Benzo[ghi]perylene Induces Morphological and Genotoxic Effects in Human Bronchial Cells Independently of The AhR Pathway

Polycyclic aromatic hydrocarbons (PAH) are organic compounds found in the contaminated atmosphere of Mexico City, where the PAH present with highest concentration is benzo[ghi]perylene [B[ghi]p]. We recently demonstrated that double-stranded breaks in DNA appear after 3h of exposure, whereas cellular changes and activation of aryl hydrocarbon receptor (AHR) pathway occur after 48h in bronchial cells under exposure to B[ghi]p, these ndings have led us to explore if the AHR pathway participates in morphological changes and genotoxic effects due to B[ghi]p in human NL-20 bronchial cells. Cells of the NL-20 human bronchial line were exposed to B[ghi]p in the presence, or absence, of the AHR receptor antagonist, CH-223191. Cell viability was quantied by the MTT assay, which revealed 76 and 66% at 6h and 24h, respectively (p<0.001), regardless of the presence of CH-223191. RT-qPCR showed an increase in the expression of the AHR and CYP1A1 cytochrome genes only after 24h of exposure, and the expression was inhibited by CH-223191. Genotoxicity assays revealed the presence of comets, nuclear buds (NB) and DNA fragmentation in cells exposed to B[ghi]p after 6h and in cells exposed to B[ghi]p plus CH-223191 at 24h. These results verify that B[ghi]p induces morphological and genotoxic effects, and these effects are independent of the AHR pathway.

B [ghi]p is a PAH of the type pericondensed. Its structure contains 6 rings and its molecular weight is 276, but it has no bay or gulf regions (Marr et al. 1999). In 1983, the IARC classi ed B[ghi]p as noncarcinogenic in animals and humans, and the most recent evaluation (2010) established that B[ghi]p does not produce cancer by cutaneous application in mouse models; a nding supported by other studies that used the cutaneous injection technique (Müller 1968). Intrapulmonary application in a rat model failed to produce conclusive ndings, since some tumors appeared, but the results were not reproducible (Deutsch-Wenzel et al. 1983, IARC. 1983). In vitro studies reported that B[ghi]p is mutagenic in the Salmonella typhimurium assay in presence of an exogenous metabolic system (rat fraction S9) (Platt and Grupe. 2005). Also, B[ghi]p correlated with dithiothreitol activity in an in vitro system designed to quantify the formation of reactive oxygen species (ROS) (Cho et al. 2005). In summary, few in vitro and in vivo studies have been conducted to evaluate the toxic properties of B[ghi]p. Recent work performed by our research group with cells of the human NL-20 bronchial line, showed that B[ghi]p after 48h of exposure produced morphological changes in cells, consisting in the formation of small vesicles throughout the cytoplasm. Also, B[ghi]p showed these cellular effects: increased damage to the cell membrane, translocation of the aryl hydrocarbon receptor (AHR) to the nucleus, and the expression of genes that included the AHR receptor and cytochrome CYP4B1. Except for damage to the cell membrane and vesicle formation, these effects were reversed in presence of the CH-223191 receptor antagonist which, in turn, increased oxidative stress, mainly of superoxide ( AHR is a ligand-dependent transcription factor that belongs to the basic helix-hoop-helix family (bHLH) Per/Arnt/Sim (PAS) and binds to a wide variety of exogenous compounds, including benzo[a]pyrene (B[a]p) and 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) (Opitz et al. 2011). AHR is found predominantly in the cytoplasm in a multiprotein complex that includes two hsp90 chaperone proteins (90-kDa heat shock proteins) that help maintain the conformation of the receptor, an AIP (auxiliary interacting protein) that stabilizes the interaction between the hsp90 and the receptor, and a p23 co-chaperone protein that stabilizes the intermediary ligand-AHR complex. In the presence of a ligand, the multiprotein complex becomes uncoupled and the AHR-PAH translocates to the nucleus, where it forms a heterodimer with ARNT (aryl hydrocarbon receptor nuclear translocator) (Larigot 2018, Opitz et al. 2011). The AHR-ARNT complex then binds to XRE (xenobiotic response elements) to induce the transcription of target genes that codify enzymes that participate in activation and detoxi cation; namely, phase I and phase II enzymes, such as CYPs and aldo-ketoreductases, AKR (among others), respectively, and proteins and enzymes that participate in the activation of cell cycle and cell adhesion, migration and differentiation.
Once released from its ligand, AHR is degraded by the proteosome pathway in the nucleus (Roberts and Whitelaw 1999). As mentioned above, our group has described that 3h of exposure to B[ghi]p in human bronchial cell line NL-20, can induce H2AX foci in the nuclei, and the activation of AHR after 48h of exposure. These ndings led us to design the present study, in which we evaluated if morphological and genotoxic effects due to B[ghi]p in human NL-20 bronchial cells, depend on AHR pathway.

Cultures of NL-20 cells
The NL-20 cell line was obtained from the ATCC (CRL-2503). These are human epithelial bronchial cells immortalized from normal (i.e., non-tumorous) bronchi. Cells were cultured in HAM-F12 culture medium supplemented with 4% fetal bovine serum, 1% L-glutamine, 1% non-essential amino acids, 5 µg/ml of insulin, 10 ng/ml of epidermal growth factor, 1 µg/ml of transferrin, 500 ng/ml of hydrocortisone, 100 µg/ml of streptomycin sulfate, and 100 µg/ml of Penicillin-G, under the following conditions: 37ºC, 5% CO 2 and relative humidity. They were detached in a solution of 0.05% trypsin and 0.01% versene for 10 min at 37°C and were centrifuged at 1800 rpm for 5 min, and then counted in a hemocytometer before performing the following exposure assays to B[ghi]p.

Exposure to B[ghi]p
Previously, we performed assays at different times to exposure NL-20 cells to B[ghi]p and after did them, we decided to continue the study at 6h and 24h of exposure to B[ghi]p. The NL-20 cells were exposed Total RNA extraction Cells from the experimental scheme were lysed by adding 100 µL of Trizol per well and stored at -70°C until the samples from three biological triplicates were collected. In the next step of the extraction procedure, the cellular lysate was incubated for 5 min at room temperature, 60 µL of cold chloroform were added, and the mixture was agitated in a vortex and left to ambient temperature for 10 min. It was then centrifuged at 12,000 rpm for 15 min at 4°C and the aqueous phase containing the RNA was recovered.
This was transferred to a microtube with 150 µL of cold isopropanol and 1 µL of glycogen and incubated overnight at -20°C to precipitate the largest quantity of total RNA possible. This was centrifuged at 12,000 rpm for 10 min at 4°C. The supernatant was removed, and the RNA was washed twice with 300 µl of ethanol 70%, mixed gently and then centrifuged at 10,000 rpm for 5 min at 4°C. Finally, the alcohol was decanted, and the RNA was left to dry, afterwards it was re-suspended in 10 µL water-RNases free. It was treated with DNase I (1U per 1µg of RNA) for 30 min. The enzyme was inactivated with EDTA at 70°C for 10 min and the concentration and purity of the RNA was quanti ed in a NanoDrop. All samples had a 260/280 relation above 1.7. cDNA synthesis cDNA synthesis was performed with 200 ng of total RNA using the First Strand cDNA Synthesis kit (Thermo Scienti c) in a thermocycler under the following conditions: at 37°C for 60 min, at 70°C for 10 min, and at 22°C for 1 min.
Expression of the AHR, CYP1A1 and HOX-1 genes by RT-qPCR Gene expression was quanti ed in real-time with SYBR green as the principle uorochrome and ROX as the passive reference colorant, as speci ed in the Maxima SYBER Green qPCR Master Mix Kit (Thermo Scienti c). The speci c primers for each gene were used (Table 1) and the reactions were carried out in a Stratagene Mx3005P thermocycler (Agilent) using the following program: at 95°C for 10 min, 40 cycles at 95°C for 15 seconds at the melting temperature of each primer for 1 min, and nalizing with one cycle at 95°C for 1 min, at 58°C for 30 sec, and at 95°C for 30 sec. Expression levels were analyzed by applying the 2 −ΔΔCt mathematical model (Pfa 2001;Livak and Schmittgen 2001) with GAPDH as the endogenous gene. Table 1 Forward (F) and Reverse (R) sequences of the GAPDH, AHR, CYP1A1 and HOX-1 genes showing size in base pairs (pb) of each ampli ed product and the optimal TM for RT-qPCR.

Genes
Oligonucleotides sequences Amplicon, size, bp MT Expression of the AHR protein by immunocytochemistry

MTT viability assay
Following the protocol by Tolosa et al. (2015) for this assay, 15,000 cells were cultured per well in 96-well plates 24h before the experiment. Exposure times were 6h and 24h. Cells were exposed to the experimental scheme conditions in triplicate, but 3 hours before ending each time, 20 µL of MTT at a concentration of 5 mg/mL (in PBS) were added to each well. At the end of the exposure times, the medium was removed carefully before adding 100 µL of acidic isopropanol to each well for 5 min under darkness. The amount of color was quanti ed in an ELISA plate at a wavelength of 540 nm as principal absorbance, and a wavelength of 620 nm to measure background absorbance. The percentage of reduced MTT was calculated using the absorbance obtained from the unexposed control-vehicle that was considered 100%.

Cell morphology
Cells were cultured and were left to adhere on sterile coverslips before being exposed to the experimental scheme. At the end of the exposure times, they were xed in absolute methanol for 5 min and left to dry before staining with 500 µL of Giemsa for 10 min. Next, they were washed to remove the excess colorant and left to dry. Finally, they were mounted with resin on slides and observed under an optical microscope at 100x. Photomicrographs were registered for any experimental scheme to reveal morphological changes.

Nuclear electrophoresis analysis: COMET assay and nuclear buds
At the end of the experimental scheme, cells were detached using trypsin and saline solution (1:9) in an incubator at 37°C for 15 min. To concentrate, cells were centrifuged at 1800 rpm for 5 min, the supernatant was discarded, and cells were re-suspended in 100 µL of complemented HAM-F12 culture medium. Next, 50 µL were extracted and mixed with 90 µL of low-fusion point agarose at 0.5% at 38°C.
Prior to this, agarose gels at 1% had been prepared in PBS on slides to empty the second layer of agarose with the cells, which was left to gel at 4°C for 7 min. To obtain nuclei, the gels were left in lysis buffer (2.5 M NaCl, 100 mM EDTA, 10 mM Tris, Triton 1% X-100 and DMSO 10%, pH 10) for 24h at 4°C.
In darkness, the gels were placed in a horizontal electrophoresis chamber with an alkaline buffer at 4°C (300 mMNaOH, 1 mM EDTA pH 13) for 20 min to allow the DNA to unroll, and then for another 20 min at 25 V to starts the migration of the DNA fragments towards the anode. At that point, the gels were neutralized by 3 washings in neutralizing buffer (0.4 M Tris, pH 7.5) for 5 min each. The gels were then xed in absolute ethanol at 4°C for 10 min and stained with 50 µL of Gel Red (10%) for analysis under uorescence microscopy. The parameters considered to measure the damage in 100 nuclei per treatment were nuclei with comets and buds, expressed as the percentage of comets and of nuclei with buds.

Detection of DNA fragmentation induced by B[ghi]p with the TUNEL assay
To conduct this assay, we used the DeadEnd Colorimetric TUNEL System kit (Promega), exposing 45,000 cells from each experimental scheme to B[ghi]p. After removing the culture medium, the cells were washed in 1 mL of saline solution to separate them with trypsin for 10 min. To concentrate cells, they were centrifuged at 1800 rpm for 5 min. The solution was decanted and 100 µL of saline solution were added for a wash with centrifugation at 1800 rpm for 5 min. The solution was decanted again, and cells were re-suspended in 100 µL of saline solution. On previously prepared slides coated with silane 0.1%, smears were taken for each experimental scheme and left to dry at 37°C. These samples were then xed with paraformaldehyde at 4% for 25 min and the presence of DNA fragmentation was revealed, following the instructions of the kit. The slides from three independent experiments were evaluated in an optical microscope, and the nuclei with DNA fragmentation were quanti ed in 100 cells and expressed as percentages of nuclei with DNA fragmentation.

Statistical analysis
The differences in the levels of expression of AHR, CYP1A1 and HOX-1, and the viability test were evaluated by ANOVA analysis. The presence of buds, the number of nuclei with DNA fragmentation, as determined by TUNEL, and the number of comets induced, were all evaluated by Ji-squared test. Differences were considered signi cant when the p value was < 0.05.

Results
The effect of B[ghi]p on cell viability is independent of the activation of the AHR pathway The viability of cells exposed to B[ghi]p decreased (p < 0.001) to 76% and 66% at 6h and 24h of exposure, respectively (Fig. 1A). Both percentages of viability were above of the 50% inhibitory concentration (CI 50 ).
The effects of the reduction of viability due to exposure to B[ghi]p were not reversed by the AHR antagonist CH-223191 (Fig. 1A), this indicates that the reduction of the viability of the cells exposed to B[ghi]p after 6h and 24h of exposure in the presence of the CH-223191 antagonist occurs independently of AHR pathway activation.
The expression of AHR pathway is veri ed at 24 of exposure of B[ghi]p The transcriptional activation of the AHR pathway was evaluated. B[ghi]p induces a signi cant increase (p < 0.05) in the expression of the AHR (Fig. 1B) and CYP1A1 (Fig. 1C) genes, but only after 24h of exposure, compared to control vehicle (DMSO). However, the high expression of AHR and CYP1A1 after 24h of exposure returned to basal levels when cells were exposed to B[ghi]p plus the AHR antagonist CH-223191 (Figs. 1B and 1C, respectively) Figure 1D shows an increase of expression of this gene (p < 0.05), in 6h and 24h of exposure, which indirectly indicates the induction of oxidative stress.

Translocation of the AHR receptor after B[ghi]p exposure
Immunocytochemistry illustrates that, after 6h of exposure to B[ghi]p, there is a weak signal of protein ( Figs. 2H and 2K), and it remains mainly in cytoplasm (Fig. 3I), also in presence of the antagonist CH-223191 (Figs. 2I and 2L). At 24h of exposure to B[ghi]p, the AHR protein is translocated mainly to the nucleus (Figs. 2T and 2W, cyan color in merge, Fig. 2W) and in presence of AHR antagonist CH223191, the protein is weakly detected (Figs. 2U and 2X). These results verify B[ghi]p activates the pathway because the protein is translocated to the nucleus

Morphological changes in cells exposed to B[ghi]p
To characterize the morphological changes produced by B[ghi]p, at 6h and 24h, the analysis of photomicrographs shows the presence of intracytoplasmic vesicles (IVs) in NL-20 cells exposed to both B[ghi]p (Figs. 3C and 3G) and in B[ghi]p plus antagonist CH-223191 (Figs. 3D and 3H). At 6h, there is not AHR gene expression (Fig. 1B) and cells exposed to B[ghi]p alone have IVs (Fig. 3C): it proves this morphological modi cation is independent of any interaction of B[ghi]p to any receptor. The presence of IVs also remains after 24 of exposure to B[ghi]p (Fig. 3G), even in presence of the antagonist CH-223191 (Fig. 3H). The IVs are formed independent of the AHR pathway.

Genotoxic effects due to B[ghi]p
To evaluate the genotoxic effects that B[ghi]p might have on the nuclei of bronchial cells, a COMET assay, quanti cation of nuclear buds (NB), and DNA fragmentation determined by the TUNEL assay were performed. COMET analysis showed low, non-signi cant differences in formation of comets induced by B[ghi]p between 6h and 24h of exposure, even in presence of the antagonist CH-223191 (p > 0.05) (Fig. 4A), so we did not do any other measurements (v.g. tail moment of the comet/width of the tail). Regardless, the number of comets induced were signi cantly higher (p < 0.05) in cells exposed to B[ghi]p compared to those observed in the negative control and medium (Fig. 4A). In contrast, there was a signi cant increase of NB in cells exposed to both conditions: B[ghi]p alone and in presence of the AHR antagonist CH-223191 at 6h and 24h (p < 0.05) (Fig. 4B). These results show that induction of comets and NB occurs independently of the AHR receptor pathway.
Finally, B[ghi]p induced DNA fragmentation in 70% of the exposed cells, compared to only 3% of controls (p < 0.05) at 6h of exposure. CH-223191 cannot lower the DNA fragmentation as it is observed in controls (Fig. 4C). However, at 24h, there was reduced percentage of cells with DNA fragmentation (18%). The presence of the CH-223191 (Fig. 4C), reduced cells with DNA fragmentation at 15%, but these differences were not signi cant (p > 0.05). This con rm that DNA fragmentation is not dependent of the AHR pathway.

Discussion
Cell viability and morphological changes Although the AHR pathway was not activated at 6h of exposure to B[ghi]p, surprising observations were performed and included the decrease in cell viability accompanied by morphological changes in the NL-20 bronchial cells. These results are similar with those of a study conducted with A549 pulmonary adenocarcinoma cells exposed to 2 µM of B[ghi]p for 14h, which reported 68% viability (Genies et al. 2016). Those ndings indicate that B[ghi]p is cytotoxic and, perhaps, that the integrity of the mitochondria of the exposed cells is compromised by a decrease in the amount of metabolically-active cells, accompanied by greater oxidative stress. The particulate material (PM) and fuel emissions associated with the PAH have been shown to interfere with the function of mitochondria in both electron transport and the transitory permeability of the aperture of the mitochondrial pore (Xia et al. 2007). The mitochondrial toxic effect of B[ghi]p is currently under study in our laboratory. In an effort to clarify this, the present study evaluated genotoxic damage and AHR pathway activation at 6h and 24h of exposure to B[ghi]p, at concentrations lower than those reported earlier by our group. The most signi cant result of this study was that the AHR pathway was not activated at 6h of exposure, although cytotoxic and genotoxic damage and morphological changes to the human NL-20 bronchial cells did occur.
In cells of the WB-T344 epithelial line, the complex mixtures derived from the urban particles did activate the AHR receptor, and that activation contributes to the carcinogenic properties of these mixtures (Andrysik et al. 2011). There is, however, controversy in this regard. Depending on cell type, activation of this pathway can produce different effects, including the fact that in LNCaP prostate carcinoma cells, AHR pathway activation plays an important role in the non-genotoxic effects of B[ghi]p, such as inhibition of the cell cycle (Hrubá et al. 2011). It is important to note that studies with other cell lines (eg. H4IIE hepatoma), have shown that this hydrocarbon is a weak AHR inducer compared to the agonist, TCDD, since 10 µM of B[ghi]p is required to achieve the same effect that is induced with only 100 pM of TCDD after 6h of exposure. However, the mutagenesis or genotoxicity processes and potential AHR activation are properties of the PAH that can occur independently (Machala et al. 2001), as has been shown using benzo[k] uoranthene, a potent AHR inducer, but weak genotoxic agent, that produces low levels of adducts in the DNA of alveolar epithelial cells of A549 adenocarcinoma (Líbalová et al. 2014).
The increase of RNAm of the CYP1A1 and AHR genes is a classic marker of the AHR pathway activation that re ects the activation of the metabolism of phase I xenobiotics (Pan et al. 2013). The present study, however, did not detect any increase in the transcription of the AHR and CYP1A1 genes at 6h of exposure of the NL-20 cells to B[ghi]p. This indicates that longer exposure times to B[ghi]p are required to activate the pathway, as was demonstrated at 24h of exposure, with the resulting transcriptional increase of AHR and CYP1A1.  2019), induces doublestrand breaks that were proved by the results of TUNEL assay in this study, that will be discussed below.
However, upon analyzing the nuclei after unicellular electrophoresis of the exposed cells, we observed a signi cant percentage of NB at 6h of exposure. Some researchers consider that these abnormalities of the nuclei are zones of ampli ed DNA that the cells removed during the S phase of the cell cycle (Fenech 2002). It is thought that almost all these NB originate from acentric, telomeric or interstitial fragments, perhaps representing DNA that is trapped in the nuclear membrane and remains there after nuclear division or, perhaps, excess DNA that is excluded from the nucleus but stays tightly-connected to it through a stem of nucleoplasmic material. These events have been associated with chromosomal instability (Lindberg et al. 2007, Fenech et al. 2011). Meanwhile, by marking all the chromosomes with uorescent probes and evaluating the genotoxic effect of zidovudine on bone marrow mesenchymal cells, Dutra et al. (2010) demonstrated that less than 50% of the NB contain positive signs for one or two chromosomes, although over 50% had no marking signal, and some even showed the presence of a centromere. This suggests that the NB may contain fragments of some chromosome, a hallmark of genetic instability and do not necessarily represent only ampli cation zones.
In addition, NB are sensitive biomarkers that indicate exposure to diverse compounds (Nersesyan. 2005). Examples of this are the PAH mixtures to which coke oven operators are exposed, which showed a 3.5fold increase in the amount of NB compared to a control group. This means that they can be used as indicators of chromosomal damage to biomonitor workers exposed to PAH, and that their frequency can be correlated with exposure time (Duan et al. 2009). While the mechanism of NB formation in this system is unknown, the fact that their appearance was observed after short exposure times during which, the AHR-mediated pathway was not activated, suggests that NB can function as early markers of exposure to property, such that no adducts are produced. This K region, which is present in other PAH as wellincluding B[ghi]p -preferentially generates deoxyguanosine (Rojas and Alexandrov 1986) and deoxyadenosine (Nair et al. 1991). In this regard, Desler et al. (2009) propose that the PAH with no bay or gulf regions are generators of oxidative stress, and that it is through this pathway that they cause oxidative damage in nitrogenous bases of DNA such as 8-oxo-dG. These PAH with K regions are very weak tumor initiators compared to the PAH that have bay or gulf regions (Berry et al. 1977

Conclusions
This work demonstrated that short-term exposure to B[ghi]p functions as a cytotoxic and genotoxic agent whose effect is independent of the AHR-mediated pathway. B[ghi]p can induce morphological changes in cells, reduces their viability, and cause comets, formation of NB and DNA fragmentation: markers of genotoxicity. This nding opens up a broad eld of study since B[ghi]p seems to affect cells in two distinct ways: independently of the AHR-pathway at short exposure times, and at 24h. This raises the following question: ¿what is the overall effect of prolonged exposure to B[ghi]p on the structure of the genome of bronchial cells? Research into these important issues is underway in our laboratory with the expectation that in the near future we will be able to provide some answers to these key questions.