Reactive Oxygen Species Generated by Polycyclic Aromatic Hydrocarbons From Ambient Particulate Matter Enhance Vascular Smooth Muscle Cell Migration Through MMP Upregulation and Actin Reorganization

Epidemiological studies have suggested that elevated concentrations of particulate matter (PM) are strongly associated with the incidence of atherosclerosis; however, the underlying cellular and molecular mechanisms of atherosclerosis by PM exposure and the components that are mainly responsible for this adverse effect remain to be established. In this investigation, we evaluated the effects of ambient PM on vascular smooth muscle cell (VSMC) behavior. In addition, the effects of polycyclic aromatic hydrocarbons (PAHs), including oxygenated PAHs (oxy-PAHs), on VSMC migration and the underlying mechanisms were examined.

molecular oxygen [15]. It is well known that PAHs and oxy-PAHs induce severe redox stress in cells and tissues, leading to the oxidation of nucleic acids, proteins, and lipids [16,17] as well as carcinogenicity [18,19]. In a previous study, we showed that in cardiomyocytes, oxy-PAHs induced higher electrophysiological instability than PAHs, which might be due to the increased generation of ROS. In addition, we showed that the amount and number of oxy-PAHs contained in WPM were higher in SPM, leading to higher electrophysiological instability [20]. However, few studies have investigated the underlying mechanisms in the vasculature, which is composed of endothelial cells (ECs) and vascular smooth muscle cells (VSMCs). For example, in ECs, cellular glutathione is protective against the cytotoxic effects induced by benzo [a]pyrene-1,6-quinone, a PAH component of air pollution [21]. In addition, benzo [a]pyrenes increase VSMC migration and invasion through the upregulation of matrix metalloproteinases (MMPs) [22]. Nevertheless, further cellular and molecular alterations in VSMCs induced by ambient PMs and the components that participate in these alterations, and their underlying mechanisms still need to be elucidated.

Ambient PM induces VSMCs migration through ROS generation
To investigate the effects of ambient PM on VSMC behavior, we analyzed the alterations in VSMC proliferation and migration. As shown in Fig. 1a, both PM collected in summer (SPM) and winter (WPM) signi cantly increased VSMC migration in both the 2-D wound healing assay and the 3-D Boyden chamber assay. The increased migration was signi cantly higher in WPM-treated VSMCs than in those treated with SPM. In addition, we observed that the increased level of migration was even higher than that of PDGF-treated VSMCs (a positive control). However, VSMC proliferation was not altered by treatment with ambient PM (Fig. 1b), demonstrating that ambient PM only changes the migratory activities of VMSCs. Based on previous reports that demonstrate that ROS may mediate VSMC migration [23,24], we evaluated ROS generation and their effects on VSMC migration. We observed that ROS generation was signi cantly increased in ambient PM-treated VSMCs, with a larger increase observed in cells treated with WPM than those treated by SPM. Furthermore, NAC treatment signi cantly reduced ambient PM-induced ROS levels (Fig. 1c) and subsequently inhibited VSMC migration (Fig. 1d). These results demonstrate that ambient PM induces VSMC migration through ROS generation.

Ambient PM regulates matrix metalloproteinase expression and activity
The altered levels of MMPs in ambient PM-treated VSMCs were investigated because MMPs are critical regulators of VSMC migration [25]. As shown in Fig. 2a, VSMCs treated either with SPM or WPM signi cantly increased their expression of MMP2 and MMP9 proteins in a dose-dependent manner, with a greater increase observed in WPM-treated VSMCs than those treated with SPM, which was consistent with the results obtained from the migratory activity experiment. However, the protein levels of MMP13, one of the effectors of cell migration, were not changed, which demonstrates that ambient PM induced VSMC migration through MMP2 and MMP9. As expected, in VSMCs treated with ambient PM, we observed a signi cant increase in the mRNA levels of MMP2 and MMP9, but also in the mRNA levels of MMP13, which is inconsistent with the lack of change in its protein levels (Fig. 2b). In addition, the extracellular proteolytic activity of MMP2 and the transcriptional levels of MMP9 were signi cantly increased in ambient PM-treated VSMCs, and these changes were blocked by NAC treatment (Fig. 2c and 2d). We also investigated the effects of ambient PM on the phosphorylation levels of focal adhesion kinase (FAK) and FAK-related signals since FAK acts as a regulator of MMP expression [26,27] and also enhances cell migratory activity itself through actin reorganization [28]. The phosphorylation levels of FAK at the Y397 and Y925 sites, and of Src, a downstream target of FAK, were increased by ambient PM; however, NAC blocked phosphorylation of p-Src and p-FAK at Y925, but not at Y397. In addition, the phosphorylated levels of Akt and ERK were signi cantly increased in ambient PMtreated VSMCs; yet, NAC inhibited the increase of p-Akt without affecting the levels of p-ERK (Fig. 2e). These results demonstrate that ambient PM enhances not only MMP expression, but also FAK phosphorylation, leading to VSMC migration.

PAHs are mediators inducing ROS generation and VSMCs migration
We have previously suggested that PAHs and oxy-PAHs that are found in ambient PM might act as ROS generators [20]. Therefore, in this study, we hypothesized that PAHs, including oxy-PAHs, comprise the main components of ambient PM that induce ROS generation, which subsequently promotes the migration of VSMCs. We investigated the effects of two types of PAHs, anthracene (ANT) and benz(a)anthracene (BaA), and their oxygenated derivatives, 9,10-anthraquinone (AQ) and 7,12benz(a)anthraquinone (BAQ), respectively, on ROS generation and VSMC migration. We rst demonstrated that PAHs or oxy-PAHs used at concentrations of 5 and 10 µM had no signi cant cytotoxicity in VSMC (data not shown). As shown in Fig. 3a, VSMCs treated with PAH or oxy-PAH signi cantly increased their ROS generation in a dose-dependent manner, compared to that of the untreated control. As expected, we observed that the alterations were signi cantly greater in VSMCs treated with oxy-PAHs than in those treated with PAHs. In addition, VSMC migration was increased by PAH and oxy-PAH treatments in both wound healing and Boyden chamber assays; however, the only signi cant difference observed between PAHs and oxy-PAHs was between VSMCs treated with BaA and those treated with BAQ, and only at a concentration of 10 mM (Fig. 3b and 3c). Interestingly, cell proliferation was signi cantly increased in AQ-, BaA-, and BAQ-treated VSMCs, but not in ANT-treated cells (Fig. 3d). Thus, we have shown that PAHs enhanced VSMC migration.

PAHs regulates matrix metalloproteinase expression and activity through ROS generation
To investigate whether the increased migration by PAHs is due to the altered MMP expression induced by ROS, we evaluated the effects of PAHs on MMP expression. We observed that following treatment with all four types of PAHs, protein expression of MMP2 and MMP9 were signi cantly increased compared to that of the control (Fig. 4a), with the resulting levels of MMP9 protein being signi cantly higher than that of MMP2. The only signi cant difference in MMP9 protein expression between PAHs and oxy-PAHs was when comparing VSMCs treated with ANT to those treated with AQ. The increase in MMP2 and MMP9 were successfully inhibited by NAC treatment (Fig. 4b). Compared to the control, the mRNA levels of MMP2, MMP9, and MMP13 were all signi cantly increased in BAQ-treated VSMCs, while only MMP9 and MMP13 mRNA were increased in BaA-treated cells. All observed increases in mRNA expression were inhibited by NAC treatment (Fig. 4c). As shown in Fig. 4d, the transcriptional activities of MMP2 and MMP9 were signi cantly increased in VSMCs treated with all four types of PAHs, and consistent with the results of protein expression, the resulting levels of MMP9 were higher than those of MMP2. In addition, these increases were signi cantly inhibited by NAC treatment. Moreover, we observed that the increase of transcriptional activity of these MMPs by PAHs was also signi cantly inhibited by a-naphtho avone (a-NF), a speci c inhibitor of AhR. The aryl hydrocarbon receptor (AhR) is a ligand-activated transcription factor that regulates biological responses to planar aromatic hydrocarbons and acts primarily as a sensor of xenobiotic chemicals [29,30]. Consistent with these ndings, the VSMC migration induced by PAHs was also successfully blocked by NAC or a-NF treatment (Fig. 4e). These results demonstrate that PAHs enhance VSMC migration through ROS generation.
PAHs regulate the expression of ROS-related genes and the dependent signaling pathways Since there is evidence that CYP1A1 and transcription factor Nrf2 are protective against PAHs and ambient PM-induced pro-oxidative damage in various cell types [31,32], we investigated their alterations in PAH-treated VSMCs. The mRNA levels of CYP1A1 were dramatically increased in BaA-and BAQ-treated VSMCs in a time-dependent manner, and these alterations were only partially, but signi cantly, inhibited by NAC treatment (Fig. 5a). In addition, PAH treatment caused translocation of Nrf2 into the nucleus, and this effect was inhibited by NAC (Fig. 5b). The mRNA levels of Nrf2 and its target genes, heme oxygenase-1 (HO-1) and NAD(P)H dehydrogenase 1 (NQO1), were signi cantly increased following treatment with both forms of PAHs (Fig. 5c). The protein expression level of NQO1 was also increased by PAH treatment (Fig. 5d). Studies have shown that HO-1 and NQO1 are expressed in an ROS-dependent manner [33], and consistent with the results of these studies, NAC inhibited the increases in mRNA and protein levels. In addition, the phosphorylation level of Src, a downstream target of Nrf2, was signi cantly increased by PAHs (Fig. 5d). These results demonstrate that the cellular uptake of PAHs generates ROS and regulates the ROS-dependent signaling pathways in VSMCs.
PAHs enhance the formation of focal adhesion complex through aryl hydrocarbon receptor The aryl hydrocarbon receptor (AhR) is the primary sensor of aromatic hydrocarbons and is involved with the regulation of focal adhesion (FA) sites [34]. AhR activation subsequently activates Src and FAK, leading to the reorganization of the actin cytoskeleton, which is an effector for the changes observed in cell migration [27]. We rst observed that in VSMCs treated with either the PAH BaA or the oxy-PAH BAQ, AhR translocated into the nucleus from the membrane in a time-dependent manner. AhR translocation was signi cantly higher in BAQ-treated VSMCs than in BaA-treated VSMCs when comparing the observations at the same time points (Fig. 6a). Further co-immunostaining showed that compared to that of controls, the degree of colocalization of FAK and paxillin at the sites of lopodia growth were higher in both forms of PAH-treated VSMCs. Furthermore, BAQ-treated VSMCs displayed more extensive colocalization with paxillin than in BaA-treated cells (Fig. 6b). In addition, we could observe the more spots of integrin b1 on edge of actin cytoskeleton. Blocking of AhR by a-NF signi cantly reduced the level of these colocalization, suggesting that PAHs affect FA formation and cytoskeletal organization in VSMCs through this receptor. In addition, we con rmed that the levels of phosphorylated FAK and FArelated proteins, such as p-Src and paxillin, were signi cantly increased in VSMCs treated by both forms of PAH (Fig. 6c). As expected, PAH treatment of VSMCs signi cantly increased the levels of p-Akt and signi cantly altered the levels of phosphorylated p38 (Fig. 6c). These alterations were successfully blocked by treatment with a-NF (Fig. 6d). Furthermore, we examined the mRNA levels of integrins aV, b1, b3, a1, and a5, which are the components of FA complexes in the lopodia. With the exception of a5 in VSMCs treated with BaA, all integrin mRNA levels were signi cantly increased compared to controls in PAH-treated cells. All of these increases were blocked by treatment with a-NF, except for aV in VSMCs treated with BaA ( Fig. 6e). Our results indicate that PAH uptake by AhR regulates FA-related biological responses in VSMCs.

Discussion
Despite the culmination of evidence suggesting that there is an association between ambient air particles and vascular dysfunction, the underlying mechanisms are complex and variable, and remain to be elucidated. Moreover, due to the complexity of PM components, the speci c components responsible for PM-induced vascular dysfunction should be investigated.
We have previously reported that ambient PM generates a substantial amount of ROS, which may induce electrophysiological instability in cardiomyocytes [20]. In addition, we revealed that this effect might be caused by PAHs, the main component of PM. The present study demonstrated that PM exposure signi cantly increased ROS generation in VSMCs, leading to enhanced migratory activity. The signi cant increase in migration was greater in VSMCs treated with WPM than in those treated with SPM, and the higher ROS generation by WPM was evidenced by the higher levels of oxy-PAHs contained in WPM than in SPM, consistent with the results of our previous report. ROS generation by BAQ, one of the oxy-PAHs tested in this study, was signi cantly higher than that of its non-oxygenated form, BaA. Furthermore, we observed that VSMC migration was positively correlated with ROS levels. However, ANT and AQ, the other set of PAH tested, did not show any signi cant difference in ROS generation and cell migration when compared to that of the BaA and BAQ pair, which might be due to their different chemical natures and experimental concentrations.
Although VSMC proliferation is also one of the key factors triggering vascular pathology, we did not observe any signi cant alterations in VSMC proliferation by ambient PM. However, the signi cant increase in VSMC proliferation following PAHs treatment (except for ANT) tested in this study suggests that other components in ambient PM might participate in regulating the pathways that lead to VSMC proliferation. Another plausible explanation for this discrepancy is that the concentration of PAHs tested in this study was slightly higher than the concentration of mobilized PAHs from PM.
Cell migration requires the ne spatial-temporal integration of many proteins that regulate the fundamental processes responsible for driving cell movement [35]. It includes the various MMPs that degrade extracellular matrix components and cytoskeleton-related proteins that enhance cell protrusion. Furthermore, there is a substantial amount of evidence to suggest that redox stress is associated with cardiovascular pathologies, such as neointima hyperplasia during restenosis [36], angiotensin II-induced hypertension [37], and impaired endothelium-dependent vasorelaxation [38]. Although the mitochondrial respiratory chain is the major source of intracellular ROS in animal cells, VSMCs contain various sources of ROS, including xanthine oxidase, lipoxygenase, nitric oxide synthases, NADPH oxidases, and heme oxygenases [39]. In addition, ROS can impact the signaling of a variety of molecular targets that participate in VSMC migration, including target proteins associated with extracellular matrix degradation [40], the formation of FA complexes [41], and cytoskeleton dynamics [42].
The ndings of this study show that ROS generation by ambient PM or PAHs is positively correlated with MMP expression. In addition, the phosphorylated levels of FAK, as well as the levels of FA complexrelated proteins, such as paxillin and p-Src, were signi cantly increased. The successful inhibition of these alterations by the ROS scavenger, NAC, demonstrates that ambient PM mediates VSMC migration through ROS generation and that PAHs contained in ambient PM are responsible for these responses. Moreover, our results showed that the mRNA and protein expression levels of MMP2 and MMP9 were signi cantly increased in ambient PM or PAH-treated VSMCs, which is consistent with previous reports that ROS can directly or indirectly activate MMP2 and MMP9 in VSMCs [40,43]. However, although we demonstrated that MMP13 participates in VSMC migration in our previous report [44], and mitochondrial ROS generation mediates MMP13 expression in unmyelinated axons [45], we did not observe signi cant alterations in protein levels of MMP13 in ambient PM-or PAH-treated VSMCs in this study. This discrepancy might be due to insu cient ROS generation or different cell speci cities. Therefore, in future studies, it will be interesting to further examine the effects of ROS on MMP13 expression and VSMC migration.
Our results also demonstrated that ambient PM or PAHs increased the phosphorylation of FAK and FArelated proteins. Indeed, FA dynamics (assembly and disassembly) involving the coordination between the FA and the actin cytoskeleton is a required process for cell migration [46]. Furthermore, integrins mediate the dynamic interactions between the extracellular matrix and the actin cytoskeleton during cell movement [47]. During this process, the interaction between integrins a5b1 and aVb3, and FAK acts as a key mediator for VSMC migration [48]. Moreover, there is convincing evidence that FA and actin cytoskeleton dynamics are directly regulated by ROS [49,50]. In agreement with the above reports, our results showed that the binding between actin and integrin b1 in the FA complex was increased in the lapodia region in PAH-treated VSMCs. In addition, the formation of FA complex was also increased in PAH-treated VSMCs.
Furthermore, we observed that PAH treatment led to signi cant activation and nuclear translocation of the transcription factor Nrf2 and subsequently increased the expression of the target proteins of Nrf2, HO-1, and NQO1. The mRNA levels of CYP1A1 were also signi cantly increased in PAH-treated VSMCs. Our results concur with a previous report that showed that exposure to diesel exhaust particulate matter led to signi cant activation of Nrf2 and expression of CYP1A1 in endothelial cells [31]. However, the correlation between ROS levels and these alterations was not evaluated in this previous study. Although Nrf2 activation and the upregulation of CYP1A1 are protective against ROS insult, the signi cant increase in VSMC migration in our study suggests that the level of ROS generation following PM or PAH treatment may be beyond their protective capacity.
The present study has some methodological limitations. First, one of the drawbacks is that PAHs are not the only components responsible for ROS generation in PM. Cationic metals, including Fe, Zn, Mn, and Cu, contained in ambient PM are known to produce ROS in biological systems. In particular, there is convincing evidence to suggest that Fe generates ROS through the Haber-Weiss and Fenton reaction [51]. Indeed, we observed that ambient PM contained these metallic compounds in our previous study [20]. Therefore, further investigation of the association between metals and PAHs that lead to ROSmediated VSMC migration is needed. Second, as this was an in vitro study, the consequences of ambient PM or PAH treatment reported here may not manifest in humans following real-world inhalation. In addition, because vascular pathology is mediated by complicated interactions between endothelial cells and VSMCs, the effects of PM or PAHs on endothelial cells need to be evaluated. Despite the limitations of this study, our results on the effects of ambient PM or PAHs on VSMC migration may still be relevant when examining mammalian vascular pathophysiology.

Conclusions
Our results provide strong evidence that ambient PM increases VSMC migration by ROS generation, and PAHs are the major contributors to this effect. We demonstrated that ROS generated by PM or PAHs affects VSMC migration in two ways. First is through the signi cant increase of MMPs, which causes the degradation of extracellular matrix, and second is the formation of the FA complex that allows for actin cytoskeleton rearrangement. Although our ndings of vascular pathology caused by ambient PM is supported by an increasing amount of clinical evidence, the in vivo and clinical relevance of these ndings remain to be elucidated.

Ambient particulate matters and preparation of organic components
Collection of ambient particulate matter (PM) at Seoul metropolitan area during summer (SPM) and winter (WPM) season and particle preparation was described in the previous study [52], which showed the detailed process for the sampling of PM10 and extraction of organic matter from the PM. The analyzed organic compounds including PAHs and oxy-PAHs and their concentrations were presented in the previous study [20].

Isolation of VSMCs
All animal experiments for the isolation of rat aortic VSMCs were conducted in accordance with the International Guide for the Care and Use of Laboratory Animals. The protocol was approved by the Animal Research Committee of the Chosun University School of Medicine (Protocol No. CIACUC2020-S0032). Rat aortic VSMCs were isolated from 6-week-old Sprague-Dawley rats as described previously [53]. Brie y, the removed aorta was freed from connective tissues and blood clots, then severed and transferred into a tube containing a mixture of collagenase type I (1 mg/mL, Sigma, St. Louis, MO, US) and elastase (0.5 mg/mL, Worthington, NJ, US) and incubated for 30 min at 37°C. The aorta was placed into a 100-mm cell culture dish and the adventitia was stripped with forceps under a binocular microscope. Each piece of the aorta was transferred into a tube containing 5 mL of enzyme dissociation mixture (containing collagenase and elastase), and the tubes were incubated for 2 h at 37°C. The dispersion of the tissue was accomplished by slow pipetting. The suspension was centrifuged (1600 ×g for 5 min), and the obtained pellet was resuspended in DMEM with 10% fetal bovine serum (FBS, WelGENE, KOREA). This dissociation step was repeated until tissues were completely dispersed. The cells were cultured at 37°C in an incubator with a humidi ed atmosphere of 95% air and 5% CO 2 and used for up to 10 passages in this study.

Measurement of cytotoxicity and cell proliferation
Cytotoxicity was determined by MTT assay using a CellTiter 96 Assay kit (Promega) according to the manufacturer's instructions. Brie y, VSMCs were seeded into a 96-well plate in triplicate and cultured for 12 h, and then cells were treated with PAHs or oxy-PAHs for 24 h. The absorbance was measured at 490 nm using an ELISA reader (TECAN, in nite M200 PRO). The proliferative rates of VSMCs were evaluated by determining BrdU incorporation using the BrdU Cell Proliferation Assay kit (Cell signaling). VSMCs were treated with seasonal ambient PM, PDGF-BB (20 ng/ml), PAHs, or oxy-PAHs for 24 h. Then cells were incubated with BrdU for 4 h and subsequent procedure was performed according to the manufacturer's instructions. The absorbance was measured at 450 nm using an ELISA reader (TECAN, in nite M200 PRO).

Cell migration assay
Cell migration was examined by three-dimensional Boyden chamber assay and two-dimensional wound healing assay. For Boyden chamber assays, cells (5×10 4 cells in 100 μl) were placed in the upper compartment of the transwell chambers coated with collagen I on the lower surface. Ambient PM or PAHs were treated with the designated concentration in each upper compartment with designated concentration. After incubation for 16 h at 37°C, cells on the lower surface of the lter were xed and stained, and ve random elds/membranes were counted at ×200 magni cations. For wound healing assay, a rectangular lesion was created using a cell scraper, and then the cells were incubated with a designated concentration of ambient PM or PAHs for designated times. The distance from the margin of the lesion to the most migrated cells was measured, and the mean value of the distances was taken as the mobility of cells in each culture dish. Then, cells were gently washed the coverslips three times in warm PBS and the coverslips were placed in the chamber, which was mounted on the stage of an inverted microscope equipped with a confocal laser-scanning system. The dye, when exposed to an excitation wavelength of 480 nm, emitted light at 535 nm only when it had been oxidized. Fluorescence images were collected using a confocal microscope (Fluoview FV1000 confocal system, Olympus) by excitation at 488 nm and emission greater than 500 nm with a long-pass barrier lter. The uorescence intensity of an equivalent eld size (3×3 mm) in the plate was measured using the Image J quanti cation software.

Zymography
Brie y, aliquots of the control and test media were electrophoresed on a 10% SDS-polyacrylamide gel containing 0.8% gelatin or 0.3 mg/ml collagen. Gels were washed with 2.5% Triton X-100 to remove SDS for 1 h, washed with D.W for 1hr and then incubated at 37°C for 48 h in developing buffer (gelatin incubation buffer : 50 mM Tris-HCL, pH 7.5, 5 mM CaCl 2 , 1 mM ZnCl 2 , 0.02% sodium azide, 1% Triton X-100 or collagen incubation buffer : 50 mM Tris-HCL, pH 7.5, 10 mM CaCl 2 , 50 mM NaCl, 0.05% Brij35, pH 7.6). After 48 h, the gel was stained with 1% Coomassie blue for 1 h and later, then destained until there is a good resolution between the bands and blue background. The gel was photographed and zymolytic area was quanti ed using Image J software (NIH, Bethesda, MD, US).

MMPs promoter luciferase activity assay
For the luciferase assay, the reporter constructs of rat MMP2 (GenBank: DQ915967.1) and rat MMP9 (GenBank: AF148065.1) promoter were cloned into the pGL3 basic vector (Promega). VSMCs were transiently transfected with a total of 100 ng each of the luciferase reporter constructs using Lipofectamine 3000 (Invitrogen). To ensure e cient transfection, all wells were also co-transfected with a Renilla luciferase vector (pRL-TK; Promega). After 24h transfection, the cells were treated with either negative control (DMSO), SPM, WPM, PAHs, or oxy-PAHs for 12h. Luciferase activities were determined using the Dual-Luciferase Reporter Assay System (Promega) as described in the manufacturer's protocol. Data were normalized by internal Renilla luciferase activity.
Quantitative real-time PCR (qRT-PCR) The expression levels of the various genes were analyzed by qRT-PCR. Cells were seeded into a 6well plate with glass coverslips at a density of 5×10 5 cells/ml and cultured for 24 h. Cells were treated with negative control (DMSO), ambient PM, or PAHs for 12 h. Total RNA was extracted using TRIzol lysis reagent (QIAGEN) according to the instructions provided by the manufacturer. The RNA concentration of each sample was measured by a spectrophotometer (Eppendorf) at 260 nm. Total RNA was subjected to reverse transcription using HelixCript™ 1st-Strand cDNA Synthesis Kit (NanoHelix). Real-time quantitative PCR with realHelix™ qPCR kit (NanoHelix) was performed by the SYBR Green method using an Applied Rotor-Gene 3000 TM . Gene expression was normalized to GAPDH. The relative mRNA expression levels were quanti ed and analyzed using Rotor-Gene 6 software (Corbett-research) using △△ Ct methods. Table   1 was primer sequences for qPCR.

Immunoblot analysis
VSMCs were seeded into a 6well plate at a density of 5×10 5  Data are expressed as mean ± SD. Statistical comparisons between the two groups were performed using Student's t-test. Statistical comparisons among multiple groups were performed using one-way ANOVA followed by Bonferroni post hoc test when the F statistic was signi cant. A two-tailed P < 0.05 was considered statistically signi cant.

Declarations
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.
Availability of data and materials The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.    The mRNA levels of MMPs were quanti ed by qPCR. Gene expression was normalized to GAPDH. (c) The Page 22/28 zymolytic activity of MMP2 was evaluated using gelatin zymography. Serum-starved VSMCs without or with NAC were treated with the designated concentrations of SPM or WPM for 24 h. PDGF treatment alone was used as a positive control. The media were then collected and used for this assay. The images shown are representative of those obtained from at least three independent experiments. (d) The effect of NAC on transcriptional activity of the MMP9 promoter in VSMCs was evaluated using a luciferase assay.
(e) The expression levels of phosphorylated ERK1/2, Akt, FAK, and Src were analyzed by western blotting. Protein levels were normalized to their total levels. All values are represented as mean ± SD. *P < 0.05, **P < 0.01, and ***P < 0.001 versus control; #P < 0.05 WPM versus SPM; $P < 0.05 and $$$P < 0.001 PM with NAC versus PM alone; NS, no signi cance.  HO-1, and NQO1 in the absence or presence of NAC were evaluated by qPCR. (d) The expression levels of NQO1 and phosphorylated Src were analyzed by western blotting. Protein levels were normalized toactin. All values are represented as mean ± SD. ***P < 0.001 versus control; $P < 0.05, $P < 0.001 PAHs with NAC versus PAHs alone.