- Email: [email protected]
The influence of water mixtures on the dermal absorption of glycol ethers
Toxicology and Applied Pharmacology 218 (2007) 128 – 134 www.elsevier.com/locate/ytaap
The influence of water mixtures on the dermal absorption of glycol ethers Matthew J. Traynor, Simon C. Wilkinson, Faith M. Williams ⁎ Toxicology Unit, Institute for Research on Environment and Sustainability and Medical School, University of Newcastle upon Tyne, Newcastle NE1 7RU, UK Received 2 August 2006; revised 28 September 2006; accepted 29 September 2006 Available online 10 November 2006
Abstract Glycol ethers are solvents widely used alone and as mixtures in industrial and household products. Some glycol ethers have been shown to have a range of toxic effects in humans following absorption and metabolism to their aldehyde and acid metabolites. This study assessed the influence of water mixtures on the dermal absorption of butoxyethanol and ethoxyethanol in vitro through human skin. Butoxyethanol penetrated human skin up to sixfold more rapidly from aqueous solution (50%, 450 mg/ml) than from the neat solvent. Similarly penetration of ethoxyethanol was increased threefold in the presence of water (50%, 697 mg/ml). There was a corresponding increase in apparent permeability coefficient as the glycol ether concentration in water decreased. The maximum penetration rate of water also increased in the presence of both glycol ethers. Absorption through a synthetic membrane obeyed Fick's Law and absorption through rat skin showed a similar profile to human skin but with a lesser effect. The mechanisms for this phenomenon involves disruption of the stratum corneum lipid bilayer by desiccation by neat glycol ether micelles, hydration with water mixtures and the physicochemical properties of the glycol ether–water mixtures. Full elucidation of the profile of absorption of glycol ethers from mixtures is required for risk assessment of dermal exposure. This work supports the view that risk assessments for dermal contact scenarios should ideally be based on absorption data obtained for the relevant formulation or mixture and exposure scenario and that absorption derived from permeability coefficients may be inappropriate for water-miscible solvents. © 2006 Elsevier Inc. All rights reserved. Keywords: Glycol ether; Dermal absorption; Water; Mixtures
Introduction Glycol ethers are widely used alone and as mixtures in industrial and household applications because their physicochemical properties, miscibility with both water and organic media make them versatile solvents. Glycol ethers penetrate the skin rapidly (Kezic et al., 1997; Filon et al., 1999) and it has recently been reported that following a 1-h exposure of the human forearm to a 50% butoxyethanol/50% water mixture, the body burden exceeded that which would be obtained from an 8h inhalation exposure at the level of the threshold limit value set by the American Conference of Governmental Industrial Hygienists (ACGIH) (Jakasa et al., 2004). The toxicity of glycol ethers to man is mainly caused by the aldehyde and acid metabolites (Ghanayem et al., 1987) after conversion by alcohol dehydrogenase and aldehyde dehydrogenase. The epidermis has been shown to contain alcohol and aldehyde dehydrogenases ⁎ Corresponding author. E-mail address: [email protected] (F.M. Williams). 0041-008X/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.taap.2006.09.019
(Kao and Carver, 1990; Lockley et al., 2004a), but local metabolism of glycol ethers during dermal penetration was not detected for skin mounted in a flow through diffusion system (Lockley et al., 2002, 2004b), suggesting that the rapid passage of the solvents through the skin limits access to the enzymes. However, using skin in short-term culture with increased sensitivity allowed detection of a small amount of conversion (Traynor et al, in preparation). Butoxyethanol has been demonstrated to cause erythrocyte haemolysis following metabolism to butoxyacetic acid (Bartnik et al., 1987; Ghanayem et al., 1989; Multinger et al., 2005). Ethoxyethanol has been shown to have haematological, developmental effects and reproductive effects (Hardin, 1983; Hardin et al., 1984) in laboratory animals. There is also some evidence for effects in exposed workers. Due to the ease with which the glycol ethers are absorbed through the skin (Johanson et al., 1988; Kezic et al., 1997; Filon et al., 1999) and the potential for the development of adverse health effects, it is important to understand factors that influence absorption. There have been a number of studies of dermal absorption of glycol
M.J. Traynor et al. / Toxicology and Applied Pharmacology 218 (2007) 128–134
ether vapours in humans in vivo (Johanson and Boman, 1991; Kezic et al., 1997) and of undiluted liquids in vitro (Johanson et al., 1988; Kezic et al., 1997). Jakasa et al. (2004) also investigated the absorption of aqueous solutions of glycol ethers through the forearm of volunteers. The ability of in vitro studies to predict in vivo absorption of butoxyethanol and ethoxyethanol neat or from methanol in the rat has been demonstrated (Lockley et al., 2002, 2004b). It has previously been suggested that glycol ethers absorb more effectively through skin when in contact as water mixtures. The aim of this study was to use in vitro methods to investigate the influence of applying ethoxyethanol and butoxyethanol as water and other mixtures on absorption through human skin. It is important to define this for risk assessment of dermal penetration of glycol ethers and to identify conditions in which there may be significant deviation from the expected absorption profile. Materials and methods [1-14C]2-butoxyethanol, specific activity 430 MBq/mmol, was obtained from Amersham. [1-14C]2-ethoxyethanol, specific activity 0.04 MBq/mg, was a gift from Unilever and 3H water, specific activity 1600 MBq/ml, was purchased from ICN radiochemicals (Basingstoke, UK). All radiochemicals had a radiochemical purity exceeding 97%. Minimal essential medium (Eagle) and gentamycin were obtained from Sigma; sodium hydrogen carbonate and teepol L were purchased from BDH; HiSafe 3 scintillation fluid was obtained from Fisher; 2-butoxyethanol (99.9% purity) was obtained from Fluka; and 2ethoxyethanol (99.5% purity) from Aldrich. The water used was sterile water for irrigation. The polydimethylsiloxane membrane was kindly donated by Dr R. Chilcott DSTL (product code 19TO.3-1000-60M1 SAMCO Silicone Products) (Chilcott et al., 2005). Skin preparation. Human breast skin was obtained following mammoplasty from a local hospital and stored at − 70 °C until required. Ethical approval for obtaining skin was given by the University of Newcastle Medical and Dental Ethics Committee and the University Hospital of South Durham Ethics Committee. A section of skin was removed from the freezer and defrosted at room temperature. The skin was then placed dermal side down on a corkboard and dermatomed to a thickness of 320 μm. All experiments were performed using human skin from at least two donors. Rat skin was obtained from male Wistar rats (28 days old), sacrificed by cervical dislocation. The dorsal region was shaved and the skin dissected. The skin was placed dermal side down on a corkboard and dermatomed to a thickness of 280 μm. A polydimethylsiloxane membrane (thickness 400 ± 13 μm) was used in some experiments. The membrane was prepared for use by soaking in sterile water for 24 h prior to use. In vitro flow through diffusion system. Teflon flow through cells of the Newcastle Scott Dick design were used. Receptor fluid (Eagle's minimal essential medium supplemented with 2.2 g/l sodium hydrogen carbonate and 200 μg/ml gentamycin) was maintained at pH 7.4 by gassing with 5% CO2/air and pumped through the cells at 1.5 ml/h. The receptor fluid reservoir and diffusion cells were maintained at approximately 37 °C using a water jacket connected to a circulating water bath. This maintained the surface of the skin at 32 °C. Skin sections were placed in the diffusion cells (exposed surface area 0.64 cm2) and secured in place using threaded nuts. The cells were partially occluded by a tight-fitting cap containing carbon filters which allowed air to pass through the filters. Dose application and determination of diffusion and distribution. The study procedure followed OECD Guidelines Butoxyethanol was used neat (900 mg/ ml) or diluted in aqueous solution at concentrations ranging from 90% v/v to 0.1% v/v (810 mg/ml to 0.9 mg/ml). Ethoxyethanol was used neat (930 mg/ml) or in aqueous solution at concentrations ranging from 90% v/v to 0.1% v/v (837 mg/ ml to 0.93 mg/ml). Test compounds (including 14C butoxyethanol or 14C ethoxyethanol) were applied at 200 μl/cm2 (infinite dose) or 20 μl/cm2 (finite dose). 14C radioactivity amounted to 3.0 kBq per diffusion cell for butoxyethanol
129
and 58 kBq per diffusion cell for ethoxyethanol. In certain studies, absorption of water was studied using 3H2O. Water was applied neat or as a 1:1 (v/v) mixture with butoxyethanol and/or ethoxyethanol for 4 h. Tritium radioactivity amounted to 2.0 MBq per diffusion cell. Receptor fluid fractions were collected every 30 min until 3 h after dose application and then every 60 min until the experiments were terminated at 20 h after dose application. In human skin experiments, the dose was removed from the skin surface after 4 h using alternate wet (soaked in 3% (v/v) teepol L) and dry tissue swabs (6 swabs in total). For studies with rat skin and membrane the dose remained in contact with the skin for the full study and the dose remaining on the skin surface at termination of the study was recovered in the same way. Weighed aliquots (250 μl) of receptor fluid samples were taken for scintillation counting, after mixing with scintillation fluid (3 ml). Tritium-labelled samples were left overnight in darkness before scintillation counting. The tissue swabs used to remove the dose solution were soaked in scintillation fluid (10 ml) and left to desorb for 72 h before scintillation counting. The carbon filters from the traps above each cell were soaked in scintillation fluid (10 ml) and left to desorb for 2 days before counting. At the end of the experiment the skin was removed from the cells and digested using 2 ml of 1.5 mol/l potassium hydroxide in 4:1 (v/v) methanol/water. Once the skin was fully digested (72 h), glacial acetic acid (70 μl) was added to quench chemiluminescence and 10 ml of scintillation fluid was added prior to counting. Scintillation counting was carried out using a Wallac 1410 LSC. HiSafe 3 liquid scintillation cocktail was used throughout. Calculation of absorption parameters. A cumulative absorption–time curve was constructed for amount of test chemical measured in the receptor fluid with time. The maximum absorption rate was calculated from the slope of the linear region of the curve (R2 ≥ 0.95). This was generally between 1 and 3.5 h for short-term 4-h applications and up to 6 h for 24-h doses. For an infinite dose the apparent permeability coefficient (kp) was calculated by dividing the maximum absorption rate by the concentration of the test compound in the dose solution. The lag time was obtained from the intercept of the linear portion of cumulative absorption–time curve with the time axis. Statistical comparisons were made using one-way ANOVA followed by post hoc testing using Bonferroni's correction. The chosen level of significance was P < 0.05.
Results Absorption of butoxyethanol from water mixtures through skin in vitro All of the studies were performed under partially occluded conditions in which the cell was covered with a charcoal trap which allowed air passage through the mesh, and the skin was exposed to either an infinite dose (large volume, concentration not depleted during the study) or a finite dose (small volume depleting during the study). The profile of absorption of butoxyethanol through the human skin comprised a short lag phase followed by increasing absorption, with the linear portion of the profile occurring between 0.5 and 4 h. In studies with human skin, the dose was washed off at 4 h to relate the results to those obtained in vivo by Jakasa et al. (2004) (Fig. 1). A similar absorption profile was obtained with rat skin, although the linear portion of the profile was more prolonged than with human skin as the dose was not washed off until the end of the study (Fig. 2). Rat skin was more permeable to neat butoxyethanol than human skin. A maximum absorption rate for neat butoxyethanol (900 mg/ml) of 0.73 ± 0.01 mg/cm2/h was measured with rat skin, compared to 0.39 ± 0.06 mg/cm2/h with human skin, following application as an infinite dose (Tables 1 and 2). The maximum absorption rate for butoxyethanol with human skin increased significantly when butoxyethanol was diluted
130
M.J. Traynor et al. / Toxicology and Applied Pharmacology 218 (2007) 128–134 Table 1 Influence of butoxyethanol concentration on percutaneous absorption of butoxyethanol (applied as a neat liquid and in aqueous solution) through dermatomed human skin (infinite dose 200 μl/cm2 and finite dose 20 μl/cm2)
Fig. 1. Cumulative absorption profiles of 14C butoxyethanol recovered in receptor fluid from a range of water mixtures applied as infinite dose (200 μl/cm2) to dermatomed human skin in flow through diffusion cells. Results are mean ± SEM (n ≥ 5).
with water. A maximum absorption rate of 2.34 ± 0.28 mg/cm2/h was measured with 675 mg/ml butoxyethanol in aqueous solution, and this increased further to 2.69 ± 0.40 mg/cm2/h with 450 mg/ml butoxyethanol in aqueous solution. This represented an increase of almost sixfold compared to neat butoxyethanol. The maximum absorption rates for butoxyethanol with aqueous solutions at 10% (v/v) (90 mg/ml) and below were not significantly greater, or less than, that measured with neat butoxyethanol. The profile of the relationship between maximum absorption rate and concentration of butoxyethanol indicated three phases: a linear relationship between absorption rate and concentration between 0% and 20% butoxyethanol, between 50% and 70% little change and then a decrease to neat butoxyethanol (Fig. 3a). The aim of the study was to elucidate the influence of mixtures on absorption through human skin but in order to investigate the underlying mechanism, and to establish whether effects were specific to human skin, absorption was also determined with rat skin and a polydimethylsiloxane
Fig. 2. Cumulative absorption profiles of butoxyethanol from water mixtures applied to dermatomed rat skin in flow through diffusion cells as infinite dose (200 μl/cm2) for 24 h. Results are mean ± SEM (n = 5).
Concentration of butoxyethanol (mg/ml)
Maximum absorption rate (mg/cm2/h)
Apparent kp (×10− 3 cm/h)
Lag time (h)
900 (neat) 810 (90% v/v) 675 (75% v/v) 450 (50% v/v) 90 (10% v/v) 45 (5% v/v) 9 (1% v/v) 4.5 (0.5% v/v) 0.9 (0.1% v/v) 900 (neat) finite 810 (90% v/v) finite 450 (50% v/v) finite
0.39 ± 0.06 0.72 ± 0.11 2.34 ± 0.28*** 2.26 ± 0.40*** 0.66 ± 0.14 0.22 ± 0.01 0.14 ± 0.03 0.05 ± 0.01 0.01 ± 0.00 0.04 ± 0.01 0.08 ± 0.02 0.19 ± 0.03**
0.44 ± 0.07 0.88 ± 0.14 3.41 ± 0.41 5.95 ± 0.89** 7.34 ± 1.52*** 4.88 ± 0.28 12.5 ± 2.3*** 10.4 ± 2.5*** 15.3 ± 2.2*** n/a n/a n/a
0.6 ± 0.07 0.6 ± 0.04 0.6 ± 0.02 0.7 ± 0.04 0.8 ± 0.06 0.8 ± 0.05 0.5 ± 0.04 0.7 ± 0.07 0.5 ± 0.09 n/a n/a n/a
Results are mean ± SEM (n ≥ 5). **P < 0.01, ***P < 0.001 when compared to neat butoxyethanol. n/a = not applicable.
membrane. The effect of dilution with water on absorption of butoxyethanol was much less marked with rat skin than human skin. The maximum absorption rate for butoxyethanol
Fig. 3. Profile of the influence of dilution of glycol ethers in water (a) butoxyethanol and (b) ethoxyethanol on maximum absorption rate (mean ± SEM) through human skin in the flow through cell.
M.J. Traynor et al. / Toxicology and Applied Pharmacology 218 (2007) 128–134
131
Table 2 Influence of butoxyethanol concentration on percutaneous absorption of butoxyethanol (applied as a neat liquid and in aqueous solution) through dermatomed rat skin (infinite dose 200 μl/cm2)
Table 3 Influence of butoxyethanol concentration on percutaneous absorption of butoxyethanol (applied as a neat liquid and in aqueous solution) through polydimethylsiloxane membrane (infinite dose 200 μl/cm2)
Concentration of butoxyethanol (mg/ml)
Maximum absorption rate (mg/cm2/h)
Apparent kp (×10− 3 cm/h)
Lag time (h)
Concentration of butoxyethanol (mg/ml)
Maximum absorption rate (mg/cm2/h)
kp (×10− 3 cm/h)
Lag time (h)
900 (neat) 810 (90% v/v) 450 (50% v/v)
0.73 ± 0.11 1.45 ± 0.09 1.31 ± 0.08
0.81 ± 0.12 1.79 ± 0.11* 2.92 ± 0.19***
0.6 ± 0.06 0.7 ± 0.05 0.8 ± 0.07
900 (neat) 810 (90% v/v) 450 (50% v/v)
9.05 ± 0.82 7.36 ± 0.59 4.89 ± 0.22***
10.06 ± 0.91 9.08 ± 0.72 10.86 ± 0.49
0.4 ± 0.05 0.6 ± 0.05 0.7 ± 0.04
Results are mean ± SEM (n ≥ 5). *P < 0.05, ***P < 0.001 when compared to neat butoxyethanol.
Results are mean ± SEM (n ≥ 5). ***P < 0.001 when compared to neat butoxyethanol.
in aqueous solution (810 mg/ml) with rat skin was 1.46 ± 0.09 mg/cm2/h, and this decreased to 1.31 ± 0.08 mg/cm2/h with 450 mg/ml butoxyethanol in aqueous solution (Fig. 2 and Table 2). These increases were not statistically significant when compared with neat butoxyethanol. The apparent kp increased significantly with decreasing concentration of butoxyethanol for both human and rat skin (Tables 1 and 2), although to a lesser extent with rat skin. There was no effect of butoxyethanol concentration on lag time with either rat or human skin (Tables 1 and 2). The recovery of butoxyethanol during the experiments was between 85% and 105%. Expressed as a percentage of the applied dose, the cumulative amount of butoxyethanol recovered in the receptor fluid after 20 h increased from 1.6% for neat butoxyethanol to 44% for 0.1 % (v/v) (0.9 mg/ml) butoxyethanol in water with human skin. Between 60% and 80% of the dose in the human skin experiments was recovered from the skin surface in the wash after 4 h, with less than 10% being recovered from the charcoal trap (Fig. 4). Histological studies performed at 24 h for neat butoxyethanol, 50% (v/v) butoxyethanol in water and water alone showed no damage to the stratum corneum. There was evidence of hydration of the epidermis with water (results not shown) and suggestion of reduction in stratum corneum stability with butoxyethanol. Lack of damage was consistent with no effect on lag time. When a finite dose was applied to human skin and removed at 4 h, the absorption profile was similar to that following an infinite dose removed at 4 h although the total recovery was lower and 20% was recovered from the carbon filters. The
maximum absorption rate was, as with an infinite dose, significantly greater for aqueous butoxyethanol (450 mg/ml) than for neat butoxyethanol (Table 1), the increase being approximately fourfold. Again, there was no significant effect on the apparent lag time. At 12 h, 2.3% of a neat dose was recovered in the receptor fluid compared to 8.7% of the aqueous butoxyethanol. The similarity of effects between finite and infinite doses indicate that the effects are independent of direct stratum corneum damage and delipidisation as this would have been expected to be greater with the infinite dose. Neat butoxyethanol and 90% (v/v) and 50% (v/v) butoxyethanol in water were applied to the membrane. Absorption of butoxyethanol through the polydimethylsiloxane membrane was greater than through rat or human skin. The maximum absorption rate through polydimethylsiloxane membrane when an infinite dose was applied was 9.05 ± .0.82 mg/cm2/h and the apparent kp was 10 ± 0.9 × 10− 3 cm/h (Table 3). In contrast to both human and rat skins, the maximum absorption rate decreased as the dose concentration was decreased by dilution with water and the apparent kp remained constant.
Fig. 4. Distribution of butoxyethanol following application as a neat liquid and as water mixtures to dermatomed human skin. Results are expressed as percentage of applied dose in each area and total recovery; mean ± SEM (n ≥ 5).
Absorption of ethoxyethanol from water mixtures through human skin in vitro To determine whether the effects of water mixtures were restricted to butoxyethanol, absorption of ethoxyethanol was studied. The cumulative absorption profile for neat ethoxy-
Fig. 5. Cumulative absorption profiles of 14C ethoxyethanol recovered in receptor fluid from a range of water mixtures applied as infinite dose (200 μl/cm2) to dermatomed human skin in flow through diffusion cells. Results are mean± SEM (n ≥ 5).
132
M.J. Traynor et al. / Toxicology and Applied Pharmacology 218 (2007) 128–134
Table 4 Influence of ethoxyethanol concentration on percutaneous absorption of ethoxyethanol (applied as a neat liquid and in aqueous solution) through dermatomed human skin (infinite dose 200 μl/cm2)
Table 5 Percutaneous absorption of water (applied as a neat liquid and as a 50% (v/v) butoxyethanol or ethoxyethanol aqueous solution) though dermatomed human skin (infinite dose 200 μl/cm2)
Concentration of ethoxyethanol (mg/ml)
Maximum absorption rate (mg/cm2/h)
kp (×10− 3 cm/h)
Lag time (h)
Concentration of water (mg/ml)
Maximum absorption rate (mg/cm2/h)
kp (×10− 3 cm/h)
Lag time (h)
930 (neat) 837 (90% v/v) 697 (75% v/v) 465 (50% v/v) 93 (10% v/v) 46.5 (5% v/v) 9.3 (1% v/v) 4.65 (0.5% v/v) 0.93 (0.1% v/v)
0.64 ± 0.09 1.36 ± 0.26 1.87 ± 0.31*** 1.51 ± 0.25* 0.05 ± 0.00 0.03 ± 0.01 0.004 ± 0.00 0.002 ± 0.00 0.004 ± 0.00
0.69 ± 0.09 1.63 ± 0.31 2.68 ± 0.45** 3.28 ± 0.54*** 0.52 ± 0.22 0.57 ± 0.15 0.5 ± 0.08 0.5 ± 0.1 0.5 ± 0.06
0.9 ± 0.05 0.9 ± 0.04 1.1 ± 0.03 0.6 ± 0.04 1.0 ± 0.01 1.0 ± 0.04 0.9 ± 0.06 1.0 ± 0.04 0.9 ± 0.05
1000 neat 500 in butoxyethanol 500 in ethoxyethanol
4.72 ± 0.52 5.17 ± 0.70 5.28 ± 0.40
4.71 ± 0.52 10.34 ± 1.40* 10.56 ± 0.80*
0.84 ± 0.04 0.64 ± 0.05 0.77 ± 0.04
Results are mean ± SEM (n ≥ 5). *P < 0.05, ***P < 0.001 when compared to neat ethoxyethanol.
ethanol through human skin in infinite dose (Fig. 5) was similar to that for neat butoxyethanol (Figs. 1 and 3b). The maximum absorption rate of neat ethoxyethanol through human skin in infinite dose conditions was 0.64 ± 0.09 mg/cm2/h. As with butoxyethanol, this increased when aqueous solutions of ethoxyethanol were applied. A maximum absorption rate of 1.87 ± 0.31 mg/cm2/h which was measured with 75% (v/v) aqueous ethoxyethanol (697 mg/ml) about threefold higher than with neat ethoxyethanol. Concentrations of ethoxyethanol in aqueous solution lower than 10% (v/v) (93 mg/ml) exhibited lower maximum absorption rates than neat ethoxyethanol (Table 4). The apparent kp increased from 0.69 ± 0.09 × 10− 3 cm/h for neat ethoxyethanol to 3.2 ± 0.5 × 10− 3 cm/h from a 465 mg/ml solution (Table 4) but there was, again, no significant effect on the lag time. The proportion of ethoxyethanol recovered in receptor fluid increased from 2.8% with neat ethoxyethanol to 6.2% with 50% (v/v) aqueous ethoxyethanol, but, in contrast to butoxyethanol, the proportion of ethoxyethanol measured in receptor fluid decreased with dose concentrations of 10% (v/v) and lower to a constant proportion of the applied dose that was similar to that for neat ethoxyethanol. As with butoxyethanol, 80–95% of the dose
Fig. 6. Distribution of ethoxyethanol following application as a neat liquid and as water mixtures to dermatomed human skin. Results are expressed as percentage of applied dose in each area and total recovery; mean ± SEM (n ≥ 5).
Results are mean ± SEM (n ≥ 5). *P < 0.05 when compared to neat water.
was recovered from the skin surface, with about 10% recovered in the charcoal traps (Fig. 6). Absorption of water from glycol ether mixtures through human skin in vitro The maximum absorption rate of an infinite dose of water through human skin was 4.72 ± 0.52 mg/cm2/h. This rate was not significantly reduced when water was mixed with butoxyethanol or ethoxyethanol, despite the proportional decrease in the concentration of the water (Table 5). The apparent kp for water was significantly greater in the presence of ethoxyethanol or butoxyethanol, but the increase was not significantly different when a mixture of ethoxyethanol and butoxyethanol was used. There was no significant effect of either solvent on the lag time for water absorption. Discussion In the current study, the highest maximum absorption rate for butoxyethanol measured in human skin (2.34 ± 0.28 mg/cm2/h) was obtained with an infinite dose of a 450 mg/ml aqueous solution (50% v/v solution). This rate was six times greater than the maximum absorption rate of neat butoxyethanol (0.39 ± 0.06 mg/cm2/h). Dilution of butoxyethanol to 90% (v/v) in water resulted in the maximum absorption rate almost doubling compared to neat butoxyethanol, while dilution to 10% (v/v) in water resulted in a maximum absorption rate that was still greater than with neat butoxyethanol. These results clearly indicated that water promoted the absorption of butoxyethanol through human skin in vitro. There was a similar effect of water on the absorption of butoxyethanol from 90% and 50% solutions through rat skin although the proportional increase was not as great as that seen with human skin. Therefore, studies in the rat are of no value in predicting the profile supporting the need for in vitro studies with human skin. Volunteer studies in which 50% butoxyethanol/water was applied to 40 cm2 of the forearm for 4 h and standardised by comparison to an inhalation study (Jakasa et al., 2004) gave measurements of dermal absorption rate of 2.7 mg/cm2/h which was consistent with the in vitro predictions here and the kp increased from 0.88 × 10− 3 to 1.75 × 10− 3 cm/h. For the rat in vivo, a previous measurement of the dermal absorption rate for a 50% butoxyethanol solution was 1.3 mg/cm2/h which was also similar to the in vitro prediction (Payan, personal communication).
M.J. Traynor et al. / Toxicology and Applied Pharmacology 218 (2007) 128–134
In contrast, the absorption of butoxyethanol through polydimethylsiloxane membrane obeyed Fick's first law of diffusion when diluted with water i.e. rate of absorption was proportional to the concentration of the penetrating chemical. The apparent kp was independent of dose concentration, in contrast to the studies with human skin, in which the apparent kp increased with decreased dose concentration. Therefore the membrane was not an ideal surrogate for skin for studying absorption of butoxyethanol/water mixtures. This observation also makes interpretation of the results obtained with skin more difficult, as it indicates that an increase in thermodynamic activity was not solely responsible for the increase in absorption measured for aqueous solutions of glycol ether. Bunge (2006) has defined the influence of mixing butoxyethanol and water on the thermodynamic activity of the two components. The profile indicated three phases: a linear relationship between thermodynamic activity and concentration between 0% and 20% butoxyethanol, between 20% and 70% little change and then a linear increase to 100%. It is reasonable to compare the profile of maximum absorption rate for butoxyethanol and water with concentration in these studies with that of thermodynamic activity of the solutions. The profiles were similar at low butoxyethanol concentrations but they deviated at high butoxyethanol concentrations where dehydration of the skin occurred. The thermodynamic activity of water remained high even at high concentrations of butoxyethanol in parallel with the observed maximum absorption rate. The maximum absorption rate for neat ethoxyethanol through human skin in this study was 0.64 ± 0.09 mg/cm2/h which was similar to the measurement by Filon et al. (1999), who reported a maximum penetration rate of 0.83 ± 0.4 mg/ cm2/h for neat ethoxyethanol when applied to abdominal skin mounted in a Franz cell. The greatest maximum absorption rate for ethoxyethanol was seen from a 50% aqueous solution, which was threefold greater than for neat ethoxyethanol. As with butoxyethanol a substantial increase in maximum absorption rate was observed when 90% (v/v) butoxyethanol in water was used. The apparent kp of ethoxyethanol was also increased in the presence of water up to a maximum with a 50% aqueous solution. The dermal absorption of water from the glycol ether/water mixtures was also increased compared to water alone. When the amount of water present in the dose solution was halved, there was a slight increase in the maximum penetration rate of water in the presence of both butoxyethanol and ethoxyethanol. The apparent kp of the water also increased. Therefore, similar effects of water mixtures were observed for two glycol ethers suggesting that this is a class effect and may have implications for risk assessment of glycol ethers. Wilkinson and Williams (2002) previously showed a higher apparent kp and an increased proportion of the applied dose absorbed of glycol ethers from aqueous solutions compared with neat butoxyethanol, although not an absolute increase in absorption rate, as measured in the present study. However, the concentration of aqueous butoxyethanol studied in the 2002 paper was low, 3 mg/ml, and this may not have been sufficient
133
to observe the increase in absorption rate reported in the present study. It has been reported that the presence of water increased the absorption of butoxyethanol through guinea pig skin in vivo (Johanson and Fernstrom, 1988). These results contribute to our understanding of the mechanism for increased penetration of butoxyethanol and ethoxyethanol into skin in aqueous vehicles. The barrier properties of the skin reside in the stratum corneum and the ordered structure of the lipid lamellae and corneocytes is responsible for low permeability to chemicals. Perturbation of this highly ordered structure may result in an increase in permeability. Effects of either butoxyethanol or ethoxyethanol on the stratum corneum structure and/or hydration of the skin may contribute to the increase in maximum absorption rate measured for glycol ethers in the presence of water. An increase in absorption rate may result from an alteration in the barrier properties of the stratum corneum and/or a greater degree of partitioning of the glycol ethers into the stratum corneum. Glycol ethers are good solvents for lipophilic as well as hydrophilic compounds, and they may have a solubilising effect on the stratum corneum lipids similar to that described for benzyl alcohol, where formation of micelles and disruption of lipid layers is associated with promoting effects (Nanayakker et al., 2005). When dehydrating solvents are applied to the skin mixed with water the disruptive effect on stratum corneum lipid structure and barrier function may increase the permeability to both the solvent and water (Van der Merwe and Riviere, 2005). It has been shown that neat solvents such as ethanol dehydrate the stratum corneum (Marjukka Suhonen et al., 1999; Pillai et al., 2004). Similarly neat glycol ethers may have a dehydrating effect on the skin contributing to less flux, so that with the addition of water and rehydration of the skin an increase in absorption could occur. Wilkinson and Williams (2002) suggested that there was increased partitioning of butoxyethanol from the aqueous vehicle into the skin. At certain concentrations, butoxyethanol in water does not behave as a perfect solution but the butoxyethanol molecules cluster together in pseudomicelles (Castillo and Dominguez, 1990) and, if these were in contact with skin, they might preferentially partition out of the water and into the lipid-rich stratum corneum. In this study there was no depletion of the concentration of the test dose in contact with the skin with time, but as absorption of water also increased, changes in concentration might not have occurred. Also it is known that the state of hydration of the skin will affect penetration. With increased hydration within a few hours, there is evidence of accumulation of water in the intercellular spaces disrupting the lamellar arrangement of the lipids (Warner et al., 2003). This can lead to increased penetration of water-soluble chemicals such as glycol ethers (Tezel et al., 2003). Application of water to the skin for 1 or 3 h before butoxyethanol increased the maximum absorption rate by three fold (results not shown), but this increase was less than that measured with butoxyethanol/water mixtures, suggesting that the hydration by water is only partly responsible. Also the absolute absorption rate of water was greater than that of butoxyethanol. Water penetration
134
M.J. Traynor et al. / Toxicology and Applied Pharmacology 218 (2007) 128–134
in butoxyethanol/water mixtures might enhance butoxyethanol penetration by solvent drag. In conclusion, this study has found that the dermal absorption of butoxyethanol and ethoxyethanol is increased in the presence of water. Mechanisms for this phenomenon may involve disruption of lipid bilayer by glycol ether and dehydration of the skin by neat solvent, which contribute to reduced flux with neat glycol ethers. Hydration following application of water mixtures resulted in increased flux. This information is important for risk assessment of exposure to glycol ethers which have industrial and household uses in aqueous solution. Some predictive models based on quantitative structure activity relationships (QSARs) are used by risk assessors to predict kp values for substances for which experimental data are not available (Fitzpatrick et al., 2004). The experimental datasets on which these models are based consist of absorption studies employing mainly aqueous vehicles in “infinite” dose. This approach should not be applied to any solvent, which is freely soluble in water, as the influence of vehicle or formulation or finite dose is not considered. This further emphasises the need to determine absorption rates using in vitro approaches with human skin and a relevant exposure. Risk assessments for dermal contact scenarios should ideally be based on absorption data obtained for the relevant mixture and particular attention focussed on abnormal absorption from aqueous solutions. It is important to understand the role of skin interactions in deviation of dermal absorption from predictions and investigation of absorption in disease states may help. Acknowledgment This work was funded by the Health and Safety Executive Magdalen House, Stanley Precinct, Bootle, Merseyside, L20 3QZ, UK. References Bartnik, F.G., Reddy, A.K., Klecak, G., Zimmermann, V., Hostynek, J.J., Kunstler, K., 1987. Percutaneous absorption, metabolism, and hemolytic activity of n-butoxyethanol. Fundam. Appl. Toxicol. 8 (1), 59–70. Bunge, A.L., 2006. Why skin permeation data from neat and aqueous solutions of butoxyethanol is not surprising. Perspectives of Percutaneous Penetration, vol. 10A, p. 24. Castillo, R.C., Dominguez, H.C., 1990. Determination of mutual diffusion coefficients in water-rich 2-butoxyethanol/water mixtures using the Taylor dispersion technique. J. Phys. Chem. 94, 8731–8734. Chilcott, R.P., Barai, N., Beezer, A.E., Brain, S.I., Brown, M.B., Bunge, A.L., Burgess, S.E., Cross, S., Dalton, C.H., Dias, M., Farinha, A., Finnin, B.C., Gallagher, S.J., Green, D.M., Gunt, H., Gwyther, R.L., Heard, C.M., Jarvis, C.A., Kamiyama, F., Kasting, G.B., Ley, E.E., Lim, S.T., McNaughton, G.S., Morris, A., Nazemi, M.H., Pellett, M.A., Du Plessis, J., Quan, Y.S., Raghavan, S.L., Roberts, M., Romonchuk, W., Roper, C.S., Schenk, D., Simonsen, L., Simpson, A., Traversa, B.D., Trottet, L., Watkinson, A., Wilkinson, S.C., Williams, F.M., Yamamoto, A., Hadgraft, J., 2005. Interand intralaboratory variation of in vitro diffusion cell measurements: an
international multicenter study using quasi-standardized methods and materials. J. Pharm. Sci. 94 (3), 632–638. Filon, F.L., Fiorito, A., Adami, G., Barbieri, P., Coceani, N., Bussani, R., Reisenhofer, E., 1999. Skin absorption in vitro of glycol ethers. Int. Arch. Occup. Environ. Health 72, 480–484. Fitzpatrick, D., Corish, J., Hayes, B., 2004. Modelling skin permeability in risk assessment—The future. Chemosphere 55 (10), 1309–1314. Ghanayem, B.I., Burka, L.T., Sanders, J.M., Matthews, H., 1987. Metabolism and disposition of ethylene glycol monobutyl ether (2-butoxyethanol) in rats. Drug Metab. Dispos. 15 (4), 478–484. Ghanayem, B.I., Burka, L.T., Matthews, H., 1989. Structure–activity relationships for the in vitro hematotoxicity of N-alkoxyacetic acids, the toxic metabolites of glycol ethers. Chem.-Biol. Int. 70 (3–4), 339–352. Hardin, B.D., 1983. Reproductive toxicity of the glycol ethers. Toxicology 26, 91–102. Hardin, B.D., Goad, P.T., Burg, J.R., 1984. Developmental toxicity of four glycol ethers applied cutaneously to rats. Environ. Health Perspect. 57, 69–74. Jakasa, I., Mohammadi, N., Kruse, J., Kezic, S., 2004. Percutaneous absorption of neat and aqueous solutions of 2-butoxyethanol in volunteers. Int. Arch. Occup. Environ. Health 77 (2), 79–84. Johanson, G., 1991. Percutaneous absorption of 2-butoxyethanol vapour in human subjects. Br. J. Ind. Med. 48 (11), 788–792. Johanson, G., Boman, A., Dynesius, B., 1988. Percutaneous absorption of 2butoxyethanol in man. Scand. J. Work, Environ. Health 14 (2), 101–109. Johanson, G., Fernstrom, P., 1988. Influence of water on the percutaneous absorption of 2-butoxyethanol in guinea pigs. Scand. J. Work, Environ. Health 14 (2), 95–100. Kao, J., Carver, M.P., 1990. Cutaneous metabolism of xenobiotics. Drug Metab. Rev. 22 (4), 363–410. Kezic, S., Mahieu, K., Monster, A.C., de Wolff, F.A., 1997. Dermal absorption of vaporous and liquid 2-methoxyethanol and 2-ethoxyethanol in volunteers. Occup. Environ. Med. 54 (1), 38–43. Lockley, D.J., Howes, D., Williams, F.M., 2002. Percutaneous penetration and metabolism of 2-ethoxyethanol. Toxicol. Appl. Pharmacol. 180 (2), 74–82. Lockley, D.J., Howes, D., Williams, F.M., 2004a. Percutaneous penetration and metabolism of 2-butoxyethanol. Arch. Toxicol. 78 (11), 617–628. Lockley, D.J., Howes, D., Williams, F.M., 2004b. Metabolism of glycol ethers. Arch. Toxicol. 78 (11), 617–628. Marjukka Suhonen, T., Bouwstra, J.A., Urtti, A., 1999. Chemical enhancement of percutaneous absorption in relation to stratum corneum structural alterations. J. Con. Release 59 (2), 149–161. Multinger, L., Catala, M., Cordier, S., Delaforge, M., Fenaux, P., Garnier, R., Rico-Lattes, I., Vasseur, P., 2005. The INSEM expert review on glycol ethers: findings and recommendations. Toxicol. Lett. 156, 29–37. Nanayakker, G.R., Bartlett, A., Forbes, B., Marriott, C., Whitfield, P.J., Brown, M.B., 2005. The effect of unsaturated fatty acids in benzyl alcohol on the permeation of model penetrants. Int. J. Pharm. 301, 129–139. Pillai, O., Nair, V., Panchagnula, R., 2004. Transdermal iontophoresis of insulin: IV. Influence of chemical enhancers. Int. J. Pharm. 269 (1), 109–120. Payan J.P., personal communication. Tezel, A., Sens, A., Mitragotri, S., 2003. Description of transdermal transport of hydrophilic solutes during low-frequency sonophoresis based on a modified porous pathway model. J. Pharm. Sci. 92 (2), 381–393. Van der Merwe, D., Riviere, J.E., 2005. Comparative studies on the effects of water, ethanol and water/ethanol mixtures on chemical partitioning into porcine stratum corneum and silastic membrane. Toxicol. in vitro 19 (1), 69–77. Wilkinson, S.C., Williams, F.M., 2002. Effects of experimental conditions on absorption of glycol ethers through human skin in vitro. Int. Arch. Occup. Environ. Health 75 (8), 519–527. Warner, R.R, Stone, K.J., Boissy, Y.I., 2003. Hydration disrupts human stratum corneum ultrastructure. J. Invest. Dermatol. 130, 275–284.
The influence of water mixtures on the dermal absorption of glycol ethers Matthew J. Traynor, Simon C. Wilkinson, Faith M. Williams ⁎ Toxicology Unit, Institute for Research on Environment and Sustainability and Medical School, University of Newcastle upon Tyne, Newcastle NE1 7RU, UK Received 2 August 2006; revised 28 September 2006; accepted 29 September 2006 Available online 10 November 2006
Abstract Glycol ethers are solvents widely used alone and as mixtures in industrial and household products. Some glycol ethers have been shown to have a range of toxic effects in humans following absorption and metabolism to their aldehyde and acid metabolites. This study assessed the influence of water mixtures on the dermal absorption of butoxyethanol and ethoxyethanol in vitro through human skin. Butoxyethanol penetrated human skin up to sixfold more rapidly from aqueous solution (50%, 450 mg/ml) than from the neat solvent. Similarly penetration of ethoxyethanol was increased threefold in the presence of water (50%, 697 mg/ml). There was a corresponding increase in apparent permeability coefficient as the glycol ether concentration in water decreased. The maximum penetration rate of water also increased in the presence of both glycol ethers. Absorption through a synthetic membrane obeyed Fick's Law and absorption through rat skin showed a similar profile to human skin but with a lesser effect. The mechanisms for this phenomenon involves disruption of the stratum corneum lipid bilayer by desiccation by neat glycol ether micelles, hydration with water mixtures and the physicochemical properties of the glycol ether–water mixtures. Full elucidation of the profile of absorption of glycol ethers from mixtures is required for risk assessment of dermal exposure. This work supports the view that risk assessments for dermal contact scenarios should ideally be based on absorption data obtained for the relevant formulation or mixture and exposure scenario and that absorption derived from permeability coefficients may be inappropriate for water-miscible solvents. © 2006 Elsevier Inc. All rights reserved. Keywords: Glycol ether; Dermal absorption; Water; Mixtures
Introduction Glycol ethers are widely used alone and as mixtures in industrial and household applications because their physicochemical properties, miscibility with both water and organic media make them versatile solvents. Glycol ethers penetrate the skin rapidly (Kezic et al., 1997; Filon et al., 1999) and it has recently been reported that following a 1-h exposure of the human forearm to a 50% butoxyethanol/50% water mixture, the body burden exceeded that which would be obtained from an 8h inhalation exposure at the level of the threshold limit value set by the American Conference of Governmental Industrial Hygienists (ACGIH) (Jakasa et al., 2004). The toxicity of glycol ethers to man is mainly caused by the aldehyde and acid metabolites (Ghanayem et al., 1987) after conversion by alcohol dehydrogenase and aldehyde dehydrogenase. The epidermis has been shown to contain alcohol and aldehyde dehydrogenases ⁎ Corresponding author. E-mail address: [email protected] (F.M. Williams). 0041-008X/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.taap.2006.09.019
(Kao and Carver, 1990; Lockley et al., 2004a), but local metabolism of glycol ethers during dermal penetration was not detected for skin mounted in a flow through diffusion system (Lockley et al., 2002, 2004b), suggesting that the rapid passage of the solvents through the skin limits access to the enzymes. However, using skin in short-term culture with increased sensitivity allowed detection of a small amount of conversion (Traynor et al, in preparation). Butoxyethanol has been demonstrated to cause erythrocyte haemolysis following metabolism to butoxyacetic acid (Bartnik et al., 1987; Ghanayem et al., 1989; Multinger et al., 2005). Ethoxyethanol has been shown to have haematological, developmental effects and reproductive effects (Hardin, 1983; Hardin et al., 1984) in laboratory animals. There is also some evidence for effects in exposed workers. Due to the ease with which the glycol ethers are absorbed through the skin (Johanson et al., 1988; Kezic et al., 1997; Filon et al., 1999) and the potential for the development of adverse health effects, it is important to understand factors that influence absorption. There have been a number of studies of dermal absorption of glycol
M.J. Traynor et al. / Toxicology and Applied Pharmacology 218 (2007) 128–134
ether vapours in humans in vivo (Johanson and Boman, 1991; Kezic et al., 1997) and of undiluted liquids in vitro (Johanson et al., 1988; Kezic et al., 1997). Jakasa et al. (2004) also investigated the absorption of aqueous solutions of glycol ethers through the forearm of volunteers. The ability of in vitro studies to predict in vivo absorption of butoxyethanol and ethoxyethanol neat or from methanol in the rat has been demonstrated (Lockley et al., 2002, 2004b). It has previously been suggested that glycol ethers absorb more effectively through skin when in contact as water mixtures. The aim of this study was to use in vitro methods to investigate the influence of applying ethoxyethanol and butoxyethanol as water and other mixtures on absorption through human skin. It is important to define this for risk assessment of dermal penetration of glycol ethers and to identify conditions in which there may be significant deviation from the expected absorption profile. Materials and methods [1-14C]2-butoxyethanol, specific activity 430 MBq/mmol, was obtained from Amersham. [1-14C]2-ethoxyethanol, specific activity 0.04 MBq/mg, was a gift from Unilever and 3H water, specific activity 1600 MBq/ml, was purchased from ICN radiochemicals (Basingstoke, UK). All radiochemicals had a radiochemical purity exceeding 97%. Minimal essential medium (Eagle) and gentamycin were obtained from Sigma; sodium hydrogen carbonate and teepol L were purchased from BDH; HiSafe 3 scintillation fluid was obtained from Fisher; 2-butoxyethanol (99.9% purity) was obtained from Fluka; and 2ethoxyethanol (99.5% purity) from Aldrich. The water used was sterile water for irrigation. The polydimethylsiloxane membrane was kindly donated by Dr R. Chilcott DSTL (product code 19TO.3-1000-60M1 SAMCO Silicone Products) (Chilcott et al., 2005). Skin preparation. Human breast skin was obtained following mammoplasty from a local hospital and stored at − 70 °C until required. Ethical approval for obtaining skin was given by the University of Newcastle Medical and Dental Ethics Committee and the University Hospital of South Durham Ethics Committee. A section of skin was removed from the freezer and defrosted at room temperature. The skin was then placed dermal side down on a corkboard and dermatomed to a thickness of 320 μm. All experiments were performed using human skin from at least two donors. Rat skin was obtained from male Wistar rats (28 days old), sacrificed by cervical dislocation. The dorsal region was shaved and the skin dissected. The skin was placed dermal side down on a corkboard and dermatomed to a thickness of 280 μm. A polydimethylsiloxane membrane (thickness 400 ± 13 μm) was used in some experiments. The membrane was prepared for use by soaking in sterile water for 24 h prior to use. In vitro flow through diffusion system. Teflon flow through cells of the Newcastle Scott Dick design were used. Receptor fluid (Eagle's minimal essential medium supplemented with 2.2 g/l sodium hydrogen carbonate and 200 μg/ml gentamycin) was maintained at pH 7.4 by gassing with 5% CO2/air and pumped through the cells at 1.5 ml/h. The receptor fluid reservoir and diffusion cells were maintained at approximately 37 °C using a water jacket connected to a circulating water bath. This maintained the surface of the skin at 32 °C. Skin sections were placed in the diffusion cells (exposed surface area 0.64 cm2) and secured in place using threaded nuts. The cells were partially occluded by a tight-fitting cap containing carbon filters which allowed air to pass through the filters. Dose application and determination of diffusion and distribution. The study procedure followed OECD Guidelines Butoxyethanol was used neat (900 mg/ ml) or diluted in aqueous solution at concentrations ranging from 90% v/v to 0.1% v/v (810 mg/ml to 0.9 mg/ml). Ethoxyethanol was used neat (930 mg/ml) or in aqueous solution at concentrations ranging from 90% v/v to 0.1% v/v (837 mg/ ml to 0.93 mg/ml). Test compounds (including 14C butoxyethanol or 14C ethoxyethanol) were applied at 200 μl/cm2 (infinite dose) or 20 μl/cm2 (finite dose). 14C radioactivity amounted to 3.0 kBq per diffusion cell for butoxyethanol
129
and 58 kBq per diffusion cell for ethoxyethanol. In certain studies, absorption of water was studied using 3H2O. Water was applied neat or as a 1:1 (v/v) mixture with butoxyethanol and/or ethoxyethanol for 4 h. Tritium radioactivity amounted to 2.0 MBq per diffusion cell. Receptor fluid fractions were collected every 30 min until 3 h after dose application and then every 60 min until the experiments were terminated at 20 h after dose application. In human skin experiments, the dose was removed from the skin surface after 4 h using alternate wet (soaked in 3% (v/v) teepol L) and dry tissue swabs (6 swabs in total). For studies with rat skin and membrane the dose remained in contact with the skin for the full study and the dose remaining on the skin surface at termination of the study was recovered in the same way. Weighed aliquots (250 μl) of receptor fluid samples were taken for scintillation counting, after mixing with scintillation fluid (3 ml). Tritium-labelled samples were left overnight in darkness before scintillation counting. The tissue swabs used to remove the dose solution were soaked in scintillation fluid (10 ml) and left to desorb for 72 h before scintillation counting. The carbon filters from the traps above each cell were soaked in scintillation fluid (10 ml) and left to desorb for 2 days before counting. At the end of the experiment the skin was removed from the cells and digested using 2 ml of 1.5 mol/l potassium hydroxide in 4:1 (v/v) methanol/water. Once the skin was fully digested (72 h), glacial acetic acid (70 μl) was added to quench chemiluminescence and 10 ml of scintillation fluid was added prior to counting. Scintillation counting was carried out using a Wallac 1410 LSC. HiSafe 3 liquid scintillation cocktail was used throughout. Calculation of absorption parameters. A cumulative absorption–time curve was constructed for amount of test chemical measured in the receptor fluid with time. The maximum absorption rate was calculated from the slope of the linear region of the curve (R2 ≥ 0.95). This was generally between 1 and 3.5 h for short-term 4-h applications and up to 6 h for 24-h doses. For an infinite dose the apparent permeability coefficient (kp) was calculated by dividing the maximum absorption rate by the concentration of the test compound in the dose solution. The lag time was obtained from the intercept of the linear portion of cumulative absorption–time curve with the time axis. Statistical comparisons were made using one-way ANOVA followed by post hoc testing using Bonferroni's correction. The chosen level of significance was P < 0.05.
Results Absorption of butoxyethanol from water mixtures through skin in vitro All of the studies were performed under partially occluded conditions in which the cell was covered with a charcoal trap which allowed air passage through the mesh, and the skin was exposed to either an infinite dose (large volume, concentration not depleted during the study) or a finite dose (small volume depleting during the study). The profile of absorption of butoxyethanol through the human skin comprised a short lag phase followed by increasing absorption, with the linear portion of the profile occurring between 0.5 and 4 h. In studies with human skin, the dose was washed off at 4 h to relate the results to those obtained in vivo by Jakasa et al. (2004) (Fig. 1). A similar absorption profile was obtained with rat skin, although the linear portion of the profile was more prolonged than with human skin as the dose was not washed off until the end of the study (Fig. 2). Rat skin was more permeable to neat butoxyethanol than human skin. A maximum absorption rate for neat butoxyethanol (900 mg/ml) of 0.73 ± 0.01 mg/cm2/h was measured with rat skin, compared to 0.39 ± 0.06 mg/cm2/h with human skin, following application as an infinite dose (Tables 1 and 2). The maximum absorption rate for butoxyethanol with human skin increased significantly when butoxyethanol was diluted
130
M.J. Traynor et al. / Toxicology and Applied Pharmacology 218 (2007) 128–134 Table 1 Influence of butoxyethanol concentration on percutaneous absorption of butoxyethanol (applied as a neat liquid and in aqueous solution) through dermatomed human skin (infinite dose 200 μl/cm2 and finite dose 20 μl/cm2)
Fig. 1. Cumulative absorption profiles of 14C butoxyethanol recovered in receptor fluid from a range of water mixtures applied as infinite dose (200 μl/cm2) to dermatomed human skin in flow through diffusion cells. Results are mean ± SEM (n ≥ 5).
with water. A maximum absorption rate of 2.34 ± 0.28 mg/cm2/h was measured with 675 mg/ml butoxyethanol in aqueous solution, and this increased further to 2.69 ± 0.40 mg/cm2/h with 450 mg/ml butoxyethanol in aqueous solution. This represented an increase of almost sixfold compared to neat butoxyethanol. The maximum absorption rates for butoxyethanol with aqueous solutions at 10% (v/v) (90 mg/ml) and below were not significantly greater, or less than, that measured with neat butoxyethanol. The profile of the relationship between maximum absorption rate and concentration of butoxyethanol indicated three phases: a linear relationship between absorption rate and concentration between 0% and 20% butoxyethanol, between 50% and 70% little change and then a decrease to neat butoxyethanol (Fig. 3a). The aim of the study was to elucidate the influence of mixtures on absorption through human skin but in order to investigate the underlying mechanism, and to establish whether effects were specific to human skin, absorption was also determined with rat skin and a polydimethylsiloxane
Fig. 2. Cumulative absorption profiles of butoxyethanol from water mixtures applied to dermatomed rat skin in flow through diffusion cells as infinite dose (200 μl/cm2) for 24 h. Results are mean ± SEM (n = 5).
Concentration of butoxyethanol (mg/ml)
Maximum absorption rate (mg/cm2/h)
Apparent kp (×10− 3 cm/h)
Lag time (h)
900 (neat) 810 (90% v/v) 675 (75% v/v) 450 (50% v/v) 90 (10% v/v) 45 (5% v/v) 9 (1% v/v) 4.5 (0.5% v/v) 0.9 (0.1% v/v) 900 (neat) finite 810 (90% v/v) finite 450 (50% v/v) finite
0.39 ± 0.06 0.72 ± 0.11 2.34 ± 0.28*** 2.26 ± 0.40*** 0.66 ± 0.14 0.22 ± 0.01 0.14 ± 0.03 0.05 ± 0.01 0.01 ± 0.00 0.04 ± 0.01 0.08 ± 0.02 0.19 ± 0.03**
0.44 ± 0.07 0.88 ± 0.14 3.41 ± 0.41 5.95 ± 0.89** 7.34 ± 1.52*** 4.88 ± 0.28 12.5 ± 2.3*** 10.4 ± 2.5*** 15.3 ± 2.2*** n/a n/a n/a
0.6 ± 0.07 0.6 ± 0.04 0.6 ± 0.02 0.7 ± 0.04 0.8 ± 0.06 0.8 ± 0.05 0.5 ± 0.04 0.7 ± 0.07 0.5 ± 0.09 n/a n/a n/a
Results are mean ± SEM (n ≥ 5). **P < 0.01, ***P < 0.001 when compared to neat butoxyethanol. n/a = not applicable.
membrane. The effect of dilution with water on absorption of butoxyethanol was much less marked with rat skin than human skin. The maximum absorption rate for butoxyethanol
Fig. 3. Profile of the influence of dilution of glycol ethers in water (a) butoxyethanol and (b) ethoxyethanol on maximum absorption rate (mean ± SEM) through human skin in the flow through cell.
M.J. Traynor et al. / Toxicology and Applied Pharmacology 218 (2007) 128–134
131
Table 2 Influence of butoxyethanol concentration on percutaneous absorption of butoxyethanol (applied as a neat liquid and in aqueous solution) through dermatomed rat skin (infinite dose 200 μl/cm2)
Table 3 Influence of butoxyethanol concentration on percutaneous absorption of butoxyethanol (applied as a neat liquid and in aqueous solution) through polydimethylsiloxane membrane (infinite dose 200 μl/cm2)
Concentration of butoxyethanol (mg/ml)
Maximum absorption rate (mg/cm2/h)
Apparent kp (×10− 3 cm/h)
Lag time (h)
Concentration of butoxyethanol (mg/ml)
Maximum absorption rate (mg/cm2/h)
kp (×10− 3 cm/h)
Lag time (h)
900 (neat) 810 (90% v/v) 450 (50% v/v)
0.73 ± 0.11 1.45 ± 0.09 1.31 ± 0.08
0.81 ± 0.12 1.79 ± 0.11* 2.92 ± 0.19***
0.6 ± 0.06 0.7 ± 0.05 0.8 ± 0.07
900 (neat) 810 (90% v/v) 450 (50% v/v)
9.05 ± 0.82 7.36 ± 0.59 4.89 ± 0.22***
10.06 ± 0.91 9.08 ± 0.72 10.86 ± 0.49
0.4 ± 0.05 0.6 ± 0.05 0.7 ± 0.04
Results are mean ± SEM (n ≥ 5). *P < 0.05, ***P < 0.001 when compared to neat butoxyethanol.
Results are mean ± SEM (n ≥ 5). ***P < 0.001 when compared to neat butoxyethanol.
in aqueous solution (810 mg/ml) with rat skin was 1.46 ± 0.09 mg/cm2/h, and this decreased to 1.31 ± 0.08 mg/cm2/h with 450 mg/ml butoxyethanol in aqueous solution (Fig. 2 and Table 2). These increases were not statistically significant when compared with neat butoxyethanol. The apparent kp increased significantly with decreasing concentration of butoxyethanol for both human and rat skin (Tables 1 and 2), although to a lesser extent with rat skin. There was no effect of butoxyethanol concentration on lag time with either rat or human skin (Tables 1 and 2). The recovery of butoxyethanol during the experiments was between 85% and 105%. Expressed as a percentage of the applied dose, the cumulative amount of butoxyethanol recovered in the receptor fluid after 20 h increased from 1.6% for neat butoxyethanol to 44% for 0.1 % (v/v) (0.9 mg/ml) butoxyethanol in water with human skin. Between 60% and 80% of the dose in the human skin experiments was recovered from the skin surface in the wash after 4 h, with less than 10% being recovered from the charcoal trap (Fig. 4). Histological studies performed at 24 h for neat butoxyethanol, 50% (v/v) butoxyethanol in water and water alone showed no damage to the stratum corneum. There was evidence of hydration of the epidermis with water (results not shown) and suggestion of reduction in stratum corneum stability with butoxyethanol. Lack of damage was consistent with no effect on lag time. When a finite dose was applied to human skin and removed at 4 h, the absorption profile was similar to that following an infinite dose removed at 4 h although the total recovery was lower and 20% was recovered from the carbon filters. The
maximum absorption rate was, as with an infinite dose, significantly greater for aqueous butoxyethanol (450 mg/ml) than for neat butoxyethanol (Table 1), the increase being approximately fourfold. Again, there was no significant effect on the apparent lag time. At 12 h, 2.3% of a neat dose was recovered in the receptor fluid compared to 8.7% of the aqueous butoxyethanol. The similarity of effects between finite and infinite doses indicate that the effects are independent of direct stratum corneum damage and delipidisation as this would have been expected to be greater with the infinite dose. Neat butoxyethanol and 90% (v/v) and 50% (v/v) butoxyethanol in water were applied to the membrane. Absorption of butoxyethanol through the polydimethylsiloxane membrane was greater than through rat or human skin. The maximum absorption rate through polydimethylsiloxane membrane when an infinite dose was applied was 9.05 ± .0.82 mg/cm2/h and the apparent kp was 10 ± 0.9 × 10− 3 cm/h (Table 3). In contrast to both human and rat skins, the maximum absorption rate decreased as the dose concentration was decreased by dilution with water and the apparent kp remained constant.
Fig. 4. Distribution of butoxyethanol following application as a neat liquid and as water mixtures to dermatomed human skin. Results are expressed as percentage of applied dose in each area and total recovery; mean ± SEM (n ≥ 5).
Absorption of ethoxyethanol from water mixtures through human skin in vitro To determine whether the effects of water mixtures were restricted to butoxyethanol, absorption of ethoxyethanol was studied. The cumulative absorption profile for neat ethoxy-
Fig. 5. Cumulative absorption profiles of 14C ethoxyethanol recovered in receptor fluid from a range of water mixtures applied as infinite dose (200 μl/cm2) to dermatomed human skin in flow through diffusion cells. Results are mean± SEM (n ≥ 5).
132
M.J. Traynor et al. / Toxicology and Applied Pharmacology 218 (2007) 128–134
Table 4 Influence of ethoxyethanol concentration on percutaneous absorption of ethoxyethanol (applied as a neat liquid and in aqueous solution) through dermatomed human skin (infinite dose 200 μl/cm2)
Table 5 Percutaneous absorption of water (applied as a neat liquid and as a 50% (v/v) butoxyethanol or ethoxyethanol aqueous solution) though dermatomed human skin (infinite dose 200 μl/cm2)
Concentration of ethoxyethanol (mg/ml)
Maximum absorption rate (mg/cm2/h)
kp (×10− 3 cm/h)
Lag time (h)
Concentration of water (mg/ml)
Maximum absorption rate (mg/cm2/h)
kp (×10− 3 cm/h)
Lag time (h)
930 (neat) 837 (90% v/v) 697 (75% v/v) 465 (50% v/v) 93 (10% v/v) 46.5 (5% v/v) 9.3 (1% v/v) 4.65 (0.5% v/v) 0.93 (0.1% v/v)
0.64 ± 0.09 1.36 ± 0.26 1.87 ± 0.31*** 1.51 ± 0.25* 0.05 ± 0.00 0.03 ± 0.01 0.004 ± 0.00 0.002 ± 0.00 0.004 ± 0.00
0.69 ± 0.09 1.63 ± 0.31 2.68 ± 0.45** 3.28 ± 0.54*** 0.52 ± 0.22 0.57 ± 0.15 0.5 ± 0.08 0.5 ± 0.1 0.5 ± 0.06
0.9 ± 0.05 0.9 ± 0.04 1.1 ± 0.03 0.6 ± 0.04 1.0 ± 0.01 1.0 ± 0.04 0.9 ± 0.06 1.0 ± 0.04 0.9 ± 0.05
1000 neat 500 in butoxyethanol 500 in ethoxyethanol
4.72 ± 0.52 5.17 ± 0.70 5.28 ± 0.40
4.71 ± 0.52 10.34 ± 1.40* 10.56 ± 0.80*
0.84 ± 0.04 0.64 ± 0.05 0.77 ± 0.04
Results are mean ± SEM (n ≥ 5). *P < 0.05, ***P < 0.001 when compared to neat ethoxyethanol.
ethanol through human skin in infinite dose (Fig. 5) was similar to that for neat butoxyethanol (Figs. 1 and 3b). The maximum absorption rate of neat ethoxyethanol through human skin in infinite dose conditions was 0.64 ± 0.09 mg/cm2/h. As with butoxyethanol, this increased when aqueous solutions of ethoxyethanol were applied. A maximum absorption rate of 1.87 ± 0.31 mg/cm2/h which was measured with 75% (v/v) aqueous ethoxyethanol (697 mg/ml) about threefold higher than with neat ethoxyethanol. Concentrations of ethoxyethanol in aqueous solution lower than 10% (v/v) (93 mg/ml) exhibited lower maximum absorption rates than neat ethoxyethanol (Table 4). The apparent kp increased from 0.69 ± 0.09 × 10− 3 cm/h for neat ethoxyethanol to 3.2 ± 0.5 × 10− 3 cm/h from a 465 mg/ml solution (Table 4) but there was, again, no significant effect on the lag time. The proportion of ethoxyethanol recovered in receptor fluid increased from 2.8% with neat ethoxyethanol to 6.2% with 50% (v/v) aqueous ethoxyethanol, but, in contrast to butoxyethanol, the proportion of ethoxyethanol measured in receptor fluid decreased with dose concentrations of 10% (v/v) and lower to a constant proportion of the applied dose that was similar to that for neat ethoxyethanol. As with butoxyethanol, 80–95% of the dose
Fig. 6. Distribution of ethoxyethanol following application as a neat liquid and as water mixtures to dermatomed human skin. Results are expressed as percentage of applied dose in each area and total recovery; mean ± SEM (n ≥ 5).
Results are mean ± SEM (n ≥ 5). *P < 0.05 when compared to neat water.
was recovered from the skin surface, with about 10% recovered in the charcoal traps (Fig. 6). Absorption of water from glycol ether mixtures through human skin in vitro The maximum absorption rate of an infinite dose of water through human skin was 4.72 ± 0.52 mg/cm2/h. This rate was not significantly reduced when water was mixed with butoxyethanol or ethoxyethanol, despite the proportional decrease in the concentration of the water (Table 5). The apparent kp for water was significantly greater in the presence of ethoxyethanol or butoxyethanol, but the increase was not significantly different when a mixture of ethoxyethanol and butoxyethanol was used. There was no significant effect of either solvent on the lag time for water absorption. Discussion In the current study, the highest maximum absorption rate for butoxyethanol measured in human skin (2.34 ± 0.28 mg/cm2/h) was obtained with an infinite dose of a 450 mg/ml aqueous solution (50% v/v solution). This rate was six times greater than the maximum absorption rate of neat butoxyethanol (0.39 ± 0.06 mg/cm2/h). Dilution of butoxyethanol to 90% (v/v) in water resulted in the maximum absorption rate almost doubling compared to neat butoxyethanol, while dilution to 10% (v/v) in water resulted in a maximum absorption rate that was still greater than with neat butoxyethanol. These results clearly indicated that water promoted the absorption of butoxyethanol through human skin in vitro. There was a similar effect of water on the absorption of butoxyethanol from 90% and 50% solutions through rat skin although the proportional increase was not as great as that seen with human skin. Therefore, studies in the rat are of no value in predicting the profile supporting the need for in vitro studies with human skin. Volunteer studies in which 50% butoxyethanol/water was applied to 40 cm2 of the forearm for 4 h and standardised by comparison to an inhalation study (Jakasa et al., 2004) gave measurements of dermal absorption rate of 2.7 mg/cm2/h which was consistent with the in vitro predictions here and the kp increased from 0.88 × 10− 3 to 1.75 × 10− 3 cm/h. For the rat in vivo, a previous measurement of the dermal absorption rate for a 50% butoxyethanol solution was 1.3 mg/cm2/h which was also similar to the in vitro prediction (Payan, personal communication).
M.J. Traynor et al. / Toxicology and Applied Pharmacology 218 (2007) 128–134
In contrast, the absorption of butoxyethanol through polydimethylsiloxane membrane obeyed Fick's first law of diffusion when diluted with water i.e. rate of absorption was proportional to the concentration of the penetrating chemical. The apparent kp was independent of dose concentration, in contrast to the studies with human skin, in which the apparent kp increased with decreased dose concentration. Therefore the membrane was not an ideal surrogate for skin for studying absorption of butoxyethanol/water mixtures. This observation also makes interpretation of the results obtained with skin more difficult, as it indicates that an increase in thermodynamic activity was not solely responsible for the increase in absorption measured for aqueous solutions of glycol ether. Bunge (2006) has defined the influence of mixing butoxyethanol and water on the thermodynamic activity of the two components. The profile indicated three phases: a linear relationship between thermodynamic activity and concentration between 0% and 20% butoxyethanol, between 20% and 70% little change and then a linear increase to 100%. It is reasonable to compare the profile of maximum absorption rate for butoxyethanol and water with concentration in these studies with that of thermodynamic activity of the solutions. The profiles were similar at low butoxyethanol concentrations but they deviated at high butoxyethanol concentrations where dehydration of the skin occurred. The thermodynamic activity of water remained high even at high concentrations of butoxyethanol in parallel with the observed maximum absorption rate. The maximum absorption rate for neat ethoxyethanol through human skin in this study was 0.64 ± 0.09 mg/cm2/h which was similar to the measurement by Filon et al. (1999), who reported a maximum penetration rate of 0.83 ± 0.4 mg/ cm2/h for neat ethoxyethanol when applied to abdominal skin mounted in a Franz cell. The greatest maximum absorption rate for ethoxyethanol was seen from a 50% aqueous solution, which was threefold greater than for neat ethoxyethanol. As with butoxyethanol a substantial increase in maximum absorption rate was observed when 90% (v/v) butoxyethanol in water was used. The apparent kp of ethoxyethanol was also increased in the presence of water up to a maximum with a 50% aqueous solution. The dermal absorption of water from the glycol ether/water mixtures was also increased compared to water alone. When the amount of water present in the dose solution was halved, there was a slight increase in the maximum penetration rate of water in the presence of both butoxyethanol and ethoxyethanol. The apparent kp of the water also increased. Therefore, similar effects of water mixtures were observed for two glycol ethers suggesting that this is a class effect and may have implications for risk assessment of glycol ethers. Wilkinson and Williams (2002) previously showed a higher apparent kp and an increased proportion of the applied dose absorbed of glycol ethers from aqueous solutions compared with neat butoxyethanol, although not an absolute increase in absorption rate, as measured in the present study. However, the concentration of aqueous butoxyethanol studied in the 2002 paper was low, 3 mg/ml, and this may not have been sufficient
133
to observe the increase in absorption rate reported in the present study. It has been reported that the presence of water increased the absorption of butoxyethanol through guinea pig skin in vivo (Johanson and Fernstrom, 1988). These results contribute to our understanding of the mechanism for increased penetration of butoxyethanol and ethoxyethanol into skin in aqueous vehicles. The barrier properties of the skin reside in the stratum corneum and the ordered structure of the lipid lamellae and corneocytes is responsible for low permeability to chemicals. Perturbation of this highly ordered structure may result in an increase in permeability. Effects of either butoxyethanol or ethoxyethanol on the stratum corneum structure and/or hydration of the skin may contribute to the increase in maximum absorption rate measured for glycol ethers in the presence of water. An increase in absorption rate may result from an alteration in the barrier properties of the stratum corneum and/or a greater degree of partitioning of the glycol ethers into the stratum corneum. Glycol ethers are good solvents for lipophilic as well as hydrophilic compounds, and they may have a solubilising effect on the stratum corneum lipids similar to that described for benzyl alcohol, where formation of micelles and disruption of lipid layers is associated with promoting effects (Nanayakker et al., 2005). When dehydrating solvents are applied to the skin mixed with water the disruptive effect on stratum corneum lipid structure and barrier function may increase the permeability to both the solvent and water (Van der Merwe and Riviere, 2005). It has been shown that neat solvents such as ethanol dehydrate the stratum corneum (Marjukka Suhonen et al., 1999; Pillai et al., 2004). Similarly neat glycol ethers may have a dehydrating effect on the skin contributing to less flux, so that with the addition of water and rehydration of the skin an increase in absorption could occur. Wilkinson and Williams (2002) suggested that there was increased partitioning of butoxyethanol from the aqueous vehicle into the skin. At certain concentrations, butoxyethanol in water does not behave as a perfect solution but the butoxyethanol molecules cluster together in pseudomicelles (Castillo and Dominguez, 1990) and, if these were in contact with skin, they might preferentially partition out of the water and into the lipid-rich stratum corneum. In this study there was no depletion of the concentration of the test dose in contact with the skin with time, but as absorption of water also increased, changes in concentration might not have occurred. Also it is known that the state of hydration of the skin will affect penetration. With increased hydration within a few hours, there is evidence of accumulation of water in the intercellular spaces disrupting the lamellar arrangement of the lipids (Warner et al., 2003). This can lead to increased penetration of water-soluble chemicals such as glycol ethers (Tezel et al., 2003). Application of water to the skin for 1 or 3 h before butoxyethanol increased the maximum absorption rate by three fold (results not shown), but this increase was less than that measured with butoxyethanol/water mixtures, suggesting that the hydration by water is only partly responsible. Also the absolute absorption rate of water was greater than that of butoxyethanol. Water penetration
134
M.J. Traynor et al. / Toxicology and Applied Pharmacology 218 (2007) 128–134
in butoxyethanol/water mixtures might enhance butoxyethanol penetration by solvent drag. In conclusion, this study has found that the dermal absorption of butoxyethanol and ethoxyethanol is increased in the presence of water. Mechanisms for this phenomenon may involve disruption of lipid bilayer by glycol ether and dehydration of the skin by neat solvent, which contribute to reduced flux with neat glycol ethers. Hydration following application of water mixtures resulted in increased flux. This information is important for risk assessment of exposure to glycol ethers which have industrial and household uses in aqueous solution. Some predictive models based on quantitative structure activity relationships (QSARs) are used by risk assessors to predict kp values for substances for which experimental data are not available (Fitzpatrick et al., 2004). The experimental datasets on which these models are based consist of absorption studies employing mainly aqueous vehicles in “infinite” dose. This approach should not be applied to any solvent, which is freely soluble in water, as the influence of vehicle or formulation or finite dose is not considered. This further emphasises the need to determine absorption rates using in vitro approaches with human skin and a relevant exposure. Risk assessments for dermal contact scenarios should ideally be based on absorption data obtained for the relevant mixture and particular attention focussed on abnormal absorption from aqueous solutions. It is important to understand the role of skin interactions in deviation of dermal absorption from predictions and investigation of absorption in disease states may help. Acknowledgment This work was funded by the Health and Safety Executive Magdalen House, Stanley Precinct, Bootle, Merseyside, L20 3QZ, UK. References Bartnik, F.G., Reddy, A.K., Klecak, G., Zimmermann, V., Hostynek, J.J., Kunstler, K., 1987. Percutaneous absorption, metabolism, and hemolytic activity of n-butoxyethanol. Fundam. Appl. Toxicol. 8 (1), 59–70. Bunge, A.L., 2006. Why skin permeation data from neat and aqueous solutions of butoxyethanol is not surprising. Perspectives of Percutaneous Penetration, vol. 10A, p. 24. Castillo, R.C., Dominguez, H.C., 1990. Determination of mutual diffusion coefficients in water-rich 2-butoxyethanol/water mixtures using the Taylor dispersion technique. J. Phys. Chem. 94, 8731–8734. Chilcott, R.P., Barai, N., Beezer, A.E., Brain, S.I., Brown, M.B., Bunge, A.L., Burgess, S.E., Cross, S., Dalton, C.H., Dias, M., Farinha, A., Finnin, B.C., Gallagher, S.J., Green, D.M., Gunt, H., Gwyther, R.L., Heard, C.M., Jarvis, C.A., Kamiyama, F., Kasting, G.B., Ley, E.E., Lim, S.T., McNaughton, G.S., Morris, A., Nazemi, M.H., Pellett, M.A., Du Plessis, J., Quan, Y.S., Raghavan, S.L., Roberts, M., Romonchuk, W., Roper, C.S., Schenk, D., Simonsen, L., Simpson, A., Traversa, B.D., Trottet, L., Watkinson, A., Wilkinson, S.C., Williams, F.M., Yamamoto, A., Hadgraft, J., 2005. Interand intralaboratory variation of in vitro diffusion cell measurements: an
international multicenter study using quasi-standardized methods and materials. J. Pharm. Sci. 94 (3), 632–638. Filon, F.L., Fiorito, A., Adami, G., Barbieri, P., Coceani, N., Bussani, R., Reisenhofer, E., 1999. Skin absorption in vitro of glycol ethers. Int. Arch. Occup. Environ. Health 72, 480–484. Fitzpatrick, D., Corish, J., Hayes, B., 2004. Modelling skin permeability in risk assessment—The future. Chemosphere 55 (10), 1309–1314. Ghanayem, B.I., Burka, L.T., Sanders, J.M., Matthews, H., 1987. Metabolism and disposition of ethylene glycol monobutyl ether (2-butoxyethanol) in rats. Drug Metab. Dispos. 15 (4), 478–484. Ghanayem, B.I., Burka, L.T., Matthews, H., 1989. Structure–activity relationships for the in vitro hematotoxicity of N-alkoxyacetic acids, the toxic metabolites of glycol ethers. Chem.-Biol. Int. 70 (3–4), 339–352. Hardin, B.D., 1983. Reproductive toxicity of the glycol ethers. Toxicology 26, 91–102. Hardin, B.D., Goad, P.T., Burg, J.R., 1984. Developmental toxicity of four glycol ethers applied cutaneously to rats. Environ. Health Perspect. 57, 69–74. Jakasa, I., Mohammadi, N., Kruse, J., Kezic, S., 2004. Percutaneous absorption of neat and aqueous solutions of 2-butoxyethanol in volunteers. Int. Arch. Occup. Environ. Health 77 (2), 79–84. Johanson, G., 1991. Percutaneous absorption of 2-butoxyethanol vapour in human subjects. Br. J. Ind. Med. 48 (11), 788–792. Johanson, G., Boman, A., Dynesius, B., 1988. Percutaneous absorption of 2butoxyethanol in man. Scand. J. Work, Environ. Health 14 (2), 101–109. Johanson, G., Fernstrom, P., 1988. Influence of water on the percutaneous absorption of 2-butoxyethanol in guinea pigs. Scand. J. Work, Environ. Health 14 (2), 95–100. Kao, J., Carver, M.P., 1990. Cutaneous metabolism of xenobiotics. Drug Metab. Rev. 22 (4), 363–410. Kezic, S., Mahieu, K., Monster, A.C., de Wolff, F.A., 1997. Dermal absorption of vaporous and liquid 2-methoxyethanol and 2-ethoxyethanol in volunteers. Occup. Environ. Med. 54 (1), 38–43. Lockley, D.J., Howes, D., Williams, F.M., 2002. Percutaneous penetration and metabolism of 2-ethoxyethanol. Toxicol. Appl. Pharmacol. 180 (2), 74–82. Lockley, D.J., Howes, D., Williams, F.M., 2004a. Percutaneous penetration and metabolism of 2-butoxyethanol. Arch. Toxicol. 78 (11), 617–628. Lockley, D.J., Howes, D., Williams, F.M., 2004b. Metabolism of glycol ethers. Arch. Toxicol. 78 (11), 617–628. Marjukka Suhonen, T., Bouwstra, J.A., Urtti, A., 1999. Chemical enhancement of percutaneous absorption in relation to stratum corneum structural alterations. J. Con. Release 59 (2), 149–161. Multinger, L., Catala, M., Cordier, S., Delaforge, M., Fenaux, P., Garnier, R., Rico-Lattes, I., Vasseur, P., 2005. The INSEM expert review on glycol ethers: findings and recommendations. Toxicol. Lett. 156, 29–37. Nanayakker, G.R., Bartlett, A., Forbes, B., Marriott, C., Whitfield, P.J., Brown, M.B., 2005. The effect of unsaturated fatty acids in benzyl alcohol on the permeation of model penetrants. Int. J. Pharm. 301, 129–139. Pillai, O., Nair, V., Panchagnula, R., 2004. Transdermal iontophoresis of insulin: IV. Influence of chemical enhancers. Int. J. Pharm. 269 (1), 109–120. Payan J.P., personal communication. Tezel, A., Sens, A., Mitragotri, S., 2003. Description of transdermal transport of hydrophilic solutes during low-frequency sonophoresis based on a modified porous pathway model. J. Pharm. Sci. 92 (2), 381–393. Van der Merwe, D., Riviere, J.E., 2005. Comparative studies on the effects of water, ethanol and water/ethanol mixtures on chemical partitioning into porcine stratum corneum and silastic membrane. Toxicol. in vitro 19 (1), 69–77. Wilkinson, S.C., Williams, F.M., 2002. Effects of experimental conditions on absorption of glycol ethers through human skin in vitro. Int. Arch. Occup. Environ. Health 75 (8), 519–527. Warner, R.R, Stone, K.J., Boissy, Y.I., 2003. Hydration disrupts human stratum corneum ultrastructure. J. Invest. Dermatol. 130, 275–284.