Cytotoxicity of trichloroethylene and perchloroethylene on normal human epidermal keratinocytes and protective role of Vitamin E

Toxicology 209 (2005) 55–67

Cytotoxicity of trichloroethylene and perchloroethylene on normal human epidermal keratinocytes and protective role of Vitamin E Qi-Xing Zhua,b,1 , Tong Shenb , Rui Dingb , Zhao-Zhao Lianga , Xue-Jun Zhanga,∗ a b

Institute of Dermatology, Anhui Medical University, 81 Meishan Road, Hefei, Anhui 230032, PR China Department of Toxicology, School of Public Health, Anhui Medical University, Hefei, Anhui, PR China Received 25 October 2004; received in revised form 25 November 2004; accepted 3 December 2004 Available online 7 January 2005

Abstract Trichloroethylene (TCE) and perchloroethylene (PERC), the most common alkenyl halides, have been extensively used in industry, and can cause skin damage. To evaluate their cytotoxic potential on skin, the effects of these agents on the normal human epidermal keratinocytes (NHEK) were investigated. Their action on cell viability, membrane integrity and lipid peroxidation (LPO) was assessed by neutral red uptake (NRU) assay, lactate dehydrogenase (LDH) release test and measurement of malondialdehyde (MDA) levels and superoxide dismutase (SOD) activity. In addition, the protective effect of antioxidatant vitamin E on the cytotoxicity was also studied. Incubation of NHEK with various concentrations (0.01–31.6 mM) of TCE or PERC caused a dose-dependent decrease in cell viability, with 80% reduction at 31.6 mM. NR50 values from the cytotoxicity assay was found to be 4.53 and 2.16 mM for TCE and PERC, respectively. A time- and concentration- dependent release of LDH were observed at 1, 2, 3, 4 h after cells were exposed to different doses of TCE or PERC. These agents also caused an increase of MDA, whilst an inhibition of SOD activity, in a concentration-dependent manner. Pre-treatment of the cells with vitamin E at 10–200 mM dose-dependently attenuated the cytotoxic effect of TCE or PERC. Pre-treatment with vitamin E also reversed subsequent TCE or PERC-induced release of LDH, elevation of lipid peroxidation and decline of anti-oxidant enzyme activities. These results suggest that TCE and PERC could induce cytotoxicity to NHEK associated with oxidative stress and antioxidatant vitamin E could effectively protect NHEK from TCE- or PERC-induced cytotoxicity, which may be associated to the superoxide scavenging effect and enhancement of anti-oxidant enzyme activities. © 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Trichloroethylene; Perchloroethylene; Cytotoxicity; Normal human epidermal keratinocytes; Vitamin E

Abbreviations: DMEM, Dulbecco’s minimum essential medium; FBS, fetal bovine serum; LDH, lactate dehydrogenase; LPO, lipid peroxidation; MDA, malondialdehyde; NAD, nicotinamide-adenine dinucleotide; NHEK, normal human epidermal keratinocyte; NRU, neutral red uptake; PBS, phosphate buffered saline; PERC, perchloroethylene; SD, standard deviation; SOD, superoxide dismutase; TCE, trichloroethylene ∗ Corresponding author. Tel.: +86 551 5161002; fax: +86 551 5161016. E-mail addresses: [email protected] (Q.-X. Zhu), [email protected] (X.-J. Zhang). 1 Tel.: +86 551 2923002; fax: +86 551 5161002. 0300-483X/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.tox.2004.12.006

56

Q.-X. Zhu et al. / Toxicology 209 (2005) 55–67

1. Introduction Trichloroethylene (TCE) and perchloroethylene (PERC, also known as tetrachloroethylene) are the most common alkenyl halides and their exposure by humans often occurs in workplace and/or ambient environment via inhalation and/or direct dermal contact. TCE is a colorless and volatile liquid, which is used in a variety of industrial processes, including metal degreasing and dry cleaning agent (IARC, 1995; IPCS, 1985). PERC has been used mainly as a chemical intermediate and a solvent in laundering clothes (IARC, 1995; IPCS, 1984). Exposure to TCE or PERC is known to cause a wide array of toxicological effects on animals and human, most notable of which are hepatic, renal- and neuro-toxicities (Briving et al., 1986; Lash and Parker, 2001; Lash et al., 2001). Dermal absorption from exposure to TCE or PERC is negligible, but direct skin contact with TCE or PERC liquid may lead to significant skin lesions. Derma-toxicity caused by TCE or PERC includes irritant reactions, dermatitis and even toxic epidermal necrosis (Phoon et al., 1984). Epidemiological investigations have shown that TCE- and PERC-induced skin lesions through direct contact are becoming more frequent with the increased quantity of these agents used and higher population exposure in developing countries including China (McLaughlin and Blot, 1997; Nakajima et al., 2003). Although there are many reports on TCE and PERCcaused skin lesions, the mode of action and cellular mechanisms underlying their dermatoxicity have received considerably less attention and remain unclear. To gain insight to dermal toxicity, it is important to study the individual contributions made by the cellular elements of the skin. The keratinocyte constitutes the main architecture of the epidermis, the outermost layer of the skin that forms an environmental barrier. This cell type is frequently used for in vitro toxicological investigations of skin lesions, as an alternative to whole-animal studies. Several areas of cutaneous research have employed cultured keratinocytes as experimental models, such as dermatotoxicology and immunotoxicology (Bernstein and Vaughan, 1999). There is increasing evidence that keratinocytes play an active role in the pathogenesis of skin diseases (Freedberg et al., 2001). They are subject to constant oxidative stress induced mainly by a variety of physical and chemical stimuli in the microenvironment, which is a major

promoter of skin damage processes. Lipid peroxidation (LPO) caused by oxidative stress is thought to be an important event contributing to the toxicity of a variety of compounds such as TCE and PERC in lung, liver and kidney. These alkenyl halides are known to produce destruction and dysfunction of cellular membranes through increased LPO and that can be effectively prevented by non-enzymatic anti-oxidants, such as vitamin E. Numerous in vitro and in vivo studies have demonstrated the remarkable potency of vitamin E in preventing oxidative cell injury by TCE and PERC in human lung cancer cells, isolated hepatocytes and nephrocytes (Ebrahim and Sakthisekaran, 1997; Chen et al., 2002). In contrast to renal and hepatic cytotoxicity, there have been no cytotoxicological studies of these agents on human epidermal keratinocytes to date and the equivalent information is not available. Knowledge on dermal toxicity of these agents at a cellular level and search for potentially useful preventive approaches would provide a rationale for developing more effective measures in prevention and treatment of the associated skin lesions. In the present study, we used an in vitro model, normal human epidermal keratinocytes, to test whether TCE or PERC would exert a potent cytotoxic effect on the keratinocytes; if so, whether this could have been mediated by membrane disruption via increased oxidative stress and the incurred cell damage could be attenuated by the anti-oxidant vitamin E.

2. Materials and methods 2.1. Chemicals TCE and PERC with a 99.5% purity (analytical grade or the highest commercial grade available), vitamin E (d-␣-tocopherol) were purchased from Sigma Chemical Co. (St. Louis, MO, USA); and KeratinocyteSFM (Cat. No. 17005), Dulbecco’s minimum essential medium (DMEM) and chemicals for cell culture were all from Gibco (Grand Island, NY); fetal bovine serum (FBS) was obtained from Hyclone Laboratories (Utah). TCE and PERC were dissolved in acetone, and the final concentration of acetone in the medium was no more than 1% (v/v), which showed little effect on cells from our control experiment; the values for cell viability, LDH, MDA and SOD in the presence of the

Q.-X. Zhu et al. / Toxicology 209 (2005) 55–67

vehicle were 98.1 ± 7.6%, 99.2 ± 7.0%, 100.6 ± 5.3% and 101.2 ± 3.0% of blank control, respectively. 2.2. Culture of normal human epidermal keratinocyte (NHEK) and TCE or PERC exposure Human foreskin samples were obtained from patients undergoing surgical treatment, with local ethics authority approval and patient consent. NHEK were produced from a pool consisting of a minimum of three foreskins by the split-skin technique. The freshly obtained tissue samples were washed twice in antibiotics-containing D-Hanks solution and three times in basic D-Hanks solution, and then cut into thin (approximately 1 mm) fragments of about 1 cm2 area with residual subcutaneous tissue removed. The skin sheets were then digested with 0.25% trypsin in two steps: first at 4 ◦ C overnight to obtain epidermis with mechanical separation and then at 37 ◦ C for 10 min to produce dissociated single keratinocytes. The cell suspension was plated onto 60 mm Petridishes coated with mouse collagen and grown in DMEM containing 10% FBS with 100 U penicillin/ml and 100 ␮g streptomycin/ml. The culture was maintained at 37 ◦ C in a humidified atmosphere with 5% CO2 (v/v). The medium was changed to Keratinocyte-SFM the next day. Cells were used for experiment after the culture reached 80% confluence. The concentration and exposure time were determined according to different experimental requirement. In view of TCE and PERC being volatile organic solvents and the problems of evaporation and partition into culture plate during the exposure, all the intervention solutions were made immediately before the experiment and also in large volume to minimize the loss and keep the concentration relatively stable. We have additionally made assessment on two representative concentrations (low and high) for each compound under experimental conditions. Our measurement showed that the concentrations measured at time 0 appeared to be accurate; these measured values were 0.123 ± 0.008 mM (0.125 mM TCE), 1.96 ± 0.041 mM (2.0 mM TCE), 0.051 ± 0.004 mM (0.05 mM PERC) and 0.814 ± 0.006 mM (0.8 mM PERC). A small and gradual decrease in the concentrations were observed over the time and following 4 h of incubation at 37 ◦ C the above concentrations became 81.3 ± 7.8%, 80.2 ± 9.1%, 82.2 ± 8.4% and 79.8 ± 6.0% of control respectively. These are mainly

57

due to evaporation and to a less extent, partition. On the whole solutions in these culture plates retained about 80% of original content. The chemical consumption due to interaction with components in the culture medium was negligible as there was little difference between the solutions made in culture medium and those in distilled water following 4 h incubation. We have not used sealed system, as it would be difficult to maintain the correct O2 and CO2 (thus pH). Considering certain loss of chemicals due to evaporation, we referred all the toxicants’ concentrations as initial concentrations. 2.3. Neutral Red uptake (NRU) assay Cytotoxicity was determined by neutral red uptake, as described by the standard operating procedure (SOP) for the normal human epidermal keratinocyte neutral red uptake cytotoxicity test, a test for basal cytotoxicity (ICCVAM, 2001). NHEK cells were seeded onto 96-well microplates with 5000 cells/well and grown in Keratinocyte-SFM for 3 days. On the 4th day, the medium was then replaced with 250 ␮l fresh Keratinocyte-SFM containing 31.6, 10, 3.16, 1.0, 0.316, 0.1, 0.0316, 0.01 mM of TCE or PERC. This concentration range was sought to generate a NR50 for both agents in NRU test and was determined through our preliminary experiments. To observe the protective effects of vitamin E, cells were pre-incubated with 200, 150, 100, 50, 10 mM vitamin E for 2 h at 37 ◦ C. Eight replicate wells were used per concentration. Culture medium, acetone and sodium lauryl sulfate (SLS) served as blank, vehicle and positive control, respectively. The assay was performed according to the procedures described by the SOP. Absorbance was measured at 540 nm with an automated microplate reader (Elx 800, Bio-Tek Instruments Inc., USA). Results were expressed as percent of vehicle control. A midpoint toxicity value, NR50 (the concentration producing 50% reduction of NR uptake), was calculated from the doseresponse curves and expressed in mM (Babich and Borenfreund, 1990). Concentrations referred in the results were initial concentrations (i.e. concentrations at time 0). 2.4. Lactate dehydrogenase (LDH) release test TCE or PERC-induced cytotoxicity leading to plasma membrane damage was also measured using the

58

Q.-X. Zhu et al. / Toxicology 209 (2005) 55–67

lactate dehydrogenase (LDH) test. This assay, which strongly correlates with the number of lyzed cells, was based on the reduction of nicotinamide-adenine dinucleotide (NAD) to NADH by LDH (Bonnekoh et al., 1990). Individual wells of 96-well culture plate were seeded with 1.0 × 104 NHEK. When cells reached 80% confluence, the medium was replaced with Keratinocyte-SFM containing toxicants. Their concentrations were chosen based on NUR results; a geometric series of 5 concentrations differing by a factor of two were chosen with medium dose about 1/10 of the NR50 for each agent, which include: 0.125, 0.25, 0.5, 1.0, 2.0 mM for TCE and 0.05, 0.1, 0.2, 0.4, 0.8 mM for PERC. The highest dose for each agent in LDH experiments was used to test the protective effect of vitamin E. In these protection experiments, cells were pre-incubated with 200, 150, 100, 50, 10 mM vitamin E for 2 h, and then treated with 2.0 mM TCE or 0.8 mM PERC. After 1, 2, 3 and 4 h exposure, LDH activity in the supernatant and in the cell homogenate were determined by measuring absorbance at 490 nm. Cytotoxicity was expressed as the percent of LDH release corresponding to the ratio between the LDH activity in the supernatant and the total LDH activity.

sodium dodecyl sulfate, 20% acetic acid (pH = 3.5) and 0.8% TBA. The mixture was then placed in a boiling water bath for 40 min. After cooling, a n-butanol and pyridine mixture (15:1, v/v) was added and the final mixture was centrifuged at 1000 × g for 10 min. The supernatant was collected and used for measuring absorbance at 532 nm. 1,1,3,3-Tetramethoxypropane was used as an external standard. The results were expressed as nanomoles of MDA formed per mg of protein. In SOD activity assay, the supernatant was used for total SOD activity quantification, analyzed in a reaction mixture containing 985 ml of 100 mM phosphate buffer (pH 7.4), 0.3 mM K2 H2 –EDTA, 0.5 mM NBT, and 0.1 mM xanthine. The mixture was pre-incubated for 3 min at 25 ◦ C and 10 ml of 0.02 U/ml xanthine oxidase was added to generate superoxide and to induce NBT reduction. Absorbance at 550 nm was recorded. SOD activity was calibrated from a standard curve of percentage inhibition of NBT reduction with standard SOD activity. Data were expressed as SOD units/mg protein. The amount of cellular protein was determined according to the method of Lowry et al. (Lowry et al., 1951). Bovine serum albumin served as standard. 2.6. Statistics analysis

2.5. Lipid peroxidation (LPO) and superoxide dismutase (SOD) activity measurement The content of MDA, served as an indicator of LPO, was determined using the thiobarbituric acid (TBA) method and performed according to the procedures described by Heath and Packer (Heath and Packer, 1968) with slight modifications. The anti-oxidative enzyme-SOD activity was quantified using the xanthine oxidase–nitroblue tetrazolium (NBT) reduction method (Nishigori et al., 1989). NHEK were seeded into individual wells of a 24well culture plate at density of 1 × 105 . When cells reached 80% confluence, they were treated with the same concentrations of TCE or PERC as in LDH test. In protection group, the vitamin E pre-treatment was carried out in the same way as in LDH experiments. After 4 h of TCE or PERC application, cells were collected and undergone the assay procedures for absorbance measurement. Four hours of exposure represent a maximum continuous exposure in a working section at a workplace. For measurement of MDA content, the supernatant was transferred into a tube containing 8%

Experiments were performed at least three times. Throughout the text, data are expressed as mean ± S.D. All statistical analysis was performed in SAS 6.12 software package. Regression analysis of does-response curves was used to determined NR50 values and 95% confidence intervals (CI). The comparison of each intervention group with the appropriate control was analyzed with one-way analysis of variance (ANOVA) followed by Student–Newman–Keuls test. P < 0.05 was considered statistically significant.

3. Results 3.1. Effect of TCE and PERC on cell viability measured as NR50 The primary experimental objective was to quantitatively assess the toxic effect of TCE and PERC on cell viability using NRU assay. SLS, an anionic detergent and the most frequently used substance in experimental irritant contact dermatitis, served as

Q.-X. Zhu et al. / Toxicology 209 (2005) 55–67

59

Fig. 1. Comparative cytotoxicity of different concentrations of TCE and PERC on NHEK as assessed using NRU assay. SLS served as positive control. The data are presented as arithmetic mean percent of the control ± S.D. Abscissa were toxicants’ concentrations at time 0.

Table 1 Time- and concentration-dependence of LDH release from NHEK cells induced by TCE Concentrations (mM)

LDH release (relative ratio) 1h

2h

3h

4h

0 0.125 0.25 0.5 1.0 2.0

15.00 ± 0.89 15.00 ± 0.67 14.99 ± 0.66 15.21 ± 0.80 15.61 ± 0.74 16.02 ± 0.98

15.45 ± 0.82 16.05 ± 1.41 16.88 ± 1.01 17.57 ± 1.10 19.35 ± 1.39* 20.56 ± 1.40*,***

15.95 ± 0.89 18.27 ± 1.10 19.16 ± 1.74 21.74 ± 1.06*,*** 23.53 ± 1.16**,**** 25.98 ± 1.10*,****

15.99 ± 1.03 20.61 ± 1.99 22.05 ± 2.62*** 23.76 ± 2.28*,*** 26.37 ± 0.93**,**** 29.64 ± 1.16**,****

The data are expressed as mean relative release ± S.D. Cells were exposed to 0.125, 0.25, 0.5, 1.0, 2.0 mM TCE for 1, 2, 3 and 4 h; a statistically significant increase was observed when TCE concentration was raised to 0.5 mM at 3 h or the duration of the exposure reached 2 h at 1.0 mM. TCE concentrations were dose at time 0. ∗ P < 0.05. ∗∗ P < 0.01 compared to control. ∗∗∗ P < 0.05. ∗∗∗∗P < 0.01 compared to 0 h (14.46 ± 0.77).

60

Q.-X. Zhu et al. / Toxicology 209 (2005) 55–67

positive control. Application of TCE or PERC produced a dose-dependent reduction in the cell viability in NHEK cells over the dose range examined (0.01–31.6 mM). The dose-response curves of cell viability in NHEK treated with TCE or PERC are shown in Fig. 1. The NR50 values of TCE, PERC and SLS were 4.53 (3.92–5.13), 2.16 (1.27–3.07) and 0.28 (0.20–0.36) mM, respectively. The positive control SLS produced the most potent effect, in accordance with the NR50 values described in SOP, and a lower potency was observed for TCE and PERC. These data thus provide experimental evidence that TCE and PERC are capable of producing cytotoxic effect to human keratinocytes. 3.2. Cytotoxicity of TCE and PERC by LDH release The cellular lesion associated with compromised plasma membrane integrity was further investigated with LDH release assay. As shown in Table 1, following exposure to 0.125, 0.25, 0.5, 1.0, 2.0 mM TCE for 1, 2, 3 and 4 h, there was a time- and concentration-dependent increase in LDH release from cultured NHEK cells. The minimum effective concentration for LDH elevation was 0.5 mM following 3 h’ exposure (P < 0.05) and the shortest exposure required to produce a change was 2 h at a TCE concentration 1.0 mM (P < 0.05). Qualitatively similar results were obtained for PERC. Table 2 shows changes to LDH release by exposure to 0.05, 0.1, 0.2, 0.4, 0.8 mM PERC for 1, 2,

3 and 4 h. Again, a time- and concentration-dependent effect was demonstrated. PERC was slightly more potent than TCE: the minimum effective concentration was 0.2 mM with 3 h incubation (P < 0.05) and the shortest exposure was 2 h at a concentration of 0.4 mM (P < 0.05). 3.3. Effect TCE and PERC on MDA generation and SOD activity To explore the possible mechanisms responsible for plasma membrane damage, experiments were performed to measure the level of lipid peroxidation and SOD activity under conditions similar to those which produced the cytotoxic effects as described above. Application of TCE and PERC indeed altered the level of MDA–a by-product of lipid peroxidation and the activity of SOD in NHEK cells. Figs. 2 and 3 demonstrate a concentration-dependent increase in MDA levels and the parallel decrease in SOD activity in NHEK following treatment of TCE at 0.125, 0.25, 0.5, 1.0, 2.0 mM for 4 h. A significant elevation of MDA levels was elicited by TCE when its concentration reached 0.5 mM (P < 0.05), whilst a corresponding reduction in SOD activity was observed at a minimum effective concentration of 0.25 mM (P < 0.05). Also shown in Figs. 2 and 3, are alterations to MDA levels and SOD activity by 4 h PERC incubation at 0.05, 0.1, 0.2, 0.4, 0.8 mM. A qualitatively similar but slightly more sensitive changes to MDA and SOD occurred, the threshold concentrations for MDA and

Table 2 Time- and concentration-dependence of LDH release from NHEK cells induced by PERC Concentrations (mM)

0 0.05 0.1 0.2 0.4 0.8

LDH release (relative ratio) 1h

2h

3h

4h

15.03 ± 1.00 14.91 ± 0.92 15.03 ± 0.93 15.30 ± 1.00 15.57 ± 1.01 15.73 ± 0.77

15.52 ± 0.92 15.80 ± 0.95 16.93 ± 1.04 17.98 ± 2.10 19.52 ± 0.91*,*** 19.94 ± 1.10*,***

15.92 ± 0.83 17.89 ± 0.95 19.35 ± 1.09 20.85 ± 1.02*,*** 22.53 ± 1.10*,*** 23.22 ± 1.16**,****

16.20 ± 0.70 20.01 ± 1.61 21.11 ± 2.29 23.57 ± 1.52*,*** 26.75 ± 1.44**,**** 30.30 ± 0.70**,****

The data are expressed as mean relative release ± S.D. Cells were incubated in 0.05, 0.1, 0.2, 0.4,0.8 mM PERC for 1, 2, 3 and 4 h; the minimum effective concentration to exhibit a significant increase was 0.2 mM at 3 h and the shortest duration to produce such effect was 2 h at 0.4 mM PERC. PERC concentrations were dose at time 0. ∗ P < 0.05. ∗∗ P < 0.01 compared to control. ∗∗∗ P < 0.05. ∗∗∗∗P < 0.01 compared to 0 h (14.54 ± 0.80).

Q.-X. Zhu et al. / Toxicology 209 (2005) 55–67

61

Fig. 2. Effects of TCE and PERC on MDA level in cultured NHEK. Cells were treated with 0.125,0.25, 0.5, 1.0, 2.0 mM TCE or 0.05, 0.1, 0.2, 0.4, 0.8 mM PERC for 4 h; the significant increase started from 0.5 mM TCE or 0.2 mM PERC. The concentrations for TCE and PERC were dose at time 0, * P < 0.05 vs. control.

SOD being 0.2 mM (P < 0.05) and 0.1 mM (P < 0.05), respectively. 3.4. Protective effect of vitamin E on cytotoxicity TCE and PERC to NHEK Results in previous sections point to lipid peroxidation as a likely mechanism underlying PERC and TCE-mediated cytotoxic effects and further experiments were therefore, performed to determine whether the anti-oxidant vitamin E would confer protection against such cellular lesions. Fig. 4 shows the effect of vitamin E pre-treatment for 2 h on cell viability. Pre-treatment with vitamin E at concentrations of 50 mM or higher resulted in a marked reversal of TCEor PERC-impaired cell viability (P < 0.05), indicating that vitamin E could offer protection against TCE- or PERC-induced cytotoxicity in NHEK cells. The effect of vitamin E on TCE- and PERC-induced LDH release is shown in Fig. 5. Consistent with the cell viability experiments, vitamin E at 50 mM or higher

markedly reduced LDH release by these alkenyl halides (P < 0.05). Figs. 6 and 7 show similar protective effects of vitamin E on the altered MDA levels and SOD activity. At 10 mM or 50 mM, vitamin E attenuated and at higher concentrations (150 mM) largely suppressed TCE- and PERC-evoked MDA elevation (P < 0.05); this effect was accompanied by a significant restoration of SOD activity, approaching the levels of control group at 150 mM. All these data not only suggest a protective role of vitamin E but also support an involvement of oxidative stress as a mechanism contributing to the cytotoxicity by these compounds.

4. Discussion Toxicity of TCE and PERC exhibits significant tissue-dependent differences. Existing studies have mainly focused on investigating the liver and kidney toxicity using cultured isolated kidney cells and hepatocytes (Lash and Parker, 2001; Lash et al., 2001). To as-

62

Q.-X. Zhu et al. / Toxicology 209 (2005) 55–67

Fig. 3. Effects of TCE and PERC on SOD activity in cultured NHEK. Cells were treated with 0.125, 0.25, 0.5, 1.0, 2.0 mM TCE or 0.05, 0.1, 0.2, 0.4,0.8 mM PERC for 4 h. A significant decrease started from 0.25 mM TCE or 0.1 mM PERC compared to control. TCE and PERC concentrations were concentrations at time 0, * P < 0.05 vs. control.

sess their skin toxicity, a type of in vitro model, freshly isolated normal human epidermal keratinocytes, was used in this study. Normal human epidermal keratinocytes offer many advantages as these cells retain the characteristics of native tissue and can also be maintained under tightly controlled and easily manipulated experimental conditions, independent of other physiological and external influences on skin function. TCE and PERC may have two modes of action on cells. They may disturb membrane structure (Lash and Parker, 2001) and induce oxidative stress (Chen et al., 2002). As cell viability depends directly on the membrane integrity, in this research cell viability was quantitatively assessed by NRU assay and the damage to the cellular membrane was detected by LDH leakage.

Moreover, the NRU assay can provide a sensitive, integrated signal of both cell integrity and growth inhibition, and allow us to set the concentration range in which the maximum and the minimum toxic effects could be observed. It has been generally accepted that the leakage of the cytosolic enzyme LDH correlates well with cellular viability, thus being a useful indicator of plasma membrane damage. By quantitative comparison, TCE and PERC were found to produce similar cytotoxicity to NHEK, but PERC (NR50 = 2.16 mM) was more potent than TCE (NR50 = 4.53 mM). These potency values for TCE and PERC may have toxicological implications as these concentrations are likely to be encountered in the real environment especially via occupational cutaneous exposure. Consistent with the

Q.-X. Zhu et al. / Toxicology 209 (2005) 55–67

63

Fig. 4. Protective effects of various concentrations vitamin E on cytoxicity to NHEK by 2.0 mM TCE and 0.8 mM PERC as assessed with NRU assay. Two dotted lines are drawn in the figure to refer the level of cell viability in TCE or PERC alone. Pre-treatment with vitamin E at 50 mM or above resulted in a marked restoration of decreased cell viability (P < 0.05 compared to the toxicants only), approaching the levels of control group at higher vitamin E concentrations; the minimum effective concentrations were 50 mM and 100 mM for PERC and TCE respectively. * P < 0.05 vs. toxicants only.

NRU experiments, TCE and PERC exhibited similar relative potencies on the LDH release, both in terms of concentration and exposure duration. These cytotoxic effects may be attributable to the structural characteristics of these chemicals, which contain chloride atoms, and a higher potency of PERC than TCE may reflect the fact that more chloride atoms are incorporated in the former. A technical problem facing all the in vitro experiments evaluating the cytotoxicity of volatile organic chemicals (VOCs) is the evaporation and, to a less extend, partitioning, which would present the difficulty of maintaining a stable chemical dosimetry. A sealed system could be a solution to the problem but would compromise O2 supply and pH stability in the culture medium. For this reason, the concentrations we

have used are quoted as concentrations at time 0 and are best referred to as apparent concentrations. The actual concentrations would be slightly lower; thus the cytotoxic potency obtained from this study would be a reserved estimation of their toxicity. Another important issue is the partitioning of chemicals into cell membrane, which would affect intracellular concentrations of the toxicant concerned and ultimately its cytotoxicity. Thus what the cells are exposed to in the medium do not necessarily reflect the cellular dose. A higher octanol/water partition coefficient of PERC (3.4) than TCE (2.4) may also partly underlie a higher apparent cytotoxic potency of the former (Mattie et al., 1994). Destruction of membrane structure and function may result from LPO due to oxidative stress within

64

Q.-X. Zhu et al. / Toxicology 209 (2005) 55–67

Fig. 5. Effects of vitamin E on TCE- and PERC-induced LDH release in cultured NHEK. Cells were exposed to 2.0 mM TCE and 0.8 mM PERC for 4 h. Vitamin E alone had no effect on LDH release. The dotted lines are drawn to show the levels of LDH release by TCE and PERC alone. Pre-treatment with vitamin E 50 mM or above resulted in a marked reduction of LDH release compared with the toxicants alone (P < 0.05); the minimum effective concentrations were 50 mM and 100 mM for TCE and PERC, respectively. * P < 0.05 vs. toxicants alone.

the tissue. TCE and PERC are known to produce membrane damage through increased LPO in some tissue types (Ebrahim and Sakthisekaran, 1997; Channel et al., 1998; Mark et al., 1999). The hypothesis that TCE and PERC exert their effects on keratinocytes via a LPO-mediated mechanism was tested in the present study. Our results demonstrate that MDA levels indeed increased significantly in NHEK following exposure to these compounds (Fig. 2). Quantitative information was obtained from the concentration-dependence of TCE- and PERC-induced MDA formation. LPO occurred over a range of effective concentrations of these agents similar to those on cytotoxicity and with same order of potency–PERC greater than TCE. These results serve as evidence that LPO may be involved in the cytotoxic effects. SOD is an endogenous enzymatic scavengers which can counterbalance the oxidative destruction of free

radicals. It has been shown that SOD may play a primary protective role against ultraviolet-B-induced injury of the human keratinocyte cell line HaCaT (Sasaki et al., 2000). In our study, the TCE- and PERC-induced increase of LPO in NHEK was mirrored by a reduction of SOD activity. This parallel reduction in SOD activity could be due to clearing the free radicals inside the cell and thus indicate a high degree of free radical production and LPO occurrence. The anti-oxidative capacity is apparently damaged in keratinocytes by these toxicants, which would exacerbate cytotoxic effects due to LPO. In addition, a compromised SOD could also be primary event as attenuation of SOD activity occurred at lower concentrations of the alkenyl halides than those producing LPO elevation. Vitamin E not only acts as an effective lipophilic anti-oxidant and radical scavenger, but also stabilizes cellular membranes occurring through loss of fluidity,

Q.-X. Zhu et al. / Toxicology 209 (2005) 55–67

65

Fig. 6. Effects of vitamin E on the increment of MDA levels in NHEK induced by TCE and PERC. Cells were exposed to 2.0 mM TCE and 0.8 mM PERC for 4 h. The levels of MDA produced by TCE and PERC without vitamin E treatment are marked by the dotted lines. Pre-treatment with vitamin E resulted in a marked inhibition of the MDA increase (P < 0.05 vs. the toxicants alone), the minimum effective concentrations being 10 mM and 50 mM for TCE and PRCE, respectively. * P < 0.05 vs. toxicants only.

which can effectively terminate the LPO chain reaction. Supplementation of the skin with anti-oxidants such as vitamin E to enhance the skin’s anti-oxidant capacity proves to be a valid approach, and the protective effect of topically applied anti-oxidants has been subject to intense investigation (Shapiro and Saliou, 2001; Kuriyama et al., 2002; Maalouf et al., 2002). We found 50–100 mM is the lowest effective dose for vitamin E to protect TCE- and PERC-induced cytotoxicity in normal human epidermal keratinocytes by NUR assay. Pre-incubation with 50 mM or higher dose of vitamin E prior to TCE and PERC exposure consistently antagonized the LDH release and MDA elevation, and restored the SOD activity. These results not only demonstrate that vitamin E can sufficiently protect normal human epidermal keratinocytes from subsequent cytotoxicity due to TCE and PERC exposure, but also confirm that lipid peroxidation

plays a role in the skin toxicity exerted by these agents. In summary, this is the first report of a quantitative comparison of cytotoxicity of TCE and PERC using an in vitro model, NHEK in serum-free culture, which has confirmed the potential of TCE and PERC to induce cutaneous cytotoxicity. These results also suggest that peroxidation of cell membrane is importantly involved, if not exclusively, in the biochemical processes leading to cell damage in TCE- and PERC-induced skin lesion. Moreover, the protective effect of vitamin E against the cytotoxic response suggests that LPO is a mechanism of cytotoxicity and not a result of cell damage. This research not only provides useful estimates of relative toxicities of the two toxicants to skin, but also suggests a potentially effective, protective approach to these skin lesions. However, it has to be emphasized that the exact mechanism of action in this in vitro model may

66

Q.-X. Zhu et al. / Toxicology 209 (2005) 55–67

Fig. 7. Effects of vitamin E on the decline in SOD activity in cultured NHEK induced by TCE and PERC. 2.0 mM TCE and 0.8 mM PERC were applied to NHEK cells for 4 h with or without vitamin E pre-treatment. The SOD activities in the presence of TCE and PERC only are indicated by the dotted lines. Pre-treatment with vitamin E produced significant restoration of SOD activity (P < 0.05 vs. the toxicants only), nearly reaching the levels of control group with the increased vitamin E concentrations; the minimum effective concentrations were 50 mM for both agents. * P < 0.05 vs. toxicants only.

be multifactorial, which is not completely understood, and goes beyond the aims of this study. It is important to note that extrapolation of these findings to in vivo conditions remains to be established.

Acknowledgments This project was partly supported by grants from “National Nature Science Foundation of China” (No. 30471469) and “Nature Science Foundation of Anhui Province” (No. 03043801).

References Babich, H., Borenfreund, E., 1990. Applications of the neutral red cytotoxicity assay to in vitro toxicology. Altern. Lab. Anim. 18 (1), 129–144.

Bernstein, I.A., Vaughan, F.L., 1999. Cultured keratinocytes in in vitro dermatotoxicological investigation: a review. J. Toxicol. Environ. Health Part B: Crit. Rev. 2 (1), 1–30. Bonnekoh, B., Farkas, B., Geisel, J., Mahrle, G., 1990. Lactate dehydrogenase release as an indicator of dithranol-induced membrane injury in cultured human keratinocytes. A time profile study. Arch. Dermatol. Res. 282 (3), 325–329. Briving, C., Jacobson, I., Hamberger, A., Kjellstrand, P., Haglid, K.G., Rosengren, L.E., 1986. Chronic effects of perchloroethylene and trichloroethylene on the gerbil brain amino acids and glutathione. Neurotoxicology 7 (1), 101–108. Channel, S.R., Latendresse, J.R., Kidney, J.K., Grabau, J.H., Lane, J.W., Steel-Goodwin, L., Gothaus, M.A., 1998. A subchronic exposure to trichloroethylene causes lipid peroxidation and hepatocellular proliferation in male B6C3F1 mouse liver. Toxicol. Sci. 43 (2), 145–154. Chen, S.J., Wang, J.L., Chen, J.H., Huang, R.N., 2002. Possible involvement of glutathione and p53 in trichloroethylene- and perchloroethylene-induced lipid peroxidation and apoptosis in human lung cancer cells. Free Radic. Biol. Med. 33 (4), 464–472. Ebrahim, A.S., Sakthisekaran, D., 1997. Effect of vitamin E and taurine treatment on lipid peroxidation and antioxidant de-

Q.-X. Zhu et al. / Toxicology 209 (2005) 55–67 fense in perchloroethylene-induced cytotoxicity in mice. J. Nutr. Biochem. 8 (5), 270–274. Freedberg, I.M., Marjana, T.C., Komine, M., Blumenberg, M., 2001. Keratins and the keratinocyte activation cycle. J. Invest. Dermatol. 116, 633–640. Heath, R.L., Packer, L., 1968. Photoperoxidation in isolated chloroplasts: I. Kinetics and stoichiometry of fatty acid peroxidation. Arch. Biochem. Biophys. 125 (2), 189–198. IARC, 1995. Dry cleaning, some chlorinated solvents and other industrial chemicals. In: IARC Monographs on The Evaluation of Carcinogenic to Human, vol. 63, No. (4–17), pp. 33–221. ICCVAM, NTP, NICEATM, 2001. Guidance Document on Using In Vitro Data to Estimate In Vivo Starting Doses for Acute Toxicity. IPCS, 1984. Tetrachloroethylene. Environmental Health Criteria 31. World Health Organization, Geneva. IPCS, 1985. Trichloroethylene. Environmental Health Criteria 50. World Health Organization, Geneva. Kuriyama, K., Shimizu, T., Horiguchi, T., Watabe, M., Abe, Y., 2002. Vitamin E ointment at high dose levels suppresses contact dermatitis in rats by stabilizing keratinocytes. Inflamm. Res. 51 (10), 483–489. Lash, L.H., Parker, J.C., 2001. Hepatic and renal toxicities associated with perchloroethylene. Pharmacol. Rev. 53 (2), 177–208. Lash, L.H., Qian, W., Putt, D.A., Hueni, S.H., Elfarra, A.A., Krause, R.J., Parker, J.C., 2001. Renal and hepatic toxicity of trichloroethylene and its glutathione-derived metabolites in rats and mice: sex-, species-, and tissue-dependent differences. J. Pharmacol. Exp. Therapeut. 297 (1), 155–164. Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J., 1951. Protein measurement with the Folin–Phenol reagent. J. Biol. Chem. 193 (2), 265–275.

67

Maalouf, S., El-Sabban, M., Darwiche, N., Gali-Muhtasib, H., 2002. Protective effect of vitamin E on ultraviolet B light-induced damage in keratinocytes. Mol. Carcinogen. 34 (3), 121–130. Mark, T., John, C., David, D., Patty, M., Steve, S., Cynthia, W., Dwight, W., 1999. Oxidative stress and DNA damage in Fisher rats following acute exposure to trichloroethylene or perchloroethylene. Toxicology 138 (1), 43–53. Mattie, D.R., Bates Jr., G.D., Jepson, G.W., Fisher, J.W., McDougal, J.N., 1994. Determination of skin:air partition coefficients for volatile chemicals: experimental method and applications. Fundam. Appl. Toxicol. 22 (1), 51–57. McLaughlin, J.K., Blot, W.J., 1997. A critical review of epidemiology studies of trichloroethylene and perchloroethylene and risk of renal-cell cancer. Int. Arch. Occup. Environ. Health 70, 222–231. Nakajima, T., Yamanoshita, O., Kamijima, M., Kishi, R., Ichihara, G., 2003. Generalized skin reactions in relation to trichloroethylene exposure: a review from the viewpoint of drug-metabolizing enzymes. J. Occup. Health 45, 8–14. Nishigori, C., Miyashi, Y., Inamura, I., Takebe, H., 1989. Reduced superoxide dismutase activity in xeroderma pigmentosum fibroblasts. J. Invest. Dermatol. 93 (3), 506–510. Phoon, W.H., Chan, M.O., Rajan, V.S., Tan, K.J., Thirumoorthy, T., Goh, C.L., 1984. Stevens-Johnson syndrome associated with occupational exposure to trichloroethylene. Contact Dermatitis 10 (2), 270–276. Sasaki, H., Akamatsu, H., Horio, T., 2000. Protective role of copper, zinc superoxide dismutase against UVB-induced injury of the human keratinocyte cell line HaCaT. J. Invest. Dermatol. 114 (3), 502–507. Shapiro, S.S., Saliou, C., 2001. Role of vitamins in skin care. Nutrition 17, 839–844.