin Medical University. The investigation conforms with the principles outlined in the Declaration of Helsinki (see Cardiovasc Res 1997;35:2–4) for use of human tissue. The human tissues were from 19 hearts of patients with AF, 10 of them with no smoking history and nine with at least 17 years of smoking history. The tissues (right atrium appendage) were obtained during heart surgery, and these tissues are considered as surgical waste (see Supplementary material online, Table S1).

2.5. Western blot

The protein samples were extracted from the right atrial tissue, with the procedures essentially the same as described in detail elsewhere.32,33 Membrane and cytosolic proteins were separated. Protein sample (∼50 µg) was fractionated by SDS–PAGE (7.5−10% polyacrylamide gels). The primary antibodies against TGF-β1 (rabbit polyclonal), TGF-βRII (rabbit polyclonal), and CTGF (goat polyclonal) were purchased from Santa Cruz Biotechnology, with GAPDH (anti-GAPDH antibody from Research Diagnostics Inc.) as an internal control.

2.6. Real-time RT–PCR

For quantification of TGF-β1 and TGF-βRII transcripts, conventional real-time RT–PCR was carried out with total RNA samples extracted from canine right atria or from cultured atrial fibroblasts as to be described below. TaqMan quantitative assay was performed with the expression level of GAPDH as an internal control.

The mirVana™ qRT–PCR microRNA (miRNA) Detection Kit (Ambion) was used in conjunction with real-time PCR with SYBR Green I for quantification of transcript abundance of miR-133 and miR-590, according the procedures detailed elsewhere.34,35 5s rRNA was used as an internal control for variations.

2.7. Fibroblast isolation and culture

Atrial fibroblasts were obtained from right atria from the dogs from the control group, using the same procedures as described by Shiroshita-Takeshita et al.36 First- and second-passage fibroblasts were used for studies. For experiments involving the TGF-β type I receptor, SB-431542 [4-[4-(1,3-benzodioxol-5-yl)-5-(2-pyridinyl)-1H-imidazol-2-yl] benzamide] (Tocris Bioscience, Ellisville, MO, USA),37,38 the compound was dissolved at a concentration of 5 mmol/L in DMSO and fibroblasts were pre-incubated for 30 min with SB-431542 at a concentration of 10 µmol/L before the addition of nicotine for 24 h.

2.8. Transfection of miRNAs

MiR-133 and miR-590 and their respective mutant constructs were synthesized by Integrated DNA Technologies, Inc. The sequences of anti-miR-133 and anti-miR-590 antisense inhibitor oligonucleotides (AMOs) used in our studies are the exact antisense copies of their respective mature miRNA sequences and the nucleotides in the AMOs all contain 2′-O-methyl modifications at every base and a 3′ C3 containing amino linker.34,35

2.9. Measurement of collagen content

Total collagen content was determined through a colorimetric reaction against picrosirius red as described previously.39,40 Briefly, cells (2 × 105) after various treatments (at 85% confluence) were lysed and the lysate (100 µL) was dehydrated and stained with 0.1% picrosirius red in saturated picric acid (w/v) in a 96-well plate. The dye was solubilized and absorbance read at 540 nm. Readings were converted to protein units using a linear calibration curve generated from standards (Vitrogen 100, Angiotech Biomaterials, Palo Alto, CA, USA) and normalized to the wet weight of each cell sample (1.2–1.5 mg).

2.10. Luciferase assay

To generate reporter vectors bearing miRNA-binding sites, we subcloned the 3′-untranslated region (3′-UTR) of TGF-β1 or TGF-βRII by PCR-based methods. The constructs were inserted into the multiple cloning sites downstream the luciferase gene (HindIII and SpeI sites) in the pMIR-REPORT™ luciferase miRNA expression reporter vector (Ambion, Inc.), with the methods described previously.34,35

For luciferase assay, cells (1 × 105/well) were transfected with 1 µg PGL3–target DNA (firefly luciferase vector) and 0.1 µg PRL-TK (TK-driven Renilla luciferase expression vector), along with using lipofectamine 2000, according to the manufacturer's instructions, after 24 h starvation in serum-free medium. Luciferase activities were measured 48 h after transfection with a dual luciferase reporter assay kit (Promega) on a luminometer (Lumat LB9507).34,35

2.11. Data analysis

Group data were expressed as mean ± SEM. Statistical comparisons among multiple groups were performed by analysis of variance (ANOVA). If significant effects are indicated by ANOVA, a t-test with Bonferroni correction or a Dunnett's test was used to evaluate the significance of differences between individual means. Otherwise, baseline and drug data were compared by Student′s t-test. A two-tailed P < 0.05 was taken to indicate a statistically significant difference.

3. Results

3.1 Nicotine induces in vivo atrial fibrosis and enhances AF vulnerability

The dogs treated with nicotine base were more prone to AF induction in response to a premature extrastimulus following a train of rapid pacing. This was reflected by the significantly increased AF vulnerability and AF duration (Figure 1) in nicotine-treated dogs relative to control sham-treated dogs. The AF-promoting effect of nicotine was clearly dose-dependent.

Figure 1

Nicotine increases atrial fibrillation (AF) vulnerability in dogs. (A) Examples of atrial activation recorded in the right atrial wall, showing sinus rhythm and AF. (B and C) AF vulnerability and AF duration, respectively. Ctl: control; Nic (5): nicotine base at a dose of 5 nmol/L kg−1; Nic (50): nicotine base at 50 nmol/L kg−1. *P < 0.05 vs. Ctl; n = 5 dogs for each bar.

Histopathological analysis demonstrated substantial structural alterations, induced by nicotine, in both right and left atria. In control specimens, narrow myocytes were well aligned and surrounded by little interstitial tissue. In contrast, section of atrium from a nicotine-treated dog was occupied by a large proportion of connective tissue that accumulated between bundles of fibres. Moreover, normal collagenous structures have a parallel arrangement of relatively broad bands of collagen fibres, whereas the abnormal interstitial fibrosis consisted of haphazardly arranged, disorganized, fine collagen fibres (Figure 2).

Figure 2

Induction of atrial fibrosis by nicotine in dogs. (A) Microscopic images showing the fibrotic tissues (blue colour) of Masson's trichrome-stained sections. (B) Percentage of atrial cross-sectional area composed of fibrous tissue. n = 5 dogs for each bar.

3.2 Nicotine stimulates production of collagens in atrial fibroblasts

Cardiac fibroblasts are known to synthesize fibrillar collagens via autocrine/paracrine action of TGF-β1 and are thus responsible for tissue fibrosis/remodelling.19,41 We thus evaluated the effects of nicotine on collagen production in cultured atrial fibroblasts isolated from healthy dogs. As shown in Figure 3, collagen content was significantly higher with the treatment of nicotine base (50 or 500 nmol/L) compared with that without nicotine treatment. The collagen production-stimulating effect of nicotine was concentration-dependent. α-Bungarotoxin (α-BTX, an antagonist selective to α7 nicotinic acetylcholine receptors, α7-nAChRs), co-applied with nicotine (500 nmol/L) nearly abolished the collagen production-promoting effect of nicotine. Hexamethonium (HEX, a non-competitive antagonist to peripherally acting or ganglionic nAChR) produced similar antagonizing actions. In contrast, mecamylamine (MA, a centrally acting antagonist selective for non-α7-nAChRs) failed to affect the effect of nicotine. The synthetic activity of procollagen type I/III mRNAs in isolated fibroblasts and canine atrial tissue was also assessed (Supplementary material online, Figure S1).

Figure 3

Effect of nicotine on collagen production in cultured canine atrial fibroblasts. For experiments involving receptor antagonists, the drugs were co-incubated with nicotine (500 nmol/L). *P < 0.05 vs. Ctl; +P < 0.05 vs. Nic (500); n = 7 independent batches of cells for each bar.

3.3 Upregulation of TGF-β1 and TGF-βRII expression by nicotine

It is known that TGF-β1 is expressed in the heart and is mainly synthesized and released from cardiac fibroblasts,42,43 which in turn results in production of collagens. In order to investigate whether TGF-β1 plays a role in nicotine-stimulated collagen production and atrial fibrosis, we assessed the effects of nicotine on TGF-β1 and TGF-βRII. Consistent with the collagen stimulation, western blot analysis revealed significant upregulation of TGF-β1 and TGF-βRII genes at the protein level in nicotine-treated atrial fibroblasts, relative to non-treated cells. Nicotine-induced increases were abrogated by α-BTX (Figure 4).

Figure 4

Effects of nicotine on expression of TGF-β1 (A) and TGF-βRII (B) in cultured canine atrial fibroblasts. *P < 0.05 vs. Ctl; +P < 0.05 vs. Nic (500); n = 7 independent samples for each bar.

Noticeably, the nicotine-induced alteration of expression of these molecules at the mRNA level was not in parallel to their expression at the protein level. Quantitative RT–PCR demonstrated much smaller increases in the transcript levels of TGF-β1 and TGF-βRII, relative to those in their protein levels. For instance, nicotine (50 nmol/L) caused a nearly four-fold (P < 0.05) elevation of TGF-β1 protein level but only ∼40% (P > 0.05) increase in its mRNA level.

Similar disparities between protein and mRNA expressions of these genes were consistently observed in the protein and RNA samples extracted from atrial tissues of dogs (Supplementary material online, Figure S2).

TGF-β1 is a specific inducer of CTGF expression by activating its promoter. To corroborate that upregulation of TGF-β1 (and its receptor) results in activation of its signalling pathway, expression of CTGF was assessed in vivo and in vitro, nicotine dose dependently upregulated CTGF expression (Supplementary material online, Figure S3).

These results suggest a post-transcriptional mechanism that mediates the expression regulation of TGF-β1 and TGF-βRII genes by nicotine.

3.4 Evidence for decreased miR-133 and miR-590 levels underlying nicotine-induced upregulated TGF-β1 and TGF-βRII expressions

MiRNAs are endogenous ∼22-nt non-coding RNAs. The action of miRNAs involves incorporation of the single-stranded miRNA into the RNA-induced silencing complex and subsequent binding of the miRNA to the 3′-UTR of its target mRNA through exact complementarity with its 5′ end 7–8 nt, and partial complementarity with rest of the sequence.4447 In this way, miRNAs produce translational inhibition. To explore the possible involvement of miRNAs in the expression regulation of TGF-β1 and TGF-βRII, we first measured the expression levels of 10 miRNAs that have the potential to target the genes encoding TGF-β1 and TGF-βRII based on computational prediction using the miRNA TargetScan hosted by Wellcome Trust Sanger Institute.48 As illustrated in Figure 5A, out of the 10 miRNAs examined, miR-133 and miR-590 were significantly downregulated in the presence of nicotine in canine atrial fibroblasts. These effects of nicotine were concentration-dependent and reversed by co-application of α-BTX (Figure 5B and C). Similar downregulation of miR-133 and miR-590 levels was also consistently observed in atrial tissues of smoking people (Figure 5D) and nicotine-treated dogs (Supplementary material online, Figure S4).

Figure 5

Nicotine-induced alteration of expression of miRNAs in cultured atrial fibroblasts. (A) Nicotine (500 nmol/L) down-regulated miR-133 and miR-590 and did not affect other miRNAs examined. Concentration-dependent decreases in miR-133 (B) and miR-590 (C) levels by nicotine and effects of α7-nAChR antagonist α-BTX (100 nmol/L). *P < 0.05 vs. Ctl; +P < 0.05 vs. Nic (500); n = 6 independent samples for each bar. (D) miR-133 and miR-590 levels were decreased in RNA samples isolated from smoking individuals with AF. We compared non-smoking (AF, n = 10) and smoking individuals with AF (AFS, n = 9). Data are expressed as mean ± SEM normalized to AF. *P < 0.05 vs. AF.

TGF-β1 gene contains multiple putative binding sites for miR-133 (Figure 6A) in the 3′-UTR and coding region; on the other hand, the 3′-UTR and coding region of TGF-βRII bears multiple putative binding sequences for miR-590 (Figure 7A). We subsequently evaluated the ability of miR-133 to repress TGF-β1 expression and of miR-590 to repress TGF-βRII expression. As depicted in Figure 6B, transfection of miR-133 into the cultured atrial fibroblasts remarkably reduced the protein level of TGF-β1, and co-transfection of the anti-miR-133 antisense inhibitor oligonucleotide (AMO-133) abolished the effect and produced an overshoot. As a negative control, transfection of miR-590 did not affect TGF-β1 expression. Moreover, transfection of miR-133 prevented the nicotine-induced upregulation of TGF-β1 protein. The ability of miR-133 to repress TGF-β1 was verified by luciferase reporter activity assays in HEK293 cells that express minimal endogenous miR-133 (Figure 6C). When the luciferase vector carrying the 3′-UTR of TGF-β1 was co-transfected with miR-133, luciferase activity was robustly diminished, compared with transfection of luciferase vector alone (control). The inhibitory effect of miR-133 was antagonized by its antisense AMO-133. miR-133 did not significantly alter the mRNA levels of TGF-β1 (Figure 6D).

Figure 6

Post-transcriptional repression of TGF-β1 by miR-133. (A) Complementarities between miR-133 and TGF-β1 gene. miR-133 is presented in the form of DNA from 3′ end to 5′ end. Matched bases are indicated by bold-face letters and ‘|’ connecting the letters, and the wobble base-pairing is indicated by ‘:’. The numbers labelled in the sequences indicate the numbers of nucleotides in the sequence of TGF-β1 gene. (B) Effects of miR-133 on TGF-β1 protein (25 kDa) expression. +AMO-133: co-transfection of miR-133 and AMO-133; +miR-133: co-application of miR-133 and nicotine (500 nmol/L). miR-590 was used as a negative control for specificity of miR-133 actions. *P < 0.05 vs. Ctl; +P < 0.05 vs. miR-133 or Nic (500); n = 5 independent samples for each bar. (C) The ability of miR-133 to repress TGF-β1 expression. *P < 0.05 vs. Ctl; +P < 0.05 vs. miR-133; n = 6 independent samples for each bar. (D) Real-time RT–PCR analysis of TGF-β1 expression at the mRNA level (n = 5 independent samples for each bar).

Figure 7

Post-transcriptional repression of TGF-βRII by miR-590. (A) Complementarities between miR-590 and TGF-βRII gene. miR-590 is presented in the form of DNA from 3′ end to 5′ end. Matched bases are indicated by bold-face letters and ‘|’ connecting the letters, and the wobble base-pairing is indicated by ‘:’. The numbers labelled in the sequences indicate the numbers of nucleotides in the sequence of TGF-βRII gene. (B) Effects of miR-590 on TGF-βRII protein (70 kDa) expression. +AMO-590: co-transfection of miR-590 and AMO-590; +miR-590: co-application of miR-590 and nicotine (500 nmol/L). miR-133 was used as a negative control for specificity of miR-590 actions. *P < 0.05 vs. Ctl; +P < 0.05 vs. miR-590 or Nic (500); n = 5 independent samples for each bar. (C) The ability of miR-590 to repress TGF-βRII expression. *P < 0.05 vs. Ctl; +P < 0.05 vs. miR-590; n = 6 independent samples for each bar. (D) Real-time RT–PCR analysis of TGF-βRII expression at the mRNA level (n = 5 independent samples for each bar).

In another series of experiments, miR-590 produced similar pattern of changes in TGF-βRII; transfection of miR-590 into the cultured atrial fibroblasts caused substantial reduction TGF-βRII expression at the protein level, the effects abrogated by co-transfection of the anti-miR-590 AMO (AMO-590) (Figure 7B). As a negative control, transfection of miR-133 failed to affect TGF-βRII expression. Moreover, transfection of miR-590 prevented the nicotine-induced upregulation of TGF-βRII protein. The ability of miR-590 to repress TGF-βRII expression was confirmed by luciferase reporter activity assays in HEK293 cells (Figure 7C). When the luciferase vector carrying the 3′-UTR of TGF-βRII was co-transfected with miR-590, luciferase activity was robustly diminished, compared with transfection of luciferase vector alone (control). The inhibitory effect of miR-590 disappeared when co-transfected with its antisense AMO-590. miR-590 did not significantly alter the mRNA levels of TGF-βRII (Figure 7D).

3.5 Effects of miR-133 and miR-590 on atrial fibroblast proliferation

If the decreases in miR-133 and miR-590 levels in the presence of nicotine mediated the nicotine-induced enhancement of atrial fibrosis, then treatment of cells with miR-133 or miR-590 alone should cause suppression of collagen production in atrial fibroblasts. This notion was tested by transfecting synthetic miR-133 or miR-590 into the cultured canine atrial fibroblasts. As shown in Figure 8, miR-133 or miR-590 indeed significantly diminished collagen content and the effects were antagonized by their respective AMOs. Moreover, transfection of AMO-133 or AMO-590 alone enhanced collagen content and either miR-133 or miR-590 abolished the nicotine-induced enhancement of collagen production. Co-application of miR-133 and miR-590 produced greater inhibitory effects on collagen production than either of these miRNAs alone.

Figure 8

Effect of miR-133 and miR-590 on collagen production in cultured canine atrial fibroblasts. Nic (500): nicotine base (500 nmol/L); +AMO-133 or AMO-590: co-transfection of miR-133 and AMO-133 or miR-590 and AMO-590; +miR-133 or +miR-590: co-application of Nic and miR-133 or Nic and miR-590; +SB: Nic+SB-431542. *P < 0.05 vs. Ctl; +P < 0.05 vs. miR-133 or miR-590; §P < 0.05 vs. Nic (500); n = 5 independent batches of cells for each bar.

It has been known that binding of TGF-β1 to constitutively activated TGF-βRII interacts with TGF-β type I receptor (TGF-βRI) and activates the latter.19 This process then activates intracellular signalling pathways and one of them can result in production of extracellular matrix proteins collagens.19,41 To investigate whether nicotine-induced TGF-β1 acts through this pathway to stimulate atrial collagen production, we tested the effect of SB-431542, a small molecule inhibitor of TGF-βRI,37,38 on nicotine-induced collagen production in atrial fibroblasts. The data in Figure 8 provide evidence in support of the notion; SB-431542 significantly weakened the collagen-promoting effect of nicotine.

4. Discussion

We demonstrate here that nicotine can induce ASR and increase AF vulnerability in dogs. Nicotine stimulates collagen production and atrial fibrosis both in vitro in atrial fibroblasts and in vivo in atrial tissues. Nicotine upregulates the protein level of TGF-β1 and TGF-βRII and downregulates miRNAs miR-133 and miR-590. We further demonstrated that miR-133 and miR-590 also decreased in smoking individuals with AF. Overexpression of miR-133 and miR-590 results in post-transcriptional repression of TGF-β1 and TGF-βRII, respectively, which reduces collagen production in cultured canine atrial fibroblasts. All these effects of nicotine are prevented by α7-nAChR antagonist α-BTX. The present study therefore revealed downregulation of critical miRNAs as a molecular mechanism by which nicotine promotes ASR for AF.

Evidence exists from clinical observations indicating that cigarette smoking/nicotine exposure promotes AF. However, there have been little animal (experimental) studies to clarify this notion. And the potential mechanisms by which cigarette smoking/nicotine exposure increases propensity of AF remained unclear. Specifically, nicotine exposure develops functional substrates and structural substrates, for AF to occur.16 The structural substrates known to modulate atrial vulnerability to inducible AF are enhanced atrial interstitial fibrosis, a phenomenon that characteristically develops during aging. Atrial interstitial fibrosis increases with age in humans and has been observed in subjects with AF in both clinical and experimental settings. Theoretical models have implicated atrial interstitial fibrosis as a substrate for AF. Enhanced fibrosis may have a role in structural remodelling and the pathogenesis of AF as a result of separation of the cells by fibrotic depositions. On another hand, previous studies reported that nicotine can directly modulate atrial electrophysiology by blocking several K+ channels.49,50 Moreover, It has been well established that neuronal mechanisms were important in maintaining cardiac function.5153 The canine heart contains nicotine-sensitive neurons which can produce either promoting or suppressive effects.54 Administration of nicotine causes modulation of intrinsic cardiac nerve activity and cardiac responses, in a spatial selective manner.5355 These effects also likely contribute to nicotine's ability to promote AF, though it, in general, requires higher concentrations of nicotine to act on ion channels, and the spatially divergent cardiac responses to nicotinic stimulation could not entirely interpret the proarrhythmic effects of nicotine. Evidently, nicotine promotes AF at least via three distinct mechanisms involving neuronal, electrical, and structural remodelling to create functional and anatomical substrates for the arrhythmia.

On the basis of our results, it appears reasonable to propose that cigarette smoking/nicotine exposure activates α7-nAChRs in atrial fibroblasts, which then results in an expression downregulation of miR-133 and miR-590 via some unknown mechanisms. Decreased miR-133 and miR-590 levels cause an upregulation of TGF-β1 and TGF-βRII proteins due to a removal of translational inhibition by these miRNAs. Increased TGF-β1 and TGF-βRII result in enhanced collagen production and fibrosis generation (Supplementary material online, Figure S5).

TGF-β- and TGF-βRII-mediated signalling cascade that promotes production and deposition of collagens in the myocardium related to the development of myocardial fibrosis, and subsequently induces atrial remodelling and fibrillation have been studied extensively.19,22,56 TGF-β is one of the most potent inducer of collagen production and fibrosis, which regulates CTGF expression, as corroborated in the present study, through a signalling cascade requiring Smads, PKC/ras/MEK/extracellular signal-regulated kinase, and an Ets-1 binding element in the CTGF promoter.57 More importantly, a recent study revealed that miR-133 and miR-30c are negative regulators of CTGF expression in the heart.58 The present study further clarifies the importance of miRNAs in regulating the expression of TGF-β and TGF-βRII at the post-transcriptional level and the contribution of this regulation to cardiac fibrogenesis.

Probably the most striking finding of this study is that miRNAs are involved as a necessary mediator in nicotine-induced atrial fibrosis. Discovery of miRNA has revolutionized our understanding of the mechanisms that regulate gene expression and we are compelled to revise our understanding of the mechanisms regulating normal and pathological cellular functions. Our data that downregulation of miR-133 and miR-590 by nicotine triggers generation of atrial fibrosis via removal of post-transcriptional repression of TGF-β1 and TGF-βRII and consequent production of collagens. In our experiments, miR-133 and miR-590 do not significantly affect the mRNA levels of TGF-β1 and TGF-βRII, possibly due to very imperfect complementarities between the miRNAs and the genes. Thus, the profibrotic response to nicotine in canine atrium is critically dependent upon downregulation of miR-133 and miR-590. It is interesting to speculate that these miRNAs are also involved in ASR for AF under other pathological settings, a notion requiring rigorous experimental investigations.

We showed that downregulation of miR-133 and miR-590 by nicotine is mediated by activation of α7-nAChR, because the antagonist of α7-nAChR α-BTX abrogates nicotine's action on the miRNAs. However, the present study does not answer the question how α7-nAChR activation leads to miR-133 and miR-590 downregulation.

Conflict of interest: none declared.

Funding

This work was supported in part by National Basic Research Program of China (973 Program) [2007CB512000 (2007CB512006) to B.Y., 2007CB516800 (2007CB516803) to Y.L.] and the National Nature Science Foundation of China [30871062].

Footnotes

  • The first three authors contributed equally to this work.

References

View Abstract