The present study was performed in accordance with the principles outlined in the Declaration of Helsinki and was approved by the research Ethics Committee of Guangdong General Hospital, Guangdong Academy of Medical Sciences (no. GDREC2013047H; Guangzhou, China). All patients provided written informed consent (version no. 20130307). Patients with pneumonia or other infectious diseases were excluded from the study. Left atrial appendages were obtained from patients undergoing open-heart surgery (from March 2013 to March 2016) in Guangdong General Hospital in Guangzhou (Guangdong, China). The specimens were collected prior to initiation of cardiopulmonary bypass, immediately snap-frozen in liquid nitrogen, and stored at −80°C until the day of the experiments. Tissues from 11 patients with chronic AF (from 32–67 years) and a control group of 10 patients (from 30–61 years) with sinus rhythm (SR) were also used in the present study. Patient data are presented in Table I.

The experiments using CFs were approved by the research Ethics Committee of Guangdong General Hospital, Guangdong Academy of Medical Sciences (no. GDREC2013047A). The CFs were isolated from 1–3-day-old Wistar rats (n=24, 14:10 male:female) which were provided by Sun Yat-sen University and fed at 20–25°C (12-h light/dark cycle), as previously described (14). Briefly, the heart tissues were finely minced and mechanically digested with 0.25% Trypsin + EDTA (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) using a rotor in a flask. The cell suspensions were plated in a cell culture flask for 90 min to separate the fibroblasts and cardiomyocytes. The majority of CFs adhered to the flask. The cells were cultured in Dulbecco's modified Eagle's medium (Gibco; Thermo Fisher Scientific, Inc.) containing 4.5 g/l D-glucose, 10% fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc.), penicillin (100 U/ml) and streptomycin (100 µg/ml) in a humidified atmosphere (95% humidity, 37°C) composed of 95% O₂, 5% CO₂. The CFs were used at passages 3–6.

The cells were treated according to three experimental designs. In the first, the cells (50–60% confluence) were treated with different doses of mouse recombinant MIF at 37°C (rMIF; 20 and 40 nM; R&D Systems Inc., Minneapolis, MN, USA) for 24 or 48 h. In the second, the cells were pretreated with PP1 (5 µM; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) for 1 h. The MIF treatment groups were pretreated with an equal quantity of dimethyl sulfoxide (DMSO) and were stimulated with 40 nM MIF for 48 h. In the third, the cells were treated with 40 nM MIF for 15, 30, 45 and 60 min, or with 20 and 40 nM MIF for 24 h.

The CFs were plated into a 96-well plate at a density of 4,500/well and proliferation was measured using the cell proliferation reagent (WST-1) kit (cat. no. 11644807001; Roche Diagnostics GmbH, Mannheim, Germany) according to the manufacturer's instructions. The signal in optical density was read at 450 nm on a plate reader (Mutiskan GO; Thermo Fisher Scientific, Inc.). The experiment was repeated three times.

The cells were lysed in 0.05 M Tris-HCl buffer (pH 8.0), containing 0.15 M sodium chloride, 0.02% sodium azide, 0.1% sodium dodecyl sulfate (SDS), 1% Nonidet P-40 (NP-40), and a protease inhibitor cocktail set (Calbiochem; EMD Millipore, Billerica, MA, USA). The cell lysates were centrifuged at 12,000 × g for 15 min at 4°C and protein concentrations were determined by bicinchoninic acid Protein Assay kit (Pierce; Thermo Fisher Scientific, Inc.). The samples were diluted with 4X loading buffer (Bio-Rad Laboratories, Inc., Hercules, CA, USA) and heated at 95°C for 5 min. The proteins (30 µg) were fractionated on 8% SDS-polyacrylamide gels for Src, phosphorylated (p-)Src, collagen type 1, α1 (Col-1A1), collagen type 3, α1 (Col-3A1), matrix metalloproteinase (MMP)-2/-9 and transforming growth factor (TGF)-β or 10% SDS polyacrylamide gels for MIF. The proteins were then transferred onto nitrocellulose membranes (GE Healthcare Life Sciences, Chalfont, UK) according to standard protocols. The membranes were blocked with dried skimmed milk powder in TBS-Tween-20 (TBST) for 1 h at room temperature, prior to overnight incubation at 4°C with primary rabbit polyclonal antibodies against Src (1:1,000, cat. no. ab47405; Abcam, Cambridge, UK), p-Src Tyr⁴¹⁶ (1:1,000, cat. no. 2101; Cell Signaling Technology, Inc., Danvers, MA, USA), Col-1A1 (1:1,000, cat. no. ab34710), Col-3A1 (1:1,000, cat. no. ab7778) (both from Abcam), MMP-9 (1:1,000, cat. no. AB19016; EMD Millipore, Billerica, MA, USA) and TGF-β (1:1,000, cat. no. 3711; Cell Signaling Technology, Inc.) and MMP-2 (1:2,000, cat. no. AB19015; EMD Millipore;). The signals were normalized to the protein levels of GAPDH (1:1,000, cat. no. 2118) or β-actin (1:3,000, cat. no. 3700) (both from Cell Signaling Technology, Inc.). Following washing in TBST, the membranes were incubated for 1 h with horseradish peroxidase-conjugated anti-rabbit or anti-mouse IgG (1:5,000-1:10,000; Abcam) in blocking solution. The protein bands were visualized using electrochemiluminescence reagents (Pierce; Thermo Fisher Scientific, Inc.), and images were evaluated densitometrically using ImageJ 1.39 U software (National Institutes of Health, Bethesda, MD, USA).

The rMIF was dissolved in sterile PBS containing 0.1% bovine serum albumin (Sigma-Aldrich; Merck KGaA). PP1 was purchased from Sigma-Aldrich; Merck KGaA. The kinase inhibitor was dissolved in DMSO (Sigma-Aldrich; Merck KGaA). The concentration of DMSO in the working solutions did not exceed 1.0%.

All data are expressed as the mean ± standard error of the mean. Student's t-test was used as evaluate pairwise statistical significances of differences between two group means, and one-way analysis of variance was used for multiple groups with SPSS 21 statistical software (IBM Corp, Armonk, NY, USA). P<0.05 was considered to indicate a statistically significant difference.

The protein levels of Col-1A1, Col-3A1, MMP-2/-9, TGF-β and MIF were detected in left atrial homogenates from patients with AF and SR controls (for patient characteristics; Table I). The protein expression levels of Col-1A1, Col-3A1, MMP-2/-9, TGF-β and MIF were normalized to those of GAPDH in each sample. With the exception of Col-1A1, the protein levels of Col-3A1, MMP-2/9, TGF-β1 and MIF were significantly higher in the left atrial tissues from patients with AF (n=11), compared with the SR (n=10) group (Col-1A1, 0.81±0.12, vs. 0.51±0.09, P=0.062; Col-3A1, 0.88±0.12, vs. 0.32±0.06, P=0.001; MMP-2, 1.06±0.09, vs. 0.73±0.05, P=0.036; MMP-9, 1.06±0.15, vs. 0.57±0.09, P=0.013; TGF-β, 0.55±0.11, vs. 0.30±0.04, P=0.049; MIF, 1.20±0.15, vs. 0.76±0.14, P=0.044), as shown in Fig. 1A-G. These results indicated that, as an important proinflammatory cytokines, MIF may be involved in the regulation of the atrial fibrosis, which contributes to the onset of AF.

The proliferation of CFs is important in myocardial fibrosis. The present study investigated the proliferation of CFs using CCK-8 assays, and it was found that 24-h MIF treatment had no effect on the proliferation of CFs, whereas cells exhibited rapid growth following stimulation for 48 h with 40 nM MIF (Fig. 2A).

Src, a member of the protein tyrosine kinase (PTK) family, has been suggested to be involved in cell proliferation. The present study examined whether Src kinases were involved in MIF-induced CF proliferation using a specific Src antagonist (PP1). The CFs were pretreated with PP1 (5 µM) for 1 h prior to rMIF stimulation. PP1 completely inhibited the MIF-induced CF proliferation (Fig. 2B). These results suggested that MIF induced cell proliferation through Src kinase-dependent mechanisms in CFs.

The results reported above suggested that the Src kinase signaling pathway may be responsible for the proliferation of CFs induced by rMIF. To confirm the specificity of the signaling pathways involved in the proliferation of CFs induced by rMIF, western blot analysis was performed to analyze the levels of Src and p-Src in CFs following treatment with MIF. The CFs were treated with 20 or 40 nM of rMIF for 24 h, or 40 nM rMIF for 0, 15, 30, 45 and 60 min, and the levels of Src and p-Src were measured. The higher concentration of rMIF (40 nM) was found to produce a marked increase in the protein levels of Src, compared with those in the unstimulated cells and the 20 nM rMIF-treated group of cells, and Src was activated by 40 nM rMIF at 30 min (Fig. 3A and B).

To investigate the role of MIF on the secretion function of CFs, the present study also examined the effects of MIF on collagen deposition, and the expression of MMP and profibrotic TGF-β in CFs. The MIF dose-response was assessed using concentrations of 20 and 40 nM for 24 h, and it was found that the expression levels of Col-1A1, Col-3A1, MMP-2/-9 and TGF-β increased significantly in the 40 nM rMIF treatment group, but not in 20 nM rMIF treatment group, with the exception of MMP-9 which did increase significantly at 20 nM (Fig. 4A-F).