AOH induces oxidative stress and DNA damage in ovarian cancer cells via modulation of GPER1 and HIF1α/PI3K/CLDNs signaling pathway

Abstract

Claudins (CLDNs) are small proteins that form tight junctions (TJs) in cells and play a key role in various biological processes in cells. Previous results have shown that CLDNs expression can be regulated by hormones and, therefore, can also be regulated by compounds that mimic their action. Such compounds belong to a large group called endocrine-disrupting chemicals (EDCs). Alternariol (AOH), an emerging mycotoxin, contaminates various food products. AOH is reported to be genotoxic and to induce oxidative stress. Previous results showed that AOH is also estrogenic and acts as an androgen receptor (AR) agonist, suggesting that it might act as one of the EDCs, however, its detailed molecular mechanism has not been fully elucidated yet. In this study, we decided to evaluate the interplay between AOH and G protein-coupled estrogen receptor 1 (GPER1) in hormone-dependent human ovarian cancer (OC) cells with a focus on the involvement of hypoxia-inducible factor 1-alpha/phosphoinositide 3-kinases/protein kinase B (HIF1α/PI3K/Akt) and claudins (CLDNs) signaling pathways in that effect. We observed that AOH induces oxidative stress and DNA damage in OC cells, and this effect is partially mediated by GPER1. We found that the GPER1 antagonist G15 partially mitigated AOH-induced ROS production and significantly reduced DNA damage, confirming the receptor’s role in mediating these effects. Furthermore, we observed the involvement of HIF1α/PI3K/Akt and CLDNs pathways in AOH effect in OC cells. In conclusion, we postulate that AOH has pro-oxidative ability and that this effect is partially mediated by GPER1. Moreover, we postulated that HIF1α/PI3K/Akt and CLDNs participate in AOH action in OC cells, which in turn provides useful information for future toxicological research studies as a new molecular mechanism of AOH.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-26042-9.

Introduction

Claudins (CLDNs) are small proteins that form tight junctions (TJs) in cells and play a key role in maintaining homeostasis, transport of substances and communication between cells¹. Nevertheless, with the number of publications, it turns out that CLDNs have a greater role than previously speculated in many different cellular processes, including oxidative stress and DNA damage^2–4. Moreover, it was shown that the expression of CLDNs is regulated by hormones¹, whereas claudin-1 (CLDN1), claudin-3 (CLDN3) and claudin-4 (CLDN4) are reported to be the most sensitive to hormone dysregulation. Since CLDNs can be regulated by hormones, there is a risk that endocrine-disrupting chemicals (EDCs) mimicking natural estrogen affect their expression¹. Especially, when taking into consideration that their expression and localization are tissue-dependent and change during disease processes¹. Especially in the case of endocrine-sensitive cancers like ovarian cancer, in which CLDN3 and CLDN4 were reported to be upregulated³.

The direct influence of EDCs on the proliferation of ovarian cells is linked with classical estrogen signaling⁵. However, EDCs may affect the cells indirectly by interaction and modulation of other signaling pathways, especially via non-genomic estrogen signaling^6,7. G protein-coupled estrogen receptor 1 (GPER1) is known to regulate the expression and activity of the NADPH oxidase (NOX1), an enzyme participating in the production of reactive oxygen species (ROS)^8,9. It is also reported that GPER modulates the activation of the phosphoinositide 3-kinase (PI3K)/protein kinase B (Akt) and extracellular signal-regulated kinases 1/2 (ERK 1/2) pathways¹⁰. The PI3K/Akt signaling pathway is associated with the modulation of many factors, also including those associated with the response to oxidative stress: hypoxia-inducible factor 1α (HIF1α)¹¹. What is more, HIF1α modulates various biological processes in cells such as apoptosis, oxidative stress, proliferation and invasion^12,13. Lastly, it has been reported that CLDNs are targeted by HIF1α¹⁴.

Some of the mycotoxins, the toxic fungal metabolites present in the daily diet, are considered EDCs¹⁵. The best-known estrogenic mycotoxin is zearalenone (ZEN), which, due to its structural similarity to estrogen, binds directly to estrogen receptors (ERs). Another mycotoxin with suggested androgenic and estrogenic properties¹⁶ is alternariol (AOH) produced by Alternaria species¹⁵. So far, we know that AOH induces oxidative stress and apoptosis and modulates cell cycle progression and invasiveness^17–19. Previous results showed that AOH possesses a higher binding affinity to estrogen receptor β (ERβ)²⁰. However, its role in the induction of oxidative stress seems to be only partial²¹, similar to estrogen receptor α (ERα)¹⁹. In that case, the non-genomic action might be an explanation, mediated by non-classical estrogen receptor- GPER1²².

Taking into consideration that mycotoxins are reported to modulate the expression of CLDNs in different cell types²³ as well as oxidative stress is one of the basic molecular mechanisms of AOH action¹⁵, in this study, we decided to evaluate if and how AOH might induce oxidative stress and DNA damage via GPER1/PI3K/HIF1α/CLDNs signaling pathway in hormone-responsive ovarian cancer (OC) cells with the main focus on the role of GPER1 in DNA damage and oxidative stress.

Materials and methods

Cell culture

OVCAR3 cells were obtained from ATCC (LGC, Teddington, England). Cells were cultured in a humidified incubator under standard conditions (5% CO₂; 37 °C) and maintained in RPMI 1640 medium supplemented with 10% fetal bovine serum (FBS), 1% penicillin/streptomycin, L-glutamine, sodium pyruvate and 4-(2-Hydroxyethyl) piperazine-1-ethanesulfonic acid (HEPES). Medium without FBS and antibiotics was used as an experimental medium. All the cell culture supplements were derived from Thermo Fisher Scientific (Waltham, MA, USA).

Treatment

AOH (LGC, Teddington, England) was dissolved in DMSO as a stock solution of 0.01 M. GPER1 inhibitor (G15) and GPER1 agonist (G1) were derived from Tocris (Bio-Techne, Bristol, GB) and dissolved in DMSO. All compounds were dissolved directly before use, in the experimental medium, to obtain solutions of the desired concentrations. The final concentration of G1 and G15 was 10 nM and 20 nM, respectively. The concentration of DMSO in the experimental medium was lower than 0.1% and did not affect the viability of cells in all experiments. Based on the results from the viability test and GPER1 activation, two doses of mycotoxin (10 µM and 1 µM) were selected for further study. Cells treated with the experimental medium were used as a control (Cnt).

Cell viability

The viability of OVCAR3 cells was determined with an MTT assay (Merck Millipore, USA). Cells were seeded onto a 96-well plate at a density of 2 × 10⁴. After attachment, cells were treated with the experimental medium containing AOH in a concentration range of 0.001–100 µM. After 20 h, 10 µl of freshly dissolved MTT (5 mg/mL) in phosphate-buffered saline (PBS) was added and incubated in standard conditions for the next 4 h. Then, the medium was removed, and the formazan crystals were dissolved in 100 µl of DMSO. Absorbance was measured at 570 nm (BioTek Synergy Neo 2 plate reader, BioTek Winooski, Vermont, U.S.). Cell viability was expressed as a percentage of Cnt.

Oxidative stress and DNA damage

All flow cytometry assays were conducted with Muse® Cell Analyzer (Merck Millipore, USA). Cells were seeded on 12-well plates and left to reach 70% confluence. Then, they were treated with AOH or a combination of AOH and G15 for 24 h. Then, the assays were conducted, for oxidative stress: Muse® Oxidative Stress Kit (Luminex®; Austin, Texas, USA) and for DNA damage: Muse® Multi-Color DNA Damage Kit (Luminex®; Austin, Texas, USA) according to the instructions provided by the producer. All experiments were conducted in triplicate.

Confocal microscopy

Cells were seeded onto an 8-well chamber slide at a density of 4 × 10⁴ cells (Nunc ™ Lab-Tek ™ II Chamber Slide™ System; Thermo Fisher Scientific Inc, Waltham, MA, USA). The following day, cells were treated with AOH or a combination of AOH and G15. After 24 h, the medium was removed, and cells were fixed with 4% paraformaldehyde (15 min; room temperature). Next, cells were washed three times in 1 × dPBS for 5 min and blocked (5% BSA, 0.3% Triton X-100 in 1 × dPBS) for 1 h at room temperature. After blocking, cells were incubated overnight at 4 °C with primary antibody phospho-histone H2A.X (1:800; #9718; Cell Signaling Technology, Leiden, WZ, The Netherlands). The next day, slides were washed three times in PBS for 5 min, then incubated with secondary antibodies for 1 h at room temperature in the darkness (1:1000; #4412; Cell Signaling Technology, Leiden, WZ, The Netherlands). Afterwards, cells were washed once again as before, and to stain nuclei, NucSpot 555/570 (Biotium, CA, USA #41,033) was used. Next, cells were washed and coverslipped with EverBrite ™ Hardset Mounting Medium (Biotium, CA, USA #23,003). Olympus iXplore SpinSR ScanR (Olympus, Tokyo, JP) was used for visualization- magnification 60 × .

RTqPCR

Cells were seeded onto 60 mm Petri dishes and cultured to achieve 80% confluence. Next, cells were treated for 24 h as explained before. TRIZOL reagent was used for RNA isolation. Then, RNA was isolated, and the concentration was measured with NanoDrop (Thermo Fisher Scientific Inc, Waltham, MA, USA). 2 µg of RNA was used for reverse transcription (#4,368,813; Applied Biosystems, Waltham, Massachusetts, USA). Primers were designed using Primer-BLAST (National Institutes of Health). The Human Reference RNA (Stratagene, San Diego, CA, USA) was used to calibrate the reactions. Ribosomal protein S17 (RPS17), ribosomal protein P0 (RPLP0), and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) were used as reference genes. Verification of the melting curve was performed for each reaction. The ΔΔCt method was used for data analysis. The experiment was conducted in triplicate with two technical replicates.

Western blot

Cells were seeded onto 100 mm Petri dishes and cultured to reach 70% confluence. Next, cells were treated and after 24 h, were harvested and the total protein was isolated in RIPA buffer with phenylmethylsulfonyl fluoride (PMSF), protease and phosphatase inhibitor cocktails (Sigma-Aldrich, St. Louis, MO, USA). The concentration of the protein was measured with BCA assay (Thermo Fisher Scientific Inc, Waltham, MA, USA). 20 μg of protein was used for electrophoresis. Akt (#9272), phospho-Akt (Ser473) (#4060), PARP1 (#46,011), SOD1 (#4266), SOD2 (#13,141), pH2AX (#9718), CLDN1 (#13,995), CLDN3 (#83,609), PI3K (#11,889), p44/42 (#4695), phospho-p44/42 (#4370), CLDN4 (STJ114778), GPER1 (STJ192629) (St John’s Laboratory Ltd, University Way, London) antibodies were used according to the manufacturer’s recommendations. The results were normalized to GAPDH (#2118) or vinculin (#13,901) (Cell Signaling Technology, Leiden, WZ, The Netherlands) as a reference protein. As secondary antibodies, anti-mouse (#7076) and anti-rabbit (#7074) were used (Cell Signaling Technology, Leiden, WZ, The Netherlands). Proteins on membranes were visualized with ChemiDoc MP (Bio-Rad Laboratories, Inc., Hercules, California, U.S.). Densitometry analyses were performed in the Image Lab software (Bio-Rad Laboratories, Inc., Hercules, California, U.S.).

Statistical analysis

Data were analysed in the GraphPad Prism program. Statistical significance was determined with One-Way ANOVA with Bonferroni post-hoc test. A p-value lower than 0.05 indicates statistically significant results.

Results

AOH modulates the expression of GPER1, induces oxidative stress and DNA damage in OVCAR3 cells

In our previous study, we evaluated how AOH affects the viability of SKOV3 cells¹⁹. In this study, we observed that AOH also decreases the viability of OVCAR3 cells in a dose-dependent manner (Fig. 1A). The IC ₅₀ value for OVCAR3 cells after 24 h of exposure to AOH was equal to 59.26 µM. Based on these results, the two doses of AOH were chosen (10 µM and 1 µM). Next, RTqPCR and western blot analysis were performed to verify whether these doses affect the expression of GPER1. We observed that AOH increases the expression of GPER1 at gene and protein levels in a dose-dependent manner (Fig. 1B and C). The role of ERα in the observed effect of AOH was excluded, due to the lack of functional expression in OVCAR3 cells used in the experiments, confirmed with RTqPCR and Western blot method, where lack of expression as well as lack of 66 kDa form of ERα was observed (Supplementary file 2).

Fig. 1.

AOH decreases the viability of OVCAR3 cells and modulates the expression of GPER1. (a) Viability of OVCAR3 cells: The viability of cells after treatment with AOH for 24 h (MTT assay, the results are expressed as % of control (non-treated cells)). (b) GPER1 gene expression: Relative expression of GPER1 after 24 h with AOH (c) GPER1 protein expression: Western Blot analysis of GPER1 expression after 24 h with AOH, the representative results. All the results are expressed as mean ± SE. One-way ANOVA with Bonferroni correction was used as the statistical analysis. P < 0.05 was considered statistically significant, *p < 0.05, **p < 0.01, ***p < 0.001. AOH alternariol, GPER1 G protein-coupled estrogen receptor 1, Vin vinculin, Cnt control (non-treated cells).

To assess the effect of AOH and reveal the role of GPER1 in AOH-induced oxidative stress, we counted ROS-positive cells with flow cytometry. We observed that AOH in both tested doses significantly increased the number of ROS-positive cells compared to non-treated cells (***p < 0.001) (Fig. 2A and B). The addition of G15 to the 10 µM of AOH resulted in a significant (*p < 0.05) reduction in the number of ROS-positive cells, indicating that blocking GPER1 partially reduces AOH-induced oxidative stress (Fig. 2A and B). Both agonist (G1) and antagonist of GPER1 (G15) stimulate the generation of ROS in cells (Fig. 2A and B) (Table 1).

Fig. 2.

AOH induces oxidative stress in OVCAR3 cells. (a) Oxidative stress: The analysis of the generation of ROS with flow cytometry. (b) Representative results of flow cytometry. (c) SOD2 protein expression: Expression of SOD2 was evaluated with Western Blot, the representative results. (d) SOD1, SOD2, PARP1 and XRCC1 gene expression: Relative expression of SOD1, SOD2, PARP1 and XRCC1. (e) DNA damage: evaluated with flow cytometry (ATM and H2A.X activation). (f) Representative results of flow cytometry. (g) p-H2A.X, PARP1 and cleaved-PARP1 protein expression: Expression of phospho-H2A.X, PARP1 and cleaved—PARP1 evaluated with Western Blot, the representative results. (h) Confocal microscopy: Bar graph showing the difference in normalized mean fluorescence intensity signal. (i) Immunofluorescent staining of phospho-H2A.X (green) and nucleus (orange, NucSpot 555 / 570). All the results are expressed as mean ± SE. One-way ANOVA with Bonferroni correction was used for statistical analysis. P < 0.05 was considered statistically significant, *p < 0.05, **p < 0.01, ***p < 0.001. AOH alternariol, G15 selective GPER1 receptor antagonist, G1 selective GPER1 agonist, SOD1 Superoxide dismutase 1, SOD2 Superoxide dismutase 2, PARP1 Poly [ADP-ribose] polymerase 1, XRCC1 X-ray repair cross-complementing protein 1, GAPDH anti-glyceraldehyde 3-phosphate dehydrogenase, pH2AX phospho-H2A histone family member X. Cnt control (non-treated cells).

Table 1.

Primers used in RTqPCR.; CLDN1 claudin 1, CLDN3 claudin 3, CLDN4 claudin 4, GAPDH glyceraldehyde 3—phosphate dehydrogenase, GPER1 G protein-coupled estrogen receptor 1, HIF1α Hypoxia—inducible factor 1—alpha, PARP 1-poly [ADP-ribose] polymerase 1; RPLP0 ribosomal protein P0; RPS17 ribosomal protein S17, SOD1 superoxide dismutase 1, SOD2 superoxide dismutase 2.

Gene	Sequence (5′—3′)	Product size [bp]
CLDN1	For AAGATGAGGATGGCTGTCA Rev ATTCGTACCTGGCATTGACTG	139
CLDN3	For CCTTCATCGGCAGCAACATCA Rev CAGCAGCGAGTCGTACACCT	113
CLDN4	For CAGACAAGCCTTACTCCGCC Rev CCAGGGAAGAACAAAGCAGA	111
GAPDH	For TTCCCGTTCTCAGCCTTGAC Rev GATTTGGTCGTATTGGGCGC	169
GPER1	For AACCCAGGTACCCAGAGAGTGAG Rev GTGGGAAGAACAGATGCTCCTCA	147
HIF1α	For TTACTCATCCATGTGACCATGA Rev AGTTCTTCCTCGGCTAGTTAG	140
PARP1	For TCTTCAAGAGCGATGCCTATT Rev TGAGGTAAGAGATTTCTCGGAA	129
RPLP0	For ACGGATTACACCTTCCCACTTGCTGAAAAGGTC Rev AGCCACAAAGGCAGATGGATCAGCCAAG	69
RPS17	For AAGCGCGTGTGCGAGGAGATCG Rev TCGCTTCATCAGATGCGTGACATAACCTG	87
SOD1	For GCGTGGCCTAGCGAGTTAT Rev ACACCTTCACTGGTCCATTACT	114
SOD2	For GGGTTGGCTTGGTTTCAATAAG Rev CTGAAGGTAGTAAGCGTGCTC	136
XRCC1	For TCAAGGCAGACACTTACCGAA Rev TCCAACTGTAGGACCACAGAG	76

Then, we decided to evaluate the expression of protein connected with oxidative stress in cells, SOD2, and observed that AOH increased the expression of SOD2, while adding G15 decreased this effect (Fig. 2C and Table 2). We observed that in OVCAR3 cells, AOH in a dose of 10 µM significantly decreases the expression of SOD1, SOD2 and PARP1 (****p < 0.0001) compared to non-treated cells, however, the addition of G15 did not induce significant changes (Fig. 2D). The difference was observed in the expression of X-ray repair cross-complementing protein 1 (XRCC1), a protein involved in the repair of DNA single-strand breaks (SSB). We noticed increased expression after treatment with AOH, and significantly higher expression of this gene after simultaneous treatment with AOH and G15 than with AOH treatment alone (Fig. 2D), suggesting increased activity of DNA repair.

Table 2.

The expression of proteins associated with oxidative stress and DNA damage was modulated by AOH. The protein expression was evaluated using a Western blot and presented in the table as mean of fold change of the density ± SE of three independent replicates. One-way ANOVA with Bonferroni correction was used for statistical analysis, p < 0.05 was considered significant. SOD2 superoxide dismutase 2, phospho-H2A.X phospho-H2A histone family member X, PARP1 poly (ADP-ribose) polymerase 1, AOH alternariol, G15 selective GPER1 receptor antagonist, G1 selective GPER1 agonist, Cnt control (non-treated cells).

In that case, we decided to verify whether the induction of oxidative stress is also associated with DNA damage and if GPER1 might have a role in it. Firstly, cells expressing phospho-ataxia telangiectasia mutated kinase (phospho-ATM) and phospho-histone H2A.X (phospho-H2A.X) were stained and counted with flow cytometry. It was observed that 10 µM of AOH significantly increased DNA damage (Fig. 2E and F). The addition of G15 significantly reduced DNA damage compared to AOH treatment alone (Fig. 2E, *p < 0.05). In line with previous results, we also observed increased expression of phospho-H2A.X after treatment with AOH in western blot analysis (Fig. 2G and Table 2). The addition of G15 also decreased this effect (Fig. 2G and Table 2). We also observed changes in the expression of poly (ADP-ribose) polymerase 1 (PARP) and its cleaved form at the protein level. 10 µM of AOH decreased the expression of PARP, and the addition of G15 reversed this effect (Fig. 2G and Table 2). In turn, 1 µM of AOH, G1 and G15 alone increased their expression (Fig. 2G and Table 2). In the case of cleaved-PARP, we observed increased expression in all tested groups compared to the control (Fig. 2G and Table 2). The addition of G15 to 1 µM and 10 µM of AOH increased this effect even more (Fig. 2G and Table 2).

Phospho-H2A.X expression and localization were also visualized with confocal microscopy. We observed that 10 µM of AOH significantly increased phospho-H2A.X expression (Fig. 2H and I). Adding G15 significantly (***p < 0.001) reduced the number of phospho-H2AX foci compared to cells treated only with AOH (Fig. 2 and I), suggesting a potential role of GPER1 in AOH- induced DNA damage.

AOH affects HIF1α/PI3K/Akt signaling pathway and modulates the expression of claudins in ovarian cancer cells

Based on the modulation of oxidative stress and DNA damage by AOH, we decided to verify the hypothesis that AOH modulates HIF1α/PI3K/Akt/CLDNs signaling pathway, due to the known interactions between these factors^14,24. In the tested cell line, treatment with AOH decreased the expression of HIF1α (Fig. 3A). We also noted that the addition of G15 lowered this expression even further at both doses of AOH (*p < 0.05; **p < 0.01) (Fig. 3A). We also observed that AOH modulated the expression of claudins (Fig. 3B). The increased expression of CLDN1 after treatment with 10 µM of AOH was observed, while the addition of G15 partially decreased this effect (Fig. 3B and Table 3). In the case of CLDN3, we observed decreased expression, and the addition of G15 to AOH did not change it (Fig. 3B and Table 3). Interestingly, G1 itself increased the expression of CLDN3, and the addition of G15 to G1 abolished this effect (Fig. 3B and Table 3). In the case of CLDN4, both doses of AOH increased its expression, similarly to G1 and G15 (Fig. 3B and Table 3). The addition of G15 to the lower dose of AOH increased this effect even more (Fig. 3B and Table 3). In the case of gene expression, we observed that AOH significantly increases CLDN1 expression itself, while the addition of G15 increases it even more, however, statistically insignificant (Fig. 3C). In the case of CLDN3, in OVCAR3 cells we observed a significant decrease after AOH treatment and the addition of G15 abolished this effect to the level of the control (Fig. 3C). In CLDN4 expression, we observed significant upregulation when OVCAR3 cells were treated with a higher dose of AOH (Fig. 3C). The same trend was observed for G1. The addition of G15 to both AOH and G1 diminished this effect (Fig. 3C).

Fig. 3.

AOH modulates the expression of HIF1α, Akt and claudins. (a) HIF1α gene expression: Relative expression of HIF1α. (b) CLDN1, CLDN3 and CLDN3 protein expression: Western Blot analysis of CLDN1, CLDN3 and CLDN3, the representative results. (c) CLDN1, CLDN3 and CLDN3 gene expression: Relative expression of CLDN1. CLDN3 and CLDN4. (d) pAkt, Akt and PI3K protein expression: Western Blot analysis of pAkt, Akt and PI3K, the representative results. (e) pp44/42 and p44/42 protein expression: Western Blot analysis of pp44/42 and p44/42, the representative results. p < 0.05 was considered statistically significant. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 as compared to the control. AOH alternariol. GPER1 G protein-coupled estrogen receptor 1. G15 selective GPER1 receptor antagonist. G1 selective GPER1 agonist, Cnt control (non-treated cells), Vin vinculin, HIF1α hypoxia-inducible factor 1-alpha, CLDN1 claudin 1, CLDN3 claudin 3, CLDN4 claudin 4, p44/42 extracellular signal-regulated kinases, phospho-p44/42 phospho-extracellular signal-regulated kinases, Akt protein kinase B, pAkt phospho-protein kinase B, PI3K phosphoinositide 3-kinase.

Table 3.

The protein expression of PI3K/Akt /CLDNs. The results are expressed as the mean of the fold change of control (non-treated cells). The protein expression was evaluated with a Western blot and presented in the table as mean density ± SE of three independent replicates. One-way ANOVA with Bonferroni correction was used for statistical analysis; p < 0.05 was considered significant, **p < 0.01. AOH alternariol, CLDN1 claudin 1, CLDN3 claudin 3, CLDN4 claudin 4, pAkt phospho-protein kinase B, Akt protein kinase B, PI3K phosphoinositide 3-kinase, p44/42 extracellular signal-regulated kinases, pp44/42 phospho-extracellular signal-regulated kinases, G15 selective GPER1 receptor antagonist, G1 selective GPER1 agonist, Cnt control (non-treated cells).

In western blot analysis, we observed that PI3K expression was also modulated by AOH (Fig. 3D and Table 3). In OVCAR3 cells, all treatments increased the expression of PI3K as compared to control, but with no apparent differences between them (Fig. 3D and Table 3). The expression of Akt was not altered (Fig. 3D and Table 3), whereas increased expression of phospho-Akt was observed after treatment with 1 µM AOH + G15, G15, G1 and G15 + G1 (Fig. 3D and Table 3). We noticed that AOH itself had no effect on p44/42 phosphorylation, while blocking GPER1 increased phospho-p44/42 expression in the cells (Fig. 3E). G15 and G1 alone also significantly increased the amount of this protein in these cells.

Discussion

GPER1 so far has not been considered as the target of AOH, possibly due to the fact that in estrogenicity assay it is not taken into consideration. It is well known that the binding affinity of AOH to ERs is much lower than natural estrogen²⁰. Previous studies carried out by us showed that in hormone-dependent cells, none of these classic ERs are fully responsible for the induction of oxidative stress triggered by AOH^19,21. Thus, it is highly possible that other mechanisms or more likely interaction with different signaling pathways might be the molecular mechanism of that mycotoxin. A previous study conducted with ZEN, the most potent estrogenic mycotoxin, in colon cells, showed that GPER1 plays a significant role in its effect²⁵. Moreover, the interaction with other signaling pathways is more obvious in the case of GPER1 than in classic ERs²⁶. In this study, we observed that AOH significantly affected the expression of GPER1 in OC cells. A similar effect was observed by Lo et. al., where expression of GPER1 was elevated upon treatment with ZEN in HT-29 and SW480 cells²⁵. Moreover, as suggested before, other environmental EDCs significantly modulate GPER1 expression, even when their estrogenicity is lower than natural estrogens, and the effect mediated by GPER1 is comparable to estrogen in estrogen- responsive cells²⁶. GPER1 was previously reported to participate in bisphenol A (BPA) induced toxicity, with modulation of the cell cycle, and activation of ERK and PI3K/Akt signaling pathway²⁷.

Oxidative stress is a well-established contributor to ovarian carcinogenesis, participating in and mediating genetic alterations, signaling pathways and transcription factors²⁸. AOH stimulates the generation of ROS and induces DNA damage in cells, mainly double-strand DNA breaks (DSB) via interaction with DNA topoisomerase¹⁵. We previously presented that AOH induces the generation of ROS in prostate and breast cancer cells^21,29. Fernández-Blanco et. al. observed increased oxidative stress in Caco-2 cells after treatment with AOH³⁰. In our present study, we observed that both 1 µM and 10 µM increased ROS production in OVCAR3 cells. However, the blocking of GPER1 only partially decreased the effect stimulated by the higher dose, which indicates that it only partially mediates this effect. We also presented that the increase in ROS-positive cells after AOH treatment was associated with modulation of SOD1, SOD2, PARP1 and XRCC1 expression, but the addition of G15 did not reverse the effect, except for XRCC1, where the addition of G15 increased its expression. This observation suggests activation of pathways involved in the repair of DNA SSB. The modulation of oxidative stress and DNA damage caused by mycotoxins exposure in ovarian cells was also observed previously: Zhang et. al. also observed increased expression of PARP1 and XRCC1 in porcine ovarian granulosa cells after treatment with ochratoxin A (OTA)³¹. On the protein level, we observed decreased expression of SOD2 after blocking of GPER1 compared to AOH treatment alone. Modulation of SOD2 expression by AOH during oxidative stress induction was also observed in the colorectal adenocarcinoma cell line Caco-2³⁰. The role of GPER1 in oxidative stress might be two-fold: it was reported that GPER1 overexpression might protect against oxidative stress in cells³², or participate in pro-apoptotic and oxidative-stress-induced effects³³. In our study, it seems that only partially GPER1 participates in AOH-induced oxidative stress. Oxidative stress induced by another EDCs, BPA, was observed in ovarian cancer stem cells in mice³⁴. Similar to our study, BPA decreased the expression of SOD1 and SOD2³⁴. Mlynarcikova et. al. presented that bisphenol S (BPS), bisphenol F and bisphenol AF (BPAF) at high concentrations also increased the generation of ROS in ovarian cancer cells³⁵.

Genotoxic effects of AOH were observed in various cell lines at relatively low concentrations (1 µM)³⁶. In this study, we observed that the DNA-damaging effect of AOH was stronger than the induction of oxidative stress. This observation seems to be in line with a previous study made by Solhaug et. al., who observed that AOH in a dose of 30 µM increases the phosphorylation of H2A.X in RAW 264.7 cells^18,37. It also seems that the DNA damage caused by AOH exposure might be cell type-sensitive, and hormone- responsive cells seem to be more sensitive than, e.g. macrophages or hepatic cells³⁸. In the case of other EDCs, Chang et. al. showed that mono-ethylhexyl phthalate (MEHP) also induces DNA damage in mammalian Chinese hamster ovary cells³⁹.

CLDNs are TJs proteins that link epithelial cells¹. They play a crucial role in carcinogenesis due to their involvement in migration, invasion, metastasis and drug resistance³. Modulation of their expression is linked with PI3K/Akt signaling pathway². Previous studies showed that AOH might regulate the expression of CLDNs in colon cancer cells⁴⁰, however, its role is also crucial in OC. Our previous results also showed that AOH modulates the expression of HIF1α⁴⁰, whereas another study confirmed the interaction between HIF1α and PI3K/Akt/ERK signaling pathway²⁴. In turn, CLDNs were reported to control pathways like PI3K/Akt and ERK 1/2⁴¹. Thus, it seems that it might be an interaction between these signaling pathways. In our study, we observed that AOH reduces HIF1α expression, while GPER1 blockade reduces its expression even further. Liu et. al. showed the association between CLDNs and HIF1α¹⁴. They may be dysregulated or/and translocated by mycotoxins, which was mainly observed in colon cancer cells²³, however, CLDNs might also be modulated by hormones¹. In our study, we observed modulation of CLDN1, CLDN3 and CLDN4. We observed increased expression of CLDN1 both on protein and gene levels after treatment with 10 µM of AOH. The addition of G15 increased this effect, however, it was statistically insignificant. This observation is consistent with the previous one made by Akimoto et. al., who observed that estrogen upregulates expression of CLDN1 and that this induction was mediated by GPER1 via ERK and/ or Akt signaling⁴². This is in line with our observation. Moreover, other mycotoxins were also reported to modulate these signaling pathways: ZEN⁴³ and deoxynivalenol (DON)⁴⁴. In the case of CLDN3, blocking of GPER1 increased the reduced expression by AOH to control levels at the gene level, while at the protein level, we observed increased expression after the addition of G15 to 1 µM of AOH. Previous studies suggested that CLDN3 overexpression modulates the malignant properties of colon cancer cells, and that effect is associated with modulation of PI3K/Akt/ERK via epidermal growth factor (EGFR)⁴⁵. Interestingly, EGFR is known to directly interact with the GPER1 signaling pathway⁴⁶, which in turn might confirm our observation that these interactions are present. Other EDCs, are also reported to have the ability to activate PI3K/Akt signaling. For instance, Li et. al. showed that bisphenol P (BPP) and bisphenol M (BPM) promote TNBC metastasis by PI3K/Akt-mediated EMT⁴⁷. Ptak et. al. showed that BPA induces migration by activating MAPK and PI3K/Akt signalling pathways in OVCAR3 cells⁴⁸. CLDN4 expression was increased after treatment with AOH, and adding G15 abolished this effect. Cuevas et. al. also observed modulation of CLDN4 expression and localisation after treatment with estradiol in endometrial cancer cells⁴⁹. Fen et. al. noticed that BPA is able to disrupt intestinal barrier function in mice by affecting the expression of, among others, CLDN1⁵⁰. These interesting observations confirm the observed broad scope of CLDNs role in both physiological and pathological processes in cells¹.

Conclusions

Taken together, this study for the first time showed that GPER1 might be involved in AOH- induced oxidative stress and DNA damage in hormone-responsive cells, in this case, OC cells, and suggested that HIF1α/ PI3K/Akt/ERK/CLDNs interaction is involved in that process. This observation shed new light on the known mechanism of AOH action and its potential effect as EDC; however, more detailed molecular studies, as well as in vitro and in vivo-based models, are needed in further research to confirm our observation.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

K.L.M.J.- analysis design, collection of data, performing analysis, writing and revision of the manuscript, providing the funding; H.-G.D.E; L.K. P.; U. K. A.- collection of the data, P.-C.A.W.- providing the funding and supervision; K.K- analysis design, concept of the manuscript, writing and revision of the manuscript.

Funding

The research was funded by the Rector of the Medical University of Lodz under the Programme of Grants Internal BRaIn, contract number 1/2023 (to K.L.M.J.) grant no. 503/0-078-03/503-90-102 and Medical University of Lodz, grant no. 503/0-078-03/503-01-001 (to P.-C.A.W.).

Data availability

The datasets generated and analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare that they have no competing interests.

Ethics approval and consent to participate

Not applicable.

Consent for publication

All authors read and approved the final version of this manuscript.