Airborne Particulate Matter (PM10) Inhibits Apoptosis through PI3K/AKT/FoxO3a Pathway in Lung Epithelial Cells: The Role of a Second Oxidant Stimulus

Outdoor particulate matter (PM10) exposure is carcinogenic to humans. The cellular mechanism by which PM10 is associated specifically with lung cancer includes oxidative stress and damage to proteins, lipids, and DNA in the absence of apoptosis, suggesting that PM10 induces cellular survival. We aimed to evaluate the PI3K/AKT/FoxO3a pathway as a mechanism of cell survival in lung epithelial A549 cells exposed to PM10 that were subsequently challenged with hydrogen peroxide (H2O2). Our results showed that pre-exposure to PM10 followed by H2O2, as a second oxidant stimulus increased the phosphorylation rate of pAKTSer473, pAKTThr308, and pFoxO3aSer253 2.5-fold, 1.8-fold, and 1.2-fold, respectively. Levels of catalase and p27kip1, which are targets of the PIK3/AKT/FoxO3a pathway, decreased 38.1% and 62.7%, respectively. None of these changes had an influence on apoptosis; however, the inhibition of PI3K using the LY294002 compound revealed that the PI3K/AKT/FoxO3a pathway was involved in apoptosis evasion. We conclude that nontoxic PM10 exposure predisposes lung epithelial cell cultures to evade apoptosis through the PI3K/AKT/FoxO3a pathway when cells are treated with a second oxidant stimulus.


Introduction
Air pollution is a problem that mainly affects large cities. It was estimated that in 2016 air pollution was responsible for 4,200,000 premature deaths [1]. Outdoor particulate matter with aerodynamic size ≤ 10 µm (PM 10 ) is an important component of air pollution. It is a complex mixture of organic and inorganic compounds, including metals and polycyclic aromatic hydrocarbons, among others. PM 10 is deposited in the upper respiratory tract, and epidemiological studies have shown PM-induced adverse effects on health such as chronic obstructive pulmonary disease, asthma, fibrosis, and lung cancer. In addition, an increase of 10 mg/m 3 of PM 2.5 and PM 10 was associated with an increase of 8% and 3.4-6% in cancer mortality, respectively [2,3]. Since 2013, PM 10 has been catalogued as carcinogenic to humans according to the International Agency for Research in Cancer (IARC) [4].

Pre-Exposure to PM 10 Followed by H 2 O 2 Treatment Induced AKT and FoxO3 Phosphorylation through PI3K Activation
First, cell cultures were exposed to PM 10 (10 µg/cm 2 ), H 2 O 2 (500 µM), or LY294002 (LY) inhibitor (50 µM), and we found that none of the concentrations tested had influence on cell viability (Table 1). The selection of the concentration used in this study was based on the dosimetric evaluation of total PM deposition in the lungs of exposed citizens from Rubidoux, California, United States. In this city, the concentration of PM is 79 µg/cm 2 over a 24 h period. Using a dosimetry approach considering variations in airway anatomy, nasal breathing, and deposition at bifurcation points, the above-mentioned PM concentration of exposed humans reconciles with in vitro models in a PM concentration ranging from 0.2 to 20 µg/cm 2 [21]. Table 1. Cell viability of lung epithelial A549 cells exposed to outdoor particulate matter with aerodynamic size ≤ 10 µm (PM 10  Control: cells cultured during 24 h with free fetal bovine serum, washed and replaced with fresh cell culture medium, and incubated for a second period of 24 h. LY294002: Cells exposed to 50 µM of LY294002 inhibitor for 48 h. PM 10 + H 2 O 2 : cells exposed to PM 10 for 24 h and exposed 500 µM H 2 O 2 for 24 h. PM 10 + LY + H 2 O 2 : cells pre-exposed to 10 µg/cm 2 PM 10  The concentration of 500 µM H 2 O 2 mimics an oxidant stimuli unable to induce cytotoxicity in lung epithelial cells [5]. The concentration of LY294002 inhibitor tested here has been used successfully for PI3K pathway inhibition in the same cell line [22]. In addition, we performed a dose-response curve for pFoxO3a Ser253 inhibition ( Figure 1). The concentration in which we observed the inhibitory effect of LY294002 in pFoxO3a Ser253 was 50 µM. Our design was based in the hypothesis that a healthy population might have undetectable alterations in the respiratory tract, but a second oxidant exposure could induce damage that is undetectable after PM 10 exposure. The second oxidant exposure could be related to infections or allergies, leading to a higher number of alterations. Specifically, we focused on a particular pathway that might partially explain the epidemiological evidence that points out the link between PM 10 exposure and lung cancer development [2,3,23]. We have seen that PM 10 exposure induces several alterations, including double-stranded DNA breaks, without affecting cell viability, which has raised the concern of apoptosis evasion. Since alterations in the PI3K/AKT/FoxO3a pathway have been described as cell survival mechanisms in several types of cancer, we evaluated the phosphorylation state of pAKT Ser473 and pAKT Thr308 after PM 10 exposure and no changes were detected ( Figure 2). However, a 2.5-fold and 1.8-fold increase in the pAKT Ser473 and pAKT Thr308 state, respectively, were detected in cell cultures pre-exposed to PM 10 followed by H 2 O 2 treatment, and this increase was completely prevented by the incubation of LY294002 inhibitor in both pAKT Ser473 and pAKT Thr308 , which highlights that PI3K inhibition plays a role in the pAKT Ser473 state mediated by PM 10 exposure ( Figure 3). The 48 h incubation with H 2 O 2 (500 µM) and the 48 h LY294002 inhibitor treatment in cells exposed to H 2 O 2 (500 µM) led to similar changes in pAKT Ser473 (2.33-fold and 2.61-fold; Figure 3A,B). Apparently, a 48 h treatment with H 2 O 2 (500 µM) causes a similar pAKT Ser473 increase, nevertheless, the LY294002 inhibitor was unable to prevent this effect, suggesting that an oxidant stimulus such as H 2 O 2 induced and increased pAKT Ser473 but by an independent PI3K signaling. However, the 48 h incubation with H 2 O 2 (500 µM) and the 48 h LY294002 inhibitor treatment in cells exposed to H 2 O 2 (500 µM) showed no changes in the levels of pAKT Thr308 ( Figure 3C,D).  Levels of pAKT Ser473 , pAKT Thr308 , and pFoxO3a Ser253 proteins. Representative blot of (A) levels of pAKT Ser473 assessed; (B) densitometry of pAKT Thr308 protein levels using ImageJ software; (C) assessed pAKT Th308 levels; (D) densitometry of pAKT Ser473 protein levels using ImageJ software; (E) assessed pFoxO3a Ser253 levels; and (F) densitometry of pFoxO3a Ser253 protein levels using ImageJ software. Lung epithelial cells were exposed to PM10 (10 µg/cm2) for 24 h, treated with H2O2 (500 µM) for 24 h, and treated with LY294002 inhibitor (LY) (50 µM) for 1 h before the H2O2 (500 µM) treatment. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) protein was used as loading control for western blot. The image is representative of three independent experiments, and values are the mean ± SD of three independent experiments.   Levels of pAKT Ser473 , pAKT Thr308 , and pFoxO3a Ser253 proteins. Representative blot of (A) levels of pAKT Ser473 assessed; (B) densitometry of pAKT Thr308 protein levels using ImageJ software; (C) assessed pAKT Th308 levels; (D) densitometry of pAKT Ser473 protein levels using ImageJ software; (E) assessed pFoxO3a Ser253 levels; and (F) densitometry of pFoxO3a Ser253 protein levels using ImageJ software. Lung epithelial cells were exposed to PM10 (10 µg/cm2) for 24 h, treated with H2O2 (500 µM) for 24 h, and treated with LY294002 inhibitor (LY) (50 µM) for 1 h before the H2O2 (500 µM) treatment. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) protein was used as loading control for western blot. The image is representative of three independent experiments, and values are the mean ± SD of three independent experiments. Figure 2. Levels of pAKT Ser473 , pAKT Thr308 , and pFoxO3a Ser253 proteins. Representative blot of (A) levels of pAKT Ser473 assessed; (B) densitometry of pAKT Thr308 protein levels using ImageJ software; (C) assessed pAKT Th308 levels; (D) densitometry of pAKT Ser473 protein levels using ImageJ software; (E) assessed pFoxO3a Ser253 levels; and (F) densitometry of pFoxO3a Ser253 protein levels using ImageJ software. Lung epithelial cells were exposed to PM 10 (10 µg/cm2) for 24 h, treated with H 2 O 2 (500 µM) for 24 h, and treated with LY294002 inhibitor (LY) (50 µM) for 1 h before the H 2 O 2 (500 µM) treatment. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) protein was used as loading control for western blot. The image is representative of three independent experiments, and values are the mean ± SD of three independent experiments. Representative blot of (A) pAKT Ser473 and total AKT, as well as (C) pAKT Thr308 and total AKT assessed by western blot and densitometry of (B) pAKT Ser473 and (D) pAKT Thr308 of levels using ImageJ software. Lung epithelial cells were pre-exposed to PM10 (10 µg/cm 2 ) for 24 h, and then cells were treated with H2O2 (500 mM) for 24 h. In lane 3 and 6 of the blot (panel A) are cells treated with LY294002 inhibitor (50 µM) for 1 h before the H2O2 (500 µM) treatment. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) protein was used as loading control for western blot. pAKT Ser473 ** p < 0.001 versus control; ** p < 0.001 versus PM10 + H2O2 versus PM10 + LY + H2O2. pAKT Thr308 *** p < 0.0001 versus control; *** p < 0.0001 versus PM10 + H2O2 versus PM10 + LY + H2O2. The image is representative of three independent experiments, and values are the mean ± SD of three independent experiments.
Then, levels of FoxO3a S253 were assessed and found a 1.2-fold increase in cell cultures exposed to PM10 plus H2O2. Moreover, this increase was prevented by inhibition of PI3K using the LY294002 inhibitor ( Figure 4A,B). By contrast, none of the other treatments had this increase, suggesting that PM10 exposure is responsible for the increase in FoxO3a Ser253 rate. Representative blot of (A) pAKT Ser473 and total AKT, as well as (C) pAKT Thr308 and total AKT assessed by western blot and densitometry of (B) pAKT Ser473 and (D) pAKT Thr308 of levels using ImageJ software. Lung epithelial cells were pre-exposed to PM 10 (10 µg/cm 2 ) for 24 h, and then cells were treated with H 2 O 2 (500 mM) for 24 h. In lane 3 and 6 of the blot (panel A) are cells treated with LY294002 inhibitor (50 µM) for 1 h before the H 2 O 2 (500 µM) treatment. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) protein was used as loading control for western blot. pAKT Ser473 ** p < 0.001 versus control; ** p < 0.001 versus PM 10 + H 2 O 2 versus PM 10 + LY + H 2 O 2 . pAKT Thr308 *** p < 0.0001 versus control; *** p < 0.0001 versus PM 10 + H 2 O 2 versus PM 10 + LY + H 2 O 2 . The image is representative of three independent experiments, and values are the mean ± SD of three independent experiments.
Then, levels of FoxO3a S253 were assessed and found a 1.2-fold increase in cell cultures exposed to PM 10 plus H 2 O 2 . Moreover, this increase was prevented by inhibition of PI3K using the LY294002 inhibitor ( Figure 4A,B). By contrast, none of the other treatments had this increase, suggesting that PM 10 exposure is responsible for the increase in FoxO3a Ser253 rate.

Figure 4.
Representative blot of (A) pFoxO3a Ser253 and total FoxO3a and (B) densitometry of levels using ImageJ software. Lung epithelial cells were pre-exposed to PM10 (10 µg/cm 2 ) for 24 h, and then cells were treated with H2O2 (500 mM) for 24 h. In lane 3 and 6 of the blot (panel A) are cells treated with LY294002 inhibitor (50 µM) for 1 h before the H2O2 (500 µM) treatment. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) protein was used as loading control for western blot. * p < 0.01 versus control; ** p < 0.001 versus PM10 + H2O2. The image is representative of three independent experiments, and values are the mean ± SD of three independent experiments.

Pre-Exposure to PM10 Decreased Catalase and p27 kip1 Protein through PI3K Activation
Catalase and p27 kip1 protein levels are modulated by AKT/FoxO3a ( Figures 5 and 6), and we found a 38.1% and 62.7% downregulation in both protein levels, respectively, in cell cultures exposed to PM10 followed by H2O2 ( Figures 5B and 6B). In both cases, PI3K inhibition completely prevented the decrease of catalase and p27 kip1 levels, while it was unaffected by 48 h H2O2 treatment or H2O2 and LY294002 treatments ( Figures 5B and 6B). Interestingly, the downregulation was higher for p27 kip1 than for catalase, which might imply that the PI3K/AKT/FoxO3a pathway has an important role in p27 kip1 expression, while for catalase other control expression mechanisms are involved. Indeed, the number of activators and repressors reported to be involved in catalase expression has been increasing and includes SP1, NF-Y, XBP1, NRF-2, and C/EBP-β, and PPARγ and MAPK signaling, respectively, among others (Revised by Glorieux et al., 2015) [24].

Pre-Exposure to PM 10 Decreased Catalase and p27 kip1 Protein through PI3K Activation
Catalase and p27 kip1 protein levels are modulated by AKT/FoxO3a ( Figures 5 and 6), and we found a 38.1% and 62.7% downregulation in both protein levels, respectively, in cell cultures exposed to PM 10 followed by H 2 O 2 ( Figures 5B and 6B). In both cases, PI3K inhibition completely prevented the decrease of catalase and p27 kip1 levels, while it was unaffected by 48 h H 2 O 2 treatment or H 2 O 2 and LY294002 treatments ( Figures 5B and 6B). Interestingly, the downregulation was higher for p27 kip1 than for catalase, which might imply that the PI3K/AKT/FoxO3a pathway has an important role in p27 kip1 expression, while for catalase other control expression mechanisms are involved. Indeed, the number of activators and repressors reported to be involved in catalase expression has been increasing and includes SP1, NF-Y, XBP1, NRF-2, and C/EBP-β, and PPARγ and MAPK signaling, respectively, among others (Revised by Glorieux et al., 2015) [24].

Inhibition of Apoptosis via PI3K/AKT/FoxO3a by Pre-Exposure to PM 10 Followed by H 2 O 2 Treatment
Cell cultures pre-exposed to PM 10 were treated with H 2 O 2 , and this combination had no influence on apoptosis. However, the LY294002 inhibitor revealed that these treatments had a 55.98% increase in apoptosis compared to cells exposed to PM 10 plus H 2 O 2 ( Figure 7). Cell cultures exposed to H 2 O 2 for 48 h had 41.8% increased apoptosis, while H 2 O 2 for 48 h plus LY294002 inhibitor had 50.8% increased apoptosis (Figure 7), and importantly, we found that none of the concentrations tested (PM 10 (10 µg/cm 2 ), H 2 O 2 (500 µM), or LY294002 (LY) inhibitor (50 µM)) had influence on cell viability (Table 1).

Figure 5.
Representative blot of (A) assessed catalase levels and (B) densitometry using ImageJ software. Lung epithelial cells were pre-exposed to PM10 (10 µg/cm 2 ) for 24 h, and then cells were treated with H2O2 (500 µM) for 24 h. In lane 3 and 6 of the blot (panel (A)) are cells treated with LY294002 inhibitor (LY) (50 µM) for 1 h before the H2O2 (500 µM) treatment. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) protein was used as loading control for western blot. * p < 0.001 versus control; ** p < 0.01 versus PM10 + H2O2. The image is representative of three independent experiments, and values are the mean ± SD of three independent experiments.    Lung epithelial cells were pre-exposed to PM 10  influence on apoptosis. However, the LY294002 inhibitor revealed that these treatments had a 55.98% increase in apoptosis compared to cells exposed to PM10 plus H2O2 (Figure 7). Cell cultures exposed to H2O2 for 48 h had 41.8% increased apoptosis, while H2O2 for 48 h plus LY294002 inhibitor had 50.8% increased apoptosis (Figure 7), and importantly, we found that none of the concentrations tested (PM10 (10 µg/cm 2 ), H2O2 (500 µM), or LY294002 (LY) inhibitor (50 µM)) had influence on cell viability (Table 1).  This result reveals that PM 10 exposure activates the PI3K/AKT/FoxO3a pathway, which prevents cell death after the second oxidant challenge. The outcome of forced apoptosis evasion induced by PM 10 + H 2 O 2 exposure might lead to replication of lung epithelial cells with unrepaired DNA. Unfortunately, the apoptosis resistance plays a central role in carcinogenesis, tumor development and progression, and chemotherapy resistance. In addition, replication of these cells could be a key point in which newly divided cells acquire a cancer-like phenotype [25].

Discussion
In this study, we demonstrated that PM 10 exposure had no influence on the PI3K/AKT/FoxO3a pathway in lung epithelial A549 cells. However, cell cultures previously exposed to PM 10 were challenged to H 2 O 2 , which induced an upregulation in AKT (pAKT Ser473 and pAKT Thr308 )/pFoxO3a Ser253 . On one hand, H 2 O 2 would mimic a prooxidant environment that can be reached in the respiratory tract by bacterial or viral infections [26,27], and on the other hand, the deregulation of AKT/FoxO3a signaling has a central role during cellular disturbances found in diseases such as leukemia [28], diabetic kidney disease [29], and breast cancer [30], among others.
Specifically, to investigate whether PM 10 exposure could activate the above-mentioned pathway, we decided to use 2-(4-morpholinyl)-8-phenyl-4H-1-benzopyran-4-one, which is commercially known as LY294002. This compound inhibits purified PI3K without inhibition of other kinases or enzymes requiring ATP [31]. This compound is an analog of quercetin, used in the past as an unspecific kinase inhibitor; however, quercetin acts as an inhibitor at IC50 = 3.8 µM while LY294002 inhibitor displays an IC50 = 1.8 µM. Based on this information, LY294002 has been widely used as a PI3K inhibitor in in vivo [32] and in vitro models, including cortical neurons [31] and lung epithelial A549 cells [22]. LY294002, a reversible inhibitor of PI3K, and wortmannin, an irreversible inhibitor of PI3K, are the most described and used PI3K inhibitors [31,33,34]. Although both are considered as specific and invaluable tools to study the PI3K pathway, the inhibitory profile of LY294002 is broader than wortmannin. LY294002 inhibit mTOR (mammalian target of rapamycin), DNA-PK (DNA-dependent protein kinase), as well as other kinases, such as CK2 (casein kinase 2) and Pim-1 [35][36][37]. Also, LY294002 activates AKT and accumulate phospho-AKT at the intracellular membrane, a condition that is abolished by treatment with wortmannin [38]. Thus, the effects in apoptosis evasion by the combination of PM 10 and H 2 O 2 can be prevented by the non-selective inhibitor of PI3K selected for our study, displaying an effect associated to cell survival in cancer cells.
To our knowledge, we demonstrate for the first time that PM 10 exposure followed by an oxidant stimulus with H 2 O 2 upregulates the pAKT Ser473 /pAKT Thr308 /pFoxO3a Ser253 , with a reduction in apoptosis mediated by PI3K in lung epithelial cells, which could mimic air pollutants exposure followed by oxidant endogenous stimulus seen, for instance, during bacterial infections. Nevertheless, a 65% downregulation of FoxO3a protein levels in the entire lung tissue from mice exposed to PM 2.5 (17.7 µg/m 3 ) inhalation over 12 weeks has been previously reported [39]. However, the rate of phosphorylation was undetermined, and the quantification performed showed the global FoxO3a levels, whereas our study reveals the response of a specific cell lineage. In addition, FoxO3a Ser253 upregulation was only found after H 2 O 2 was used as a second oxidant stimulus, which suggests that PM 10 predisposes to deregulation in AKT/FoxO3a signaling that could evade apoptosis.
We also found that AKT (pAKT Ser473 and pAKT Thr308 )/pFoxO3a Ser253 activation was associated with a decrease in catalase protein, leading to a reduction in the antioxidant defense since this enzyme is responsible for H 2 O 2 detoxification [40]. Thus, cells with downregulated catalase might accumulate higher amounts of H 2 O 2 that would lead to a highly oxidant environment. In addition, H 2 O 2 is the precursor of hydroxyl radical, a well-known species with strong DNA affinity that has no enzymatic defense for detoxification. Indeed, we have already demonstrated that PM 10 exposure induced a decrease in the antioxidant enzymatic activity of catalase, superoxide dismutase, glutathione reductase, and glutathione S-transferase without affecting cell death, and this effect was not reverted by hydroxyl radical chelation [5], which highlights that PM 10 induces irreversible enzymatic activity damage in lung epithelial cells.
The decrease of p27 kip1 levels in this study suggests a deregulation in the progression from the G 1 to the S phase of the cell cycle that can be associated with the inhibition of apoptosis mediated by the AKT (pAKT Ser473 and pAKT Thr308 )/pFoxO3a Ser253 pathway activated by PM 10 exposure in combination with H 2 O 2 in lung epithelial cells. However, what is concerning is the strong evidence of decrease or loss of p27 kip1 in several tumors, including breast, colon, and prostate adenocarcinomas [20]. In addition, Liu and colleagues recently suggested that some cell-cycle regulators exert important independent-cycle functions, and p27 kip1 is listed in their work, emphasizing that this protein has a role in postmitotic neurons and immune T cells [41]. Therefore, deregulation of p27 kip1 levels could have an impact on pulmonary T cells [42] and immune cells from bloodstream [43,44] and brain [45]. Perhaps exposure solely to PM 10 is not enough to induce deregulation of some proteins such as p27 kip1 , but a second oxidant challenge could trigger tissue dysfunction. Besides infections as a possible second oxidant challenge, H 2 O 2 is detected in exhaled breath of patients with non-small cell lung cancer but also in healthy cigarette smokers [46], which highlights the possibility that PM 10 exposure can really be followed by a H 2 O 2 oxidant environment with unknown consequences.
Regarding the PM 10 exposure model, findings of this study represent the effect of the complete particle (PM 10 ) with all the organic and non-organic components, and the contribution of each component cannot be dissected. However, the link between lung cancer and PM 10 exposure is attributed to the inhalation of the entire particle plus smaller fractions. Importantly, because PM 10 is derived from anthropogenic activity and each city has non-identical PM 10 chemical compositions, a different apoptotic evasion footprint would be expected according to the city where the PM 10 is located. In this case, PM 10 collected from Mexico City is derived from traffic emissions from 5,000,000 vehicles, volatile organic compounds derived from thinners, degreasers, cleaners, lubricants, and liquid fuels used in shops like drycleaners, and liquefied petroleum gas leaks in houses, while metals come from the industrial sector located on the north side of Mexico City. Unfortunately, wind transports other contaminants from adjacent cities to Mexico City's atmosphere.
Based on the IARC Scientific Publication No. 161 published in 2013, a complete PM 10 particle is spatially and temporally heterogeneous, which means that there is geographical variability of composition and concentration among the PM analyzed from Europe, Africa, and the United States. However, epidemiological studies clearly show an association between PM exposure and the risk of cancer, while the experimental research provides a basis for the plausibility of a risk of cancer regardless of the city in which the PM was collected. Beyond the chemical heterogeneity, air pollution's classification as a group 1 carcinogen is attributed to the fact that some carcinogenic or highly toxic compounds derived from fuel combustion are common, regardless of the combustion source. For instance, some countries can have higher fuel combustion for industrial purposes, whereas others might have combustion for heating, cooking, and heavy automobile traffic, but these sources might generate formaldehyde, acrolein, benzene, toluene, 1,3-butadiene, benzo[a]pyrene, iron, and sulfates, among others.
We recognize that the findings of this study were identified in A549 cells, which are already transformed cells. Nevertheless, if this cell line was able to display such important alterations, perhaps normal lung epithelial cells might have higher susceptibility to stress induced by PM 10 exposure followed by oxidant stimulus. Indeed, even if the results can only be restricted to lung epithelial cells, we cannot discard the notion that a similar response could be found in bronchial cells, which are also targeted by PM 10 exposure.
Finally, we cannot dismiss that other apoptosis-evasion mechanisms could be involved and were not explored in this study. For instance, the mTOR pathway, mitochondrial disturbances, and immune cell inactivation could act together at the tissue level to explain why PM 10 exposure is classified as carcinogenic to humans. In addition, early PM 10 exposure time points must be investigated because PI3K activation is an immediate cellular response. As an example, viral infection activates pAKT Ser473 after 15 min with a sustained increase until 4 h [47]. Besides, it is probable that the unseen decrease of pAKT Ser473 levels could be related to negative regulatory feedbacks after PI3K inhibition with LY294002. For example, it has been reported that insulin feedback induced by PI3K inhibitors could reactivate the PI3K-mTOR signaling in tumors [48]. Further studies considering a time course evaluation of pAKT Ser473 under the context of pre-exposure of PM 10 and treatment with H 2 O 2 need to be performed in order to elucidate if treatment with LY294002 is capable of reduce AKT phosphorylation.

PM 10 Sampling
PM 10 was collected in a residential-urban area in Mexico City. A large volume particle collector was used with 1.13 m 3 /min flow (GMW model 1200 VFCHVPM10 Sierra Andersen, Smyrna, GA, USA) and equipped with cellulose filters (Sartorius AG, Goettingen, Germany). After collection, the nitrocellulose filters were kept in a desiccator at 4 • C in the dark. The particles were recovered from the membranes and stored in a vial of glass, sterile and free of endotoxins. The PM 10 was kept at 4 • C in a desiccator and in the dark until used [49].

Lung Epithelial A549 Cell Culture
The lung epithelial A549 cell line derived from a human lung adenocarcinoma used in this work was obtained from the American Type Culture Collection (CCL-185; ATCC, Manassas, Virginia, United States). Cell cultures were grown with F12 Kaighn's culture medium (21127-022; Gibco, Grand Island, New York, United States) supplemented with 10% fetal bovine serum (FBS;16000-044; Gibco, Carlsbad, California, United States) at 37 • C and in a 5% CO 2 atmosphere. The cells were growing to 70% confluence in early passages.

PM 10 Exposure, H 2 O 2 Treatment, and LY294002 Inhibitor in Cell Cultures
The experiments were carried out in 6-well plates; 125,000 cells were seeded and after 24 h, cell cultures were exposed to the different treatments. For PM 10 exposure, 95 µg of PM 10 (solid particles) were weighed in a sterilized and free-endotoxin vial. Then, 200 µL of F12 Kaighn's medium supplemented with 10% FBS were added to the vial of particles and gently mixed by pipetting for 10 s. Next, this suspension (200 µL containing the PM 10 particles) was added to a 6-well plate with seeded cells, which had 1800 µL of F12 Kaighn's medium supplemented with 10% FBS in each well. The area of each well from the 6-well plate is 9.5 cm 2 ; therefore, the final PM 10 concentration is 10 µg/cm 2 . Cells were exposed to PM 10 for 24 h with this procedure when PM 10 treatment is indicated in this study. For combination with H 2 O 2 , the culture medium was removed, and cells were washed with PBS and treated with 500 µM H 2 O 2 (TA-125-HP; Thermo Fisher, Fremont, California, United States) for 24 h. In addition, we used LY294002 (9901S; Cell Signaling, Danvers, Massachusetts, United States) as an inhibitor of the AKT/Fox3a pathway at 50 µM in DMSO before the H 2 O 2 treatment. The LY294002 concentration was determined through a dose-response curve over inhibition of pFoxO3a Ser253 . Briefly, cells A549 were cultured in F12 Kaighn's medium supplemented with 10% of FBS. After 24 h, the cells were incubated with LY294002 inhibitor (0.0, 0.5, 1, 2, 5, 10, 25, and 50 µM) for 24 h, then cells were lysed (20 mM Tris, 150 mM NaCl, 1% NP-40) using phosphatase and protease inhibitors (78440; Thermo Fisher, Rockford, Illinois, United States). The inhibitory effect of LY294002 over pFoxO3a Ser253 was evaluated through western blot of pFoxO3 Ser253 , FoxO3a, and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as housekeeping control (two independent experiments). Control group was supplemented with DMSO in order to evaluate the effect caused by treatments and avoid any bias in the experimental design. Treatments labeled as H 2 O 2 + H 2 O 2 were incubated for 24 h with 500 µM H 2 O 2 , then cells were extensively washed with cell culture medium without serum, and, again, cells were treated with 500 µM H 2 O 2 for 24 h.

Cellular Viability
After cell cultures were exposed to the different treatments (Section 4.3), cell viability was evaluated using the trypan blue dye exclusion assay. Briefly, the cells' exposure to different treatments were collected using 0.25% trypsin-EDTA (25200055; Thermo Fisher, Carlsbad, California, United States) and dyed with trypan blue (0.4%), and then 500 cells were counted using an inverted microscope (Leica). Cells that did not incorporate the trypan blue dye were considered viable cells. The results are presented as the percentage of cells from three independent experiments.

Determination of Protein Levels
After treatments, cell cultures were lysed (20 mM Tris, 150 mM NaCl, 1% NP-40) and phosphatase and protease inhibitors (78440; Thermo Fisher, Rockford, Illinois, United States) were added. Protein content was analyzed using the bicinchoninic acid reagent (SIGMA: B9643) using bovine serum albumin as standard curve. The principle is based on the reduction of copper (2+) to copper (1+) by proteins, and then copper (1+) reacts with bicinchoninic acid that absorbs light at 562 nm [50]. The protein content in lysates from cell cultures was compared with a standard solution of bovine serum albumin (23209; Thermo Fisher, Carlsbad, California, United States).

Apoptosis Measurement
The apoptosis was determined by staining cells with Annexin V-fluorescein isothiocyanate (Ex/Em at 495 nm/529 nm) [52]. A549 cells (8 × 10 5 /well) were seeded in plates of 6-well culture slides with F12K medium supplemented with 10% FBS for 24 h. After the cells were treated as described in Section 4.3, they were washed with PBS and centrifuged at 200 g for 5 min. The pellet obtained was resuspended in 100 µL of Annexin V solution (Annexin V FLUOS dying kit, 1858777; Roche Diagnostics GmbH, Mannheim, Germany) and incubated 15 min at 15 • C. FACSDiva software v.1.1 was used for data acquisition and analysis. The early and late apoptosis were considered and were used as control undyed cells and cells dyed with Annexin-V and propidium iodide.

Statistical Analysis
The results are presented as the means ± standard deviations of three independent experiments. The statistical analysis of variance was applied with the multiple comparisons test with the Prism Program version 5 (GraphPad Software). A value of p < 0.05 was considered statistically significant.

Conclusions
In this study, we demonstrated that lung epithelial A549 cells pre-exposed to PM 10 had higher susceptibility to developing an upregulation of the AKT/FoxO3a pathway mediated by PI3K activity that leads to apoptosis evasion when cells are challenged to a second oxidant stimulus, such as H 2 O 2 .