Mechanisms of PM10 Disruption of the Nrf2 Pathway in Cornea

We have previously shown that PM10 exposure causes oxidative stress and reduces Nrf2 protein levels, and SKQ1 pre-treatment protects against this damage in human corneal epithelial cells (HCE-2). The current study focuses on uncovering the mechanisms underlying acute PM10 toxicity and SKQ1-mediated protection. HCE-2 were pre-treated with SKQ1 and then exposed to 100 μg/mL PM10. Cell viability, oxidative stress markers, programmed cell death, DNA damage, senescence markers, and pro-inflammatory cytokines were analyzed. Nrf2 cellular location and its transcriptional activity were determined. Effects of the Nrf2 inhibitor ML385 were similarly evaluated. Data showed that PM10 decreased cell viability, Nrf2 transcriptional activity, and mRNA levels of antioxidant enzymes, but increased p-PI3K, p-NFκB, COX-2, and iNOS proteins levels. Additionally, PM10 exposure significantly increased DNA damage, phosphor-p53, p16 and p21 protein levels, and β-galactosidase (β-gal) staining, which confirmed the senescence. SKQ1 pre-treatment reversed these effects. ML385 lowered the Nrf2 protein levels and mRNA levels of its downstream targets. ML385 also abrogated the protective effects of SKQ1 against PM10 toxicity by preventing the restoration of cell viability and reduced oxidative stress. In conclusion, PM10 induces inflammation, reduces Nrf2 transcriptional activity, and causes DNA damage, leading to a senescence-like phenotype, which is prevented by SKQ1.


Introduction
Air pollution is composed of particles with a diameter < 10 microns (PM 10 ) [1] and is hazardous to human health [2].PM 10 exposure has been associated with cardiovascular, pulmonary diseases, and cancer [1].Studies have largely focused on the effects of PM 10 on the lung [3], but recent studies have emerged with a focus on its harmful effects on the cornea, especially in diseases such as dry eye [3,4].Studies using in vitro and in vivo modeling have shown that PM 10 exposure causes apoptosis [5], autophagy [6], inflammation [7], and oxidative stress [8].However, the exact mechanism underlying PM 10 -induced damage to the cornea is not clear.
Oxidative stress has been implicated in several cancerous and non-cancerous diseases [14 -17].Oxidative stress also activates the nuclear factor erythroid-2-related factor 2 (Nrf2) antioxidant pathway [18] which is critical in regulating antioxidant defenses that are protective.Normally, Nrf2 is bound to Kelch-like erythroid cell-derived protein with CNC homology-associated protein 1 (Keap-1), thereby keeping the basal levels of Nrf2 in the cytosol low.However, oxidative stress causes changes in Keap1 that result in Nrf2 being moved to the nucleus, where it binds the antioxidant response element (ARE) and activates phase II detoxifying enzymes [18].Studies have shown that particulate matter (PM)-mediated effects on Nrf2 are dependent on the dose of PM exposure: at a low dose, Nrf2 expression is activated, but at a high dose, the capacity of PM to activate Nrf2 is compromised [19].Decreased Nrf2 protein levels have been linked to different pathological conditions [20].In the eye, Nrf2 has been tested as a potential therapeutic target preventing ocular diseases including dry eye, cataract, uveitis, and various retinopathies [21].In fact, the protective activity of Nrf2 in eyes subjected to smoke generated from tobacco has been well documented [22].To this end, we have recently provided evidence that PM 10 exposure reduces Nrf2 protein levels and causes mitochondrial dysfunction by increasing mitochondrial ROS and reducing ATP levels [23] in the cornea.SKQ1 (10-(6 ′ -plastoquinonyl) decyltriphenylphosphonium) is an antioxidant that accumulates in the inner mitochondrial membrane [24].SKQ1 is protective against damage induced by oxidative stress in various animal models of disease [25][26][27][28].Recently, we provided evidence that SKQ1 safeguards against PM 10 -induced oxidative damage in HCE-2 cells [23].In regard to this observation, an ophthalmic formulation of SKQ1 (Visomitin) has been used to successfully inhibit the pathology of anesthetic-induced dry eye syndrome after surgery or lengthy general anesthesia [29].
The aim of the current work is to better understand how PM 10 induces damage to the cornea and to investigate the role of Nrf2 in SKQ1-mediated protection.ML385 was used as it prevents Nrf2 activity by inhibiting its DNA binding capacity [30].

PM 10 Decreases Cell Viability but Does Not Induce Apoptosis
PM 10 treatment showed a concentration-dependent reduction (p < 0.001) in HCE-2 cells (Figure 1A).At the lowest concentration of 100 µg/mL, 80% of the cells were viable.However, at the highest concentration (1200 µg/mL), measured by MTT assay, more than 50% of the cells were non-viable.Therefore, 100 µg/mL was chosen for all experiments.PM 10 and SKQ1 effects on cell viability, analyzed by MTT assay, are shown in Figure 1B.Cell viability after exposure to 100 µg/mL PM 10 was significantly lowered (p < 0.001) vs. control or SKQ1 groups.Pre-treatment with SKQ1 before PM 10 prevented (p < 0.001) the loss in cell viability.
To analyze if PM 10 induced apoptosis, FACS analysis was employed to quantitate the number of apoptotic (Annexin V-FITC + ) and necrotic (PI + ) cells among the four groups.Representative scatterplots indicating the Annexin V/PI staining in HCE-2 cells (Figure 2) show that apoptotic (Annexin V + /PI − ) and necrotic cells (Annexin V − /PI + ) in the control, SKQ1, PM 10 , and PM 10 + SKQ1-exposed groups are not significantly different after 24 h of PM 10 exposure, indicating that a 100 µg/mL dose was not cytotoxic, but more likely cytostatic.dependent reduction in cell viability was observed after PM10 treatment.At the highest do µg/mL), nearly 50% of the cells were non-viable compared to 100 µg/mL, at which 80% of were viable (A).Cell viability was significantly reduced after PM10 exposure (100 µg/mL) vs or SKQ1.Pre-treatment with SKQ1 before PM10 prevented a loss in cell viability (B).Data ar as the mean + SD. (*** p < 0.001, n = 3).

PM10 Induces DNA Damage and Causes Cell Cycle Arrest
DNA damage was examined by γ-H2AX staining in the nuclei of cells, and images and quantification are shown in Figure 3A,B.Compared to control or SKQ1 groups, the PM10-treated group had significantly more γ-H2AX staining indicative of DSBs in the nuclei (indicated by arrows, Figure 3A) and a significantly higher number of cells exhibiting DSBs (Figure 3B, p < 0.001).SKQ1 pre-treated cells (before PM10 exposure) showed significantly fewer cells with nuclear DSBs (Figure 3A,B, p < 0.001).Levels of phosphorylated p53, p21, and p16 were measured after PM10 exposure using Western blot analysis and the IDVs are shown in Figure 3C-E

PM10 Effects Activation of NFκB by PI3K Upregulating COX2 and iNOS
To examine the effects of PM10 and SKQ1 on PI3K-regulated activation of NFκB, their phosphorylated levels were examined.IDV were analyzed and are shown in Figure 5. PM10 exposure increased the levels of phosphorylated PI3K (Figure 5A, p < 0.001) and phosphorylated NFκB (Figure 5B, p < 0.001) vs. control or SKQ1 alone.However, SKQ1 pre-treatment before PM10 significantly lessened both proteins (Figure 5A

PM 10 Effects Activation of NFκB by PI3K Upregulating COX2 and iNOS
To examine the effects of PM 10 and SKQ1 on PI3K-regulated activation of NFκB, their phosphorylated levels were examined.IDV were analyzed and are shown in Figure 5. PM 10 exposure increased the levels of phosphorylated PI3K (Figure 5A, p < 0.001) and phosphorylated NFκB (Figure 5B, p < 0.001) vs. control or SKQ1 alone.However, SKQ1 pretreatment before PM 10 significantly lessened both proteins (Figure 5A, p < 0.001, Figure 5B, p < 0.05).The effects of PM10 and SKQ1 on transcript and protein levels of COX2 and iNOS are shown in Figure 6A-D.The PM10-exposed vs. control or SKQ1-alone groups show a significant increase (p < 0.001) in both COX2 mRNA (Figure 6A) and protein (Figure 6C).Pretreatment with SKQ1 before PM10 significantly decreased (p < 0.001) COX2 mRNA and protein levels.Similarly, PM10 exposure caused a significant upregulation of both iNOS mRNA (Figure 6B, p < 0.001) and protein (Figure 6D, p < 0.01), while SKQ1 pre-treatment significantly reduced (p < 0.001) both.The effects of PM 10 and SKQ1 on transcript and protein levels of COX2 and iNOS are shown in Figure 6A-D.The PM 10 -exposed vs. control or SKQ1-alone groups show a significant increase (p < 0.001) in both COX2 mRNA (Figure 6A) and protein (Figure 6C).Pre-treatment with SKQ1 before PM 10 significantly decreased (p < 0.001) COX2 mRNA and protein levels.Similarly, PM 10 exposure caused a significant upregulation of both iNOS mRNA (Figure 6B, p < 0.001) and protein (Figure 6D, p < 0.01), while SKQ1 pre-treatment significantly reduced (p < 0.001) both.

PM10 Causes Nuclear Translocation of Nrf2 Not Colocalized with γ-H2AX (DSBs)
The effects of PM10 on translocation of Nrf2 are presented in Figure 7A-H     To confirm the immunostaining data, Western blot analysis of Nrf2 in the nuclear and cytosolic fractions was performed, and the results are shown in Figure 8A,B.Data show that in the nuclear fractions of PM10-exposed vs. control or SKQ1-alone cell fractions, Nrf2 protein levels were significantly increased (Figure 8A, p < 0.001).SKQ1 pre-treatment before PM10 exposure significantly reduced (p < 0.05) Nrf2 protein in the nucleus.In contrast, cytosolic fraction showed significantly reduced Nrf2 protein levels (Figure 8B, p < 0.05) after PM10 exposure, which was reversed significantly (Figure 8B, p < 0.05) by pretreatment with SKQ1.To test if Nrf2 colocalizes with the DSBs in the nucleus, cells were immunostained for Nrf2 and γ-H2AX, and the data are shown in Figure 8C.Cells exposed to PM10 exhibit no colocalization of the two proteins in the nucleus, as indicated by the

ML385 Reduces Protein Levels and Transcriptional Activity of Nrf2
The effects of ML385, a novel and specific Nrf2 inhibitor [31], were tested on HCE-2 cell viability by MTT assay (Figure 10A,B).Changes in cell viability were not observed after treatment of cells with 0-10 µM ML385 at 24 h (Figure 10A) and 48 h (Figure 10B).The protein levels of Nrf2 after treatment with ML385 are shown in Figure 10C.Data provide evidence that a reduction (p < 0.001) in Nrf2 occurred in the presence of ML385 vs.

ML385 Reduces Protein Levels and Transcriptional Activity of Nrf2
The effects of ML385, a novel and specific Nrf2 inhibitor [31], were tested on HCE-2 cell viability by MTT assay (Figure 10A,B).Changes in cell viability were not observed after treatment of cells with 0-10 µM ML385 at 24 h (Figure 10A) and 48 h (Figure 10B).The protein levels of Nrf2 after treatment with ML385 are shown in Figure 10C.Data provide evidence that a reduction (p < 0.001) in Nrf2 occurred in the presence of ML385 vs. control.In the presence of ML385, PM 10 exposure further reduced Nrf2 levels significantly (p < 0.001).Effects of ML385 on Nrf2 transcriptional activity were analyzed by measuring the mRNA levels of downstream targets: GPX4 and NQO1.Levels of GPX4 (Figure 10D) and NQO1 (Figure 10E) were significantly lowered (p < 0.001) by ML385 compared to the control.
control.In the presence of ML385, PM10 exposure further reduced Nrf2 levels significantly (p < 0.001).Effects of ML385 on Nrf2 transcriptional activity were analyzed by measuring the mRNA levels of downstream targets: GPX4 and NQO1.Levels of GPX4 (Figure 10D) and NQO1 (Figure 10E) were significantly lowered (p < 0.001) by ML385 compared to the control.

ML385 Treatment Abrogates the Protective Effects of SKQ1
To determine if SKQ1 mediates its protective effects via the Nrf2 pathway, the effects of ML385 on viability, MDA, and GSH levels were measured, and the data are shown in Figure 11A-C.Data provided evidence that without ML385, SKQ1 prevented (p < 0.001) PM 10 -induced loss in cell viability (Figure 11A).However, in the presence of ML385, SKQ1 did not alter the effects of PM 10 on viability.The MDA levels (Figure 11B) were significantly reduced (p < 0.001) by SKQ1 after PM 10 exposure without ML385; and with it, they were not affected.Similarly, without ML385, SKQ1 increased (p < 0.001) GSH levels (Figure 11C) after PM 10 exposure.However, in the presence of ML385, SKQ1 failed to do so.

ML385 Treatment Abrogates the Protective Effects of SKQ1
To determine if SKQ1 mediates its protective effects via the Nrf2 pathway, the effects of ML385 on viability, MDA, and GSH levels were measured, and the data are shown in Figure 11A-C.Data provided evidence that without ML385, SKQ1 prevented (p < 0.001) PM10-induced loss in cell viability (Figure 11A).However, in the presence of ML385, SKQ1 did not alter the effects of PM10 on viability.The MDA levels (Figure 11B) were significantly reduced (p < 0.001) by SKQ1 after PM10 exposure without ML385; and with it, they were not affected.Similarly, without ML385, SKQ1 increased (p < 0.001) GSH levels (Figure 11C) after PM10 exposure.However, in the presence of ML385, SKQ1 failed to do so.

Discussion
Epidemiological studies have linked increased exposure to air pollution to cardiopulmonary diseases such as lung cancer, stroke, and asthma [32][33][34].Recent studies indicate that air pollution can also cause neurodegenerative diseases [35] and reproductive issues such as infertility [36,37], and can affect the health of the eye [38].In fact, many studies have linked PM exposure to increased visits for several ocular diseases [39][40][41].Current studies have reported that PM2.5 induces oxidative stress, inflammation [42], apoptosis [43], and autophagy [44] in the cornea.However, the signaling pathways underlying PM10 induced corneal damage are not well understood.
Recently, we showed that exposing animals to PM10 disrupted Nrf2-regulated defenses in the cornea [23].The aim of the current in vitro study is to begin to gain an

Discussion
Epidemiological studies have linked increased exposure to air pollution to cardiopulmonary diseases such as lung cancer, stroke, and asthma [32][33][34].Recent studies indicate that air pollution can also cause neurodegenerative diseases [35] and reproductive issues such as infertility [36,37], and can affect the health of the eye [38].In fact, many studies have linked PM exposure to increased visits for several ocular diseases [39][40][41].Current studies have reported that PM 2 .5 induces oxidative stress, inflammation [42], apoptosis [43], and autophagy [44] in the cornea.However, the signaling pathways underlying PM 10 induced corneal damage are not well understood.
Recently, we showed that exposing animals to PM 10 disrupted Nrf2-regulated defenses in the cornea [23].The aim of the current in vitro study is to begin to gain an understanding of the mechanisms underlying PM 10 -induced damage to corneal epithelial cells and to investigate if Nrf2 signaling is essential for SKQ1-mediated protection.To this end, first, we tested the effects of PM 10 on HCE-2 cell viability using an MTT assay.This is a common tool used to measure overall health of live cells [45], and we observed that PM 10 decreased cell viability and that it was prevented by SKQ1.Our data are compatible with others who used PM 2 .5 exposure of human [46,47] and rat corneal epithelial (RCE) cells [48] and reported decreased cell viability.Another antioxidant, N-acetyl cysteine (NAC), was used to prevent the effects of PM 2 .5 on viability in RCE [48].
Reduced cell viability can occur due to a reduction in cellular proliferation or activation of cell death [49].To determine if the loss in cell viability was due to programmed cell death (e.g., apoptosis), we performed flow cytometry using Annexin V, a common staining method used to detect apoptotic cells, and established that PM 10 used at an acute dose of 100 µg/mL did not induce apoptosis.Similar findings have been reported before in a study that tested acute doses of PM 2 .5 (0-96 µg/cm 3 ) in lung epithelial cells, where 96 µg/cm 3 did not induce apoptosis [50].Another study in lung epithelial cells used a range of doses (0-100 µg/mL) and found that at 100 µg/mL, PM 2 .5 induced apoptosis [51].These studies show that contrasting reports regarding the effects of particulates on cells exist in the literature [50][51][52].These differences in experimental outcomes could be due to the size of the particulate, their source, and the time of exposure.It is also possible that PM 10 induced cytostatic stress at 100 µg/mL, which was not severe enough to induce apoptosis or necrosis.It has been established that cells undergo two distinct pathways: apoptosis following overwhelming stress and senescence resulting from low/transient stress [53].Some of the key drivers of senescence include various stressors, mitochondrial dysfunction, DNA damage and inflammation [54].In this study, we observed that PM 10 exposure caused increased DSBs (γ-H2AX staining), while SKQ1 pre-treatment before PM 10 prevented this.Similar effects of PM 10 on DNA damage foci have been previously reported in A549 cells and the use of Trolox, an antioxidant, was able to prevent most of the DSBs [55].Classical markers of senescence include cell cycle arrest (upregulation of p16 and p21 and phosphorylated p53) and increased β-gal staining [54].When we tested for these, we observed an upregulation of p21, p16 and phosphorylated p53, and accumulation of β-gal in HCE-2 cells exposed to PM 10 ; SKQ1 pre-treatment before PM 10 blocked these effects.Our data concur with those from other studies, in which PM 2 .5 /PM 10 exposure at a sublethal dose induced senescence in HCE-2 and A459 cells, which was reversed using antioxidants [46,55].Taken together, our data suggest that PM 10 causes senescence by inducing DNA damage, which leads to activation/nuclear localization of γ-H2AX, followed by the phosphorylation of protein p53 and cell cycle arrest.Furthermore, SKQ1, an antioxidant, prevented these events, suggesting that oxidative stress is important.
Another characteristic of senescence is the induction of an inflammatory response [56].We found that PM 10 induced the phosphorylation of NFκB and PI3K, providing evidence for PI3K/AKT/NFκB activation; this was prevented by SKQ1.It has been established that NFκB regulates gene-associated inflammatory responses such as iNOS, COX2, and TNF-α secretion [57].Exploring this, we found that PM 10 mediated an increase in COX2 and iNOS, which was prevented by SKQ1.A study using human embryonic lung-derived diploid fibroblasts (WI-38) showed that oxidative stress activated NFκB, COX2, and iNOS, inducing senescence, and complimented our work.They used cyanidin, a natural compound [58], which inhibited NFκB activation to downregulate COX2 and iNOS levels [58].
Previously, we showed that PM 10 increased oxidative stress and reduced ATP levels in HCE-2 cells, which were prevented by SKQ1 pre-treatment [23].We also tested the effects of PM 10 and SKQ1 on Nrf2 protein levels using primary (HCEC) and transformed (HCE-2) corneal epithelial cells.SKQ1 prevented a PM 10 -induced decline in Nrf2 levels irrespective of whether they were primary or transformed [23].Nrf2 is a transcription factor, a master regulator of antioxidant defenses, whose activation by particulates (dependent on dosage and exposure time) [19] has been reported in different cell types [59,60].Reports have shown that exposure to a high dose of PM 10 decreased Nrf2 levels [61] and failed to activate Nrf2-mediated antioxidant defense responses in the lung [61,62].Normally, the activation of Nrf2 is controlled by an adapter protein Keap1 which binds to cytoplasmic Nrf2, precipitating its degradation, but under stress, Nrf2 moves into the nucleus and begins to transcribe antioxidant genes [63].In this regard, we observed that PM 10 caused nuclear translocation of Nrf2, but decreased its transcriptional capacity and reduced transcript levels of downstream targets, which were prevented by SKQ1.Our data concur with another study involving alveolar epithelial cells, in which PM 10 exposure lowered Nrf2 levels, causing its nuclear translocation, which was prevented by a natural phenolic compound, gallic acid [61], a strong natural antioxidant agent with anti-inflammatory capabilities.
Current studies are focused on finding therapeutics to mitigate the damage caused by PM [4,61,64,65].One such compound is SKQ1, a mitochondria-targeted antioxidant that has been shown to be effective against PM 10 -mediated damage in the cornea both in vivo and in vitro [23].An ophthalmic formulation of SKQ1 called Visomitin has been successfully tested in a phase 3 clinical trial conducted in the USA for the treatment of dry eye disease [66].Recently, we showed that SKQ1 ameliorated the effects of PM 10 by inhibiting oxidative stress, restoring ATP levels, and modulating the Nrf2 pathway [23].
In the current study, to test if Nrf2 was required for SKQ1-mediated protection, we used ML385 (10 µM), an inhibitor of Nrf2 activity [30].Our data showed that ML385 effectively reduced Nrf2 and transcript levels of downstream targets.ML385 has been used and provided data compatible with ours in a sepsis model [67].
These data provide evidence to suggest that activation of Nrf2 is needed for SKQ1 protection.Our data also concur with a study in cortical neurons that underwent ischemia/reoxygenation where ML385 prevented the beneficial effects of kampferol [68].Therefore, Nrf2 involvement is critical for the activity of many therapeutic compounds [67][68][69][70][71].
The size of PM is important to consider when determining its effects on the eye [72].For example, large-sized PMs, such as PM 10 , are more able to exert a direct impact on the secretion of tears and on the barrier function of the epithelium compared to smaller-sized PMs such as PM 2 .5 [72].The size of PM also influences the prevalence of keratoconus [73].A recent study showed a strong correlation between the prevalence of keratoconus with PM 10, but only a moderate correlation with PM 2 .5 [73].In contrast, studies from our lab have shown that exposure to either PM 2 .5 [74] or PM 10 (unpublished data) can lead to worsening of Pseudomonas aeruginosa-induced keratitis, including early corneal perforation and thinning.
Most in vitro work has focused on PM 2 .5 and showed that it causes a dose-dependent loss in cell viability and increased oxidative stress [46], triggering inflammation [75] and p65-NFκB [48].It also impairs cell migration by altering FAK/RhoA signaling [76] and induces autophagy [44], DNA damage and senescence [46].In contrast, few studies have tested the in vitro effects of PM 10 .Future studies, including transcriptomics, and analysis of the epigenetic effects of the particulate, are planned to clarify the role of PM 10 vs. 2.5 in cornea.
In conclusion, we have shown that in corneal epithelial cells, PM 10 precipitates oxidative stress, disrupts Nrf2-mediated antioxidant defenses by reducing its transcriptional ability, and induces DNA damage and cell cycle arrest, leading to a senescence-like phenotype.SKQ1 pre-treatment reverses these effects.Furthermore, inhibition of Nrf2 using ML385 negates the protective effects of SKQ1 against PM 10 toxicity.

Materials and Methods
4.1.Tissue Culture, PM 10 , and SKQ1 Treatments HCE-2 ([50.B1], catalog# CRL-11135, ATCC, Gaithersburg, MA, USA) were cultured as reported previously [19].PM 10 (SRM 2787) was purchased from the National Institute of Standards and Technology (NIST).The powder was resuspended in 1X sterile phosphatebuffered saline (PBS) to obtain a stock solution at a concentration of 10 mg/mL which was then diluted in the media to obtain the final concentration of 100 µg/mL.SKQ1 (catalog# 934826-68-3, BOC Sciences, Shirley, NY, USA) was dissolved in 50% ethanol to obtain a 162 mM stock solution and serially diluted to 50 µM working solution in sterile 1X PBS, which was then diluted 1:1000 in media to obtain 50 nM SKQ1 final concentration.Cells were exposed to 100 µg/mL PM 10 for 24 h, as reported previously [23,77].To test the efficacy of SKQ1, a subset of cells was treated with 50 nM SKQ1 for one hour before addition of PM 10 .The cells without PM 10 exposure and SKQ1 pre-treatment are referred to as control.For all experiments, cells were harvested from each of the four groups (control, SKQ1, PM 10 , and PM 10 + SKQ1) after 24 h and were tested.To study Nrf2-mediated effects, cells were treated with the Nrf2 inhibitor ML385 (catalog# SML1833,10 µM, EMD Millipore, St. Louis, MO, USA) for 18 h before incubation with SKQ1 and PM 10 for an additional 24 h.ML385 was present in the culture medium throughout the course of these experiments.

Flow Cytometry
Flow cytometry was performed per the manufacturer's protocol.Annexin V conjugated to FITC (catalog# 51-65874, BD Pharmingen, San Diego, CA, USA) or propidium Iodide (PI; catalog# Bi51-66211E, BD Pharmingen, San Diego, CA, USA) were used with binding buffer (10 mM HEPES, 140 mM NaCl, 2.5 mM CaCl 2 , pH 7.4).Briefly, cells (control, SKQ1, and PM 10 -± SKQ1-exposed) were harvested after 24 h exposure using TrypLE™ Express Enzyme (catalog# 12604013, Thermo Fisher Scientific, Waltham, MA, USA).The TripLE was inactivated using media (Thermo Fisher Scientific, Waltham, MA, USA) containing 10% fetal bovine serum (FBS), washed twice, and resuspended to create a cell suspension composed of single cells at a concentration of 1 × 10 6 cells/mL in the binding buffer.A 100 µL aliquot of each sample was then transferred into 5 mL fluorescence associated cell-sorting (FACS) tubes.Unstained cells treated with Annexin V-FITC or PI only were used as controls.Next, 5 µL of Annexin V-FITC and 5 µL of PI were added to cells, vortexed gently, and incubated for 15 min in the dark at room temperature.Lastly, 0.4 mL of binding buffer was added to each sample and analyzed using the Accuri C6 system (BD Pharmingen, San Diego, CA, USA).

Western Blot Analysis
Western blot analysis was performed as described previously [23].Briefly, control, SKQ1, and PM 10 -± SKQ1-exposed or PM 10 ± SKQ1 ± ML385 cells were harvested after 24 h and lysed in RIPA buffer with protease and phosphatase inhibitors (catalog# sc-24948, Santa Cruz Biotech, Dallas, TX, USA) and the resulting supernatants were collected.To evaluate the Nrf2 protein levels in the nuclear extracts, nuclear fractions were obtained utilizing a nuclear extraction kit (catalog# 10009277, Cayman Chemical, Ann Arbor, MI, USA).Protein concentrations were determined using a bicinchoninic acid assay (BCA) kit (catalog# 23235, Thermo Fisher Scientific, Waltham, MA, USA) per the manufacturer's protocol.Samples were run on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) in Tris-glycine-SDS buffer, and electro-blotted onto nitrocellulose membranes (Bio-Rad, Hercules, CA, USA).Tris-buffered saline with 0.05% Tween 20 (TBST) and 5% nonfat milk (MTBST) was used for blocking (1 h) and the membranes were incubated with primary antibodies (1:1000, for all antibodies, see Table 1) in 2% MTBST overnight at 4 • C.After being washed 3 times with TBST, the membranes were incubated with horseradish peroxidase (HRP)-conjugated anti-rabbit secondary antibody (catalog# 7074, 1:2000; Cell Signaling Technology, Danvers, MA, USA) diluted in 5% MTBST at room temperature for 2 h and washed 3X in TBST.Bands were developed with Super signal West Femto Chemiluminescent Substrate (Thermo Fisher Scientific, Waltham, MA, USA), visualized using an iBright™ CL1500 Imaging System (Thermo Fisher Scientific, Waltham, MA, USA), and normalized to β-tubulin or β-actin (1:1000, for both).Phosphorylated proteins were normalized to their respective unphosphorylated proteins and further normalized to housekeeping proteins.The intensity of the bands was quantified using Image lab software version 6.1 and the data are shown as mean integrated density values (IDV) + SD.

β-Gal Staining
β-gal staining was performed using a BetaBlue TM staining Kit (catalog# 71074-3, EMD Millipore, St. Louis, MO, USA), per the manufacturer's protocol.Briefly, cells were grown on 12 mm coverslips and treated as described above in Section 4.1 for 24 h.Control, SKQ1, and PM 10 ± SKQ1-exposed cells were washed 2X with PBS, fixed in 4% paraformaldehyde (PFA) for 15 min at room temperature, rinsed 4X with PBS and incubated with BetaBlue staining solution for 3 h.Color progression was stopped by washing coverslips 4X with PBS.All images were obtained with a Leica DM4000B microscope at similar magnification and processed similarly in Adobe Photoshop 7.0.1.Cells were counted from 5 different fields for each coverslip (n = 2) and the percentage of β-gal positive cells was determined and plotted.

Immunofluorescence
Cells were grown on 12 mm coverslips in 6-well plates and treated with PM 10 ± SKQ1 as described above in Section 4.1 for 24 h.Control, SKQ1, and PM 10 ± SKQ1-exposed cells were then fixed in 4% paraformaldehyde (PFA) for 10 min, washed thrice with PBS.For Nrf2 localization, coverslips were treated with citrate buffer (10 mM citric acid, 0.1% Tween 20 pH 6.0) for antigen retrieval as previously published [78] and incubated with rabbit anti-Nrf2 antibody (Table 1) diluted 1:100 (Cell Signaling Technology) for 1 h at room temperature.To detect DNA double-strand breaks, cells were blocked with blocking buffer (1X PBS containing 5% normal goat serum and 0.3% TritonX-100) for 1 h and then incubated with anti-phospho-histone H2A.X antibody (Table 1) diluted in antibody-dilution buffer (1:1000, 1X PBS containing 1% BSA and 0.05% TritonX-100) overnight at 4 • C. To detect for colocalization of Nrf2 with DNA damage foci, cells were blocked as described above and then incubated with mouse-anti Nrf2 (1:100, SantCruz Biotechnology, Table 1) and anti-phospho-histone H2A.X antibody (1:800) diluted in antibody dilution buffer overnight at 4 • C. Coverslips were then washed and probed with goat anti-rabbit IgG coupled with Alexa 488 and/or goat anti-mouse IgG coupled with Alexa 594 (1:1500 in antibody dilution buffer, Table 1) for 2 h, washed 3X with PBS and mounted to slides using an antifade mounting medium (Vectashield) containing DAPI (Vector Laboratories Inc., Burlingame, CA, USA).A Leica TCS SP8 microscope (Deerfield, IL, USA) or Zeiss Axiophot 200M Apotome (White Plains, NY, USA) was used to image the cells and all images were processed in a similar manner via Adobe Photoshop 7.0.1.For colocalization studies, Pearson's correlation coefficient was determined using Image J2 software version 2.9.0 and interpreted as previously described [79].Cells were counted from 5 different fields (n = 4) for each coverslip, and the percentage of cells positive for nuclear and cytosolic Nrf2 staining was determined and plotted.

Transcriptional Activity Analysis of Nrf2
This assay was performed using Nrf2 transcription factor assay kit (catalog# 600590, Cayman Chemical, Ann Arbor, MI, USA) per the manufacturer's protocol and as reported previously [80].The assay kit analyses specific transcription factor DNA-binding activity in nuclear extracts.Briefly, nuclear fractions were obtained from control, SKQ1, and PM 10 ± SKQ1-exposed cells (24 h) obtained utilizing a nuclear extraction kit (catalog# 10009277, Cayman Chemical, Ann Arbor, MI, USA) and protein concentrations determined using a Bradford assay (catalog# 5000001, Bio-Rad, Hercules, CA, USA).Nuclear protein (20 µg) was incubated with a specific dsDNA sequence containing the Nrf2 response element, which was bound onto the wells of a 96-well plate.The binding of Nrf2 to the Nrf2 response element was detected by the addition of a specific primary anti-Nrf2 antibody.A secondary antibody conjugated to HRP was added to provide a colorimetric readout at 450 nm.
Table 2.Nucleotide sequence of the specific primers used for PCR amplification (Human).

Gene
Nucleotide Sequence Primer GenBank

GSH Assay
The GSH levels were analyzed by a glutathione assay kit (catalog# 703002, Cayman Chemical, Ann Arbor, MI, USA) per the manufacturer's protocol and as described previously [23,82].Briefly, control, PM 10 ± SKQ1, PM 10 + ML385, PM 10 + SKQ1 + ML385- exposed cells were harvested in 0.5 mL of ice cold 50 mM MES (2-(N-morpholino) ethanesulphonic acid) containing 1 mM EDTA, then homogenized, and the resulting supernatants were collected.Ellman's reagent was used to determine total GSH levels per the manufacturer's protocol and then normalized to total protein.Final GSH data were expressed as mean + SD.

Statistical Analysis
A one-way ANOVA followed by Bonferroni's multiple comparison test (GraphPad Prism version 8) was used to test the significance for MTT, flow cytometry, RT-PCR, ELISA, Western blot, GSH, and MDA assays.RT-PCR data from ML385 experiments were analyzed with an unpaired Student's t-test.Data were considered to be significant at p < 0.05.All experiments were repeated to ensure reproducibility and the combined data from these experiments are shown as mean + SD.

Figure 1 .
Figure 1.PM10 dose curve and effects of SKQ1 on HCE-2 cells exposed to PM10 for 24 h.dependent reduction in cell viability was observed after PM10 treatment.At the highest do µg/mL), nearly 50% of the cells were non-viable compared to 100 µg/mL, at which 80% of were viable (A).Cell viability was significantly reduced after PM10 exposure (100 µg/mL) vs or SKQ1.Pre-treatment with SKQ1 before PM10 prevented a loss in cell viability (B).Data ar as the mean + SD. (*** p < 0.001, n = 3).

Figure 1 .
Figure 1.PM 10 dose curve and effects of SKQ1 on HCE-2 cells exposed to PM 10 for 24 h.A dose-dependent reduction in cell viability was observed after PM 10 treatment.At the highest dose (1200 µg/mL), nearly 50% of the cells were non-viable compared to 100 µg/mL, at which 80% of the cells were viable (A).Cell viability was significantly reduced after PM 10 exposure (100 µg/mL) vs. control or SKQ1.Pre-treatment with SKQ1 before PM 10 prevented a loss in cell viability (B).Data are shown as the mean + SD. (*** p < 0.001, n = 3).

Figure 2 .
Figure 2. PM10 exposure did not induce apoptosis after 24 h.Flow cytometry data showed that representative scatter plots of Annexin V/PI staining in control, SKQ1, PM10 and PM10 + SKQ1 exposed groups were not significantly different.(n = 3).

Figure 2 .
Figure 2. PM 10 exposure did not induce apoptosis after 24 h.Flow cytometry data showed that representative scatter plots of Annexin V/PI staining in control, SKQ1, PM 10 and PM 10 + SKQ1 exposed groups were not significantly different.(n = 3).

2. 2 .
PM 10 Induces DNA Damage and Causes Cell Cycle Arrest DNA damage was examined by γ-H2AX staining in the nuclei of cells, and images and quantification are shown in Figure 3A,B.Compared to control or SKQ1 groups, the PM 10 -treated group had significantly more γ-H2AX staining indicative of DSBs in the nuclei (indicated by arrows, Figure 3A) and a significantly higher number of cells exhibiting DSBs (Figure 3B, p < 0.001).SKQ1 pre-treated cells (before PM 10 exposure) showed significantly fewer cells with nuclear DSBs (Figure 3A,B, p < 0.001).Levels of phosphorylated p53, p21, and p16 were measured after PM 10 exposure using Western blot analysis and the IDVs are shown in Figure 3C-E.PM 10 exposure vs. control or SKQ1 alone significantly increased the levels of phosphorylated p53 (Figure 3C, p < 0.001), p21 (Figure 3D, p < 0.001) and p16 (Figure 3E, p < 0.001); pre-treatment with SKQ1 before PM 10 significantly reduced (Figure 3C-E, p < 0.001) the levels of all proteins tested.

2. 3 .
SKQ1 Prevents a Senescence-like Phenotype Induced by PM10 β-gal staining was performed as another indicator of cell senescence.Representative images are shown in Figure 4A, and quantitation is shown in Figure 4B.Data show that compared to control or SKQ1 alone, PM10-exposed cells displayed a significant increase (Figure 4A,B, p < 0.001) in the number of β-gal positive cells (arrows).Pre-treatment with SKQ1 before PM10 significantly reduced the number of β-gal-stained cells (Figure 4A,B).

2. 3 .
SKQ1 Prevents a Senescence-like Phenotype Induced by PM 10 β-gal staining was performed as another indicator of cell senescence.Representative images are shown in Figure 4A, and quantitation is shown in Figure 4B.Data show that compared to control or SKQ1 alone, PM 10 -exposed cells displayed a significant increase (Figure 4A,B, p < 0.001) in the number of β-gal positive cells (arrows).Pre-treatment with SKQ1 before PM 10 significantly reduced the number of β-gal-stained cells (Figure 4A,B).

Figure 5 .
Figure 5. PM10 effects on NFκB activation by PI3K in HCE-2 cells after 24 h exposure.Data showed that PM10 exposure significantly increased the protein levels of p-PI3K (A) and p-NFκB (B) compared to control or SKQ1 alone.These effects are prevented in cells pre-treated with SKQ1 before PM10.Data are shown as mean + SD. (* p < 0.05, *** p < 0.001, n = 3).

Figure 5 .
Figure 5. PM 10 effects on NFκB activation by PI3K in HCE-2 cells after 24 h exposure.Data showed that PM 10 exposure significantly increased the protein levels of p-PI3K (A) and p-NFκB (B) compared to control or SKQ1 alone.These effects are prevented in cells pre-treated with SKQ1 before PM 10 .Data are shown as mean + SD. (* p < 0.05, *** p < 0.001, n = 3).

2. 5 .
PM 10 Causes Nuclear Translocation of Nrf2 Not Colocalized with γ-H2AX (DSBs) The effects of PM 10 on translocation of Nrf2 are presented in Figure 7A-H and the quantitation for nuclear and cytosolic Nrf2 staining is shown in Figure 7I,J.PM 10 exposure (Figure 7C,G) vs. control (Figure 7A,E) or SKQ1 alone (Figure 7B,F) caused nuclear translocation of Nrf2 (about 50% of the cells, Figure 7I, p < 0.001).The other 50% of the cells also exhibited cytosolic Nrf2 staining (Figure 7J, p < 0.001).SKQ1 pre-treated cells exposed to PM 10 (Figure 7D,H) exhibited similar Nrf2 expression as the control (Figure 7A,E) or SKQ1 alone (Figure 7B,F) groups.To confirm the immunostaining data, Western blot analysis of Nrf2 in the nuclear and cytosolic fractions was performed, and the results are shown in Figure8A,B.Data show that in the nuclear fractions of PM 10 -exposed vs. control or SKQ1-alone cell fractions, Nrf2 protein levels were significantly increased (Figure8A, p < 0.001).SKQ1 pre-treatment before PM 10 exposure significantly reduced (p < 0.05) Nrf2 protein in the nucleus.In contrast, cytosolic fraction showed significantly reduced Nrf2 protein levels (Figure8B, p < 0.05) after PM 10 exposure, which was reversed significantly (Figure8B, p < 0.05) by pre-treatment with SKQ1.To test if Nrf2 colocalizes with the DSBs in the nucleus, cells were immunostained for Nrf2 and γ-H2AX, and the data are shown in Figure8C.Cells exposed to PM 10 exhibit no colocalization of the two proteins in the nucleus, as indicated by the low Pearson's correlation coefficient (0.241).The control (0.098), SKQ1 (0.113) and PM 10 + SKQ1 (0.187) groups show low correlation coefficients (Figure8C), indicating no colocalization between Nrf2 and γ-H2AX.

Figure 11 .
Figure 11.Effects of ML385 on SKQ1-mediated protection after PM10 exposure on cell viability, MDA, and GSH levels.ML385 inhibited the ability of SKQ1 to rescue cell viability after PM10 exposure (A).Elevated MDA levels induced by PM10 exposure were not lowered by SKQ1 in the presence of ML385 (B).ML385 prevented the ability of SKQ1 to rescue lowered GSH levels (C).Data are presented as mean + SD. (** p < 0.01, *** p < 0.001, n = 3).

Figure 11 .
Figure 11.Effects of ML385 on SKQ1-mediated protection after PM 10 exposure on cell viability, MDA, and GSH levels.ML385 inhibited the ability of SKQ1 to rescue cell viability after PM 10 exposure (A).Elevated MDA levels induced by PM 10 exposure were not lowered by SKQ1 in the presence of ML385 (B).ML385 prevented the ability of SKQ1 to rescue lowered GSH levels (C).Data are presented as mean + SD. (** p < 0.01, *** p < 0.001, n = 3).

Table 1 .
Antibody information and gel percentage.