- Email: [email protected]
Enhancement of percutaneous penetration of aniline and o-toluidine in vitro using skin barrier creams
Available online at www.sciencedirect.com
Toxicology in Vitro 22 (2008) 812–818 www.elsevier.com/locate/toxinvit
Brief communication
Enhancement of percutaneous penetration of aniline and o-toluidine in vitro using skin barrier creams Gintautas Korinth *, Lars Lu¨ersen, Karl Heinz Schaller, Ju¨rgen Angerer, Hans Drexler Institute and Out-Patient Clinic of Occupational, Social and Environmental Medicine, University Erlangen-Nuremberg, Schillerstrasse 25/29, D-91054 Erlangen, Germany Received 6 March 2007; accepted 1 November 2007 Available online 12 November 2007
Abstract Aniline (ANI) and the human carcinogen o-toluidine (OT) are released at the workplace during the production and processing of rubber. Recently, we showed in rubber industry workers that a frequent use of skin barrier creams (SBC) increased the internal exposure of ANI and OT. In the present study, diffusion cells were used to investigate the effects of two SBC and one skin care cream (SCC) on percutaneous penetration of neat ANI and OT as well as of OT from a mixture with a workplace specific lubricant. The experiments were carried out with untreated and with skin creams treated human skin. A considerable percutaneous penetration enhancement of test compounds was observed for treated skin compared with untreated skin; the highest enhancement (mean factors 6.2–12.3) was found for SBC (based on oil in water emulsion) treated skin. The lowest penetration enhancement showed SCC treated skin (mean factors 4.2–9.7). The in vitro data support our findings in workers that the percutaneous absorption of aromatic amines significantly increases in presence of skin creams. The efficacy of skin creams to protect the percutaneous penetration of aromatic amines is not confirmed by our own experiments. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: Percutaneous absorption; Diffusion cell; Aromatic amines; Aniline; o-Toluidine; Skin barrier creams; Skin care creams; Penetration enhancement
1. Introduction In the rubber industry, the aromatic amines (AA), aniline (ANI), which is not a workplace chemical, but released during rubber processing, and o-toluidine (OT), released during vulcanisation are chemicals of interest in occupational medical health surveillance examinations due to their carcinogenic effects. ANI shows a carcinogenic potential Abbreviations: AA, aromatic amines; ANI, aniline; ANOVA, analysis of variance; GC, gas chromatography; MS, mass spectroscopy; OT, otoluidine; o/w, oil in water emulsion; PFAA, pentafluoropropionic acid anhydride; SBC, skin barrier cream; SCC, skin care cream; SD, standard deviation; SEM, standard error of the mean; w/o, water in oil emulsion. * Corresponding author. Tel.: +49 9131 8526102; fax: +49 9131 8522317. E-mail address: [email protected] (G. Korinth). 0887-2333/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.tiv.2007.11.006
and OT has been classified as a human carcinogen (DFG, 2006) causing urinary bladder cancer and/or renal pelvis cancer (Bartsch, 1981; ACGIH, 2006). Recently, we could show that personal protective equipment does not prevent workers in the German rubber industry from a relevant systemic incorporation of ANI and OT (Korinth et al., 2007b). Using a multiple linear regression analysis, it was evident that skin barrier creams (SBC) play an important role to enhance the percutaneous uptake of ANI and OT. It is known that ANI and OT penetrate well through human skin (Barry et al., 1985; Lu¨ersen et al., 2006) and can significantly contribute to overall exposure (Lu¨ersen et al., 2006; T. Wellner, unpublished data). The influence of skin creams on percutaneous absorption of chemicals at the workplace is insufficient investigated. One reason is that there are no methods available allowing a quantification of percutaneously absorbed
G. Korinth et al. / Toxicology in Vitro 22 (2008) 812–818
chemicals in workers. Due to ethical reasons, the percutaneous absorption of carcinogens can be tested only in animal experiments or in vitro. The aim of our in vitro study was to investigate the percutaneous penetration behaviour of ANI and OT using excised human skin untreated and treated with SBC or with a skin care cream (SCC). Furthermore, we tested the percutaneous penetration of OT mixed with a mandrel release agent (lubricant), which is used in the rubber industry during the vulcanisation process of rubber tubes.
813
absorption. All creams contained among others, water, antimicrobial agents (e.g., various parabenes), emulsifiers (e.g., glycerol stearate or stearic acid) and some known penetration enhancers, such as glycerine and urea. 2.3. Preparation of skin membranes and experimental design Percutaneous penetration experiments were carried out with the system of Franz static diffusion cell (FDC-400 9FF; Crown Glass, Somerville, NJ, USA) using excised skin of three female donors from abdominal (38 and 52 years, white and 40 years, black ethnic skin) and of one female donor from femoral area (50 years, white ethnic skin). The skin was obtained from a local hospital immediately after cosmetic reduction surgery, removed from subcutaneous fat tissue, wrapped in aluminium foil, put in plastic bags and stored at 20 °C, according to van de Sandt et al. (2004), for a period of about 1.5, 3, 7 and 8 months. For the experiments, skin was thawed at room temperature and the dermis was splitted with a scalpel to obtain a skin thickness of 0.9 mm measured with a precision vernier caliper. Integrity of skin membranes was assessed visually. Diffusion cells were heated by a thermostatic circulatingwater bath (MV-4; Julabo, Seelbach, Germany) at the temperature of 37 °C. Skin surface temperatures were measured with a digital surface thermometer (GMH 3230; Greisinger electronic, Regenstauf, Germany) immediately before applying skin creams. Percutaneous penetration of neat ANI and OT as well as the mixture of OT and mandrel release agent was compared for untreated and with skin creams treated skin. SBC and SCC were applied uniformly with dry cotton swabs as a thin layer on the skin mounted on diffusion cells. After 10 min ANI, OT and OT/lubricant mixture were applied on the skin surface with a volume of 500 ll/cm2 skin (=infinite dose) under occlusion. Receptor fluid (0.9% NaCl solution) was stirred continuously; receptor fluid samples were collected at 1, 2, 3 and 4 h after application of test compounds and stored at 20 °C until analysis. Sampling volume was immediately replaced by adding fresh receptor fluid.
2. Materials and methods 2.1. Test compounds Non-radiolabeled ANI (CAS No. 62-53-3) and OT (CAS No. 95-53-4) of purity greater than 99% were purchased from Sigma–Aldrich and Fluka (Deisenhofen, Germany), respectively. The physicochemical properties of both compounds are presented in Table 1. Furthermore, we mixed OT in a ratio of 75:25 v/v, with a mandrel release agent (OT/lubricant mixture). The mandrel release agent Struktol MR 161 (Schill + Seilacher, Hamburg, Germany) is a mixture of synthetic liquid water soluble lubricants based on silicone-free polymers. It is used in rubber plants to reduce the friction between the uncured rubber tubes and the metallic mandrel and later in the production process between the cured tube and the hot mandrel. 2.2. Skin barrier creams We tested two commercially available SBC and one SCC from two manufacturers. According to the manufacturers instructions for use one SBC (LindesaÒ, Faweco, Darmstadt, Germany) and the SCC (StokolanÒ, Stockhausen, Krefeld, Germany) base on o/w (oil in water) emulsion; the other tested SBC (TaktosanÒ Emulsion, Stockhausen, Krefeld, Germany) was a w/o (water in oil) emulsion. The emulsions are the basis of the creams formulated for skin protection against water-based or waterinsoluble workplace agents. One feature of all creams is, according to the manufacturers, their rapid percutaneous
Table 1 Physicochemical properties of aniline and o-toluidine Test compound
Chemical structure
Molecular weighta (g/mol)
Water solubilitya (g/l)
Log Pa
Melting point (aggregation at RT)a,b
93.13
36
0.90
6 °C (liquid)
107.16
16
1.32
16.3 °C (liquid)
NH2 Aniline
CH3 o-Toluidine
NH2 a b
Values are obtained from SRC database (Syracuse research corporation, http://www.syrres.com). Log P is the octanol/water partition coefficient. RT, room temperature.
814
G. Korinth et al. / Toxicology in Vitro 22 (2008) 812–818
2.4. Analysis
2.6. Data presentation and statistics
Aliquots of ANI and OT receptor fluid samples were diluted with 0.9% NaCl solution to a range from 0.05 to 1 mg/l. Hundred microliter aliquots were mixed with 25 ll of internal standard solutions (c = 2 mg/l) ANI-d5 and OT-d7 (CDN Isotopes, Que., Canada) and with 1 ml of 0.9% NaCl solution (pH 6.4). Samples were extracted with 3 ml hexane and separated by centrifugation at 3200 rpm for 10 min. ANI and OT, enriched in the organic phase, were derivatised with 25 ll of pentafluoropropionic acid anhydride (PFAA) for 1 h at 80 °C in water bath. After cooling to room temperature, the excess of PFAA was removed by adding 1 ml phosphate buffer (pH 8). Phases were separated by centrifugation at 3000 rpm for 10 min. The organic solvent was transferred to GC-vials containing 50 ll toluene and evaporated with a speed vac to 200 ll. Residues were transferred into a micro insert and evaporated to 20 ll. The determination of the AA was carried out with GC/MS, according to a slightly modified method described by Weiss and Angerer (2002). Detection limit for both AA was 0.01 mg/l. The relative analytical recovery (extraction efficiency) of ANI and OT from the receptor fluid was 102% and 98%, respectively.
Percutaneous penetration data are presented as mean and standard error of the mean (SEM) for each test series obtained from two skin membranes and two donors (n = 4 for each test compound and each sampling time). The influence of skin surface temperature on percutaneous penetration parameters was tested by Spearman’s rank correlation coefficient with SPSS 14.0 for WindowsÒ software (SPSS Inc., Chicago, IL, USA). Statistical differences of the penetrated amounts of ANI and OT between untreated and with skin creams treated skin are determined using the analysis of variance (one-way ANOVA) with Excel 2003Ò software (Microsoft Co., Redmond, WA, USA). As level of significance p 6 0.05 was considered.
2.5. Calculation of percutaneous penetration parameters Penetrated amounts were calculated from the concentrations of ANI and OT in the receptor fluid samples corrected for the dilution factor due to the sampling procedure. Percutaneous penetration rate was derived as maximum flux from the slope of the linear part of cumulative absorbed amount per square cm of skin versus time (lg cm 2 h 1). The lag time (h) is the intersection of the linear part of the graphs with the time (x) axis within the same exposure phase as the maximum fluxes. Maximum fluxes and lag times were calculated in each case from penetration data of four test series.
3. Results The range of skin surface temperature was 32.3–35.2 °C (mean ± SD: 34.0 ± 0.7 °C), indicating that the exposure conditions between experiments were comparable. A significant influence of the variation of skin surface temperature on the penetration parameters (penetrated amount, flux and lag time) was not observed (p > 0.05). Both ANI and OT were detected in all receptor fluid samples. The percutaneous penetration kinetics of neat ANI and OT are presented in Fig. 1. The kinetic course of cumulative penetrated amount between both AA is similar for untreated as well as for with SBC and SCC treated skin in all test settings. It is to assume that in experiments with the neat AA as well as the OT/lubricant mixture using untreated skin a pseudo steady-state penetration as described by Jewell et al. (2000) was achieved 2-h postexposure. However, all test series using treated skin show a continuous penetration enhancement until the end of experiments. From kinetic data with skin creams a penetration enhancement of test compounds is evident from the first hour of exposure, which further increases up to the
Fig. 1. Cumulative penetrated amounts of aniline (a) and o-toluidine (b) given as means ± SEM (n = 4 for each test series and each sampling time). Penetrated amounts are compared for untreated skin (without SC) and with two skin barrier creams (SBC) as well as with a skin care cream (SCC) treated skin. w/o is water in oil emulsion, o/w is oil in water emulsion and is the basis of the creams.
G. Korinth et al. / Toxicology in Vitro 22 (2008) 812–818
Fig. 2. Cumulative penetrated amounts of o-toluidine from a lubricant mixture (75:25 v/v) given as means ± SEM (n = 4 for each test series and each sampling time). The percutaneous penetration is compared for untreated skin (without SC) and with two skin barrier creams (SBC) as well as with a skin care cream (SCC) treated skin. w/o is water in oil emulsion, o/w is oil in water emulsion and is the basis of the creams.
end of exposure. The similar percutaneous penetration behaviour was observed also for OT from the OT/lubricant mixture (Fig. 2). Table 2 shows the parameters of percutaneous penetration of neat ANI and OT as well as OT from the OT/lubricant mixture. The maximum flux in experiments with untreated skin determined over an exposure period of 4 h, which represents usually the cumulative exposure of a half work shift, was for the neat ANI 298.5 ± 39.4 lg cm 2 h 1 and for neat OT 66.6 ± 18.9 lg cm 2 h 1. The maximum flux of OT from the OT/lubricant mixture was somewhat higher (85.7 ± 9.9 lg cm 2 h 1) than the flux of neat OT. Since also the concentration of OT in the lubricant mixture was lower (75%), the result indicates certain penetration enhancement of OT by the lubricant vehicle, which is, however, statistically not significant
815
(p = 0.51). Comparing the results between untreated and with SBC or SCC treated skin, significant differences of total penetrated amounts of neat AA and of OT from the OT/lubricant mixture (p 6 0.05) could be observed. Solely for ANI, comparing the penetrated amounts between untreated and with SCC treated skin, the difference was not significant, but a strong statistical trend was recognised (p = 0.07). However, the differences of the penetrated amounts of both neat AA as well as of OT from the OT/ lubricant mixture between different treatments with skin creams were not statistically significant (p > 0.05). We calculated factors of penetration enhancement dividing the maximum fluxes from the same skin donor in the same experimental series for untreated and treated skin. These factors demonstrate the differences of percutaneously absorbed neat ANI and OT as well as of OT from the OT/lubricant mixture between untreated and with SBC or SCC treated skin. Fig. 3 presents the range of percutaneous penetration enhancement by skin creams. The greatest factor of penetration enhancement was observed using SBC (o/w emulsion) in experiments with neat OT ranging from 7.2 to 18.5 (mean 12.3). The mean factors of penetration enhancement due to the application of skin creams averaged from 4.2 to 12.3. This range represents also the intra- and inter-individual variability of percutaneous penetration. 4. Discussion In Germany, the efficacy of SBC is announced by the respective manufacturer or the manufacturers associations of the creams. According to the industrial instructions for use, the efficacy of SBC against chemicals based on the principle of w/o or o/w emulsion forming a physical protective layer on the skin surface (IKW and BVH, 2002). The aim is that SBC should be capable to reduce or to
Table 2 Parameters of percutaneous penetration of neat aniline and o-toluidine as well as of o-toluidine from a lubricant mixture calculated from data using untreated and with skin barrier (SBC) or skin care creams (SCC) treated skin (means ± SEM) Test series
Penetrated amount over 4 h (lg/0.64 cm2)*
Maximum flux (lg cm
2
h 1)
Aniline Untreated skin Skin treated with SBC (w/o) Skin treated with SBC (o/w) Skin treated with SCC (o/w)
438.8 ± 64.0 2137.4 ± 180.1 2474.0 ± 415.84 1992.5 ± 592.4
298.5 ± 39.4 1404.9 ± 180.3 1719.9 ± 301.7 1233.3 ± 341.7
1.66 ± 0.22 1.58 ± 0.18 1.70 ± 0.22 1.52 ± 0.06
o-Toluidine Untreated skin Skin treated with SBC (w/o) Skin treated with SBC (o/w) Skin treated with SCC (o/w)
115.3 ± 40.4 877.0 ± 74.8 1190.5 ± 269.1 779.1 ± 156.2
66.6 ± 18.9 491.4 ± 51.8 694.3 ± 100.9 520.4 ± 86.8
1.40 ± 0.32 1.19 ± 0.18 1.44 ± 0.23 1.71 ± 0.13
o-Toluidine/lubricant mixture (75:25 v/v) Untreated skin 164.1 ± 19.2 Skin treated with SBC (w/o) 1051.0 ± 191.2 Skin treated with SBC (o/w) 1180.0 ± 243.7 Skin treated with SCC (o/w) 806.5 ± 249.5
85.7 ± 9.9 570.9 ± 55.8 667.5 ± 105.4 423.2 ± 114.0
1.01 ± 0.06 1.20 ± 0.25 1.33 ± 0.16 1.16 ± 0.19
Lag time (h)
The differences between untreated and treated skin for each test compound are significant using one-way ANOVA (p 6 0.05) with exception of aniline values between untreated and with SCC treated skin (p = 0.07).
816
G. Korinth et al. / Toxicology in Vitro 22 (2008) 812–818
Fig. 3. Percutaneous penetration enhancement factors for aniline (ANI), o-toluidine (OT) and OT in a lubricant mixture given as ranges and means (closed squares). The factors were calculated by dividing maximum fluxes from the same skin donor in the same experimental series. On this basis, we were able to compare the maximum fluxes of neat ANI, OT and OT in a lubricant mixture between untreated and with skin barrier creams (SBC) as well as with a skin care cream (SCC) treated skin. w/o is water in oil emulsion, o/w is oil in water emulsion and is the basis of the creams.
prevent the percutaneous uptake of chemicals. Since there is a lack of studies investigating the efficacy of SBC, occupational hygienists have often to rely on the information of manufacturers. In fact, the protection capability of w/o and o/w emulsions of the human skin is scientifically not proven. In workers of the rubber industry, we observed that the use of SBC leads to a percutaneous penetration enhancement of ANI and OT resulting in a higher systemic exposure (Korinth et al., 2007b). In the present study, the diffusion cell model was selected to investigate the efficacy of SBC. Neat ANI and OT as well as OT dissolved in a mandrel release agent (lubricant) were tested due to their occurrence at workplaces during the vulcanisation process of rubber tubes. Since the concentration of OT in the mandrel release agent under in-use conditions was unknown, the in our experiments chosen OT/lubricant mixture had a model character. Both tested SBC and the SCC were also applied in workers participated in our field study (Korinth et al., 2007b). The exposure time of 4 h in our experiments corresponds well with real occupational exposure at industrial workplaces, since the skin often can be cleaned first at the work break. The skin surface temperature in the experiments was similar to that of humans on hands (Korinth et al., 2003b), but our study design was rather comparable with worst case scenario due to the exposure to high concentrated AA under occlusion. On the other hand, an
occlusion minimises the influence on the results of different environmental conditions, such as room temperature and air humidity. Our results show that percutaneous penetration of ANI and OT is a relevant path of intake (Table 2). The absorption is comparable with that of glycol ethers (Venier et al., 2004), which are generally considered being percutaneously well absorbed. Comparing maximum fluxes the percutaneous penetration of neat ANI is by a factor of about 4.5 higher than that of neat OT and by a factor of 3.5 higher for OT from the OT/lubricant mixture (Table 2). However, the lag time for neat OT and for OT from the OT/lubricant mixture is shorter than that of ANI (1.4 vs. 1.0 h, respectively, 1.7 h). Such a flux/lag time constellation can indicate a greater skin depot of neat OT compared with neat ANI. For aqueous solutions of ANI and OT, (0.003%) the maximum fluxes of both compounds are similar 0.30 vs. 0.37 lg cm 2 h 1 and the lag time is with 0.8 h equal (T. Wellner, unpublished data). In order to determine the risk of percutaneous absorption of AA it is important to know the influence of skin creams on percutaneous absorption. The results of our in vitro experiments using SBC and SCC were consistent. The treatment of the skin with skin creams lead to a substantial penetration enhancement of ANI and OT through the skin, which is compared to untreated skin in almost all experiments statistically significant (Table 2). We expected that SCC would show a higher percutaneous penetration enhancement of the test compounds than SBC, but surprisingly the SCC showed in all test settings with except of one a lower penetration enhancement. ANI and OT are lipophilic compounds with similar physicochemical properties (Table 1). According to manufacturer information, the best protection against absorption of compounds like ANI and OT through the skin are creams based on o/w emulsions (IKW and BVH, 2002). However, the SBC based on w/o emulsion showed a lower penetration enhancement than SBC based on o/w emulsion. Our in vitro model shows a consistent percutaneous penetration behaviour using three different skin creams recommended for the use to protect the workers. These findings confirm and explain the results of our field study in the rubber industry, which demonstrated considerable percutaneous absorption enhancement of ANI and OT in workers when frequently using SBC (Korinth et al., 2007b). However, in that study, the use of SCC in workers in the rubber industry did not show a penetration enhancement of ANI and OT. At workplaces SCC are usually applied after the end of the work shift if an exposure to chemicals is not longer present. Therefore, the effect of penetration enhancement cannot appear. Our experimental design (infinite dose exposure) shows advantages for the use in comparison studies. However, this study design has only model character, since the chemicals applied in workplaces mostly are low concentrated mixtures of finite dose. Therefore, the enhancement factors
G. Korinth et al. / Toxicology in Vitro 22 (2008) 812–818
determined in our study cannot be directly transferred to different occupational scenarios. The non-uniform percutaneous penetration enhancement in our experiments makes it evident that the dogma of w/o and o/w emulsions is in general not valid. Our results are in good agreement with literature using the diffusion cell model to investigate the efficacy of SBC with excised human skin. Already in the eighties it was demonstrated that various SBC did not reduce the percutaneous penetration of formaldehyde and benzene (Lode´n, 1986). van der Bijl et al. (2002) demonstrated an about 1.5-fold penetration enhancement of the human carcinogen benzo[a]pyrene through excised human skin treated with SBC compared to untreated skin. In our previous study, we showed that two SBC enhanced the percutaneous penetration of solvents (Korinth et al., 2003a). Based on the maximum fluxes from this study the penetration of 1,2,4-trimethylbenzene, ethylene glycol and isopropyl alcohol was enhanced by SBC by factors of 4.1, 2.9 and 1.8, respectively. The in vitro observed percutaneous penetration enhancement due to the use of SBC is confirmed by a few studies on workers exposed to polycyclic aromatic hydrocarbons (Adams et al., 1999) and to ANI and OT (Korinth et al., 2007b). In Germany, manufacturers of skin creams are legally obligated to declare the ingredients. The ingredients list for SCC used in our experiments contained known penetration enhancers like glycerine and urea. The SBC based on w/o emulsion contained also glycerine, and the SBC based on o/w emulsion did not contain well-known penetration enhancers. All skin creams contained the emulsifiers glyceryl stearate or stearic acid. Besides the direct effects of enhancers on percutaneous penetration also the role of emulsifiers on penetration of chemicals must be considered with caution. Although the applied amount of skin creams in our experiments was little, nevertheless, they build certainly a mixture with the test compound in the exposure chamber due to the lower penetration of creams into the skin in vitro compared with the situation in vivo. In contrast to the penetrated amount, there is no significant difference in the lag time between treated and untreated skin (Table 2). This might imply that skin creams does not alter the diffusion of chemicals through the skin, but the partitioning into stratum corneum. It is unknown whether the considerable percutaneous penetration enhancement of hazardous substances by skin creams in diffusion cell experiments can be transferred without any restrictions to human beings. Ethical reasons do not allow the testing of the percutaneous absorption of carcinogenic chemicals in humans. Data of animal studies, however, show large discrepancies to the results gained for human skin (Korinth et al., 2007a). In vitro studies are increasingly accepted to test percutaneous absorption (OECD, 2004; Williams, 2006). Our findings should be carefully considered for the decision to use SBC or SCC for workers exposed to hazardous substances. The skin surface of hands is large (1000 cm2) and the dermal uptake of workplace chemicals is primarily determined by the
817
exposed skin area and exposure duration. The penetration enhancement of OT under use of SBC is up to factor 18 as shown in our experiments (Fig. 3). Workers following manufacturer instructions for use of SBC may be particularly endangered. The manufacturers of skin creams should inform the workers on this hazard. In particular, the manufacturers should verify the influence of their skin creams on percutaneous absorption of certain model chemicals or of occupationally relevant chemical mixtures using scientifically accepted methods and to publish the results. Based on the results of our study it is imperatively necessary to revise the recommended principles of efficacy of SBC, which do not confirm the current scientific knowledge. Future parallel in vitro and in vivo studies with chemicals of similar physicochemical properties, but of lower toxicity, might be useful to support the results of this study. Acknowledgements These experiments are a part of a study sponsored by grants of the institution for statutory accident insurance of the German chemical industry (German: Berufsgenossenschaft der Chemischen Industrie). References ACGIH (American Conference of Governmental Industrial Hygienists), 2006. Threshold limit values for chemical substances and physical agents & biological exposure indices. ACGIH worldwide, Cincinnati. Adams, A., Gu¨ndel, J., Strunk, P., Angerer, J., 1999. Zur Effektivita¨t prima¨rpra¨ventiver Maßnahmen bei beruflicher PAH-Exposition. Arbeitsmedizin Sozialmedizin Umweltmedizin 34, 97–100 (in German). Barry, B.W., Harrison, S.M., Dugard, P.H., 1985. Vapour and liquid diffusion of model penetrants through human skin; correlation with thermodynamic activity. The Journal of Pharmacy and Pharmacology 37, 226–236. Bartsch, H., 1981. Metabolic activation of aromatic amines and azo dyes. IARC Scientific Publications 40, 13–30. DFG (Deutsche Forschungsgemeinschaft), 2006. List of MAK and BAT values 2006, Report no. 42. Wiley–VCH, Weinheim, pp. 18–177. Industrieverband Ko¨rperpflege- und Waschmittel e.V. (IKW) und Bundesverband Handschutz (BVH) e.V., 2002. Gruppenmerkbla¨tter fu¨r den beruflichen Hautschutz, Hautmittel. Bro¨nners Druckerei, Frankfurt/Main (in German). Jewell, C., Heylings, J., Clowes, H.M., Williams, F.M., 2000. Percutaneous absorption and metabolism of dinitrochlorobenzene in vitro. Archives of Toxicology 74, 356–365. Korinth, G., Geh, S., Schaller, K.H., Drexler, H., 2003a. In vitro evaluation of the efficacy of skin barrier creams and protective gloves on percutaneous absorption of industrial solvents. International Archives of Occupational and Environmental Health 76, 382–386. Korinth, G., Go¨en, T., Lakemeyer, M., Broding, H.C., Drexler, H., 2003b. Skin strain and its influence on systemic exposure to a glycol ether in offset printing workers. Contact Dermatitis 49, 248–254. Korinth, G., Go¨en, T., Schaller, K.H., Drexler, H., 2007a. Discrepancies between different rat models for the assessment of percutaneous penetration of hazardous substances. Archives of Toxicology (doi:10.1007/s00204-007-0221-6). Korinth, G., Weiss, T., Penkert, S., Schaller, K.H., Angerer, J., Drexler, H., 2007b. Percutaneous absorption of aromatic amines in rubber industry workers – impact of impaired skin and skin barrier creams. Occupational and Environmental Medicine 64, 366–372.
818
G. Korinth et al. / Toxicology in Vitro 22 (2008) 812–818
Lode´n, M., 1986. The effect of 4 barrier creams on the absorption of water, benzene, and formaldehyde into excised human skin. Contact Dermatitis 14, 292–296. Lu¨ersen, L., Wellner, T., Koch, H.M., Angerer, J., Drexler, H., Korinth, G., 2006. Penetration of b-naphthylamine and o-toluidine through human skin in vitro. Archives of Toxicology 80, 644–646. OECD (Organisation for Economic Co-operation and Development), 2004. Guidance document for the conduct of skin absorption studies. OECD environmental health and safety publications, Testing and Assessment Number 28, Paris. van der Bijl, P., Gareis, A., Lee, H., van Eyk, A.D., Stander, I.A., Cilliers, J., 2002. Effects of two barrier creams on the diffusion of benzo[a]pyrene across human skin. Journal of the South African Dental Association 57, 49–52. van de Sandt, J.J., van Burgsteden, J.A., Cage, S., Carmichael, P.L., Dick, I., Kenyon, S., Korinth, G., Larese, F., Limasset, J.C., Maas, W.J.,
Montomoli, L., Nielsen, J.B., Payan, J.P., Robinson, E., Sartorelli, P., Schaller, K.H., Wilkinson, S.C., Williams, F.M., 2004. In vitro predictions of skin absorption of caffeine, testosterone, and benzoic acid: a multi-centre comparison study. Regulatory Toxicology and Pharmacology 39, 271–281. Venier, M., Adami, G., Larese, F., Maina, G., Renzi, N., 2004. Percutaneous absorption of 5 glycol ethers through human skin in vitro. Toxicology in Vitro 18, 665–671. Weiss, T., Angerer, J., 2002. Simultaneous determination of various aromatic amines and metabolites of aromatic nitro compounds in urine for low level exposure using gas chromatography–mass spectrometry. Journal of Chromatography. B, Analytical Technologies in the Biomedical and Life Sciences 778, 179–192. Williams, F.M., 2006. In vitro studies – how good are they at replacing in vivo studies for measurement of skin absorption? Environmental Toxicology and Pharmacology 21, 199–203.
Toxicology in Vitro 22 (2008) 812–818 www.elsevier.com/locate/toxinvit
Brief communication
Enhancement of percutaneous penetration of aniline and o-toluidine in vitro using skin barrier creams Gintautas Korinth *, Lars Lu¨ersen, Karl Heinz Schaller, Ju¨rgen Angerer, Hans Drexler Institute and Out-Patient Clinic of Occupational, Social and Environmental Medicine, University Erlangen-Nuremberg, Schillerstrasse 25/29, D-91054 Erlangen, Germany Received 6 March 2007; accepted 1 November 2007 Available online 12 November 2007
Abstract Aniline (ANI) and the human carcinogen o-toluidine (OT) are released at the workplace during the production and processing of rubber. Recently, we showed in rubber industry workers that a frequent use of skin barrier creams (SBC) increased the internal exposure of ANI and OT. In the present study, diffusion cells were used to investigate the effects of two SBC and one skin care cream (SCC) on percutaneous penetration of neat ANI and OT as well as of OT from a mixture with a workplace specific lubricant. The experiments were carried out with untreated and with skin creams treated human skin. A considerable percutaneous penetration enhancement of test compounds was observed for treated skin compared with untreated skin; the highest enhancement (mean factors 6.2–12.3) was found for SBC (based on oil in water emulsion) treated skin. The lowest penetration enhancement showed SCC treated skin (mean factors 4.2–9.7). The in vitro data support our findings in workers that the percutaneous absorption of aromatic amines significantly increases in presence of skin creams. The efficacy of skin creams to protect the percutaneous penetration of aromatic amines is not confirmed by our own experiments. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: Percutaneous absorption; Diffusion cell; Aromatic amines; Aniline; o-Toluidine; Skin barrier creams; Skin care creams; Penetration enhancement
1. Introduction In the rubber industry, the aromatic amines (AA), aniline (ANI), which is not a workplace chemical, but released during rubber processing, and o-toluidine (OT), released during vulcanisation are chemicals of interest in occupational medical health surveillance examinations due to their carcinogenic effects. ANI shows a carcinogenic potential Abbreviations: AA, aromatic amines; ANI, aniline; ANOVA, analysis of variance; GC, gas chromatography; MS, mass spectroscopy; OT, otoluidine; o/w, oil in water emulsion; PFAA, pentafluoropropionic acid anhydride; SBC, skin barrier cream; SCC, skin care cream; SD, standard deviation; SEM, standard error of the mean; w/o, water in oil emulsion. * Corresponding author. Tel.: +49 9131 8526102; fax: +49 9131 8522317. E-mail address: [email protected] (G. Korinth). 0887-2333/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.tiv.2007.11.006
and OT has been classified as a human carcinogen (DFG, 2006) causing urinary bladder cancer and/or renal pelvis cancer (Bartsch, 1981; ACGIH, 2006). Recently, we could show that personal protective equipment does not prevent workers in the German rubber industry from a relevant systemic incorporation of ANI and OT (Korinth et al., 2007b). Using a multiple linear regression analysis, it was evident that skin barrier creams (SBC) play an important role to enhance the percutaneous uptake of ANI and OT. It is known that ANI and OT penetrate well through human skin (Barry et al., 1985; Lu¨ersen et al., 2006) and can significantly contribute to overall exposure (Lu¨ersen et al., 2006; T. Wellner, unpublished data). The influence of skin creams on percutaneous absorption of chemicals at the workplace is insufficient investigated. One reason is that there are no methods available allowing a quantification of percutaneously absorbed
G. Korinth et al. / Toxicology in Vitro 22 (2008) 812–818
chemicals in workers. Due to ethical reasons, the percutaneous absorption of carcinogens can be tested only in animal experiments or in vitro. The aim of our in vitro study was to investigate the percutaneous penetration behaviour of ANI and OT using excised human skin untreated and treated with SBC or with a skin care cream (SCC). Furthermore, we tested the percutaneous penetration of OT mixed with a mandrel release agent (lubricant), which is used in the rubber industry during the vulcanisation process of rubber tubes.
813
absorption. All creams contained among others, water, antimicrobial agents (e.g., various parabenes), emulsifiers (e.g., glycerol stearate or stearic acid) and some known penetration enhancers, such as glycerine and urea. 2.3. Preparation of skin membranes and experimental design Percutaneous penetration experiments were carried out with the system of Franz static diffusion cell (FDC-400 9FF; Crown Glass, Somerville, NJ, USA) using excised skin of three female donors from abdominal (38 and 52 years, white and 40 years, black ethnic skin) and of one female donor from femoral area (50 years, white ethnic skin). The skin was obtained from a local hospital immediately after cosmetic reduction surgery, removed from subcutaneous fat tissue, wrapped in aluminium foil, put in plastic bags and stored at 20 °C, according to van de Sandt et al. (2004), for a period of about 1.5, 3, 7 and 8 months. For the experiments, skin was thawed at room temperature and the dermis was splitted with a scalpel to obtain a skin thickness of 0.9 mm measured with a precision vernier caliper. Integrity of skin membranes was assessed visually. Diffusion cells were heated by a thermostatic circulatingwater bath (MV-4; Julabo, Seelbach, Germany) at the temperature of 37 °C. Skin surface temperatures were measured with a digital surface thermometer (GMH 3230; Greisinger electronic, Regenstauf, Germany) immediately before applying skin creams. Percutaneous penetration of neat ANI and OT as well as the mixture of OT and mandrel release agent was compared for untreated and with skin creams treated skin. SBC and SCC were applied uniformly with dry cotton swabs as a thin layer on the skin mounted on diffusion cells. After 10 min ANI, OT and OT/lubricant mixture were applied on the skin surface with a volume of 500 ll/cm2 skin (=infinite dose) under occlusion. Receptor fluid (0.9% NaCl solution) was stirred continuously; receptor fluid samples were collected at 1, 2, 3 and 4 h after application of test compounds and stored at 20 °C until analysis. Sampling volume was immediately replaced by adding fresh receptor fluid.
2. Materials and methods 2.1. Test compounds Non-radiolabeled ANI (CAS No. 62-53-3) and OT (CAS No. 95-53-4) of purity greater than 99% were purchased from Sigma–Aldrich and Fluka (Deisenhofen, Germany), respectively. The physicochemical properties of both compounds are presented in Table 1. Furthermore, we mixed OT in a ratio of 75:25 v/v, with a mandrel release agent (OT/lubricant mixture). The mandrel release agent Struktol MR 161 (Schill + Seilacher, Hamburg, Germany) is a mixture of synthetic liquid water soluble lubricants based on silicone-free polymers. It is used in rubber plants to reduce the friction between the uncured rubber tubes and the metallic mandrel and later in the production process between the cured tube and the hot mandrel. 2.2. Skin barrier creams We tested two commercially available SBC and one SCC from two manufacturers. According to the manufacturers instructions for use one SBC (LindesaÒ, Faweco, Darmstadt, Germany) and the SCC (StokolanÒ, Stockhausen, Krefeld, Germany) base on o/w (oil in water) emulsion; the other tested SBC (TaktosanÒ Emulsion, Stockhausen, Krefeld, Germany) was a w/o (water in oil) emulsion. The emulsions are the basis of the creams formulated for skin protection against water-based or waterinsoluble workplace agents. One feature of all creams is, according to the manufacturers, their rapid percutaneous
Table 1 Physicochemical properties of aniline and o-toluidine Test compound
Chemical structure
Molecular weighta (g/mol)
Water solubilitya (g/l)
Log Pa
Melting point (aggregation at RT)a,b
93.13
36
0.90
6 °C (liquid)
107.16
16
1.32
16.3 °C (liquid)
NH2 Aniline
CH3 o-Toluidine
NH2 a b
Values are obtained from SRC database (Syracuse research corporation, http://www.syrres.com). Log P is the octanol/water partition coefficient. RT, room temperature.
814
G. Korinth et al. / Toxicology in Vitro 22 (2008) 812–818
2.4. Analysis
2.6. Data presentation and statistics
Aliquots of ANI and OT receptor fluid samples were diluted with 0.9% NaCl solution to a range from 0.05 to 1 mg/l. Hundred microliter aliquots were mixed with 25 ll of internal standard solutions (c = 2 mg/l) ANI-d5 and OT-d7 (CDN Isotopes, Que., Canada) and with 1 ml of 0.9% NaCl solution (pH 6.4). Samples were extracted with 3 ml hexane and separated by centrifugation at 3200 rpm for 10 min. ANI and OT, enriched in the organic phase, were derivatised with 25 ll of pentafluoropropionic acid anhydride (PFAA) for 1 h at 80 °C in water bath. After cooling to room temperature, the excess of PFAA was removed by adding 1 ml phosphate buffer (pH 8). Phases were separated by centrifugation at 3000 rpm for 10 min. The organic solvent was transferred to GC-vials containing 50 ll toluene and evaporated with a speed vac to 200 ll. Residues were transferred into a micro insert and evaporated to 20 ll. The determination of the AA was carried out with GC/MS, according to a slightly modified method described by Weiss and Angerer (2002). Detection limit for both AA was 0.01 mg/l. The relative analytical recovery (extraction efficiency) of ANI and OT from the receptor fluid was 102% and 98%, respectively.
Percutaneous penetration data are presented as mean and standard error of the mean (SEM) for each test series obtained from two skin membranes and two donors (n = 4 for each test compound and each sampling time). The influence of skin surface temperature on percutaneous penetration parameters was tested by Spearman’s rank correlation coefficient with SPSS 14.0 for WindowsÒ software (SPSS Inc., Chicago, IL, USA). Statistical differences of the penetrated amounts of ANI and OT between untreated and with skin creams treated skin are determined using the analysis of variance (one-way ANOVA) with Excel 2003Ò software (Microsoft Co., Redmond, WA, USA). As level of significance p 6 0.05 was considered.
2.5. Calculation of percutaneous penetration parameters Penetrated amounts were calculated from the concentrations of ANI and OT in the receptor fluid samples corrected for the dilution factor due to the sampling procedure. Percutaneous penetration rate was derived as maximum flux from the slope of the linear part of cumulative absorbed amount per square cm of skin versus time (lg cm 2 h 1). The lag time (h) is the intersection of the linear part of the graphs with the time (x) axis within the same exposure phase as the maximum fluxes. Maximum fluxes and lag times were calculated in each case from penetration data of four test series.
3. Results The range of skin surface temperature was 32.3–35.2 °C (mean ± SD: 34.0 ± 0.7 °C), indicating that the exposure conditions between experiments were comparable. A significant influence of the variation of skin surface temperature on the penetration parameters (penetrated amount, flux and lag time) was not observed (p > 0.05). Both ANI and OT were detected in all receptor fluid samples. The percutaneous penetration kinetics of neat ANI and OT are presented in Fig. 1. The kinetic course of cumulative penetrated amount between both AA is similar for untreated as well as for with SBC and SCC treated skin in all test settings. It is to assume that in experiments with the neat AA as well as the OT/lubricant mixture using untreated skin a pseudo steady-state penetration as described by Jewell et al. (2000) was achieved 2-h postexposure. However, all test series using treated skin show a continuous penetration enhancement until the end of experiments. From kinetic data with skin creams a penetration enhancement of test compounds is evident from the first hour of exposure, which further increases up to the
Fig. 1. Cumulative penetrated amounts of aniline (a) and o-toluidine (b) given as means ± SEM (n = 4 for each test series and each sampling time). Penetrated amounts are compared for untreated skin (without SC) and with two skin barrier creams (SBC) as well as with a skin care cream (SCC) treated skin. w/o is water in oil emulsion, o/w is oil in water emulsion and is the basis of the creams.
G. Korinth et al. / Toxicology in Vitro 22 (2008) 812–818
Fig. 2. Cumulative penetrated amounts of o-toluidine from a lubricant mixture (75:25 v/v) given as means ± SEM (n = 4 for each test series and each sampling time). The percutaneous penetration is compared for untreated skin (without SC) and with two skin barrier creams (SBC) as well as with a skin care cream (SCC) treated skin. w/o is water in oil emulsion, o/w is oil in water emulsion and is the basis of the creams.
end of exposure. The similar percutaneous penetration behaviour was observed also for OT from the OT/lubricant mixture (Fig. 2). Table 2 shows the parameters of percutaneous penetration of neat ANI and OT as well as OT from the OT/lubricant mixture. The maximum flux in experiments with untreated skin determined over an exposure period of 4 h, which represents usually the cumulative exposure of a half work shift, was for the neat ANI 298.5 ± 39.4 lg cm 2 h 1 and for neat OT 66.6 ± 18.9 lg cm 2 h 1. The maximum flux of OT from the OT/lubricant mixture was somewhat higher (85.7 ± 9.9 lg cm 2 h 1) than the flux of neat OT. Since also the concentration of OT in the lubricant mixture was lower (75%), the result indicates certain penetration enhancement of OT by the lubricant vehicle, which is, however, statistically not significant
815
(p = 0.51). Comparing the results between untreated and with SBC or SCC treated skin, significant differences of total penetrated amounts of neat AA and of OT from the OT/lubricant mixture (p 6 0.05) could be observed. Solely for ANI, comparing the penetrated amounts between untreated and with SCC treated skin, the difference was not significant, but a strong statistical trend was recognised (p = 0.07). However, the differences of the penetrated amounts of both neat AA as well as of OT from the OT/ lubricant mixture between different treatments with skin creams were not statistically significant (p > 0.05). We calculated factors of penetration enhancement dividing the maximum fluxes from the same skin donor in the same experimental series for untreated and treated skin. These factors demonstrate the differences of percutaneously absorbed neat ANI and OT as well as of OT from the OT/lubricant mixture between untreated and with SBC or SCC treated skin. Fig. 3 presents the range of percutaneous penetration enhancement by skin creams. The greatest factor of penetration enhancement was observed using SBC (o/w emulsion) in experiments with neat OT ranging from 7.2 to 18.5 (mean 12.3). The mean factors of penetration enhancement due to the application of skin creams averaged from 4.2 to 12.3. This range represents also the intra- and inter-individual variability of percutaneous penetration. 4. Discussion In Germany, the efficacy of SBC is announced by the respective manufacturer or the manufacturers associations of the creams. According to the industrial instructions for use, the efficacy of SBC against chemicals based on the principle of w/o or o/w emulsion forming a physical protective layer on the skin surface (IKW and BVH, 2002). The aim is that SBC should be capable to reduce or to
Table 2 Parameters of percutaneous penetration of neat aniline and o-toluidine as well as of o-toluidine from a lubricant mixture calculated from data using untreated and with skin barrier (SBC) or skin care creams (SCC) treated skin (means ± SEM) Test series
Penetrated amount over 4 h (lg/0.64 cm2)*
Maximum flux (lg cm
2
h 1)
Aniline Untreated skin Skin treated with SBC (w/o) Skin treated with SBC (o/w) Skin treated with SCC (o/w)
438.8 ± 64.0 2137.4 ± 180.1 2474.0 ± 415.84 1992.5 ± 592.4
298.5 ± 39.4 1404.9 ± 180.3 1719.9 ± 301.7 1233.3 ± 341.7
1.66 ± 0.22 1.58 ± 0.18 1.70 ± 0.22 1.52 ± 0.06
o-Toluidine Untreated skin Skin treated with SBC (w/o) Skin treated with SBC (o/w) Skin treated with SCC (o/w)
115.3 ± 40.4 877.0 ± 74.8 1190.5 ± 269.1 779.1 ± 156.2
66.6 ± 18.9 491.4 ± 51.8 694.3 ± 100.9 520.4 ± 86.8
1.40 ± 0.32 1.19 ± 0.18 1.44 ± 0.23 1.71 ± 0.13
o-Toluidine/lubricant mixture (75:25 v/v) Untreated skin 164.1 ± 19.2 Skin treated with SBC (w/o) 1051.0 ± 191.2 Skin treated with SBC (o/w) 1180.0 ± 243.7 Skin treated with SCC (o/w) 806.5 ± 249.5
85.7 ± 9.9 570.9 ± 55.8 667.5 ± 105.4 423.2 ± 114.0
1.01 ± 0.06 1.20 ± 0.25 1.33 ± 0.16 1.16 ± 0.19
Lag time (h)
The differences between untreated and treated skin for each test compound are significant using one-way ANOVA (p 6 0.05) with exception of aniline values between untreated and with SCC treated skin (p = 0.07).
816
G. Korinth et al. / Toxicology in Vitro 22 (2008) 812–818
Fig. 3. Percutaneous penetration enhancement factors for aniline (ANI), o-toluidine (OT) and OT in a lubricant mixture given as ranges and means (closed squares). The factors were calculated by dividing maximum fluxes from the same skin donor in the same experimental series. On this basis, we were able to compare the maximum fluxes of neat ANI, OT and OT in a lubricant mixture between untreated and with skin barrier creams (SBC) as well as with a skin care cream (SCC) treated skin. w/o is water in oil emulsion, o/w is oil in water emulsion and is the basis of the creams.
prevent the percutaneous uptake of chemicals. Since there is a lack of studies investigating the efficacy of SBC, occupational hygienists have often to rely on the information of manufacturers. In fact, the protection capability of w/o and o/w emulsions of the human skin is scientifically not proven. In workers of the rubber industry, we observed that the use of SBC leads to a percutaneous penetration enhancement of ANI and OT resulting in a higher systemic exposure (Korinth et al., 2007b). In the present study, the diffusion cell model was selected to investigate the efficacy of SBC. Neat ANI and OT as well as OT dissolved in a mandrel release agent (lubricant) were tested due to their occurrence at workplaces during the vulcanisation process of rubber tubes. Since the concentration of OT in the mandrel release agent under in-use conditions was unknown, the in our experiments chosen OT/lubricant mixture had a model character. Both tested SBC and the SCC were also applied in workers participated in our field study (Korinth et al., 2007b). The exposure time of 4 h in our experiments corresponds well with real occupational exposure at industrial workplaces, since the skin often can be cleaned first at the work break. The skin surface temperature in the experiments was similar to that of humans on hands (Korinth et al., 2003b), but our study design was rather comparable with worst case scenario due to the exposure to high concentrated AA under occlusion. On the other hand, an
occlusion minimises the influence on the results of different environmental conditions, such as room temperature and air humidity. Our results show that percutaneous penetration of ANI and OT is a relevant path of intake (Table 2). The absorption is comparable with that of glycol ethers (Venier et al., 2004), which are generally considered being percutaneously well absorbed. Comparing maximum fluxes the percutaneous penetration of neat ANI is by a factor of about 4.5 higher than that of neat OT and by a factor of 3.5 higher for OT from the OT/lubricant mixture (Table 2). However, the lag time for neat OT and for OT from the OT/lubricant mixture is shorter than that of ANI (1.4 vs. 1.0 h, respectively, 1.7 h). Such a flux/lag time constellation can indicate a greater skin depot of neat OT compared with neat ANI. For aqueous solutions of ANI and OT, (0.003%) the maximum fluxes of both compounds are similar 0.30 vs. 0.37 lg cm 2 h 1 and the lag time is with 0.8 h equal (T. Wellner, unpublished data). In order to determine the risk of percutaneous absorption of AA it is important to know the influence of skin creams on percutaneous absorption. The results of our in vitro experiments using SBC and SCC were consistent. The treatment of the skin with skin creams lead to a substantial penetration enhancement of ANI and OT through the skin, which is compared to untreated skin in almost all experiments statistically significant (Table 2). We expected that SCC would show a higher percutaneous penetration enhancement of the test compounds than SBC, but surprisingly the SCC showed in all test settings with except of one a lower penetration enhancement. ANI and OT are lipophilic compounds with similar physicochemical properties (Table 1). According to manufacturer information, the best protection against absorption of compounds like ANI and OT through the skin are creams based on o/w emulsions (IKW and BVH, 2002). However, the SBC based on w/o emulsion showed a lower penetration enhancement than SBC based on o/w emulsion. Our in vitro model shows a consistent percutaneous penetration behaviour using three different skin creams recommended for the use to protect the workers. These findings confirm and explain the results of our field study in the rubber industry, which demonstrated considerable percutaneous absorption enhancement of ANI and OT in workers when frequently using SBC (Korinth et al., 2007b). However, in that study, the use of SCC in workers in the rubber industry did not show a penetration enhancement of ANI and OT. At workplaces SCC are usually applied after the end of the work shift if an exposure to chemicals is not longer present. Therefore, the effect of penetration enhancement cannot appear. Our experimental design (infinite dose exposure) shows advantages for the use in comparison studies. However, this study design has only model character, since the chemicals applied in workplaces mostly are low concentrated mixtures of finite dose. Therefore, the enhancement factors
G. Korinth et al. / Toxicology in Vitro 22 (2008) 812–818
determined in our study cannot be directly transferred to different occupational scenarios. The non-uniform percutaneous penetration enhancement in our experiments makes it evident that the dogma of w/o and o/w emulsions is in general not valid. Our results are in good agreement with literature using the diffusion cell model to investigate the efficacy of SBC with excised human skin. Already in the eighties it was demonstrated that various SBC did not reduce the percutaneous penetration of formaldehyde and benzene (Lode´n, 1986). van der Bijl et al. (2002) demonstrated an about 1.5-fold penetration enhancement of the human carcinogen benzo[a]pyrene through excised human skin treated with SBC compared to untreated skin. In our previous study, we showed that two SBC enhanced the percutaneous penetration of solvents (Korinth et al., 2003a). Based on the maximum fluxes from this study the penetration of 1,2,4-trimethylbenzene, ethylene glycol and isopropyl alcohol was enhanced by SBC by factors of 4.1, 2.9 and 1.8, respectively. The in vitro observed percutaneous penetration enhancement due to the use of SBC is confirmed by a few studies on workers exposed to polycyclic aromatic hydrocarbons (Adams et al., 1999) and to ANI and OT (Korinth et al., 2007b). In Germany, manufacturers of skin creams are legally obligated to declare the ingredients. The ingredients list for SCC used in our experiments contained known penetration enhancers like glycerine and urea. The SBC based on w/o emulsion contained also glycerine, and the SBC based on o/w emulsion did not contain well-known penetration enhancers. All skin creams contained the emulsifiers glyceryl stearate or stearic acid. Besides the direct effects of enhancers on percutaneous penetration also the role of emulsifiers on penetration of chemicals must be considered with caution. Although the applied amount of skin creams in our experiments was little, nevertheless, they build certainly a mixture with the test compound in the exposure chamber due to the lower penetration of creams into the skin in vitro compared with the situation in vivo. In contrast to the penetrated amount, there is no significant difference in the lag time between treated and untreated skin (Table 2). This might imply that skin creams does not alter the diffusion of chemicals through the skin, but the partitioning into stratum corneum. It is unknown whether the considerable percutaneous penetration enhancement of hazardous substances by skin creams in diffusion cell experiments can be transferred without any restrictions to human beings. Ethical reasons do not allow the testing of the percutaneous absorption of carcinogenic chemicals in humans. Data of animal studies, however, show large discrepancies to the results gained for human skin (Korinth et al., 2007a). In vitro studies are increasingly accepted to test percutaneous absorption (OECD, 2004; Williams, 2006). Our findings should be carefully considered for the decision to use SBC or SCC for workers exposed to hazardous substances. The skin surface of hands is large (1000 cm2) and the dermal uptake of workplace chemicals is primarily determined by the
817
exposed skin area and exposure duration. The penetration enhancement of OT under use of SBC is up to factor 18 as shown in our experiments (Fig. 3). Workers following manufacturer instructions for use of SBC may be particularly endangered. The manufacturers of skin creams should inform the workers on this hazard. In particular, the manufacturers should verify the influence of their skin creams on percutaneous absorption of certain model chemicals or of occupationally relevant chemical mixtures using scientifically accepted methods and to publish the results. Based on the results of our study it is imperatively necessary to revise the recommended principles of efficacy of SBC, which do not confirm the current scientific knowledge. Future parallel in vitro and in vivo studies with chemicals of similar physicochemical properties, but of lower toxicity, might be useful to support the results of this study. Acknowledgements These experiments are a part of a study sponsored by grants of the institution for statutory accident insurance of the German chemical industry (German: Berufsgenossenschaft der Chemischen Industrie). References ACGIH (American Conference of Governmental Industrial Hygienists), 2006. Threshold limit values for chemical substances and physical agents & biological exposure indices. ACGIH worldwide, Cincinnati. Adams, A., Gu¨ndel, J., Strunk, P., Angerer, J., 1999. Zur Effektivita¨t prima¨rpra¨ventiver Maßnahmen bei beruflicher PAH-Exposition. Arbeitsmedizin Sozialmedizin Umweltmedizin 34, 97–100 (in German). Barry, B.W., Harrison, S.M., Dugard, P.H., 1985. Vapour and liquid diffusion of model penetrants through human skin; correlation with thermodynamic activity. The Journal of Pharmacy and Pharmacology 37, 226–236. Bartsch, H., 1981. Metabolic activation of aromatic amines and azo dyes. IARC Scientific Publications 40, 13–30. DFG (Deutsche Forschungsgemeinschaft), 2006. List of MAK and BAT values 2006, Report no. 42. Wiley–VCH, Weinheim, pp. 18–177. Industrieverband Ko¨rperpflege- und Waschmittel e.V. (IKW) und Bundesverband Handschutz (BVH) e.V., 2002. Gruppenmerkbla¨tter fu¨r den beruflichen Hautschutz, Hautmittel. Bro¨nners Druckerei, Frankfurt/Main (in German). Jewell, C., Heylings, J., Clowes, H.M., Williams, F.M., 2000. Percutaneous absorption and metabolism of dinitrochlorobenzene in vitro. Archives of Toxicology 74, 356–365. Korinth, G., Geh, S., Schaller, K.H., Drexler, H., 2003a. In vitro evaluation of the efficacy of skin barrier creams and protective gloves on percutaneous absorption of industrial solvents. International Archives of Occupational and Environmental Health 76, 382–386. Korinth, G., Go¨en, T., Lakemeyer, M., Broding, H.C., Drexler, H., 2003b. Skin strain and its influence on systemic exposure to a glycol ether in offset printing workers. Contact Dermatitis 49, 248–254. Korinth, G., Go¨en, T., Schaller, K.H., Drexler, H., 2007a. Discrepancies between different rat models for the assessment of percutaneous penetration of hazardous substances. Archives of Toxicology (doi:10.1007/s00204-007-0221-6). Korinth, G., Weiss, T., Penkert, S., Schaller, K.H., Angerer, J., Drexler, H., 2007b. Percutaneous absorption of aromatic amines in rubber industry workers – impact of impaired skin and skin barrier creams. Occupational and Environmental Medicine 64, 366–372.
818
G. Korinth et al. / Toxicology in Vitro 22 (2008) 812–818
Lode´n, M., 1986. The effect of 4 barrier creams on the absorption of water, benzene, and formaldehyde into excised human skin. Contact Dermatitis 14, 292–296. Lu¨ersen, L., Wellner, T., Koch, H.M., Angerer, J., Drexler, H., Korinth, G., 2006. Penetration of b-naphthylamine and o-toluidine through human skin in vitro. Archives of Toxicology 80, 644–646. OECD (Organisation for Economic Co-operation and Development), 2004. Guidance document for the conduct of skin absorption studies. OECD environmental health and safety publications, Testing and Assessment Number 28, Paris. van der Bijl, P., Gareis, A., Lee, H., van Eyk, A.D., Stander, I.A., Cilliers, J., 2002. Effects of two barrier creams on the diffusion of benzo[a]pyrene across human skin. Journal of the South African Dental Association 57, 49–52. van de Sandt, J.J., van Burgsteden, J.A., Cage, S., Carmichael, P.L., Dick, I., Kenyon, S., Korinth, G., Larese, F., Limasset, J.C., Maas, W.J.,
Montomoli, L., Nielsen, J.B., Payan, J.P., Robinson, E., Sartorelli, P., Schaller, K.H., Wilkinson, S.C., Williams, F.M., 2004. In vitro predictions of skin absorption of caffeine, testosterone, and benzoic acid: a multi-centre comparison study. Regulatory Toxicology and Pharmacology 39, 271–281. Venier, M., Adami, G., Larese, F., Maina, G., Renzi, N., 2004. Percutaneous absorption of 5 glycol ethers through human skin in vitro. Toxicology in Vitro 18, 665–671. Weiss, T., Angerer, J., 2002. Simultaneous determination of various aromatic amines and metabolites of aromatic nitro compounds in urine for low level exposure using gas chromatography–mass spectrometry. Journal of Chromatography. B, Analytical Technologies in the Biomedical and Life Sciences 778, 179–192. Williams, F.M., 2006. In vitro studies – how good are they at replacing in vivo studies for measurement of skin absorption? Environmental Toxicology and Pharmacology 21, 199–203.