FGF21 attenuates pulmonary arterial hypertension via downregulation of miR‐130, which targets PPARγ

Abstract The proliferation, migration and apoptotic resistance of pulmonary artery smooth muscle cells (PASMCs) are central to the progression of pulmonary arterial hypertension (PAH). Our previous study identified that fibroblast growth factor 21 (FGF21) regulates signalling pathway molecules, such as peroxisome proliferator‐activated receptor gamma (PPARγ), to play an important role in PAH treatment. However, the biological roles of miRNAs in these effects are not yet clear. In this study, using miRNA sequencing and real‐time PCR, we found that FGF21 treatment inhibited miR‐130 elevation in hypoxia‐induced PAH in vitro and in vivo. Dual luciferase reporter gene assays showed that miR‐130 directly negatively regulates PPARγ expression. Inhibition of miR‐130 expression suppressed abnormal proliferation, migration and apoptotic resistance in hypoxic PASMCs, and this effect was corrected upon PPARγ knockdown. Both the ameliorative effect of FGF21 on pulmonary vascular remodelling and the inhibitory effect on proliferation, migration and apoptotic resistance in PASMCs were observed following exogenous administration of miR‐130 agomir. In conclusion, this study revealed the protective effect and mechanism of FGF21 on PAH through regulation of the miR‐130/PPARγ axis, providing new ideas for the development of potential drugs for PAH based on FGF21.

Fibroblast growth factor 21 (FGF21) is an endocrine hormone with multiple biological functions that has demonstrated multiple biological effects. [5][6][7] A growing body of data suggests that FGF21 is an important vasoprotective factor. 8,9 Our previous studies reported that FGF21 can inhibit hypoxia-induced PASMC proliferation and migration by upregulating Peroxisome proliferator-activated receptor gamma (PPARγ), promoting apoptosis and thereby downregulating inflammatory cytokine levels, improving collagen deposition in the lung and attenuating hypoxia-induced PAH. 10,11 However, the exact mechanism of action remains unclear.
MicroRNAs (miRNAs) are highly conserved non-coding RNAs that are expressed endogenously, and their dysregulation is an important factor in the development of cardiovascular injury, such as congestive heart failure and PAH. 12,13 Studies have shown that
Hyp groups were maintained in an atmosphere of 5% O 2 , 90% N 2 and 5% CO 2 at 37 °C for 48 h. All groups were given treatment at the beginning of modelling.

| Western blot analysis
Lung tissues were homogenized in cold radioimmunoprecipitation assay (RIPA) lysis buffer using an automatic homogenizer (FastPrep-24 5G, MP Biomedicals) and lysed using an ultrasonic disruptor. The supernatants were collected after the homogenates were centrifuged (12, (1:1000). They were then incubated with goat anti-rabbit secondary antibody (1:10000) labelled with HRP at room temperature for 1 h. To detect Cyt C release, mitochondrial and cytosolic pellets were isolated using a Mitochondria Isolation Kit and immunoblotted with antibodies against Cyt C (1:1,000); COX IV served as the mitochondrial marker, and β-actin was used as the cytosolic marker. After the pellets were triple washed with phosphate-buffered saline, the protein bands were examined using the Bio-Rad ChemiDoc MP imaging system and the quantification was performed using Image Lab software (Bio-Rad).

| Proliferation assays
The cells were seeded into 96-well plates at a density of 1 × 10 4 cells/ well for 48 h. Then, 10 µl of cell counting kit-8 (CCK8, Dojindo) reagent were added to each well, the cells were incubated at 37°C for 2 h and absorbance was measured at 450 nm. A 5-ethynyl-20-deoxyuridine cell proliferation assay kit (Abcam) was used to analyse cell proliferation.

| Detection of apoptosis
A One-Step TUNEL Apoptosis Assay Kit was used to detect apoptosis. The cells were seeded into 24-well plates at a density of 2 × 10 4 cells/well for 48 h. After washing with PBS three times, the cells were immobilized in 4% paraformaldehyde for 30 min and permeabilized with 0.1% Triton X-100. The cells were then incubated with the TUNEL reaction mixture, as instructed. Fluorescence images were captured using a fluorescence microscope (Leica DMi8).

| Wound-healing assay
The cells were seeded into six-well plates at a density of 2 × 10 5 cells/ well for 24 h. The cells were then serum-deprived for 24 h.
Afterwards, the cell monolayers were scratched with 200µl pipette tips in the centre of the well and incubated with the experimental treatments for 48 h. ImageJ (NIH, USA) was used to determine the level of wound-healing cover.

| Animal models
Male C57BL/6 mice (8-12 weeks old, 20-25 g) were purchased from Vital River Laboratory Animal Technology. The experimental scheme and the animal housing were approved by the Animal Ethics Committee of Wenzhou Medical University. The mice were raised in a humidity range of 55%-65% and 20-24°C. To verify the function of FGF21 and the change in miR-130 expression, mice were divided into three groups randomly: normoxia (Nor, saline-treated) group, hy-

| Ethics approval
All animal procedures conformed to the guidelines from Directive 2010/63/EU of the European Parliament on the protection of animals used for scientific purposes or the current NIH guidelines and were approved by the Animal Ethics Committee of Wenzhou Medical University. All animals were handled with care and euthanized humanely during the study.

| Echocardiography
Transthoracic echocardiography was performed using a Visual Sonics Vevo 3100 ultrasound machine. After anaesthetization with continuous isoflurane inhalation (1.5%-3.0%), mice were placed on a heated pad in a supine position, and the fur on their chest was removed with a chemical hair remover. The pulmonary artery acceleration time (PAAT) and velocity time integral (VTI) were obtained from the modified parasternal long-axis view using the pulsed Doppler mode.

| Measurement of haemodynamic and right ventricular hypertrophy
The mice were anaesthetized with 20% urethane (1 ml/100 g, i.p.), and catheters were inserted into the right ventricular (RV) and left carotid arteries. 10 Right ventricular systolic pressure (RVSP) was re-

| Measurement of pulmonary arterial remodelling and arterial collagen deposition
Pulmonary arterial remodelling was assessed using the method described by Cai et al. 10 Briefly, connective tissue around the lung was removed, and the lung tissue was fixed in 4% paraformaldehyde for 24 h. Then, the lung tissue was cut into 5µm thick slices after being embedded in paraffin. Sections were then stained with haematoxylin-eosin (H&E), and an optical microscope was used to evaluate the structure of the pulmonary arteries with external diameters of 25-100 µm. Image-Pro Plus 6.0, (Media Cybernetics) was used to analyse the ratio of the wall thickness (WT) to the total thickness (TT, WT/TT) and the pulmonary artery wall area (WA) to the total area (TA, WA/TA). To assess arterial collagen deposition, sections were stained with Masson staining. An optical microscope was used to observe arteries with external diameters of 25-100 µm.
The collagen fibres were blue, and the erythrocytes, cytoplasm and muscle fibres were stained red. The ratios of the collagen fibre area to the pulmonary artery WA were analysed using ImageJ (NIH).

| sRNA sequencing
Total RNA was extracted using the TRIzol reagent (Invitrogen).
The quantity and integrity of the products were evaluated using

| qRT-PCR
MiRNA was extracted from PASMCs using a SanPrep Column mi-croRNA Extraction Kit (Sangon Biotech, Shanghai, China) following the manufacturer's instructions. RNA was quantified using an ultraviolet spectrophotometer, and only samples with an OD260/ OD280 ratio greater than 1.8, were used for experiments. The cDNA was synthesized using a miRNA First-Strand cDNA Synthesis

Kit (stem-loop method) from Sangon Biotech (Shanghai, China).
Real-time PCR was performed using a MicroRNA qPCR Kit (SYBR Green; Sangon Biotech, Shanghai, China) according to the manufacturer's protocol. U6 small nuclear RNA (snRNA) was selected as an endogenous control for miR-130, and the 2 −ΔΔCt method was used to calculate the relative expression of the detected gene. The primer sequences used for miRNA and mRNA analysis are shown in Table S1 and S2.
Firefly luciferase was selected as a reporter, and Renilla luciferase was selected as a control. PASMCs were transfected with 50 nM miR-130 mimic and 100 ng pmirGLO vector using the Attractene Transfection Reagent (Qiagen). Luciferase activity was measured using a Dual-Luciferase ® Reporter Assay System (E1910, Promega) after incubation for 48 h.

| Drawing software
Statistical histograms were drawn using GraphPad Prism 7.0 (Graph Pad Software Inc.), and the overall layout was constructed using Adobe Illustrator CS6 (Adobe Illustrator Software Inc).

| Statistical analysis
Statistical analyses were performed using GraphPad Prism 7.0 (Graph Pad Software Inc.). The results are presented as the means ±standard deviation (SD). Student's t-test was used to compare the two groups. Multiple comparisons were performed using one-way ANOVA. Statistical significance was set at p < 0.05.

| FGF21 can alleviate hypoxia-induced pulmonary arterial remodelling and reverse hypoxiainduced PAH
As shown in Figure 1A-C, RVSP and RV/(LV+S) were significantly increased in the Hyp group compared with the Nor group (p < 0.01), and FGF21 significantly inhibited the hypoxia-induced increase in pulmonary artery pressure and ventricular remodelling level (p < 0.05, p < 0.01). There was no significant difference in heart rate and body weight between the groups ( Figure 1D).
Moreover, sections stained with H&E showed that hypoxiainduced pulmonary arterial remodelling was alleviated in the Hyp+FGF21 group, compared with the Hyp group and WA/TA (%) and WT/TT (%) showed the same trend (p < 0.01; Figure 1E,G).
In conclusion, consistent with previous findings, 10,11 FGF21 can improve hypoxia-induced pulmonary remodelling and alleviate hypoxia-induced PH.

| The expression of miR-130 increases in PAH mice, and FGF21 can significantly downregulate miR-130 expression
To explore the internal mechanism of FGF21 in the treatment of PAH, lung samples from the Hyp and Hyp+FGF21 groups were collected for miRNA high-throughput sequencing. We found a significant reduction in miRNAs after FGF21 intervention and identified the top 15 KEGG pathways in the lung (Figure 2A-C). PCR was used to verify the screened differentially expressed miRNAs, and it was found that miR-130 expression in the hypoxia group was significantly higher than in the normoxia group, and FGF21 treatment significantly downregulated its level (p < 0.01; Figure 2D,E). This evidence suggests that FGF21 may alleviate PAH by downregulating the expression of miR-130.

| MiR-130 can directly target PPARγ and inhibit its expression
Based on the bioinformatics database, miR-130 was predicted to target the PPARγ gene ( Figure 2F). This correlated with the expectations.
Accumulating evidence suggests that PPARγ reverses pulmonary artery remodelling, PAH and RV hypertrophy (RVH). [20][21][22] Our study also demonstrated that PPARγ inhibits hypoxia-induced PASMC proliferation and migration while promoting apoptosis (p < 0.05/p < 0.01; Figure S1-S3). To further confirm this hypothesis, we performed a luciferase reporter gene assay to test the direct interaction between PPARγ and miR-130. The results showed that the fluorescence activity of miR-130 mimic transfection was significantly reduced compared with control miRNA transfection (p < 0.01; Figure 2G); however, no

| MiR-130 promotes hypoxia-induced PASMC proliferation and migration and inhibits apoptosis by regulating PPARγ expression
The functional effects of miR-130 on PPARγ were also studied. In PASMCs, siRNA-knockdown efficiency was first analysed using a  Figure 4G). These results indicate that miR-130 can inhibit apoptosis by regulating PPARγ expression.

| FGF21 reduces proliferation and migration and enhances apoptosis of hypoxia-induced PASMCs by inhibiting the negative regulatory effects of miR-130 on PPARγ
To confirm whether FGF21 regulates hypoxia-induced PASMC pro- Upon FGF21 treatment, the Bax/Bcl-2 ratio was increased in hypoxic PASMCs and mice (p < 0.01), which was completely abolished by the addition of miR-130 (p < 0.01; Figure 6A,B). The cleaved caspase-3/caspase-3 ratio, AIF expression and the release of Cyt C from the mitochondria to the cytoplasm showed the same trend both in vivo and in vitro (p < 0.01/p < 0.05; Figure 6C-L). The TUNEL assay also showed the same results (p < 0.01; Figure 6M).
These results indicate that FGF21 enhances apoptosis in hypoxiainduced PASMCs by inhibiting the negative regulatory effects of miR-130 on PPARγ.

| FGF21 attenuates chronic hypoxia-induced PAH by inhibiting the negative regulatory effects of miR-130 on PPARγ
To further verify that FGF21 regulates PPARγ expression by regulating miR-130, mice were exposed to hypoxia for three weeks and  Figure 7D). There was no significant difference in heart rate and body weight between the groups ( Figure 7E,F). Transthoracic echocardiography showed that FGF21 could inhibit hypoxia-induced downregulation of PAAT and PAVTI, and miR-130 agomir reversed these changes (p < 0.01; Figure 7B,G,H). Immunofluorescence results showed that PASMC proliferation was obviously stimulated under hypoxic conditions, while FGF21 prominently reduced this increased proliferation, and this effect could be weakened by exogenous miR-130 ( Figure 7I). Moreover, H&E-stained sections showed that hypoxia-induced pulmonary arterial remodelling was alleviated in the Hyp+FGF21 group, compared with the Hyp group, which could be reversed by exogenous miR-130 agomir administration ( Figure 7J). As expected, FGF21 reversed the hypoxia-induced increase in the pulmonary artery WT/TT ratio (%) and WA/TA ratio (%), while these effects were substantially alleviated in miR-130-treated mice (p < 0.01; Figure

| DISCUSS ION
PAH is a chronic progressive disease characterized by pulmonary vascular remodelling, ultimately leading to RVH and even death. [23][24][25] Abnormal proliferation, migration and apoptotic resistance in PASMCs are central aspects of pulmonary vascular remodelling. The absence of specific drugs that directly target the vascular remodelling process has led to high end-stage annual mortality in PAH. 26,27 Therefore, the search for potential drugs that modulate PASMCs has been a research hotspot in PAH in recent years. In this study, Data are presented as the mean ±SD. *p < 0.05, **p < 0.01 muscle cell proliferation. 28,29 Our previous study found that exogenous FGF21 upregulated the PPARγ expression, inhibited hypoxia-induced PASMC proliferation, improved pulmonary artery remodelling and reduced pulmonary artery pressure, 10 A recent report claimed that miR-130 could exacerbate myocardial injury caused by acute myocardial infarction by targeting PPARγ. 31 We also predicted that PPARγ is one of the direct targets of miR-130 through bioinformatics analysis. This result is expected, as our studies 10,11 and those of others 22  This study also has certain shortcomings, for example, the relationship between miR-130 and PPARγ was predicted based on previous experimental findings and the bioinformatics approach. Although the association was illustrated to some extent, direct evidence needs to be obtained by constructing lung-specific PPARγ-knockout mice.
In addition, the exact role of the FGF21/miR-130/PPARγ axis in PAH needs to be further validated in combination with clinical samples.

| CON CLUS IONS
This study identified a new mechanism, by which FGF21 attenuates hypoxia-induced PAH by inhibiting the negative regulation of miR-130 on PPARγ (Figure 8), further enriching the understanding of the regulatory molecular mechanism of FGF21 in PAH. This provides a new idea for epigenetics-based PAH treatment.

CO N FLI C T O F I NTE R E S T
The authors have declared that no competing interest exists.
F I G U R E 6 FGF21 enhances apoptosis in hypoxia-induced PASMCs by inhibiting the negative regulatory effects of miR-130 on PPARγ.
(A, C, E) Western blotting for Bax, Bcl-2, cleaved caspase-3, caspase-3, AIF expression in PASMCs in Nor+MC, Hyp+MC, Hyp+MC+FGF21, Hyp+FGF21+miR-130 mimic groups, β-actin was used as a loading control (n = 4). (B, D, I) Western blotting for Bax, Bcl-2, cleaved caspase-3, caspase-3 and AIF in lung homogenates of mice (n = 6). (F-H) The expression levels of Cyt C in mitochondrial and cytosol pellets in PASMCs were examined by western blotting with antibodies against Cyt C with COX IV as a mitochondria marker and β-actin as the internal control (n = 4). (J-L) The expression levels of Cyt C in mitochondrial and cytosol pellets in lung homogenates of mice were examined by western blotting with antibodies against Cyt C with COX IV as a mitochondria marker and β-actin as the internal control (n = 6). (M) The apoptosis index of PASMCs in each group was measured by TUNEL assay (n = 10) (×200; scale bars indicate 100 µm) and is shown as the ratio of TUNEL positive cells (red) to total cells (blue). Data are presented as the mean ±SD. *p < 0.05, **p < 0.01