Metformin alleviates oxidative stress‐induced senescence of human lens epithelial cells via AMPK activation and autophagic flux restoration

Abstract Cataracts are the leading cause of blindness worldwide owing to the increasing proportion of elderly individuals in the population. The purpose of this study was to investigate whether metformin could alleviate the occurrence and development of age‐related cataract (ARC) and the underlying mechanism. In the present study, we established a senescence model induced by oxidative stress, which was confirmed by measuring β‐galactosidase activity, qRT‐PCR and Western blotting. In addition, we showed that metformin alleviated the oxidative stress‐induced senescence of HLE‐B3 cells via the activation of AMPK. Next, we provided evidence that oxidative stress impaired autophagic flux and induced lysosomal dysfunction. Subsequently, we found that metformin restored autophagic flux that had been impaired by oxidative stress by activating AMPK. Additionally, we found that metformin suppressed HLE‐B3 cell senescence by improving lysosomal function and inactivating mTOR. Furthermore, the inactivation of AMPK, impairment of autophagic flux and lysosomal dysfunction were observed in the human lens epithelium of ARC. In summary, our data suggest that the activation of AMPK may be a potential strategy for preventing ARC, and metformin may be an emerging candidate to alleviate the formation and development of ARC.

iris damage, ligament damage, glaucoma and posterior cataracts may occur. Some patients even need to undergo reoperation. 8 Although surgery is the main treatment for ARC, the complications associated with surgery cannot be ignored. Therefore, it is urgent to explore the pathogenesis of ARC and identify target genes and effective drugs for the treatment of ARC. Numerous studies have demonstrated that genes and environmental factors, including ultraviolet rays, ionizing radiation, chemicals and DNA damage, contribute to ARC. 9,10 One of the mechanisms of the above-mentioned interactions is that oxidative stress is triggered and further leads to the development of ARC. 10,11 In the past few years, research on anti-ageing treatment has discovered many drugs that can prolong life, among which metformin (MET) is the most notable. It has been found that MET not only lowers blood sugar, prevents macrovascular and microvascular diseases, and improves hyperinsulinaemia and insulin resistance 12 but also delays ageing, inhibits age-related pathological changes and reduces oxidative stress damage. 13,14 If the mechanism of action of MET can be clearly explained, its clinical application will be expanded to treat certain age-related diseases, including ARC, and increase life expectancy.
The majority of MET studies have demonstrated that MET can prolong the life span of mice and C. elegans and plays an important role in enhancing the health of these organisms. MET increases the average life span of C. elegans by 40%, prolongs the health of C. elegans, slows the accumulation of lipofuscin and prolongs young motor ability in a dose-dependent manner. 15 Similarly, at the age of 12 months, C57BL6 mice were fed 0.1% MET for 6 months. The average life span of C57BL6 mice was prolonged by 5.83%. The average life span of B6C3F1 mice was prolonged by 4.15% when the mice were fed in the same way and for the same time as C57BL6 mice. 16 Furthermore, a 10-year randomized clinical trial used MET to treat overweight/obese patients with type 2 diabetes and showed that long-term use of MET is beneficial to human health and survival. 17 Lens epithelial cells, a single layer of epithelial cells on the lens' anterior surface, are the most metabolically active part of the lens.
They provide basic materials and metabolic energy for the growth, differentiation and damage repair of the lens. The normal construction and function of lens epithelial cells is essential for the maintenance of the transparency and metabolic homeostasis of the entire lens. 18 Previous studies have revealed that oxidative stress, especially H 2 O 2, could cause excessive accumulation of reactive oxygen species (ROS), resulting in dysfunction and irreversible damage of normal lens epithelial cells, contributing to the modification, denaturation, aggregation of lens proteins including enzyme and crystallins, initiating early cataract formation. 19,20 Therefore, the mechanisms of protecting the normal lens epithelial cells against oxidative stress are an ongoing focus in the field of ARC research.
In the present study, we clarified the role and specific mechanism of MET in our system. Our results indicate that (a) MET can delay the hydrogen peroxide-induced senescence of lens epithelial cells by activating the AMPK pathway, (b) MET restores autophagic flux by activating the AMPK pathway, and (c) MET restores autophagic flux associated with the amelioration of lysosomal function and mammalian target of rapamycin (mTOR) inactivation. Our study provides a rationale for cellular senescence-based therapeutics for the protection of the eye and for the treatment of ARC in the elderly population.

| Study participants and preparation of human anterior lens capsules
Sixty-four patients aged 50-55 years with ARC participated in this study. Human lens epithelium specimens (approximately 5 mm in diameter) were collected during cataract surgery. Normal lens anterior capsule specimens with the adherent epithelium (approximately 5 mm in diameter) were donated by the patients and served as controls in the study. 21 We declared that the study followed the tenets of the Declaration of Helsinki and was approved by the Ethics Committee of Harbin Medical University. All subjects in this study knew and understood the content and risk of the research and signed the informed consent form.
According to Lens Opacities Classification System III (LOCS III), patients whose lenses had a score of C2-C3, N2-N3, or P2-P3, including eight patients in each ARC category, and eight age-matched controls from donor eyes from the eye bank with a LOCS III score of ≤C1, ≤N1 or ≤P1 were enrolled. 22

| Haematoxylin and eosin (H&E)
Fresh anterior lens capsules (human) were immediately fixed with 4% paraformaldehyde at room temperature and embedded in paraffin. For histological examination, the tissues were cut into 4μm slices. The sections were deparaffinized with 100% xylene and then rehydrated with gradient alcohol (100% ethanol, 90% aqueous ethanol, 80% aqueous ethanol, 70% aqueous ethanol and distilled water).
Then, the sections were stained with H&E (5 min and 2 min, respectively, at room temperature), and dehydrated. The morphological changes were observed under a microscope (Nikon, Eclipse).

| Immunohistochemistry (IHC)
Paraffin sections were deparaffinized and hydrated through a xylene and graded alcohol series. The sections were rinsed with water, boiled in 0.1 M citric acid (pH 6.1) for 30 min, allowed to cool to room temperature and treated with 3% hydrogen peroxide/deionized water buffer to inhibit endogenous peroxidase. Then, the fixed capsules were blocked with foetal bovine serum (FBS) for 30 min at room temperature and incubated with anti-p21, anti-p53, anti-phospho-AMPKa (Thr172), anti-phospho-ACC (Ser79), anti-LC3, anti-SQSTM1/p62 and antibodies in PBS overnight at 4℃. The secondary antibody conjugated to horseradish peroxidase (Cell Signaling Technology, USA) was then applied for 1 h at 37℃. Immunoreactivity was detected using diaminobenzidine (DAB; Cell Signaling Technology), and then, coverslips were added with Permount mounting medium. Immunostained images were captured using a Nikon Eclipse microscope (Nikon, Eclipse).

| Western blot analysis
Total protein was extracted from human lens epithelial tissue or cultured cells using RIPA lysis buffer with a protease inhibitor cocktail (Pierce), and a BCA kit (Thermo Scientific) was used to quantify the protein concentration. Forty micrograms of protein was loaded on an SDS-PAGE gel and separated by electrophoresis, followed by blotting onto a PVDF membrane (Millipore). The target proteins were probed with the corresponding primary antibodies against P21, P53, AMPKa1, phospho-AMPKa (Thr172), phospho-ACC (Ser79), LC3, SQSTM1/p62, CTSB, phospho-mTOR (Ser2448), phospho-p70S6K under optimized conditions and then incubated with the secondary antibody. Immunological signals were visualized via the electrochemiluminescence method using an Immobile Western Chemiluminescence HRP substrate kit (Millipore).

| Quantitative real-time polymerase chain reaction (qRT-PCR)
Total RNA was isolated from cultured cells using TRIzol reagent (Invitrogen) and reverse transcribed using the PrimeScript RT reagent kit (Takara) according to the manufacturer's instructions. qRT-PCR was performed using a SYBR Green Supermix kit (Takara), with β-actin as an endogenous control. The primer sequences used for PCR are shown in Table S1 2.9 | Statistical analysis Student's t-tests and ANOVA were used to calculate the statistical significance of the experimental data. The significance level was set as *p < 0.05 and **p < 0.01. Error bars denote SD.

| Identification of senescent HLE cells in the human lens epithelium of ARCs and controls
To investigate the senescence of lens epithelial cells in the lens epithelium of ARCs and the controls, we performed a series of measurements. H&E and SAβ-Gal staining were used to detect morphological differences in lens epithelial cells in ARC and control samples. The results indicated that lens epithelial cells became flat and sparse in the ARC samples compared to the control samples ( Figure 1A). Moreover, SAβ-Gal staining showed that senile lens epithelial cells were strongly positive ( Figure 1B). The IHC results demonstrated significant differences in senescence-related genes (P53, P21) between ARCs and the controls ( Figure 1C,D). Western blot analysis confirmed the IHC results showing the expression of senescence-related genes (P53, P21) in ARCs were significantly higher than those in the control samples ( Figure 1E).

| Oxidative stress-induced senescence in HLE-B3 cells in vitro
Exogenous H 2 O 2 has been widely used as a potent inducer of cellular senescence, which is commonly referred to as oxidative stressinduced senescence. 23

| The AMPK pathway was inactivated in the human lens epithelium of ARCs and in HLE-B3 cells with oxidative stress-induced senescence
Accumulating evidence indicates that inactivation of the AMPK pathway is closely related to ageing. 24,25 Based on this theory, we measured the expression of AMPK pathway components in human lens epithelium and HLE-B3 cells in vitro. The protein levels of phosphorylated AMPKα (Thr172) and phosphorylated ACC (Ser79) were markedly decreased in ARCs, while the total protein levels of AMPKα remained unchanged ( Figure 3C). Analogously, IHC examination showed that the expression levels of the critical genes p-AMPKα (Thr172) and p-ACC (Ser79) were decreased in the anterior capsule of the lens in ARCs compared with the controls (Figure 3A,B). Next, we found that the mRNA expression of FAS, an important gene that is downstream of the AMPK pathway, was significantly decreased in oxidative stress-induced HLE-B3 cells ( Figure 3D). Unsurprisingly, the protein levels of phosphorylated AMPKα (Thr172) and phosphorylated ACC (Ser79) were also significantly reduced in the senescent HLE-B3 cells ( Figure 3E). These results revealed that the AMPK pathway was inactivated in both the lens epithelium of ARCs and HLE-B3 cells with oxidative stress-induced senescence.

| MET alleviated oxidative stress-induced senescence in HLE-B3 cells
To evaluate the effects of the known AMPK pathway activator MET on oxidative stress-induced senile lens epithelial cells, different concentrations of MET were added to the culture medium after H 2 O 2 treatment. As shown in Figure 4A In addition, the protein levels of P21 and P53 were markedly decreased ( Figure 4E). Thus, we selected 2 mM MET as the optimal concentration for our subsequent experiments. Moreover, in order to better illustrate the anti-ageing mechanism of MET, we performed three experiments between control and metformin alone groups in HLE-B3 cells: SAβ-Gal staining, qRT-PCR and Western blot analysis.
These data showed that there was no significant difference in the expression of senescence-associated genes (P21, P16, P53) between control and MET alone groups ( Figure S1). Therefore, MET may prevent the senescence of HLE-B3 cells against oxidative stress.

| MET prevented oxidative stress-induced senescence in HLE-B3 cells via AMPK activation
MET is known to activate the AMPK pathway, which plays an important role in delaying senescence. 26

| Impairment of autophagic flux and lysosomal dysfunction in the human lens epithelium of ARCs and in HLE-B3 cells with oxidative stressinduced senescence
Several studies have shown that impairment of autophagic flux is a feature of cellular senescence. 27 Thus, we examined whether autophagic flux was impaired in the lens epithelium of ARCs and in HLE-B3 cells with oxidative stress-induced senescence. As indicators of autophagy, the conversion of LC3-I to LC3-II and the expression of P62 were assessed using Western blotting. The results showed an increase in the LC3-II/I ratio and P62 expression in the lens epithelium of ARCs compared to the controls ( Figure 6C,E). In addition, the IHC results suggested that there was a significant difference in the LC3-II/I ratio and expression of P62 between ARCs and the controls ( Figure 6A,B). As shown in Figure 6D

| MET restored the oxidative stress-impaired autophagic flux associated with improvements in lysosomal function and mTOR inactivation
As previously described, MET could partially prevent H 2 O 2induced senescence in HLE-B3 cells by improving autophagic flux.
Additionally, the association between lysosomes and senescence has been extensively reported. It is generally acknowledged that mTOR signalling is involved in senescence. To determine the mechanism by which MET restored H 2 O 2 -impaired autophagic flux in the context of lysosomal function, CTSB was used to evaluate the function of lysosomes via Western blotting. CTSB activation was evidently increased when AMPK was activated by MET and was obviously downregulated when AMPK was blocked by CC ( Figure 8A

| DISCUSS ION
Numerous scientific investigations have confirmed that MET can alleviate senescence by activating the AMPK pathway. 15,26,28 In the present study, we provided convincing evidence that MET alleviated oxidative stress-induced senescence stimulated via AMPK activation in vitro ( Figure 5). Interestingly, we also discovered that MET restored the autophagic flux that was impaired by oxidative stress via AMPK activation in vitro (Figure 7), and this restoration was also related to the promotion of lysosomal function and mTOR inactivation ( Figure 8). Notably, we detected AMPK inactivation and impaired autophagic flux in the lens epithelium of ARC (Figures 3,6).
Our findings will be instructive for more intensive studies of METmediated senescence inhibition. As an AMPK pathway activator, MET is expected to be an important target for new strategies against the formation and development of ARC.
Increasing evidence has indicated that oxidative damage is a predominant contributor to the pathogenesis of ARC. 29 induce HLE-B3 cell senescence ( Figure 2). The ageing human lens epithelium of ARC is also characterized by common hallmarks of cellular senescence, including increased expression of P21 and P53 and increased SAβ-gal activity (Figure 1). In addition, the expression pattern of ageing genes in senescent HLE-B3 cells and in the human lens epithelium of ARC is consistent (Figures 1,2). Successful establishment of human lens epithelial cell ageing was the basis for the success of the subsequent experiments.
It is widely accepted that MET stimulates AMPK, which is a key regulator of metabolic homeostasis in cells. 31 Dysregulation of the AMPK pathway is a serious problem for cells and organisms. A series of studies focused on the link between AMPK and cellular senescence showed that AMPK activation plays a positive role in the anti-senescence effect. 15,24,28 In our study, under oxidative stress, which is generally acknowledged as one of the causes of cataracts, the anti-senescence effect of the AMPK pathway was particularly prominent in HLE-B3 cells ( Figure 5). More interestingly, we observed that the AMPK pathway was activated in the human lens epithelium of ARC compared to the controls (Figure 3). These results provide meaningful evidence that the activation of AMPK may be an important target for delaying the occurrence of ARC. The AMPK pathway influences cellular senescence by modulating a complex network that includes P53, mTOR and FOXOs. 32 On the basis of the above results, the molecular mechanisms linking AMPK activation to senescence prevention are worth further investigation.
Autophagy is a catabolic process to destroy and remove unnecessary or damaged components in cells and is regulated by conditions such as oxidative stress and starvation. 33 The association between autophagy and senescence has been widely reported, and the role of MET in senescence inhibition is generally due to its effects on the induction of autophagy. 34 There is a growing body of evidence, demonstrating that MET can directly prevent senescence by restoring autophagic flux. 35 However, some scholars insist that MET can alleviate the senescence by promoting autophagy activity via AMPK activation. 36 Our results strongly support the latter notion that MET significantly enhances autophagic flux by activating the AMPK pathway to delay oxidative stress-induced senescence in HLE-B3 cells (Figure 7). Our results show that oxidative stress damages the metabolism of lysosomes and leads to inhibition of P62 degradation in both senescent HLE-B3 cells in vitro and the human lens epithelium of ARC ( Figure 6B,C,D,E). Moreover, we discovered marked impairment of autophagic flux in the human lens epithelium of ARC ( Figure 6A,B,C,E). In addition, we measured the expression of CTSB to explore the function of lysosomes in autophagosome degradation, which was associated with the late phase of autophagy ( Figure 6F,G,H). As previously mentioned, lysosomal function was noticeably damaged in both senescent HLE-B3 cells and the human lens epithelium of ARC. Our results further confirmed that MET restored autophagic flux to promote lysosomal function ( Figure 8A,B). MET enhanced the lysosomal degradation of autophagosomes, which is recognized as the late stage of autophagy. However, these data were different from previous results showing that MET could prevent oxidative stress-induced senescence by restoring autophagic flux and mitochondrial functions in HLE-B3 cells. 37 Moreover, it is worth noting that the effect of MET on autophagic flux was similar to that of RAPA, which is also known as an mTOR inhibitor. Therefore, we believe that MET slowed oxidative stress-induced senescence in association with mTOR inactivation ( Figure 8C). In fact, consistent results have been detected by others. 38 In addition, our results showed that the protein levels of p-AMPK, p-mTOR and p-p70S6K were higher in the control group than in the other groups ( Figure 8C). These findings suggest that the data presented here could explain the role of MET in delaying senescence under oxidative stress injury conditions. Whether MET can prevent normal cells from ageing still needs further study.
In conclusion, in our H 2 O 2 -induced senescence model, our results suggested that the anti-senescence effect of MET depended on activation of the AMPK pathway, which then activated autophagy. Furthermore, we also demonstrated that MET delays senescence by activating the AMPK pathway, improving lysosomal function and downregulating mTOR expression.
Moreover, activation of the AMPK pathway is critical to the antisenescence effect of MET. Moreover, we found that senescence of the human lens epithelium in cataracts led to the inactivation of AMPK and inhibition of autophagy. Overall, these findings support MET as an emerging candidate for alleviating the formation and development of ARC and indicate a direction for our next experiments.

ACK N OWLED G EM ENTS
This work was supported by the National Natural Science Foundation of China (No: 81770912) awarded to Yanhua Qi.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest. Writing-review & editing (lead).

DATA AVA I L A B I L I T Y S TAT E M E N T
All data generated or used during the study are available from the corresponding author by request.