Methylparaben-induced regulation of estrogenic signaling in human neutrophils

Parabens, including the most common methylparaben (MeP), are popular preservatives, which possess estrogenic activity. The aims of this study were to assess the impact of MeP on estrogen receptors (ERs) and/or NF-κB-dependent generation of IL-8 and production of nitric oxide (NO), and also to verify the hypothesis about the crosstalk of ERs with NF-κB in xenoestrogen-exposed neutrophils. Human neutrophils were incubated for 20-h with MeP (0.06 μM) and/or ER antagonist (1 μM) and/or NF-κB inhibitor (100 μM). After the isolation of cell lysates and cytoplasmic and nuclear fraction, the expression of ERα, ERβ, p-IKKα/β, p65 NF-κB, and inducible nitric oxide synthase (iNOS) was measured by Western blot analysis, The concentration of NO was evaluated by Griess reaction, and that of IL-8 was measured by ELISA. The results showed that MeP modulated the expression of ERα, but not ERβ. Exposure to paraben activated iKKα/β-dependent NF-κB pathway, but translocation of p65 NF-κB into the cell nucleus was inhibited by ERs. MeP also decreased the iNOS-dependent production of NO, but did not influence the secretion of IL-8 by neutrophils. The study indicates that MeP may affect the functioning of human neutrophils by modulating intracellular signal transduction pathways, including ERs and NF-κB pathway.


Introduction
Preservatives are substances added to cosmetics, pharmaceuticals, and personal care and food products to prolong their shelf-life and inhibit the growth of microorganisms. Parabens are one of the compounds used as preservatives because of their antibacterial and antifungal properties, of which methylparaben (MeP) is the most widely used. The presence of MeP in food must be marked on the product label (E218). Most recent studies have shown the presence of MeP in frozen fruits, meat, olives, chips, rice, and dairy products. It should be noted that any perfect antimicrobial agent should be effective, efficient, and relatively cheap, while at the same time it should not be harmful to humans and the environment. In the case of parabens, issues on their safety strongly polarize the scientific and consumer communities (Gálvez-Ontiveros et al., 2021;Matwiejczuk et al., 2020;Nowak et al., 2018Nowak et al., , 2020a. Parabens have been regarded as xenoestrogens (XEs)-exogenous compounds mimicking the action of estrogens. Other compounds that are widespread in the human environment, such as bisphenols including bisphenol A (BPA), F, AF, and S, nonylphenol (NP), octylphenol (OP), and organochlorine compounds, are also categorized as XEs (Acconcia et al., 2017;Nowak et al., 2019). The main mode of action of XEs is their interaction with hormone receptors such as estrogen receptor (ER)-α (NR3A1) and ERβ (NR3A2). The natural ligands of ERs are estrogens, which on binding to these receptors lead to conformational changes and chaperon dissociation and result in the formation of homodimers (ERα+ERα and ERβ+ERβ) and heterodimers (ERα+ERβ). These dimers translocate into the cell nucleus, where they bind to estrogen response elements and control estrogen-regulated genes. They may also activate/inhibit other intracellular signaling pathways and affect the expression of nuclear transcription factors (Acconcia et al., 2017;Nowak et al., 2019;Paterni et al., 2017;Toporova and Balaguer, 2020).
ERs are found in various types of cells, which make these cells sensitive to estrogen's action and therefore potential targets of XEs. ERα and ERβ have also been detected in human neutrophils-immunocompetent cells responsible for inducing immediate immunological response against pathogenic microorganisms (Garley and Jabłońska, 2018;Janiuk et al., 2021;Martín-Millán and Castañeda, 2013;Molero et al., 2002;Nowak et al., 2019).
Neutrophils are the first immune cells to arrive at the site of an ongoing inflammatory process. They can directionally migrate in response to a gradient of chemotactic factors such as IL-8. Another feature of neutrophils is their ability to generate cytotoxic nitric oxide (NO) in an enzymatic, inducible nitric oxide synthase (iNOS)-related reaction (Ratajczak-Wrona et al., 2020;Ratajczak-Wrona and Jabłońska, 2019). All these processes occur as a result of activation or inhibition of intracellular signaling pathways such as NF-κB. In neutrophils, the transcriptional factor NF-κB controls numerous transcriptional programs, such as the expression of genes involved in the generation of inflammatory mediators. Interestingly, scientists have identified ER dimers among the modulators of NF-κB Xing et al., 2012).
This study aimed to analyze the impact of MeP on intracellular signal transduction pathways in human neutrophils. In particular, it focused on investigating the following: (I) Does MeP affect estrogenic signaling involving ERα and ERβ? (II) Does MeP exposure directly cause NF-κB modulation? (III) Does potential MeP-induced modification of the activity of NF-κB pathway occur via ERs? The study also examined whether MeP exposure influences iNOS-dependent production of NO and generation of IL-8 by neutrophils and whether the generation of inflammatory factors is coregulated via ERs/NF-κB signaling pathways.
The research was conducted on 30 participants: healthy male volunteer blood donors from the Regional Centre for Transfusion Medicine, Bialystok, Poland. All donors gave written informed consent prior to blood donations. Donors were non-smokers and non-alcohol consuming at least 48 h before blood donation. All males were healthy, without any immunological disorders and not under any hormonal treatment. Donors white blood cell count was counted in Bürker hemocytometer after incubation with Türk's solution (crystal violet and acetic acid) (AQUA-MED, Lodz, Poland). May-Grünwald-Giemza (AQUA-MED, Łódź, Poland)-stained blood smears were also assessed. Results are presented in Table 1.
Neutrophils were isolated from heparinized (10 U/ml heparin) blood by centrifuge in density gradient -Polymorphoprep™ (Axis-Shield Diagnostics, Dundee, United Kingdom) ( Fig. 1). Using of density gradient enables separation of two peripheral white blood cell fractions: mononuclear cells (PBMCs) and polymorphonuclear leukocytes (PMNs -93% pure). To obtain highly pure (99%) neutrophils fraction, PMNs were positively selected in Midi-MACS magnetic separation system (Miltenyi Biotec, Bergisch Gladbach, Germany) with use MicroBeads conjugated to monoclonal anti-human CD16 antibodies. Neutrophils survival was assessed in a light microscope after incubation with 0.5% solution of trypan blue (Lachema, Praga, Czech Republic) and were equal to 98%. After each step of isolation, PMNs/neutrophils purity was assessed in a light microscope in May-Grünwald-Giemza-stained so-called "thick drop" smears.
To assess the role of NF-κB or ERs, some neutrophils were 1-h preincubated with NF-κB-inhibitor: ammonium pyrrolidinedithiocarba- After 1-h pre-incubation with or without PDTC and/or ICI 182.780, neutrophils were 20-h incubated with or without methylparaben (methyl 4-hydroxybenzoate, MeP; Merck, Burlington, U.S.A.). The concentration of MeP -0.06 μMwas established according to Sandanger et al., 2011 results, as average MeP concentration measured in the human blood. In our previous study (Nowak et al., 2020b), we determined that 0.06 μM MeP is not cytotoxic for human neutrophils.
MeP was dissolved in DMSO then the stock was diluted in PBS. The final concentration of DMSO in cell culture was up to 0.1%.
After cell culture, neutrophils supernatant was frozen in − 80 • C. The cells' precipitates were collected and immediately used in further steps of research.

Isolation of whole-cell lysates
Neutrophils precipitates were collected from each well and suspended in a Protease Inhibitor Cocktail (Sigma-Aldrich, MERCK) and sonicated (SONICS Vibra Cell). Samples were centrifuged at 15 300×g. In cell lysate supernatants, the amount of protein was measured by Qubit™ Protein Assay Kit and a Qubit 2.0 Fluorometer (Invitrogen™, Thermo Fisher Scientific, Waltham, U.S.A.). Then, samples were prepared for western blot procedure.

Isolation of cytoplasmic and nuclear cellular fraction
Neutrophils precipitates were collected from each well and suspended in PBS 1 × without CaCl 2 and MgCl 2 ions (Gibco, Thermo Fisher Scientific, Waltham, U.S.A.). Isolation of neutrophils cytoplasmic and nuclear fractions was performed according to the NucBuster™ Protein Extraction Kit (Novagen®, Merck, Darmstadt, Germany) procedure.
The first step of isolation was to add the extraction reagent, vortexing, incubation in ice and samples renewed vortexing. After high-speed centrifuge (16 000×g) at 4 • C, the supernatant (cytoplasmic fraction) was held out, while precipitate was used for isolation of nuclear fraction. The protease inhibitors, reducing agent dithiothreitol (DTT), and extraction reagent were added to each sample. Then, cells precipitates were vortexing, incubated in ice and once again vortexing. High-speed centrifuge (16 000×g) at 4 • C allowed to obtain of nuclear fraction (supernatant) of neutrophils.
The concentration of protein was measured by the fluorometric method in Qubit™ 2.0 Fluorometer (Invitrogen™, Thermo Fisher Scientific, Waltham, U.S.A.). Use of Qubit™ Protein Assay Kit (Invi-trogen™, Thermo Fisher Scientific, Waltham, U.S.A.) allowed quantifying protein concentration. Each sample of the cytoplasmic or nuclear fraction was frozen until Western blot analysis.

Assessment of total NO concentration in supernatants by Griess reaction
Total nitric oxide (NO) concentration in cells culture supernatants was measured by an indirect method based on the measurement of nitrite concentration based on Griess reaction. The generation of NO was determined as a sum of the nitrite and nitrate concentrations. Prior to the mensuration, the nitrates were reduced by cadmium. Next, the addition of Griess reagent (Merck, Burlington, USA) induces a colorful reaction of nitrite conversion into the nitric acid. NO concentration was measured in a spectrophotometer (ASYS Hitech GmbH microplate reader, Biogenet) in 540 nm light-length. Preparation of the standard curve allows reading the NO concentration in each sample. Total NO concentration was expressed as μM/10 6 cells in 270 μL.

Measurement of IL-8 generation by ELISA
The concentration of IL-8 in cells supernatant was measured by enzyme-linked immunosorbent assays (ELISA). ELISA Kit for IL-8 was manufactured by Invitrogen, Carlsbad, U.S.A. Kit for human IL-8 uses quantitative sandwich reaction and was made according to the manufacturer's instructions. The absorbance of each sample was measured in spectrophotometer (ASYS Hitech GmbH microplate reader, Biogenet) in 450 nm light-length. IL-8 concentration was calculated from standard curve, and presented as pg/mL.

Statistics
Results were analyzed in Microsoft® Excel® 2016 and Statsoft Statistica version 13.3. Distribution in each parameter groups was checked Shapiro-Wilk test. Results were analyzed in one-way ANOVA test. A p value ≤ 0.05 was accepted as statistically significant. Data are presented as mean ± standard deviation (SD).
Higher expression of ERα was observed in the nuclear fraction than in the cytoplasmic fraction of nonexposed cells, but the expression of ERβ was lower. However, p65 NF-κB was equally distributed in both cellular fractions.
Cells exposed to MeP (0.06 μM) showed an increase in the total expression of ERα and p65 NF-κB (but not ERβ) compared with cells incubated without MeP. The expression of ERα was higher in the nuclear fraction than in the cytoplasmic fraction of these cells, while the expression of ERβ and p65 NF-κB was lower in nuclear fraction than in the cytoplasm. Furthermore, the expression of p-IKKα/β was higher and the expression of iNOS was lower in these cells compared with neutrophils incubated without MeP.
The study analyzed the role of NF-κB pathway by culturing neutrophils with PDTC, an inhibitor of NF-κB pathway. Simultaneous incubation of cells with PDTC and MeP decreased the total expression of ERα in cell lysates compared with nonexposed cells and cells incubated only with tested paraben. However, the expression of ERβ in MeP-and PDTCexposed cells was higher compared to control cells. The addition of PDTC to MeP-cultured cells reduced the total expression of p65 NF-κB compared with neutrophils incubated only with MeP. Higher expression of ERα (but not ERβ and p65 NF-κB) was observed in the nuclear fraction than in the cytoplasmic fraction of cells incubated with MeP and PDTC.
To investigate the role of ERs in signal transduction, we incubated neutrophils with ICI 182.780, an antagonist of ER. Simultaneous incubation of cells with ICI 182.780 and MeP led to a decrease in MePinduced total expression of ERα. Expression of ERα and p65 NF-κB in neutrophils incubated with MeP and ICI 182.780 was significantly higher compared with cells incubated with PDTC and MeP. However, the total expression of ERβ in cells incubated simultaneously with MeP and ICI 182.780 was lower compared with the cells incubated with MeP and PDCT. Higher expression of ERα and p65 NF-κB in the nuclear fraction than in the cytoplasmic fraction of cells incubated with MeP and ICI 182.780 was observed, while ERβ was equally distributed between the cytoplasmic and nuclear fractions of these cells.
No differences in the expression of ERα, ERβ, and p65 NF-κB were observed in cells simultaneously incubated with PDTC and ICI 182.780 compared with nonexposed cells. However, cells incubated with PDTC and ICI 182.780 showed increased levels of ERα expression compared with cells incubated with MeP and PDTC and with MeP and ICI 182.780.
To investigate whether MeP modulates the expression of cytoplasmic proteins such as p-IKKα/β and iNOS via NF-κB pathway and/or ERs, cells were cultured with paraben and PDTC (inhibitor of NF-κB pathway) or with MeP and ICI 182.780 (ER antagonist). In neutrophils cultured with MeP and PDTC or with MeP and ICI 182.780, the expression of p-IKKα/β was higher compared to cells incubated without MeP. In cells simultaneously incubated with MeP and PDTC and in those incubated with MeP and ICI 182.780, the expression of iNOS was decreased compared with the nonexposed neutrophils.

Effect of MeP on NO production
The total concentration of NO in supernatants of neutrophils incubated with MeP and in cells cultured with MeP and PDTC was decreased compared with the concentration of NO in cells not exposed to MeP (Fig. 3). However, in neutrophils simultaneously incubated with MeP and ICI 182.780, no statistical difference was observed in NO production compared with cells cultured without MeP.

Effect of MeP on IL-8 generation
In this study, no observed any statistical difference was found in the concentration of IL-8 in the supernatants of neutrophils incubated with MeP compared with cells incubated without MeP (Fig. 4). Similarly, no statistical differences were observed in cytokine concentration in the supernatants of cells incubated simultaneously with MeP and PDTC or with MeP and ICI 182.780 compared with neutrophils cultured without MeP.

Discussion
This study investigated whether parabens induce a biological effect in neutrophils by modulating estrogenic signaling, including ERα and ERβ. The results showed that only ERα was involved in MeP-induced signal transduction in human neutrophils, which was confirmed by an increase in MeP-induced total ERα expression (Fig. 2). The expression returned to normal levels after the addition of ICI 182.780, an ER antagonist, to the neutrophils. Moreover, ERα is predominantly located in cell nucleus which indicates that MeP is involved in ERα dimerization and activation. It was also observed that the total ERβ expression remained unchanged in cells treated with MeP and ICI 182.780. Furthermore, due to the predominant localization of ERβ in cytoplasm, we excluded the fact that ERβ is activated in paraben-exposed cells. The obtained results indicated that MeP may show estrogenic activity on human neutrophils by activating ERα but not ERβ. Crystallographic and computational experiments confirm that parabens act in a liganddependent way by binding ERs in the same ligand-binding pocket in which endogenous estrogens also bind (but with lower affinity than estradiol). Moreover, the affinity of XEs toward ERs may vary, which might be the reason for the preference of MeP to act via ERα (Delfosse et al., 2015;Sasaki and Terasaki, 2018).
In this study, MeP (0.06 μM) activated the classical NF-κB pathway by increasing the p-IKKα/β-dependent total expression of p65 NF-κB (Fig. 2). Classical NF-κB activation involves the action of IKKα/β, which phosphorylates IκB inhibitor, thereby causing its removal from NF-κB dimers. The released NF-κB dimers translocate into the cell nucleus, where, as a transcriptional factor, they bind to the promotor sequences of genes (Kandasamy, 2021). Use of the PDTC inhibitor prevents the ubiquitination of phosphorylated IκB and precludes the releases of NF-κB dimers (Schreck et al., 1992). In MeP-and PDTC-exposed cells, p65 NF-κB was accumulated in the cytoplasm. However, in MeP-treated cells, with inhibited ER signaling, the localization of p65 NF-κB changed in favor of the nuclear fraction and p65 NF-κB activation when ERs were blocked. Therefore, there may be some interaction, which inhibits the translocation of dimers into the nucleus of neutrophils and prevents their activation in MeP-exposed cells. Comparing the effect of paraben on ERα, but not on ERβ, observed in the present study with those of the previous studies, it could be postulated that MeP-activated ERα is responsible for blocking NF-κB translocation in neutrophils. The results of this study also support the previous reports with regard to the presence of crosstalk between ERs and NF-κB signaling in immune cells treated with XEs. The suppressive effect of estrogenic signaling on the regulator of inflammatory response, NF-κB, is one of the well-known mechanisms of the anti-inflammatory action of estrogens (Smart et al., 2020;Stein and Yang, 1995;Š imková et al., 2020). As demonstrated by Yoshitake et al. (2008), the decrease of p65 NF-κB expression in macrophages exposed to XEs depends on ERs. However, in MeP-exposed neutrophils, blockade of NF-κB signaling increased the total expression of ERβ, but its cellular localization was predominant, which indicates that ERβ signaling remains inactive.
In previous research with MeP (0.06 μM) (Nowak et al., 2020b), no statistically significant differences were noted in the generation of NO by neutrophils that were incubated for 30 min with paraben. However, in the current study, the period of incubation of neutrophils with MeP was prolonged (20 h) and a decrease in the iNOS-dependent production of NO was observed. This shows that NO synthesized by neutrophils plays an important role in invading pathogens, and decreased levels of NO may be observed in people using a range of products that contain parabens as preservatives (Fig. 3). Based on the results obtained in the research on ER antagonist and NF-κB inhibitor, it could be suggested that MeP-induced decrease of NO production is independent of NF-κB and, at least partially, remains under the control of ERs.
In vitro, MeP suppresses the functions of neutrophils, including their ability of chemotaxis (Nowak et al., 2020b)-a process regulated by a gradient of chemotactic factors such as IL-8. Neutrophils not only respond to IL-8 but may also produce cytokines, thereby regulating the immune cell influx to the site of inflammation. The results of this study showed no association of MeP exposure with IL-8 production, as well as with signaling pathways (Fig. 4), which are in accordance with the results obtained from previous in vitro studies. Based on the available scientific data (Aung et al., 2019;Berger et al., 2018;Watkins et al., 2015), the association between the exposure to parabens and intensification/inhibition of cytokine generation could be excluded. In in vivo studies on women participants Š imková et al. (2020) did not find any correlation between the concentration of parabens and 27 cytokines measured in blood serum. Similarly, Watkins et al. (2015) and Berger et al. (2018) did not observe any correlation between the presence of parabens in the urine and concentrations of IL-1β, IL-6, IL-10, and TNF-α in the serum of pregnant women. However, Aung et al. (2019) demonstrated a correlation between the concentration of MeP in the urine and increased level of IL-6 in the serum of pregnant women, but the researchers did not observe any correlation between MeP concentration and other cytokines such as IL-1β, IL-10, and TNF-α. Moreover, no correlation was found between the concentrations of propyl-and butylparaben and levels of cytokines, whereas in the case of ethylparaben, its high concentration was linked with decreased levels of TNF-α.

Conclusions
To sum up, this study showed the impact of MeP on intracellular signal transduction in human neutrophils. MeP may cause multidirectional health implications, including disturbance in antibacterial immune responses in people using cosmetics, food, and pharmaceuticals that contain parabens as preservatives (Fig. 5). The obtained results confirmed that MeP affects ERα receptor but not ERβ and, therefore, paraben cannot directly modulate the activity of NF-κB pathway. However, via ERα, MeP may disrupt the NF-κB pathway-a key pathway that regulates the functioning and survival of neutrophils. Exposure to paraben impairs the ability of neutrophils to produce NO-an important element in the elimination of pathogens. However, fortunately, this study showed that MeP did not affect the secretion of IL-8, a cytokine that plays a key role in regulating neutrophil chemotaxis, the deregulation of which could inhibit the recruitment of immune cells.

Declaration of interest
Authors declare no commercial or financial conflict of interest.  but not ERβ expression. MeP exposure activate iKKα/β-dependent NF-κB pathway, but its translocation into the cells' nucleus was inhibited by ERα regulation. Moreover, MeP decrease iNOS-dependent nitric oxide (NO) production. Legend: ↑ -increase, ↓ -decrease, ↔ no change.