This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

Author's personal copy

N-acetyl cysteine directed detoxification of 2-hydroxyethyl methacrylate byadduct formation

Giuseppina Nocca a,1, Vincenzo D’Anto b,1, Claudia Desiderio c, Diana Valeria Rossetti a, Rosa Valletta b,Adriana Marquez Baquala a, Helmut Schweikl d, Alessandro Lupi c, Sandro Rengo b, Gianrico Spagnuolo b,*

a Institute of Biochemistry and Clinical Biochemistry, Catholic University, Rome, Italyb Department of Oral and Maxillofacial Sciences, University of Naples ‘‘Federico II’’, Napoli, Italyc Molecular Recognition Chemistry Institute, C.N.R, Rome, Italyd Department of Operative Dentistry and Periodontology, University of Regensburg, Regensburg, Germany

a r t i c l e i n f o

Article history:Received 24 November 2009Accepted 3 December 2009Available online 22 December 2009

Keywords:HEMADental monomersCytotoxicityN-acetyl cysteine

a b s t r a c t

Cytotoxicity of the dental resin monomer 2-hydroxyethyl methacrylate (HEMA) and the protective effectsof N-acetyl cysteine (NAC) on monomer-induced cell damage are well demonstrated. The aim of our studywas to analyze the hypothesis that the protection of NAC from HEMA cytotoxicity might be due to directNAC adduct formation. To this end, using HPLC we first measured the actual intracellular HEMA concen-trations able to cause toxic effects on 3T3-fibroblasts and then determined the decrease in intracellular andextracellular HEMA levels in the presence of NAC. In addition, by capillary electrophoresis coupled withmass spectrometry analysis (CE-MS), we evaluated NAC–HEMA adduct formation. HEMA reduced 3T3 cellvitality in a dose- and time-dependent manner. The concentration of HEMA inside the cells was 15–20times lower than that added to the culture medium for cell treatment (0–8 mmol/L). In the presence of10 mmol/L NAC, both intracellular and extracellular HEMA concentrations greatly decreased in conjunc-tion with cytotoxicity. NAC–HEMA adducts were detected both in the presence and absence of cells. Ourfindings suggest that the in vitro detoxification ability of NAC against HEMA-induced cell damage occursthrough NAC adduct formation. Moreover, we provide evidence that the actual intracellular concentrationof HEMA able to cause cytotoxic effects is at least one magnitude lower than that applied extracellularly.

� 2009 Elsevier Ltd. All rights reserved.

1. Introduction

Resin based materials (RBM) are routinely used in restorativedentistry, prosthodontics and orthodontics. Many in vitro studieshave shown that the polymerization reaction of RBM is nevercomplete and non-polymerized monomers, including triethyleneglycol methacrylate (TEGDMA) or 2-hydroxyethyl methacrylate(HEMA), might be released in the oral environment [1,2].

Dentin bonding techniques use adhesives or primers containingHEMA as their major component. Un-reacted free monomers seemto be directly responsible for the cytotoxicity of these materials onpulp and gingival cells, and are probably also implicated in theirallergic potential [3–5].

Despite the introduction of new HEMA-free composites into themarket in recent years, and the attempt to enhance the biocom-patibility of bonding and adhesive materials without affecting

their mechanical and clinical proprieties [6–10], the current majorchallenge in the field is the elucidation of the mechanismsresponsible for the induction of dental monomer cell damage. Inparticular, researchers have focused on the HEMA-modulatedintracellular pathways leading to cell death or survival in order toidentify molecules such as N-acetyl cysteine (NAC) which are ableto reduce monomer toxicity and to find ways to increase its usefuleffects [4].

It has been well documented that HEMA induces in vitro celldeath via apoptosis phenomena [11–13] and that various intracel-lular pathways might be involved in the toxicity of this monomer [4].It has also been shown that an increase in reactive oxygen species(ROS) levels induced by HEMA plays a major role in causing celldamage or cell cycle arrest [14,15]. Apoptosis induced by HEMAmight be mediated by ROS through a differential activation of themitogen-activated protein kinases (MAPKs) p38, JNK, and ERK[16,17].

Moreover, it has been reported that HEMA affects survivalpathways and transcription factors, including PI3k/AKT and NFkB(nuclear factor kB), leading to cell death. Mechanisms of HEMAtoxicity might be, at least partially, exerted by inhibiting the

* Corresponding author. Tel./fax: þ39 817462080.E-mail address: gspagnuo@unina.it (G. Spagnuolo).

1 These authors equally contributed to the work.

Contents lists available at ScienceDirect

Biomaterials

journal homepage: www.elsevier .com/locate/biomater ia ls

0142-9612/$ – see front matter � 2009 Elsevier Ltd. All rights reserved.doi:10.1016/j.biomaterials.2009.12.015

Biomaterials 31 (2010) 2508–2516

Author's personal copy

activation of PI3k/AKT and NFkB pathways [17,18]. In this regard,either activation or inhibition of NFkB by HEMA confirms theprotective role of this transcription factor in monomer cytotoxicity[12,18].

Recent reports have demonstrated a protective effect of NACagainst the cytotoxicity of dental resins or cements containingHEMA, TEGDMA and poly(methyl methacrylate) (PMMA); NAC isan anti-oxidant cysteine derivative that can be incorporated intocells and then deacetylated into L-cysteine. The latter representsa precursor to glutathione (GSH), a principal molecule in the non-enzymatic anti-oxidant system [4,19,20]. NAC acts not only asa direct oxidant scavenger [21], but also improves the intracellularglutathione systems compromized by oxidative stress [21,22].Recently, it has been shown that NAC exerts its effects on RBM,monomers and PMMA cell damage mainly through its anti-oxidantactivity [14,23–26], whereas other authors have suggested anadditional protective mechanism of NAC against HEMA cell damageacting through NFkB activation and cell differentiation induction[18,27]. NAC protective action can, however, include extracellularand intracellular detoxification mechanisms by removing themethacrylates [28] through adduct formation: a Michael addition(Fig. 1), due to the NAC nucleophilic properties, is in fact alsopossible under physiological conditions [29].

Hence, we hypothesized that the downstream protective effectsof NAC might also be due to a direct chemical binding between NACand HEMA, resulting in the formation of a HEMA–NAC adduct. Tothis end, using HPLC we measured the actual HEMA intracellularconcentrations able to cause toxic effects in vitro as well as thedecrease in intracellular and extracellular HEMA levels caused byNAC, and subsequently checked the HEMA–NAC adduct formationusing capillary electrophoresis coupled with mass spectrometry(CE-MS).

2. Materials and methods

2.1. Chemicals and reagents

2-Hydroxy ethyl methacrylate (HEMA), 3-(4,5 dimethyiazol-2-1)-2-5-diphenyltetrazolium bromide (MTT), NAC, cell culture medium and reagents were purchasedfrom Sigma Chemical Co. (Milan, Italy); Annexin V and a Propidium Iodide (PI) Kitwas obtained from MBL Medical & Biological Laboratories Co., Ltd. (Nagoya, Japan).

Methanol (HPLC grade, Prolabo, France) and ultra pure water (obtained bya P.Nix Power System apparatus, Human, Seoul, Korea) were used for HPLC analyses.Ammonium hydroxide solution (28%, Carlo Erba, Milan, Italy), formic acid (98%,Mallinckrodt Baker B.V., Deventer, Holland) and sodium hydroxide pellets (Merck,Darmstadt, Germany) were also utilized.

2.2. Cell culture

Mouse 3T3-fibroblasts (Swiss albino mouse cell line, Istituto Zooprofilattico,Brescia, Italy) were grown in a 5% CO2 atmosphere at 37 �C in DMEM (Dulbecco’smodified Eagle medium) with HEPES (10 mM), glucose (1.0 g/L), NaHCO3 (3.7 g/L),penicillin (100 units/mL), streptomycin (100 mg/mL) and 10% fetal calf serum.

2.3. Effect of HEMA on cell vitality

Cytotoxic concentrations of HEMA were identified by the MTT assay as previ-ously reported [30]. 3T3 cells were plated in a 96-well tissue culture dish at8000 cells/well and after 24 h of incubation, the medium was removed and the cellmonolayer was incubated with HEMA (0–8 mmol/L) for 2, 4, 8, 16 and 24 h. Themedium was replaced by 100 mL/well of a solution of MTT (0.5 mg/mL) in phos-phate-buffered saline (PBS), and the cells were incubated at 37 �C for 1 h in a 5% CO2

atmosphere. The MTT solution was replaced with 100 mL/well of DMSO and gentlyswirled for 10 min. The optical density in each well was immediately measured bya plate reader at 540 nm (Sunrise, Tecan, Switzerland). The results were expressed asthe percentage of untreated cultures. Each experiment was performed at least fivetimes in quadruplicate. Moreover, in order to determine the protective effect of NAC,cells were incubated with NAC (10 mmol/L) for 2 h and then treated with HEMA inthe presence or absence of NAC (10 mmol/L). After 8 h and 24 h of incubation time,an MTT test was performed.

In addition, flow cytometry (FACScan, Becton–Dickinson, San Jose, CA, USA) wasused to detect viability and cell death. Fibroblasts (1�105) were plated in culture

dishes (35 mm) and incubated at 37 �C for 24 h. Cells were then exposed to HEMA(0–8 mmol/L) for 24 h in the presence or absence of NAC (10 mmol/L). After treat-ment, floating and adherent cells were collected, harvested by centrifugation andthen washed once with phosphate-buffered saline. Cells were next suspended inbinding buffer (500 mL); treated and untreated cells were stained with AnnexinV – fluorescein isothiocyanate and PI and then analyzed by flow cytometry. Viablecells (no staining), apoptotic cells (Annexin Vþ), and necrotic cells (both PI/AnnexinVþ or PI alone) were detected and quantified as a percentage of the entire pop-ulation [31]. The sum of apoptotic and necrotic cells was considered to be the cell

Fig. 1. Mechanism of the reaction between N-acetylcysteine and 2-hydroxyethylmethacrylate: the attack of the nucleofilic sulphydrilic group on the terminal carbon ofthe carbon–carbon double bond causes the p electrons to move towards oxygengenerating the adduct in enolic form; the adduct then turns out to be in equilibriumwith the predominant ketonic structure.

G. Nocca et al. / Biomaterials 31 (2010) 2508–2516 2509

Author's personal copy

death population [32]. Data were analyzed using WinMDI Version 2.8 (The ScrippsResearch Institute, San Diego, CA, USA). Data from at least five independent exper-iments, performed in duplicate, were pooled to determine the mean viable cellpopulation.

2.4. Determination of intracellular and extracellular concentrations of HEMA

Sub-confluenting 3T3-fibroblasts in DMEM were incubated with HEMA(0–8 mmol/L) for various time periods (0, 2, 4, 8, 16 and 24 h); the cells were thenwashed in PBS solution and stored (�80 �C), cellular lysates were centrifuged (20,000g,15 min, 4 �C) and the supernatant was collected. The extracellular media were alsocentrifuged and filtered through a 0.45 mm syringe filter (Whatman, Maidstone Kent,UK). Both samples were then analyzed using a JASCO HPLC system (2 PU-980 pumps,UV-970 UV/VIS detector and AS-1555 autosampler). The analyses were performed ata wavelength of 220 nm with a C-18 (3 mm) Supelco reversed phase column(150� 4.7 mm) using an elution gradient of water (A) and methanol (B) starting from75% to 35% of A (30 min), 0.6 mL/min flow, 50 mL injected volume. The concentration ofHEMA in each sample was quantified using a calibration line performed with standardsolutions before each analysis. Each analysis was performed five times.

In order to determine the effect of NAC on intracellular and extracellularconcentrations of HEMA, cells were incubated with NAC (10 mmol/L) for 2 h, and thentreated with HEMA (4 and 8 mmol/L, respectively) in the presence or absence of NAC(10 mmol/L). After 8 h or 24 h of incubation time, cytosol from fibroblasts and extra-cellular media was collected as described above and analyzed by HPLC and MicellarElectrokinetic Capillary Chromatography (MEKC). Moreover, mixtures of HEMA (4 or8 mmol/L) and NAC (10 mmol/L) in DMEM without cells were prepared. Aliquoteswere prelevated after incubation (8 h or 24 h, 37 �C) and analyzed by HPLC and MEKC.

2.5. Analysis of NAC–HEMA adducts in vitro

A solution of NAC in H2O (10 mmol/L, 3 mL) was mixed with a 0.3% NH3 solution toreach pH 7.0. Subsequently, the neutralized NAC solution was incubated withcommercially pure HEMA corresponding to a final concentration of 8 mmol/L

(reaction mixture) at 37 �C [33]. This reaction mixture was then analyzed immediatelyafter mixing and after 24 h by HPLC (as described above), CE-UV and CE-MS; NAC andHEMA solutions were also analyzed under the same experimental conditions.

2.5.1. Capillary electrophoresis-UV detection (CE-UV)CE-UV analyses were performed on a capillary electrophoresis system (P/ACE�

MDQ, Beckman, Richmond, CA, USA), equipped with a UV detector (214 nm wave-length), using a fused silica capillary (50 cm effective and 60 cm total length, 50 mminner Ø, Beckman) thermostated at 35 �C. Analyses were performed both in freezone electrophoresis (CZE) and MEKC separation modes using different conditionsas specified below.

CZE background electrolyte (BGE) was 50 mmol/L ammonium acetate at pH 7.0.The running voltage was 28 kV. The samples were diluted (1:15) with water andinjected by pressure (0.5 psi) for 20 s at the anodic end of the capillary.

MEKC analyses were performed using a BGE containing sodium phosphate/sodium tetraborate buffer (44.4 and 11.1 mmol/L, respectively), pH 7.19, with36.7 mmol/L SDS as micellar phase and a final content of 10% acetonitrile (v/v).Samples were diluted (1:1) with sodium phosphate/sodium tetraborate (7.2 and1.8 mmol/L, respectively) and injected by pressure (0.5 psi) for 10 s at the anodic endof the capillary. The applied voltage was 18 kV.

After each run (both CZE and MEKC) the capillary was washed at 20 psi with H2O(4 min), NaOH (0.1 mmol/L, 2 min), acetonitrile: H2O (9:1, 5 min), NaOH (0.1 mmol/L, 2 min) and finally H2O (4 min).

2.5.2. Capillary electrophoresis - mass spectrometry (CE-MS)Flow injection (FIA-MS) and free zone electrophoresis (CZE-MS) analyses were

performed using a capillary electrophoresis automated system (Agilent Technolo-gies, Waldbronn, Germany), equipped with a diode array UV detector and externalnitrogen, coupled to the Esquire 3000 plus mass spectrometer (Bruker Daltonics,Bremen, Germany) via a coaxial sheath liquid electrospray-ionization (ESI) interface(Agilent Technologies, Waldbronn, Germany). The sheath liquid was delivered by anexternal syringe pump (Cole Palmer, Vernon Hills, IL, USA) at a constant flow rate(180 or 240 mL/h for CZE-MS or FIA experiments, respectively). Both FIA and CZE-MS

2 16 24

0

20

40

60

80

100

120

140

0 mmol/L HEMA1 mmol/L HEMA2 mmol/L HEMA

4 mmol/L HEMA6 mmol/L HEMA8 mmol/L HEMA

Time [hours]

MT

T (%

o

f c

on

tro

l)

***

***

***

***

***

***

***

*

***

***

B

A

4 8

0.6

0.4

0.2

0.0

2 4 8 16 24

Time [hours]

1 mmol/L HEMA2 mmol/L HEMA4 mmol/L HEMA6 mmol/L HEMA8 mmol/L HEMA

mm

ol/L

Fig. 2. - HEMA cytotoxic effect and intracellular concentration in 3T3-fibroblasts. (A) After 2, 4, 8, 16 and 24 h incubation period to various concentrations of HEMA, exposed BALB3T3 cells showed differences in mitochondrial dehydrogenase activity compared to control cell cultures. Each value represents the mean� SEM for 4 independent experiments.* (p< 0.05) and ** (p< 0.01) indicates significant statistical differences between treated and untreated cells. (B) HEMA intracellular concentration, after exposure to increasingquantities of monomer, was determined by HPLC after 2, 4, 8, 16 and 24 h. Concentration of HEMA was quantified using calibration curves performed with standard solutions beforeeach analysis. Bars represent mean values� standard error of the mean (SEM) from repeated experiments (n¼ 5).

G. Nocca et al. / Biomaterials 31 (2010) 2508–25162510

Author's personal copy

analyses were performed using a 50 mm I.D., 375 mm O.D. fused silica uncoatedcapillary (Composite Metal Services, Hallow, Worcs., UK, 89 cm total length). Thecapillary effective lengths were 21.5 cm for UV detection and 79.5 cm for MSdetection. New capillaries were conditioned using the following procedure: (1)water (5 bar, 2 min); (2) 0.1 mol/L sodium hydroxide (5 bar, 20 min); (3) water(5 bar, 5 min). The temperature of the CE-MS assembly cartridge was set at 25 �C.

2.5.2.1. FIA-MS analysis. The capillary electrophoresis (CE) apparatus was used asa miniaturized delivery system for introducing the sample in MS for FIA/MSexperiments as previously reported [34]. The sample was introduced into the sourcevia pressure application (1 bar applied pressure, instrumental flush function) at theinlet side of the fused silica capillary. The sheath liquid (water/methanol 30:70 v/vand formic acid 0.1% v/v) was delivered at 240 mL/h in positive ionization. Nebulizingand drying gas (nitrogen) were set at 6.0 psi and 4.0 L/min, respectively. Dry gastemperature was 300 �C. The MS scan range was 50–400 m/z using a maximumaccumulation time of 200 ms and a set target value of 50,000 and by activating theion charge control (ICC) function. Mass spectrometry capillary voltage was 4000 V inpositive ESI ionization mode.

2.5.2.2. CZE-MS analysis. CZE-MS experiments were performed by applyinga running voltage of 25 kV. Samples were injected at the anodic end at 50 mbar (10or 20 s) followed by BGE injection at 50 mbar (15 s). The BGE was constituted of25 mmol/L acetic titrated with ammonium hydroxide to pH 7.0. The sheath liquid(50 mmol/L ammonium acetate/methanol, 30:70 v/v) was delivered at 180 mL/h flowrate in positive ESI ionization mode. Nebulizing and drying gas (nitrogen) were set at5.0 psi and 4.0 L/min. Dry gas temperature was 250 �C. The MS scan range was50–350 m/z using a maximum accumulation time of 200 ms and a set target value of20,000 and activating the ion charge control (ICC) function. Mass spectrometrycapillary voltage was 4500 V in positive ESI ionization mode. Between runs thecapillary in use was rinsed with water (5 bar, 1 min) and BGE (5 bar, 1.5 min). Afterevery four runs the capillary was rinsed with water (5 bar, 0.5 min), 0.1 mol/Lsodium hydroxide (5 bar, 0.7 min) and water (5 bar, 0.7 min).

2.6. Statistical analysis

Values were expressed as the mean� SEM and the data were analyzed by one-way analysis of variance (ANOVA) followed by Tukey’s test for multiple comparisons.The level of significance was set at p< 0.05.

3. Results

3.1. Effect of HEMA on cell vitality

HEMA-treated cells did not show any significant decrease in cellvitality at incubation times lower than 8 h. After 8 h treatment, cellsincubated with 6 mmol/L HEMA showed a significant decrease(30%) in cell vitality compared to untreated cells, reaching values of50% with 8 mmol/L HEMA. However, cell vitality was decreased byHEMA in a dose- and time-dependent manner with increasingexposure periods. After 24 h incubation with 4–8 mmol/L HEMA,cell vitality was reduced from about 50% to 10% (Fig. 2A). Theanalysis of viability and cell death by flow cytometry showeda dose-dependent reduction in cell viability (Fig. 4B). After 24 h,HEMA (4–8 mmol/L) caused a statistically significant decrease inthe percentage of viable cells to about 60% (Fig. 4B).

3.2. Determination of the intracellular concentration of HEMA

Quantification of HEMA levels by HPLC over time showed thatthe concentration of HEMA inside the cells was 15–20 times lowerthan that added to the culture medium for cell treatment (Fig. 2B).

The intracellular amounts of HEMA increased in a dose-dependent manner after incubation periods from 2 h to 24 h(Fig. 2B). Sub-toxic concentrations of HEMA (1–2 mmol/L) causedonly a slight increase in the intracellular levels of the monomer(0.10–0.15 mmol/L). Moreover, compared to these sub-toxic values,a statistically significant increase in the HEMA intracellularconcentrations (0.25, 0.35, and 0.45 mmol/L) was caused by 4, 6,and 8 mmol/L HEMA, respectively after a 4 h exposure period(Fig. 2B and Table 1).

8 24

0

2

4

6

8

10

4 mmol/L HEMA4 mmol/L HEMA + NAC (24h)8 mmol/L HEMA8 mmol/L HEMA + NAC (24h)

***

***

***

***

Time [hours]

HE

MA

[m

mo

l/L

]

8 24

0.0

0.1

0.2

0.3

0.4

0.5

4 mmol/L HEMA4 mmol/L HEMA + NAC (2h)4 mmol/L HEMA + NAC (24h)8 mmol/L HEMA8 mmol/L HEMA + NAC (2h)8 mmol/L HEMA + NAC (24h)

**

**

*** ***

*

Time [hours]

HE

MA

[m

mo

l/L

]

A

B

Fig. 3. HEMA extracellular and intracellular concentration in 3T3-fibroblasts in the presence of NAC. (A) HPLC analysis of monomer extracellular concentration after treatment with4.0 or 8.0 mmol/L HEMA in the absence or presence of NAC for 8 or 24 h. (B) HPLC analysis of monomer intracellular concentration – both in the continuous presence of NAC and for2 h only – after treatment with 4.0 or 8.0 mmol/L HEMA. HEMA concentration was quantified using calibration curves performed with standard solutions before each analysis. Barsrepresent mean values� standard error of the mean (SEM) from repeated experiments (n¼ 5). *p< 0.05. **p< 0.01.

G. Nocca et al. / Biomaterials 31 (2010) 2508–2516 2511

Author's personal copy

3.3. Effect of NAC on intracellular and extracellular HEMAconcentrations

Cells pre-treated with NAC (10 mmol/L, 2 h) and then incubatedwith HEMA (4 mmol/L, 8 h) in a fresh medium without re-addingNAC showed a slight but significant decrease in the intracellularmonomer concentration (Fig. 3B). In contrast, when NAC is added tothe medium in combination with HEMA, a significant decrease inboth extracellular (Fig. 3A) and intracellular (Fig. 3B) monomerconcentrations was observed under all experimental conditions.NAC (10 mmol/L) reduced the extracellular HEMA concentrationfrom 4 to 1 mmol/L and from 8 to about 3 mmol/L after an exposureperiod of 8 h, whereas the monomer levels dropped to 0.6 and1.66 mmol/L, respectively after 24 h (Fig. 3A). Likewise, the intra-cellular HEMA concentration was reduced from 0.2 mmol/L to

0.12 mmol/L when HEMA (4 mmol/L) was added to the medium inthe presence of NAC (10 mmol/L) for 8 h; furthermore, a reductionfrom 0.45 mmol/L to 0.21 mmol/L was observed with 8 mmol/LHEMA under the same conditions, showing that the extracellulardepletion process is more efficient than the intracellular one (Fig. 3Aand B). After 24 h, 10 mmol/L NAC caused a more definite decreasein the intracellular HEMA levels (2.3-fold and 4.1-fold) when cellcultures were treated with 4 and 8 mmol/L HEMA, respectively(Fig. 3B). The effect of NAC on the kinetics of HEMA depletion inDMEM was independent of the presence of 3T3-fibroblasts (data notshown). The MTT assay and flowcytometric analysis showeda protective effect of NAC against HEMA cytotoxicity (Fig. 4A and B)when 10 mmol/L NAC was present in combination with HEMA for24 h. In contrast, a 2 h pre-incubation of cell cultures with 10 mmol/L NAC and subsequent exposure to HEMA in the absence of NAC didnot prevent monomer toxicity (Fig. 4A and B).

3.4. Detection of NAC–HEMA adducts

A NAC–HEMA adduct, obtained from a reaction mixtureprepared as indicated in the Materials and methods section, wasisolated by HPLC and identified by FIA-MS; the full scan MS spec-trum reported in Fig. 5 shows the presence of a compound with m/z316.2 [MþNa]þ. In addition, the CZE-UV electrophoretic profile ofthe reaction mixture after 24 h of incubation shows peaks corre-sponding to NAC, HEMA and the NAC–HEMA adduct (Fig. 6A). Theleft panels of Fig. 6B show the same reaction mixture analyzed byCZE-MS, indicating the extractive ion current plots. The full scan MSspectra of HEMA (m/z 131.0 [MþH]þ), NAC–HEMA adduct (m/z294.1 [MþH]þ) and NAC (m/z 164.1, [MþH]þ) are presented in theright panels. The MS spectrum of the adduct (Fig. 6B) showedadditional ions at m/z 276.1 and m/z 316.1 over the ion at m/z 294.1,presumably corresponding to the loss of a water moiety [MþH–H2O]þ and to the addition of an Naþ ion [MþNa]þ, respectively.Finally, the compounds present in the reaction mixture were usedas standards for the MECK analysis; the electrophoretic profileobtained was similar to that observed using CZE (Fig. 7A). Fig. 7Band C depict the MEKC electropherograms of extracellular mediumand cellular lysates showing the signals of HEMA, NAC and NAC–HEMA adduct. The presence of NAC in a solution containing HEMAcaused a clear reduction in the HEMA signal in the extracellularmedium (Fig. 7B). In contrast, such a reduction was not so prom-inent in cell lysates (Fig. 7C). Moreover, the signal from the NAC–HEMA adduct detected in extracellular medium was higher thanthat present in the analysis of the intracellular fluid (Fig. 7B and C).

4. Discussion

In recent years, major efforts have been made by researches inthe field of dental materials toxicology to elucidate the mechanismsinvolved in the adverse effects caused by monomers released fromresin based materials, and to find effective strategies to reducethese negative effects. Particular attention has focused on mole-cules such as NAC or anti-oxidants, which are able to protect thecells from monomer-induced cell damage. HEMA has been themost studied methacrylic dental monomer since it is a majorcomponent in RBM, adhesives and bondings.

Here, as in previous studies, we confirmed the cytotoxicity ofHEMA using 3T3-fibroblasts as the cell model since it is recom-mended by the International Standard Organization (ISO) for the invitro evaluation of materials [35]. As expected, our results showeddose-dependent toxic effects of HEMA on cell cultures in bothassays used. To date, different mechanisms underlying the adverseeffects of RBM and dental monomers have been suggested. HEMAreleased from resin materials might cause DNA damage that results

0 4 8

0

50

100

150

no NACNAC 10 mM 2hours

***

**

NAC 10 mM 24hours

HEMA [mM]

MT

T (%

o

f c

on

tro

l)

0 4 8 10

0

20

40

60

80

100

no NACNAC 10 mM 2hours

*** ***

NAC 10 mM 24hours

HEMA [mM]

% v

ia

ble

c

ells

A

B

Fig. 4. Effect of NAC on HEMA toxicity. 3T3-fibroblasts were treated with 0–10 mmHEMA in the presence or absence of various NAC concentrations for 2 h or 24 h. (A)Mitochondrial dehydrogenase activity and (B) viability were quantified as described inSection 2. Results represent the means� SEM of at least four independent experi-ments. ** (p< 0.01) and *** (p< 0.001) indicates significant differences betweentreated and untreated cell cultures.

Table 1Statistical analysis of the above-reported intracellular HEMA concentrations.p< 0.05; 0.01 and 0.001 indicate significantly different from control.

2 h p 4 h p 8 h p 16 h p 24 h p

1 vs. 2 mmol/L > 0.05 > 0.05 > 0.05 > 0.05 > 0.051 vs. 4 mmol/L < 0.05 < 0.01 < 0.01 < 0.001 < 0.011 vs. 6 mmol/L < 0.001 < 0.001 < 0.001 < 0.001 < 0.0011 vs. 8 mmol/L < 0.001 < 0.001 < 0.001 < 0.001 < 0.0012 vs. 4 mmol/L > 0.05 < 0.05 > 0.05 < 0.05 > 0.052 vs. 6 mmol/L < 0.001 < 0.001 < 0.001 < 0.001 < 0.0012 vs. 8 mmol/L < 0.001 < 0.001 < 0.001 < 0.001 < 0.0014 vs. 6 mmol/L > 0.05 < 0.01 > 0.05 < 0.01 < 0.014 vs. 8 mmol/L < 0.001 < 0.001 < 0.001 < 0.001 < 0.0016 vs. 8 mmol/L > 0.05 < 0.01 < 0.05 < 0.01 < 0.05

G. Nocca et al. / Biomaterials 31 (2010) 2508–25162512

Author's personal copy

in decreased cell proliferation [15,23,36]. Resin monomers causedan increase in intracellular ROS as hydrogen peroxide, superoxideanions and hydroxyl radicals, followed by a subsequent reductionin the intracellular level of anti-oxidant molecules like GSH[37–39]. Such phenomena may induce cytotoxicity by modulatingthe signaling pathways leading to cell death through necrosis orapoptosis [4,12,40].

However, no information about the effective intracellularconcentrations of HEMA which cause these diverse cell responseshas been available thus far. The present investigation providesinteresting experimental evidence regarding the in vitro behavior ofHEMA and its cellular uptake. The highest intracellular concentra-tion of HEMA (variable as a function of the quantity of monomerused) was always reached after 2 h of incubation, whereas the toxic

86.4 132.4 252.6

316.2

338.1

+MS, 0.5-0.7min (#85-#133), 100%=1256407

0.00

0.25

0.50

0.75

1.00

1.25

x106

Inte

ns.

50 100 150 200 250 300 350m/z

Fig. 5. Relative full scan MS spectrum of the adduct with m/z (316.2 [Hþ] Naþ ion), isolated through HPLC.

EIC 164 +All

EIC 131 +All

EIC 294 +All

0.000.250.50x105

Inte

ns.

012

x105

0.000.250.500.75x106

2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0Time [min]

164.1

+MS, 13.9min131.0

+MS, 7.1min

276.1 294.1316.1

338.1

+MS, 11.0-11.1min

0

2x104

Inte

ns.

0.0

0.5

x105

0

1

x105

50 100 150 200 250 300m/z

NAC

B

HEMA

A

ADDUCT

HEMA

NAC

ADDUCT

0.014

0.012

0.010

0.008

0.006

0.004

Au

0.002

0.000

-0.0020

Relative -- time: -4.858 Minutes

5 10 15

Fig. 6. CZE-UV electrophoretic profile of the reaction mixture after 24 h of incubation (A). Extracted ion current CZE-MS electropherograms and full scan MS spectra of HEMA(m/z 131.0 [Hþ]), HEMA–NAC adduct (m/z 294.1 [Hþ]) and NAC (m/z 164.1, [Hþ]) (B).

G. Nocca et al. / Biomaterials 31 (2010) 2508–2516 2513

Author's personal copy

Fig. 7. (A) Electrophoretic profiles of NAC (brown), HEMA (blue) and mixture of reaction (red) after 24 h of incubation obtained by MEKC. (B) Electrophoretic profiles of 3T3extracellular medium obtained after 2 h of incubation with HEMA in the presence (red) or absence (blue) of NAC. (C) Electrophoretic profiles of 3T3 intracellular fluid obtained after2 h of incubation with HEMA in the presence (red) or absence (blue) of NAC.

Author's personal copy

effects were detected after 8 h for monomer concentrations from4 mmol/L. The HEMA concentration inside 3T3-fibroblasts wasabout 15–20 times lower than that available in the extracellularculture medium. Interestingly, the toxic effects occurred whenintracellular HEMA concentrations reached 0.2 mmol/L which,therefore, can be considered as a threshold. Obviously, thisthreshold concentration is related to 3T3 cells and could be differentfor other cell lines. Previous in vitro investigations on HEMA andother dental monomers showed that the concentrations whichcaused a toxic effect depend on the cells used for testing [41]. Thesedifferences between cell lines might be related to structural phys-ical–chemical properties of each cell phenotype as well as theirintracellular ability to detoxify HEMA.

We also showed that the toxicity of HEMA on 3T3 cells wasstrongly reduced in the presence of 10 mmol/L NAC. This observa-tion is consistent with the ability of NAC to decrease monomer-induced cell damage as widely demonstrated in previous in vitrostudies using various cell types [4]. Yet, the mechanism behind theprotective role of NAC on cell viability in the presence of resinmonomers is still unknown at this time. In general, NAC can act asthe supply source of GSH to compensate for its depletion caused byexcessive amounts of ROS, maintaining the balance of the anti-oxidant redox system [42]. In addition, the NAC sulfhydryl groupplays a critical role in direct scavenging of free radicals, includingROS, and serves as a nucleophilic center that can react with theelectrophilic b-carbon in methacrylates through a Michael-typereaction [43,44]. NAC has been shown to inhibit cell death medi-ated by HEMA through its anti-oxidant activity since the involve-ment of ROS and oxidative stress in HEMA-mediated apoptotic celldeath has been previously demonstrated [4,14,23]. However, it hasalso been reported that NAC protects various types of cells fromdeath or loss of function induced by HEMA, in conjunction withmodulation of NFkB and c-Jun N-terminal kinase (JNK) [4,18].

Recently, incorporation of NAC into PMMA-based dental resin hasbeen shown to restore the suppressed viability and function of dentalpulp cells or oral fibroblasts on the resin substrate to a biologicallysignificant degree [45–47]. Moreover, it has been demonstrated thatpre-reacting the MMA-based bone cement monomer with NACcompletely reduced adverse biological events caused by the resinmonomers after a reaction time of 24 h [25]. In our study, theprotective effects of NAC on cell viability were correlated toa decrease in both intracellular and extracellular HEMA concentra-tions (Fig. 3). In particular, the decrease in extracellular HEMAconcentrations indicates that NAC reacts with this monomer prob-ably through Michael addition, reducing its entry into the cells. Incontrast, pre-incubation of 3T3 cells with NAC for 2 h followed bya treatment with HEMA in the absence of NAC did not cause anysignificant reduction in intracellular or extracellular monomerconcentrations. This result can explain the lack of protection exertedby NAC when it is only used for 2 h pre-incubation (Fig. 4A and B),suggesting that after removing NAC, its intracellular amount is nothigh enough to act as a cellular glutathione supply source to enhancethe resistance to monomer-induced oxidative stress. Conversely, thepresence of NAC together with HEMA caused a strong decrease inintracellular and extracellular HEMA concentrations, indicatinga direct and continuous interaction of NAC with the monomer. Ourresults are in line with a previous report which demonstrated thatthe ability of NAC to act as an anti-oxidant decreased with increasingmonomer-NAC reaction time. It was concluded that a time-depen-dent depletion of NAC’s sulfhydryl group was caused by a directchemical reaction between NAC and the monomer [25].

Based on our findings, we provide the first evidence of a directinteraction between NAC and a dental resin monomer resulting inthe formation of an adduct. Products of the HEMA–NAC reactionwere analyzed in the presence and absence of cells by various

analytical methodologies including HPLC and CE using both UV andMS detection. The CE conditions were optimized according to thephysico-chemical properties of HEMA, NAC and sample matrix.

The first analysis, performed in CZE mode and positive polarity,allowed a successful separation of the reaction mixture compo-nents and revealed a signal that was possible to ascribe to theHEMA–NAC adduct by the mass spectrometry detection both in FIAand in coupling with CZE separation. Unfortunately, these oper-ating conditions were not suitable for the analysis of the NAC–HEMA reaction products in culture medium due to the presence ofmatrix interference in the electropherogram.

However, the use of MECK allowed the successful separation ofthe products in the reaction mixture, the cell lysate and finally inthe 3T3 extracellular medium. As a matter of fact, with sucha technique, the presence of the micellar phase in the bufferintroduces an additional separation mechanism and a solubiliza-tion effect on the biological materials.

Since MECK is not compatible with mass spectrometry and inorder to confirm the formation of the adduct also in culture medium,the samples were analyzed by HPLC. The fraction presumably con-taining the adduct was collected and then injected directly into massspectrometry by FIA. In this way it was possible to confirm the pres-ence of a NAC–HEMA adduct even in culture medium. The samefraction was used also as a standard for MECK analysis. These resultsdemonstrated the formation of the adduct in the NAC–HEMA mixtureeither in the presence or absence of cells. The NAC–HEMA reaction israpid [25] since the adduct signal appears after only 2 h of incubation,both in extracellular medium and in cell lysates (data not shown).Interestingly, the NAC–HEMAextracellular adduct signal was strongerthan the intracellular one, indicating a pre-reaction between NAC andHEMA in the extracellular environment and suggesting a NAC-directed detoxification of HEMA predominantly outside the cells.

5. Conclusions

To date, several protective biological downstream effects ofNAC have been suggested, whereas no experimental evidence forthe presence of a direct chemical interaction between NAC andmethacrylates has been reported. Our results demonstrate that thein vitro detoxification ability of NAC against HEMA-induced celldamage might occur through the formation of a NAC–HEMAadduct. Moreover, under these experimental conditions, weprovide original information regarding the actual intracellularconcentrations of HEMA able to cause cytotoxic effects. Theseresults, together with related observations presented by others,have shed light on the mechanisms underlying the detoxificationof HEMA by NAC. These findings can aid in the reasonable design ofmaterials with considerably improved characteristics of safety andbiocompatibility.

Appendix

Figure with essential color discrimination. Figs. 2–4 and 7 in thisarticle have parts that are difficult to interpret in black and white.The full color images can be found in the on-line version, at doi:10.1016/j.biomaterials.2009.12.015.

References

[1] Gerzina TM, Hume WR. Diffusion of monomers from bonding resin–resincomposite combinations through dentine in vitro. J Dent 1996;24:125–8.

[2] Hume WR, Gerzina T. Bioavailability of components of resin-based materialswhich are applied to teeth. Crit Rev Oral Biol Med 1996;7:172–9.

[3] Ferracane JL. Elution of leachable components from composites. J Oral Rehabil1994;21:441–52.

G. Nocca et al. / Biomaterials 31 (2010) 2508–2516 2515

Author's personal copy

[4] Schweikl H, Spagnuolo G, Schmalz G. Genetic and cellular toxicology of dentalresin monomers. J Dent Res 2006;85:870–7.

[5] Kanerva L, Pelttari M, Jolanki R, Alanko K, Estlander T, Suhonen R. Occupa-tional contact urticaria from diglycidyl ether of bisphenol a epoxy resin.Allergy 2002;57:1205–7.

[6] Torkabadi S, Nakajima M, Ikeda M, Foxton RM, Tagami J. Bonding durability ofHEMA-free and HEMA-containing one-step adhesives to dentine surroundedby bonded enamel. J Dent 2008;36:80–6.

[7] Ikeda T, De Munck J, Shirai K, Hikita K, Inoue S, Sano H, et al. Effect of air-drying and solvent evaporation on the strength of HEMA-rich versus HEMA-free one-step adhesives. Dent Mater 2008;24:1316–23.

[8] De Munck J, Arita A, Shirai K, Van Landuyt KL, Coutinho E, Poitevin A, et al.Microrotary fatigue resistance of a HEMA-free all-in-one adhesive bonded todentin. J Adhes Dent 2007;9:373–9.

[9] Xie D, Chung ID, Wu W, Mays J. Synthesis and evaluation of HEMA-free glass-ionomer cements for dental applications. Dent Mater 2004;20:470–8.

[10] Ohba M, Manabe A, Itoh K, Hisamitsu H, Wakumoto S. 2-HEMA-free dentinbonding system to prevent contraction gap. Dent Mater J 1998;17:83–8.

[11] Paranjpe A, Bordador LC, Wang MY, Hume WR, Jewett A. Resin monomer2-hydroxyethyl methacrylate (HEMA) is a potent inducer of apoptotic celldeath in human and mouse cells. J Dent Res 2005;84:172–7.

[12] Spagnuolo G, Mauro C, Leonardi A, Santillo M, Paterno R, Schweikl H, et al. NF-kappaB protection against apoptosis induced by HEMA. J Dent Res2004;83:837–42.

[13] Samuelsen JT, Holme JA, Becher R, Karlsson S, Morisbak E, Dahl JE. HEMA reducescell proliferation and induces apoptosis in vitro. Dent Mater 2008;24:134–40.

[14] Spagnuolo G, D’Anto V, Cosentino C, Schmalz G, Schweikl H, Rengo S. Effect ofN-acetyl-L-cysteine on ROS production and cell death caused by HEMA inhuman primary gingival fibroblasts. Biomaterials 2006;27:1803–9.

[15] Chang HH, Guo MK, Kasten FH, Chang MC, Huang GF, Wang YL, et al. Stim-ulation of glutathione depletion, ROS production and cell cycle arrest ofdental pulp cells and gingival epithelial cells by HEMA. Biomaterials 2005;26:745–53.

[16] Samuelsen JT, Dahl JE, Karlsson S, Morisbak E, Becher R. Apoptosis induced bythe monomers HEMA and TEGDMA involves formation of ROS and differ-ential activation of the MAP-kinases p38, JNK and ERK. Dent Mater 2007;23:34–9.

[17] Spagnuolo G, D’Anto V, Valletta R, Strisciuglio C, Schmalz G, Schweikl H, et al.Effect of 2-hydroxyethyl methacrylate on human pulp cell survival pathwaysERK and AKT. J Endod 2008;34:684–8.

[18] Paranjpe A, Cacalano NA, Hume WR, Jewett A. N-acetyl cysteine mediatesprotection from 2-hydroxyethyl methacrylate induced apoptosis via nuclearfactor kappa B-dependent and independent pathways: potential involvementof JNK. Toxicol Sci 2009;108:356–66.

[19] Lean JM, Davies JT, Fuller K, Jagger CJ, Kirstein B, Partington GA, et al. A crucialrole for thiol antioxidants in estrogen-deficiency bone loss. J Clin Invest2003;112:915–23.

[20] Meister A, Anderson ME. Glutathione. Annu Rev Biochem 1983;52:711–60.[21] Gillissen A, Bartling A, Schoen S, Schultze-Werninghaus G. Antioxidant func-

tion of ambroxol in mononuclear and polymorphonuclear cells in vitro. Lung1997;175:235–42.

[22] Gillissen A, Nowak D. Characterization of N-acetylcysteine and ambroxol inanti-oxidant therapy. Respir Med 1998;92:609–23.

[23] Schweikl H, Hartmann A, Hiller KA, Spagnuolo G, Bolay C, Brockhoff, et al.Inhibition of TEGDMA and HEMA-induced genotoxicity and cell cycle arrest byN-acetylcysteine. Dent Mater 2007;23:688–95.

[24] Yamada M, Kojima N, Att W, Hori N, Suzuki T, Ogawa T. N-Acetyl cysteinerestores viability and function of rat odontoblast-like cells impaired by poly-methylmethacrylate dental resin extract. Redox Rep 2009;14:13–22.

[25] Yamada M, Ogawa T. Chemodynamics underlying N-acetyl cysteine-mediatedbone cement monomer detoxification. Acta Biomater 2009;5:2963–73.

[26] Tsukimura N, Yamada M, Aita H, Hori N, Yoshino F, Chang-Il Lee M, et al. N-acetyl cysteine (NAC)-mediated detoxification and functionalization of pol-y(methyl methacrylate) bone cement. Biomaterials 2009;30:3378–89.

[27] Paranjpe A, Cacalano NA, Hume WR, Jewett A. N-acetylcysteine protects dentalpulp stromal cells from HEMA-induced apoptosis by inducing differentiationof the cells. Free Radic Biol Med 2007;43:1394–408.

[28] De Flora S, Cesarone CF, Balansky RM, Albini A, D’Agostini F, Bennicelli C, et al.Chemopreventive properties and mechanisms of N-acetylcysteine. Theexperimental background. J Cell Biochem 1995;22:33–41.

[29] Barshteyn N, Elfarra AA. Formation of three N-acetyl-L-cysteine monoadductsand one diadduct by the reaction of S-(1,2-dichloroviny l)-L-cysteine sulfoxidewith N-acetyl-L-cysteine at physiological conditions: chemical mechanismsand toxicological implications. Chem Res Toxicol 2007;20:1563–9.

[30] Mosmann T. Rapid colorimetric assay for cellular growth and survival:application to proliferation and cytotoxicity assays. J Immunol Methods1983;65:55–63.

[31] Vermes I, Haanen C, Steffens-Nakken H, Reutelingsperger C. A novel assay forapoptosis: flow cytometric detection of phosphatidylserine expression onearly apoptotic cells using fluorescein labelled annexin V. J Immunol Methods1995;184:39–51.

[32] Zamai L, Canonico B, Luchetti F, Ferri P, Melloni E, Guidotti L, et al. Supravitalexposure to propidium iodide identifies apoptosis on adherent cells. Cytom-etry 2001;44:57–64.

[33] McCarthy TJ, Hayes EP, Schwartz CS, Witz G. The reactivity of selected acry-lated esters toward glutathione and deoxyribonucleosidi in vitro: structure–activity relationships. Fundam Appl Toxicol 1994;22:543–8.

[34] Desiderio C, Mancinelli A, De Rossi A, Rossetti DV, Inzitari R, Messana I, et al.Rapid determinetion of short chain carnitines in human plasma by electro-spray ionisation-ion trap mass spectrometry using capillary electrophoresisinstrumnet as sampler. J Chromatogr A 2007;1150:320–6.

[35] International Organization for Standardization, ISO 7450: dentistry-preclinicalevaluation of biocompatibility of medical devices used in dentistry-testmethods for dental materials. Geneve, 1997.

[36] Schweikl H, Schmalz G, Spruss T. The induction of micronuclei in vitro byunpolymerized resin monomers. J Dent Res 2001;80:1615–20.

[37] Stanislawski L, Lefeuvre M, Bourd K, Soheili-Majd E, Goldberg M, Perianin A.TEGDMA-induced toxicity in human fibroblasts is associated with an early anddrastic GSH depletion with subsequent production of oxygen reactive species.J Biomed Mater Res A 2003;66:476–82.

[38] Engelmann J, Leyhausen G, Leibfritz D, Geurtsen W. Effect of TEGDMA on theintracellular glutathione concentration of human gingival fibroblasts.J Biomed Mater Res 2002;63:746–51.

[39] Volk J, Engelmann J, Leyhausen G, Geurtsen W. Effects of three resin mono-mers on the cellular glutathione concentration of cultured human gingivalfibroblasts. Dent Mater 2006;22:499–505.

[40] Spagnuolo G, Galler K, Schmalz G, Cosentino C, Rengo S, Schweikl H. Inhibitionof phosphatidylinositol 3-kinase amplifies TEGDMA induced apoptosis inprimary human pulp cells. J Dent Res 2004;83:703–7.

[41] Geurtsen W, Lehmann F, Spahl W, Leyhausen G. Cytotoxicity of 35 dental resincomposite monomers/additives in permanent 3T3 and three human primaryfibroblast cultures. J Biomed Mater Res 1998;41:474–80.

[42] Zafarullah M, Li WQ, Sylvester J, Ahmad M. Molecular mechanisms ofN-acetylcysteine actions. Cell Mol Life Sci 2003;60:6–20.

[43] Schweikl H, Schmalz G. Triethylene glycol dimethacrylate induces largedeletions in the hprt gene of V79 cells. Mutat Res 1999;438:71–8.

[44] Geurtsen W, Leyhausen G. Chemical–biological interactions of the resinmonomer triethyleneglycol-dimethacrylate (TEGDMA). J Dent Res 2001;80:2046–50.

[45] Yamada M, Kojima N, Paranjpe A, Att W, Aita H, Jewett A, et al. N-acetyl cysteine(NAC)-assisted detoxification of PMMA resin. J Dent Res 2008;87:372–7.

[46] Kojima N, Yamada M, Paranjpe A, Tsukimura N, Kubo K, Jewett A, et al. Restoredviability and function of dental pulp cells on polymethylmethacrylate (PMMA)-based dental resin supplemented with N-acetyl cysteine (NAC). Dent Mater2008;24:1686–93.

[47] Att W, Yamada M, Kojima N, Ogawa T. N-acetyl cysteine prevents suppressionof oral fibroblast function on poly(methylmethacrylate) resin. Acta Biomater2009;5:391–8.

G. Nocca et al. / Biomaterials 31 (2010) 2508–25162516