- Email: [email protected]
Epidermal Enzymes as Penetration Enhancers in Transdermal Drug Delivery?
March 1996 Volume 85, Number 3
RESEARCH ARTICLES
Epidermal Enzymes as Penetration Enhancers in Transdermal Drug Delivery? SUNITA PATIL, PARMINDER SINGH, CHRISTIANE SZOLAR-PLATZER,
AND
HOWARD MAIBACHX
Received May 30, 1994, from the Surge 110, Department of Dermatology, University of California, San Francisco, CA 94143. revised manuscript received October 9, 1995. Accepted for publication November 14, 1995X. Abstract 0 Epidermal enzymes play an important role in the process of differentiation of keratinocytes. The present preliminary in vitro study was undertaken to observe if topical enzyme treatment influenced permeation of compounds across the skin. Due to the noted function and importance of phosphatidylcholine metabolism during maturation of the barrier lipids, the effects of topical application of the phosphatidylcholine dependent phospholipase C enzyme (not present in epidermis) on skin penetration of three model drugs, viz. benzoic acid, mannitol and testosterone, were studied. Similar studies were also carried out using epidermal enzymes like triacylglycerol hydrolase, acid phosphatase, and phospholipase A2 (present in epidermis). Pretreatment of skin with phospholipase C significantly enhanced permeation of benzoic acid, mannitol, and testosterone relative to untreated skin. Triacylglycerol hydrolase (neutral) increased the penetration of mannitol 3-fold and had no effect on benzoic acid penetration. Topical application of acid phosphatase did not alter the permeation of any of these compounds. Phospholipase A2 significantly enhanced permeation of benzoic acid and mannitol while it did not have any effect on the penetration of testosterone. These results for the first time demonstrate that enzymes may remarkably affect and/or regulate the permeation of topically applied compounds.
Introduction Phospholipids are present in the epidermis especially in the viable layers and the stratum corneum-stratum granulosum (SC-SG) interface and to a lesser extent in the lower most layers of the stratum corneum.1 Major epidermal phospholipids are phosphatidylcholine, phosphatidylethanolamine, and sphingomyelin. During keratinocyte differentiation, epidermal phospholipids are subject to enzymatic conversion into more nonpolar species, which in turn contribute to the barrier function of the stratum corneum.2 Choline-containing lipids are found to be in excess in certain pathological disorders,3,4 although their significance is not yet known. X
Abstract published in Advance ACS Abstracts, January 1, 1996.
© 1996, American Chemical Society and American Pharmaceutical Association
Final
Phosphatidylcholine metabolism plays a significant role during maturation of the epidermal barrier lipids and keratinocyte differentiation. The presence of phosphatidylcholine dependent phospholipase C (PLC) in the epidermis has not yet been reported. Acid phosphatase is known to take part in the process of dyshesion of corneocytes. Lipases, sphingomyelinases, and phospholipases play a central role in the conversion of predominantly polar lipids into the nonpolar neutral lipids and sphingolipids.5 Triacylglycerolhydrolase (lipases), acid phosphatase, and phospholipase A2 are reported to be associated with the lamellar bodies, their secreted contents, and also within the intercellular corneocyte membranes. Arachidonic acid, the precursor of prostaglandins and leukotrienes, can be directly liberated from membrane phospholipids by phospholipase A2 or indirectly by phosphatidylinositol dependent phospholipase C.6-9 These enzymes do not directly influence the barrier function of skin but trigger a signal transduction mechanism by activation of lipid and calcium sensitive protein kinase c and by mobilization of intracellular calcium stores.10 Such processes include cell proliferation and differentiation of keratinocytes.11 Triacylglycerol hydrolase (lipase) is classified as either an acid or neutral lipase on the basis of its pH optimum. It has been identified in human epidermis both in the epidermal lamellar bodies and the corneocyte intercellular domains.5 Lipase is therefore important for the conversion of the predominantly polar contents of the lamellar bodies into the nonpolar neutral lipids. New technologies for topical drug delivery and challenging methods to alter and overcome the primary barrier formed by the SC intercellular lipids are constantly being developed.12 The presence of the above enzymes and their importance in the skin barrier function has been elucidated.5-9 No information is however available as to the role, if any, of these enzymes in transdermal drug delivery. Given the above scenario, the focus of the present work was to investigate the role of topically applied epidermal enzymes (most of them known to be natural constituents of epidermis) in drug
0022-3549/96/3185-0249$12.00/0
Journal of Pharmaceutical Sciences / 249 Vol. 85, No. 3, March 1996
Table 1sSteady State Flux Data of Different Solutes across Excised Human Skin after Various Enzyme Treatments Steady State Fluxa (nmol/cm2/h) Compound Benzoic acid Mannitol Testosterone Benzoic acid Mannitol Testosterone Benzoic acid Mannitol Testosterone Benzoic acid Mannitol
Control
Treated
Phospholipase C 1.33 ± 0.04 7.56 ± 1.66 0.014 ± 0.012 0.062 ± 0.031 0.240 ± 0.035 0.401 ± 0.101 Phospholipase A2 0.215 ± 0.049 0.423 ± 0.013 0.029 ± 0.016 0.097 ± 0.009 0.77 ± 0.12 0.81 ± 0.48 Triacylglycerol Hydrolase 0.061 ± 0.018 0.040 ± 0.007 0.013 ± 0.003 0.045 ± 0.016 0.241 ± 0.035 0.269 ± 0.16 Acid Phosphatase 0.267 ± 0.062 0.358 ± 0.109 0.056 ± 0.009 0.054 ± 0.03
p Value
Enhancement Ratiob
<0.001 <0.002 <0.03
5.7 4.4 1.7
<0.005 <0.0001 NSc
2.0 3.3
NS <0.008 NS
3.5
NS NS
a Mean ± SD (n ) 4−6). b Flux across enzyme pretreated skin/flux across untreated skin. c Not significant.
permeation. As a first step, the role of these enzymes was studied in vitro using an excised human skin model. The effects of four enzymes, namely, phosphatidylcholine dependent phospholipase C, lipase (neutral), acid phosphatase, and phospholipase A2 (all reported to be present in skin except phospholipase C), on the skin permeation of benzoic acid, mannitol, and testosterone were studied. The effects of PLC on the structural arrangement of lipid bilayers in the stratum corneum were also evaluated using electron microscopic techniques.5
Materials and Methods MaterialssThe enzymes phospholipase C, triacylglycerol hydrolase, acid phosphatase and phospholipase-A2 were obtained from Sigma Chemical Co. (St. Louis, MO). Mannitol and [14C]mannitol were also obtained from Sigma Chemical Co., benzoic acid was from J. T. Baker Chemical Co. (Phillipsburg, PA), 14C benzoic acid and 14C testosterone were from NEN Research Products (Boston, MA). Other chemicals and solvents were of reagent grade and obtained commercially. Skin PreparationsDermatomed skin (approximate thickness of 600 µm) was obtained from local hospital sources. The tissue was kept frozen at -20 °C, and one or more tissue sample sources were used to study permeation of different drugs. The control and treatment procedures for a particular enzyme were always conducted on a single tissue source to avoid interindividual variability. It should also be noted that the penetration studies of all three drugs after a single enzyme treatment in a single tissue source was not possible. Frozen skin samples are widely used in in vitro skin permeation experiments and are reported to be acceptable tissue preparations.13,14 No tissue more than 20 days old was used. Before the experiment, 2 cm2 pieces of skin were excised and washed in MEM Eagle’s tissue culture medium (University of California, San Francisco, Cell Culture Facility) and immediately mounted on the permeability cells. Permeation StudiessVertical, Marzulli-Bronaugh type flow through diffusion cells consisting of two water-jacketed cylindrical half-cells, each having a volume of 2.8 mL and a diameter of 1 cm2, were used for the study. A 1.5 cm2 skin sample was mounted on the diffusion cells. The temperature of the diffusion assembly was maintained at 37 °C by circulating water jacket which retained a temperature of 32 °C on the surface of the skin. The fluid in the receptor cells was constantly stirred by Teflon-coated magnetic bars at about 200 rpm. After mounting on the diffusion cells, the skin samples were treated with appropriate enzyme in respective buffer for a period of 2 h. Control samples were pretreated with the buffer in the absence of enzyme. The receiver compartment was continuously pumped with
250 / Journal of Pharmaceutical Sciences Vol. 85, No. 3, March 1996
phosphate-buffered saline (pH 7.4) at a steady flow rate of 2 mL/h. After pretreatment, excess enzyme/buffer solution was removed and the skin washed four or five times with deionized water to remove any residual enzyme/buffer solution left on the skin surface. Thereafter, the diffusion studies were initiated by charging the donor compartment with 200 µL of either 1 mM benzoic acid (containing ∼1 µCi of [14C]benzoic acid), 1 mM mannitol (containing ∼1 µCi [14C]mannitol), or ∼1 µCi of [14C]testosterone, in deionized water (pH 7.4). The donor compartment was covered with aluminum foil to avoid evaporation of the donor fluid. The samples were collected on a fraction collector over a period of 48 h. Permeation of tritiated water was simultaneously monitored to check the integrity of the skin for each experiment. The following enzymes and buffer systems were used: (1) phospholipase C (EC 3.1.4.3, B. cereus) in Tris HCl buffer (pH 7.2) at concentrations of 10.3 units of solid/100 µL and 20.6 units of solid/ 100 µL of buffer solution; (2) triacylglycerol hydrolase (EC 3.1.1.3, porcine pancreas) in Tris HCl buffer (pH 7.2) at a concentrationof 1000 units of solid/100 µL of buffer solution; (3) acid phosphatase (EC 3.1.3.2, human serum) in Tris maleate buffer (pH 4.8) at a concentration of 1unit of solid/100 µL of buffer solution; (4) phospholipase A2 (EC3.1.1.4, bovine pancreas) in Tris HCl buffer (pH 8.0) at a concentration of 12 units of solid/100 µL of buffer solution. The diffusion experiments were also performed by varying the duration of phospholipase C pretreatment (3 h), the concentration of phospholipase C, and the coapplication of phospholipase C (along with benzoic acid) to study their effect on benzoic acid permeation. The cumulative amount of solute penetrating into the receptor compartment was plotted against time. The steady state flux was estimated from the slope of the linear portion of the cumulative-time profile. Statistical comparisons were made using the paired or unpaired Student’s t-test. The level of significance was taken as p ) 0.05. Electron MicroscopysAfter pretreatment with PLC and with the control buffer, the skin was cut into fine pieces of approximately 1 mm3 and fixed in glutaraldehyde fixative (0.1 M cacodylate buffer, pH 7.2) for 1 h at room temperature. The samples were then left overnight in the same fixative at 4 °C. Tissues were later washed repeatedly with 0.1 M cacodylate buffer and postfixed in 0.5% buffered ruthenium tetroxide (RuO4), containing 2% potassium ferrocyanide for 1 h at 4 °C.15 Tissues were routinely dehydrated in ethanol and embedded in Spurr’s low-viscosity resin. Ultrathin sections were viewed under a transmission electron microscope (TEM, Zeiss 10 A EM, operated at 60 kV) after staining with uranyl acetate and lead citrate. Both control and PLC-treated tissues were processed in triplicate to confirm the structural observations in the skin.
Results and Discussion Table 1 lists the steady state flux data for the three model solutes across human skin pretreated for 2 h with different enzymes. The integrity of the skin during penetration studies was checked by simultaneously quantifying the flux of tritiated water across the skin. The data were rejected if the permeability coefficient for water exceeded 0.0035 cm/h. The enhancement ratio (i.e. flux across enzyme-treated skin/flux across untreated skin) for each treatment is also shown in Table 1. There was apparent variability in the steady state flux data of the same compound across untreated skin from different sources. Such interspecimen variability has been demonstrated earlier by different workers.16 As a result, the transport studies (control and treated) for any particular solute and enzyme were conducted across the skin of the same donor. The enhancement ratios in Table 1 for any solute, therefore, represent data from the skin of the same donor. There was no effect on the permeation of tritiated water with any of the enzymes as compared to no enzyme treatment. The enzyme solutions were prepared in buffer at the pH best suited for their stability.17 The enzyme treatment time was 2 h, which again ensures their stability in aqueous solution. Phospholipase C significantly enhanced skin permeation of benzoic acid, mannitol, and testosterone with the maximum effect observed for benzoic acid penetration. In-
Figure 1sElectron micrograph of phospholipase C-treated skin showing no alterations of the lamellar lipid bilayers in the stratum corneum (arrow; 125000×). Also note a normal secreting lamellar body (lb) (inset; 160000×).
creasing the duration of phospholipase C treatment from 2 to 3 h did not effect skin permeability of benzoic acid (p > 0.05). The coapplication of phospholipase C and benzoic acid gave flux values comparable to those from the 2 h enzyme pretreatment study. Increasing the concentration of phospholipase C had no effect on the transport coefficients of benzoic acid. Concentrations of the enzyme lower than 0.1 units of solid/µL of buffer were not studied. Triacylglycerol hydrolase only enhanced the transport of mannitol with an insignificant effect on benzoic acid and testosterone delivery. Acid phosphatase did not significantly alter the skin permeation behavior of either benzoic acid or mannitol. Phospholipase A2 also significantly enhanced skin permeation of benzoic acid and mannitol. Pretreatment with this enzyme did not affect the transport of testosterone through skin. The lag time was also estimated by extrapolating the straight line portion of the cumulative solute penetratingtime profile to the x-axis. No apparent changes in the lag time were observed for any of the three drugs with or without enzyme pretreatments. Electron microscopic observations demonstrated no disruption of the lipid bilayers in the stratum corneum interstices after pretreating the skin with PLC (Figure 1). The outermost layer of SC also noted a normal pattern of lamellar sheets (Figure 2) compared to the buffer-treated controls (Figure 3).
The lipid secreting lamellar body is also shown (Figure 1, inset). The permeation-enhancing action of PLC may be due to either (1) its effects on the keratinocyte membrane lipids or (2) possibly by altering the configuration of the lipid bilayered sheets. Further analysis of the stratum corneum lipids using spectroscopic and/or chromatographic techniques may provide an additional insight into the action of this enzyme, if any, on the epidermis. The presents results suggest that topical application of epidermal enzymes can facilitate transdermal delivery of compounds. The effects are more pronounced for mannitol as compared to benzoic acid and relatively lipophilic testosterone. Phospholipase C was the most effective of all the enzymes studied in the sense that it facilitated the transport of all the solutes and to the greatest extent. The skin barrier properties are believed to be more unfavorable to the transport of hydrophilic compounds as compared to lipophilic solutes.18 Any disruption of skin lipid infrastructure would be expected to better facilitate the delivery of hydrophilic solutes. Although not yet proven, these effects of topical enzymes appear to be mediated through their effects on skin lipids. Preliminary studies (on only three volunteers) using standard patch tests with phospholipase C and triacylglycerol hydrolase solutions did not show any skin irritation as assessed by transepidermal water loss and local blood flow measurements. Further skin irritation studies using various enzymes with
Journal of Pharmaceutical Sciences / 251 Vol. 85, No. 3, March 1996
Figure 2sOutermost layer of the stratum corneum showing normal lipid lamellae after 2 h treatment with phospholipase C (arrow; 200000×).
References and Notes
Figure 3sElectron micrograph of the buffer-treated (control) skin showing normal bilayered lipid lamellar sheets in the stratum corneum interstices (arrow; 128000×).
different concentrations and durations of application will be one of the aims of the future projects in our laboratory. The enzymes used here are known to be important mediators in lipid biotransformation. These enzymes are distributed in viable epidermis and play an important role in the conversion of phospholipids into more nonpolar neutral lipids, which in turn are principally responsible for the barrier properties of human skin. Such metabolic activity however occurs in viable cells. Further studies to ascertain whether these macromolecular enzymes actually partition into the stratum corneum and to observe the effects of topically applied enzymes on drug permeation in vivo are presently under investigation in our laboratory. Although promising, these results are yet to be investigated in vivo. 252 / Journal of Pharmaceutical Sciences Vol. 85, No. 3, March 1996
1. Gray, G. M.; Yardley, H. J. J. Lipid Res. 1975, 16, 434. 2. Lampe, M. A.; Williams, M. L.; Elias, P. M. J. Lipid Res. 1983, 4, 131-140. 3. Tsambaos, D.; Kalofoutis, A.; Stratigos, J.; Miras, C.; Capetanakis, J. Br. J. Dermatol. 1977, 97, 135-138. 4. Tsambaos, D.; Mahrle, G. Arch. Dermatol. Res. 1979, 266, 177180. 5. Menon, G. K.; Ghadially, R.; Williams, M. L.; Elias, P. M. Br. J. Dermatol. 1992, 126 (4), 337-345. 6. Hammarstrom, S.; Hamberg, M.; Samuelsson, B.; Duell, E. A.; Stawiski, M.; Voorhees, J. J. Proc. Natl. Acad. Sci. U.S.A. 1975, 72, 5130-5134. 7. Hammarstrom, S.; Lindgren, J. A.; Marcelo, C.; Duell, E. A.; Anderson, T. F.; Voorhees, J. J. J. Invest. Dermatol. 1979, 73, 180-183. 8. Ziboh, V.; Casebolt, T. L.; Marcelo, C. L.; Voorhees, J. J. J. Invest. Dermatol. 1984, 83, 426-430. 9. Duell, E. A.; Fortune, J.; Petersen, C.; Ellis, C.; Voorhees, J. J. J. Invest. Dermatol. 1986, 87, 137. 10. Hokin, L. E. Annu. Rev. Biochem. 1985, 54, 205-235. 11. Marcelo, C. L.; Voorhees, J. J. In Advances in Cyclic Nucleotides Research, Hamet, P., Sands, H., Eds.; Raven Press: New York, 1980; Vol. 12, pp 129-137. 12. Ranade, V. V. J. Clin. Pharmacol. 1991, 31, 401-418. 13. Franz, T. J. J. Invest. Dermatol. 1975, 64, 190-195. 14. Hawkins, G. S.; Reifenrath, W. G. Fundam. Appl. Toxicol. 1984, 4, 5133-5144. 15. McNutt, N. S.; Crain, W. L. Cancer 1981, 47, 698-709. 16. Southwell, D.; Barry, B. W.; Woodford, R. Int. J. Pharm. 1984, 18, 299-304. 17. Sigma Catalog (1995).
Acknowledgments We would like to thank the Department of Pathology, University of California, San Francisco, for their assistance in electron microscopy.
JS950220O
RESEARCH ARTICLES
Epidermal Enzymes as Penetration Enhancers in Transdermal Drug Delivery? SUNITA PATIL, PARMINDER SINGH, CHRISTIANE SZOLAR-PLATZER,
AND
HOWARD MAIBACHX
Received May 30, 1994, from the Surge 110, Department of Dermatology, University of California, San Francisco, CA 94143. revised manuscript received October 9, 1995. Accepted for publication November 14, 1995X. Abstract 0 Epidermal enzymes play an important role in the process of differentiation of keratinocytes. The present preliminary in vitro study was undertaken to observe if topical enzyme treatment influenced permeation of compounds across the skin. Due to the noted function and importance of phosphatidylcholine metabolism during maturation of the barrier lipids, the effects of topical application of the phosphatidylcholine dependent phospholipase C enzyme (not present in epidermis) on skin penetration of three model drugs, viz. benzoic acid, mannitol and testosterone, were studied. Similar studies were also carried out using epidermal enzymes like triacylglycerol hydrolase, acid phosphatase, and phospholipase A2 (present in epidermis). Pretreatment of skin with phospholipase C significantly enhanced permeation of benzoic acid, mannitol, and testosterone relative to untreated skin. Triacylglycerol hydrolase (neutral) increased the penetration of mannitol 3-fold and had no effect on benzoic acid penetration. Topical application of acid phosphatase did not alter the permeation of any of these compounds. Phospholipase A2 significantly enhanced permeation of benzoic acid and mannitol while it did not have any effect on the penetration of testosterone. These results for the first time demonstrate that enzymes may remarkably affect and/or regulate the permeation of topically applied compounds.
Introduction Phospholipids are present in the epidermis especially in the viable layers and the stratum corneum-stratum granulosum (SC-SG) interface and to a lesser extent in the lower most layers of the stratum corneum.1 Major epidermal phospholipids are phosphatidylcholine, phosphatidylethanolamine, and sphingomyelin. During keratinocyte differentiation, epidermal phospholipids are subject to enzymatic conversion into more nonpolar species, which in turn contribute to the barrier function of the stratum corneum.2 Choline-containing lipids are found to be in excess in certain pathological disorders,3,4 although their significance is not yet known. X
Abstract published in Advance ACS Abstracts, January 1, 1996.
© 1996, American Chemical Society and American Pharmaceutical Association
Final
Phosphatidylcholine metabolism plays a significant role during maturation of the epidermal barrier lipids and keratinocyte differentiation. The presence of phosphatidylcholine dependent phospholipase C (PLC) in the epidermis has not yet been reported. Acid phosphatase is known to take part in the process of dyshesion of corneocytes. Lipases, sphingomyelinases, and phospholipases play a central role in the conversion of predominantly polar lipids into the nonpolar neutral lipids and sphingolipids.5 Triacylglycerolhydrolase (lipases), acid phosphatase, and phospholipase A2 are reported to be associated with the lamellar bodies, their secreted contents, and also within the intercellular corneocyte membranes. Arachidonic acid, the precursor of prostaglandins and leukotrienes, can be directly liberated from membrane phospholipids by phospholipase A2 or indirectly by phosphatidylinositol dependent phospholipase C.6-9 These enzymes do not directly influence the barrier function of skin but trigger a signal transduction mechanism by activation of lipid and calcium sensitive protein kinase c and by mobilization of intracellular calcium stores.10 Such processes include cell proliferation and differentiation of keratinocytes.11 Triacylglycerol hydrolase (lipase) is classified as either an acid or neutral lipase on the basis of its pH optimum. It has been identified in human epidermis both in the epidermal lamellar bodies and the corneocyte intercellular domains.5 Lipase is therefore important for the conversion of the predominantly polar contents of the lamellar bodies into the nonpolar neutral lipids. New technologies for topical drug delivery and challenging methods to alter and overcome the primary barrier formed by the SC intercellular lipids are constantly being developed.12 The presence of the above enzymes and their importance in the skin barrier function has been elucidated.5-9 No information is however available as to the role, if any, of these enzymes in transdermal drug delivery. Given the above scenario, the focus of the present work was to investigate the role of topically applied epidermal enzymes (most of them known to be natural constituents of epidermis) in drug
0022-3549/96/3185-0249$12.00/0
Journal of Pharmaceutical Sciences / 249 Vol. 85, No. 3, March 1996
Table 1sSteady State Flux Data of Different Solutes across Excised Human Skin after Various Enzyme Treatments Steady State Fluxa (nmol/cm2/h) Compound Benzoic acid Mannitol Testosterone Benzoic acid Mannitol Testosterone Benzoic acid Mannitol Testosterone Benzoic acid Mannitol
Control
Treated
Phospholipase C 1.33 ± 0.04 7.56 ± 1.66 0.014 ± 0.012 0.062 ± 0.031 0.240 ± 0.035 0.401 ± 0.101 Phospholipase A2 0.215 ± 0.049 0.423 ± 0.013 0.029 ± 0.016 0.097 ± 0.009 0.77 ± 0.12 0.81 ± 0.48 Triacylglycerol Hydrolase 0.061 ± 0.018 0.040 ± 0.007 0.013 ± 0.003 0.045 ± 0.016 0.241 ± 0.035 0.269 ± 0.16 Acid Phosphatase 0.267 ± 0.062 0.358 ± 0.109 0.056 ± 0.009 0.054 ± 0.03
p Value
Enhancement Ratiob
<0.001 <0.002 <0.03
5.7 4.4 1.7
<0.005 <0.0001 NSc
2.0 3.3
NS <0.008 NS
3.5
NS NS
a Mean ± SD (n ) 4−6). b Flux across enzyme pretreated skin/flux across untreated skin. c Not significant.
permeation. As a first step, the role of these enzymes was studied in vitro using an excised human skin model. The effects of four enzymes, namely, phosphatidylcholine dependent phospholipase C, lipase (neutral), acid phosphatase, and phospholipase A2 (all reported to be present in skin except phospholipase C), on the skin permeation of benzoic acid, mannitol, and testosterone were studied. The effects of PLC on the structural arrangement of lipid bilayers in the stratum corneum were also evaluated using electron microscopic techniques.5
Materials and Methods MaterialssThe enzymes phospholipase C, triacylglycerol hydrolase, acid phosphatase and phospholipase-A2 were obtained from Sigma Chemical Co. (St. Louis, MO). Mannitol and [14C]mannitol were also obtained from Sigma Chemical Co., benzoic acid was from J. T. Baker Chemical Co. (Phillipsburg, PA), 14C benzoic acid and 14C testosterone were from NEN Research Products (Boston, MA). Other chemicals and solvents were of reagent grade and obtained commercially. Skin PreparationsDermatomed skin (approximate thickness of 600 µm) was obtained from local hospital sources. The tissue was kept frozen at -20 °C, and one or more tissue sample sources were used to study permeation of different drugs. The control and treatment procedures for a particular enzyme were always conducted on a single tissue source to avoid interindividual variability. It should also be noted that the penetration studies of all three drugs after a single enzyme treatment in a single tissue source was not possible. Frozen skin samples are widely used in in vitro skin permeation experiments and are reported to be acceptable tissue preparations.13,14 No tissue more than 20 days old was used. Before the experiment, 2 cm2 pieces of skin were excised and washed in MEM Eagle’s tissue culture medium (University of California, San Francisco, Cell Culture Facility) and immediately mounted on the permeability cells. Permeation StudiessVertical, Marzulli-Bronaugh type flow through diffusion cells consisting of two water-jacketed cylindrical half-cells, each having a volume of 2.8 mL and a diameter of 1 cm2, were used for the study. A 1.5 cm2 skin sample was mounted on the diffusion cells. The temperature of the diffusion assembly was maintained at 37 °C by circulating water jacket which retained a temperature of 32 °C on the surface of the skin. The fluid in the receptor cells was constantly stirred by Teflon-coated magnetic bars at about 200 rpm. After mounting on the diffusion cells, the skin samples were treated with appropriate enzyme in respective buffer for a period of 2 h. Control samples were pretreated with the buffer in the absence of enzyme. The receiver compartment was continuously pumped with
250 / Journal of Pharmaceutical Sciences Vol. 85, No. 3, March 1996
phosphate-buffered saline (pH 7.4) at a steady flow rate of 2 mL/h. After pretreatment, excess enzyme/buffer solution was removed and the skin washed four or five times with deionized water to remove any residual enzyme/buffer solution left on the skin surface. Thereafter, the diffusion studies were initiated by charging the donor compartment with 200 µL of either 1 mM benzoic acid (containing ∼1 µCi of [14C]benzoic acid), 1 mM mannitol (containing ∼1 µCi [14C]mannitol), or ∼1 µCi of [14C]testosterone, in deionized water (pH 7.4). The donor compartment was covered with aluminum foil to avoid evaporation of the donor fluid. The samples were collected on a fraction collector over a period of 48 h. Permeation of tritiated water was simultaneously monitored to check the integrity of the skin for each experiment. The following enzymes and buffer systems were used: (1) phospholipase C (EC 3.1.4.3, B. cereus) in Tris HCl buffer (pH 7.2) at concentrations of 10.3 units of solid/100 µL and 20.6 units of solid/ 100 µL of buffer solution; (2) triacylglycerol hydrolase (EC 3.1.1.3, porcine pancreas) in Tris HCl buffer (pH 7.2) at a concentrationof 1000 units of solid/100 µL of buffer solution; (3) acid phosphatase (EC 3.1.3.2, human serum) in Tris maleate buffer (pH 4.8) at a concentration of 1unit of solid/100 µL of buffer solution; (4) phospholipase A2 (EC3.1.1.4, bovine pancreas) in Tris HCl buffer (pH 8.0) at a concentration of 12 units of solid/100 µL of buffer solution. The diffusion experiments were also performed by varying the duration of phospholipase C pretreatment (3 h), the concentration of phospholipase C, and the coapplication of phospholipase C (along with benzoic acid) to study their effect on benzoic acid permeation. The cumulative amount of solute penetrating into the receptor compartment was plotted against time. The steady state flux was estimated from the slope of the linear portion of the cumulative-time profile. Statistical comparisons were made using the paired or unpaired Student’s t-test. The level of significance was taken as p ) 0.05. Electron MicroscopysAfter pretreatment with PLC and with the control buffer, the skin was cut into fine pieces of approximately 1 mm3 and fixed in glutaraldehyde fixative (0.1 M cacodylate buffer, pH 7.2) for 1 h at room temperature. The samples were then left overnight in the same fixative at 4 °C. Tissues were later washed repeatedly with 0.1 M cacodylate buffer and postfixed in 0.5% buffered ruthenium tetroxide (RuO4), containing 2% potassium ferrocyanide for 1 h at 4 °C.15 Tissues were routinely dehydrated in ethanol and embedded in Spurr’s low-viscosity resin. Ultrathin sections were viewed under a transmission electron microscope (TEM, Zeiss 10 A EM, operated at 60 kV) after staining with uranyl acetate and lead citrate. Both control and PLC-treated tissues were processed in triplicate to confirm the structural observations in the skin.
Results and Discussion Table 1 lists the steady state flux data for the three model solutes across human skin pretreated for 2 h with different enzymes. The integrity of the skin during penetration studies was checked by simultaneously quantifying the flux of tritiated water across the skin. The data were rejected if the permeability coefficient for water exceeded 0.0035 cm/h. The enhancement ratio (i.e. flux across enzyme-treated skin/flux across untreated skin) for each treatment is also shown in Table 1. There was apparent variability in the steady state flux data of the same compound across untreated skin from different sources. Such interspecimen variability has been demonstrated earlier by different workers.16 As a result, the transport studies (control and treated) for any particular solute and enzyme were conducted across the skin of the same donor. The enhancement ratios in Table 1 for any solute, therefore, represent data from the skin of the same donor. There was no effect on the permeation of tritiated water with any of the enzymes as compared to no enzyme treatment. The enzyme solutions were prepared in buffer at the pH best suited for their stability.17 The enzyme treatment time was 2 h, which again ensures their stability in aqueous solution. Phospholipase C significantly enhanced skin permeation of benzoic acid, mannitol, and testosterone with the maximum effect observed for benzoic acid penetration. In-
Figure 1sElectron micrograph of phospholipase C-treated skin showing no alterations of the lamellar lipid bilayers in the stratum corneum (arrow; 125000×). Also note a normal secreting lamellar body (lb) (inset; 160000×).
creasing the duration of phospholipase C treatment from 2 to 3 h did not effect skin permeability of benzoic acid (p > 0.05). The coapplication of phospholipase C and benzoic acid gave flux values comparable to those from the 2 h enzyme pretreatment study. Increasing the concentration of phospholipase C had no effect on the transport coefficients of benzoic acid. Concentrations of the enzyme lower than 0.1 units of solid/µL of buffer were not studied. Triacylglycerol hydrolase only enhanced the transport of mannitol with an insignificant effect on benzoic acid and testosterone delivery. Acid phosphatase did not significantly alter the skin permeation behavior of either benzoic acid or mannitol. Phospholipase A2 also significantly enhanced skin permeation of benzoic acid and mannitol. Pretreatment with this enzyme did not affect the transport of testosterone through skin. The lag time was also estimated by extrapolating the straight line portion of the cumulative solute penetratingtime profile to the x-axis. No apparent changes in the lag time were observed for any of the three drugs with or without enzyme pretreatments. Electron microscopic observations demonstrated no disruption of the lipid bilayers in the stratum corneum interstices after pretreating the skin with PLC (Figure 1). The outermost layer of SC also noted a normal pattern of lamellar sheets (Figure 2) compared to the buffer-treated controls (Figure 3).
The lipid secreting lamellar body is also shown (Figure 1, inset). The permeation-enhancing action of PLC may be due to either (1) its effects on the keratinocyte membrane lipids or (2) possibly by altering the configuration of the lipid bilayered sheets. Further analysis of the stratum corneum lipids using spectroscopic and/or chromatographic techniques may provide an additional insight into the action of this enzyme, if any, on the epidermis. The presents results suggest that topical application of epidermal enzymes can facilitate transdermal delivery of compounds. The effects are more pronounced for mannitol as compared to benzoic acid and relatively lipophilic testosterone. Phospholipase C was the most effective of all the enzymes studied in the sense that it facilitated the transport of all the solutes and to the greatest extent. The skin barrier properties are believed to be more unfavorable to the transport of hydrophilic compounds as compared to lipophilic solutes.18 Any disruption of skin lipid infrastructure would be expected to better facilitate the delivery of hydrophilic solutes. Although not yet proven, these effects of topical enzymes appear to be mediated through their effects on skin lipids. Preliminary studies (on only three volunteers) using standard patch tests with phospholipase C and triacylglycerol hydrolase solutions did not show any skin irritation as assessed by transepidermal water loss and local blood flow measurements. Further skin irritation studies using various enzymes with
Journal of Pharmaceutical Sciences / 251 Vol. 85, No. 3, March 1996
Figure 2sOutermost layer of the stratum corneum showing normal lipid lamellae after 2 h treatment with phospholipase C (arrow; 200000×).
References and Notes
Figure 3sElectron micrograph of the buffer-treated (control) skin showing normal bilayered lipid lamellar sheets in the stratum corneum interstices (arrow; 128000×).
different concentrations and durations of application will be one of the aims of the future projects in our laboratory. The enzymes used here are known to be important mediators in lipid biotransformation. These enzymes are distributed in viable epidermis and play an important role in the conversion of phospholipids into more nonpolar neutral lipids, which in turn are principally responsible for the barrier properties of human skin. Such metabolic activity however occurs in viable cells. Further studies to ascertain whether these macromolecular enzymes actually partition into the stratum corneum and to observe the effects of topically applied enzymes on drug permeation in vivo are presently under investigation in our laboratory. Although promising, these results are yet to be investigated in vivo. 252 / Journal of Pharmaceutical Sciences Vol. 85, No. 3, March 1996
1. Gray, G. M.; Yardley, H. J. J. Lipid Res. 1975, 16, 434. 2. Lampe, M. A.; Williams, M. L.; Elias, P. M. J. Lipid Res. 1983, 4, 131-140. 3. Tsambaos, D.; Kalofoutis, A.; Stratigos, J.; Miras, C.; Capetanakis, J. Br. J. Dermatol. 1977, 97, 135-138. 4. Tsambaos, D.; Mahrle, G. Arch. Dermatol. Res. 1979, 266, 177180. 5. Menon, G. K.; Ghadially, R.; Williams, M. L.; Elias, P. M. Br. J. Dermatol. 1992, 126 (4), 337-345. 6. Hammarstrom, S.; Hamberg, M.; Samuelsson, B.; Duell, E. A.; Stawiski, M.; Voorhees, J. J. Proc. Natl. Acad. Sci. U.S.A. 1975, 72, 5130-5134. 7. Hammarstrom, S.; Lindgren, J. A.; Marcelo, C.; Duell, E. A.; Anderson, T. F.; Voorhees, J. J. J. Invest. Dermatol. 1979, 73, 180-183. 8. Ziboh, V.; Casebolt, T. L.; Marcelo, C. L.; Voorhees, J. J. J. Invest. Dermatol. 1984, 83, 426-430. 9. Duell, E. A.; Fortune, J.; Petersen, C.; Ellis, C.; Voorhees, J. J. J. Invest. Dermatol. 1986, 87, 137. 10. Hokin, L. E. Annu. Rev. Biochem. 1985, 54, 205-235. 11. Marcelo, C. L.; Voorhees, J. J. In Advances in Cyclic Nucleotides Research, Hamet, P., Sands, H., Eds.; Raven Press: New York, 1980; Vol. 12, pp 129-137. 12. Ranade, V. V. J. Clin. Pharmacol. 1991, 31, 401-418. 13. Franz, T. J. J. Invest. Dermatol. 1975, 64, 190-195. 14. Hawkins, G. S.; Reifenrath, W. G. Fundam. Appl. Toxicol. 1984, 4, 5133-5144. 15. McNutt, N. S.; Crain, W. L. Cancer 1981, 47, 698-709. 16. Southwell, D.; Barry, B. W.; Woodford, R. Int. J. Pharm. 1984, 18, 299-304. 17. Sigma Catalog (1995).
Acknowledgments We would like to thank the Department of Pathology, University of California, San Francisco, for their assistance in electron microscopy.
JS950220O