Fluorofenidone inhibits epithelial-mesenchymal transition in human lens epithelial cell line FHL 124: a promising therapeutic strategy against posterior capsular opacification

This content is licensed under a Creative Commons Attributions 4.0 International License. ABSTRACT | Purpose: The present study aimed to in­ vestigate the inhibitory effect of fluorofenidone against transforming growth factor β2­induced proliferation and epithelial­mesenchymal transition in human lens epithelial cell line FHL 124 and its potential mechanism. Methods: We evaluated the effect of fluorofenidone on proliferation and epithelial­mesenchymal transition of human lens epithelial cell line FHL 124 in vitro. After treatment with 0, 0.1, 0.2, 0.4, 0.6, and 1.0 mg/mL fluorofenidone, cell proliferation was measured via MTT assay. Cell viability was evaluated by lactate dehydrogenase activity from damaged cells. FHL 124 cells were treated with different transforming growth factor β2 concentrations (0­10 ng/mL) for 24 h and the expression of CTGF, α­SMA, COL­I, E­cadherin, and Fn were detected via quantitative polymerase chain reaction and Western blot analysis. After treatment with 0, 0.2, and 0.4 mg/ml fluorofenidone, the expressions of transforming growth factor β2 and SMADs were detected with real­time polymerase chain reaction and Western blot analysis. Expressions of CTGF, α­SMA, COL­I, and Fn were analyzed by immunocytochemistry assay. Results: The viability of FHL 124 cells was not inhibited when the fluorofenidone concentration was ≤0.4 mg/mL after the 24h treatment. Cytotoxicity was not detected via lactate dehydrogenase assay after the 24h and 36h treatment with 0.2 and 0.4 mg/mL fluorofenidone. Transforming growth factor β2 increased mRNA and protein expression of CTGF, α­SMA, COL­I, and Fn. However, fluorofenidone significantly suppressed expression of SMADs, CTGF, α­SMA, COL­I, and Fn in the absence or presence of transforming growth factor β2 stimulation. Conclusions: Fluorofenidone significantly inhibited expres­ sion of SMADs, CTGF, α­SMA, COL­I, and Fn in FHL 124 cells. Due to noncompliance in infants, fluorofenidone may become a novel therapeutic drug against posterior capsular opacification in infants.


INTRODUCTION
Posterior capsular opacification (PCO) frequently de velops after extracapsular cataract extraction or phacoe mulsification surgery, which significantly compromises visual outcomes. Furthermore, the postoperative recur rent rate of PCO in infants is 100%. The existing pharma cological treatments are unsatisfactory and have toxic side effects. PCO manifests Elschnig pearl, peripheral Soemmering's ring (central PCO in the visual axis), cor tex proliferation, and cholesterol crystal in some cases. It responds to neodymiumdoped yttrium aluminium garnet (YAG) laser capsulotomy quite well, and vision can be restored effectively and permanently in adults. However, in infants, laser capsulotomy is accompanied by unexpected visionrelated complications, such damage to intraocular lens, which is much more common in infants due to their noncompliance and vigorous prolife ration of human lens epithelial cells (HLECs). Even if the posterior capsule is perfectly removed, it is also much more likely for residual lens cells to migrate and proli ferate into the vitreous cavity in children than in adults.
As a consequence, retinal detachment may occur, which is definitely detrimental to children's visual health, and additional vitrectomy to eliminate vitreous lens cells and repair the retina increases the risks in patients. Briefly, in infant PCO, drug conservative therapy is much safer and need to be developed than laser treatment and surgery.
It is widely known that the lens epithelialmesen chymal transition (EMT) and migration from equator of anterior capsular to the center of posterior capsule are the common cytological bases of PCO (1) . EMT and collagen deposition are also the pathological processes in PCO. EMT is characterized by decreased expression of Ecadherin and increased expression of αSMA, and αSMA is an important sign of EMT and extracellular matrix (ECM) synthesis in HLECs (2) . Both TGFβ and CTGF play crucial roles in ECM synthesis and tissue fibrosis by combining with their respective receptors and promo ting cell differentiation and ECM (3) . Moreover, fibrosis and transdifferentiation of intraocular LECs, trabecular meshwork cells, and retinal pigment epithelial (RPE) cells induced by transforming growth factor β2 (TGFβ) and connective tissue growth factor (CTGF) have been considered pathological processes for many eye disea ses (4) . They also enhance cell proliferation, differen tiation, adhesion, and other important physiological activities (5) . Additionally, both TGFβ and CTGF induce EMT, αSMA, Fn, COLI, and collagen type IV but inhibit Ecadherin expression (6) .
Recently, a very promising drug, fluorofenidone, also known as AKFPD, is identified as an antiproliferative agent (7) . AKFPD is a pyridyl ketone compound with a broad spectrum of antifibrosis activities (8,9) . Moreover, previous studies have revealed its utility in inhibition of mouse renal fibrosis caused by diabetes and unilateral ureter obstruction (8) . AKFPD delivers satisfactory results in the treatment of renal interstitial fibrosis in preclinical research (10) . This drug has been demonstrated to possess multiorgan antifibrotic activities, such as in the kid ney (11) , lung (12) , and liver (13) . In the field of ophthalmology, the effect of AKFPD on the HLECs remains unexplored till now. This study aimed to determine the therapeutic effects of AKFPD in PCO and explore the related mole cular mechanism in vitro.

Culture and treatment of HLEC line
HLEC line FHL 124 was provided by ATCC (Manassas, VA, USA). The cells in this study have 99.5% homology with native human lens epithelium (14) . They were cultiva ted in culture dishes with DMEM containing 5% fetal bo vine serum (FBS). The cells were synchronized by repla cing the nutrient medium with serumfree DMEM and cultured for 24 h when the cells had 75% confluence.

Cell viability assay
The cell viability test was performed using two diffe rent methods: MTT assay and lactate dehydrogenase (LDH) activity.
The cell viability of lens cells in different groups was determined via MTT assay (Beyotime Company, Shanghai, China) to evaluate the effect on cell proliferation. Lens cells were cultivated in DMEM medium with 0 (control), 0.1, 0.2, 0.4, 0.6, or 1 mg/mL AKFPD (Beyotime Company, Shanghai, China) for 4, 8, 12, 24, or 36 h, respectively. The mixture was seeded into 96well plates at a density of 5000 cells/well. The cells were washed twice with PBS (HyClone, USA). A total of 25 μL of MTT (50 mg/mL) was added to each well, and the cells were incubated at 37°C for 4 h. Then, the culture medium of each well was replaced with 150 μL of dimethyl sulfoxide (Sigma) and shaken for 15 min. The absorbance was measured at 490 nm by a microplate reader (Thermo).
LDH is a glycolytic enzyme involved in pyruvate to lactic acid metabolism, which is present in almost all tissues or cytoplasm in the body. When the cell membrane is damaged, LDH is rapidly released. The detection of LDH was performed following the methods of Deng et al. (15) .
In LDH assay, LDH activity released from the dama ged cells was measured after treatment with 0 (control), 0.2 mg/mL, and 0.4 mg/mL AKFPD. The degree of cell damage was determined by detecting LDH activity in cell culture supernatant using a commercial LDH kit (Roche, Mannheim, Germany). Cell culture medium was proces sed, and the optical density (OD) was measured using a microplate reader at the wavelength of 450 nm. Cytoly sis percentage (% cytotoxicity ) was calculated: exp. (va lue − low control):(high control − low control).

Quantitative real-time PCR (qPCR)
The cells were treated with TGFβ2 at concentrations of 0 (control group), 1 ng/mL, 5 ng/mL, or 10 ng/mL for 24h. Total cell RNAs were extracted using a TRIzol total RNA extraction kit (Invitrogen Company, Shanghai, China) following the manufacturer's instructions. Then, reverse transcription was performed using cDNA syn thesis kit from Fermentas Co., Ltd. (Lithuania). The pri mer pairs used are presented in table 1. qPCR was performed on BioRad IQ5 thermal cycler (BioRad, California, USA). The results were analyzed with BioQ software to obtain Ct value for each PCR, and the △△Ct method was used to quantify the levels of gene expression.

Western blot analysis
Lens cells were treated with TGFβ2 at concentra tions of 0 (control group), 0.5 ng/mL, 1 ng/mL, 5 ng/mL, or 10 ng/mL for 24 h. The monolayer cultures were collected with cell scrapers and then lysed with 100 μL of cell lysis buffer on ice for 30 min. The cell lysates were centrifuged, and supernatants were collected. To tal protein was prepared from each group. The protein concentrations in the supernatants were aliquoted and kept using the BCA method (Biocolor, Shanghai, China) for further experiments. A total of 50 μg protein per sample was electrophoresed by 10% polyacrylamide gel electrophoresis and transferred to nitrocellulose mem brane (Millipore, Billerica, MA). It was blocked with 5% skimmed milk for 1 h at room temperature and incuba ted overnight at 4 o C with mouse monoclonal antibodies specific to CTGF (Millipore), αSMA (Millipore, USA), Fn (Abcam, UK), Col1 (Proteintech, USA), and Ecadherin (Millipore) at 4°C. After washing, the membrane was in ACATTGGATGGGAGGCTTCA GATCAGGCCACCTCCA GAGA cubated with secondary antibodies (antimouse antibo dy conjugated, Abcam, Cambridge, UK). The membrane was immersed in enhanced chemiluminescence solution and then exposed to an Xray film. After hybridization of secondary antibodies, the resulting images were analyzed with ChemiImager 4000 (Alpha Innotech Cor poration, California, USA).

Inhibitory effect of AKF-PD on FHL 124 cells
Inhibitory effect of AKF-PD on TGF-β2, SMAD3, and SMAD4 in lens cells The cells were treated with AKFPD at concentrations of 0 (control group), 0.2, or 0.4 mg/mL for 24h. qPCR qPCR were performed as mentioned above. The pri mer pairs that we used were presented in table 1.

Inhibitory effect of AKF-PD on TGF-β2-induced SMAD3 and SMAD4 expression in lens cells
Lens cells were treated with AKFPD at concentrations of 0 (control group), 0.2, or 0.4 mg/mL in the presence of 10 ng/mL TGFβ2 for 24 h. qPCR and Western blot analysis were performed as mentioned above.

Morphology of lens cells in vitro
Lens cells were seeded into culture dishes with DMEM containing 5% FBS. The HLECs were synchroni zed by replacing the nutrient medium with serumfree DMEM and cultured for 24 h when the cells had 75% confluence.
The morphology changes were observed after the treatment.

Immunocytochemistry
FHL 124 cells were cultivated at a density of 6×104 cells/mL. The cells were fixed with 4% paraformaldehyde for 16 min. Then, the fixed cells were treated with 0.1% Triton X100 for 10 min. The cells were subsequently incubated in 3% H2O2 for 10 min. The cells were blo cked in 5% goat serum for 20 min and then incubated with mouse antihuman COLI, Fn, CTGF, and αSMA (1:100 dilution) for one night. Following three washes with PBS, the cells were treated with secondary antibody (polymer helper and polyperoxidaseantimouse IgG) for 30 min at 37 o C. The cells were treated with DAB reagent box (ZSGBBIO Company, Beijing, China). FHL 124 cells were stained with hematoxylin for 20 s. The slides were embedded in neutral balsam. The slides were observed through a microscope. Representative images were cap tured with the incorporated digital camera (Olympus image analysis system, Japan). The average positive OD was determined and analyzed by image analysis system.

Image acquisition and statistical analysis
SPSS 13.0 statistical software was employed to con duct all statistical analyses. After treatment of lens cells with different TGFβ2 and AKFPD concentrations, the overall comparison of protein and mRNA expressions with control group was analyzed using oneway analysis of variance, while the difference between groups was compared using Tukey honestly significant difference test. Differences with pvalue <0.05 were considered statistically significant.

FHL 124 cell proliferation inhibited by AKF-PD was detected via MTT assay and LDH activity (Figure 1)
In MTT assay, there was no statistically significant difference in the control, 0.1 mg/mL, 0.2 mg/mL, and 0.4 mg/mL groups at all time points. However, cell proli feration was inhibited when AKFPD concentration was ≥0.6 mg/mL compared with the control group (*p <0.05; **p<0.01) at all time points. The effect was more re markable in the 1.0 mg/mL group at 36 h (**p<0.01). From the discussion above, the following studies were performed with 0.2 and 0.4 mg/mL AKFPD after the 24h treatment.
In LDH assay, after treated with AKFPD for 24h, the cytolysis percentages were 13.90 ± 2.9, 13.74 ± 3.2, and 15.41 ± 4.7 for the control, 0.2, and 0.4 mg/mL In MTT assay, cell proliferation was inhibited when AKF-PD concentration was ≥0.6 mg/mL compared with the control group. In LDH assay, after the 24-or 36-h treatment with AKF-PD (0, 0.2, and 0.4 mg/mL), no statistically significant difference was found between the percentage of cytolysis in AKF-PD-treated groups or control group (p>0.05). *p<0.05 and **p<0.01 were obtained comparisons between treatment and control groups at different time points. groups, respectively, and the cytolysis percentages after 36h were 14.10 ± 3.0, 13.84 ± 2.9, and 15.27 ± 4.5 for the control, 0.2, and 0.4 mg/mL AKFPD groups. There was also no significant difference among the groups at 24 and 36 h, respectively (p>0.05).
Effect of TGF-β2 on the mRNA and protein expression of CTGF, α-SMA, COL-I,

E-cadherin, and Fn in lens cells (Figure 2)
It is confirmed in our experiment that, after the 24h treatment of lens cells with TGFβ2, mRNA (*p<0.05; **p<0.01 compared with the control group) and pro

Effect of AKF-PD on the mRNA and protein expression of TGF-β2, SMAD3, and SMAD4 in lens cells (Figure 3)
Our experiment showed that, after the 24h treatment of lens cells with AKFPD, the mRNA and protein levels of TGFβ2, SMAD3, and SMAD4 were dose dependently downregulated. The mRNA and protein levels of TGFβ2, SMAD3, and SMAD4 were significantly inhibited at 0.4 mg/mL (*p<0.05, **p<0.01) compared with the control group. However, the inhibitory effects on TGFβ2 and SMAD4 by Western blot analysis were not statistically significant in the 0.2 mg/mL AKFPD groups (p>0.05).

Inhibitory effect of AKF-PD on TGF-β2 induced expression of SMAD mRNA and protein in lens cells (Figure 4)
Further experiments revealed that, in the presence of TGFβ2, AKFPD also decreased the expression of the mRNA and protein levels of SMAD3 and SMAD4 dose dependently (*p<0.05; **p<0.01). However, the inhi bitory effect on mRNA and protein expression of SMADs was detected but was not statistically significant in the 0.2 mg/mL AKFPD groups (p>0.05).

Morphology observation of lens cells in vitro
There was no morphological change in lens cells trea ted with 0.4 mg/mL AKFPD (C) compared with control cells (B) after 24 h. Cells were spindled, starred, elongated, or irregular in shape after treatment with 10 ng/mL TGFβ2 (D), which were inhibited by 0.4 mg/mL AKFPD (A) ( Figure 5).

Protein expression of COL-I, Fn, CTGF,
and α-SMA was also analyzed by immunocytochemistry assay (Figure 6) After treatment with 0.4 mg/mL AKFPD for 24h, the protein expression of COLI, Fn, CTGF, and αSMA was significantly decreased (*p<0.05). The brown stains were highly densified by TGFβ2; however, they were faded by AKFPD (*p<0.05).  of pulmonary hypertension induced by hypoxia in rats through its regulation of TGFβ expression and syn thesis of ECM (16) . Besides targeting TGFβSMADs signal pathway, AKFPD attenuates inflammation by inhibiting the NFкB pathway in human proximal tubule cells (17) . Furthermore, through blockage of the Fas/Fas L path way (18) , AKFPD inhibits AngIIinduced apoptosis of re nal tubular cells, which can be initiated by the binding of lethal ligands, such as FAS/CD95 ligand, tumor necrosis factor (TNF)α, and TNF (ligand) superfamily member 10 (best known as TNFrelated apoptosisinducing li gand), to various death receptors (i.e., FAS/CD95, TNFα receptor 1, and TNFrelated apoptosisinducing ligand receptors 1 and 2, respectively) (19) . AKFPD is a newly developed drug with antifibrotic activities through inhi biting various signal pathways and sheds new light on treatments of progressive fibrotic diseases (20) .
AKFPD shows inhibitory effects through SMAD signal pathways (20) . The function of TGFβ in promoting fibrosis is principally mediated by the SMAD signaling pathway (21,22) . As an activator, TGFβ unites with TβRII (one of the TGFβ ligands) to form the TβRII TGFβTβRI tripolymer. The binding of TGFβ with its receptor II (TβRII) activates the kinase of TGFβ receptor I (TβRI). TβRI is phosphorylated and then phosphorylates SMAD2 and SMAD3. Subsequently, phosphorylated SMAD2 and SMAD3 bind to SMAD4 to constitute a SMAD complex. Then, the complex is shifted to the nucleus to regulate the transcription of target genes (23) . Finally, downstream biological effects were activated by SMAD pathway, where TGFβ signaling is activated and fibrosis of related tissue is enhanced by CTGF (24) . Moreover, there were increased expressions of CTGFmRNA accompanied with stronger synthesis of collagen I and αSMA in the residual debris of PCO (25) . Furthermore, αSMA is an important sign of EMT and ECM synthesis in HLECs. Mo reover, both αSMA and Ecadherin, which are involved in EMT in HLECs (26) , help mediate cellmatrix adherence and myofibroblast (27) .
There are other signaling pathways, including ERK1/2, p38 MAPK, JNK, STAT3, and PKC. These pa thways are also involved in the TGFβinduced upregu lation of CTGF expression in other cell types (28) . Many other transcription factors and microRNAs also regulate CTGF expression (29) . CTGF can promote cell mitosis and proliferation of fibroblasts and synthesize collagen, mediate cell adhesion, enhance fibrosis, and regulate ECM synthesis (30) .

DISCUSSION
Previous studies have proven that AKFPD functions as an antifibrotic agent in the pulmonary and renal fibrosis models (16) . It also ameliorates the progression In this study, the authors initially demonstrated an inhibitory effect of AKFPD on TGFβ2induced prolife ration and EMT of HLEC line FHL 124. The effect acted in a dosedependent manner. The authors found that cell proliferation was significantly suppressed at 0.6 mg/mL AKFPD, and inhibition reached its climax at 1.0 mg/mL. Moreover, LDH assay indicated that there were no significant toxic effects at the concentrations of 0, 0.2, and 0.4 mg/mL. This result showed a hopeful prospect that AKFPD may become a new therapeutic drug for PCO. The authors provided new evidence that TGFβ2 increased the expression of CTGF, αSMA, COLI, and Fn but decreased the expression of Ecadherin in the cell line. In contrast, AKFPD showed its inhibitory effect by depressing TGFβ2SMAD signaling pathway: AKFPD suppressed expression of SMADs, CTGF, αSMA, COLI, and Fn, in the absence or presence of TGFβ2 stimula tion, dose dependently.
Although AKFPD showed remarkable suppression on proliferation and EMT of HLECs, there are still other issues that need to be managed. Cytotoxicity to corneal endothelial cell needs to be further detected. Due to nonregeneration of these cells, drugs that interfere with cell metabolism must be safe and nontoxic. The metabolism of RPE cells after AKFPD treatment needs to be elucidated through the following studies. RPE is an important barrier for the retinal vessel and nerve cells. The RPE layer provides a stable microenvironment that prevents the leakage from choroid vessel and is extre mely vital for normal retinal metabolism. In vivo studies are also urgent for the whole AKFPD experimental se ries. Since lens cells are exposed to the aqueous humor in vivo, whether the liquid environment influences the effects of AKFPD on lens must be investigated in the following steps. In the future use of eyedrops, nanopar ticles could be applied to prolong and enhance the drug retention (31) . Nanoparticles are ideal and successful con trol release carriers for many medications (31) . An in vivo study on rabbit eyes confirmed that chitosansodium alginate nanoparticles could increase the 5FU level in the aqueous humor compared to the native 5FU solu tion (32) . Whether nanoparticlesAKFPD can effectively inhibit HLEC EMT in vivo and how it influences corneal endothelium need to be explored in future experiments.
Therefore, AKFPD exhibited powerful inhibitory effect on EMT of FHL 124 cells in the absence or pre sence of TGFβ2 stimulation in vitro. In addition, the inhibition on lens cells was not mediated by cytotoxicity. It may be a new strategy to inhibit EMT and prevent or treat PCO. Since laser capsulectomy is unavailable for infants, this strategy provides a promising therapy for infant PCO.