HEMA-induced oxidative stress inhibits NF-κB nuclear translocation and TNF release from LTA- and LPS-stimulated immunocompetent cells



HEMA-induced oxidative stress inhibits NF-κB nuclear translocation and TNF release from LTA- and LPS-stimulated immunocompetent cells




Dental Materials, 2021-01-01, Volume 37, Issue 1, Pages 175-190, Copyright © 2020 The Academy of Dental Materials


Abstract

Objective: The release of inflammatory cytokines from antigen-stimulated cells of the immune system is inhibited by resin monomers such as 2-hydroxyethyl methacrylate (HEMA). Although the formation of oxidative stress in cells exposed to HEMA is firmly established, the mechanism behind the inhibited cytokine secretion is only partly known. The present investigation presents evidence regarding the role of HEMA-induced oxidative stress in the secretion of the pro-inflammatory cytokine TNFα from cells exposed to the antigens LTA (lipoteichoic acid) or LPS (lipopolysaccharide) of cariogenic microorganisms using BSO (L-buthionine sulfoximine) or NAC (N-acetyl cysteine) to inhibit or stabilize the amounts of the antioxidant glutathione.

Method: RAW264.7 mouse macrophages were treated with LTA, LPS or HEMA in the presence of BSO or NAC for 1h or 24h. Secretion of TNFα from cell cultures was analyzed by ELISA, and the formation of reactive oxygen (ROS) or nitrogen species (RNS) was determined by flow cytometry. Protein expression was detected by Western blotting.

Results: The release of TNFα in both LTA- and LPS-exposed cells was decreased by HEMA, and this concentration-dependent inhibitory effect was amplified by BSO or NAC. LTA- and LPS-stimulated expression of the redox-sensitive transcription factor NF-αB (p65) in cell nuclei decreased in the presence of HEMA because the translocation of p65 from the cytosol was prevented by oxidative stress specifically increased by the monomer.

Conclusions: A disturbance of the cellular redox balance, particularly induced by HEMA, is a crucial factor in the inhibition of LTA- and LPS-stimulated signalling pathways leading to TNFα secretion.

Introduction

Dental caries is an infectious disease generated by a plethora of Gram-positive and Gram-negative oral microorganisms in diversified biofilms on tooth surfaces. Invasion of dentinal hard tissue by pathogenic bacteria or their compounds LPS (lipopolysaccharide) or LTA (lipoteichoic acid) across dentinal tubules triggers the activation of cells of the innate immune system in the dental pulp complex [ , ]. While LPS is a glycolipid and a pathogenic molecule of Gram-negative cariogenic bacteria, LTA is a glycolipid connected to polyglycerolphosphate in the cell wall of Gram-positive microbes. As potent inflammatory PAMPs (pathogen-associated molecular patterns), LPS and LTA bind to Toll-like receptors (TLR) on the surface of cells of the dental pulp, such as odontoblasts at the dentin-pulp interface, and tissue resident immune cells such as fibroblasts, dendritic cells and macrophages. Binding of LPS to TLR4 and LTA to TLR2/TLR6 receptors triggers signaling cascades through MyD88-dependent pathways, finally activating the redox-sensitive transcription factor NF-κB (nuclear factor-κB) or MAPKs, and resulting in the production and release of pro- or anti-inflammatory cytokines [ ]. Although the immediate cell response to LPS or LTA exposure varies depending on the cell type and distinct signaling components, both molecules stimulate the release of cytokines such as TNFα in macrophages [ , ]. The release of cytokines as a result of signaling through NF-κB is positively or negatively regulated by reactive oxygen (ROS) or nitrogen species (RNS) such as hydrogen peroxide or nitric oxide (NO), and a related overabundance of reactive molecules. The formation of ROS and RNS under proinflammatory conditions, in turn, is mostly a result of the enhanced NFκB-regulated expression of Nox2 and iNOS enzymes [ , ]. Therefore, it seems as if LPS- or LTA-stimulated activation of NFκ-B and the formation of inflammatory cytokines was the cause and effect of oxidative stress [ , ]. Downregulation of LPS- or LTA-stimulated inflammatory processes, on the other hand, is related to the NFκB- and stress-induced expression of the antioxidant protein HO-1 (hemeoxygenase 1), which protects cells from ROS and RNS [ , , ]. In addition to NFκB, the expression of HO-1 is directly controlled by the transcription factor Nrf2 (nuclear factor erythroid 2 [NF-E2]-related factor 2), which is a master regulator of cellular redox homeostasis. Consequently, HO-1 is considered to be the mediator of an NFκB-Nrf2 crosstalk [ , ].

As a basic mechanism, dental resin monomers such as TEGDMA or HEMA induce intracellular oxidative stress by the enhanced formation of ROS and RNS as repeatedly shown [ ]. It is also firmly established that HO-1, catalase and related enzymes are activated under the control of Nrf2 as an adaptive response in order to reestablish cellular redox homeostasis and vital cell functions [ ]. Monomer-induced oxidative stress beyond the capacities of the Nrf2-coordinated antioxidant network results in cytotoxic effects via apoptosis, at least in part as a consequence of oxidative DNA damage [ , ]. The crucial importance of Nrf2 activity for maintaining redox homeostasis and cell viability was impressively demonstrated through its stimulation by the artificial compound tBHQ ( tert -butylhydroquinone) [ ]. It has also been suggested that HEMA might activate Nrf2 via its electrophilic properties [ ]. Still, using tBHQ as an inducer of Nrf2 activation also provided evidence that monomer-induced oxidative stress was also associated with the inhibition of the release of pro- and anti-inflammatory cytokines from LPS-stimulated immunecompetent macrophages [ ].

It has repeatedly been shown that the release of TNFα, IL-6 and IL-10 from LPS-stimulated cells was inhibited in the presence of increasing concentrations of HEMA [ , ]. The mechanism behind this is the inhibition of the LPS-stimulated translocation of NF-κB from the cytosol to the cell nucleus in the presence of the monomer [ ]. Analysis of the formation of reactive oxygen and nitrogen species suggested that molecules such as H 2 O 2 and NO enhanced in HEMA-treated cells prevented the activation of NF-κB and subsequent cytokine production [ ]. Therefore, in the present investigation we analyzed in more detail the role of oxidative stress in the inhibition of the release of cytokines relevant for vital immune responses to invading cariogenic pathogens by resin monomers. To this end, we modified the capacity of cells to control the intracellular redox balance through the modification of the concentration of glutathione (GSH) as a major non-enzymatic antioxidant. We have previously shown that oxidative stress increased when GSH synthesis was inhibited by buthionine sulfoximine (BSO), but decreased when GSH levels were maintained at high levels in the presence of N -acetyl cysteine (NAC), a substrate in GSH synthesis [ ]. Here, the effects of LPS and LTA on cytokine secretion, formation of ROS and RNS, as well as expression levels of related proteins including the redox-sensitive transcription factors NF-κB and Nrf2 were investigated. These parameters have not been analyzed thus far in cells treated with LTA. The secretion of the cytokine TNFα from RAW264.7 mouse macrophages was used as a reliable and approved model of immune cell function as previously shown [ ]

Materials and methods

Reagents and antibodies

2-Hydroxyethyl methacrylate (HEMA; CAS-No. 868-779) was obtained from Merck (Darmstadt, Germany). RPMI 1640 medium ( l -glutamine, 2.0 g/l NaHCO 3 ) was purchased from PAN Biotech (Aidenbach, Germany). Fetal bovine serum (FBS) was obtained from Life Technologies, Gibco BRL (Eggenstein, Germany). Lipopolysaccharide (LPS; E. coli , serotype 055:B5), dihydroethidium (DHE), lipoteichoic acid from Staphylococcus aureus (LTA; CAS-No. 56411-57-5) dihydrorhodamine 123 (DHR123), l -buthionine sulfoximine (BSO; CAS-No. 83730-53-4), N -acetyl cysteine (NAC; CAS-No. 616-91-1), 3-(4,5 dimethyiazol-2-1)-2-5-diphenyl tetrazolium bromide (MTT), crystal violet (C0775), and a bicinchoninic acid protein assay kit came from Sigma (Taufkirchen, Germany). 2′7′-dichlorodihydrofluorescin diacetate (DCFH 2 -DA; CAS-No. 4091-99-0) was obtained from MoBiTec (Göttingen, Germany). Amido black 10B was obtained from Merck (Darmstadt, Germany). Anti-catalase (H-300, sc-50508), anti-heme oxygenase 1 (HO-1, M-19, sc-1797), anti-Nrf2 (H-300, sc-13032), anti-p47-phox (sc-17845) monoclonal antibodies, and a protease inhibitor cocktail were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Anti-NF-κB p65 (no. 6956), anti-phospho-NF-κB p65 (no. 3033), anti-iNOS (D6B6S, no. 3120), plus anti-rabbit IgG HRP-linked antibodies (no. 7074) came from cell signaling (NEB Frankfurt, Germany). Anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) monoclonal antibody (clone 6C5) was obtained from Millipore (Schwalbach, Germany).

Goat anti-mouse IgG (H + L)-HRP conjugate was purchased from Bio-Rad Laboratories (Munich, Germany), and Amersham hyperfilm ECL came from GE Healthcare (Munich, Germany). Cell Meter™ fluorimetric intracellular peroxynitrite (no. 16317) and Cell Meter™ fluorimetric intracellular nitric oxide (NO) (no. 16356) assay kits were purchased from AAT Bioquest (Sunnyvale, CA, USA). The protease inhibitor cocktail (complete mini) was obtained from Roche Diagnostics (Mannheim, Germany), and a tumor necrosis factor-α (TNF-α) ELISA kit was obtained from BD Biosciences (San Diego, CA, USA).

Cell culture

The murine RAW264.7 mouse macrophages cell line (ATCC TIB71) was maintained in RPMI 1640 medium supplemented with l -glutamine, sodium-pyruvate, 2.0 g/l NaHCO 3 and 10% fetal bovine serum (FBS) following our standard culture techniques as previously described [ ].

Determination of TNFα

Cells (5 × 10 4 cells/well) from routine culture were incubated in RPMI 1640 medium in 6-well plates at 37 °C for 48 h. For the analysis of concentrations of LPS or LTA effective in the release of tumor necrosis factor-α (TNFα) from RAW264.7 mouse macrophages, cell cultures were exposed to LTA or LPS in the range of 0–50 μg/ml in 2 ml culture media. Next, cell cultures were treated with 50 μM BSO (L-buthionine sulfoximine) or 10 mM NAC ( N -acetyl cysteine) to analyze for the role of glutathione in the release of TNFα from HEMA-exposed cells. To this end, cell cultures were initially preincubated with 50 μM BSO for 18 h or with 10 mM NAC ( N -acetyl cysteine) for 1 h. Then, preincubation media were removed and cells were exposed to HEMA (0–1–4–8 mM), both in the presence or absence of LTA (10 μg/ml) or LPS (0.1 μg/ml), or 50 μM BSO or 10 mM NAC in 2 ml culture media as described [ ]. Cell culture supernatants were collected after 1 h or 24 h exposure periods and analyzed for the formation of tumor necrosis factor-α (TNF-α). In addition, amounts of proteins present in adherent cells were analyzed using an Amido Black assay as described below (Section 2.4 ). BSO and NAC concentrations appropriate under the current experimental conditions were established in a previous investigation [ ]. Concentrations of LTA and LPS were chosen because of their effectiveness in the release of TNF-α observed in range finding experiments shown in Fig. 1 .

Release of TNFα from RAW264.7 mouse macrophages. Cell cultures were exposed to a wide range of concentrations of lipoteichoic acid (LTA) (A, B) or lipopolysacharide (LPS) (C, D) for 1 h or 24 h. Levels of TNFα were detected by ELISA as described in Section 2 . Dashed lines indicate the TNFα detection limit. Bars represent median values (25% and 75% percentiles) combined from duplicates in five independent experiments (n = 10). Cell survival (E, F) was determined using a crystal violet assay. * Indicates significant differences of median values between LTA- or LPS-treated cultures and untreated controls. o Indicates significant differences of median values between LTA- or LPS-exposed cultures.
Fig. 1
Release of TNFα from RAW264.7 mouse macrophages. Cell cultures were exposed to a wide range of concentrations of lipoteichoic acid (LTA) (A, B) or lipopolysacharide (LPS) (C, D) for 1 h or 24 h. Levels of TNFα were detected by ELISA as described in Section
2 . Dashed lines indicate the TNFα detection limit. Bars represent median values (25% and 75% percentiles) combined from duplicates in five independent experiments (n = 10). Cell survival (E, F) was determined using a crystal violet assay. * Indicates significant differences of median values between LTA- or LPS-treated cultures and untreated controls.
o Indicates significant differences of median values between LTA- or LPS-exposed cultures.

Amounts of TNF-α (pg/ml) were determined using a standard ELISA kit (BD Pharmingen). Absorbance was read at 450 nm in a spectrophotometer (Infinite F200, TECAN, Mainz, Germany), and optical density values were collected (Magellan software; version 6.2). Levels of TNF-α in untreated cell cultures below the detection limit were set to the lowest detectable levels (15.6 pg/ml) in standard curves. Absolute amounts of TNFα were summarized from duplicate values in 4–5 independent experiments as specified in the figure legend.

Determination of protein concentration

Adherent cells in 6-well plates employed for the analysis of TNF-α release in cultures treated with HEMA and LTA/LPS were washed with phosphate-buffered saline (PBS/5 mM EDTA), and then lysed in 500 μl RIPA lysis buffer/well for 10 min in the cold. The cell lysate was then transferred to an assay tube. Each well was washed with 400 μl RIPA lysis buffer and the buffer solution was then combined with the cell lysate. Protein concentration in each sample was then determined with Amido Black as described earlier [ ]. 500 μl of reagent (13 μg/ml Amido Black/ml, dissolved in 10% acetic acid, 90% methanol) was added to 80 μl of protein sample and incubated for 5 min at room temperature. Then, the solution was centrifuged at 16,000× g for 5 min, the supernatant discarded, the pellet washed in 500 μl of 10% acetic acid, 90% methanol, and again centrifuged at 16,000× g for 5 min. The washing procedure was repeated twice, and finally the pellet was dissolved in 400 μl of 0.1 N NaOH. An aliquot of 180 μl of each sample was individually transferred to wells in a 96-well plate, and the absorbance was determined at 615 nm in a spectrophotometer (Tecan). Absolute amounts of proteins were calculated from standard curves established with bovine serum albumin (0–30 μg).

Detection of reactive oxygen species (ROS)

Cell cultures (5 × 10 4 /well) were maintained in routine culture medium in 6-well plates for 48 h, and then preincubated with 50 μM BSO for 18 h, or with 10 mM NAC ( N -acetyl cysteine) for 1 h. Subsequently, cell cultures were treated with HEMA (0–1–4–8 mM), both in the presence or absence of LTA (10 μg/ml) or LPS (0.1 μg/ml), or 50 μM BSO or 10 mM NAC in 2 ml culture media [ ]. The exposure of cells was stopped by discarding the exposure media, and the formation of ROS was analyzed by flow cytometry after staining the cells with the fluorescent dyes 2′7′-dichlorodihydrofluorescin diacetate (DCFH 2 -DA), dihydroethidium (DHE) or dihydrorhodamine 123 (DHR123). It has been previously reported that DCFH 2 -DA indicated the formation of general oxidative stress, while DHE or DHR123 detected the formation of superoxide anions or hydrogen peroxide production as discussed in our recent investigation [ ]. Favorable properties of these dyes for the investigation of monomer-induced oxidative stress have been outlined in detail in our previous publication [ ]. Cells were incubated with 10 μM DCFH 2 -DA, 5 μM DHE or 5 μM DHR123 in culture medium 30 min prior to harvesting in phosphate-buffered saline (PBS)/5 mM EDTA. Then, the cells were collected by centrifugation and the cell pellet was resuspended in 200 μl CMF–PBS. DCF and DHR123 fluorescence was measured in the FL-1 channel (488/515–545) using a BD FACSCanto (Becton Dickinson) flow cytometer, and DHE fluorescence intensities were detected in Fl-2 (488/546–606). Mean fluorescence intensities (MFI) were established using histogram statistics and FACSDiva™ 5.0.2 (Becton Dickinson) software. Individual fluorescence intensities measured in treated cell cultures were related to fluorescence quantified in untreated control cultures (= 1.0) as described [ ].

Analysis of nitric oxide and peroxynitrite formation

The cells (5 × 10 4 /well) were grown in routine culture medium in 6-well plates for 48 h, and then preincubated with 50 μM BSO or 10 mM NAC ( N -acetyl cysteine), exposed to HEMA (0–1–4–8 mM) in the presence or absence of LPS (0.1 μg/ml) or LTA (10 μg/ml), or 50 μM BSO or 10 mM NAC as specifically described in Section 2.5 . Exposure of cells was stopped by discarding the exposure media, and cells were then harvested in phosphate-buffered saline (PBS)/5 mM EDTA, collected by centrifugation and resuspended in 200 μl CMF–PBS. The intracellular formation of nitric oxide (NO ) and peroxynitrite (ONOO) was determined using Cell Meter™ fluorimetric assay kits (AAT Bioquest). Peroxynitrite (ONOO ) or nitric oxide (NO) was examined by flow cytometry (FACSCanto) in channels FL-1 (488/519 nm, ONOO ) or FL-5 (633/660, NO) following the manufacturer’s instructions and as described elsewhere [ ]. Mean fluorescence intensities were determined, and individual fluorescence intensities normalized as described in Section 2.5 .

Analysis of protein expression by immunoblotting

Cells (1.5 × 10 6 cells) were grown in routine culture medium on plates (150 mm diameter) at 37 °C for 48 h. Next, cell cultures were preincubated with 50 μM BSO or 10 mM NAC ( N -acetyl cysteine), exposed to HEMA (0–1–4–8 mM) in the presence or absence of LPS (0.1 μg/ml) or LTA (10 μg/ml), or 50 μM BSO or 10 mM NAC in culture medium. Exposure of cells was stopped by discarding the exposure media. Floating and adherent cells were harvested in ice-cold PBS and combined after centrifugation. Nuclear and cytosolic cell fractions were prepared using different lysis buffers and various centrifugation procedures as described in our previous paper [ ]. Briefly, cell pellets were resuspended in 1 ml ice-cold PBS, mixed with 0.5 ml buffer A (10 mM Tris HCl, 60 mM KCl, 1 mM Na 2 EDTA, 1 mM DTT, pH 7.4), and incubated on ice. The cell suspension was centrifuged again, and the cell pellet was solubilized in buffer B (buffer A plus 0.4% NP 40, 5 mM NaF, 1 mM NaVO 4 ) in the cold. After centrifugation (14,000× g ) the supernatant was collected as a cytoplasmic cell fraction. The cell pellet containing cell nuclei in the pellet was washed in buffer A, resuspended in buffer C (20 mM Tris-Cl pH 8.0, 400 mM NaCl, 1.5 mM MgCl 2 , 1.5 mM Na 2 EDTA, 25% glycerol, 1 mM DTT, 5 mM NaF, 1 mM NaVO 4 ), and centrifugated (14,000× g ) again in the cold. The resulting supernatant was collected as a fraction of nuclear proteins. Protein amounts in the cytosolic and nuclear cell fractions were measured using a BCA protein assay.

Protein expression was identified by routine immunoblotting as previously described [ ]. Proteins (10–15 μg per lane) were first separated by SDS-PAGE and then transferred to a nitrocellulose membrane. After washing in TBS (25 mM Tris-Cl, 150 mM NaCl, pH 7.4) and blocking with 5% nonfat milk in TBST (TBS plus 0.1% Tween 20, pH 7.4) at room temperature, the membrane was incubated with primary antibodies to detect protein expression as specified in the legends of Figs. 4 and 7 . Primary antibodies bound to membranes were visualized by horseradish peroxidase-conjugated secondary antibodies and enhanced chemiluminescence (ECL) using a blot scanner and Image Studio software (LI-COR Biotechnology, Bad Homburg, Germany). Membranes were stripped with CHEMICON re-blot stripping solution (Millipore, Schwalbach, Germany) for reprobing.

Statistical analyses

Individual values from repeated independent experiments as specified in the figure legends were summarized as medians (25–75% quartiles). The Mann–Whitney U -test (SPSS Statistics 23, IBM, Armonk, NY, USA) with an alpha level of 0.05 was used to calculate statistically significant differences between median values.

Results

Effectiveness of LPS and LTA in the stimulation of TNFα secretion

The effectiveness of LTA (lipoteichoic acid) and LPS (lipopolysaccharide) in the triggering of innate immune responses was initially compared by the stimulation of TNFα secretion in RAW264.7 mouse macrophages through a wide range of LTA or LPS concentrations. While the amounts of TNFα detected in untreated cell cultures were close to the detection limit, both LTA and LPS were differentially effective in the stimulation of TNFα release depending on concentrations and the exposure period. LTA induced a concentration-dependent increase in TNFα release after 1 h or 24 h exposure periods. LTA concentrations in the range of 0.1–50 μg/ml significantly increased TNFα secretion more than 30-fold to about 200–300 pg/ml after a 1 h exposure, and TNFα release further increased about 3-fold when cells were exposed to high LTA concentration for a long exposure period (24 h) ( Fig. 1 A, B). LPS concentrations from 0.1 to 50 μg/ml significantly, and almost equally, increased TNFα secretion to about 500–600 pg/ml after a 1 h exposure, and TNFα release further increased about 5–10-fold after a long exposure period (24 h) ( Fig. 1 C, D).

Despite the clear influence on TNFα secretion, no influence on cell survival was detected with all LTA or LPS concentrations after a 1 h exposure. Cell survival was moderately reduced to about 70–75% by LPS (0.1–50 μg/ml), while no decrease in cell survival was observed with LTA (0.1–50 μg/ml) after a 24 h exposure ( Fig. 1 E, F). Concentrations of 10 μg/ml LTA and 0.1 μg/ml LPS were then selected to further investigate the influence of the monomer HEMA on LPS- or LTA-stimulated TNFα secretion.

LTA-stimulated secretion of TNFα and oxidative stress conditions

LTA-stimulated secretion of TNFα

No increase in the amounts of TNFα set to the detection limit in untreated cell cultures was detected after the exposure of cells to increasing HEMA concentrations for 1 h or 24 h. On the contrary, the considerable increase in the secretion of absolute amounts of 374 pg/ml or 465 pg/ml TNFα found in cultures stimulated with 10.0 μg/ml LTA for 1 h or 24 h was significantly inhibited in the presence of HEMA depending on its concentration ( Fig. 2 A, B). Both buthionine sulfoximine (BSO) and N -acetyl cysteine (NAC) were ineffective alone, but decreased LTA-stimulated TNFα secretion and affected the inhibitory effectiveness of HEMA as presented in detail below ( Fig. 2 A, B).

Release of TNFα from LTA-stimulated RAW264.7 mouse macrophages. Cell cultures were preincubated with 50 μM BSO for 18 h, or with 10 mM NAC for 1 h and then treated with LTA (10 μg/ml) or HEMA (0–1–4–8 mM) for 1 h or 24 h with or without 50 μM BSO or 10 mM NAC (A, B). Absolute amounts of TNFα detected in culture supernatants are presented in panels A and B. Amounts of proteins detected after amido black staining of lysed adherent cells are shown in panels C and D, and levels of TNFα related to protein content are presented in panels E and F. Bars show medians (25% and 75% percentiles) summarized from duplicates obtained from 4 to 5 independent experiments (n = 8–10).
Fig. 2
Release of TNFα from LTA-stimulated RAW264.7 mouse macrophages. Cell cultures were preincubated with 50 μM BSO for 18 h, or with 10 mM NAC for 1 h and then treated with LTA (10 μg/ml) or HEMA (0–1–4–8 mM) for 1 h or 24 h with or without 50 μM BSO or 10 mM NAC (A, B). Absolute amounts of TNFα detected in culture supernatants are presented in panels A and B. Amounts of proteins detected after amido black staining of lysed adherent cells are shown in panels C and D, and levels of TNFα related to protein content are presented in panels E and F. Bars show medians (25% and 75% percentiles) summarized from duplicates obtained from 4 to 5 independent experiments (n = 8–10).
Significant differences between median values of amounts of TNF, protein or TNF/protein ratios found in the various experimental conditions are as follows:
a = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of LTA.
b = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of LTA and BSO.
c = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of LTA and NAC.
d = Significant differences between median values obtained in cultures treated with LTA and HEMA and cultures exposed to HEMA in the presence of LTA and BSO.
e = Significant differences between median values obtained in cultures treated with LTA and HEMA and cultures exposed to HEMA in the presence of LTA and NAC.
f = Significant differences between median values obtained in cultures treated with LTA, HEMA and BSO and cultures exposed to HEMA in the presence of LTA and NAC.
Other statistical analyses are not shown for reasons of clarity.
Significant differences between median values calculated from cultures treated with LTA and cultures exposed to LTA in the presence of 1, 4, or 8 mM HEMA are shown by asterisks (*).

In order to normalize the absolute amounts of TNFα secreted under the various experimental conditions, we determined the amounts of protein present in exposed cell cultures ( Fig. 2 C, D). No significant influence of HEMA concentrations (0–8 mM) was found on the amounts of protein (60–84 μg) after a 1 h exposure period. While levels of proteins were only slightly reduced in cultures exposed to LTA in the presence of 4 or 8 mM HEMA, a decrease in protein content from about 84 μg (0 mM HEMA) to about 55–70 μg following a consistent pattern was observed in cultures treated with NAC, both in the presence or absence of LTA and HEMA ( Fig. 2 C). After a 24 h exposure the amount of protein increased in untreated cultures almost 4-fold to around 320 μg and was gradually decreased to 104 μg in cultures treated with 8 mM HEMA only ( Fig. 2 D). Noteworthy is that the protein levels increased to 134 μg in cultures co-treated with 8 mM HEMA and LTA compared to those exposed to 8 mM HEMA only. A continuous decrease in protein levels was observed with cultures treated with BSO in the presence or absence of LTA and HEMA down to about 50 μg in cells exposed to 8 mM HEMA. In comparison, 10 mM NAC alone reduced the amount of protein to about 190 μg, but unlike the effect of BSO, NAC increased protein levels when cell cultures were treated with a high concentration HEMA (8 mM) ( Fig. 2 D).

Related to these amounts of proteins, LTA-stimulated levels of TNFα decreased more than 3-fold from 9.2 pg LTA/μg protein after a 1 h exposure period to 2.8 pg LTA/μg protein after a 24 h exposure. Yet, secretion of TNFα was drastically and equally reduced with increasing HEMA concentrations. For instance, 2.8 pg LTA/μg protein detected in LTA-stimulated cultures were reduced to 1.1 and 0.4 or 0.3 pg LTA/μg protein in the presence of 1, 4 or 8 mM HEMA after a 24 h exposure. A similar pattern of TNFα secretion was detected after a 1 h exposure ( Fig. 2 E, F). The concentrations of BSO and NAC chosen here were highly effective in the inhibition the LTA-stimulated TNFα secretion after a 24 h exposure in particular. For instance, levels of TNFα in LTA-exposed cells decreased from 2.8 pg/μg protein to 1.5 (BSO) or 0.2 (NAC) pg/μg protein ( Fig. 2 F). Both substances seemed to intensify the inhibitory effectiveness of HEMA after long exposure (24 h), although no such effect was detectable with BSO after a 1 h exposure ( Fig. 2 E, F).

LTA-stimulated oxidative and nitrosative stress

The influence of BSO and NAC on the release of TNFα from LTA- and HEMA-stimulated cells after long exposure in particular suggested a crucial role of GSH and the production of ROS or RNS. Thus, the formation of oxidative stress, nitric oxide (NO) and peroxynitrite was further analyzed. Oxidative stress indicated by DCF fluorescence significantly increased in RAW264.7 mouse macrophages after exposure to increasing concentrations of HEMA for 1 h ( Fig. 3 A). A 1.3-fold increase compared to untreated cell cultures (= 1.0) was detected with 8 mM HEMA, and DCF fluorescence further increased more than 2−3-fold in the presence of BSO in cultures co-treated with HEMA ( Fig. 3 A). This effect of BSO was not detectibly influenced by 10 μg/ml LTA. LTA, however, slightly reduced DCF fluorescence alone and in the presence of HEMA. Furthermore, DCF fluorescence was also significantly reduced by NAC in cultures treated with HEMA, and this inhibitory effect was reproducibly intensified by LTA ( Fig. 3 A). Most relevant, a 5-fold increase in LTA-stimulated DCF-fluorescence after a 24 h exposure was continuously and significantly decreased by HEMA. This pattern of LTA-induced DCF fluorescence and inhibition in the presence of HEMA was drastically increased in the presence of BSO. LTA-stimulated DCF-fluorescence was considerably reduced in the presence of NAC, which then further intensified the inhibitory effect of HEMA ( Fig. 3 B)

Formation of oxidative stress, nitric oxide and peroxynitrite in RAW264.7 mouse macrophages. Cell cultures were first preincubated with 50 μM BSO for 18 h, or with 10 mM NAC for 1 h and then stimulated with LTA (10 μg/ml) and HEMA (0–1–4–8 mM) in the presence or absence of 50 μM BSO or 10 mM NAC. Formation of oxidative stress was detected after staining cells with DCFH 2 (A, B), and production of nitric oxide (NO) (C, D) or peroxynitrite (E, F) after 1 h or 24 h exposure periods was analyzed by flow cytometry. Bars show medians (25% and 75% percentiles) summarized from individual values in independent experiments (n = 4–5).
Fig. 3
Formation of oxidative stress, nitric oxide and peroxynitrite in RAW264.7 mouse macrophages. Cell cultures were first preincubated with 50 μM BSO for 18 h, or with 10 mM NAC for 1 h and then stimulated with LTA (10 μg/ml) and HEMA (0–1–4–8 mM) in the presence or absence of 50 μM BSO or 10 mM NAC. Formation of oxidative stress was detected after staining cells with DCFH
2 (A, B), and production of nitric oxide (NO) (C, D) or peroxynitrite (E, F) after 1 h or 24 h exposure periods was analyzed by flow cytometry. Bars show medians (25% and 75% percentiles) summarized from individual values in independent experiments (n = 4–5).
Significant differences between median values of DCF, NO or ONOO fluorescence are as follows:
a = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of LTA.
b = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of BSO.
c = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of LTA and BSO.
d = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of NAC.
e = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of LTA and NAC.
f = Significant differences between median values obtained in cultures treated with LTA and HEMA and cultures exposed to HEMA in the presence of LTA and BSO.
g = Significant differences between median values obtained in cultures treated with LTA and HEMA and cultures exposed to HEMA in the presence of LTA and NAC.
Other statistical analyses are not shown for reasons of clarity.
Significant differences between median values calculated from cultures treated with LPS and cultures exposed to LPS in the presence of 1, 4, or 8 mM HEMA are shown by asterisks (*).

Levels of nitric oxide (NO) were reduced to 0.9 in the presence of LTA compared to untreated cell cultures (= 1.0), and this effect was slightly but not significantly enhanced by 1 or 4 mM HEMA after a short exposure period (1 h) ( Fig. 3 C). While no influence of HEMA on the formation of NO · was detected in the presence or absence of BSO, NAC reduced the amounts of NO in control cultures and cells exposed to a low concentration of HEMA ( Fig. 3 C). After long exposure (24 h), LTA stimulated the formation of NO about 2.7-fold compared to untreated controls. This effect was reduced by 1 and 4 mM HEMA, but increased again with 8 mM HEMA to levels induced by LTA alone. HEMA itself significantly and continuously increased the formation of NO up to 3-fold compared to untreated cell cultures. The presence of BSO, which was ineffective alone and on LTA, even intensified the increase in NO formation in cell cultures treated with HEMA. In contrast, NAC reduced both LTA-stimulated and HEMA-induced production of NO ( Fig. 3 D). The levels of peroxynitrite (ONOO ) [ ], as a product of superoxide anion and nitric oxide, were slightly enhanced only in cultures treated with BSO for a short period of time ( Fig. 3 E, F). In contrast, ONOO formation was enhanced about 2-fold in cells stimulated with LTA compared to untreated cultures after a 24 h exposure. This LTA-induced effect, however, was moderately but significantly reduced with increasing concentrations of HEMA ( Fig. 3 F). The formation of peroxynitrite by LTA was independent of BSO, but BSO alone enhanced the ONOO levels about 1.3-fold, and it considerably (2-fold) intensified a slight increase in ONOO levels caused by HEMA. NAC reduced LTA-stimulated and HEMA-induced production of ONOO ( Fig. 3 F).

LTA-stimulated expression of proteins related to TNFα release and oxidative or nitrosative stress

Expression of proteins directly related to oxidative and nitrosative stress was detected in cell cultures exposed to LTA and HEMA for 24 h. The redoxsensitive transcription factor Nrf2 was expressed in the cytosol of untreated cell cultures, and Nrf2 expression was also detected in the nucleus in cells exposed to HEMA. Moreover, high levels of Nrf2, as detected in the nucleus of cells exposed to 10 μg/ml LTA possibly after translocation from the cytosol to the cell nucleus, were decreased in the presence of HEMA. Most relevant, expression of Nrf2 in the cell nucleus was drastically decreased in LTA-stimulated cultures co-exposed to the antioxidant NAC, while nuclear Nrf2 expression was moderately reduced in the presence of BSO ( Fig. 4 ). The expression of the transcription factor NF-κB, as indicated by subunit p65, was also stimulated in nuclei of cells treated with 10 μg/ml LTA, and these increased levels of NF-κB were reduced in the presence of HEMA. LTA-stimulated NF-κB expression also slightly decreased in the presence of NAC, as shown by the enhanced NF-κB levels in the cytosolic fraction compared to cultures treated with LTA alone. Although BSO alone strongly increased the expression of p65, the induction of nuclear NF-κB expression by LTA was even lower in cultures pretreated with BSO ( Fig. 4 ). Both HEMA and LTA induced expression of the stress protein HO-1, which was further enhanced in the presence of BSO. In contrast, co-treatment with NAC reduced HO-1 expression to low levels in cell cultures exposed to LTA or HEMA.

Protein expression in RAW264.7 mouse macrophages. Cell cultures were first preincubated with 50 μM BSO for 18 h, or with 10 mM NAC for 1 h and then stimulated with LTA (10 μg/ml) and HEMA (0–1–4–8 mM) in the presence or absence of 50 μM BSO or 10 mM NAC. Protein expression was analyzed by Western blotting. Expression of Nrf2 and subunit p65 of NF-κB was detected in the cytosol (C) as well as in cell nuclei (N), and subunit p47 phox of Nox2, inducible nitric oxide synthase (iNOS), and heme oxyganase1 (HO-1) were identified in the cytosol. Equal loading of proteins across the gel was controlled by the expression of lamin A/C or GAPDH as markers for cell nuclei or the cytosol, respectively (not shown).
Fig. 4
Protein expression in RAW264.7 mouse macrophages. Cell cultures were first preincubated with 50 μM BSO for 18 h, or with 10 mM NAC for 1 h and then stimulated with LTA (10 μg/ml) and HEMA (0–1–4–8 mM) in the presence or absence of 50 μM BSO or 10 mM NAC. Protein expression was analyzed by Western blotting. Expression of Nrf2 and subunit p65 of NF-κB was detected in the cytosol (C) as well as in cell nuclei (N), and subunit p47
phox of Nox2, inducible nitric oxide synthase (iNOS), and heme oxyganase1 (HO-1) were identified in the cytosol. Equal loading of proteins across the gel was controlled by the expression of lamin A/C or GAPDH as markers for cell nuclei or the cytosol, respectively (not shown).

Expression of the inducible form of nitric oxide synthase (iNOS) was extremely low or absent under the current experimental conditions in cultures exposed to HEMA, NAC or BSO. Notably, a drastically increased expression of iNOS in LTA-stimulated cultures was effectively inhibited in the presence of HEMA. While LTA alone even lowered p47 expression, a remarkable increase in the basic expression of p47, as a regulatory subunit of NADPH oxidase, was detected in cells exposed to NAC. An increase in catalase expression in cell cultures exposed to HEMA was moderately reduced in the presence of LTA independent of the presence of NAC or BSO ( Fig. 4 ).

LPS-stimulated secretion of TNFα and oxidative stress conditions

You're Reading a Preview

Become a DentistryKey membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here

Was this article helpful?

HEMA-induced oxidative stress inhibits NF-κB nuclear translocation and TNF release from LTA- and LPS-stimulated immunocompetent cells Helmut Schweikl , Margaritha Birke , Marialucia Gallorini , Christine Petzel , Carola Bolay , Claudia Waha , Karl-Anton Hiller and Wolfgang Buchalla Dental Materials, 2021-01-01, Volume 37, Issue 1, Pages 175-190, Copyright © 2020 The Academy of Dental Materials Abstract Objective: The release of inflammatory cytokines from antigen-stimulated cells of the immune system is inhibited by resin monomers such as 2-hydroxyethyl methacrylate (HEMA). Although the formation of oxidative stress in cells exposed to HEMA is firmly established, the mechanism behind the inhibited cytokine secretion is only partly known. The present investigation presents evidence regarding the role of HEMA-induced oxidative stress in the secretion of the pro-inflammatory cytokine TNFα from cells exposed to the antigens LTA (lipoteichoic acid) or LPS (lipopolysaccharide) of cariogenic microorganisms using BSO (L-buthionine sulfoximine) or NAC (N-acetyl cysteine) to inhibit or stabilize the amounts of the antioxidant glutathione. Method: RAW264.7 mouse macrophages were treated with LTA, LPS or HEMA in the presence of BSO or NAC for 1h or 24h. Secretion of TNFα from cell cultures was analyzed by ELISA, and the formation of reactive oxygen (ROS) or nitrogen species (RNS) was determined by flow cytometry. Protein expression was detected by Western blotting. Results: The release of TNFα in both LTA- and LPS-exposed cells was decreased by HEMA, and this concentration-dependent inhibitory effect was amplified by BSO or NAC. LTA- and LPS-stimulated expression of the redox-sensitive transcription factor NF-αB (p65) in cell nuclei decreased in the presence of HEMA because the translocation of p65 from the cytosol was prevented by oxidative stress specifically increased by the monomer. Conclusions: A disturbance of the cellular redox balance, particularly induced by HEMA, is a crucial factor in the inhibition of LTA- and LPS-stimulated signalling pathways leading to TNFα secretion. 1 Introduction Dental caries is an infectious disease generated by a plethora of Gram-positive and Gram-negative oral microorganisms in diversified biofilms on tooth surfaces. Invasion of dentinal hard tissue by pathogenic bacteria or their compounds LPS (lipopolysaccharide) or LTA (lipoteichoic acid) across dentinal tubules triggers the activation of cells of the innate immune system in the dental pulp complex [ , ]. While LPS is a glycolipid and a pathogenic molecule of Gram-negative cariogenic bacteria, LTA is a glycolipid connected to polyglycerolphosphate in the cell wall of Gram-positive microbes. As potent inflammatory PAMPs (pathogen-associated molecular patterns), LPS and LTA bind to Toll-like receptors (TLR) on the surface of cells of the dental pulp, such as odontoblasts at the dentin-pulp interface, and tissue resident immune cells such as fibroblasts, dendritic cells and macrophages. Binding of LPS to TLR4 and LTA to TLR2/TLR6 receptors triggers signaling cascades through MyD88-dependent pathways, finally activating the redox-sensitive transcription factor NF-κB (nuclear factor-κB) or MAPKs, and resulting in the production and release of pro- or anti-inflammatory cytokines [ ]. Although the immediate cell response to LPS or LTA exposure varies depending on the cell type and distinct signaling components, both molecules stimulate the release of cytokines such as TNFα in macrophages [ , ]. The release of cytokines as a result of signaling through NF-κB is positively or negatively regulated by reactive oxygen (ROS) or nitrogen species (RNS) such as hydrogen peroxide or nitric oxide (NO), and a related overabundance of reactive molecules. The formation of ROS and RNS under proinflammatory conditions, in turn, is mostly a result of the enhanced NFκB-regulated expression of Nox2 and iNOS enzymes [ , ]. Therefore, it seems as if LPS- or LTA-stimulated activation of NFκ-B and the formation of inflammatory cytokines was the cause and effect of oxidative stress [ , ]. Downregulation of LPS- or LTA-stimulated inflammatory processes, on the other hand, is related to the NFκB- and stress-induced expression of the antioxidant protein HO-1 (hemeoxygenase 1), which protects cells from ROS and RNS [ , , ]. In addition to NFκB, the expression of HO-1 is directly controlled by the transcription factor Nrf2 (nuclear factor erythroid 2 [NF-E2]-related factor 2), which is a master regulator of cellular redox homeostasis. Consequently, HO-1 is considered to be the mediator of an NFκB-Nrf2 crosstalk [ , ]. As a basic mechanism, dental resin monomers such as TEGDMA or HEMA induce intracellular oxidative stress by the enhanced formation of ROS and RNS as repeatedly shown [ ]. It is also firmly established that HO-1, catalase and related enzymes are activated under the control of Nrf2 as an adaptive response in order to reestablish cellular redox homeostasis and vital cell functions [ ]. Monomer-induced oxidative stress beyond the capacities of the Nrf2-coordinated antioxidant network results in cytotoxic effects via apoptosis, at least in part as a consequence of oxidative DNA damage [ , ]. The crucial importance of Nrf2 activity for maintaining redox homeostasis and cell viability was impressively demonstrated through its stimulation by the artificial compound tBHQ ( tert -butylhydroquinone) [ ]. It has also been suggested that HEMA might activate Nrf2 via its electrophilic properties [ ]. Still, using tBHQ as an inducer of Nrf2 activation also provided evidence that monomer-induced oxidative stress was also associated with the inhibition of the release of pro- and anti-inflammatory cytokines from LPS-stimulated immunecompetent macrophages [ ]. It has repeatedly been shown that the release of TNFα, IL-6 and IL-10 from LPS-stimulated cells was inhibited in the presence of increasing concentrations of HEMA [ , ]. The mechanism behind this is the inhibition of the LPS-stimulated translocation of NF-κB from the cytosol to the cell nucleus in the presence of the monomer [ ]. Analysis of the formation of reactive oxygen and nitrogen species suggested that molecules such as H 2 O 2 and NO enhanced in HEMA-treated cells prevented the activation of NF-κB and subsequent cytokine production [ ]. Therefore, in the present investigation we analyzed in more detail the role of oxidative stress in the inhibition of the release of cytokines relevant for vital immune responses to invading cariogenic pathogens by resin monomers. To this end, we modified the capacity of cells to control the intracellular redox balance through the modification of the concentration of glutathione (GSH) as a major non-enzymatic antioxidant. We have previously shown that oxidative stress increased when GSH synthesis was inhibited by buthionine sulfoximine (BSO), but decreased when GSH levels were maintained at high levels in the presence of N -acetyl cysteine (NAC), a substrate in GSH synthesis [ ]. Here, the effects of LPS and LTA on cytokine secretion, formation of ROS and RNS, as well as expression levels of related proteins including the redox-sensitive transcription factors NF-κB and Nrf2 were investigated. These parameters have not been analyzed thus far in cells treated with LTA. The secretion of the cytokine TNFα from RAW264.7 mouse macrophages was used as a reliable and approved model of immune cell function as previously shown [ ] 2 Materials and methods 2.1 Reagents and antibodies 2-Hydroxyethyl methacrylate (HEMA; CAS-No. 868-779) was obtained from Merck (Darmstadt, Germany). RPMI 1640 medium ( l -glutamine, 2.0 g/l NaHCO 3 ) was purchased from PAN Biotech (Aidenbach, Germany). Fetal bovine serum (FBS) was obtained from Life Technologies, Gibco BRL (Eggenstein, Germany). Lipopolysaccharide (LPS; E. coli , serotype 055:B5), dihydroethidium (DHE), lipoteichoic acid from Staphylococcus aureus (LTA; CAS-No. 56411-57-5) dihydrorhodamine 123 (DHR123), l -buthionine sulfoximine (BSO; CAS-No. 83730-53-4), N -acetyl cysteine (NAC; CAS-No. 616-91-1), 3-(4,5 dimethyiazol-2-1)-2-5-diphenyl tetrazolium bromide (MTT), crystal violet (C0775), and a bicinchoninic acid protein assay kit came from Sigma (Taufkirchen, Germany). 2′7′-dichlorodihydrofluorescin diacetate (DCFH 2 -DA; CAS-No. 4091-99-0) was obtained from MoBiTec (Göttingen, Germany). Amido black 10B was obtained from Merck (Darmstadt, Germany). Anti-catalase (H-300, sc-50508), anti-heme oxygenase 1 (HO-1, M-19, sc-1797), anti-Nrf2 (H-300, sc-13032), anti-p47-phox (sc-17845) monoclonal antibodies, and a protease inhibitor cocktail were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Anti-NF-κB p65 (no. 6956), anti-phospho-NF-κB p65 (no. 3033), anti-iNOS (D6B6S, no. 3120), plus anti-rabbit IgG HRP-linked antibodies (no. 7074) came from cell signaling (NEB Frankfurt, Germany). Anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) monoclonal antibody (clone 6C5) was obtained from Millipore (Schwalbach, Germany). Goat anti-mouse IgG (H + L)-HRP conjugate was purchased from Bio-Rad Laboratories (Munich, Germany), and Amersham hyperfilm ECL came from GE Healthcare (Munich, Germany). Cell Meter™ fluorimetric intracellular peroxynitrite (no. 16317) and Cell Meter™ fluorimetric intracellular nitric oxide (NO) (no. 16356) assay kits were purchased from AAT Bioquest (Sunnyvale, CA, USA). The protease inhibitor cocktail (complete mini) was obtained from Roche Diagnostics (Mannheim, Germany), and a tumor necrosis factor-α (TNF-α) ELISA kit was obtained from BD Biosciences (San Diego, CA, USA). 2.2 Cell culture The murine RAW264.7 mouse macrophages cell line (ATCC TIB71) was maintained in RPMI 1640 medium supplemented with l -glutamine, sodium-pyruvate, 2.0 g/l NaHCO 3 and 10% fetal bovine serum (FBS) following our standard culture techniques as previously described [ ]. 2.3 Determination of TNFα Cells (5 × 10 4 cells/well) from routine culture were incubated in RPMI 1640 medium in 6-well plates at 37 °C for 48 h. For the analysis of concentrations of LPS or LTA effective in the release of tumor necrosis factor-α (TNFα) from RAW264.7 mouse macrophages, cell cultures were exposed to LTA or LPS in the range of 0–50 μg/ml in 2 ml culture media. Next, cell cultures were treated with 50 μM BSO (L-buthionine sulfoximine) or 10 mM NAC ( N -acetyl cysteine) to analyze for the role of glutathione in the release of TNFα from HEMA-exposed cells. To this end, cell cultures were initially preincubated with 50 μM BSO for 18 h or with 10 mM NAC ( N -acetyl cysteine) for 1 h. Then, preincubation media were removed and cells were exposed to HEMA (0–1–4–8 mM), both in the presence or absence of LTA (10 μg/ml) or LPS (0.1 μg/ml), or 50 μM BSO or 10 mM NAC in 2 ml culture media as described [ ]. Cell culture supernatants were collected after 1 h or 24 h exposure periods and analyzed for the formation of tumor necrosis factor-α (TNF-α). In addition, amounts of proteins present in adherent cells were analyzed using an Amido Black assay as described below (Section 2.4 ). BSO and NAC concentrations appropriate under the current experimental conditions were established in a previous investigation [ ]. Concentrations of LTA and LPS were chosen because of their effectiveness in the release of TNF-α observed in range finding experiments shown in Fig. 1 . Fig. 1 Release of TNFα from RAW264.7 mouse macrophages. Cell cultures were exposed to a wide range of concentrations of lipoteichoic acid (LTA) (A, B) or lipopolysacharide (LPS) (C, D) for 1 h or 24 h. Levels of TNFα were detected by ELISA as described in Section 2 . Dashed lines indicate the TNFα detection limit. Bars represent median values (25% and 75% percentiles) combined from duplicates in five independent experiments (n = 10). Cell survival (E, F) was determined using a crystal violet assay. * Indicates significant differences of median values between LTA- or LPS-treated cultures and untreated controls. o Indicates significant differences of median values between LTA- or LPS-exposed cultures. Amounts of TNF-α (pg/ml) were determined using a standard ELISA kit (BD Pharmingen). Absorbance was read at 450 nm in a spectrophotometer (Infinite F200, TECAN, Mainz, Germany), and optical density values were collected (Magellan software; version 6.2). Levels of TNF-α in untreated cell cultures below the detection limit were set to the lowest detectable levels (15.6 pg/ml) in standard curves. Absolute amounts of TNFα were summarized from duplicate values in 4–5 independent experiments as specified in the figure legend. 2.4 Determination of protein concentration Adherent cells in 6-well plates employed for the analysis of TNF-α release in cultures treated with HEMA and LTA/LPS were washed with phosphate-buffered saline (PBS/5 mM EDTA), and then lysed in 500 μl RIPA lysis buffer/well for 10 min in the cold. The cell lysate was then transferred to an assay tube. Each well was washed with 400 μl RIPA lysis buffer and the buffer solution was then combined with the cell lysate. Protein concentration in each sample was then determined with Amido Black as described earlier [ ]. 500 μl of reagent (13 μg/ml Amido Black/ml, dissolved in 10% acetic acid, 90% methanol) was added to 80 μl of protein sample and incubated for 5 min at room temperature. Then, the solution was centrifuged at 16,000× g for 5 min, the supernatant discarded, the pellet washed in 500 μl of 10% acetic acid, 90% methanol, and again centrifuged at 16,000× g for 5 min. The washing procedure was repeated twice, and finally the pellet was dissolved in 400 μl of 0.1 N NaOH. An aliquot of 180 μl of each sample was individually transferred to wells in a 96-well plate, and the absorbance was determined at 615 nm in a spectrophotometer (Tecan). Absolute amounts of proteins were calculated from standard curves established with bovine serum albumin (0–30 μg). 2.5 Detection of reactive oxygen species (ROS) Cell cultures (5 × 10 4 /well) were maintained in routine culture medium in 6-well plates for 48 h, and then preincubated with 50 μM BSO for 18 h, or with 10 mM NAC ( N -acetyl cysteine) for 1 h. Subsequently, cell cultures were treated with HEMA (0–1–4–8 mM), both in the presence or absence of LTA (10 μg/ml) or LPS (0.1 μg/ml), or 50 μM BSO or 10 mM NAC in 2 ml culture media [ ]. The exposure of cells was stopped by discarding the exposure media, and the formation of ROS was analyzed by flow cytometry after staining the cells with the fluorescent dyes 2′7′-dichlorodihydrofluorescin diacetate (DCFH 2 -DA), dihydroethidium (DHE) or dihydrorhodamine 123 (DHR123). It has been previously reported that DCFH 2 -DA indicated the formation of general oxidative stress, while DHE or DHR123 detected the formation of superoxide anions or hydrogen peroxide production as discussed in our recent investigation [ ]. Favorable properties of these dyes for the investigation of monomer-induced oxidative stress have been outlined in detail in our previous publication [ ]. Cells were incubated with 10 μM DCFH 2 -DA, 5 μM DHE or 5 μM DHR123 in culture medium 30 min prior to harvesting in phosphate-buffered saline (PBS)/5 mM EDTA. Then, the cells were collected by centrifugation and the cell pellet was resuspended in 200 μl CMF–PBS. DCF and DHR123 fluorescence was measured in the FL-1 channel (488/515–545) using a BD FACSCanto (Becton Dickinson) flow cytometer, and DHE fluorescence intensities were detected in Fl-2 (488/546–606). Mean fluorescence intensities (MFI) were established using histogram statistics and FACSDiva™ 5.0.2 (Becton Dickinson) software. Individual fluorescence intensities measured in treated cell cultures were related to fluorescence quantified in untreated control cultures (= 1.0) as described [ ]. 2.6 Analysis of nitric oxide and peroxynitrite formation The cells (5 × 10 4 /well) were grown in routine culture medium in 6-well plates for 48 h, and then preincubated with 50 μM BSO or 10 mM NAC ( N -acetyl cysteine), exposed to HEMA (0–1–4–8 mM) in the presence or absence of LPS (0.1 μg/ml) or LTA (10 μg/ml), or 50 μM BSO or 10 mM NAC as specifically described in Section 2.5 . Exposure of cells was stopped by discarding the exposure media, and cells were then harvested in phosphate-buffered saline (PBS)/5 mM EDTA, collected by centrifugation and resuspended in 200 μl CMF–PBS. The intracellular formation of nitric oxide (NO ) and peroxynitrite (ONOO) was determined using Cell Meter™ fluorimetric assay kits (AAT Bioquest). Peroxynitrite (ONOO ) or nitric oxide (NO) was examined by flow cytometry (FACSCanto) in channels FL-1 (488/519 nm, ONOO ) or FL-5 (633/660, NO) following the manufacturer’s instructions and as described elsewhere [ ]. Mean fluorescence intensities were determined, and individual fluorescence intensities normalized as described in Section 2.5 . 2.7 Analysis of protein expression by immunoblotting Cells (1.5 × 10 6 cells) were grown in routine culture medium on plates (150 mm diameter) at 37 °C for 48 h. Next, cell cultures were preincubated with 50 μM BSO or 10 mM NAC ( N -acetyl cysteine), exposed to HEMA (0–1–4–8 mM) in the presence or absence of LPS (0.1 μg/ml) or LTA (10 μg/ml), or 50 μM BSO or 10 mM NAC in culture medium. Exposure of cells was stopped by discarding the exposure media. Floating and adherent cells were harvested in ice-cold PBS and combined after centrifugation. Nuclear and cytosolic cell fractions were prepared using different lysis buffers and various centrifugation procedures as described in our previous paper [ ]. Briefly, cell pellets were resuspended in 1 ml ice-cold PBS, mixed with 0.5 ml buffer A (10 mM Tris HCl, 60 mM KCl, 1 mM Na 2 EDTA, 1 mM DTT, pH 7.4), and incubated on ice. The cell suspension was centrifuged again, and the cell pellet was solubilized in buffer B (buffer A plus 0.4% NP 40, 5 mM NaF, 1 mM NaVO 4 ) in the cold. After centrifugation (14,000× g ) the supernatant was collected as a cytoplasmic cell fraction. The cell pellet containing cell nuclei in the pellet was washed in buffer A, resuspended in buffer C (20 mM Tris-Cl pH 8.0, 400 mM NaCl, 1.5 mM MgCl 2 , 1.5 mM Na 2 EDTA, 25% glycerol, 1 mM DTT, 5 mM NaF, 1 mM NaVO 4 ), and centrifugated (14,000× g ) again in the cold. The resulting supernatant was collected as a fraction of nuclear proteins. Protein amounts in the cytosolic and nuclear cell fractions were measured using a BCA protein assay. Protein expression was identified by routine immunoblotting as previously described [ ]. Proteins (10–15 μg per lane) were first separated by SDS-PAGE and then transferred to a nitrocellulose membrane. After washing in TBS (25 mM Tris-Cl, 150 mM NaCl, pH 7.4) and blocking with 5% nonfat milk in TBST (TBS plus 0.1% Tween 20, pH 7.4) at room temperature, the membrane was incubated with primary antibodies to detect protein expression as specified in the legends of Figs. 4 and 7 . Primary antibodies bound to membranes were visualized by horseradish peroxidase-conjugated secondary antibodies and enhanced chemiluminescence (ECL) using a blot scanner and Image Studio software (LI-COR Biotechnology, Bad Homburg, Germany). Membranes were stripped with CHEMICON re-blot stripping solution (Millipore, Schwalbach, Germany) for reprobing. 2.8 Statistical analyses Individual values from repeated independent experiments as specified in the figure legends were summarized as medians (25–75% quartiles). The Mann–Whitney U -test (SPSS Statistics 23, IBM, Armonk, NY, USA) with an alpha level of 0.05 was used to calculate statistically significant differences between median values. 3 Results 3.1 Effectiveness of LPS and LTA in the stimulation of TNFα secretion The effectiveness of LTA (lipoteichoic acid) and LPS (lipopolysaccharide) in the triggering of innate immune responses was initially compared by the stimulation of TNFα secretion in RAW264.7 mouse macrophages through a wide range of LTA or LPS concentrations. While the amounts of TNFα detected in untreated cell cultures were close to the detection limit, both LTA and LPS were differentially effective in the stimulation of TNFα release depending on concentrations and the exposure period. LTA induced a concentration-dependent increase in TNFα release after 1 h or 24 h exposure periods. LTA concentrations in the range of 0.1–50 μg/ml significantly increased TNFα secretion more than 30-fold to about 200–300 pg/ml after a 1 h exposure, and TNFα release further increased about 3-fold when cells were exposed to high LTA concentration for a long exposure period (24 h) ( Fig. 1 A, B). LPS concentrations from 0.1 to 50 μg/ml significantly, and almost equally, increased TNFα secretion to about 500–600 pg/ml after a 1 h exposure, and TNFα release further increased about 5–10-fold after a long exposure period (24 h) ( Fig. 1 C, D). Despite the clear influence on TNFα secretion, no influence on cell survival was detected with all LTA or LPS concentrations after a 1 h exposure. Cell survival was moderately reduced to about 70–75% by LPS (0.1–50 μg/ml), while no decrease in cell survival was observed with LTA (0.1–50 μg/ml) after a 24 h exposure ( Fig. 1 E, F). Concentrations of 10 μg/ml LTA and 0.1 μg/ml LPS were then selected to further investigate the influence of the monomer HEMA on LPS- or LTA-stimulated TNFα secretion. 3.2 LTA-stimulated secretion of TNFα and oxidative stress conditions 3.2.1 LTA-stimulated secretion of TNFα No increase in the amounts of TNFα set to the detection limit in untreated cell cultures was detected after the exposure of cells to increasing HEMA concentrations for 1 h or 24 h. On the contrary, the considerable increase in the secretion of absolute amounts of 374 pg/ml or 465 pg/ml TNFα found in cultures stimulated with 10.0 μg/ml LTA for 1 h or 24 h was significantly inhibited in the presence of HEMA depending on its concentration ( Fig. 2 A, B). Both buthionine sulfoximine (BSO) and N -acetyl cysteine (NAC) were ineffective alone, but decreased LTA-stimulated TNFα secretion and affected the inhibitory effectiveness of HEMA as presented in detail below ( Fig. 2 A, B). Fig. 2 Release of TNFα from LTA-stimulated RAW264.7 mouse macrophages. Cell cultures were preincubated with 50 μM BSO for 18 h, or with 10 mM NAC for 1 h and then treated with LTA (10 μg/ml) or HEMA (0–1–4–8 mM) for 1 h or 24 h with or without 50 μM BSO or 10 mM NAC (A, B). Absolute amounts of TNFα detected in culture supernatants are presented in panels A and B. Amounts of proteins detected after amido black staining of lysed adherent cells are shown in panels C and D, and levels of TNFα related to protein content are presented in panels E and F. Bars show medians (25% and 75% percentiles) summarized from duplicates obtained from 4 to 5 independent experiments (n = 8–10). Significant differences between median values of amounts of TNF, protein or TNF/protein ratios found in the various experimental conditions are as follows: a = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of LTA. b = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of LTA and BSO. c = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of LTA and NAC. d = Significant differences between median values obtained in cultures treated with LTA and HEMA and cultures exposed to HEMA in the presence of LTA and BSO. e = Significant differences between median values obtained in cultures treated with LTA and HEMA and cultures exposed to HEMA in the presence of LTA and NAC. f = Significant differences between median values obtained in cultures treated with LTA, HEMA and BSO and cultures exposed to HEMA in the presence of LTA and NAC. Other statistical analyses are not shown for reasons of clarity. Significant differences between median values calculated from cultures treated with LTA and cultures exposed to LTA in the presence of 1, 4, or 8 mM HEMA are shown by asterisks (*). In order to normalize the absolute amounts of TNFα secreted under the various experimental conditions, we determined the amounts of protein present in exposed cell cultures ( Fig. 2 C, D). No significant influence of HEMA concentrations (0–8 mM) was found on the amounts of protein (60–84 μg) after a 1 h exposure period. While levels of proteins were only slightly reduced in cultures exposed to LTA in the presence of 4 or 8 mM HEMA, a decrease in protein content from about 84 μg (0 mM HEMA) to about 55–70 μg following a consistent pattern was observed in cultures treated with NAC, both in the presence or absence of LTA and HEMA ( Fig. 2 C). After a 24 h exposure the amount of protein increased in untreated cultures almost 4-fold to around 320 μg and was gradually decreased to 104 μg in cultures treated with 8 mM HEMA only ( Fig. 2 D). Noteworthy is that the protein levels increased to 134 μg in cultures co-treated with 8 mM HEMA and LTA compared to those exposed to 8 mM HEMA only. A continuous decrease in protein levels was observed with cultures treated with BSO in the presence or absence of LTA and HEMA down to about 50 μg in cells exposed to 8 mM HEMA. In comparison, 10 mM NAC alone reduced the amount of protein to about 190 μg, but unlike the effect of BSO, NAC increased protein levels when cell cultures were treated with a high concentration HEMA (8 mM) ( Fig. 2 D). Related to these amounts of proteins, LTA-stimulated levels of TNFα decreased more than 3-fold from 9.2 pg LTA/μg protein after a 1 h exposure period to 2.8 pg LTA/μg protein after a 24 h exposure. Yet, secretion of TNFα was drastically and equally reduced with increasing HEMA concentrations. For instance, 2.8 pg LTA/μg protein detected in LTA-stimulated cultures were reduced to 1.1 and 0.4 or 0.3 pg LTA/μg protein in the presence of 1, 4 or 8 mM HEMA after a 24 h exposure. A similar pattern of TNFα secretion was detected after a 1 h exposure ( Fig. 2 E, F). The concentrations of BSO and NAC chosen here were highly effective in the inhibition the LTA-stimulated TNFα secretion after a 24 h exposure in particular. For instance, levels of TNFα in LTA-exposed cells decreased from 2.8 pg/μg protein to 1.5 (BSO) or 0.2 (NAC) pg/μg protein ( Fig. 2 F). Both substances seemed to intensify the inhibitory effectiveness of HEMA after long exposure (24 h), although no such effect was detectable with BSO after a 1 h exposure ( Fig. 2 E, F). 3.2.2 LTA-stimulated oxidative and nitrosative stress The influence of BSO and NAC on the release of TNFα from LTA- and HEMA-stimulated cells after long exposure in particular suggested a crucial role of GSH and the production of ROS or RNS. Thus, the formation of oxidative stress, nitric oxide (NO) and peroxynitrite was further analyzed. Oxidative stress indicated by DCF fluorescence significantly increased in RAW264.7 mouse macrophages after exposure to increasing concentrations of HEMA for 1 h ( Fig. 3 A). A 1.3-fold increase compared to untreated cell cultures (= 1.0) was detected with 8 mM HEMA, and DCF fluorescence further increased more than 2−3-fold in the presence of BSO in cultures co-treated with HEMA ( Fig. 3 A). This effect of BSO was not detectibly influenced by 10 μg/ml LTA. LTA, however, slightly reduced DCF fluorescence alone and in the presence of HEMA. Furthermore, DCF fluorescence was also significantly reduced by NAC in cultures treated with HEMA, and this inhibitory effect was reproducibly intensified by LTA ( Fig. 3 A). Most relevant, a 5-fold increase in LTA-stimulated DCF-fluorescence after a 24 h exposure was continuously and significantly decreased by HEMA. This pattern of LTA-induced DCF fluorescence and inhibition in the presence of HEMA was drastically increased in the presence of BSO. LTA-stimulated DCF-fluorescence was considerably reduced in the presence of NAC, which then further intensified the inhibitory effect of HEMA ( Fig. 3 B) Fig. 3 Formation of oxidative stress, nitric oxide and peroxynitrite in RAW264.7 mouse macrophages. Cell cultures were first preincubated with 50 μM BSO for 18 h, or with 10 mM NAC for 1 h and then stimulated with LTA (10 μg/ml) and HEMA (0–1–4–8 mM) in the presence or absence of 50 μM BSO or 10 mM NAC. Formation of oxidative stress was detected after staining cells with DCFH 2 (A, B), and production of nitric oxide (NO) (C, D) or peroxynitrite (E, F) after 1 h or 24 h exposure periods was analyzed by flow cytometry. Bars show medians (25% and 75% percentiles) summarized from individual values in independent experiments (n = 4–5). Significant differences between median values of DCF, NO or ONOO fluorescence are as follows: a = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of LTA. b = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of BSO. c = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of LTA and BSO. d = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of NAC. e = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of LTA and NAC. f = Significant differences between median values obtained in cultures treated with LTA and HEMA and cultures exposed to HEMA in the presence of LTA and BSO. g = Significant differences between median values obtained in cultures treated with LTA and HEMA and cultures exposed to HEMA in the presence of LTA and NAC. Other statistical analyses are not shown for reasons of clarity. Significant differences between median values calculated from cultures treated with LPS and cultures exposed to LPS in the presence of 1, 4, or 8 mM HEMA are shown by asterisks (*). Levels of nitric oxide (NO) were reduced to 0.9 in the presence of LTA compared to untreated cell cultures (= 1.0), and this effect was slightly but not significantly enhanced by 1 or 4 mM HEMA after a short exposure period (1 h) ( Fig. 3 C). While no influence of HEMA on the formation of NO · was detected in the presence or absence of BSO, NAC reduced the amounts of NO in control cultures and cells exposed to a low concentration of HEMA ( Fig. 3 C). After long exposure (24 h), LTA stimulated the formation of NO about 2.7-fold compared to untreated controls. This effect was reduced by 1 and 4 mM HEMA, but increased again with 8 mM HEMA to levels induced by LTA alone. HEMA itself significantly and continuously increased the formation of NO up to 3-fold compared to untreated cell cultures. The presence of BSO, which was ineffective alone and on LTA, even intensified the increase in NO formation in cell cultures treated with HEMA. In contrast, NAC reduced both LTA-stimulated and HEMA-induced production of NO ( Fig. 3 D). The levels of peroxynitrite (ONOO ) [ ], as a product of superoxide anion and nitric oxide, were slightly enhanced only in cultures treated with BSO for a short period of time ( Fig. 3 E, F). In contrast, ONOO formation was enhanced about 2-fold in cells stimulated with LTA compared to untreated cultures after a 24 h exposure. This LTA-induced effect, however, was moderately but significantly reduced with increasing concentrations of HEMA ( Fig. 3 F). The formation of peroxynitrite by LTA was independent of BSO, but BSO alone enhanced the ONOO levels about 1.3-fold, and it considerably (2-fold) intensified a slight increase in ONOO levels caused by HEMA. NAC reduced LTA-stimulated and HEMA-induced production of ONOO ( Fig. 3 F). 3.2.3 LTA-stimulated expression of proteins related to TNFα release and oxidative or nitrosative stress Expression of proteins directly related to oxidative and nitrosative stress was detected in cell cultures exposed to LTA and HEMA for 24 h. The redoxsensitive transcription factor Nrf2 was expressed in the cytosol of untreated cell cultures, and Nrf2 expression was also detected in the nucleus in cells exposed to HEMA. Moreover, high levels of Nrf2, as detected in the nucleus of cells exposed to 10 μg/ml LTA possibly after translocation from the cytosol to the cell nucleus, were decreased in the presence of HEMA. Most relevant, expression of Nrf2 in the cell nucleus was drastically decreased in LTA-stimulated cultures co-exposed to the antioxidant NAC, while nuclear Nrf2 expression was moderately reduced in the presence of BSO ( Fig. 4 ). The expression of the transcription factor NF-κB, as indicated by subunit p65, was also stimulated in nuclei of cells treated with 10 μg/ml LTA, and these increased levels of NF-κB were reduced in the presence of HEMA. LTA-stimulated NF-κB expression also slightly decreased in the presence of NAC, as shown by the enhanced NF-κB levels in the cytosolic fraction compared to cultures treated with LTA alone. Although BSO alone strongly increased the expression of p65, the induction of nuclear NF-κB expression by LTA was even lower in cultures pretreated with BSO ( Fig. 4 ). Both HEMA and LTA induced expression of the stress protein HO-1, which was further enhanced in the presence of BSO. In contrast, co-treatment with NAC reduced HO-1 expression to low levels in cell cultures exposed to LTA or HEMA. Fig. 4 Protein expression in RAW264.7 mouse macrophages. Cell cultures were first preincubated with 50 μM BSO for 18 h, or with 10 mM NAC for 1 h and then stimulated with LTA (10 μg/ml) and HEMA (0–1–4–8 mM) in the presence or absence of 50 μM BSO or 10 mM NAC. Protein expression was analyzed by Western blotting. Expression of Nrf2 and subunit p65 of NF-κB was detected in the cytosol (C) as well as in cell nuclei (N), and subunit p47 phox of Nox2, inducible nitric oxide synthase (iNOS), and heme oxyganase1 (HO-1) were identified in the cytosol. Equal loading of proteins across the gel was controlled by the expression of lamin A/C or GAPDH as markers for cell nuclei or the cytosol, respectively (not shown). Expression of the inducible form of nitric oxide synthase (iNOS) was extremely low or absent under the current experimental conditions in cultures exposed to HEMA, NAC or BSO. Notably, a drastically increased expression of iNOS in LTA-stimulated cultures was effectively inhibited in the presence of HEMA. While LTA alone even lowered p47 expression, a remarkable increase in the basic expression of p47, as a regulatory subunit of NADPH oxidase, was detected in cells exposed to NAC. An increase in catalase expression in cell cultures exposed to HEMA was moderately reduced in the presence of LTA independent of the presence of NAC or BSO ( Fig. 4 ). 3.3 LPS-stimulated secretion of TNFα and oxidative stress conditions 3.3.1 LPS-stimulated secretion of TNFα The secretion of TNFα in LPS-stimulated cell cultures, with due regard to the selected concentrations, essentially followed the pattern observed with LTA with only a few exceptions. Amounts of TNFα set to the detection limit in untreated cell cultures were not enhanced after exposure to increasing HEMA concentrations. A considerable increase in the secretion of absolute amounts of TNFα found in cultures stimulated with 0.1 μg/ml LPS for 1 h or 24 h was significantly inhibited depending on the concentrations of HEMA ( Fig. 5 A,B). Both buthionine sulfoximine (BSO) and N -acetyl cysteine (NAC) affected the inhibitory effectiveness of HEMA as described with LTA. Noteworthy, the 5-fold increase in the absolute amounts of TNFα, as detected in cultures exposed to LPS for 24 h compared to TNFα levels, was not found after a 1 h exposure period ( Fig. 5 A, B). The amount of protein found in cultures exposed to HEMA, LPS, BSO or NAC for a 1 h exposure period were not different from those detected in experiments with LTA ( Fig. 5 C). While the influence of BSO or NAC on the amounts of protein in cultures exposed to HEMA or LPS for 24 h was identical to the effects observed in the presence of LTA, protein levels were significantly reduced in the presence of 0.1 μg/ml LPS. Noteworthy, protein levels significantly increased in cultures co-treated with 8 mM HEMA and LPS compared to those exposed to 8 mM HEMA only ( Fig. 5 D). Fig. 5 Release of TNFα from LPS-stimulated RAW264.7 mouse macrophages. Cell cultures were preincubated with 50 μM BSO for 18 h, or with 10 mM NAC for 1 h and then treated with LPS (0.1 μg/ml) or HEMA (0–1–4–8 mM) for 1 h or 24 h with or without 50 μM BSO or 10 mM NAC (A, B). Absolute amounts of TNFα detected in culture supernatants are presented in panels A and B. Amounts of proteins detected after amido black staining of lysed adherent cells are shown in panels C and D, and levels of TNFα related to protein content are presented in panels E and F. Bars show medians (25% and 75% percentiles) summarized from duplicates obtained from 4 to 5 independent experiments (n = 8–10). Significant differences between median values of amounts of TNF, protein or TNF/protein ratios found in the various experimental conditions are as follows: a = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of LPS. b = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of LPS and BSO. c = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of LPS and NAC. d = Significant differences between median values obtained in cultures treated with LPS and HEMA and cultures exposed to HEMA in the presence of LPS and BSO. e = Significant differences between median values obtained in cultures treated with LPS and HEMA and cultures exposed to HEMA in the presence of LPS and NAC. f = Significant differences between median values obtained in cultures treated with LPS, HEMA and BSO and cultures exposed to HEMA in the presence of LPS and NAC. Other statistical analyses are not shown for reasons of clarity. Significant differences between median values calculated from cultures treated with LPS and cultures exposed to LPS in the presence of 1, 4, or 8 mM HEMA are shown by asterisks (*). Related to the amounts of proteins, LPS-stimulated TNFα secretion increased more than 3-fold with prolonged periods, but was drastically decreased with increasing HEMA concentrations ( Fig. 5 E, F). For instance, 54 μg TNFα/mg protein detected in LPS-stimulated cultures were reduced to 1.4 and 3.5 μg TNFα/mg protein in the presence of 4 or 8 mM HEMA, respectively after a 24 h exposure. A similar pattern of TNFα secretion was detected after a 1 h exposure. BSO or NAC were highly effective in the inhibition the LPS-stimulated TNFα secretion, identical to the effects shown with LTA ( Fig. 5 E, F). 3.3.2 LPS-stimulated oxidative and nitrosative stress As shown for the release of TNFα, the formation of oxidative or nitrosative stress in LPS-stimulated cell cultures was very similar to the results obtained with LTA. As the only exception, DCF fluorescence produced by HEMA, and enhanced in the in the presence of BSO, was further increased when cultures were co-treated with 0.1 μg/ml LPS for 1 h ( Fig. 6 A). In addition, a slight reduction in DCF fluorescence found with LTA was not detected in cultures exposed to LPS. Apart from this and highly relevant, a 5-fold increase in LPS-stimulated DCF-fluorescence was decreased by HEMA after a 24 h exposure period, similar to the finding with LTA. Likewise, DCF fluorescence induced by LPS was further enhanced in the presence of BSO, but drastically reduced in cultures co-exposed to NAC, which amplified the inhibitory effect of HEMA ( Fig. 6 B). LPS also intensified the formation of NO about 3-fold compared to untreated controls, and the increase in levels of NO induced by HEMA was slightly enhanced in the presence BSO and LPS. NAC effectively reduced LPS- and HEMA-induced formation of NO ( Fig. 6 D). Levels of peroxynitrite (ONOO ), which increased about 3-fold in LPS-treated cultures, were markedly reduced with increasing HEMA concentrations, and this inhibitory effect was intensified by NAC ( Fig. 6 F). Fig. 6 Formation of oxidative stress, nitric oxide and peroxynitrite in RAW264.7 mouse macrophages. Cell cultures were first preincubated with 50 μM BSO for 18 h, or with 10 mM NAC for 1 h and then stimulated with LPS (0.1 μg/ml) and HEMA (0–1–4–8 mM) in the presence or absence of 50 μM BSO or 10 mM NAC. Formation of oxidative stress was detected after staining cells with DCFH 2 (A, B), and production of nitric oxide (NO) (C, D) or peroxynitrite (E, F) after 1 h or 24 h exposure periods was analyzed by flow cytometry. Bars show medians (25% and 75% percentiles) summarized from individual values in independent experiments (n = 4–5). Significant differences between median values of DCF, NO or ONOO fluorescence are as follows: a = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of LPS. b = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of BSO. c = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of LPS and BSO. d = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of NAC. e = Significant differences between median values obtained in untreated cultures (medium) and cultures exposed to HEMA in the presence of LPS and NAC. f = Significant differences between median values obtained in cultures treated with LPS and HEMA and cultures exposed to HEMA in the presence of LPS and BSO. g = Significant differences between median values obtained in cultures treated with LPS and HEMA and cultures exposed to HEMA in the presence of LPS and NAC. Other statistical analyses are not shown for reasons of clarity. Significant differences between median values calculated from cultures treated with LPS and cultures exposed to LPS in the presence of 1, 4, or 8 mM HEMA are shown by asterisks (*). 3.3.3 LPS-stimulated expression of proteins related to TNFα release and oxidative or nitrosative stress Again, the expression of proteins regulating or indicating oxidative or nitrosative stress in cell cultures exposed to LPS or HEMA for 24 h revealed identical patterns as detected with LTA. The expression of Nrf2 was drastically increased in nuclei of cells exposed to 0.1 μg/ml LPS, and decreased in LPS-stimulated cultures co-exposed to HEMA in the presence of NAC, while levels of Nrf2 increased in the cytosol in parallel ( Fig. 7 ). LPS-induced Nrf2 expression was reduced in the presence of BSO, and further decreased in cultures co-exposed to HEMA. As also observed with cultures treated with LTA, the expression of NF-κB was reduced in the presence of HEMA in LPS-treated cultures and controls. HEMA- and LPS-induced expression of HO-1 increased even more in the presence of BSO, but HO-1 expression was hardly detected in cultures co-exposed to NAC ( Fig. 7 ). LPS-stimulated expression of iNOS was radically decreased in the presence of HEMA. A considerable increase in the expression of p47 was counteracted by LPS, and in part by high concentrations of HEMA. HEMA-stimulated expression of catalase was reduced in the presence of LPS similarly to the observation with LTA ( Fig. 7 ). Fig. 7 Protein expression in RAW264.7 mouse macrophages. Cell cultures were first preincubated with 50 μM BSO for 18 h, or with 10 mM NAC for 1 h and then stimulated with LPS (0.1 μg/ml) and HEMA (0–1–4–8 mM) in the presence or absence of 50 μM BSO or 10 mM NAC. Protein expression was analyzed by Western blotting. Expression of Nrf2 and subunit p65 of NF-κB was detected in the cytosol (C) as well as in cell nuclei (N), and subunit p47 phox of Nox2, inducible nitric oxide synthase (iNOS), and heme oxyganase1 (HO-1) were identified in the cytosol. Equal loading of proteins across the gel was controlled by the expression of lamin A/C or GAPDH as markers for cell nuclei or the cytosol, respectively (not shown). 4 Discussion 4.1 LTA and LPS stimulation of TNFα secretion in mouse macrophages The variety of cells forming dental pulp tissue are exposed to invading Gram-positive and Gram-negative cariogenic microorganisms in deep carious lesions in particular. These bacteria release immunogenic components such as LTA or LPS from the cell envelope. Binding of these components to cell membrane receptors TLR2 or TLR4 in immunocompetent cells of the dental pulp leads to the activation of inflammatory responses, including the release of pro-inflammatory cytokines such TNFα [ , ]. In initial range-finding experiments we demonstrated that LTA, as a component of Gram-positive microorganisms, was active in mouse macrophages under the current experimental conditions as described earlier, and similar to the known effectiveness of LPS [ , ]. Levels of TNFα were differentially produced in the cells depending on LTA or LPS concentrations. Although slightly diverse after different exposure periods, amounts of TNFα were detected in a similar range in cultures exposed to 0.1 μg LPS/ml or 10 μg LTA/ml. This observation indicated that LPS was more than 100-fold more effective than LTA in this respect in mouse macrophages. Comparable differences in innate immune responses caused by LPS and LTA were reported with epithelial cells [ ]. Likewise, LTA, obviously isolated from Streptococcus mutans as a Gram-positive microorganism associated with dental caries, stimulated the release of TNFα from RAW264.7 mouse macrophages, but less efficiently than LPS [ ]. The reason for these differences in the effectiveness of the LTA and LPS preparations used here remains to be shown. Based on the present observations in preliminary range-finding experiments, and considering the fact that severe cytotoxicity of both LTA and LPS in mouse macrophages was not detected, equally effective concentrations of 10 μg/ml LTA and 0.1 μg/ml LPS were selected to investigate the influence of the resin monomer HEMA on LTA- or LPS-stimulated TNFα secretion. 4.2 Differential inhibition of LTA- and LPS-stimulated TNFα secretion by HEMA There is ample evidence that the LPS-stimulated secretion of TNFα, as a protective mechanism against oral pathogens, is antagonized by resin monomers released from dental adhesives such as HEMA or TEGDMA [ , ]. Recent investigations on the mechanism behind this process suggested a role of the thiol reactivity of HEMA [ ]. Alternatively, a crucial role of oxidative stress in the inhibition of LPS-stimulated cytokine release by resin monomers is more than likely. The formation of reactive oxygen (ROS) or nitrogen (RNS) species is firmly documented, and a major function of the redox-sensitive transcription factor Nrf2 in the inhibition of LPS-stimulated cytokine release by HEMA has been shown [ ]. Here we used N -acetyl cysteine (NAC) as an antioxidant and l -buthionine sulfoximine (BSO) as an inhibitor of glutathione synthesis to detect direct functions of ROS or RNS in monomer-induced inhibitory processes [ , ]. To improve experimental accurateness in the interpretation of the present findings, absolute amounts of TNFα found in cell cultures were related to the cell mass expressed as amounts of protein. We found that the thus far undescribed effectiveness of LTA in cells simultaneously co-treated with increasing concentrations of HEMA induced a pattern of cell responses identical to those observed with LPS. LTA-stimulated TNFα release was inhibited by HEMA after short (1 h) and long (24 h) exposure periods exactly as detected with LPS here and as recently shown [ ]. While 1 mM HEMA decreased TNFα release in LTA- or LPS-stimulated cell cultures by about half, the secretion of TNFα was completely inhibited by 4 mM HEMA. Noteworthy, cell masses detected in treated cell cultures were not significantly reduced after a short exposure to LTA or LPS in the presence or absence of HEMA, indicating that TNFα release was not affected by simple cytotoxic effects. The effectiveness of 1 mM HEMA in particular and a concentration-dependent effect of HEMA on TNFα release after a 24 h exposure also indicate physiological responses of the cells. Remarkable as well was the fact that the absolute amounts of TNFα detected in LTA-stimulated cell cultures remained constant between short (1 h) and long exposure periods (24 h), although protein levels increased about 3-fold. Thus, in contrast to the observations with LPS, it seemed that LTA-stimulated TNFα release reached saturation after 1 h exposure under the current experimental conditions. While LTA and LPS were both effective on TNFα release in the same number of cells in a similar range after a short exposure, the amounts of TNFα drastically increased with time in LPS-treated cell cultures. Currently we have no explanation for the diverse effects of LTA and LPS on TNFα release after long exposure. We suggested earlier that TNFα released after LPS stimulation of cells might persistently activate NF-κB in an amplifying feed-forward-loop leading to large amounts of TNFα through transcriptional activation [ ]. It is possible that LTA, which is active on TLR2 might negatively interfere with the TNFα-responsive TLR4 receptor and signalling downstream [ ]. An intriguing finding is the influence of HEMA, BSO and NAC on LTA- or LPS-stimulated TNFα release. Although each substance disrupts the cellular redox homeostasis through different mechanisms, they all consistently inhibit LTA- or LPS-stimulated TNFα release. Following the current hypothetical model, HEMA probably increases oxidative stress as a consequence of its binding to GSH, and the subsequent rise in the amounts of reactive oxygen species is mostly due to hydrogen peroxide [ , ]. While BSO is a specific inhibitor of GSH synthesis, creating an immediate increase in oxidative stress, NAC, as a thiol reagent, maintains in reverse normal levels of intracellular GSH, acts as an antioxidant itself, and consequently decreases oxidative stress [ , ]. LTA- or LPS-stimulated TNFα release was only slightly influenced by BSO after a short exposure, and no influence on the effects of HEMA was detected. In contrast, BSO inhibited the secretion of TNFα in LTA- or LPS-treated cell cultures, and apparently amplified the inhibitory effect of HEMA after a long exposure (24 h). In this case, BSO probably enhanced the production of ROS or RNS, as indicated by DCF fluorescence in particular, beyond the tolerable level for TNFα release. The relevance of redox regulation on cytokine release in cells treated with LPS has also been reported in alveolar cells, since BSO seemed to increase the effect of LPS after a long exposure [ ]. Similar to our findings, a reduction in the LPS-stimulated release of the cytokine IL-1b in the presence of the GSH modulator BSO and HEMA was reported [ ]. However, due to the magnitude and dynamics of the effects of HEMA and BSO it was suggested that GSH depletion might not explain the inhibition of IL-1b release. Alternatively, as a thiol-reactive electrophilic molecule, HEMA could bind to protein-cysteines and thus explain the persistent and sustained effects of HEMA [ ]. The secretion of TNFα in cultures exposed to both LTA and LPS was also inhibited in the presence of the antioxidant NAC. A relatively small reduction in TNFα release by half after a 1 h exposure indicated that NAC was active as an antioxidant, while a strong inhibition in cells exposed for 24 h additionally suggested support of GSH synthesis by NAC. LTA- or LPS-stimulated TNFα secretion inhibited in the presence of HEMA was further reduced by NAC. It is possible that LTA- or LPS-induced nuclear translocation of NF-κB is reduced in the presence of NAC. A relatively higher amount of the NF-κB subunit p65 remains in the cytosol in cultures treated with LTA and NAC, and only a weak signal of p65 was detected in cell nuclei in cultures exposed to LPS and NAC. NAC is known as an effective immuno-modulator, and it has been previously observed that downregulation of TNFα release is associated with the inhibition of NF-κB and inducible NO synthase (iNOS) [ , ]. In conclusion, it is obvious that the influence of both BSO and NAC suggests the disturbance of the cellular redox homeostasis as a crucial factor in the inhibition of LTA- and LPS-stimulated TNFα secretion by HEMA. The type of ROS or RNS relevant for this imbalance of the steady state, however, remains to be identified. 4.3 Formation of ROS and RNS in LTA- and LPS-stimulated cells Oxidative stress based on the formation of the same types of ROS or RNS was increased with similar effectiveness and pattern by a 100-fold difference in the concentrations of LTA and LPS in the present investigation. DCF fluorescence and the formation of nitric oxide and peroxynitrite were enhanced to the same levels after the exposure of cell cultures to LPS or LTA for 24 h. However, 0.1 μg/ml LPS seem to be a stronger inducer of DCF fluorescence than 10 μg/ml LTA after a short exposure period. Formation of oxidative stress as indicated by DCF fluorescence essentially depends on the activation of the redox-sensitive transcription factor NFκ-B. It is well established that the binding of LPS to TLR4 ultimately stimulates NFκ-B activity, which among others regulates NADPH oxidase (Nox2) and nitric oxide synthase (iNOS) expression. Activation of Nox2 and iNOS enhances the formation of superoxide anions and nitric oxide (NO) , which combine to peroxynitrite (ONOO ) to finally build a plethora of ROS and RNS [ , ]. LTA binds to TLR2 and interferes with CD14, suggesting an identical mode of NFκ-B activation and the generation of oxidative stress as described with LPS [ , ]. The influence of both BSO and NAC on LTA- or LPS-induced oxidative stress indicates a crucial role of glutathione in these processes. BSO increased HEMA-induced DCF fluorescence after a short exposure period (1 h), and even amplified the effect of LPS. Most interesting, the increase in LTA- and LPS-stimulated DCF fluorescence further amplified by BSO was inhibited in the presence of increasing concentrations of HEMA only after a long exposure period (24 h). An increase in DCF fluorescence caused by BSO alone was independent of the presence of HEMA after a long exposure period. We have shown the effectiveness of the current BSO concentration on the reduction of glutathione levels in RAW264.7 mouse macrophages in a recent investigation [ ]. Although it is obvious that a reduction of glutathione levels by BSO, and the treatment of cultures with LTA or LPS lead to an increase in the levels of different types of ROS or RNS, their precise nature still remains obscure. LTA- or LPS-stimulated DCF fluorescence is apparently caused by ROS or RNS rather than NO or ONOO because the formation of these species in the presence of BSO was not reduced by HEMA. Moreover, BSO was reported to induce the accumulation of hydrogen peroxide, superoxide anions or hydroxyl radicals [ ]. It has also been recently suggested that hydrogen peroxide might not contribute to DCF florescence because the formation of hydrogen peroxide increased rather than decreased in cultures exposed to HEMA or LPS after a long exposure period [ ]. In contrast to the effects of BSO, NAC consistently reduced both LTA- and LPS-stimulated DCF fluorescence, and increased the effectiveness of HEMA after a long exposure period in particular. The protective efficiency of NAC on HEMA-induced physiological cell responses other than inhibition of immune function has been repeatedly shown [ , ]. It has also been suggested that HEMA-induced oxidative stress was a result of the formation of a wide variety of reactive oxygen and nitrogen species [ ]. Nonetheless, the type of ROS or RNS the antioxidant NAC preferably reacts with remains unclear. Rate constants with superoxide anions, hydrogen peroxide or peroxynitrite were considered low, but NAC might neutralize highly oxidizing radicals very rapidly [ ]. The differentiation of the type of ROS or RNS interfering with the inhibition of TNFα release in LTA- or LPS-stimulated cells in the presence of HEMA is difficult due to the seemingly contradictory generation of oxidative stress indicated by the increase in DCF fluorescence as a measure of general oxidative stress. Our recent findings suggested the interaction of the redox-sensitive transcription factors NF-κB and Nrf2 in the regulation of HEMA-induced inhibition of LPS-stimulated cytokine release, probably through the generation of oxidative stress [ ]. The expression of pro- and anti-inflammatory cytokines is related to oxidative stress as a result of the NF-κB-regulated expression of NADPH oxidase (Nox2) and nitric oxide synthase (iNOS), and superoxide anions and NO· finally lead to the generation of peroxynitrite (ONOO ) [ , ]. In the present investigation, the formation of NO increased in cultures exposed to LTA or LPS for a long exposure period, and a considerable increase in the levels of NO caused by HEMA was amplified by BSO, but completely inhibited by NAC. These results imply the production of NO · through distinct mechanisms in the presence of HEMA or LTA and LPS, as shown earlier with LPS [ ]. Moreover, these observations also indicate that glutathione might be of minor importance for the LTA- or LPS-regulated formation of NO ·, but NAC is relevant for the reduction in NO· levels caused by both HEMA and LTA/LPS. We cannot provide a plausible interpretation for the effects detected with NAC at this time, but it has been reported that NAC does not react with NO ·, and maintaining levels of intracellular GSH is unlikely as well [ ]. NAC might negatively interfere with the expression of nitric oxide synthase (iNOS) [ ], yet this phenomenon was not detected in our Western blot analyses. Peroxynitrite (ONOO ) is created by NO and superoxide anions are produced by both iNOS and Nox2. Both LTA and LPS enhanced the formation or peroxynitrite, possibly through the activation of NF-κB which regulates iNOS and Nox2 [ ]. Although ONOO increased in cultures exposed to BSO alone, the constant high levels of peroxynitrite in cultures exposed to LTA or LPS in the presence of BSO support the hypothesis that LTA- or LPS-stimulated formation of ONOO occurs independently of glutathione levels. NAC, on the other hand, reduced LTA- or LPS-induced generation of ONOO , likely a consequence of the inhibition of iNOS activity. HEMA apparently inhibited LTA- or LPS-induced ONOO generation according to a pattern similar to the reduction of DCF fluorescence. Remarkably, NAC amplified the inhibitory effect of HEMA, which was most pronounced in cultures treated with LPS, thus supporting the hypothesis of the inhibition of iNOS. The current results indicate that LTA and LPS induce a wide variety of reactive oxygen and nitrogen species including NO and ONOO. It seems that glutathione was less important for the generation of NO or ONOO, but the inhibition of GSH synthesis by BSO strongly enhanced other ROS or RNS detected by DCF fluorescence. Moreover, the present findings clearly show that our first hypothesis suggesting that NAC counteracts HEMA-induced inhibition of LPS-stimulated cytokine release by reducing oxidative stress was extremely oversimplified. TNFα release was not enhanced but even more inhibited by HEMA in the presence of NAC. These results point to a complex scenario involving various types and concentrations of ROS and RNS created and regulated over time by a network of transcription factors and executing proteins. 4.4 LTA/LPS-stimulated expression of proteins related to oxidative or nitrosative stress The disturbance of the cellular redox homeostasis resulting from increased oxidative or nitrosative stress leads to the activation of proteins in the control of inflammatory processes regulated by signalling proteins such as the redox-sensitive transcription factors NF-κB and Nrf2 [ ]. Referring to our initial work on the function of NF-κB and Nrf2 in the inhibition of LPS-stimulated cytokine release by HEMA, the role of stress-activated MAPKs was disregarded here to reduce the complexity of the topic [ ]. While NF-κB function at the transcriptional level, activated by LTA- or LPS-induced oxidative stress, results in the formation of cytokines, this process is counteracted by the activation of the Nrf2 system. Therefore, inflammatory processes such as the formation and release of TNFα are controlled by an NF-κB/Nrf2 crosstalk [ , ]. Detection of the catalytic subunit p65 was used to trace the expression of the transcription factor NF-κB in the present investigation [ ]. The most important finding here was that both LTA- and LPS-stimulated expression of p65 in the cell nucleus decreased in the presence of HEMA, most likely because of an inhibited translocation from the cytosol. As a consequence, we assume that the formation of TNFα was downregulated at the transcriptional level in the presence of HEMA. Primarily as a result of the immediate cell response to monomers observed earlier, alternative and supporting pathways at the non-transcriptional level might be activated as well [ , , ]. Inhibition of the nuclear translocation of p65 is sensitive to oxidative stress. The LPS-stimulated expression of p65 in the nucleus decreased in the presence of NAC, and p65 levels were reduced in the nucleus, or retained in the cytosol after stimulation with LTA as well. The effects observed with BSO are difficult to explain at this level of the investigation. Interestingly, however, oxidative stress increased in cell cultures exposed to BSO, but an increase in TNFα release was not detected although p65 expression in the cell nucleus was also intensified. It is possible that, under these experimental conditions, the increased levels of NF-κB in the cell nucleus may in fact support cell survival rather than act on cytokine formation [ ]. NF-κB-stimulated responses of immune cells are controlled by its counterpart Nrf2, a redox-sensitive transcription factor activating a network of antioxidative enzymes [ , ]. It has been previously shown that activation of Nrf2 by LPS-stimulation and related oxidative stress inhibits pro-inflammatory cytokine release [ , ]. In the present investigation both LTA and LPS stimulated the nuclear translocation of Nrf2, suggesting a need for the expression of antioxidant enzymes, especially HO-1 (heme oxygenase) [ ]. This hypothesis is supported by a decrease in nuclear levels of Nrf2 and inhibited HO-1 expression in LTA- or LPS-treated cells simultaneously treated with the antioxidant NAC. These findings on nuclear Nrf2 translocation also correlate with our analyses of ROS and RNS formation. In contrast to the influence of NAC, a possible slight inhibitory effect of BSO on LTA- or LPS-induced nuclear translocation of Nrf2 is currently difficult to explain. This particular aspect of the present study needed more detailed investigation. Yet, an understanding of this observation might be intimately related to the seemingly contradictory results obtained in the presence of HEMA. Although HEMA alone induced the nuclear translocation of Nrf2, the monomer seemed to inhibit this LTA-stimulated effect. An inhibition of the LPS-induced Nrf2 translocation by HEMA became observable only in the presence of NAC, possibly due to the drastic LPS-stimulated increase in Nrf2 levels without NAC. Expression of HO-1 as an anti-oxidant protein is controlled by both NF-κB and Nrf2, and links the activities of both transcription factors. It has been reported that Nrf2-induced expression of HO-1 inhibits NF-κB activity, indicating its anti-inflammatory function [ , ]. The expression of HO-1 under the current experimental conditions strictly depended on oxidative or nitrosative stress as demonstrated with cell cultures exposed to LTA/LPS or HEMA. Since LTA/LPS stimulate NF-κB activity, the induced HO-1 expression is probably a consequence of the formation of RNS through ONOO , which amplified the expression of Nrf2, while HEMA induced Nrf2 expression possibly occurs through the formation of hydrogen peroxide [ ]. Moderate HEMA concentrations (4 mM) inhibit LTA/LPS-induced HO-1 expression because they reduced nitrosative stress through ONOO . As an indicator of oxidative stress based on the formation of peroxides, catalase expression was mostly enhanced in cultures exposed to HEMA. This effect has been shown previously [ , ]. Notably, even if basic levels of catalase expression were slightly affected by LTA or LPS, HEMA-stimulated induction of catalase expression was reduced in the presence of LTA or LPS. These findings suggest that LTA/LPS reduced HEMA-stimulated H 2 O 2 levels. The underlying mechanism behind this process is probably the rapid formation of ONOO in LTA/LPS-treated cultures through the reaction of nitric oxide (NO) with superoxide anions, which are then available at reduced levels to form hydrogen peroxide. Expression of Nox2 represented by its subunit p47 and iNOS is controlled by NF-κB [ ]. The concentration of NO· increased as a consequence of the LTA/LPS-induced expression of iNOS. However, expression of Nox2 (p47), which produces superoxide anions, is suppressed probably through a feedback mechanism activated by hydrogen peroxide and increased concentrations of ONOO . This suggestion is supported by the increase in p47 expression in HEMA-exposed cell cultures in the presence of NAC. 5 Conclusion Considering the release of inflammatory cytokines such as TNFα, the antigens LPS and LTA expressed in cariogenic microorganisms were differentially effective in mouse macrophages. LPS- or LTA-stimulated TNFα secretion, as a crucial function in host defense against invading pathogens, strongly decreased in the presence of the dental monomer HEMA because of a cellular redox imbalance. There is evidence that HEMA-induced oxidative stress causes the inhibition of the nuclear translocation of NF-κB as a result of the interplay between a multitude of reactive oxygen and nitrogen species produced by NF-κB- and Nrf2-regulated proteins. It is reasonable to hypothesize that concentrations of monomers such as HEMA used in the present investigation in vitro are active in a more complex clinical situation as well. Considering the concentration of the pure monomer (8.2 mol/l), an unpolymerized model adhesive containing 10% HEMA would result in the exposure of dentin to 800 mmol/l HEMA. Thus, it is plausible to presume that resin monomers will interact effectively with odontoblast processes and pulp tissues locally and temporarily after diffusion through dentinal tubules when the preparation of deep cavities leaves dentin at a thickness of 0.5 mm or less and with rare partial pulp exposure [ ]. Dynamic processes in the interaction of a wide range of concentrations of LPS, LTA and monomers such as HEMA lead to complex adaptive reactions in affected pulp cells and tissues initiating inflammatory processes. Acknowledgments The authors are grateful for the expert advice on the manuscript by Christine Ross-Cavanna. The project was supported by the Deutsche Forschungsgemeinschaft (DFG, Schw 431/15-1). References 1. Bleicher F.: Odontoblast physiology. Exp Cell Res 2014; 325: pp. 65-71. 2. Staquet M.J., Carrouel F., Keller J.F., Baudouin C., Msika P., Bleicher F., et. al.: Pattern-recognition receptors in pulp defense. Adv Dent Res 2011; 23: pp. 296-301. 3. Schmalz G., Krifka S., Schweikl H.: Toll-like receptors, LPS, and dental monomers. Adv Dent Res 2011; 23: pp. 302-306. 4. Schweikl H., Gallorini M., Pöschl G., Urmann V., Petzel C., Bolay C., et. al.: Functions of transcription factors NF-κB and Nrf2 in the inhibition of LPS-stimulated cytokine release by the resin monomer HEMA. Dent Mater 2018; 34: pp. 1661-1678. 5. Jung S., Park O.J., Kim A.R., Ahn K.B., Lee D., Kum K.Y., et. al.: . J Microbiol 2019; 57: pp. 310-315. 6. Morgan M.J., Liu Z.G.: Crosstalk of reactive oxygen species and NF-κB signaling. Cell Res 2011; 21: pp. 103-115. 7. Marinho H.S., Real C., Cyrne L., Soares H., Antunes F.: Hydrogen peroxide sensing, signaling and regulation of transcription factors. Redox Biol 2014; 2: pp. 535-562. 8. Blaser H., Dostert C., Mak T.W., Brenner D.: TNF and ROS crosstalk in inflammation. Trends Cell Biol 2016; 26: pp. 249-261. 9. Gozzelino R., Jeney V., Soares M.P.: Mechanisms of cell protection by heme oxygenase-1. Annu Rev Pharmacol Toxicol 2010; 50: pp. 323-354. 10. Blancou P., Tardif V., Simon T., Remy S., Carreno L., Kalergis A., et. al.: Immunoregulatory properties of heme oxygenase-1. Methods Mol Biol 2011; 677: pp. 247-268. 11. Schweikl H., Buchalla W., Krifka S.: Cell responses to cariogenic microorganisms and dental resin materials — crosstalk at the dentin-pulp interface?. Dent Mater 2017; 33: pp. 514-524. 12. Engelmann J., Volk J., Leyhausen G., Geurtsen W.: ROS formation and glutathione levels in human oral fibroblasts exposed to TEGDMA and camphorquinone. J Biomed Mater Res B Appl Biomater 2005; 75: pp. 272-276. 13. Samuelsen J.T., Dahl J.E., Karlsson S., Morisbak E., Becher R.: Apoptosis induced by the monomers HEMA and TEGDMA involves formation of ROS and differential activation of the MAP-kinases p38, JNK and ERK. Dent Mater 2007; 23: pp. 34-39. 14. Gallorini M., Petzel C., Bolay C., Hiller K.A., Cataldi A., Buchalla W., et. al.: Activation of the Nrf2-regulated antioxidant cell response inhibits HEMA-induced oxidative stress and supports cell viability. Biomaterials 2015; 56: pp. 114-128. 15. Durner J., Dębiak M., Bürkle A., Hickel R., Reichl F.X.: Induction of DNA strand breaks by dental composite components compared to X-ray exposure in human gingival fibroblasts. Arch Toxicol 2011; 85: pp. 143-148. 16. Schweikl H., Petzel C., Bolay C., Hiller K.A., Buchalla W., Krifka S.: 2-Hydroxyethyl methacrylate-induced apoptosis through the ATM- and p53-dependent intrinsic mitochondrial pathway. Biomaterials 2014; 35: pp. 2890-2904. 17. Becher R., Valen H., Olderbø B.P., Bølling A.K., Samuelsen J.T.: The dental monomer 2-hydroxyethyl methacrylate (HEMA) causes transcriptionally regulated adaptation partially initiated by electrophilic stress. Dent Mater 2019; 35: pp. 125-134. 18. Bølling A.K., Samuelsen J.T., Morisbak E., Ansteinsson V., Becher R., Dahl J.E., et. al.: Dental monomers inhibit LPS-induced cytokine release from the macrophage cell line RAW264.7. Toxicol Lett 2013; 216: pp. 130-138. 19. Krifka S., Hiller K.A., Spagnuolo G., Jewett A., Schmalz G., Schweikl H.: The influence of glutathione on redox regulation by antioxidant proteins and apoptosis in macrophages exposed to 2-hydroxyethyl methacrylate (HEMA). Biomaterials 2012; 33: pp. 5177-5186. 20. Schweikl H., Klein U., Schindlbeck M., Wieczorek H.: A vacuolar-type ATPase, partially purified from potassium transporting plasma membranes of tobacco hornworm midgut. J Biol Chem 1989; 264: pp. 11136-11142. 21. Pacher P., Beckman J.S., Liaudet L.: Nitric oxide and peroxynitrite in health and disease. Physiol Rev 2007; 87: pp. 315-424. 22. Fischer R., Maier O.: Interrelation of oxidative stress and inflammation in neurodegenerative disease: role of TNF. Oxid Med Cell Longev 2015; 2015: 23. Farges J.C., Alliot-Licht B., Renard E., Ducret M., Gaudin A., Smith A.J., et. al.: Dental pulp defence and repair mechanisms in dental caries. Mediators Inflamm 2015; 2015: 24. Hong S.W., Baik J.E., Kang S.S., Yun C.H., Seo D.G., Han S.H.: . Mol Immunol 2014; 57: pp. 284-291. 25. Bulgari O., Dong X., Roca A.L., Caroli A.M., Loor J.J.: Innate immune responses induced by lipopolysaccharide and lipoteichoic acid in primary goat mammary epithelial cells. J Anim Sci Biotechnol 2017; 8: pp. 29. 26. Bølling A.K., Solhaug A., Morisbak E., Holme J.A., Samuelsen J.T.: The dental monomer hydroxyethyl methacrylate (HEMA) counteracts lipopolysaccharide-induced IL-1β release-possible role of glutathione. Toxicol Lett 2017; 270: pp. 25-33. 27. Griffith O.W., Meister A.: Potent and specific inhibition of glutathione synthesis by buthionine sulfoximine (S-n-butyl homocysteine sulfoximine). J Biol Chem 1979; 254: pp. 7558-7560. 28. Samuni Y., Goldstein S., Dean O.M., Berk M.: The chemistry and biological activities of N-acetylcysteine. Biochim Biophys Acta 2013; 1830: pp. 4117-4129. 29. Samuelsen J.T., Kopperud H.M., Holme J.A., Dragland I.S., Christensen T., Dahl J.E.: Role of thiol-complex formation in 2-hydroxyethyl- methacrylate-induced toxicity in vitro. J Biomed Mater Res A 2011; 96: pp. 395-401. 30. Schweikl H., Gallorini M., Forstner M., Petzel C., Bolay C., Hiller K.A., et. al.: Flavin-containing enzymes as a source of reactive oxygen species in HEMA-induced apoptosis. Dent Mater 2017; 33: pp. e255-e271. 31. Haddad J.J., Saadé N.E., Safieh-Garabedian B.: Redox regulation of TNF-alpha biosynthesis: augmentation by irreversible inhibition of gamma-glutamylcysteine synthetase and the involvement of an IkappaB-alpha/NF-kappaB-independent pathway in alveolar epithelial cells. Cell Signal 2002; 14: pp. 211-218. 32. Victor V.M., Rocha M., De la Fuente M.: N-acetylcysteine protects mice from lethal endotoxemia by regulating the redox state of immune cells. Free Radic Res 2003; 37: pp. 919-929. 33. Hardy M., Zielonka J., Karoui H., Sikora A., Michalski R., Podsiadły R., et. al.: Detection and characterization of reactive oxygen and nitrogen species in biological systems by monitoring species-specific products. Antioxid Redox Signal 2018; 28: pp. 1416-1432. 34. Schröder N.W., Morath S., Alexander C., Hamann L., Hartung T., Zähringer U., et. al.: . J Biol Chem 2003; 278: pp. 15587-15594. 35. Bergamini S., Rota C., Canali R., Staffieri M., Daneri F., Bini A., et. al.: N-acetylcysteine inhibits in vivo nitric oxide production by inducible nitric oxide synthase. Nitric Oxide 2001; 5: pp. 349-360. 36. Zhang J., Wang X., Vikash V., Ye Q., Wu D., Liu Y., et. al.: ROS and ROS-mediated cellular signaling. Oxid Med Cell Longev 2016; 2016: 37. Buelna-Chontal M., Zazueta C.: Redox activation of Nrf2 & NF-κB: a double end sword?. Cell Signal 2013; 25: pp. 2548-2557. 38. Ahmed S.M., Luo L., Namani A., Wang X.J., Tang X.: Nrf2 signaling pathway: pivotal roles in inflammation. Biochim Biophys Acta Mol Basis Dis 2017; 1863: pp. 585-597. 39. Napetschnig J., Wu H.: Molecular basis of NF-κB signaling. Annu Rev Biophys 2013; 42: pp. 443-468. 40. Krifka S., Petzel C., Hiller K.A., Frank E.M., Bosl C., Spagnuolo G., et. al.: Resin monomer-induced differential activation of MAP kinases and apoptosis in mouse macrophages and human pulp cells. Biomaterials 2010; 31: pp. 2964-2975. 41. Nakajima S., Kitamura M.: Bidirectional regulation of NF-κB by reactive oxygen species: a role of unfolded protein response. Free Radic Biol Med 2013; 65: pp. 162-174. 42. Ma Q.: Role of Nrf2 in oxidative stress and toxicity. Annu Rev Pharmacol Toxicol 2013; 53: pp. 401-426. 43. Guéguen Y., Bontemps A., Ebrahimian T.G.: Adaptive responses to low doses of radiation or chemicals: their cellular and molecular mechanisms. Cell Mol Life Sci 2019; 76: pp. 1255-1273. 44. Wardyn J.D., Ponsford A.H., Sanderson C.M.: Dissecting molecular cross-talk between Nrf2 and NF-κB response pathways. Biochem Soc Trans 2015; 43: pp. 621-626.

Related Articles

Leave A Comment?

You must be logged in to post a comment.