TP53INP2 Contributes to TGF-β2-Induced Autophagy during the Epithelial–Mesenchymal Transition in Posterior Capsular Opacification Development

Background: Posterior capsule opacification (PCO) is the most common complication after cataract surgery, in which increased levels of transforming growth factor-beta 2 (TGF-β2) accelerate PCO formation; however, the pathological mechanisms are not fully understood. This study aims to explore the regulation mechanism of TGF-β2 in PCO formation via its autophagic functions. Methods: The autophagic effect of TGF-β2 was detected by transmission electron microscopy (TEM), Western blotting, and immunofluorescence analysis. The association between autophagy and the epithelial–mesenchymal transition (EMT) was evaluated by qPCR and Western blotting. The transcriptome analysis was used to uncover the molecular mechanism of TGF-β2-induced PCO formation. Results: TGF-β2 specifically promotes autophagy flux in human lens epithelial cells. The activation of autophagy by rapamycin can promote EMT marker synthesis and improve cell migration. However, the inhibition of autophagy by 3-MA attenuates EMT. To uncover the molecular mechanisms, we performed RNA sequencing and found that TGF-β2 elevated tumor protein p53-inducible nuclear protein2 (TP53INP2) expression, which was accompanied by a nuclear-to-cytoplasm translocation. Moreover, the knockdown of TP53INP2 blocked the TGF-β2-induced autophagy and EMT processes, revealing that TP53INP2 plays an important role in TGF-β2-induced autophagy during EMT. Conclusions: Taken together, the results of this study suggested that TP53INP2 was a novel regulator of PCO development by TGF-β2, and notably, TP53INP2, may be a potential target for the pharmacological treatment of PCO.


Introduction
Posterior capsule opacification (PCO) is the most common complication after cataract surgery, causing decreased vision in 20-40% of elderly patients [1] and 100% of pediatric patients [1,2]. While the specific pathological mechanisms are not fully understood, it is widely accepted that the epithelial-mesenchymal transition (EMT) process is a pivotal cause. During the EMT, residual lens epithelial cells lose adhesion and connexin. The epithelial cells are subsequently trans-differentiated into mesenchymal cells and rapidly migrated to the posterior capsule [3,4]. Multiple cytokines are reported to trigger EMT, including FGF-2, EGF, and TGF-β2 (transforming growth factor-β2) [5,6]. TGF-β2 levels, as a major inducer of EMT, are significantly increased after cataract surgery and accelerate PCO formation. Currently, there is no ideal pharmacological treatment for PCO.
Autophagy is an evolutionally conserved degradative process that engulfs damaged, degenerative or aging proteins and organelles into phagophores, which mature into doublemembrane autophagosomes and then fuse with lysosomes for degradation [7]. Basal autophagy serves as a quality control system and is required for maintaining cellular homeostasis [8]. Consequently, imbalanced autophagy may induce cancer, neurodegenerative diseases, and ocular disorders, including cataracts [9]. However, the precise effect of autophagy on the lens is an open area of inquiry. Our previous research had found that rapamycin (rapa), an activator of autophagy, effectively rescued lens opacities in GJA8 −/− zebrafish [10]. Conversely, hyperactive autophagy may result in cell viability deprivation and death [11]. These studies confirmed the important role of autophagy in regulating lens homeostasis. Currently, there is accumulating evidence that autophagy is regarded as a crucial regulator of EMT in some tumor cells [12]. However, the roles of TGF-β2, EMT, and autophagy involved in the development of PCO are not fully understood.
In this study, we have reported the autophagic response of human lens epithelial (HLE) cells during the EMT process induced by TGF-β2, as well as the regulation of TP53INP2, to investigate the potential mechanism in the formation of PCO.

RNA Sequencing
The HLE cells were seeded onto 10 cm dishes overnight and then treated with or without TGF-β2 (10 ng/mL) for 24 h. Total RNA was extracted by TRIzol Reagent (Invitrogen, Carlsbad, CA, USA), and the RNA purity was detected by spectrophotometer (NanoDrop). The RNA integrity number (RIN) was measured using the Agilent 2200 TapeStation (Agilent Technologies, Palo Alto, CA, USA) and the sample with RIN ≥ 7 was chosen for sequencing. Subsequently, library preparation and RNA sequencing were performed by RiboBio Co., Ltd. (Guangzhou, China) using sequencing (2 × 150 bp) HiSeq3000.
The detailed information on sequencing was described previously [13], and the differential expression was estimated by DEseq2 v1.16.1 using read counts as input. The correction method of the Benjamini-Hochberg multiple test was utilized. Differentially expressed genes (DEGs) were selected according to the criteria of log2|Fold-change| > 1.2, adjusted p-value < 0.05, for further transcriptomic analysis.

Transmission Electron Microscopy
The cells were collected and fixed in 2.5% glutaraldehyde (SPI Chem, 02608-BA) in PBS at 4 • C overnight. The cells were rinsed by PBS and fixed with 1% OsO4 and 0.1% K3Fe(CN)6. The cells were then dehydrated by a graded concentration of ethanol and embedded in Spurr resin. Ultrathin sections were stained with 2% uranyl acetate and alkaline lead citrate and observed using a Hitachi Model H-7650 TEM.

Wound Healing Assays
The cells were cultured in two 6 cm dishes and transfected with TP53INP2 siRNA and NC siRNA for 24 h. Subsequently, the above cells were seeded in a 6-well plate to grow to 90% confluence. The cells were scratched with a 100 µL pipette tip and washed twice using PBS, and then incubated in DMEM containing 2% FBS with or without TGF-β2, rapa, or 3-MA for 48 h. The images were taken by microscope (Olympus) and analyzed by Image J software. The quantitative wound healing area was calculated by the equation: [1 − empty area X h/empty area 0 h] × 100.

Subcellular Fractionation
The cells were harvested with cold PBS and lysed in buffer A (Beyotime Biotechnology, Shanghai, China) containing 1% PMSF on ice for 20 min. The cells were added to buffer B and then centrifuged at 13,000 rpm for 5 min. The resultant supernatant was transferred to a new Eppendorf tube and used as cytosolic fractions. The sediment was then homogenized with the buffer of nuclear protein extraction reagent with 1% PMSF. After 30 min of vortexing, the cells were centrifuged for 10 min at 16,000 rpm and the resultant supernatants were used as nuclear fractions.

Statistical Analysis
The data were analyzed using GraphPad Prism 8.0 software and are presented as means ± SD. Student's t-tests were performed to compare the statistical difference between two groups and one-way ANOVA was conducted to compare the data of multiple groups. A value of p < 0.05 was considered significant.

Transcriptome Analysis Identified the TGF-β2 Involved in Autophagy and EMT Processes in HLE Cells
To reveal the molecular mechanisms of autophagy in TGF-β2-exposed HLE cells, we performed transcriptome profiling to compare the genes involved in PCO development in TGF-β2-treated or -untreated HLE cells. A total of 998 differentially expressed genes (DEGs) were found in the two groups ( Figure 1A,B), of which 527 genes were up-regulated and 471 genes were down-regulated (log2|fold-change| > 1.2; adjusted p-value < 0.05). The gene ontology (GO) term and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway were conducted based on up-related DEGs. The top sixteen KEGG pathways are presented in Figure 1C, and most of them were pathways related to EMT and autophagy. For example, (ECM)-receptor interaction, focal adhesion, and the PI3K/Akt pathway were involved in the EMT process and the PI3K/Akt pathway, Hippo pathway, and Wnt pathway were related to autophagy. Similarly, the results of GO analysis showed that the DEGs not only were significantly enriched in the EMT but also in autophagy ( Figure 1D). Subsequently, we conducted a protein-protein interaction (PPI) analysis of upregulated DEGs annotated to GO terms correlated with autophagy ( Figure 1D). Because it had the highest increase of 4.99-fold, the tumor protein p53 inducible nuclear protein 2 (TP53INP2) was chosen for further exploration as it was related to both autophagy and EMT (Table 1 depicts all significantly changed autophagy genes).

TGF−β2 Promotes Autophagy Flux in HLE Cells
Previous studies and RNA sequencing have indicated that TGF−β2 has the ability to regulate autophagy. To determine the influence of TGF−β2 on autophagy flux in HLE cells, we observed the autophagic vacuoles (AVs) in TGF−β2 treated or untreated cells by transmission electron microscopy (TEM), which is an intuitively morphological examination. Under the stimulation of TGF−β2, the number of AVs markedly increased ( Figure  2A), which might be caused by the activation of autophagy but also could be due to the block of autophagosome degradation. This result laid the foundation for our further evaluation of autophagic flux. Therefore, we confirmed the autophagy flux by GFP−mCherry−LC3 puncta assay and found that the number of autophagosomes (yellow) and autolysosomes (red) were both increased in TGF−β2−exposed HLE cells. In autolysosome formation, GFP signal disappearance indicated the formation of autolysosomes and only red spots were visible ( Figure 2C,D), suggesting an enhanced role of TGF−β2 in autophagy flux. These findings were further verified by the protein pattern of LC3II and p62 using Bafilomycin A1, a specific V−ATPase inhibitor, which blocked the fusion of lysosomes and autophagosomes. The immunoblotting results revealed that LC3 and p62, the

TGF-β2 Promotes Autophagy Flux in HLE Cells
Previous studies and RNA sequencing have indicated that TGF-β2 has the ability to regulate autophagy. To determine the influence of TGF-β2 on autophagy flux in HLE cells, we observed the autophagic vacuoles (AVs) in TGF-β2 treated or untreated cells by transmission electron microscopy (TEM), which is an intuitively morphological examination. Under the stimulation of TGF-β2, the number of AVs markedly increased (Figure 2A), which might be caused by the activation of autophagy but also could be due to the block of autophagosome degradation. This result laid the foundation for our further evaluation of autophagic flux. Therefore, we confirmed the autophagy flux by GFP-mCherry-LC3 puncta assay and found that the number of autophagosomes (yellow) and autolysosomes (red) were both increased in TGF-β2-exposed HLE cells. In autolysosome formation, GFP signal disappearance indicated the formation of autolysosomes and only red spots were visible ( Figure 2C,D), suggesting an enhanced role of TGF-β2 in autophagy flux. These findings were further verified by the protein pattern of LC3II and p62 using Bafilomycin A1, a specific V-ATPase inhibitor, which blocked the fusion of lysosomes and autophagosomes. The immunoblotting results revealed that LC3 and p62, the recognized autophagosome markers, accumulated under the BafA1 stimulation. However, in the TGF-β2 + BafA1 co-treated group, the LC3-II level was higher than in the BafA1 treated group due to the constant generation. Taken together, these results suggest that TGF-β2 may promote autophagy flux in HLE cells.
recognized autophagosome markers, accumulated under the BafA1 stimulation. However, in the TGF−β2 + BafA1 co−treated group, the LC3−II level was higher than in the BafA1 treated group due to the constant generation. Taken together, these results suggest that TGF−β2 may promote autophagy flux in HLE cells.

TGF−β2 Simultaneously Induced Autophagy and Directly Regulated TP53INP2 Expression and Subcellular Localization in HLE Cells
According to the RNA sequencing, the high expression of TP53INP2 was first found in TGF−β2−treated HLE cells. To validate our hypothesis, we treated the cells with different doses of TGF−β2, ranging from 1 ng/mL to 20 ng/mL for 24 h. Western blot testing revealed that the autophagy−related protein levels of LC3−II, Atg5, and Beclin1 were increased. P62(SQSTM1), meanwhile, gradually degraded, which suggests enhanced autophagy. EMT markers, such as fibronectin (FN) and vimentin expression levels were also increased, indicating the formation of the EMT process. It is worth noting that the TP53INP2 expression level was also obviously elevated ( Figure 3A,B).
We consistently observed an increased number of LC3 puncta and a reduced number of P62 puncta in TGF−β2−treated HLE cells via immunofluorescence, which verified the active autophagic response ( Figure 3C,D). However, we found that the mRNA and protein levels of autophagy (LC3, ATG5, and Beclin1), EMT−related molecules (vimentin and fibronectin), and TP53INP2 were significantly increased under the stimulation of TGF−β2.

TGF-β2 Simultaneously Induced Autophagy and Directly Regulated TP53INP2 Expression and Subcellular Localization in HLE Cells
According to the RNA sequencing, the high expression of TP53INP2 was first found in TGF-β2-treated HLE cells. To validate our hypothesis, we treated the cells with different doses of TGF-β2, ranging from 1 ng/mL to 20 ng/mL for 24 h. Western blot testing revealed that the autophagy-related protein levels of LC3-II, Atg5, and Beclin1 were increased. P62(SQSTM1), meanwhile, gradually degraded, which suggests enhanced autophagy. EMT markers, such as fibronectin (FN) and vimentin expression levels were also increased, indicating the formation of the EMT process. It is worth noting that the TP53INP2 expression level was also obviously elevated ( Figure 3A,B). nificant increase in cytoplasmic TP53INP2, yet the levels of nuclear TP53INP2 were either unaltered or even decreased. This result was subsequently confirmed by immunofluorescence assay and it was observed that TGF−β2 could facilitate TP53INP2 movement from the nucleus to the cytoplasm to participate in autophagy−related processes ( Figure 3G,H). Taken together, the above data confirm that autophagy was activated in TGF−β2−exposed HLE cells during the EMT process and that the TGF−β2 had a direct regulatory effect on TP53INP2′s expression and subcellular location. We consistently observed an increased number of LC3 puncta and a reduced number of P62 puncta in TGF-β2-treated HLE cells via immunofluorescence, which verified the active autophagic response ( Figure 3C,D). However, we found that the mRNA and protein levels of autophagy (LC3, ATG5, and Beclin1), EMT-related molecules (vimentin and fibronectin), and TP53INP2 were significantly increased under the stimulation of TGF-β2. While co-treated with SB431532, a specific inhibitor of TGF-β type I receptors, the activatedautophagy effect, active EMT process, and raised TP53INP2 expression level significantly declined. This suggests that TGF-β2 specifically induced autophagy and directly moderated TP53INP2 expression ( Figure 3E,F). Furthermore, we explored the molecular mechanism and found that TGF-β2 promoted TP53INP2 expression through a significant increase in cytoplasmic TP53INP2, yet the levels of nuclear TP53INP2 were either unaltered or even decreased. This result was subsequently confirmed by immunofluorescence assay and it was observed that TGF-β2 could facilitate TP53INP2 movement from the nucleus to the cytoplasm to participate in autophagy-related processes ( Figure 3G,H). Taken together, the above data confirm that autophagy was activated in TGF-β2-exposed HLE cells during the EMT process and that the TGF-β2 had a direct regulatory effect on TP53INP2's expression and subcellular location.

Autophagy Had a Positive Regulatory Effect on the EMT Process in HLE Cells
Given the dual regulatory role of autophagy on EMT in other types of cells, we examined the functional connection between autophagy and the EMT process in HLE cells. Two pharmacological agents, rapamycin (rapa) and 3-methyladenine (3-MA), were used. We co-treated the HLE cells with TGF-β2 and rapa, an activator of autophagy, which primarily blocked the MTORC1 signal pathway. As shown in Figure 4A-D, rapa effectively downregulated E-cadherin (epithelial marker) mRNA and protein levels and upregulated N-cadherin (mesenchymal marker) in HLE cells. Intriguingly, TGF-β2-treated cells and TGF-β2 + rapa co-treated cells were migrated faster than the NC group in a wound healing assay ( Figure 4E,F). These results suggest that increased autophagy induced the EMT process in HLE cells. Coincidently, when autophagy inhibitor 3-MA was added to HLE cells, the opposite phenomenon was observed ( Figure 4G-J): the suppression of autophagy inhibits TGF-β2-induced EMT and lowers the rate of cell migration ( Figure 4K,L). Thus, the above results indicate that autophagy had a regulated effect on the EMT process.

TP53INP2 Is Essential for TGF-β2-Mediated Autophagy and the EMT Process
To further investigate the functions of TP53INP2 in HLE cells, we transfected TP53INP2specific siRNA to HLE cells and observed that the effect of RNA interference was nearly 91% ( Figure 5A). The TP53INP2 knockdown decreased the autophagy-related proteins, including LC3-II conversion, as well as the Beclin1 expression. Simultaneously, N-cadherin and vimentin, two of the mesenchymal phenotype markers, were also downregulated ( Figure 5B,C). This suggests that a lack of TP53INP2 abated the TGF-induced autophagic response and EMT process. The wound-healing assay also showed that the TP53INP2 siRNA group slowed down the rate of cell migration. At the same time, the co-treatment of TP53INP2 siRNA with TGF-β2 cells caused the migration rate to be slower than that of TGF-β2-treated cells ( Figure 5C). Taken together, the above findings indicate that TP53INP2 played an important role in TGF-mediated autophagy and the EMT process.

TP53INP2 is Essential for TGF−β2−Mediated Autophagy and the EMT Process
To further investigate the functions of TP53INP2 in HLE cells, we transfected TP53INP2−specific siRNA to HLE cells and observed that the effect of RNA interference was nearly 91% ( Figure 5A). The TP53INP2 knockdown decreased the autophagy−related proteins, including LC3−II conversion, as well as the Beclin1 expression. Simultaneously, N−cadherin and vimentin, two of the mesenchymal phenotype markers, were also downregulated ( Figure 5B,C). This suggests that a lack of TP53INP2 abated the TGF−induced autophagic response and EMT process. The wound−healing assay also showed that the TP53INP2 siRNA group slowed down the rate of cell migration. At the same time, the co−treatment of TP53INP2 siRNA with TGF−β2 cells caused the migration rate to be slower than that of TGF−β2−treated cells ( Figure 5C). Taken together, the above findings indicate that TP53INP2 played an important role in TGF−mediated autophagy and the EMT process.

Discussion
In this study, we observed the essential role of TP53INP2 in regulating TGF−β2−induced autophagy during the EMT process in HLE cells. TP53INP2, at the crossroad of autophagy and EMT in PCO pathogenesis, is likely a novel medical target for PCO therapy.
Autophagy is one of the main intracellular degradation systems for maintaining lens homeostasis [15,16]. Studies on lens−specific Atg5 −/− mice and Pik3c3/Vps34 −/− mice have discovered that impaired autophagy in lenses generates cataracts [17]. However, excessive autophagy may lead to the autophagic death of HLE cells, suggesting that the balance of autophagy is crucial for keeping the lens transparent. Given the importance of autophagy in maintaining the normal physiologic function of HLE cells, we explored the autophagic response in an abnormal state of PCO and found that TGF−β2 induced autophagy and promoted autophagy flux in HLE cells.

Discussion
In this study, we observed the essential role of TP53INP2 in regulating TGF-β2-induced autophagy during the EMT process in HLE cells. TP53INP2, at the crossroad of autophagy and EMT in PCO pathogenesis, is likely a novel medical target for PCO therapy.
Autophagy is one of the main intracellular degradation systems for maintaining lens homeostasis [15,16]. Studies on lens-specific Atg5 −/− mice and Pik3c3/Vps34 −/− mice have discovered that impaired autophagy in lenses generates cataracts [17]. However, excessive autophagy may lead to the autophagic death of HLE cells, suggesting that the balance of autophagy is crucial for keeping the lens transparent. Given the importance of autophagy in maintaining the normal physiologic function of HLE cells, we explored the autophagic response in an abnormal state of PCO and found that TGF-β2 induced autophagy and promoted autophagy flux in HLE cells.
Further evidence has suggested that autophagy plays a fundamental role in the EMT in cancer [18]. Given the complex relationship between the two processes, we examined the influence of autophagy on EMT in HLE cells through two specific autophagy modulators, rapa and 3-MA. We observed that the activated autophagy promoted TGF-β2-mediated EMT, while blocking autophagy inhibited the EMT. 3-MA is an unspecific autophagy inhibitor and mainly blocks autophagy by inhibiting class III phosphoinositide 3-kinase to suppress autophagy induction and prevent the autophagy nucleation phase [19]. However, there are other compensatory pathways still in operation, thus the TGF-β2-induced EMT process was not completely reversed by the treatment of 3-MA. The diverse regulatory effects between autophagy and the EMT depend on different cell types and various other factors [20]. Using autophagy inhibitors as potential therapeutic targets provides a new direction for the pharmacological treatment of PCO.
TP53INP2, a nuclear protein, is highly expressed in actively metabolizing tissues and is involved in cellular processes including obesity, transcription, autophagy, and apoptosis [21,22]. Under normal conditions, TP53INP2 is predominantly localized in the nucleus and regulates transcriptional activity by binding to various nuclear hormone receptors [23]. It can also enter the nucleolus to stimulate rDNA transcription [24]. It has been reported that the translocation of LC3 from the nucleus to the cytoplasm is vital for autophagosome formation [25,26]. In response to nutrient deprivation, TP53INP2 could take LC3 to enter the cytoplasm in order to participate in phagophore development and LC3 lipidation [27][28][29]. The cytoplasmic TP53INP2 serves as a scaffold for the promotion of LC3 and ATG7 interaction and the facilitation of LC3B-PE conjugation [30].
In this study, we found that TGF-β2 could directly promote the entry of TP53INP2 into the cytoplasm and elevate TP53INP2 expression to activate autophagy and the EMT. Interestingly, the subcellular location of TP53INP2 acts as a switch to determine its biological behavior [23]. We also discovered when silencing TP53INP2 that both LC3-II conversion and the EMT process were blocked. This result was consistent with a study on bladder cancer that showed that TP53INP2 could regulate the EMT through GSK-3β signaling, supporting the idea that it has autophagy-inducing and EMT regulatory functions in HLE cells.
Given that no effective drug treatment is available, patients are still suffering from the threat of vision loss caused by PCO. Therefore, it is of great significance to find the potential drug targets. In this respect, our study provided novel insights into the interplay between autophagy and EMT induced by TGF-β2 in HLE cells and suggested that TP53INP2 could work as a molecular switch to modulate autophagy induction to fine-tune the EMT process. Of note, the role of TP53INP2 in PCO formation has not been fully clarified yet. Thus, further investigations are needed to explore the underlying mechanism of autophagy regulation by TP53INP2, and there may exist some unknown post-translational modifications and protein-protein interactions in vivo. Besides, the function of TP53INP2 may be a new target for PCO treatment.

Conclusions
In conclusion, this study focused on the role of TP53INP2 in the modulation of TGF-β2-induced autophagy and the EMT in HLE cells. During this physiological process, active autophagy is pivotal in regulating the EMT process to affect PCO development. Our findings provide new clues for understanding the pathogenesis of PCO and suggest that TP53INP2 inhibitors may be potential targets for the pharmacological treatment of PCO.