2',3'‐Cyclic‐nucleotide 3'‐phosphodiesterase contributes to epithelial‐mesenchymal transition of lens epithelial cells through the notch signalling pathway

Abstract Objectives Fibrosis is a complex process involved in multiple diseases that result in organ injury and failure. Cataract, one common form of ocular fibrosis, is a main cause of blindness worldwide, and surgery may be the only cure. In this regard, epithelial‐mesenchymal transition (EMT) of lens epithelial cells (LECs) is the primary cause of anterior subcapsular cataract (ASC). This study aimed to investigate the mechanism by which 2',3'‐cyclic‐nucleotide 3'‐phosphodiesterase (CNPase) regulates the function of EMT in LECs. Materials and Methods A mouse model of ASC was used to observe the expression of CNPase in the lens and correlate its expression changes with lens EMT. Furthermore, the effects of CNPase on cell migration and cell proliferation were evaluated by transwell migration, wound healing and EdU staining assays. Finally, Western blotting and immunofluorescence were used to assess the mechanical properties potentially involved in the regulation of EMT by CNPase. Results The expression of CNPase was upregulated in LECs during the EMT process in mice with ASC. Notably, CNPase significantly promoted the proliferation, migration and EMT of LECs in vitro. Interestingly, the EMT‐promoting mechanism of CNPase may be achieved by targeting the Notch signalling pathway. Conclusions Considering the involvement of EMT in ASC, both CNPase and the Notch signalling pathway may be therapeutic targets for the treatment of cataracts.

common mechanism underlying most cataracts, which are responsible for a large proportion of blindness worldwide. Cataract is a common age-related disease characterized by a progressive increase in lens opacity and light obstruction and a gradual loss of vision. 9,10 It has been documented that 96% of the population over 60 years of age have varying degrees of lens opacity. With the ageing of the world's population, the incidence of cataracts is increasing rapidly. 11,12 Cataracts are basically divided into two types by the location of fibrosis: anterior subcapsular cataract (ASC) and posterior capsular opacification (PCO). The proliferation and migration of lens epithelial cells (LECs) to the acellular posterior capsule and their ultimate conversion to myofibroblast-like cells, known as EMT, are the common causes of ASC and PCO. TGF-β is considered the pivotal factor in inducing EMT during development, cancer and other pathological conditions. In some epithelial cell lines cultured in vitro, EMT can be induced by TGF-β stimulation alone; furthermore, EMT is mediated by the Notch, Wnt/β-catenin and integrin signalling pathways. 13,14 Therefore, the regulation of EMT has been considered a therapeutic strategy to prevent the occurrence of ASC and PCO and to reduce the incidence of postoperative complications. 7,15,16 Abnormal Notch signalling has been found in a variety of fibrotic diseases. 16,17 Inactivation of Notch signalling plays a key role in reversing the EMT phenotype, improving symptoms and overall survival in patients with fibrosis. 17 It has currently been established that there are four cell surface transmembrane receptors (Notch 1-4) in mammals. The Notch receptor is cleaved by γ-secretase, releasing the Notch intracellular domain (NICD), which is then transferred to the nucleus to regulate downstream transcription factors, including Hes and Hey. 18,19 Previous studies have suggested that Notch signalling interacts with multiple signalling pathways, in which adenosine induces Notch signalling through corresponding receptors, suggesting that the specific regulation of adenosine metabolism may treat fibrotic diseases through the Notch pathway.
2',3'-Cyclic-nucleotide 3'-phosphodiesterase (CNPase) is a key enzyme involved in the process of adenosine metabolism that catalyses the hydrolysis of 2',3'-cyclic nucleotides to the corresponding 2'-adenosine monophosphate (AMP). These AMPs are then converted to adenosine. 20 Studies have shown that traumatic brain injury triggers the abnormal expression of CNPase, which is related to an increase in AMP and adenosine in the cerebrospinal fluid (CSF).
Activation of the mitochondrial permeability transition pore (MPTP) by AMP is hypothesized to lead to apoptosis and necrosis. 21,22 In the brain, a lack of CNPase leads to increased susceptibility to brain damage and neurological diseases. 23 Surprisingly, the lack of CNPase in the kidney protects against kidney damage, possibly by preventing the formation of 2'-AMP and reducing mitochondrial damage. 24,25 Therefore, the exact roles of 2',3'-cNMP and its downstream metabolites remain to be fully elucidated.
In this study, we focused on the role of CNPase in lens epithelial cell EMT. We used a mouse ASC model to observe the expression of CNPase in the lens and correlate its expression changes with lens EMT. Here, we show a possibly unreported mechanism of CNPase that regulates the function of EMT by directly targeting Notch signalling. These results suggest that CNPase and the drug-targeted Notch pathway may hold promise in the design of therapeutic strategies for preventing and treating ASC.

| ASC cataract mouse model
All animal experiments were approved by the Animal Ethics Committee of Kunming Medical University. The mouse model of injury-induced ASC was established as described previously. 26 Before the surgical procedure, the mice (body weight [BW] = 30-50 g, n = 6, male) were given a single intraperitoneal injection of pentobarbital sodium (70 mg/kg). The pupil was dilated with tropicamide eye drops. A small incision was then made in the central anterior capsule of the lens in the right eye of the mouse with a 0.26-mm thick needle blade. The damage depth was approximately 280 μm, which is approximately 1/4 of the partial length of the needle. Animals were allowed to heal for 5, 7 and 14 days before being sacrificed. The lens was removed and paraffin-embedded, and coronal sections were then taken for immunofluorescence staining haematoxylin-eosin (HE) and Masson staining.

| Haematoxylin-eosin staining and Masson staining
Paraffin sections were stained with haematoxylin solution (Solarbio, Beijing, China, No.: G1120) for 5 minutes, dipped in a hydrochloric acid-alcohol solution for 5 seconds and washed with distilled water.
Then, slices were stained with eosin solution (Solarbio, No.: G1120) for 3 minutes. Next, sections were dehydrated in a graded alcohol series and cleared in xylene. Finally, sections were observed and imaged under an optical microscope (Olympus).
Masson staining was performed according to the manufacturer's protocol (Solarbio, No.: G1340). Slices were dewaxed conventionally in water, stained with Weigert's iron haematoxylin for 5-10 minutes, differentiated in a hydrochloric acid-alcohol solution for 10 seconds and washed with distilled water. After being stained with Biebrich scarlet-acid fuchsin solution for 5-10 minutes, sections were washed with 2% aqueous glacial acetic acid solution for 1 minute and were then treated with an aqueous solution of phosphomolybdic acid for 1-2 minutes. Sections were then stained directly with aniline blue solution for 1-2 minutes and washed again with acid solution. A 95% ethanol solution was used for dehydration, and slides were sealed with neutral resin. Slides were then observed and imaged under a microscope (Olympus).

| Silencing of CNPase with small interfering RNA (siRNA)
To knock down CNPase, either non-silencing siRNA or three kinds of

| Lentivirus-mediated overexpression of CNPase
Lentivirus overexpressing CNPase was purchased from GenePharma. Prior to transfection, SRA 01/04 cells were inoculated in 6-well plates, at 70%-80% confluence, the cells were infected with lentivirus. After 48 hours, the infected cells were supplemented with fresh normal medium. Cells with stable CNPase overexpression were obtained by puromycin selection. Transfected cells were further treated with 10 μmol/L DAPT (a specific inhibitor of Notch receptor cleavage, Sigma-Aldrich) to explore the relationship between CNPase and the Notch signalling pathway.
Total cellular protein contents were determined by the BCA method.   For paraffin-embedded slides, sections were deparaffinized and rehydrated according to standard procedures. Antigen was retrieved by heating for 10 minutes in 0.01 mol/L sodium citrate buffer in a pressure cooker. After treatment as described above, immunofluorescence staining was performed to observe changes in the expression of vimentin and CNPase in the ASC mouse model. Fluorescence labelling was visualized and quantified with the tools described above.

| Wound healing assay
Transfected SRA 01/04 cells were seeded into 12-well plates and cultured until 100% confluency. The cells were scratched using a sterile plastic pipette tip (10 μL). Following this, the cells were washed with PBS to remove debris. Then, fresh medium was added to each well with or without TGF-β2 (15 ng/mL). After incubation at 37°C for 24 hours, the wounded areas were imaged. The wounded areas were measured by ImageJ. The data were quantified as follows: cell migration percentage (%) = (new scratch area-original scratch area)/original scratch area × 100%. The wound healing assay was repeated in triplicate.

| Cell proliferation assay
The real-time cell analysis (RTCA) iCELLigence system (CA 92121, ACEA Biosciences, Inc) based on electrical impedance measurements was used to detect cell proliferation. The RTCA iCELLigence system was placed in the incubator at 37°C and with 5% CO 2 . First, 200 μL of DMEM was added to the 8-well E-plates, and the background reading was determined. Then, 300 μL of the cell suspension was added to each well, and the cell density was 3 × 10 4 /well. After 6 hours of culture, TGF-β2 was added, and the cell proliferation rate was recorded every hour. The data are expressed as a normalized unit index.

| EdU staining assay
To determine the effect of CNPase on cell proliferation, a Cell-  After an additional 10 minutes of treatment with the penetrant, the cells were washed twice with methanol. Signals were detected using a confocal microscope (Olympus). Images of positively stained cells were captured and counted by ImageJ.

| Statistical analysis
Statistical Product and Service Solutions 16.0 software (IBM SPSS) was used for statistical analysis. All measurement data are presented as the mean ± standard deviation (SD) from at least three independent experiments. Student's t test was used for two-sample analysis, and one-way analysis of variance (ANOVA) was applied to compare the mean among three or more groups. A P value <.05 was considered to indicate a statistically significant difference.

| Upregulation of CNPase in lens epithelial cells of the ASC mouse model
To explore the biological functions of CNPase in the lens, we observed the expression of CNPase in healthy lenses and changes in its expression in ASC ( Figure 1A). On days 5, 7 and 14, lenses developed marked multilayered lens epithelial cell opacity beneath the anterior lens capsule compared with the control left eyes ( Figure 1B and 1).
Masson staining indicated significant fibrosis in the lens capsules of mice with ASC ( Figure 1D). Immunofluorescence labelling showed that CNPase was virtually undetected in the mouse lens fibres but was moderately expressed in epithelial cells of untreated lenses in vivo ( Figure 1E). We next examined the changes in CNPase expression in the mouse model of ASC induced by injury. Interestingly, compared with the normal left lenses, the lens epithelial cells in the mouse model exhibited augmented CNPase immunofluorescence ( Figure 1F).

| CNPase may be involved in the EMT process of lens epithelial cells
EMT is a crucial pathophysiological mechanism of ASC. Because Overall, the expression of CNPase was positively correlated with that of vimentin, thus supporting the idea that CNPase may mediate EMT to participate in the occurrence of ASC in vivo and in vitro.

| Silencing of CNPase inhibits the proliferation of lens epithelial cells
To CNPase siRNA-transfected cells in S phase (23%) was lower than that in lens epithelial cells after TGF-β2 treatment (39%) (Figure 2A and 2). Immunofluorescence staining further confirmed the above results of Western blotting ( Figure 3J-O). In summary, these results suggest that CNPase is involved in the regulation of the EMT phenotype in lens epithelial cells.

| The Notch signalling pathway participates in EMT induced by CNPase
In light of the critical role of the Notch signalling pathway in the pathogenesis of EMT, we next explored whether CNPase was involved in TGF-β2-induced EMT by regulating the Notch pathway. As shown in Figure 4, TGF-β2 markedly induced the expression of the

| D ISCUSS I ON
Pathological EMT has been considered the main driver of ASC.
Clinical cataract surgery and intraocular lens design highlight the importance of lens epithelial cell EMT in the pathogenesis of ASC. 27 Multiple signalling pathways, including the Notch and Wnt signalling pathways, have historically been considered to be involved in the EMT process in ASC. 28 However, the specific mechanism that triggers and maintains EMT has remained obscure. In this study, we focused on the roles of CNPase in the process of regulating  Notch signalling has been reported to play a crucial role in EMT in various fibrotic diseases and cancer metastasis, and its function in EMT of lens epithelial cells has been confirmed. 28  at least in part, regulates EMT in lens epithelial cells in a manner mediated by regulation of the Notch signalling pathway ( Figure 6).
In conclusion, an ASC mouse model was used to show for the first time that CNPase is highly expressed and related to EMT marker protein expression in lens epithelial cells. Very strikingly, CNPase promotes the proliferation, migration and EMT of lens epithelial cells by targeting Notch signalling in vitro. Therefore, clarifying the regulatory relationship between CNPase and the occurrence of EMT is desirable, as it not only would improve the understanding of ASC pathogenesis but also may suggest the effectiveness of blocking CNPase and the Notch signalling pathway as an effective strategy for preventing and treating ASC.

ACK N OWLED G EM ENTS
This study was supported by grants from the National Natural

CO N FLI C T O F I NTE R E S T
The authors have no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.