Fd Chem. Toxic. Vol. 30, No. 2, pp. 145 153, 1992
0278-6915/92 $5.00 + 0.00 Copyright ~50 1992 Pergamon Press plc
Printed in Great Britain. All rights reserved
P E R C U T A N E O U S ABSORPTION OF BENZYL ACETATE T H R O U G H RAT SKIN I N VITRO. 2. EFFECT OF VEHICLE A N D OCCLUSION S. A. M. HOTCHKISS, J. M. MILLER a n d J. CALDWELL Department of Pharmacology and Toxicology, St Mary's Hospital Medical School, Imperial College of Science Technology and Medicine, London W2 IPG, UK (Accepted 15 November 1991)
Abstract--The effect of vehicle and occlusion on the/n vitro percutaneous absorption of [methylene-~4C] benzyl acetate (1.7-16.6mg/cm 2) has been studied in diffusion cells using full thickness skin from male Fischer 344 rats. Absorption of neat benzyl acetate through rat skin occluded with parafilm was 49.3 _+ 2.0% (mean + SD, n = 4) after 48 hr. When benzyl acetate in ethanol was applied to the skin and the skin was occluded with parafilm, the extent of absorption at 48 hr was not significantly different from that after neat application. However at 6 hr, as the ethanol content of the application mixture was increased, the absorption of benzyl acetate through occluded skin was enhanced proportionally (r = 0.999). When phenylethanol was used as a vehicle, the extent of the benzyl acetate absorption through occluded skin at 48 hr was enhanced (P < 0.05) compared with that after application neat; with 50% (v/v) phenylethanol, absorption at 48 hr was 56.3 + 4.9%. However, this enhanced absorption did not correlate with the proportion of phenylethanol in the application mixture. When dimethylsulphoxide was used as a vehicle, the extent of absorption of benzyl acetate through occluded skin at 48 hr was enhanced (P < 0.05) compared with that after application neat (absorption was 59.3 + 3.7% of the applied dose when 50% (v/v) dimethylsulphoxide was used). As the dimethylsulphoxide content of the application mixture was increased, the absorption of benzyl acetate was enhanced proportionally. Occlusion of the skin surface with parafilm often significantly enhanced absorption (P < 0.05), although the effect varied with time and vehicle. In general, the degree of any enhanced absorption caused by the use of a vehicle or occlusion of the skin was small, and, in most cases, would be unlikely to be toxicologically significant.
INTRODUCTION The a w l alkyl ester, benzyl acetate, is a naturally occurring fragrance and flavouring agent found in the oils o f j a s m i n , hyacinth a n d gardenia a n d used in the m a n u f a c t u r e of perfumes, household chemicals and foodstuffs. Because o f the widespread h u m a n exposure to benzyl acetate, mainly t h r o u g h the skin, there is currently considerable interest in its safety evaluation. In a 2-yr carcinogenicity bioassay, benzyl acetate was reported to cause pancreatic acinar a d e n o m a s in rats after a d m i n i s t r a t i o n by gavage in corn oil (National Toxicology Program, 1986), a l t h o u g h the t u m o u r s have been ascribed to the corn oil vehicle since they were also present in vehicle-fed control animals (Longnecker et al., 1986). In further studies, dietary benzyl acetate was observed to act as a weak p r o m o t e r (but not initiator) o f pancreatic pre-neoplastic loci in the rat (Longnecker et al., 1990). In terms of local toxicity, benzyl acetate is not irritant to rabbit skin (Research Institute for Fragrance Materials, personal c o m m u n i c a t i o n , 1991), a n d is not a skin-sensitizing agent in h u m a n s (Greif, 1967). However, there is a lack o f data on
Abbreviation: DMSO = dimethylsulphoxide.
systemic toxicity after dermal application, a n d there is d o u b t a b o u t the relevance of data from rodent studies using high oral doses to the effects of exposure of h u m a n s to low doses via the dermal route. Since there is significant h u m a n exposure to benzyl acetate t h r o u g h the skin, it is i m p o r t a n t to investigate the rate a n d extent of percutaneous a b s o r p t i o n of the chemical, and the factors that may influence this absorption, in order to o b t a i n i n f o r m a t i o n that will contribute towards the safer use of the chemical in humans. The skin a b s o r p t i o n of benzyl acetate has been studied in the rat, b o t h in vitro (Hotchkiss et al., 1990) and in t,ivo (Chidgey et al., 1987). However, the effect of vehicle a n d occlusion of the skin surface u p o n the a b s o r p t i o n of the c o m p o u n d has not been systematically examined. Various vehicles are currently in use for the formulation of products that come into contact with skin and for the study of percutaneous a b s o r p t i o n and dermal toxicity. In the light of data from m a n y workers (Barry, 1987 and 1989; G u y et al., 1990; Idson, 1983) regarding the penetration e n h a n c i n g effect of such vehicles, it is i m p o r t a n t to determine the effect that these formulation c o m p o n e n t s may have upon the a b s o r p t i o n kinetics of topically applied compounds such as benzyl acetate. Ethanol and phenylethanol are c o m m o n l y used as m a j o r ingredients (up to 90%, v/v) in the 145
146
S.A.M. HOTCHKISSet al.
composition of fragrance products, whilst dimethylsulphoxide (DMSO) is widely used in dermatological studies as a solvent and is the archetypal penetration enhancer. Such chemicals may not be inert vehicles, but may act to enhance percutaneous absorption by a combination of mechanisms including fluidization of stratum corneum lipids, solvent effects, hydration or lipid extraction or by other means of barrier perturbation. Under conditions of dermal exposure to fragrances, drugs and other chemicals, the site of application may be left open to the atmosphere, or it may be occluded by clothing or dressings. Occlusion is known to enhance the percutaneous absorption of many, but not all. topically applied chemicals (Bronaugh et al., 1985 and 1990: Bucks et al., 1989; Feldmann and Maibach, 1965; Wester and Maibach, 1983), and therefore it is important to investigate its effect on the extent of percutaneous penetration of chemicals such as benzyl acetate. The mechanism for this enhancement is complex and not completely understood, but may involve a combination of factors including the reduction of epidermal water loss (which results in an increase in skin hydration) and, in some cases, an increase in skin temperature. In addition, occlusion may reduce the surface loss of compounds by evaporation and abrasion. We have previously established an in vitro model for percutaneous absorption which may be used to predict the in rit,o absorption of benzyl acetate in the rat (Hotchkiss et al., 1990). We now report on the use of this technique to study the effect of two variables, the nature of the vehicle of application and occlusion of the application site, on the percutaneous absorption of benzyl acetate though rat skin in ritro.
MATERIALS AND M E T H O D S
Compounds. [methylene-~4C]Benzyl acetate (sp. act. 53 mCi/mmol; radiochemical purity > 96%) was a custom synthesis from Amersham International plc (Amersham, Bucks., UK). Unlabelled benzyl acetate and DMSO were obtained from Aldrich Chemical Co. (Gillingham, Dorset, UK). Ethanol was purchased from James Burrough Ltd (Witham, Essex, UK) and 2-phenylethanol was obtained from Sigma Chemical Co. (Poole, Dorset, UK). All other chemicals were obtained from standard suppliers of laboratory chemicals and were of analytical grade. h7 vitro diffusion system. Percutaneous absorption was assessed in diffusion cells using full-thickness skin from male Fischer 344 rats, as previously described (Hotchkiss et al., 1990). The system, based on that of Bronaugh and Stewart (1985), consisted of seven teflon diffusion cells and a fraction collector (Crown Glass Co. lnc, Somerville, N J, USA), a thermostatically controlled Churchill water circulator and
a model 202V/AA16 cassette peristaltic pump (Watson-Marlow, Falmouth, Cornwall, UK). Skin preparation. Male Fischer 344 rats, weighing approximately 200 g, were obtained from Harlem Olac (Oxford, Oxon., UK), and were maintained on a diet of Biosure (Manea, Cambs., UK) CRM pellets and water ad lib. The rats were killed by cervical dislocation following light ether anaesthesia, and then the dorsal region was shaved with animal clippers and the shaved area was cut out with dissecting scissors. The skin was placed, epidermal side uppermost, on a plastic dissecting board and circles of full-thickness skin, 1.7 cm in diameter, were cut out using a circular sharpened steel cutter. Excess subcutaneous tissue was removed with a scalpel. The skin was placed in diffusion cells in the flow-through apparatus, and allowed to equilibrate for about 30 min. The skin surface temperature was maintained at 32C, and the underside of the skin was in contact with a continuously flowing receptor fluid of 0.9% saline (degassed before use by filtration through a 0.5 Itm filter) at a flow rate of 1.5 ml/hr. Dose application and determination of absorption. Aliquots of 5/~1 [methylene-~4C]benzyl acetate (1.7 16.6 mg/cm 2) were applied to the exposed skin surface (0.32 cm2) using a microsyringe fitted with a bluntended needle. The compound was applied either neat or in a vehicle of ethanol, phenylethanol or DMSO, so that the vehicle content varied from 0 to 90% (v ~v). The skin was then either occluded with a layer of standard self-sealing Nescofilm (Solmedia, Romford, Essex, UK), 1.3 cm above the skin surface, or left open to the atmosphere. In some experiments, teflon caps (Crown Glass, Somerville, N J, USA), positioned 2.9 cm above the skin surface, were used to occlude the skin instead of the layer of Nescofilm. The receptor fluid (1.5 ml) was collected hourly into 20-ml scintillation vials for up to 72hr and then 10ml scintillation fluid (Ecoscint, National Diagnostics. Manville, N J, USA) was added and levels of ~(" were determined by liquid scintillation spectrometry (Packard Tri-Carb Model 4640). Determination o[residual benzyl acetate. At the end of the experiment the skin was washed by swabbing twice with lint soaked in ethanol, and the washings were counted to determine radioactivity remaining on the skin surface. Radioactivity remaining within the skin was determined by digesting the skin circles in 1 ml of a solution of 20 g sodium hydroxide in 150 ml water, 75 ml methanol and 25 ml Triton 405, in a shaking water-bath for 1 hr at 70C. After incubation, the mixture was cooled and 0.5 ml 4.4 m-nitric acid was added, followed by 10 ml scintillant. Radioactivity was counted by liquid scintillation spectrometry, using a quench curve obtained with untreated skin spiked with a known quantity of radioactivity. Statistical evaluation. Where appropriate, results were compared using Student's t-test for unpaired data. The level of significance was taken as P < 0.05.
Percutaneous absorption of benzyl acetate. 2
147
Table 1. Effect of vehicle on the percutaneous absorption of benzyl acetate though occluded (parafilm) rat skin in vitro Vehicle content (v/v)
Vehicle None Ethanol
Phenylethanol
Dimethylsulphoxide
10% 20% 50% 80% 90% 10% 20% 50% 80% 90% 10% 20% 50% 80% 90%
Absorption (% applied dose)¢ at: 6 hr 4.6 ± 0.6 5.6 ± 0.9 7.1 + 1.5' 10.8 ± 2.7* 14.5+_4.0" 15.4 + 4.9* 6.0 ± 0.6* 7.4 + 0.6* 6.6 + 0.2* 7.7 +- 0.2* 5.8 ± 0.8* 6.9 + 1.6" 10.0+ 1.1" 11.8 ± 3.5* 23.1 +_6.4* 31.1 ± 5.8*
24 hr
48 hr
35.3 ± 3.4 35.0 +- 3.8 33.4±2.7 38.9 ± 1.6" 41.1 +_4.2* 39.8 ± 2.5* 36.5 ± 1.9 39.5 + 0.9* 38.1 ± 3.9 42.1 ± 1.4" 37.6 ± 3.2 39.0 ± 1.6" 43.2+_3.8* 44.7 +_4.5* 50.4 ± 6.9* 55.9 ± 5.5*
49.3 ± 2.0 51.9 ± 4.2 47.1 ±4.4 50.0 ± 1.0 51.5±2.6 52.2 + 5.3 54.4 ± 2.3* 57.8 ± 1.5" 56.3 + 4.9* 59.2 ± 2.5* 53.2 ± 4.8* 55.9 + 1.1" 57.0±2.8* 59.3 ± 3.7* 63.8 __.5.7* 63.6 ± 4.6*
"tValues are means + SD for four diffusion cells run concurrently, and those marked with an asterisk differ significantly (Student's t-test) from the corresponding control (no vehicle) values (*P < 0.05).
RESULTS
Percutaneous absorption o f benzyl acetate applied in ethanol I n a g r e e m e n t with o u r p r e v i o u s l y p u b l i s h e d results ( H o t c h k i s s et al., 1990), the a b s o r p t i o n o f neat benzyl acetate t h r o u g h rat skin u n d e r o c c l u s i o n w i t h p a r a -
film in vitro was extensive, r e a c h i n g 4 9 . 3 % o f the applied dose at 48 hr ( T a b l e 1; Fig 1). W h e n benzyl acetate w a s applied to the skin in e t h a n o l , at vehicle c o n t e n t s r a n g i n g f r o m 10 to 9 0 % (v/v), the extent o f a b s o r p t i o n t h r o u g h occluded skin at 48 h r w a s n o t significantly different f r o m t h a t after a p p l i c a t i o n neat. A b s o r p t i o n w a s 50.0% w h e n 50% (v/v) e t h a n o l
60
50
e~ 0
~"
40
m
30
o
20
N~ II1
10
0
~ 0
8
16
24
32
40
48
Time (hr) Fig. l. Effect of ethanol vehicle and parafilm occlusion on the absorption of benzyl acetate through rat skin in vitro after application neat (C)--unoccluded, O---occluded) or in 50% (v/v) ethanol (V]--unoccluded, I---occluded). Values are means + SD, n = 3 (unoccluded), n = 4 (occluded).
S. A. M, HOTCHKISS et al.
148
60
*
*
*
*
== 50
40
f,.}
o.
30
g
20
e- v 10
I
0
I
20
I
40
Vehicle
I
60
I
80
concentration
(%
100
v/v)
Fig. 2. Relationship between vehicle concentration and extent of benzyl acetate absorption through occluded rat skin in vitro at 6 hr (open symbols) and 48 hr (closed symbols) after application in ethanol ([Z, I ) , phenylethanol ((3, I ) or dimethylsulphoxide (z~, at). Values are means (n = 4). Error bars are omitted for clarity. Values marked with an asterisk differ significantly (Student's /-test) from the corresponding value for application neat (*P < 0.05). 70
60
"a
50
e-,
40
"el
Ila (lJ
*" o
30
¢1 O
~g
20
e- "--" m 10
0
~
0
8
16
24 Time
32
40
48
(hr)
Fig. 3, Effect of phenylethanol vehicle and parafilm occlusion on the absorption of benzyl acetate through rat skin in ritro after application neat ( © - - u n o c c l u d e d , O--occluded) or in 50% (v/v) phenylethanol ([~ --unoccluded, I - - o c c l u d e d ) . Values are means + SD, n = 3 (unoccluded), n = 4 (occluded).
Percutaneous absorption of benzyl acetate. 2
149
70
60
50
o: A o
¢~ '1~
40
"o
::t 0 o
20 m
10
0
~ 0
8
16
24
32
40
48
Time (hr) Fig. 4. Effect of dirnethylsulphoxide vehicle and parafilm occlusion on the absorption of benzyl acetate through rat skin in vitro after application neat (O--unoccluded, O--occluded) or in 50% (v/v) dimethylsulphoxide ([]--unoccluded, I---occluded). Values are means + SD, n = 3 (unoccluded), n = 4 (occluded). was used (Table l; Fig. 1). However, at 6 h r the absorption of benzyl acetate was increased by increasing the ethanol content of the application mixture, rising from 4.6% at 0% vehicle to 15.4% at 90% (v/v) vehicle (r = 0.999; Table 1; Fig. 2).
Percutaneous absorption of benzyl acetate applied in phenylethanol When benzyl acetate was applied to the skin in phenylethanol, at vehicle contents ranging from 10 to 90% (v/v), the extent of absorption through parafilmoccluded skin at 48 hr was statistically significantly enhanced compared with that after application neat. When 50% (v/v) phenylethanol was used absorption was 56.3% of the applied dose (Table 1; Fig. 3). However, at no time after application did this enhanced absorption correlate with the proportion of phenylethanol in the application mixture (r = 0.171 at 6hr; r = 0 . 1 7 6 at 48hr; Table 1; Fig. 2).
Percutaneous absorption of benzyl acetate applied in DMSO When benzyl acetate was applied to the skin in DMSO, at vehicle contents ranging from 10 to 90% (v/v), the extent of absorption through occluded skin at 48 hr was statistically significantly enhanced corn-
pared with that after application neat. Absorption was 59.3% when 50% (v/v) DMSO was used (Table 1; Fig. 4). As the DMSO content of the application mixture was increased over the range 0 to 90% (v/v), the absorption of benzyl acetate at 6 hr rose from 4.6% at 0% vehicle to 31.1% at 90% (v/v) vehicle (r = 0.917). Similarly, the absorption at 48 hr also increased proportionally, from 49.3% at 0% vehicle to 63.6% at 90% (v/v) vehicle (r = 0.871; Table I; Fig. 2). At virtually all time points, DMSO enhanced the absorption of benzyl acetate to a statistically significantly greater extent than either ethanol or phenylethanol. The maximum penetration enhancement effect for all vehicles was observed with 90% (v/v) DMSO at 6 hr, where absorption was enhanced by 26.5% of the applied dose.
Effect of parafilm occlusion absorption
on benz)'l acetate
The absorption of neat benzyl acetate at 48 hr was not significantly affected by occlusion with parafilm (Table 2). However, when benzyl acetate was applied to the skin in a vehicle of ethanol, phenylethanol or DMSO occlusion usually resulted in a statistically significant increase in the absorption of benzyl acetate at 48 hr, although this difference was not always
S. A. M. HOTCHKISS et al.
150
Table 2. Effect of parafilm occlusion on the percutaneous absorption of benzyl acetate though rat skin m t itr, Absorption (% applied dose)t at: Vehicle content (v/v)
Vehicle None Ethanol
Phenylethanol
Dimethylsulphoxide
. . Occluded
6 hr . . .
4.6 + 0.6 5.6 + 0.9 7.1 -: I 5 10.8 ! 2.7 14.5±4.0 15.4 + 4.9 6.0 + 0.6 7.4 + 0.6 6.6 + 0.2 7.7 + 0.2 5,8 + 0 . 8 6 . 9 + 1.6 10.0 -+ h i 11.8 ~ 3.5 23.1 ~- 6.4 31.1 j_5.8
10% 20% 50% 80% 90% 10% 20% 50% 8(/% 90% 10% 20% 50% 80% 90%
48 hr . . . . . . Unoccluded 4.2 + 0.5 4. L ± 1.3" 7.1 + 2.1 10A + 1.5 10.5±1.3" 9,7 ± 0.7" 5,3 ± 0.7* 6.8 ± 0.9 5.9 ± 1.9 6.4 +- 0.1" 5.1 + 0 . 9 6.2-+2.1 8.2 ± 0.8* 9.9 Z 5,1 23.0 ± 5.8 27.5±3.7
. Occluded 49.3 ± 2.0 51,9 ± 4.2 47.1 ± 4.4 50.0 ± h0 51.5±2.6 52.2 ~ 5.3 54.4 -+ 2.3 57.8 -+ 1.5 56.3 i 4.9 59.2 ± 2.5 53,2+_4,8 5 5 . 9 2 l.I 57.0 ± 2.8 59.3 ± 3.7 63.8 +- 5.7 63.6±4.6
Unoccludcd 4 7 9 4 19 4 5 5 4 4 8" 45. ~, ~ I ,~ 44.7 + ~ 6 ' 41 I ~ 3.8" 4 ~ 6 ~ ~;1' 50.9 ~ I.~* 526 ~ 25* 5(k8 :? t~.4 ~0.3 - 0 9* 48.5 ~ 49* 491 ~ 4 5 " 4~).I f I 2* 48 6 ,- ~.4" 530 ~; 2.9* 52.6 ~ 3.(1"
*Values are means ± SD for four (occluded) or three (unoccluded) diffusion cells run concurrently. Thosc marked with an asterisk differ significantly(Student's t-test) from the corresponding occluded values (*P < 0.0s)
a p p a r e n t at 6 hr. At 6 h r , the m a x i m u m effect of occlusion was observed with 9 0 % (v/v) ethanol, where occlusion e n h a n c e d the a b s o r p t i o n by 5.7% of the applied dose. At 48 hr, the m a x i m u m effect of occlusion was observed with 90% (v/v) D M S O , where the a b s o r p t i o n of benzyl acetate was e n h a n c e d by 11.0% of the applied dose (Table 2). Effect o J teflon absorption
cap
occlusion
on
benzy[
acetate
Limited data were o b t a i n e d with teflon caps rather t h a n parafilm as the occlusive devise. As with parafilm, occlusion of the skin site with teflon caps after the application of neat benzyl acetate did not e n h a n c e the extent of a b s o r p t i o n o f the chemical t h r o u g h rat skin. There was no significant difference between the a b s o r p t i o n of neat benzyl acetate at 48 hr t h r o u g h skin occluded with parafilm and that t h r o u g h skin occluded with teflon caps (Table 3).
Table 3, Effect of occlusion by parafilm or teflon caps on the recovery of radioactivity after topical application of neat benzyl acetate to rat
Site Receptor fluid (0 48hr) Receptor fluid (0-72 hr) Skin surface Skin Diffusion cell Parafilm/cap
Total
skin in vitro Recovery ofradioactivity(%applieddose)~, ........... Occlud-~d- . . . . Unoccluded P a r a f i l m Tefloncap 47.0+3.1
47.2+4.1
49.7+_2.5
52.4±3.4 0.4 + 0.1 11.7 + 5.0 0.6 ± 0.3 --
54.7_+4.3 0.3 ~ 0.1 11.2 z 3.6 0.7 +_ 0.5 0.2 i 0.2
54.5+3.0 0.4 ± 0,1 11.2 +_ 3.3 0.9 ± 0.5 26.1 ± 4.6*
64.8+_5.7
68.4±3.9
93,0±3.6"t
~Values are means ± SD for four diffusion cells run concurrently. Values marked with an asterisk differ significantly (Student's t-test) from the corresponding values without occlusion (*P < 0.05), and those marked with a dagger differ significantly (Student's t-test) from the corresponding values with parafilm occlusion (tP < 0.05).
Balance ~t" radioactivity
At 72 hr after the application of benzyl acetate, the recovery of radioactivity in the skin was a r o u n d 1 0 - 1 5 % of the applied dose, and was not related to the vehicle of application or occlusiom In all cases, less t h a n 1% was recovered from the skin surface by washing with ethanol, and less than I % was recovered on the parafilm that was used to occlude the skin surface. The total recovery of radioactivity u n d e r the occluded conditions described was approximately 70%. These data are in accordance with our previous results (Hotchkiss et al., 1990). In studies with neat benzyl acetate, the recovery of radioactivity from the skin surface and from within the tissue was not significantly different following occlusion with parafilm or teflon caps, a l t h o u g h the total recovery of radioactivity was significantly increased after occlusion with the caps, reaching more than 90% of the applied dose (Table 3).
DISCUSSION
The percutaneous absorption of benzyl acetate has been studied previously in the rat, both in cit'o (Chidgey et al., 1987) and in Ntro (Hotchkiss et al., 1990), and a b o u t 50% of the applied dose was a b s o r b e d after 48 hr. After topical application to rat skin, the extent o f a b s o r p t i o n in citrt, was found to correlate well with that observed m Nco. hence validating the in vitro system as a model for the study of factors affecting benzyl acetate a b s o r p t i o n (Hotchkiss et al., 1990). F u r t h e r in ciro and m citro studies showed that the application of benzyl acetate to rat skin in a vehicle of 50% ethanol did not affect the total a m o u n t o f c o m p o u n d a b s o r b e d at 48 hr (Chidgey et al., 1987; Hotchkiss et al., 1990). However, a wide variety o f vehicles, at various concentrations, are used in the formulation of products that
Percutaneous absorption of benzyl acetate. 2 come into contact with skin and for the study of percutaneous absorption and dermal toxicity, and it is known that certain vehicles have penetrationmodulating effects (Barry, 1987 and 1989; Grasso and Lansdown, 1972; Guy et al., 1990; Idson, 1983). Therefore we undertook a more detailed and systematic investigation of the effect of such vehicles on the percutaneous absorption of the model fragrance chemical, benzyl acetate. For the absorption studies described in this paper, only total radioactivity has been reported, and the analysis of specific metabolites has not been included. Studies previously conducted in our laboratory (Chidgey et al., 1987) have shown that benzyl acetate, administered topically to rats in ~'it,o, undergoes rapid and extensive hydrolysis to benzyl alcohol, followed by oxidation to benzoic acid and then conjugation with glycine to form hippuric acid. A small amount of benzoic acid is conjugated with glucuronic acid to form benzoyl glucuronide, and some benzylmercapturic acid is also formed. The skin itself is known to possess substantial esterase activity and we have preliminary data that suggest that the first step in this detoxication pathway (hydrolysis) occurs almost exclusively in the skin during percutaneous absorption (S.A.M. Hotchkiss, unpublished observations, 1991). It is therefore likely that the majority of the radiolabelled compound detected after absorption is benzyl alcohol, and not the parent ester, benzyl acetate. The use of ethanol, phenylethanol or DMSO as vehicles for the topical application of benzyl acetate generally resulted in a statistically significant enhancement of penetration, although the biological relevance of the relatively small increases observed (up to 27% of the applied dose, hut usually less than 10%) is uncertain. From the viewpoint of safety evaluation, if these differences were observed to occur in L,ivo in the rat. they could be either toxicologically insignificant or significant, depending upon the study conditions. The effect is more likely to be of importance in long-term studies with repeated dermal applications at relatively short intervals, where the use of a vehicle that enhances bioavailability (even by 10%) may result in enhanced exposure and possibly toxicity. However, in order to extrapolate these data from rats to the effects of exposure in humans, information on the percutaneous absorption of benzyl acetate through human skin is required. Such studies are currently under way in our laboratory, and preliminary evidence suggests that the absorption through human skin (about 12%) is significantly less than through rat skin (about 50%) (S.A.M. Hotchkiss, unpublished observations, 1991). The relatively poor permeability of human skin compared with that of other species is in broad agreement with reports in the literature by other workers using a variety of compounds (Bartek et al., 1972). The relatively small increases in absorption caused by the vehicles used in our studies are difficult to reconcile with the large (up to 50-fold) increases often
151
observed by other workers. However, in the studies described in our paper, a finite dose of benzyl acetate was selected (on the basis of a more accurate representation of true conditions of exposure in vivo), which was rapidly and extensively absorbed, even in the absence of any vehicle. The rapid disappearance of this source of benzyl acetate from the skin surface into the receptor fluid (and into the air by evaporation) would make it impossible for any vehicle to enhance absorption by the magnitude observed by other workers. It should be noted that the enhanced absorption afforded by these vehicles was obtained despite the possible decrease in thermodynamic activity of benzyl acetate caused by its dilution with the vehicle. In order for a topically applied chemical to be absorbed into the systemic circulation, it must first traverse the stratum corneum (the major skin barrier), the viable epidermis and the portion of the upper dermis which contains the cutaneous microcirculation. Passage through the densely packed keratinocyte layer which constitutes the stratum corneum may be by transcellular or intercellular routes, or, less significantly, through sweat glands or hair follicles. There is evidence to suggest that for lipophilic molecules such as benzyl acetate (log P = 1.96), absorption through the stratum corneum may be through either the intercellular or intracellular lipid regions of the keratinocytes (Barry, 1987; Scheuplein, 1972). However, as the volume of intracellular lipid is relatively small compared with the intercellular volume, the latter may be the dominant route for dermal absorption (Elias, 1981). It has been postulated that polar penetration enhancers such as ethanol and phenylethanol, and aprotic solvents such as DMSO, may interact with the polar head groups of the intercellular lipid bilayer to modify hydrogen and ionic bonds and so disturb lipid packing. This may result in increased fluidity of the lipid bilayer, hence enhancing the movement of lipophilic molecules. In addition, solvents such as ethanol, phenylethanol and DMSO may penetrate in significant quantities into the intercellular aqueous layer, thus enhancing the solubility of lipophilic chemicals in this region (Barry, 1987). However, studies by Guy et al. (1990) suggest that ethanol acts as a penetration enhancer in human skin, not by ftuidization of stratum corneum intercellular lipids, but by lipid extraction. At high concentrations, these solvents may actually damage the skin, resulting in a loss of barrier function. Other mechanisms for the mode of action of DMSO have been postulated, including denaturation of structural keratin (Scheuplein and Ross, 1970), displacement of bound water from proteins (Scheuplein and Blank, 1971) and elution of stratum corneum lipids (Embery and Dugard, 1971). After topical application of benzyl acetate in either ethanol, phenylethanol or DMSO, the fate of the vehicle may not necessarily be the same as the chemical under investigation, although in order for a vehicle to affect the skin barrier it must presumably
152
S.A.M. HO'rCHK1SSet al.
partition into the tissue to some extent. Even under conditions of occlusion, it is likely that the alcohols, ethanol and phenylethanol, being volatile, will evaporate from the skin surface to some degree. We have found that approximately 70% of a dose of phenylethanol applied to rat skin in vitro is absorbed when the skin is occluded, compared with 64% when the skin is not occluded (S.A.M. Hotchkiss, unpublished data, 1991). This indicates that the degree of evaporation throughout the experiment is insignificant compared with the extent of percutaneous absorption. Thus it is likely that when benzyl acetate is applied to the skin in such vehicles, there will be considerable concomitant absorption of the vehicle through the skin into the receptor fluid. Occlusion of the skin surface with parafilm after the application to rat skin in vitro of benzyl acetate in any of the vehicles, generally, but not always, resulted in enhanced percutaneous absorption, although the maximum increase was only 11% of the applied dose. This variable effect is in broad agreement with the existing literature on the in vivo absorption of a wide range of compounds (Bronaugh et al., 1985 and 1990; Bucks et al., 1989; Feldmann and Maibach. 1965). Our data for rat skin in vitro are in partial agreement with the data of Bronaugh et al, (1990) for rhesus monkeys in vivo; these workers determined that occlusion, using a glass chamber capped with parafilm, enhanced more than two-fold the absorption of benzyl acetate (applied in acetone) through rhesus monkey skin in vivo, although a two-fold reduction in absorption was observed when the skin was occluded by direct contact with plastic wrap (Table 4). It is possible that the use of an occlusive layer placed directly upon the skin surface results in the uptake of the test chemical into the occluding material, hence reducing the amount in contact with the skin that is available for percutaneous absorption. In our studies, the absorption of neat benzyl acetate was not enhanced by occlusion with parafilm or teflon caps placed 1.3cm and 2.9cm above the skin surface, respectively. Since the occlusive devises were not placed directly upon the skin, these experimental conditions may differ from conditions in vivo, where Table 4. Effect of occlusion on the percutaneous absorption of benzyl acetate: comparison with data from other workers Absorption (% applied dosel
Conditions Unoccluded Occluded (plastic wrap) Occluded (glass chamber) Occluded (parafilm) Occluded (teflon cap) *Bronaugb R. Sarason R. ingredients Toxicology
Bronaugh et al. (1990)* (rhesus monkey in vivo, 24 hr; acetone vehicle)
This study (rat in vitro, 48 hr: no vehicle)
35 17 79 ND ND
47 ND ND 47 50
ND = not done L., Wester R. C.~ Bucks D., Maibach H. 1. and (1990) In vivo percutaneous absorption of fragrance in rhesus monkeys and humans. Food and Chemical 28, 369-373.
clothing or a dressing may be in direct contact with the skin. In the light of these relatively unpredictable effects, it is important to select conditions of occlusion that best mirror actual in vivo exposure, in order that extrapolations may be made from in vitro results to the conditions of human use. In physiological terms, occlusion of the skin surface prevents transepidermal water loss and results in hydration of the stratum corneum, the water content of which may increase from approximately 10 to 50%. Water is an excellent penetration enhancer, interacting with the polar head groups of the intercellular lipid to cause an increase in stratum corneum fluidity. In addition, certain methods of occlusion increase skin temperalure from 32 to 37:C, and this increases stratum corneum permeability by increasing intercellular lipid disorder. However in our experiments skin temperature was unchanged. Occlusion also prevenls surface loss of the chemical by evaporation, desquamation or abrasion. The total recovery of radioactivity at the end or each experiment was more than 70% when the skin surface was occluded with parafilm. As indicated previously (Hotchkiss et al., 1990), approximately 30% of the dose was lost due to evaporation of the chemical at the time of application (10%) and throughout the course of the experiment (20%). The present data indicate that the total recovery was significantly enhanced, to 90% of the applied dose, when teflon caps, rather than parafilm, were used to occlude the skin surface. The caps form a tighter seal which prevents the loss of the volatile benzyl acetate from the skin surface throughout the experiment. We recommend that such caps are used when investigating the percutaneous absorption or metabolism of volatile compounds. In conclusion, the results presented lit this paper clearly indicate that the in vitro topical disposition of benzyl acetate in rat skin is affected by the vehicle of application and occlusion of the skin surface. However, in terms of safety evaluation, the signilicance of the relatively small changes observed is unclear. Nevertheless, these factors should be considered when designing and interpreting percutaneous absorption studies in order to ensure that the experimental conditions that are selected best represent the likely conditions of human exposure. Taken together with data on percutaneous absorption in humans, such information will ultimately be of value for risk assessment, in order to ensure the safe use of benzyl acetate by humans. are grateful to the Research Institute for Fragrance Materials, USA, ~\~r suppor~ for this research programme. Acknowledgement--We
REFERENCES
Barry B. W. (1987) Penetration enhancers. Mode of action in human skin. In Pharmacology and the Skin. Vol. l Edited by B. Shroot and H. Schaefer. pp. 12l 137. Karger, Basel.
Percutaneous absorption of benzyl acetate. 2 Barry B. W. (1989) Optimizing percutaneous absorption. In Percutaneous Absorption: Mechanisms, Methodology, Drug Deliver)'. Edited by R. L. Bronaugh and H. I. Maibach. pp. 531-554. Marcel Dekker, New York. Bartek M. J., LaBudde J. A. and Maibach H. I. (1972) Skin permeability in vivo: comparison in rat, rabbit, pig and man. Journal of Investigative Dermatology 58, 114-123. Bronaugh R. L. and Stewart R. F. (1985) Methods for in vitro percutaneous absorption studies IV. The flowthrough diffusion cell. Journal o f Pharmaceutical Science 74, 64-67. Bronaugh R. L., Stewart R. F., Wester R. C., Bucks D. and Maibach H. I. (1985) Comparison of percutaneous absorption of fragrances by humans and monkeys. Food and Chemical Toxicology 23, 111-114. Bronaugh R. L., Wester R. C., Bucks D., Maibach H. I. and Sarason R. (1990) In vivo percutaneous absorption of fragrance ingredients in rhesus monkeys and humans. Food and Chemical Toxicology 28, 369-373. Bucks D. A. W., Maibach H. I. and Guy R. H. (1989) Occlusion does not uniformly enhance penetration in vivo. In Percutaneous Absorption. Mechanisms, Methodology, Drug Delivery. Edited by R. L. Bronaugh and H. I. Maibach. 2nd Ed. pp. 77-93. Marcel Dekker, New York. Chidgey M. A. J., Kennedy J. F. and Caldwell J. (1987) Studies on benzyl acetate. III. The percutaneous absorption and disposition of [methylene-lac] benzyl acetate in the rat. Food and Chemical Toxicology 25, 521 525. Elias P. M. (1981) Epidermal lipids, membranes, and keratinization. International Journal of Dermatology 20, 1 -19.
Embery G. and Dugard P. H. (1971) The isolation of dimethyl sulfoxide soluble components from human epidermal preparations: a possible mechanism for dimethyl sulfoxide in effecting percutaneous migration phenomena. Journal of Investigative Dermatology 57, 308 311. Feldmann R. J. and Maibach H. I. (1965) Penetration of ~4C hydrocortisone through normal skin. Archives ~/ Dermatology 91, 661~666. Grasso P. and Lansdown A. B. G. (1972) Methods of
153
measuring and factors affecting percutaneous absorption. Journal of the Society of Cosmetic Chemists 23, 481-521. Greif N. (1967) Cutaneous safety of fragrance materials as measured by the maximization test. American Perfumer and Cosmetics 82, 54-57. Guy R. H., Mak V. H. W., Kai T., Bommannan D. and Potts R. O. (1990) Percutaneous penetration enhancers: mode of action. In Prediction of Percutaneous Penetration. Methods, Measurements and Modelling. Edited by R. C. Scott, R. H. Guy and J. Hadgraft. pp. 213-223. IBC, London. Hotchkiss S. A., Chidgey M. A. J., Rose S. and Caldwell J. (1990) Percutaneous absorption of benzyl acetate through rat skin in vitro. 1. Validation of an in vitro model against in vivo data. Food and Chemical Toxicology 28, 443 447. ldson B. (1983) Vehicle effects in percutaneous absorption. Drug Metabolism Reviews 14, 207-222. Longnecker D. S., Roebuck B. D., Curphey T. J., Lhoste E., Coon C. 1. and MacMillan D. (1986) Effects of corn oil and benzyl acetate on number and size of azaserineinduced foci in the pancreas of LEW and F344 rats. Environmental Health Perspectit~es 68, 197 201. Longnecker D. S., Roebuck B. D., Curphey T. J. and MacMillan D. L. (1990) Evaluation of promotion of pancreatic carcinogenesis in rats by benzyl acetate. Food and Chemical Toxicology 28, 665 668. National Toxicology Program (1986) Toxicology and carcinogenesis studies of benzyl acetate (CAS No. 140-11-4) in F344/N rats and B6C3F1 mice (gavage studies). NTP Technical Report Series, no. 250. Scheuplein R. J. (1972) Properties of the skin as a membrane. Advances in Biology of the Skin 12, 125 152. Scheuplein R. J. and Blank I. H. (1971) Permeability of the skin. Physiological Retiews 51, 702 747. Scheuplein R. J. and Ross L. (1970) Effects of surfactants and solvents on the permeability of epidermis. Journal qf the Society of Cosmetic Chemists 21, 853-873. Wester R. C. and Maibach H. 1. (1983) Cutaneous pharmacokinetics: 10 stages to percutaneous absorption. Drug Metabolism Reviews 14, 169 205.
0278-6915/92 $5.00 + 0.00 Copyright ~50 1992 Pergamon Press plc
Printed in Great Britain. All rights reserved
P E R C U T A N E O U S ABSORPTION OF BENZYL ACETATE T H R O U G H RAT SKIN I N VITRO. 2. EFFECT OF VEHICLE A N D OCCLUSION S. A. M. HOTCHKISS, J. M. MILLER a n d J. CALDWELL Department of Pharmacology and Toxicology, St Mary's Hospital Medical School, Imperial College of Science Technology and Medicine, London W2 IPG, UK (Accepted 15 November 1991)
Abstract--The effect of vehicle and occlusion on the/n vitro percutaneous absorption of [methylene-~4C] benzyl acetate (1.7-16.6mg/cm 2) has been studied in diffusion cells using full thickness skin from male Fischer 344 rats. Absorption of neat benzyl acetate through rat skin occluded with parafilm was 49.3 _+ 2.0% (mean + SD, n = 4) after 48 hr. When benzyl acetate in ethanol was applied to the skin and the skin was occluded with parafilm, the extent of absorption at 48 hr was not significantly different from that after neat application. However at 6 hr, as the ethanol content of the application mixture was increased, the absorption of benzyl acetate through occluded skin was enhanced proportionally (r = 0.999). When phenylethanol was used as a vehicle, the extent of the benzyl acetate absorption through occluded skin at 48 hr was enhanced (P < 0.05) compared with that after application neat; with 50% (v/v) phenylethanol, absorption at 48 hr was 56.3 + 4.9%. However, this enhanced absorption did not correlate with the proportion of phenylethanol in the application mixture. When dimethylsulphoxide was used as a vehicle, the extent of absorption of benzyl acetate through occluded skin at 48 hr was enhanced (P < 0.05) compared with that after application neat (absorption was 59.3 + 3.7% of the applied dose when 50% (v/v) dimethylsulphoxide was used). As the dimethylsulphoxide content of the application mixture was increased, the absorption of benzyl acetate was enhanced proportionally. Occlusion of the skin surface with parafilm often significantly enhanced absorption (P < 0.05), although the effect varied with time and vehicle. In general, the degree of any enhanced absorption caused by the use of a vehicle or occlusion of the skin was small, and, in most cases, would be unlikely to be toxicologically significant.
INTRODUCTION The a w l alkyl ester, benzyl acetate, is a naturally occurring fragrance and flavouring agent found in the oils o f j a s m i n , hyacinth a n d gardenia a n d used in the m a n u f a c t u r e of perfumes, household chemicals and foodstuffs. Because o f the widespread h u m a n exposure to benzyl acetate, mainly t h r o u g h the skin, there is currently considerable interest in its safety evaluation. In a 2-yr carcinogenicity bioassay, benzyl acetate was reported to cause pancreatic acinar a d e n o m a s in rats after a d m i n i s t r a t i o n by gavage in corn oil (National Toxicology Program, 1986), a l t h o u g h the t u m o u r s have been ascribed to the corn oil vehicle since they were also present in vehicle-fed control animals (Longnecker et al., 1986). In further studies, dietary benzyl acetate was observed to act as a weak p r o m o t e r (but not initiator) o f pancreatic pre-neoplastic loci in the rat (Longnecker et al., 1990). In terms of local toxicity, benzyl acetate is not irritant to rabbit skin (Research Institute for Fragrance Materials, personal c o m m u n i c a t i o n , 1991), a n d is not a skin-sensitizing agent in h u m a n s (Greif, 1967). However, there is a lack o f data on
Abbreviation: DMSO = dimethylsulphoxide.
systemic toxicity after dermal application, a n d there is d o u b t a b o u t the relevance of data from rodent studies using high oral doses to the effects of exposure of h u m a n s to low doses via the dermal route. Since there is significant h u m a n exposure to benzyl acetate t h r o u g h the skin, it is i m p o r t a n t to investigate the rate a n d extent of percutaneous a b s o r p t i o n of the chemical, and the factors that may influence this absorption, in order to o b t a i n i n f o r m a t i o n that will contribute towards the safer use of the chemical in humans. The skin a b s o r p t i o n of benzyl acetate has been studied in the rat, b o t h in vitro (Hotchkiss et al., 1990) and in t,ivo (Chidgey et al., 1987). However, the effect of vehicle a n d occlusion of the skin surface u p o n the a b s o r p t i o n of the c o m p o u n d has not been systematically examined. Various vehicles are currently in use for the formulation of products that come into contact with skin and for the study of percutaneous a b s o r p t i o n and dermal toxicity. In the light of data from m a n y workers (Barry, 1987 and 1989; G u y et al., 1990; Idson, 1983) regarding the penetration e n h a n c i n g effect of such vehicles, it is i m p o r t a n t to determine the effect that these formulation c o m p o n e n t s may have upon the a b s o r p t i o n kinetics of topically applied compounds such as benzyl acetate. Ethanol and phenylethanol are c o m m o n l y used as m a j o r ingredients (up to 90%, v/v) in the 145
146
S.A.M. HOTCHKISSet al.
composition of fragrance products, whilst dimethylsulphoxide (DMSO) is widely used in dermatological studies as a solvent and is the archetypal penetration enhancer. Such chemicals may not be inert vehicles, but may act to enhance percutaneous absorption by a combination of mechanisms including fluidization of stratum corneum lipids, solvent effects, hydration or lipid extraction or by other means of barrier perturbation. Under conditions of dermal exposure to fragrances, drugs and other chemicals, the site of application may be left open to the atmosphere, or it may be occluded by clothing or dressings. Occlusion is known to enhance the percutaneous absorption of many, but not all. topically applied chemicals (Bronaugh et al., 1985 and 1990: Bucks et al., 1989; Feldmann and Maibach, 1965; Wester and Maibach, 1983), and therefore it is important to investigate its effect on the extent of percutaneous penetration of chemicals such as benzyl acetate. The mechanism for this enhancement is complex and not completely understood, but may involve a combination of factors including the reduction of epidermal water loss (which results in an increase in skin hydration) and, in some cases, an increase in skin temperature. In addition, occlusion may reduce the surface loss of compounds by evaporation and abrasion. We have previously established an in vitro model for percutaneous absorption which may be used to predict the in rit,o absorption of benzyl acetate in the rat (Hotchkiss et al., 1990). We now report on the use of this technique to study the effect of two variables, the nature of the vehicle of application and occlusion of the application site, on the percutaneous absorption of benzyl acetate though rat skin in ritro.
MATERIALS AND M E T H O D S
Compounds. [methylene-~4C]Benzyl acetate (sp. act. 53 mCi/mmol; radiochemical purity > 96%) was a custom synthesis from Amersham International plc (Amersham, Bucks., UK). Unlabelled benzyl acetate and DMSO were obtained from Aldrich Chemical Co. (Gillingham, Dorset, UK). Ethanol was purchased from James Burrough Ltd (Witham, Essex, UK) and 2-phenylethanol was obtained from Sigma Chemical Co. (Poole, Dorset, UK). All other chemicals were obtained from standard suppliers of laboratory chemicals and were of analytical grade. h7 vitro diffusion system. Percutaneous absorption was assessed in diffusion cells using full-thickness skin from male Fischer 344 rats, as previously described (Hotchkiss et al., 1990). The system, based on that of Bronaugh and Stewart (1985), consisted of seven teflon diffusion cells and a fraction collector (Crown Glass Co. lnc, Somerville, N J, USA), a thermostatically controlled Churchill water circulator and
a model 202V/AA16 cassette peristaltic pump (Watson-Marlow, Falmouth, Cornwall, UK). Skin preparation. Male Fischer 344 rats, weighing approximately 200 g, were obtained from Harlem Olac (Oxford, Oxon., UK), and were maintained on a diet of Biosure (Manea, Cambs., UK) CRM pellets and water ad lib. The rats were killed by cervical dislocation following light ether anaesthesia, and then the dorsal region was shaved with animal clippers and the shaved area was cut out with dissecting scissors. The skin was placed, epidermal side uppermost, on a plastic dissecting board and circles of full-thickness skin, 1.7 cm in diameter, were cut out using a circular sharpened steel cutter. Excess subcutaneous tissue was removed with a scalpel. The skin was placed in diffusion cells in the flow-through apparatus, and allowed to equilibrate for about 30 min. The skin surface temperature was maintained at 32C, and the underside of the skin was in contact with a continuously flowing receptor fluid of 0.9% saline (degassed before use by filtration through a 0.5 Itm filter) at a flow rate of 1.5 ml/hr. Dose application and determination of absorption. Aliquots of 5/~1 [methylene-~4C]benzyl acetate (1.7 16.6 mg/cm 2) were applied to the exposed skin surface (0.32 cm2) using a microsyringe fitted with a bluntended needle. The compound was applied either neat or in a vehicle of ethanol, phenylethanol or DMSO, so that the vehicle content varied from 0 to 90% (v ~v). The skin was then either occluded with a layer of standard self-sealing Nescofilm (Solmedia, Romford, Essex, UK), 1.3 cm above the skin surface, or left open to the atmosphere. In some experiments, teflon caps (Crown Glass, Somerville, N J, USA), positioned 2.9 cm above the skin surface, were used to occlude the skin instead of the layer of Nescofilm. The receptor fluid (1.5 ml) was collected hourly into 20-ml scintillation vials for up to 72hr and then 10ml scintillation fluid (Ecoscint, National Diagnostics. Manville, N J, USA) was added and levels of ~(" were determined by liquid scintillation spectrometry (Packard Tri-Carb Model 4640). Determination o[residual benzyl acetate. At the end of the experiment the skin was washed by swabbing twice with lint soaked in ethanol, and the washings were counted to determine radioactivity remaining on the skin surface. Radioactivity remaining within the skin was determined by digesting the skin circles in 1 ml of a solution of 20 g sodium hydroxide in 150 ml water, 75 ml methanol and 25 ml Triton 405, in a shaking water-bath for 1 hr at 70C. After incubation, the mixture was cooled and 0.5 ml 4.4 m-nitric acid was added, followed by 10 ml scintillant. Radioactivity was counted by liquid scintillation spectrometry, using a quench curve obtained with untreated skin spiked with a known quantity of radioactivity. Statistical evaluation. Where appropriate, results were compared using Student's t-test for unpaired data. The level of significance was taken as P < 0.05.
Percutaneous absorption of benzyl acetate. 2
147
Table 1. Effect of vehicle on the percutaneous absorption of benzyl acetate though occluded (parafilm) rat skin in vitro Vehicle content (v/v)
Vehicle None Ethanol
Phenylethanol
Dimethylsulphoxide
10% 20% 50% 80% 90% 10% 20% 50% 80% 90% 10% 20% 50% 80% 90%
Absorption (% applied dose)¢ at: 6 hr 4.6 ± 0.6 5.6 ± 0.9 7.1 + 1.5' 10.8 ± 2.7* 14.5+_4.0" 15.4 + 4.9* 6.0 ± 0.6* 7.4 + 0.6* 6.6 + 0.2* 7.7 +- 0.2* 5.8 ± 0.8* 6.9 + 1.6" 10.0+ 1.1" 11.8 ± 3.5* 23.1 +_6.4* 31.1 ± 5.8*
24 hr
48 hr
35.3 ± 3.4 35.0 +- 3.8 33.4±2.7 38.9 ± 1.6" 41.1 +_4.2* 39.8 ± 2.5* 36.5 ± 1.9 39.5 + 0.9* 38.1 ± 3.9 42.1 ± 1.4" 37.6 ± 3.2 39.0 ± 1.6" 43.2+_3.8* 44.7 +_4.5* 50.4 ± 6.9* 55.9 ± 5.5*
49.3 ± 2.0 51.9 ± 4.2 47.1 ±4.4 50.0 ± 1.0 51.5±2.6 52.2 + 5.3 54.4 ± 2.3* 57.8 ± 1.5" 56.3 + 4.9* 59.2 ± 2.5* 53.2 ± 4.8* 55.9 + 1.1" 57.0±2.8* 59.3 ± 3.7* 63.8 __.5.7* 63.6 ± 4.6*
"tValues are means + SD for four diffusion cells run concurrently, and those marked with an asterisk differ significantly (Student's t-test) from the corresponding control (no vehicle) values (*P < 0.05).
RESULTS
Percutaneous absorption o f benzyl acetate applied in ethanol I n a g r e e m e n t with o u r p r e v i o u s l y p u b l i s h e d results ( H o t c h k i s s et al., 1990), the a b s o r p t i o n o f neat benzyl acetate t h r o u g h rat skin u n d e r o c c l u s i o n w i t h p a r a -
film in vitro was extensive, r e a c h i n g 4 9 . 3 % o f the applied dose at 48 hr ( T a b l e 1; Fig 1). W h e n benzyl acetate w a s applied to the skin in e t h a n o l , at vehicle c o n t e n t s r a n g i n g f r o m 10 to 9 0 % (v/v), the extent o f a b s o r p t i o n t h r o u g h occluded skin at 48 h r w a s n o t significantly different f r o m t h a t after a p p l i c a t i o n neat. A b s o r p t i o n w a s 50.0% w h e n 50% (v/v) e t h a n o l
60
50
e~ 0
~"
40
m
30
o
20
N~ II1
10
0
~ 0
8
16
24
32
40
48
Time (hr) Fig. l. Effect of ethanol vehicle and parafilm occlusion on the absorption of benzyl acetate through rat skin in vitro after application neat (C)--unoccluded, O---occluded) or in 50% (v/v) ethanol (V]--unoccluded, I---occluded). Values are means + SD, n = 3 (unoccluded), n = 4 (occluded).
S. A. M, HOTCHKISS et al.
148
60
*
*
*
*
== 50
40
f,.}
o.
30
g
20
e- v 10
I
0
I
20
I
40
Vehicle
I
60
I
80
concentration
(%
100
v/v)
Fig. 2. Relationship between vehicle concentration and extent of benzyl acetate absorption through occluded rat skin in vitro at 6 hr (open symbols) and 48 hr (closed symbols) after application in ethanol ([Z, I ) , phenylethanol ((3, I ) or dimethylsulphoxide (z~, at). Values are means (n = 4). Error bars are omitted for clarity. Values marked with an asterisk differ significantly (Student's /-test) from the corresponding value for application neat (*P < 0.05). 70
60
"a
50
e-,
40
"el
Ila (lJ
*" o
30
¢1 O
~g
20
e- "--" m 10
0
~
0
8
16
24 Time
32
40
48
(hr)
Fig. 3, Effect of phenylethanol vehicle and parafilm occlusion on the absorption of benzyl acetate through rat skin in ritro after application neat ( © - - u n o c c l u d e d , O--occluded) or in 50% (v/v) phenylethanol ([~ --unoccluded, I - - o c c l u d e d ) . Values are means + SD, n = 3 (unoccluded), n = 4 (occluded).
Percutaneous absorption of benzyl acetate. 2
149
70
60
50
o: A o
¢~ '1~
40
"o
::t 0 o
20 m
10
0
~ 0
8
16
24
32
40
48
Time (hr) Fig. 4. Effect of dirnethylsulphoxide vehicle and parafilm occlusion on the absorption of benzyl acetate through rat skin in vitro after application neat (O--unoccluded, O--occluded) or in 50% (v/v) dimethylsulphoxide ([]--unoccluded, I---occluded). Values are means + SD, n = 3 (unoccluded), n = 4 (occluded). was used (Table l; Fig. 1). However, at 6 h r the absorption of benzyl acetate was increased by increasing the ethanol content of the application mixture, rising from 4.6% at 0% vehicle to 15.4% at 90% (v/v) vehicle (r = 0.999; Table 1; Fig. 2).
Percutaneous absorption of benzyl acetate applied in phenylethanol When benzyl acetate was applied to the skin in phenylethanol, at vehicle contents ranging from 10 to 90% (v/v), the extent of absorption through parafilmoccluded skin at 48 hr was statistically significantly enhanced compared with that after application neat. When 50% (v/v) phenylethanol was used absorption was 56.3% of the applied dose (Table 1; Fig. 3). However, at no time after application did this enhanced absorption correlate with the proportion of phenylethanol in the application mixture (r = 0.171 at 6hr; r = 0 . 1 7 6 at 48hr; Table 1; Fig. 2).
Percutaneous absorption of benzyl acetate applied in DMSO When benzyl acetate was applied to the skin in DMSO, at vehicle contents ranging from 10 to 90% (v/v), the extent of absorption through occluded skin at 48 hr was statistically significantly enhanced corn-
pared with that after application neat. Absorption was 59.3% when 50% (v/v) DMSO was used (Table 1; Fig. 4). As the DMSO content of the application mixture was increased over the range 0 to 90% (v/v), the absorption of benzyl acetate at 6 hr rose from 4.6% at 0% vehicle to 31.1% at 90% (v/v) vehicle (r = 0.917). Similarly, the absorption at 48 hr also increased proportionally, from 49.3% at 0% vehicle to 63.6% at 90% (v/v) vehicle (r = 0.871; Table I; Fig. 2). At virtually all time points, DMSO enhanced the absorption of benzyl acetate to a statistically significantly greater extent than either ethanol or phenylethanol. The maximum penetration enhancement effect for all vehicles was observed with 90% (v/v) DMSO at 6 hr, where absorption was enhanced by 26.5% of the applied dose.
Effect of parafilm occlusion absorption
on benz)'l acetate
The absorption of neat benzyl acetate at 48 hr was not significantly affected by occlusion with parafilm (Table 2). However, when benzyl acetate was applied to the skin in a vehicle of ethanol, phenylethanol or DMSO occlusion usually resulted in a statistically significant increase in the absorption of benzyl acetate at 48 hr, although this difference was not always
S. A. M. HOTCHKISS et al.
150
Table 2. Effect of parafilm occlusion on the percutaneous absorption of benzyl acetate though rat skin m t itr, Absorption (% applied dose)t at: Vehicle content (v/v)
Vehicle None Ethanol
Phenylethanol
Dimethylsulphoxide
. . Occluded
6 hr . . .
4.6 + 0.6 5.6 + 0.9 7.1 -: I 5 10.8 ! 2.7 14.5±4.0 15.4 + 4.9 6.0 + 0.6 7.4 + 0.6 6.6 + 0.2 7.7 + 0.2 5,8 + 0 . 8 6 . 9 + 1.6 10.0 -+ h i 11.8 ~ 3.5 23.1 ~- 6.4 31.1 j_5.8
10% 20% 50% 80% 90% 10% 20% 50% 8(/% 90% 10% 20% 50% 80% 90%
48 hr . . . . . . Unoccluded 4.2 + 0.5 4. L ± 1.3" 7.1 + 2.1 10A + 1.5 10.5±1.3" 9,7 ± 0.7" 5,3 ± 0.7* 6.8 ± 0.9 5.9 ± 1.9 6.4 +- 0.1" 5.1 + 0 . 9 6.2-+2.1 8.2 ± 0.8* 9.9 Z 5,1 23.0 ± 5.8 27.5±3.7
. Occluded 49.3 ± 2.0 51,9 ± 4.2 47.1 ± 4.4 50.0 ± h0 51.5±2.6 52.2 ~ 5.3 54.4 -+ 2.3 57.8 -+ 1.5 56.3 i 4.9 59.2 ± 2.5 53,2+_4,8 5 5 . 9 2 l.I 57.0 ± 2.8 59.3 ± 3.7 63.8 +- 5.7 63.6±4.6
Unoccludcd 4 7 9 4 19 4 5 5 4 4 8" 45. ~, ~ I ,~ 44.7 + ~ 6 ' 41 I ~ 3.8" 4 ~ 6 ~ ~;1' 50.9 ~ I.~* 526 ~ 25* 5(k8 :? t~.4 ~0.3 - 0 9* 48.5 ~ 49* 491 ~ 4 5 " 4~).I f I 2* 48 6 ,- ~.4" 530 ~; 2.9* 52.6 ~ 3.(1"
*Values are means ± SD for four (occluded) or three (unoccluded) diffusion cells run concurrently. Thosc marked with an asterisk differ significantly(Student's t-test) from the corresponding occluded values (*P < 0.0s)
a p p a r e n t at 6 hr. At 6 h r , the m a x i m u m effect of occlusion was observed with 9 0 % (v/v) ethanol, where occlusion e n h a n c e d the a b s o r p t i o n by 5.7% of the applied dose. At 48 hr, the m a x i m u m effect of occlusion was observed with 90% (v/v) D M S O , where the a b s o r p t i o n of benzyl acetate was e n h a n c e d by 11.0% of the applied dose (Table 2). Effect o J teflon absorption
cap
occlusion
on
benzy[
acetate
Limited data were o b t a i n e d with teflon caps rather t h a n parafilm as the occlusive devise. As with parafilm, occlusion of the skin site with teflon caps after the application of neat benzyl acetate did not e n h a n c e the extent of a b s o r p t i o n o f the chemical t h r o u g h rat skin. There was no significant difference between the a b s o r p t i o n of neat benzyl acetate at 48 hr t h r o u g h skin occluded with parafilm and that t h r o u g h skin occluded with teflon caps (Table 3).
Table 3, Effect of occlusion by parafilm or teflon caps on the recovery of radioactivity after topical application of neat benzyl acetate to rat
Site Receptor fluid (0 48hr) Receptor fluid (0-72 hr) Skin surface Skin Diffusion cell Parafilm/cap
Total
skin in vitro Recovery ofradioactivity(%applieddose)~, ........... Occlud-~d- . . . . Unoccluded P a r a f i l m Tefloncap 47.0+3.1
47.2+4.1
49.7+_2.5
52.4±3.4 0.4 + 0.1 11.7 + 5.0 0.6 ± 0.3 --
54.7_+4.3 0.3 ~ 0.1 11.2 z 3.6 0.7 +_ 0.5 0.2 i 0.2
54.5+3.0 0.4 ± 0,1 11.2 +_ 3.3 0.9 ± 0.5 26.1 ± 4.6*
64.8+_5.7
68.4±3.9
93,0±3.6"t
~Values are means ± SD for four diffusion cells run concurrently. Values marked with an asterisk differ significantly (Student's t-test) from the corresponding values without occlusion (*P < 0.05), and those marked with a dagger differ significantly (Student's t-test) from the corresponding values with parafilm occlusion (tP < 0.05).
Balance ~t" radioactivity
At 72 hr after the application of benzyl acetate, the recovery of radioactivity in the skin was a r o u n d 1 0 - 1 5 % of the applied dose, and was not related to the vehicle of application or occlusiom In all cases, less t h a n 1% was recovered from the skin surface by washing with ethanol, and less than I % was recovered on the parafilm that was used to occlude the skin surface. The total recovery of radioactivity u n d e r the occluded conditions described was approximately 70%. These data are in accordance with our previous results (Hotchkiss et al., 1990). In studies with neat benzyl acetate, the recovery of radioactivity from the skin surface and from within the tissue was not significantly different following occlusion with parafilm or teflon caps, a l t h o u g h the total recovery of radioactivity was significantly increased after occlusion with the caps, reaching more than 90% of the applied dose (Table 3).
DISCUSSION
The percutaneous absorption of benzyl acetate has been studied previously in the rat, both in cit'o (Chidgey et al., 1987) and in Ntro (Hotchkiss et al., 1990), and a b o u t 50% of the applied dose was a b s o r b e d after 48 hr. After topical application to rat skin, the extent o f a b s o r p t i o n in citrt, was found to correlate well with that observed m Nco. hence validating the in vitro system as a model for the study of factors affecting benzyl acetate a b s o r p t i o n (Hotchkiss et al., 1990). F u r t h e r in ciro and m citro studies showed that the application of benzyl acetate to rat skin in a vehicle of 50% ethanol did not affect the total a m o u n t o f c o m p o u n d a b s o r b e d at 48 hr (Chidgey et al., 1987; Hotchkiss et al., 1990). However, a wide variety o f vehicles, at various concentrations, are used in the formulation of products that
Percutaneous absorption of benzyl acetate. 2 come into contact with skin and for the study of percutaneous absorption and dermal toxicity, and it is known that certain vehicles have penetrationmodulating effects (Barry, 1987 and 1989; Grasso and Lansdown, 1972; Guy et al., 1990; Idson, 1983). Therefore we undertook a more detailed and systematic investigation of the effect of such vehicles on the percutaneous absorption of the model fragrance chemical, benzyl acetate. For the absorption studies described in this paper, only total radioactivity has been reported, and the analysis of specific metabolites has not been included. Studies previously conducted in our laboratory (Chidgey et al., 1987) have shown that benzyl acetate, administered topically to rats in ~'it,o, undergoes rapid and extensive hydrolysis to benzyl alcohol, followed by oxidation to benzoic acid and then conjugation with glycine to form hippuric acid. A small amount of benzoic acid is conjugated with glucuronic acid to form benzoyl glucuronide, and some benzylmercapturic acid is also formed. The skin itself is known to possess substantial esterase activity and we have preliminary data that suggest that the first step in this detoxication pathway (hydrolysis) occurs almost exclusively in the skin during percutaneous absorption (S.A.M. Hotchkiss, unpublished observations, 1991). It is therefore likely that the majority of the radiolabelled compound detected after absorption is benzyl alcohol, and not the parent ester, benzyl acetate. The use of ethanol, phenylethanol or DMSO as vehicles for the topical application of benzyl acetate generally resulted in a statistically significant enhancement of penetration, although the biological relevance of the relatively small increases observed (up to 27% of the applied dose, hut usually less than 10%) is uncertain. From the viewpoint of safety evaluation, if these differences were observed to occur in L,ivo in the rat. they could be either toxicologically insignificant or significant, depending upon the study conditions. The effect is more likely to be of importance in long-term studies with repeated dermal applications at relatively short intervals, where the use of a vehicle that enhances bioavailability (even by 10%) may result in enhanced exposure and possibly toxicity. However, in order to extrapolate these data from rats to the effects of exposure in humans, information on the percutaneous absorption of benzyl acetate through human skin is required. Such studies are currently under way in our laboratory, and preliminary evidence suggests that the absorption through human skin (about 12%) is significantly less than through rat skin (about 50%) (S.A.M. Hotchkiss, unpublished observations, 1991). The relatively poor permeability of human skin compared with that of other species is in broad agreement with reports in the literature by other workers using a variety of compounds (Bartek et al., 1972). The relatively small increases in absorption caused by the vehicles used in our studies are difficult to reconcile with the large (up to 50-fold) increases often
151
observed by other workers. However, in the studies described in our paper, a finite dose of benzyl acetate was selected (on the basis of a more accurate representation of true conditions of exposure in vivo), which was rapidly and extensively absorbed, even in the absence of any vehicle. The rapid disappearance of this source of benzyl acetate from the skin surface into the receptor fluid (and into the air by evaporation) would make it impossible for any vehicle to enhance absorption by the magnitude observed by other workers. It should be noted that the enhanced absorption afforded by these vehicles was obtained despite the possible decrease in thermodynamic activity of benzyl acetate caused by its dilution with the vehicle. In order for a topically applied chemical to be absorbed into the systemic circulation, it must first traverse the stratum corneum (the major skin barrier), the viable epidermis and the portion of the upper dermis which contains the cutaneous microcirculation. Passage through the densely packed keratinocyte layer which constitutes the stratum corneum may be by transcellular or intercellular routes, or, less significantly, through sweat glands or hair follicles. There is evidence to suggest that for lipophilic molecules such as benzyl acetate (log P = 1.96), absorption through the stratum corneum may be through either the intercellular or intracellular lipid regions of the keratinocytes (Barry, 1987; Scheuplein, 1972). However, as the volume of intracellular lipid is relatively small compared with the intercellular volume, the latter may be the dominant route for dermal absorption (Elias, 1981). It has been postulated that polar penetration enhancers such as ethanol and phenylethanol, and aprotic solvents such as DMSO, may interact with the polar head groups of the intercellular lipid bilayer to modify hydrogen and ionic bonds and so disturb lipid packing. This may result in increased fluidity of the lipid bilayer, hence enhancing the movement of lipophilic molecules. In addition, solvents such as ethanol, phenylethanol and DMSO may penetrate in significant quantities into the intercellular aqueous layer, thus enhancing the solubility of lipophilic chemicals in this region (Barry, 1987). However, studies by Guy et al. (1990) suggest that ethanol acts as a penetration enhancer in human skin, not by ftuidization of stratum corneum intercellular lipids, but by lipid extraction. At high concentrations, these solvents may actually damage the skin, resulting in a loss of barrier function. Other mechanisms for the mode of action of DMSO have been postulated, including denaturation of structural keratin (Scheuplein and Ross, 1970), displacement of bound water from proteins (Scheuplein and Blank, 1971) and elution of stratum corneum lipids (Embery and Dugard, 1971). After topical application of benzyl acetate in either ethanol, phenylethanol or DMSO, the fate of the vehicle may not necessarily be the same as the chemical under investigation, although in order for a vehicle to affect the skin barrier it must presumably
152
S.A.M. HO'rCHK1SSet al.
partition into the tissue to some extent. Even under conditions of occlusion, it is likely that the alcohols, ethanol and phenylethanol, being volatile, will evaporate from the skin surface to some degree. We have found that approximately 70% of a dose of phenylethanol applied to rat skin in vitro is absorbed when the skin is occluded, compared with 64% when the skin is not occluded (S.A.M. Hotchkiss, unpublished data, 1991). This indicates that the degree of evaporation throughout the experiment is insignificant compared with the extent of percutaneous absorption. Thus it is likely that when benzyl acetate is applied to the skin in such vehicles, there will be considerable concomitant absorption of the vehicle through the skin into the receptor fluid. Occlusion of the skin surface with parafilm after the application to rat skin in vitro of benzyl acetate in any of the vehicles, generally, but not always, resulted in enhanced percutaneous absorption, although the maximum increase was only 11% of the applied dose. This variable effect is in broad agreement with the existing literature on the in vivo absorption of a wide range of compounds (Bronaugh et al., 1985 and 1990; Bucks et al., 1989; Feldmann and Maibach. 1965). Our data for rat skin in vitro are in partial agreement with the data of Bronaugh et al, (1990) for rhesus monkeys in vivo; these workers determined that occlusion, using a glass chamber capped with parafilm, enhanced more than two-fold the absorption of benzyl acetate (applied in acetone) through rhesus monkey skin in vivo, although a two-fold reduction in absorption was observed when the skin was occluded by direct contact with plastic wrap (Table 4). It is possible that the use of an occlusive layer placed directly upon the skin surface results in the uptake of the test chemical into the occluding material, hence reducing the amount in contact with the skin that is available for percutaneous absorption. In our studies, the absorption of neat benzyl acetate was not enhanced by occlusion with parafilm or teflon caps placed 1.3cm and 2.9cm above the skin surface, respectively. Since the occlusive devises were not placed directly upon the skin, these experimental conditions may differ from conditions in vivo, where Table 4. Effect of occlusion on the percutaneous absorption of benzyl acetate: comparison with data from other workers Absorption (% applied dosel
Conditions Unoccluded Occluded (plastic wrap) Occluded (glass chamber) Occluded (parafilm) Occluded (teflon cap) *Bronaugb R. Sarason R. ingredients Toxicology
Bronaugh et al. (1990)* (rhesus monkey in vivo, 24 hr; acetone vehicle)
This study (rat in vitro, 48 hr: no vehicle)
35 17 79 ND ND
47 ND ND 47 50
ND = not done L., Wester R. C.~ Bucks D., Maibach H. 1. and (1990) In vivo percutaneous absorption of fragrance in rhesus monkeys and humans. Food and Chemical 28, 369-373.
clothing or a dressing may be in direct contact with the skin. In the light of these relatively unpredictable effects, it is important to select conditions of occlusion that best mirror actual in vivo exposure, in order that extrapolations may be made from in vitro results to the conditions of human use. In physiological terms, occlusion of the skin surface prevents transepidermal water loss and results in hydration of the stratum corneum, the water content of which may increase from approximately 10 to 50%. Water is an excellent penetration enhancer, interacting with the polar head groups of the intercellular lipid to cause an increase in stratum corneum fluidity. In addition, certain methods of occlusion increase skin temperalure from 32 to 37:C, and this increases stratum corneum permeability by increasing intercellular lipid disorder. However in our experiments skin temperature was unchanged. Occlusion also prevenls surface loss of the chemical by evaporation, desquamation or abrasion. The total recovery of radioactivity at the end or each experiment was more than 70% when the skin surface was occluded with parafilm. As indicated previously (Hotchkiss et al., 1990), approximately 30% of the dose was lost due to evaporation of the chemical at the time of application (10%) and throughout the course of the experiment (20%). The present data indicate that the total recovery was significantly enhanced, to 90% of the applied dose, when teflon caps, rather than parafilm, were used to occlude the skin surface. The caps form a tighter seal which prevents the loss of the volatile benzyl acetate from the skin surface throughout the experiment. We recommend that such caps are used when investigating the percutaneous absorption or metabolism of volatile compounds. In conclusion, the results presented lit this paper clearly indicate that the in vitro topical disposition of benzyl acetate in rat skin is affected by the vehicle of application and occlusion of the skin surface. However, in terms of safety evaluation, the signilicance of the relatively small changes observed is unclear. Nevertheless, these factors should be considered when designing and interpreting percutaneous absorption studies in order to ensure that the experimental conditions that are selected best represent the likely conditions of human exposure. Taken together with data on percutaneous absorption in humans, such information will ultimately be of value for risk assessment, in order to ensure the safe use of benzyl acetate by humans. are grateful to the Research Institute for Fragrance Materials, USA, ~\~r suppor~ for this research programme. Acknowledgement--We
REFERENCES
Barry B. W. (1987) Penetration enhancers. Mode of action in human skin. In Pharmacology and the Skin. Vol. l Edited by B. Shroot and H. Schaefer. pp. 12l 137. Karger, Basel.
Percutaneous absorption of benzyl acetate. 2 Barry B. W. (1989) Optimizing percutaneous absorption. In Percutaneous Absorption: Mechanisms, Methodology, Drug Deliver)'. Edited by R. L. Bronaugh and H. I. Maibach. pp. 531-554. Marcel Dekker, New York. Bartek M. J., LaBudde J. A. and Maibach H. I. (1972) Skin permeability in vivo: comparison in rat, rabbit, pig and man. Journal of Investigative Dermatology 58, 114-123. Bronaugh R. L. and Stewart R. F. (1985) Methods for in vitro percutaneous absorption studies IV. The flowthrough diffusion cell. Journal o f Pharmaceutical Science 74, 64-67. Bronaugh R. L., Stewart R. F., Wester R. C., Bucks D. and Maibach H. I. (1985) Comparison of percutaneous absorption of fragrances by humans and monkeys. Food and Chemical Toxicology 23, 111-114. Bronaugh R. L., Wester R. C., Bucks D., Maibach H. I. and Sarason R. (1990) In vivo percutaneous absorption of fragrance ingredients in rhesus monkeys and humans. Food and Chemical Toxicology 28, 369-373. Bucks D. A. W., Maibach H. I. and Guy R. H. (1989) Occlusion does not uniformly enhance penetration in vivo. In Percutaneous Absorption. Mechanisms, Methodology, Drug Delivery. Edited by R. L. Bronaugh and H. I. Maibach. 2nd Ed. pp. 77-93. Marcel Dekker, New York. Chidgey M. A. J., Kennedy J. F. and Caldwell J. (1987) Studies on benzyl acetate. III. The percutaneous absorption and disposition of [methylene-lac] benzyl acetate in the rat. Food and Chemical Toxicology 25, 521 525. Elias P. M. (1981) Epidermal lipids, membranes, and keratinization. International Journal of Dermatology 20, 1 -19.
Embery G. and Dugard P. H. (1971) The isolation of dimethyl sulfoxide soluble components from human epidermal preparations: a possible mechanism for dimethyl sulfoxide in effecting percutaneous migration phenomena. Journal of Investigative Dermatology 57, 308 311. Feldmann R. J. and Maibach H. I. (1965) Penetration of ~4C hydrocortisone through normal skin. Archives ~/ Dermatology 91, 661~666. Grasso P. and Lansdown A. B. G. (1972) Methods of
153
measuring and factors affecting percutaneous absorption. Journal of the Society of Cosmetic Chemists 23, 481-521. Greif N. (1967) Cutaneous safety of fragrance materials as measured by the maximization test. American Perfumer and Cosmetics 82, 54-57. Guy R. H., Mak V. H. W., Kai T., Bommannan D. and Potts R. O. (1990) Percutaneous penetration enhancers: mode of action. In Prediction of Percutaneous Penetration. Methods, Measurements and Modelling. Edited by R. C. Scott, R. H. Guy and J. Hadgraft. pp. 213-223. IBC, London. Hotchkiss S. A., Chidgey M. A. J., Rose S. and Caldwell J. (1990) Percutaneous absorption of benzyl acetate through rat skin in vitro. 1. Validation of an in vitro model against in vivo data. Food and Chemical Toxicology 28, 443 447. ldson B. (1983) Vehicle effects in percutaneous absorption. Drug Metabolism Reviews 14, 207-222. Longnecker D. S., Roebuck B. D., Curphey T. J., Lhoste E., Coon C. 1. and MacMillan D. (1986) Effects of corn oil and benzyl acetate on number and size of azaserineinduced foci in the pancreas of LEW and F344 rats. Environmental Health Perspectit~es 68, 197 201. Longnecker D. S., Roebuck B. D., Curphey T. J. and MacMillan D. L. (1990) Evaluation of promotion of pancreatic carcinogenesis in rats by benzyl acetate. Food and Chemical Toxicology 28, 665 668. National Toxicology Program (1986) Toxicology and carcinogenesis studies of benzyl acetate (CAS No. 140-11-4) in F344/N rats and B6C3F1 mice (gavage studies). NTP Technical Report Series, no. 250. Scheuplein R. J. (1972) Properties of the skin as a membrane. Advances in Biology of the Skin 12, 125 152. Scheuplein R. J. and Blank I. H. (1971) Permeability of the skin. Physiological Retiews 51, 702 747. Scheuplein R. J. and Ross L. (1970) Effects of surfactants and solvents on the permeability of epidermis. Journal qf the Society of Cosmetic Chemists 21, 853-873. Wester R. C. and Maibach H. 1. (1983) Cutaneous pharmacokinetics: 10 stages to percutaneous absorption. Drug Metabolism Reviews 14, 169 205.
