Methylparaben potentiates UV-induced damage of skin keratinocytes

Osamu Handa , Satoshi Kokura

d a a, ,

, Satoko Adachi , Tomohisa Takagi , Yuji Naito , Toru Tanigawa ,

b

Norimasa Yoshida and Toshikazu Yoshikawa

a b c

Department of Biomedical Safety Science, Kyoto Prefectural University of Medicine, Kyoto 602-8566, Japan Department of Inflammation and Immunology, Kyoto Prefectural University of Medicine, Kyoto 602-8566, Japan

Department of Medical Proteomics, Kyoto Prefectural University of Medicine, Kyoto 602-8566, Japan Department of Molecular Gastroenterology and Hepatology, Kyoto Prefectural University of Medicine, Kyoto Department of Environmental Systems Science, Faculty of Engineering, Doshisha University, Kyoto 610-0321,

602-8566, Japan
e

Japan Received 16 May 2006; revised 11 July 2006; accepted 12 July 2006. Available online 28 July 2006.

Abstract
For many years, methylparaben (MP) has been used as a preservative in cosmetics. In this study, we investigated the effects of ultraviolet-B (UVB) exposure on MP-treated human skin keratinocytes. HaCaT keratinocyte was cultured in MP-containing medium for 24 h, exposed to UVB (15 or 30 mJ/cm ) and further cultured for another 24 h. Subsequent cellular viability was quantified by MTT-based assay and cell death was qualified by fluorescent microscopy and flow cytometry. Oxidative stress, nitric oxide (NO) production and cellular lipid peroxidation were measured using fluorescent probes. In addition, activation of nuclear factor kappa B and activator protein-1 was assessed by electro-mobility gel-shift assay. Practical concentrations of MP (0.003%) had a little or no effect on cellular viability, oxidative stress, NO production, lipid peroxidation and activation of nuclear transcription factors in HaCaT keratinocytes. Low-dose UVB also had little or no effect on these parameters in HaCaT keratinocytes. However, UVB exposure significantly increased cell death, oxidative stress, NO production, lipid peroxidation and activation of transcription factors in MP-treated HaCaT keratinocytes. These results indicate that MP, which has been considered a safe preservative in cosmetics, may have harmful effects on human skin when exposed to sunlight. Keywords: Methylparaben; Ultra-violet; Transcription factor; Apoptosis; Oxidative stress Abbreviations: AP-1, activator protein-1; DAF-DA, 4-amino-5-methylamino-2,7-difluorescein diacetate; DCFDA, 5-(and 6)-carboxy-2,7-dichlorodihydrofluorescein diacetate; DHR123, dihydrorhodamine-123; DMEM, Dulbecco's modified essential medium; DPPP, diphenyl-1-pyrenylphosphine; EMSA, electro-mobility gel shift assay; FBS, fetal bovine serum; FITC, fluorescein isothiocyanate; HO342, Hoechst 33432; MP, methylparaben; NFB, nuclear factor kappa B; NO, nitric oxide; PBS, phosphate-buffered saline; PI, propidium iodide; ROS, reactive oxygen species; UVA, ultra-violet A; UVB, ultra-violet B; UVC, ultra-violet C
2

Article Outline
1. Introduction 2. Materials and methods 2.1. Reagents 2.2. Human normal keratinocytes

2.3. UVB irradiation 2.4. Cell viability assay 2.5. Microscopic analysis for dead cells 2.6. Flowcytometric quantification of dead cells 2.7. Production of reactive oxygen species (ROS) and nitric oxide (NO) 2.8. Electro-mobility gel-shift assay (EMSA) for nuclear transcription factors 2.9. Lipid peroxidation in HaCaT keratinocytes 2.10. Electron paramagnetic resonance (EPR) study 2.11. Statistical analysis 3. Results 3.1. Cytotoxic effects of MP on HaCaT keratinocytes 3.2. MP-induced production of ROS and NO 3.3. Cytotoxic effects of UVB irradiation on HaCaT keratinocytes 3.4. MP enhances UVB irradiation-induced cell death 3.5. MP enhances ROS and NO production in HaCaT keratinocytes induced by UVB irradiation 3.6. MP enhances NFB and AP-1 activation of HaCaT keratinocytes induced by UVB irradiation 3.7. MP enhances lipid peroxidation of HaCaT keratinocytes induced by UVB irradiation 3.8. EPR study 4. Discussion References

1. Introduction
Various additives and preservatives are used in cosmetics, foods and medicines in order to prevent deterioration. The parabens have been widely used as preservatives in cosmetics as well as in foods and medicines for more than 50 years due to its broad antimicrobial function. Methylparaben (MP) is a methyl ester of p-hydroxybenzoic acid (Fig. 1) and is one of a homologous series of parabens (including methyl, ethyl, butyl, heptyl and benzyl parabens). MP is one of the most commonly used preservatives in cosmetics (Mowad, 2000). MP meets several of the criteria for an ideal preservative; it exhibits a broad spectrum of antimicrobial activity, is colorless, odorless (slight burnt flavor), stable over the pH range, resistant to hydrolysis in hot and cold water (autoclaveable) (Alexander et al., 1978), non-volatile, has a low tendency towards absorption in commonly used plastics and primary packing materials, and is cheap. Moreover, MP has been shown to have low acute- and long-term toxicity, has no carcinogenic activity ([Rodrigues et al., 1986] and [Matthews et al., 1956]) and is not mutagenic (Kawachi et al., 1980). Therefore, it was concluded that practical concentrations of MP have little or no toxicity either in vivo or in vitro (Soni et al., 2002). Although MP has been reported to be readily and completely absorbed through the skin and from the gastrointestinal tract, and shows no evidence of accumulation (Soni et al., 2002), recent studies have reported that small amounts of MP remain unhydrolyzed in the epidermis ([Oh et al., 2002] and [Cross and Roberts, 2000]) and body tissue (Darbre et al., 2004). Therefore, it is possible that MP remaining in the epidermis influence keratinocytes.

Full-size image (4K)

Fig. 1. Methylparaben (MP) is a methyl ester of p-hydroxybenzoic acid.

The skin is more or less exposed to sunlight as part of daily life, in which the effects of ultraviolet radiation are not negligible, and many people use cosmetics that contain MP. Although MP has been considered to be safe and is used widely as a preservative in cosmetics, the effects of MP on skin under the sunlight is unknown. In the

present study, we investigated the effects of ultraviolet-B (UVB) exposure on MP-treated human skin keratinocytes in vitro.

2. Materials and methods

2.1. Reagents
Methyl 4-hydroxybenzoate (methylparaben; MP), Hoechst 33432 (HO342) and propidium iodide (PI) were purchased from Sigma (St. Louis, MO, USA). Dihydrorhodamine-123 (DHR123), 5-(and 6)-carboxy-2,7dichlorodihydrofluorescein diacetate (DCF-DA) and 4-amino-5-methylamino-2,7-difluorescein diacetate (DAFDA) were from Molecular Probes (Eugene, OR, USA). Diphenyl-1-pyrenylphosphine (DPPP) was from Dojindo (Kumamoto, Japan). Ethanol and other reagents were from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). MP was dissolved in 99.5% ethanol and added at a final concentration of 0.003%, 0.03% or 0.3% (0.197, 1.97 or 19.7 mM, respectively) to cell culture medium (final concentration of ethanol was 0.1%, which had no effect on cell viability within 72 h; data not shown), as described below. As a control group, cell culture medium containing 0.1% ethanol was used.

2.2. Human normal keratinocytes

HaCaT keratinocytes are a spontaneously immortalized human keratinocyte line (Boukamp et al., 1988), and were kindly supplied by Dr. Kato of the Department of Dermatology of Kyoto Prefectural University of Medicine (Kyoto, Japan). HaCaT keratinocytes were grown in 80 cm cell culture flasks and maintained in Dulbecco's modified essential medium (DMEM; Gibco-BRL, Gaithersburg, MD, USA) supplemented with 5% fetal bovine serum (FBS; Equitech-Bio Inc., Kerrville, TX, USA), 2 mM glutamine and 100 U/ml penicillin/streptomycin (Gibco) at 37 C in a humidified atmosphere containing 5% CO2. Cells were passaged every 7 days at a 1:10 split.
2

2.3. UVB irradiation

The UVB source was a bank of six fluorescent sunlamps (FL-20SE-30; Toshiba Medical Supply, Tokyo, Japan) with an emission spectrum of 275375 nm, mainly in the UVB range, peaking at 305 nm and including a small amount of UVA and UVC (energy: UVA, 30%; UVB, 54%; UVC, 0.2%). The irradiance of UVB was measured by a UV-radiometer (UVR-3036/S2; TOPCON corporation, Tokyo, Japan). In some experiments, MP-treated or untreated HaCaT keratinocytes were exposed to UVB (15 or 30 mJ/cm ). Prior to UVB irradiation, medium was decanted and replaced with phosphate-buffered saline (PBS). After irradiation, cells were incubated in cell culture medium without MP at 37 C for various time periods.
2

2.4. Cell viability assay

Cell viability was measured using a MTT-based WST-1 assay kit (Dojin Laboratory, Kumamoto, Japan) according to the manufacturer's instructions. Briefly, HaCaT keratinocytes were cultured in 96-well plates until confluence and were incubated for 6 or 24 h in the absence or presence of MP. Cells were washed twice with PBS and WST-1 solution was added. Cells were then incubated for 1 h at 37 C and the optical density of each well was read at 455 nm using a micro-plate reader (MPR-A4i; Tosoh corporation, Tokyo, Japan). Cell viability was expressed as a percentage of untreated control cultures.

2.5. Microscopic analysis for dead cells

The type of cell death (apoptosis or necrosis) induced by MP and/or UVB was determined by fluorescent microscopy after staining with Hoechst 33342 (HO342) and propidium iodide (PI), as described by Naito et al. (2002). Cells were grown in 35 mm cell culture dishes until confluence. After treatment with MP for 24 h, cells were exposed to UVB and then incubated for a further 24 h. The medium was removed and cells were washed twice with PBS and incubated with 10 g/ml HO342 dye for 15 min at 37 C and with 10 g/ml PI for 10 min at 37 C. Dual-stained cells were examined using an IX70-23FL/DIC-SP inverted fluorescence microscope (Olympus, Tokyo, Japan). Photographic images (MPEG format) were taken from four random fields. Live cells, viable cells and early stage apoptotic cells, which have cell membrane function, take up blue dye (HO342).

Apoptosis was characterized morphologically by condensed chromatin. Red-stained cells (PI) were considered late apoptotic (condensed chromatin) or necrotic cells.

2.6. Flowcytometric quantification of dead cells

Quantitative analysis of cell death was performed by flow cytometry. HaCaT keratinocytes were cultured in 35 mm cell culture dishes until confluence. After treatment with MP for 24 h, cells were exposed to UVB and incubated for a further 24 h. Subsequently, medium was removed and cells were washed twice with PBS, collected by centrifugation and pellets were resuspended in PBS. Cells were stained with 10 g/ml PI and were analyzed using a FACS Calibur (Becton Dickinson and Company: Franklin Lakes, NJ, USA). PI positive cells include late apoptotic cells and necrotic cells.

2.7. Production of reactive oxygen species (ROS) and nitric oxide (NO)
HaCaT keratinocytes were cultured in 35 mm cell culture dishes until confluence. After treatment with MP for 24 h, cells were exposed to UVB and were incubated for a further 1 h. Subsequently, medium was removed and cells were washed twice with PBS. Cells were then incubated with dihydrorhodamine-123 (DHR123) at a final concentration of 20 M for 15 min, with 5-(and 6)-carboxy-2,7-dichlorodihydrofluorescein diacetate (DCF-DA) at a final concentration of 50 M for 30 min or with 4-amino-5-methylamino-2,7-difluorescein diacetate (DAF-DA) at a final concentration of 20 M for 15 min. Quantitative analysis was performed using a fluorescent micro-plate reader (Fluoromark; Bio-Rad Laboratories, Hercules, CA, USA) or by flow cytometry (FACS Calibur). The excitation/emission wavelengths for DCF-DA, DHR-123 and DAF-DA were 504/530, 505/529 and 495/515 nm, respectively. At the same time, fluorescent images were observed by fluorescent microscopy.

2.8. Electro-mobility gel-shift assay (EMSA) for nuclear transcription factors

HaCaT keratinocytes were cultured in 35 mm cell culture dishes until confluence. After treatment with MP for 24 h, cells were exposed to UVB and were incubated for a further 2 h. Subsequently, medium was removed and cells were washed twice with PBS. Nuclear protein from HaCaT keratinocytes was extracted as described previously (Cepinskas et al., 2003). For EMSA, 3 g of total nuclear proteins was incubated with 1.0 pmol of double-stranded [ P] ATP end-labeled oligonucleotides containing consensus binding sequences for NFB (sense strand 5-AGGGACTTTCCGCTGGGGACTTTCC-3) or for activator protein-1 (AP-1; sense strand 5CGCTTGATGAGTCAGCCGGAA-3) in binding buffer (10 mM Hepes, pH 7.9, 80 mM NaCl, 3 mM MgCl2, 0.1 mM EDTA, 1 mM DTT, 1 mM PMSF and 10% glycerol) for 30 min. After electrophoresis under non-denaturing conditions (0.5 TBE buffer), gels were dried and radioactive bands were visualized on X-ray films.
32

2.9. Lipid peroxidation in HaCaT keratinocytes

HaCaT keratinocytes were cultured in 96-well plates until confluence. After treatment with MP for 24 h, cells were exposed to UVB and were incubated for a further 1 h. Subsequently, medium was removed and cells were washed twice with PBS. Cells were then incubated with diphenyl-1-pyrenylphosphine (DPPP) at a final concentration of 20 M for 15 min. Cell fluorescence was quantified using a fluorescent micro-plate reader. The excitation/emission wavelengths for DPPP were 352/380 nm.

2.10. Electron paramagnetic resonance (EPR) study

EPR was performed using an FR-80 ESR Spectrometer (JEOL, Akishima, Japan) equipped with a UV-irradiation system (RUVF-203S, Radical Research, Tokyo, Japan). The UV spectrum consists of UVB and UVA, and total UV power was around 100 mW/cm . Methylparaben was dissolved in PBS at 0.15% (w/v). Dimethylpyrroline-Noxide (Dojindo, Kumamoto, Japan) at 1 M was used as a spin trapping agent. Samples were introduced into a flat quart EPR cuvette (Radical Research, Tokyo, Japan) and were placed in the EPR cavity. UV was introduced to the cavity via optical fiber. EPR conditions were as follows: microwave frequency, 9.4 GHz; microwave power, 5 mW; modulation width, 0.1 mT; sweep rate, 10 mT/min; response time, 0.03 s.
2

2.11. Statistical analysis

Results are presented as means S.E.M. Data were compared by two-way analysis of variance. Differences were considered significant if the P-value was less than 0.05 based on Fisher's protected least significant difference tests. Statistical analysis was performed using Stat View 5.0-J (Abacus Concepts, Berkeley, CA).

3. Results
3.1. Cytotoxic effects of MP on HaCaT keratinocytes
MP at 0.003% had no effect on cellular viability of HaCaT keratinocytes within 24 h, whereas higher concentrations of MP caused significant reductions in cellular viability within 6 h, as assessed by MTT-based WST-1 assay (Fig. 2). To evaluate the type of cell death in HaCaT keratinocytes, we employed a double-staining method using Hoechst 33342 (HO342) and propidium iodide (PI). The obtained images indicated that small basal amounts of apoptosis were present and that lower concentrations of MP had no effect on cell death, whereas higher concentrations of MP caused significant increases in necrotic cells but not apoptotic cells within 6 h (Fig. 3).

Full-size image (13K)

Fig. 2. Effect of MP on cell viability of HaCaT keratinocytes. Cellular viability was measured by MTT-based WST1 assay. Values represent mean (% of control) S.E.M. of three experiments. P < 0.05 compared with MPuntreated control group.
#

Full-size image (150K)

Fig. 3. MP-induced cell death in HaCaT keratinocytes. HaCaT keratinocytes were stained with HO342 (blue) and PI (red). Blue stained cells have intact function of cell membrane. Red stained cells are necrotic cells. Images are representative of three independent experiments.

3.2. MP-induced production of ROS and NO

It is known that excessive amounts of ROS and NO production are a major cause of cell death (Naito, 2002). Therefore, we assessed the MP-induced production of ROS and NO using two oxidant-sensitive fluorescent probes (dihydrorhodamine-123; DHR123 and 5-(and 6)-carboxy-2,7-dichlorodihydrofluorescein diacetate; DCFDA) and one NO-sensitive fluorescent probe (4-amino-5-methylamino-2,7-difluorescein diacetate; DAF-DA). In general, DHR123 and DCF-DA exhibit no fluorescence without ROS, and become fluorescent (as an oxidized form; rhodamine 123 and DCF, respectively) when interacting with ROS. DAF-DA exhibits no fluorescence without NO and becomes fluorescent as a nitrated form, DAF, when interacting with NO. In the present study, fluorescent microscopy revealed that incubation of HaCaT keratinocytes in DMEM containing 0.003% or 0.03% of MP resulted in a little oxidation of either DHR-123 or DCF-DA, as well as a little nitration of DAF-DA within 24 h compared to control (data not shown). These findings were further confirmed using a fluorescent micro-plate reader (Fig. 4). Note that fluorescent intensity did not change, irrespective of whether MP was added to HaCaT

keratinocytes. HaCaT keratinocytes were also stimulated with hydrogen peroxide (H 2O2: 100 M) for 20 min as a positive control, and HaCaT produced significant amount of ROS and NO, respectively (data not shown).

Full-size image (35K)

Fig. 4. MP-induced oxidative stress and NO production in HaCaT keratinocytes. ROS production was measured based on fluorescent intensity of DHR123 and DCF-DA oxidation. NO production was measured based on fluorescent intensity of DAF-DA nitration. Values represent mean (% of control) S.E.M. of three experiments. P < 0.05 compared with MP-untreated control group.
*

3.3. Cytotoxic effects of UVB irradiation on HaCaT keratinocytes

We then focused on the effects of UVB irradiation on HaCaT keratinocytes. Cells were exposed to UVB (15 or 30 mJ/cm ) and the type of cell death was evaluated by double staining with HO342 and PI fluorescent dye. As shown in Fig. 5, UVB irradiation at 15 mJ/cm did not induce cell death, whereas UVB irradiation at 30 mJ/cm induced a small amount of apoptosis (mainly late apoptosis). This was consistent with the results obtained by flow cytometry. As shown in Fig. 6, UVB irradiation at 30 mJ/cm , but not UVB irradiation at 15 mJ/cm , induced a small amount of late apoptosis and necrosis (PI positive cells).
2 2 2 2 2

Full-size image (155K)

Fig. 5. UVB-induced apoptosis in MP-treated HaCaT keratinocytes. HaCaT keratinocytes were stained with HO342 (blue) and PI (red). Blue stained cells have intact function of cell membrane. Cells with condensed nucleus are early apoptotic cells (blue) and late apoptotic cells (red). Images are representative of three independent experiments.

Full-size image (25K)

Fig. 6. Quantification of PI-positive cells in UVB-exposed and MP-treated HaCaT keratinocytes. HaCaT keratinocytes were stained with PI and fluorescent intensity was evaluated by flowcytometer. Each value represents the mean S.E.M. of three experiments. P < 0.05 compared with MP-untreated and UVB-unexposed control group.
*

3.4. MP enhances UVB irradiation-induced cell death

As described above, UVB irradiation itself induced little or no necrosis in HaCaT keratinocytes and practical concentrations of MP (0.003%) alone had no effect on HaCaT cell viability. However, 0.003% MP significantly enhanced UVB-induced cell death of HaCaT keratinocytes as assessed by immunocytochemistry (Fig. 5) and flow cytometry (Fig. 6).

3.5. MP enhances ROS and NO production in HaCaT keratinocytes induced by UVB irradiation
As shown in Table 1, UVB irradiation alone induced little or no increase in ROS and NO production in HaCaT keratinocytes as assessed by flow cytometry. Although MP alone induced no significant or very small amount of ROS and NO production, MP significantly enhanced UVB-induced ROS and NO production in HaCaT keratinocytes. Table 1. UVB-induced ROS and NO production in MP-treated HaCaT keratinocytes

UVB (mJ/cm2) () 15 30

(A) Fluorescent intensity of DHR123 MP () 0.003% 0.03% 100.0 39.3 109.8 29.5 137.7 50.8*

101.6 18.0 123.0 60.7* 100.0 72.1 109.8 44.3 145.9 55.7* 762.3 136.1*

(B) Fluorescent intensity of DCF-DA MP () 0.003% 0.03% 100.0 5.2 114.4 3.1 120.8 9.3* 124.6 6.2 179.8 4.9* 187.7 8.9* 170.4 11.2* 184.2 5.5* 213.8 14.4*

(C) Fluorescent intensity of DAF-DA MP () 0.003% 0.03% 100.0 12.0 100.5 22.4 97.5 20.0 97.7 35.5 105.6 12.4 120.4 10.2* 136.1 21.0*

145.8 30.1* 155.5 10.8*

ROS production was measured based on fluorescent intensity of DHR123 and DCF-DA oxidation using flow cytometry. NO production was measured by fluorescent intensity of DAF-DA nitration using flow cytometry. Each value represents the mean (% of control) S.E.M. of three experiments. P < 0.05 compared with MP-untreated and UVB-unexposed control group.
*

3.6. MP enhances NFB and AP-1 activation of HaCaT keratinocytes induced by UVB irradiation
Redox-sensitive nuclear transcription factors, such as NFB and AP-1, play an important role in cell signal transduction (Handa et al., 2004). As shown in Fig. 7 and Fig. 8, UVB alone induced a little or no activation of NFB and AP-1 in HaCaT keratinocytes as assessed by electro-mobility gel-shift Assay (EMSA). Although MP alone induced no significant NFB and AP-1 activation, MP significantly enhanced UVB-induced NFB and AP-1 activation.

Full-size image (39K)

Fig. 7. UVB-induced NFB activation in MP-treated HaCaT keratinocytes. Nuclear NFB level was assessed by EMSA. The results for untreated (lane 1: control), UVB-exposed (15 mJ/cm ) (lane 2), UVB (30 mJ/cm )-exposed (lane 3), MP (0.003%)-treated (lane 4), MP (0.003%)-treated and UVB (15 mJ/cm )-exposed (lane 5), MP (0.003%)-treated and UVB (30 mJ/cm )-exposed (lane 6), MP (0.03%)-treated (lane 7), MP (0.03%)-treated and UVB (15 mJ/cm )-exposed (lane 8), MP (0.03%)-treated and UVB (30 mJ/cm )-exposed (lane 9), are shown. A representative result (EMSA) of three experiments is shown in the lower panel and quantitative result (densitometry) is shown in the upper panel. Each value represents the mean S.E.M. (% of control). P < 0.01 compared with MP-untreated corresponding group. P < 0.01 compared with MP (0.003%)-treated corresponding group, respectively.
# * 2 2 2 2 2 2

Full-size image (41K)

Fig. 8. UVB-induced AP-1 activation in MP-treated HaCaT keratinocytes. Nuclear AP-1 level was assessed by EMSA. The results for untreated (lane 1), UVB-exposed (15 mJ/cm ) (lane 2), UVB (30 mJ/cm )-exposed (lane 3), MP (0.003%)-treated (lane 4), MP (0.003%)-treated and UVB (15 mJ/cm )-exposed (lane 5), MP (0.003%)treated and UVB (30 mJ/cm )-exposed (lane 6), MP (0.03%)-treated (lane 7), MP (0.03%)-treated and UVB (15 mJ/cm )-exposed (lane 8), MP (0.03%)-treated and UVB (30 mJ/cm )-exposed (lane 9), are shown. A representative result (EMSA) of three experiments is shown in the lower panel and quantitative result (densitometry) is shown in the upper panel. Each value represents the mean S.E.M. (% of control). P < 0.01 compared with MP-untreated corresponding group. P < 0.01 compared with MP (0.003%)-treated corresponding group, respectively.
# * 2 2 2 2 2 2

3.7. MP enhances lipid peroxidation of HaCaT keratinocytes induced by UVB irradiation

As cell death is strongly associated with lipid peroxidation of cell membrane, we assessed lipid peroxidation of HaCaT keratinocytes by fluorescent probe DPPP. UVB alone induced slight lipid peroxidation in HaCaT keratinocytes. Although MP alone also induces slight lipid peroxidation, MP significantly enhanced UVB-induced increase of lipid peroxidation of HaCaT keratinocyte (Fig. 9). In order to determine whether the effect of MP on UVB-induced lipid peroxidation is additive or mutually potentiating effect, the mean fluorescent value of UVBand MP- treated groups subtracted by UVB-treated and MP-untreated group was statistically compared with the mean value of UVB-untreated corresponding group. As a result, in UVB-treated and MP (0.03%)-treated group, the effect of MP on UVB-induced lipid peroxidation was considered to be more than additive one.

Full-size image (16K)

Fig. 9. UVB-induced lipid peroxidation in MP-treated HaCaT keratinocytes. Lipid peroxidation was measured using a fluorescent probe, DPPP. Each value represents the mean (% of control) S.E.M. of three experiments. P < 0.05 compared with MP-untreated and UVB-unexposed control group. P < 0.05 compared with UVB-unexposed corresponding group, respectively.
* #

3.8. EPR study

In the absence of MP, a four-line signal for DMPO-OH was observed, which suggests the generation of hydroxyl radicals (Fig. 10a). This was confirmed by the addition of dimethylsulfoxide (DMSO, 1.4 M), a potent hydroxyl radical inhibitor, which inhibited the four-line signal and gave a six-line signal (Fig. 10b), considered to correspond to the methyl radical-adduct of DMPO. In the presence of MP, DMPO-OH signal intensity was slightly reduced and new signals were observed (Fig. 10c). The new signals were assigned to the DMPO-H and DMPO adducts of carbon-centered radicals. Addition of DMSO also inhibited DMPO-OH in the paraben-containing solution, but had no influence on DMPO-H (Fig. 10d), thus, suggesting that the generation of DMPO-H is not dependent on hydroxyl radicals.

Full-size image (21K)

Fig. 10. EPR spectra of UV-induced radical adducts of DMPO. (a) DMPO + UV; (b) (a) + DMSO; (c) (a) + MP; (d) (c) + DMSO. The inverted signals at either side are for the Mn
2+

in MnO, which was used as an internal marker.

4. Discussion
For many years, MP has been used as a safety preservative in cosmetics, however in the present study, we showed detrimental potential of MP on UVB-exposed human normal skin keratinocytes in vitro. The use of parabens in cosmetic products up to a maximum concentration of 0.8% (w/w), calculated as phydroxybenzoic acid, is permitted by the Danish and EEC regulations ([Howlett, 1992] and [Rastogi et al., 1995])

and typical concentrations of MP in cosmetics are less than 0.32% (Rastogi et al., 1995). In our preliminary study, approximately 1% of the applied MP reaches the basal layer of the skin epidermis of Yucatan mini pigs (data not shown). Together with reported data that small amounts of MP remain unhydrolyzed in the epidermis (Cross and Roberts, 2000), we employed MP at concentrations of 0.30.003% in the present study. We first examined the cytotoxic effects of MP on HaCaT keratinocytes. As has been reported previously (Pomerat and Leake, 1954), we found that practical concentrations of MP (0.003%) have no effect on cellular viability of HaCaT keratinocytes, while higher concentrations of MP decrease cellular viability within 6 h, thus, indicating that the detrimental potential of MP is dose and time dependent. Solar radiation, particularly UVA and UVB, ranging from 290 to 400 nm is accountable for numerous biological effects in the skin. UVB radiation may only reach the epidermis and upper dermis, whereas UVA is a more penetrating radiation that elicits its effects in the dermis. In this study, we focused mainly on the effects of UVB irradiation on skin keratinocytes, the main cell forming epidermis that is the front line against external injurious stress. The UVB doses employed in this study, 15 and 30 mJ/cm , are similar to the average UVB dose delivered by 1 min of sunlight exposure on a fine day on January and that by 30 s of sunlight exposure on a fine day in August, respectively, in Europe (Lebert et al., 2002). UV irradiation of skin has been shown to increase levels of ROS and NO in cells and to result in oxidative damage to lipids, proteins and DNA ([Cunningham et al., 1985],[Vile and Tyrrell, 1995], [Hattori et al., 1996] and [Sander et al., 2004]). In addition, the pro-inflammatory and redox-sensitive transcription factor, NFB (Flohe et al., 1997), has been identified among the primary molecules targeted during signal transduction initiated by UV irradiation of human skin (Fisher et al., 1996). As a result, UV stimulates the expression of wide variety of pro-inflammatory genes and causes inflammatory skin responses, photo-aging and skin cancer depending on the amount of UV exposure. In the present study, low-dose UVB irradiation alone and practical concentrations of MP alone did not induce cell death in HaCaT keratinocytes. Moreover, MP or UVB alone induced little or no ROS and NO production and NFB and AP-1 activation in HaCaT keratinocytes. However, practical concentrations of MP significantly enhanced ROS and NO production, NFB and AP-1 activation, and apoptosis of UVB exposed HaCaT keratinocytes. Because ROS and NO production has been reported to modulate transcription factors (Handa et al., 2004), UVB irradiation of MP-treated HaCaT keratinocytes may activate these transcription factors via ROS and NO production, thus inducing production of various cytokines and chemokines. This, in turn, could result in a vicious circle of inflammation, ultimately resulting in cell death. UVB irradiation can induce apoptotic, necrotic and differentiation pathways in human normal keratinocytes. In the present study, very little degree of apoptosis was induced in UVB-irradiated cells. This finding is consistent with a previous study by Henseleit et al. (1996). In contrast, Mammone et al. reported that low-dose UVB (520 mJ/cm ) increased apoptosis, while high-dose UVB (>20 mJ/cm ) induced necrosis in HaCaT keratinocytes (Mammone et al., 2000). This discrepancy may be the result of differences in the sunlamps employed in each study. The sunlamp used in the present study emitted a broad band of UV, including UVB (ranging from 280 to 315 nm) as well as small amounts of UVA (315400 nm) and the more harmful UVC (200280 nm). However, the FS40 sunlamps (Westinghouse Corp., Pittsburgh, PA) used by Mammone et al. emit a smaller UV range (250360 nm) than those used in our study (275375 nm). Therefore, the sunlamps used by Mammone et al. emitted more UVC than those used in the present study. In addition, the amount of UVA emitted in our study (6.96 and 13.92 mJ/cm for UVB irradiation at 15 and 30 mJ/cm , respectively) was almost a 100 times smaller than the level reported to affect cellular viability. Although the precise mechanisms by which cell death was induced are unknown in this study, lipid peroxidation (Nagano et al., 2005), or ROS and NO production (Naito, 2002) has been proposed as one of the mechanisms of cell apoptosis. Together with our findings that MP pre-treatment significantly enhanced the UVB-induced lipid peroxidation, ROS and NO production, and cell death, it might be possible that the enhanced effect of UVB on cell death in MP pre-treated HaCaT keratinocytes is oxidative stress-dependent event. EPR study indicated that MP might be a source of free radicals when irradiated with UVB. Hydrogen atoms (=hydrogen radicals), carbon-centered radicals and hydroxyl radicals were observed and the former two were MP-dependent. Hydrogen atoms may scavenge other biologically toxic radicals, and thus may act as a protective factor. However, little is known about the biological significance of this radical to date. Hydroxyl radicals are
2 2 2 2 2

highly reactive and react with biological molecules in a diffusion-limited manner. Thus, even if MP acts as a hydroxyl radical scavenger, its biological significance is negligible. Carbon-centered radicals may act as initiators of lipid peroxidation. Although MP has been considered a safe preservative, the present study indicates that MP may have detrimental effects on human skin exposed to sunlight. These effects would be mediated by the mechanisms dependent on ROS, NO, lipid peroxidation and redox-sensitive transcription factors. To our knowledge, this is the first report discussing this issue. However, the physiological situation is more complex because keratinocytes are surrounded by various types of cells, all of which interact. In addition, Qin et al. reported that HaCaT keratinocytes are more susceptible to UV-induced apoptosis when pretreated with interferon plus phorbol esters than human normal keratinocytes via NFB dependent mechanism (Qin et al., 1999), using an UV lamp emitting more toxic UVC, not confluent but more susceptible proliferating cells. Moreover, parabens are strongly suspected to act as endocrine disrupters based on the parabens ability to interact with steroid hormone receptors (Pugazhendhi et al., 2005). To clarify these complex issues, in vitro study using normal human keratinocytes and in vivo experiments investigating the effects of UVB irradiation on MP-treated animal skin are now underway in our laboratory.

References
Alexander et al., 1978 K.S. Alexander, B. Laprade, J.W. Mauger and A.N. Paruta, Thermodynamics of aqueous solutions of parabens, J. Pharm. Sci. 67 (1978), pp. 624627. View Record in Scopus | Cited By in Scopus (11) Boukamp et al., 1988 P. Boukamp, R.T. Petrussevska, D. Breitkreutz, J. Hornung, A. Markham and N.E. Fusenig, Normal keratinization in a spontaneously immortalized aneuploid human keratinocyte cell line, J. Cell. Biol. 106 (1988), pp. 761771. View Record in Scopus | Cited By in Scopus (1579) Cepinskas et al., 2003 G. Cepinskas, J. Savickiene, C.V. Ionescu and P.R. Kvietys, PMN transendothelial migration decreases nuclear NFkappaB in IL-1beta-activated endothelial cells: role of PECAM-1, J. Cell. Biol. 161 (2003), pp. 641651. View Record in Scopus | Cited By in Scopus (22) Cross and Roberts, 2000 S.E. Cross and M.S. Roberts, The effect of occlusion on epidermal penetration of parabens from a commercial allergy test ointment, acetone and ethanol vehicles, J. Invest. Dermatol. 115 (2000), pp. 914918. View Record in Scopus | Cited By in Scopus (23) Cunningham et al., 1985 M.L. Cunningham, N.I. Krinsky, S.M. Giovanazzi and M.J. Peak, Superoxide anion is generated from cellular metabolites by solar radiation and its components, J. Free Radic. Biol. Med. 1 (1985), pp. 381385. Abstract | PDF (424 K) | View Record in Scopus | Cited By in Scopus (40) Darbre et al., 2004 P.D. Darbre, A. Aljarrah, W.R. Miller, N.G. Coldham, M.J. Sauer and G.S. Pope, Concentrations of parabens in human breast tumours, J. Appl. Toxicol.24 (2004), pp. 513. View Record in Scopus | Cited By in Scopus (100) Fisher et al., 1996 G.J. Fisher, S.C. Datta, H.S. Talwar, Z.Q. Wang, J. Varani, S. Kang and J.J. Voorhees, Molecular basis of sun-induced premature skin ageing and retinoid antagonism, Nature 379 (1996), pp. 335 339. View Record in Scopus | Cited By in Scopus (553) Flohe et al., 1997 L. Flohe, R. Brigelius-Flohe, C. Saliou, M.G. Traber and L. Packer, Redox regulation of NFkappa B activation, Free Radic. Biol. Med. 22 (1997), pp. 11151126. Abstract | Record in Scopus | Cited By in Scopus (451) Handa et al., 2004 O. Handa, Y. Naito, T. Takagi, M. Shimozawa, S. Kokura, N. Yoshida, H. Matsui, G. Cepinskas, P.R. Kvietys and T. Yoshikawa, Tumor necrosis factor-alpha-induced cytokine-induced neutrophil chemoattractant-1 (CINC-1) production by rat gastric epithelial cells: role of reactive oxygen species and nuclear factor-kappaB,J. Pharmacol. Exp. Ther. 309 (2004), pp. 670676. View Record in Scopus | Cited By in Scopus (23) Hattori et al., 1996 Y. Hattori, C. Nishigori, T. Tanaka, K. Uchida, O. Nikaido, T. Osawa, H. Hiai, S. Imamura and S. Toyokuni, 8-Hydroxy-2-deoxyguanosine is increased in epidermal cells of hairless mice after chronic ultraviolet B exposure, J. Invest. Dermatol. 107 (1996), pp. 733737. View Record in Scopus | Cited By in Scopus (118) PDF (1110 K) | View

Henseleit et al., 1996 U. Henseleit, T. Rosenbach and G. Kolde, Induction of apoptosis in human HaCaT keratinocytes, Arch. Dermatol. Res. 288 (1996), pp. 676683.View Record in Scopus | Cited By in Scopus (37) Howlett, 1992 J.F. Howlett, The regulation of preservatives in the European Community, Food Addit. Contam. 9 (1992), pp. 607614. View Record in Scopus | Cited By in Scopus (4) Kawachi et al., 1980 T. Kawachi, T. Yahagi, T. Kada, Y. Tazima, M. Ishidate, M. Sasaki and T. Sugiyama, Cooperative programme on short-term assays for carcinogenicity in Japan, IARC Sci. Publ. (1980), pp. 323 330. View Record in Scopus | Cited By in Scopus (21) Lebert et al., 2002 M. Lebert, M. Schuster and D.P. Hader, The European Light Dosimeter Network: 4 years of measurements, J. Photochem. Photobiol. B 66 (2002), pp. 8187. Abstract | Article | Record in Scopus | Cited By in Scopus (21) Mammone et al., 2000 T. Mammone, D. Gan, D. Collins, R.A. Lockshin, K. Marenus and D. Maes, Successful separation of apoptosis and necrosis pathways in HaCaT keratinocyte cells induced by UVB irradiation, Cell. Biol. Toxicol. 16 (2000), pp. 293302. View Record in Scopus | Cited By in Scopus (20) Matthews et al., 1956 C. Matthews, J. Davidson, E. Bauer, J.L. Morrison and A.P. Richardson, p-Hydroxybenzoic acid esters as preservatives. II. Acute and chronic toxicity in dogs, rats, and mice, J. Am. Pharm. Assoc. Am. Pharm. Assoc. (Baltim) 45 (1956), pp. 260267. View Record in Scopus | Cited By in Scopus (15) Mowad, 2000 C.M. Mowad, Allergic contact dermatitis caused by parabens: 2 case reports and a review, Am. J. Contact Dermat. 11 (2000), pp. 5356. Abstract | View Record in Scopus | Cited By in Scopus (22) Nagano et al., 2005 Y. Nagano, H. Matsui, M. Muramatsu, O. Shimokawa, T. Shibahara, A. Yanaka, A. Nakahara, Y. Matsuzaki, N. Tanaka and Y. Nakamura, Rebamipide significantly inhibits indomethacin-induced mitochondrial damage, lipid peroxidation, and apoptosis in gastric epithelial RGM-1 cells, Dig. Dis. Sci. 50 (Suppl. 1) (2005), pp. S76S83. View Record in Scopus | Cited By in Scopus (18) Naito, 2002 Y. Naito, Neutrophil-dependent oxidative stress in gastrointestinal inflammation. In: T. Yoshikawa, Editor, Oxidative Stress and Digestive Disease, Kager, Basel (2002), pp. 2440. Naito et al., 2002 Y. Naito, O. Handa, T. Takagi, T. Ishikawa, E. Imamoto, S. Nakagawa, T. Yamaguchi, N. Yoshida, H. Matsui and T. Yoshikawa, Ubiquitin-proteasome inhibitor enhances tumour necrosis factor-alphainduced apoptosis in rat gastric epithelial cells, Aliment Pharmacol. Ther. 16 (Suppl. 2) (2002), pp. 5966. View Record in Scopus | Cited By in Scopus (7) Oh et al., 2002 S.Y. Oh, M. Fujii, Y. Takeda, K. Yoda, N. Utoguchi, M. Matsumoto and Y. Watanabe, The effect of ethanol on the simultaneous transport and metabolism of methyl p-hydroxybenzoate in excised skin of Yucatan micropig, Int. J. Pharm. 236 (2002), pp. 3542. Abstract | Article | Scopus | Cited By in Scopus (8) Pomerat and Leake, 1954 C.M. Pomerat and C.D. Leake, Short term cultures for drug assays: general considerations, Ann. NY Acad. Sci. 58 (1954), pp. 11101128. Pugazhendhi et al., 2005 D. Pugazhendhi, G.S. Pope and P.D. Darbre, Oestrogenic activity of p-hydroxybenzoic acid (common metabolite of paraben esters) and methylparaben in human breast cancer cell lines, J. Appl. Toxicol. 25 (2005), pp. 301309. View Record in Scopus | Cited By in Scopus (22) Qin et al., 1999 J.Z. Qin, V. Chaturvedi, M.F. Denning, D. Choubey, M.O. Diaz and B.J. Nickoloff, Role of NFkappaB in the apoptotic-resistant phenotype of keratinocytes, J. Biol. Chem. 274 (1999), pp. 3795737964. View Record in Scopus | Cited By in Scopus (82) Rastogi et al., 1995 S.C. Rastogi, A. Schouten, N. de Kruijf and J.W. Weijland, Contents of methyl-, ethyl-, propyl-, butyl- and benzyl-paraben in cosmetic products, Contact Dermatitis 32 (1995), pp. 2830. View Record in Scopus | Cited By in Scopus (66) Rodrigues et al., 1986 C. Rodrigues, E. Lok, E. Nera, F. Iverson, D. Page, K. Karpinski and D.B. Clayson, Shortterm effects of various phenols and acids on the Fischer 344 male rat forestomach epithelium, Toxicology 38 (1986), pp. 103117. View Record in Scopus | Cited By in Scopus (11) Sander et al., 2004 C.S. Sander, H. Chang, F. Hamm, P. Elsner and J.J. Thiele, Role of oxidative stress and the antioxidant network in cutaneous carcinogenesis, Int. J. Dermatol. 43 (2004), pp. 326335. View Record in Scopus | Cited By in Scopus (124) PDF (211 K) | View Record in PDF (466 K) | View

Soni et al., 2002 M.G. Soni, S.L. Taylor, N.A. Greenberg and G.A. Burdock, Evaluation of the health aspects of methyl paraben: a review of the published literature, Food Chem. Toxicol. 40 (2002), pp. 1335 1373. Abstract | Article | PDF (371 K) | View Record in Scopus | Cited By in Scopus (67) Vile and Tyrrell, 1995 G.F. Vile and R.M. Tyrrell, UVA radiation-induced oxidative damage to lipids and proteins in vitro and in human skin fibroblasts is dependent on iron and singlet oxygen, Free Radic. Biol. Med. 18 (1995), pp. 721730. Abstract | Article | PDF (1005 K) | View Record in Scopus | Cited By in Scopus (154)

Corresponding author. Tel.: +81 75 251 5505; fax: +81 75 252 3721.

Toxicology Volume 227, Issues 1-2, 3 October 2006, Pages 62-72