TWEAK/Fn14 mediates atrial‐derived HL‐1 myocytes hypertrophy via JAK2/STAT3 signalling pathway

Abstract Atrial myocyte hypertrophy is one of the most important substrates in the development of atrial fibrillation (AF). The TWEAK/Fn14 axis is a positive regulator of cardiac hypertrophy in cardiomyopathy. This study therefore investigated the effects of Fn14 on atrial hypertrophy and underlying cellular mechanisms using HL‐1 atrial myocytes. In patients with AF, Fn14 protein levels were higher in atrial myocytes from atrial appendages, and expression of TWEAK was increased in peripheral blood mononuclear cells, while TWEAK serum levels were decreased. In vitro, Fn14 expression was up‐regulated in response to TWEAK treatment in HL‐1 atrial myocytes. TWEAK increased the expression of ANP and Troponin T, and Fn14 knockdown counteracted the effect. Inhibition of JAK2, STAT3 by specific siRNA attenuated TWEAK‐induced HL‐1 atrial myocytes hypertrophy. In conclusion, TWEAK/Fn14 axis mediates HL‐1 atrial myocytes hypertrophy partly through activation of the JAK2/STAT3 pathway.

Fn14 is a tightly regulated receptor that has been associated with various downstream signalling cascades. 4,5 Although Fn14 usually is expressed at very low levels under physiological conditions, 6 it can be induced in cardiovascular disease, 7,8 and the TWEAK/Fn14 axis is involved in the pathogenesis of various cardiovascular diseases. [4][5][6][7][8] Adult TWEAK transgenic mice exhibit cardiomyocyte hypertrophy and cardiac dilation, while Fn14-/-mice are protected against TWEAK-induced cardiac dilation. 6 Activation of TWEAK/Fn14 signalling in adult rat primary cardiomyocytes causes cardiomyocyte hypertrophy. 9 However, atrial expression of Fn14 in patients with AF and its role in atrial structural remodelling remains undefined.
Janus kinase (JAK)/signal transducer and activator of transcription (STAT) pathway was initially discovered as a major cytokine signal transduction pathway. 10 JAK-STAT signalling pathway plays an important role in cardiac pathophysiology and has been implicated in pressure overload-induced cardiac hypertrophy and remodelling. 11 The activation of JAK/STAT pathway by IL-6 may contribute to the pathogenesis of cardiac hypertrophy, 12 and phospho-STAT3 levels are elevated in atrial tissues from AF patients. 1 TWEAK/Fn14 upregulates pro-inflammatory cytokine secretion via STAT3 pathways in hepatic stellate cells. 13 We hypothesized that up-regulated Fn14 expression may facilitate TWEAK-induced atrial myocyte hypertrophy and that, conversely, Fn14 inhibition may have a protective role on atrial structural remodelling. This study therefore explored the potential role and underlying mechanism of Fn14 in HL-1 atrial myocyte hypertrophy under conditions of TWEAK stimulation.

| Western blotting
Total protein was extracted from HL-1 cells or PBMCs using standard protocols in a RIPA buffer with 1 mmol/L PMSF. Equal amounts of protein from different experimental groups were separated by 8%-12% SDS-PAGE and transferred to PVDF membranes (Millipore, Eschborn, Germany), which were blocked for 1 hour with 5% non-fat milk in TBST at room temperature, then incubated over-

| Histology and immunohistochemistry
Human atrial appendages were dissected and fixed immediately in 4% paraformaldehyde. Tissue was paraffin-embedded and sectioned (5 lm) for subsequent analyses. Atrial myocyte area (lm 2 ) was measured in images from H&E-stained sections from human atrial appendages. Immunohistochemistry was performed for detecting Fn14 expression, and sections were incubated overnight at 4°C with primary antibodies against Fn14 (1:100 dilution; OmnimAbs, Alhambra, CA, USA). Goat anti-rabbit antibody was used as secondary antibody.
All the results were analysed by Image-Pro Plus 6.0 (Media Cybernatics, Houston, TX, USA).

| Enzyme-linked immunosorbent assay
Human serum samples were collected from the same patients as isolation of PBMCs and stored at À80°C, and levels of TWEAK were determined using enzyme-linked immunosorbent assay (ELISA) kits (eBioscience, Vienna, Austria) following the manufacturer's protocols.
The sensitivity for the assay was 9.7 pg/mL. The intra-assay and inter-assay coefficient of variation was 7.9% and 9.2%, respectively.

| Statistical analysis
All statistical analyses were performed using SPSS 17.0 (SPSS, Chicago, IL, USA). Continuous variables are represented as means AE SD and were compared using unpaired t test and one-way ANOVA. Categorical variables were presented as counts and percentages then analysed using either a chi-square test or Fisher's exact test. Logistic regression analysis was used to identify independent predictors of AF. A difference with a P value of <.05 (2-sided) was considered statistically significant.

| Lower TWEAK serum levels and up-regulated TWEAK expression in PBMCs from patients with AF
To determine whether AF affected TWEAK expression, we examined the serum levels of TWEAK in AF and NSR subjects. White blood cells count and left atrial size were significantly higher in patients with AF than those in patients with NSR (6.35 AE 1.41 9 10 12 /L vs 5.61 AE 1.26 9 10 12 /L, P < .05 and 40.31 AE 5.08 mm vs 36.10 AE 4.84 mm, P < .05, respectively). There were no other significant differences between the two groups. The characteristics of the studied population were showed in Table 1. ELISA assay revealed significantly lower TWEAK serum levels in the AF group than NSR group (418.56 AE 127.28 pg/mL vs 500.5 AE 107.67 pg/mL, P < .05; Figure 1A). We have analysed the possible biomarkers of AF by univariate and multivariate regression analysis and found serum TWEAK level was an independent biomarker of AF (Table S1). To further confirm the effects of AF on TWEAK expression, we isolated PBMCs from blood samples. TWEAK expression was significantly higher in PBMCs from AF subjects as compared to NSR subjects (P < .05; Figure 1B).

| Increased Fn14 expression in atrial appendages of patients with AF
To confirm the effects of AF on Fn14 expression, we used Western blot analysis and immunohistochemistry of atrial appendages from AF and NSR subjects. Fn14 protein expression was higher in atrial appendages from AF than NSR subjects by both Western blot analysis and immunohistochemistry (P < .05; Figure 2A,B).

| H&E staining showed larger atrial myocytes area in atrial appendages from patients with AF
In H&E-stained sections, atrial myocyte area (lm 2 ) was increased in atrial appendages from AF patients (P < .05; Figure 2C,E).

| TWEAK-induced hypertrophy of HL-1 atrial myocytes and increased Fn14 expression
Based on previous studies, we chose a series of TWEAK concentrations to explore the effect of TWEAK on hypertrophy of HL-1 atrial  Figure 3A,C), and at the latter 3 concentration also Troponin T expression (P < .05; Figure 3A,D  Figure 3A,B). Thus, Fn14 expression was positively regulated by TWEAK in HL-1 cells in vitro, and the concentration of 100 ng/mL as used as TWEAK stimulation in the following vitro experiments.

| Fn14 inhibition attenuated the increased hypertrophy induced by TWEAK stimulation in HL-1 atrial myocytes
To determine the effect of Fn14 on atrial myocytes hypertrophy, we used Fn14-specific siRNA to knock down Fn14 expression in HL-1s.
Fn14 expression significantly decreased after transfection with Fn14specific siRNA as compared to siNC treatment (P < .05; Figure 4A).
Immunofluorescence assay also revealed greater Fn14 protein level with TWEAK stimulation than control conditions, and reduced Fn14 protein expression following Fn14 inhibition for 24 hours (P < .05; Figure 4C). TWEAK stimulation significantly increased HL-1 atrial myocytes hypertrophy, while inhibition of Fn14 attenuated the increased ANP and Troponin T protein expression induced by TWEAK (P < .05; Figure 4B). Therefore, TWEAK-induced HL-1 atrial myocytes hypertrophy was reduced following the inhibition of Fn14, rendering Fn14 responsible for TWEAK-induced HL-1s hypertrophy.

| Activation of JAK2/STAT3 pathway by TWEAK/Fn14 in HL-1 atrial myocytes
To assess whether JAK/STAT pathway was associated with TWEAK/Fn14 activation in HL-1 atrial myocytes, we explored JAK/ STAT expression with TWEAK treatment. p-JAK2 and p-STAT3 expression significantly increased after TWEAK treatment for 24 hours (P < .05; Figure 5A,B,D,G). p-JAK1, p-TYK2, p-STAT1 protein expression were also evaluated, and no obvious changes were observed after TWEAK stimulation ( Figure 5A-C,E,F). In addition, Western blot analysis showed that inhibition of Fn14 effectively decreased TWEAK-induced p-JAK2 and p-STAT3 expression (P < .05; Figure 6A,B).

| DISCUSSION
Atrial myocyte hypertrophy can lead to conduction heterogeneity and is one of the most important substrates in AF. 16,17 TWEAK/Fn14 axis is a positive regulator of cardiac hypertrophy in cardiomyopathy. 6,9 The present study focused on the potential role and mechanism of Fn14 in TWEAK-induced atrial myocyte hypertrophy. The major findings were that in patients with AF, Fn14 protein levels were increased in atrial myocytes and TWEAK expression was up-regulated in PMBCs while TWEAK serum levels were decreased; in vitro, TWEAK However, it has been reported that decreased concentrations of sTWEAK have been observed in diseases related to chronic low-grade inflammation, 26,27 and TWEAK may have evolved to guard against the development of a potentially excessive inflammatory response. 22,26 Previous studies have demonstrated that subjects with carotid stenosis showed a reduced plasma sTWEAK level compared with healthy subjects, and sTWEAK concentrations negatively correlated with the carotid intima-media thickness. 26 Decreased sTWEAK concentration is significantly and independently associated with long-term cardiovascular mortality in patients with peripheral arterial disease. 27  was required for TWEAK-induced HL-1 atrial myocytes hypertrophy.
In addition, we found that the expression of Fn14 protein was elevated in atrial appendages from patients with AF. These results suggest that inhibition of Fn14 may protect atrial myocytes against hypertrophy and even prevent the occurrence and progression of AF.
We further investigated the potential mechanism of TWEAK- The present study included some limitations. The major limitation of the study is that there is no AF animal model for validation in vivo. The number of human samples was limited, although statistical significance had been reached. Another limitation is that the serum levels of scavenger receptor CD163 were not detected.
In conclusion, we provide strong evidence that TWEAK induces

CONFLI CTS OF INTEREST
The authors confirm that there are no conflicts of interest. F I G U R E 8 Effects of STAT3 interference by siRNA on the expression of atrial hypertrophy in HL-1 myocytes. A, Protein expression of p-STAT3, STAT3, ANP and Troponin T after treatment with STAT3 siRNA. B, C, Quantitative analysis of A. *P < .05 vs control without treatment and # P < .05 vs only TWEAK