- Email: [email protected]
across intact and diseased skins? Atopic dermatitis as a model
International Journal of Pharmaceutics 497 (2016) 277–286
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
What is the discrepancy between drug permeation into/across intact and diseased skins? Atopic dermatitis as a model Yi-Ping Fanga , Sien-Hung Yangb,c , Chih-Hung Leed,e, Ibrahim A. Aljuffalif , Hsiao-Ching Kaog, Jia-You Fangg,h,i,* a
School of Pharmacy, Kaohsiung Medical University, Kaohsiung, Taiwan School of Traditional Chinese Medicine, Chang Gung University, Kweishan, Taoyuan, Taiwan Department of Traditional Chinese Medicine, Chang Gung Memorial Hospital at Linkou, Taoyuan, Taiwan d Department of Dermatology, College of Medicine, Chang Gung University, Taoyuan, Taiwan e Department of Dermatology, Chang Gung Memorial Hospital at Kaohsiung, Kaohsiung, Taiwan f Department of Pharmaceutics, College of Pharmacy, King Saud University, Riyadh, Saudi Arabia g Pharmaceutics Laboratory, Graduate Institute of Natural Products, Chang Gung University, Kweishan, Taoyuan, Taiwan h Chinese Herbal Medicine Research Team, Healthy Aging Research Center, Chang Gung University, Kweishan, Taoyuan, Taiwan i Research Center for Industry of Human Ecology, Chang Gung University of Science and Technology, Kweishan, Taoyuan, Taiwan b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 16 September 2015 Received in revised form 12 November 2015 Accepted 3 December 2015 Available online 4 December 2015
The discrepancy in drug absorption between healthy and diseased skins is an issue that needs to be elucidated. The present study attempted to explore the percutaneous absorption of drugs via lesional skin by using atopic dermatitis (AD) as a model. Tape-stripping and ovalbumin (OVA) sensitization induced AD-like skin. The lesions were evaluated by physiological parameters, histology, cytokines, and differentiation proteins. The permeants of tacrolimus, 8-methoxypsoralen, methotrexate, and dextran were used to examine in vitro and in vivo cutaneous permeation. Transepidermal water loss (TEWL) increased from 5.2 to 27.4 g/m2/h by OVA treatment. AD-like lesions were characterized by hyperplasia, skin redness, desquamation, and infiltration of inflammatory cells. Repeated OVA challenge produced a Thelper 2 (Th2) hypersensitivity accompanied by downregulation of filaggrin, involucrin, and integrin b. Tacrolimus, the most lipophilic permeant, revealed an increase of cutaneous deposition by 2.7-fold in ADlike skin compared to intact skin. The transdermal flux of methotrexate and dextran, the hydrophilic permeants, across AD-like skin increased about 18 times compared to the control skin. Surprisingly, ADlike skin showed less skin deposition of 8-methoxypsoralen than intact skin. This may be because the deficient lipids in the atopic-affected stratum corneum (SC) diminished drug partitioning into the superficial skin layer. The fluorescence and confocal microscopic images demonstrated a broad and deep passage of small-molecular and macromolecular dyes into AD-like skin. The results obtained from this report were advantageous for showing how the lesional skin influenced percutaneous absorption. ã 2015 Elsevier B.V. All rights reserved.
Keywords: Atopic dermatitis Diseased skin Percutaneous absorption Stratum corneum Tight junction
1. Introduction Skin diseases are the most common human disorders, affecting 30–70% of populations (Hay et al., 2014). The epidemiologic data report an increased incidence of skin diseases over the decades (Andersen and Davis, 2013). The barrier function is usually compromised in diseased conditions. Atopic dermatitis (AD) is a relapsing chronic inflammatory skin disorder with signs of eczema,
* Corresponding author at: Pharmaceutics Laboratory, Graduate Institute of Natural Products, Chang Gung University, 259 Wen-Hwa 1st Road, Kweishan, Taoyuan 333, Taiwan. Fax: +886 3 2118236. E-mail address: [email protected] (J.-Y. Fang). http://dx.doi.org/10.1016/j.ijpharm.2015.12.006 0378-5173/ ã 2015 Elsevier B.V. All rights reserved.
erythema, pruritus, and lichenification. The T-helper 2 (Th2) immune response plays a key role in the pathogenesis of AD (Schneider et al., 2013). The prevalence of AD is measured at 15– 30% for children and 2–10% for adults (Bieber, 2010). The increase of AD incidence by 2–3-fold during the last 3 decades results from increased exposure to environmental allergens, life-style change, and reduced immune tolerance. Skin-barrier deterioration is always accompanied by the occurrence of AD. Previous studies (Wollenberg et al., 2003; Miyagaki and Sugaya, 2015) demonstrated the facile entrance of viruses and bacteria into AD skin due to the defective barrier. There is no information available pertaining to the impact of AD on cutaneous absorption of drugs and actives. It is expected that drug permeation via disrupted skin is greater as compared to that
278
Y.-P. Fang et al. / International Journal of Pharmaceutics 497 (2016) 277–286
viaintact skin. However, the level of this discrepancy between AD and healthy skins is not known. Whether the overabsorption of the drugs induces the adverse risk is also unrevealed. The aim of this study was to evaluate the cutaneous penetration of drugs and macromolecules into/across AD skin in contrast to normal skin. Balb/c mice were used as the animal model by repeated epicutaneous sensitization with ovalalbumin (OVA) to induce AD-like skin. This method has proven to be successful for eliciting the Th2 cell-infiltrating immune response (Oyoshi et al., 2012). Three anti-AD drugs were selected as the permeants in this work for comparing the permeation between diseased and intact skins. These included tacrolimus, 8-methoxypsoralen, and methotrexate. These drugs have various molecular sizes and polarities that are beneficial to exploring the penetration mechanisms of ADlike skin. A macromolecule of fluorescein isothiocyanate (FITC)conjugated dextran with a molecular weight (MW) of 4 kDa (FD4) was also used as a permeant. The skin after OVA treatment was examined by macroscopic and microscopic observations, physiological parameters, and levels of Th2 cytokines. The proteins involved in barrier function such as filaggrin, involucrin, and integrin b were determined by quantitative real-time polymerase chain reaction (qRT-PCR). The Franz diffusion cell was the in vitro skin permeation platform. The in vivo permeant distribution in the skin was monitored by fluorescence and confocal microscopies for vertical and horizontal views, respectively. The present study’s findings may have critical implications for dissecting how the cutaneous diseases influenced skin permeability and the subsequent drug absorption.
2.4. Macroscopic visualization After OVA challenge or vehicle treatment, the macroscopic images of mouse skin were observed using a digital camera (Nikon, Tokyo, Japan) and a handheld digital magnifier (Mini Scope-V, M&T Optics, Taipei, Taiwan). A magnification of 100 was used to capture the images for the magnifier. 2.5. Physiological parameters After OVA or vehicle application, the mouse skin was examined for physiological parameters, including transepidermal water loss (TEWL), erythema (a*), and cutaneous surface pH. A Tewameter (TM300, Courage and Khazaka, Köln, Germany) was employed for estimating TEWL (g/m2/h). A spectrocolorimeter (CD100, Yokogawa, Tokyo, Japan) was used to quantify cutaneous erythema. The pH was determined by Skin-pH-Meter PH905 (Courage and Khazaka). 2.6. Microscopic visualization The dorsal skin was excised from the mouse after sacrifice. The excised skin was immersed in a 10% buffered formaldehyde using ethanol, embedded in paraffin wax, and sliced at a thickness of 5 mm. The specimens were stained with hemoxylin and eosin (H&E) and viewed under light microscopy (IX81, Olympus, Tokyo, Japan). 2.7. Cytokine measurements
2. Materials and methods 2.1. Materials 8-Methoxypsoralen, methotrexate, FITC, FD4, polyethylene glycol 400 (PEG400), and OVA from chicken egg white were purchased from Sigma–Aldrich (St. Louis, MO, USA). Tacrolimus was supplied by United States Pharmacopeia (Rockville, MD, USA). 2.2. Animals Seven-week-old Balb/c mice were supplied by National Laboratory Animal Center (Taipei, Taiwan). This study was carried out in strict accordance with the recommendation in the Guidelines for the Care and Use of Laboratory Animals of Chang Gung University of Science and Technology. All efforts were made to minimize suffering. 2.3. OVA sensitization The procedures of epicutaneous sensitization for the mice were modified from previous studies (Kim et al., 2012; Yanaba et al., 2013). At first, the mice were sensitized with 1 mg/ml OVA (0.1 ml) by intraperitoneal administration every other day for 10 days. At day 8, the dorsal hair of the mice was shaved using an electric clipper and tape-stripped 6 times, mimicking cutaneous injury produced by scratching in AD patients. Next, 100 mg of OVA in 100 ml normal saline was pipetted onto a 1 1 cm sterile gauze, which was topically applied to the dorsal skin. Normal saline without OVA was employed as the vehicle control. The gauze was fixed with Tegaderm adhesive dressing (3M, Maplewood, MN, USA) and a Silkypore stretch bandage (Alcare, Tokyo, Japan). After 24 h, the gauze was removed and new gauze was applied. This procedure was repeated for 7 consecutive days.
The mouse skin was removed 24 h after the last challenge of OVA. The excised skin was suspended in PBS and homogenized by MagNA Lyser (Roche, Indianapolis, IN, USA). After centrifugation, levels of interleukin (IL)-4, IL-5, IL-10, and IL-13 in the tissue supernatant were measured using an enzyme immunoassay kit according to the manufacturer’s instructions (BioLegend). 2.8. Quantitative real-time PCR (qRT-PCR) The levels of filaggrin, involucrin, and integrin b were determined for their mRNA expression. Total RNA was extracted with Trizol (Invitrogen), and cDNA was synthesized by reverse transcription using an iScript cDNA Synthesis Kit (Bio-Rad). The qRT-PCR was performed using iQ SYBR Green Supermix (Bio-Rad) based on the manufacturer’s instructions. The housekeeping gene GAPDH was used as an internal control, and gene expression was calculated using the comparative threshold cycle method. 2.9. In vitro cutaneous permeation The in vitro drug permeation into/across the skin was conducted with the Franz cell. The excised OVA- or vehicletreated skin was mounted between the donor and receptor compartments with the stratum corneum (SC) facing upward toward the donor. The receptor medium was 30% ethanol in pH 7.4 buffer for tacrolimus, 8-methoxypsoralen, and methotrexate to maintain the sink condition. The pH 7.4 buffer was the receptor medium for FD4. The donor (0.5 ml) was loaded with tacrolimus (4% w/v in 30% PEG400/water), 8-methoxypsoralen (0.8% in 40% PEG400/pH 7.4 buffer), methotrexate (1% in water), or FD4 (0.56% in pH 7.4 buffer). The effective diffusion region for the permeants was 0.785 cm2. The receptor temperature and stirring rate of the stirrer were kept at 37 C and 600 rpm, respectively. A 300-ml medium was withdrawn from the receptor at determined intervals. After a 24-h application, the skin was removed from the Franz cell. The skin was washed by double-distilled water for three times to
Y.-P. Fang et al. / International Journal of Pharmaceutics 497 (2016) 277–286
remove the residual drug on skin surface. The drug deposition within the skin was extracted by methanol for tacrolimus, and 8methoxypsoralen and 0.1 N HCl for methotrexate and FD4, respectively. A glass homogenizer (Chin-Fa, Hsinchu, Taiwan) was used to extract the permeants from skin. The high-performance liquid chromatography (HPLC) methods for analyzing the three small-molecule drugs were described previously (Fang et al., 2008; Lin et al., 2010, 2015). FD4 was analyzed by fluorescence spectrophotometer as described by Fang et al. (2004). 2.10. In vivo cutaneous permeation A glass cylinder with a hollow area of 0.785 cm2 was attached to the mouse back by cyanoacrylate superglue. An aliquot of 0.2 ml FITC or FD4 in the same condition as in the in vitro experiment was pipetted into the cylinder. The administration period was 6 h. The animal was then sacrificed, and the treated skin area was excised. The skin species were monitored by fluorescence and confocal microscopies. The fluorescence microscopy was used to observe the vertical section of the skin. The biopsies were cut vertically, embedded in OCT, and frozen at 70 C. The samples were sectioned in a cryostat microtome and mounted with glycerin and gelatin. The slices were monitored by inverted microscope (IX81, Olympus) using a filter set at 450–490 and 515–565 nm for excitation and emission. A confocal laser scanning microscopy (TCS, SP2, Leica, Wetzlar, Germany) was utilized to observe the horizontal section of the skin. The thickness of the skin was scanned at 5 mm increments from the surface via the z-axis. Images were taken by summing 10 fragments at different depths.
279
2.11. Statistical analysis Statistical analysis of differences between the OVA and the vehicle control groups was performed using the Kruskal–Wallis test. The post hoc test for checking individual differences was Dunn’s test. All data were analyzed using WINKS statistical software (Texasoft, Cedar Hill, TX, USA). 3. Results 3.1. Macroscopic visualization and physiological parameters OVA was used to induce AD-like symptoms in the mouse skin. As shown in the left panel of Fig. 1A, the gross appearance of the vehicle-treated mouse skin exhibited an intact surface without disruption. After the repeated challenge by OVA, dryness, redness, edema, hemorrhage and excoriation could be detected on the skin surface as illustrated in the right panel of Fig. 1A. Fig. 1B shows the close-up pictures. As compared with the control skin, the OVAchallenged mouse skin displayed dry, scaly skin with a shiny and erythema tone. Some deep furrows were also observed on the ADlike skin. We next examined whether OVA treatment was associated with alterations in cutaneous physiology, including TEWL, erythema, and pH. As shown in Fig. 1C, the vehicle-treatment group revealed a constant water evaporation rate of 5.2 g/m2/h. The OVA-treated group had a significantly increased TEWL level (27.4 g/m2/h) compared to the control. OVA application resulted in a tendency toward an increase in a* as shown in Fig. 1D. After a treatment of OVA, the skin surface pH increased by about 1 unit as compared to the control (6.3 vs. 5.4) as depicted in Fig. 1E.
Fig. 1. Gross observation and physiological parameters of mouse skin treated with and without OVA: (A) the gross imaging by digital camera; (B) the close-up imaging by handheld digital microscope; (C) transepidermal water loss (TEWL); (D) erythema; and (E) skin surface pH value. The control group corresponds to the mice treated with the vehicle. The data of physiological assessment are presented as the mean of 12 experiments S.D. *, p < 0.05 between the control and AD-like groups.
280
Y.-P. Fang et al. / International Journal of Pharmaceutics 497 (2016) 277–286
Fig. 2. Histological examination and the levels of cytokines and differentiation proteins in mouse skin treated with and without OVA: (A) hemoxylin and eosin (H&E) staining; (B) expression of IL-4, IL-5, IL-10, and IL-13; AND (C) mRNA levels of filaggrin, involucrin, and integrin b. The data of cytokines are presented as the mean of 10– 14 experiments S.D. The data of differentiation proteins are presented as the mean of 4–5 experiments S.D.
Y.-P. Fang et al. / International Journal of Pharmaceutics 497 (2016) 277–286
3.2. Microscopic visualization and skin protein measurements The skin sections stained by H&E were examined to clarify whether OVA influenced cutaneous architecture. As demonstrated in the left panel of Fig. 2A, no pathological changes were found in the vehicle-treated group. OVA sensitization had thickened both the epidermal and dermal layers (the right panel of Fig. 2A). Some detachments of SC were observed in the AD-like lesion. The epidermis deepened into the upper layer of the dermis. There was a visible granular layer in the upper epidermis. We found an excessive infiltration of inflammatory cells into the dermis. These included lymphocytes, neutrophils, macrophages, and mast cells. IL-4, IL-5, IL-10, and IL-13 were analyzed to check the effect of OVA on the production of Th2 cytokines. The OVA-treated skin demonstrated a significant elevation (p < 0.05) of the cytokines compared with the normal skin as shown in Fig. 2B. OVA generally increased these cytokines by 2-fold. This result suggested the involvement of the Th2 immune response with the epicutaneous sensitization of OVA. Some skin proteins related to epidermal junctions were determined as shown in Fig. 2C. The result of qRTPCR demonstrated that mRNA levels of these proteins were significantly downregulated (p < 0.05) in the OVA group. OVA treatment lessened filaggrin, involucrin, and integrin b by about 3, 9-, and 4-fold, respectively. 3.3. In vitro cutaneous permeation We compared the in vitro skin delivery of tacrolimus, 8methoxypsoralen, methotrexate, and FD4 between intact and ADlike skins. Table 1 summarizes the partition coefficient (log P) and MW of the permeants employed in this study. Tacrolimus revealed the greatest lipophilicity, followed by 8-methoxypsoralen, methotrexate, and FD4. Fig. 3 depicts the skin uptake and flux across the skin of the four permeants. The flux was computed from the slope of the cumulative amount in the receptor–time curve. The flux predicted transdermal penetration to the systemic circulation, whereas the skin deposition suggested the drug uptake into the cutaneous reservoir. When tacrolimus was applied to the healthy and diseased skins, the diffusion amount into the receptor was below the detection limit of the analytical method. Fig. 3A shows the tacrolimus deposition within the skin. Tacrolimus accumulation increased significantly (p < 0.05) from 0.25 to 0.66 mg/mg by OVA sensitization. The ratio between the skin deposition of AD-like and normal skins is displayed above the column for AD-like skin in Fig. 3A (2.65). Contrary to the results of tacrolimus, AD-like skin deposition of 8-methoxypsoralen was lower by a factor of 0.6 as compared to that in intact skin as shown in Fig. 3B. On the other hand, increased 8-methoxypsoralen flux was achieved with the OVA treatment. The AD-like lesion significantly promoted (p < 0.05) the flux by 2.4-fold. Fig. 3C shows that methotrexate deposition into AD-like skin was about two folds higher (p < 0.05) than that into vehicle-treated skin. A great increment of methotrexate flux by 17.5-fold was detected for AD-like skin compared to normal skin. No deposition was formed for FD4 delivery in the control group. OVA application Table 1 Physicochemical properties of the permeants tested in this study. Permeant
Partition coefficient (log P)
Molecular weight (Da)
Tacrolimus 8-Methoxypsoralen Methotrexate FD4 FITC
3.3 1.7 0.9 2.0 2.0
822.0 216.2 454.4 4000 389.4
The physicochemical properties of all drugs were obtained from DrugBank website (www.drugbank.ca/).
281
could facilitate FD4 deposition from 0 to 0.46 mg/mg (Fig. 3D). OVA also largely enhanced the transdermal flux of FD4 compared to the control. The increment ratio of FD4 flux (17.8) was comparable to that of methotrexate flux. 3.4. In vivo cutaneous permeation Imaging techniques were employed to examine in vivo cutaneous distribution of fluorescent dyes in normal and lesional skins. FITC was classified as a small-molecule and hydrophilic model permeant (Table 1). No fluorescence was found in the control and AD-like skins without dye administration (data not shown). Fig. 4A presents the fluorescence images of the sectioned skin with or without OVA challenge. A faint green signal was observed in the superficial epidermis of the vehicle-treated skin with a discontinuous distribution. Some FITC molecules accumulated in hair follicles (arrow in the left panel of Fig. 4A), which acted as a reservoir. This indicates the penetration pathway of appendages for topically applied FITC. The fluorescence distribution from FITC was also limited to epidermis and hair follicles of AD-like skin. A deeper and stronger FITC signal was observed in AD-like skin than in the control skin (arrow in the right panel of Fig. 4A). As shown in the left panel of Fig. 4B, FD4 distribution was constrained to the follicles with very few signals in the epidermis. AD-like skin showed a broad distribution of FD4 in the upper epidermis (arrows in the right panel of Fig. 4B). The fluorescence in hair follicles was comparable in intact and diseased skins. Fig. 5 illustrates that the separate x,y-sections gained at each zaxis of the skin by confocal imaging. Each photograph is a planar image taken at every 5-mm depth from the skin surface. The last photograph is the summary of ten separate sections. Increased FITC fluorescence in AD-like skin compared to vehicle-treated skin can clearly be detected as shown in Fig. 5A. An extensive distribution of FITC was observed in lesional skin. A similar result was found in the case of topically applied FD4 (Fig. 5B). The FD4 signal, more pronounced with AD-like skin than with normal skin, could be seen in all skin layers examined. A deeper fluorescence signal was detected in the horizontal image but not in the vertical fluorescence image. This may be due to the higher resolution and sensitivity of confocal microscopy. 4. Discussion AD can be categorized into two forms: an extrinsic type associated with elevated immunoglobulin E (IgE) by environmental allergens and an intrinsic type without IgE sensitization. Approximately 70–80% of AD patients present the extrinsic type (Jin et al., 2009). It is assumed that Balb/c mice exposed to allergens such as OVA reveal the extrinsic form (Shiohara et al., 2004), reflecting most of AD patients. The OVA-challenged model exhibits the advantages of good reproducibility, a short time for AD-like lesion induction, and ease of quantitative evaluation. It is important to assess how the change of barrier function by AD influences skin permeability. Previous studies (Mildner et al., 2010; Mathes et al., 2014) have utilized an organotypic skin culture model to examine the permeation profiles of lucifer yellow and biotin for atopic skin. The dermatitis model developed in this study provided a platform for allowing the assessment of atopic skin permeability. This is the first investigation reporting the effect of AD-like lesion on percutaneous absorption of anti-AD drugs in animals. The classic findings to characterize AD are dryness, erythema, hemorrhage, and scaling. These symptoms were observed in our OVA-induced AD model. The scratching behavior was also revealed in our model. Scaling is a result of premature detachment of a cornified layer in the SC, which can produce a fragile skin structure
282
Y.-P. Fang et al. / International Journal of Pharmaceutics 497 (2016) 277–286
(A)
(C)
(B)
(D)
Fig. 3. Skin deposition and transdermal flux of the permeants into/across mouse skin treated with and without OVA: (A) tacrolimus; (B) 8-methoxypsoralen; (C) methotrexate; and (D) FD4. The data are presented as the mean of 4–5 experiments S.D.
(Kawasaki et al., 2012). TEWL can be a reflection of skin dryness and barrier function. The experimental result indicated an elevated TEWL by OVA challenge, suggesting the disruption of the
cutaneous barrier. The pH of SC is essential for controlling skin homeostasis and barrier function (Schreml et al., 2012). The skin surface pH in healthy humans ranges between 5.4 and 5.9 (Braun-
Y.-P. Fang et al. / International Journal of Pharmaceutics 497 (2016) 277–286
283
Fig. 4. Fluorescence microscopic images of mouse skin treated with and without OVA: (A) the skin topically applied with FITC and (b) the skin topically applied with FD4. Scale bar = 50 mm.
Fig. 5. Confocal microscopic images of mouse skin treated with and without OVA: (A) the skin topically applied with FITC and (b) the skin topically applied with FD4. Both fragments from the skin surface to the depth of 45 mm and a summary of 10 fragments at various skin depths are displayed in this figure. Scale bar = 300 mm.
Falco and Korting, 1986), approximating the skin pH of a normal mouse (5.4) in the present study. The pH value increased in AD-like lesions. This is a common phenomenon observed in AD patients (Levin et al., 2013). The elevation of pH is associated with the
enhanced protease activity and reduced SC lipid lamellae synthesis (Cork et al., 2009), causing the barrier dysfunction. The increase of SC pH and protease activity provides a mechanistic basis of desquamation, as observed in our H&E-stained histology. The
284
Y.-P. Fang et al. / International Journal of Pharmaceutics 497 (2016) 277–286
hyperplasia, keratosis, and influx of neutrophils, macrophages, and mast cells detected in microscopic visualization in response to combined tape stripping and OVA challenge also confirmed the successful dermatitis induction (Oyoshi et al., 2012). The overexpression of IL-4, IL-5, IL-10, and IL-13 was in line with the findings observed in AD. Th2 cytokines are dominant in the acute inflammation phase of AD (Savinko et al., 2012). Our model can be regarded as the acute skin lesion of AD. Th2 cytokines are proven to affect epidermal morphology and barrier characteristics. Overexpressed IL-4 and IL-13 lead to the reduction of filaggrin and involucrin (Hänel et al., 2013). Filaggrin and involucrin synthesized by the differentiating keratinocytes in the basal layer are associated with the cornified envelope in the SC (Danso et al., 2014). Filaggrin is a predominant structural protein in the SC, contributing to a critical role for SC barrier formation. It is reported that 50% of AD patients have the inherited genetic mutation that encodes filaggrin (Guttman-Yassky et al., 2011). The deficiency of filaggrin decreases natural moisture factors (NMFs) in the SC, which results in the increased TEWL and SC pH (Kezic et al., 2008). Our result was similar to this demonstration. Involucrin is a major precursor protein for the cornified envelope. The consequences of involucrin abnormality are barrier deterioration and SC lipid loss (Agrawal and Woodfolk, 2014). Integrin b in the basal layer mediates cell–cell and cell-extracellular matrix interactions, especially the epidermal–dermal adhesion (Has et al., 2012). The suppressed expression of these proteins caused the impairment of barrier function in the AD-like lesion in our model. Both the SC layer and tight junctions (TJs) in the epidermis contribute to the main barrier for drug penetration (Lehman and Franz, 2014). The elevation of TEWL and pH in AD-like skin implied SC damage. The SC abnormality in atopic skin relates to the diminished ceramide amount and impaired lipid lamellar secretion (Man et al., 2008; Sahle et al., 2015). The ceramides and fatty acids in atopic SC have a shorter alkyl chain length than healthy skin (Agrawal and Woodfolk, 2014), suggesting a larger space for permeant penetration. An effortless entrance of allergen and irritant into AD skin via the creation of gaps between the corneocytes has been previously reported (Cork et al., 2009). The facile penetration of the drugs into/across atopic skin can be anticipated through similar pathways. Our permeation profiles confirmed this supposition. The in vivo topical application examined by vertical and horizontal views demonstrated the extensive and deep permeation of small molecules and macromolecules in AD-like skin. Local immune response occurs mainly in the epidermal and dermal layers. It is desirable to target the antiAD drugs to deeper skin strata where inflammation occurs (Pople and Singh, 2013). The follicular route is an important pathway for hydrophilic permeants and macromolecules. According to fluorescence microscopic images, the AD-like lesion did not seem to affect follicular entrance. 8-Methoxypsoralen combined with ultraviolet A irradiation is effective in the treatment of severe chronic AD (Der-Petrossian et al., 2000). Although an AD lesion is expected to accelerate drug diffusion into the skin, it is surprising that 8-methoxypsoralen deposition was decreased with AD-like skin. The basic drug transport into the skin can be controlled by partitioning, solubility, and MW. Hadgraft (2001) has indicated that the ability of drug partitioning into the SC is proportional to the drug amount within the skin. The first principal step in the cutaneous delivery is the partitioning into the SC for the drugs with high lipophilicity such as 8-methoxypsoralen. Although the SC disruption could reduce the inherent barrier effect for promoting drug penetration, the partitioning to the disrupted SC should be decreased because of the deficient ceramides and fatty acids in atopic SC to reduce SC lipophilicity. Another possibility was that the hyperplasia of ADlike skin created a long-distance path for 8-methoxypsoralen
diffusion. Our previous study (Fang et al., 2008) had demonstrated that the pathway length largely influenced 8-methoxypsoralen transport. Further study is needed to confirm this hypothesis about the drug absorption decrease by lesional skin. Tacrolimus shows a higher log P than 8-methoxypsoralen; however, the OVA challenge increased skin uptake of tacrolimus. The partitioning cannot simply explain the permeation of lipophilic permeants through atopic skin. Tacrolimus is a macrolide immunosuppressant for the first-line drug therapy of AD (Svensson et al., 2011). The large molecular volume of tacrolimus (MW = 822 Da) may hinder the passage into/across intact skin. The expansive space for permeation due to ceramide and fatty acid deficiency was beneficial for macrolide penetration. Nevertheless, the magnification of the tacrolimus skin deposition increment by OVA treatment was only 2.7. This indicates that the difficulty of partitioning into AD-like skin still governed tacrolimus permeation. The magnitude of the flux enhancement by AD-like skin of hydrophilic permeants such as methotrexate and FD4 was much higher than that of lipophilic drugs. Oral methotrexate has been demonstrated to be efficient for treating moderate-to-severe AD and psoriasis (Deo et al., 2014). Topical delivery of methotrexate was already developed for treating psoriasis (Vemulapalli et al., 2008). The rate-limiting step for methotrexate transport is at the level of the SC due to its polar and ionic characteristics. SC and TJ damage in AD-like skin led to a great augmentation of methotrexate flux. Methotrexate penetration via AD-like skin displayed a higher enhancement in flux than skin deposition because the hydrophilic permeants can be forced into lipophilic SC compartments, and then quickly continue to the hydrophilic receptor medium (Nielsen, 2005). The lengthened path produced by hyperproliferative skin did not influence the methotrexate passage as shown in the previous report (Lee et al., 2008). FD4 penetration into/across vehicle-treated skin was negligible because of the high hydrophilicity and MW. Both the SC and TJs manifest the obstruction for macromolecule delivery (Hung et al., 2015). Once the skin was disturbed by AD, the macromolecule penetration could be largely increased. This is reasonable since the lesional skin of AD favors the entrance of high-molecular size structures, including bacteria and allergens (Bieber, 2010). It is a general concept that the transport pathways for lipid-soluble molecules through the SC are mainly the extracellular lipid bilayers, whereas the intracellular keratin and interfacial corneocyte cell envelope are the primary routes for water-soluble molecule permeation (Norlén, 2013). Our result suggests that OVA affected both pathways to change the cutaneous drug absorption. The more hydrophilic permeants exhibited higher passage enhancement by AD-like skin. The SC is a skin layer presenting a lipophilic nature. Percutaneous absorption of more hydrophilic drugs would be more affected by skin with a compromised SC barrier. Our result indicates that the SC damage was a principal factor influencing drug permeation via AD-like skin. TJ disruption in viable skin may be the secondary factor. The increased absorption should help in reaching effective drug concentration at the target site; however, it also results in the increased possibility of drug penetration across the skin and into the bloodstream. This risk may be severe for methotrexate since AD-like skin largely facilitated methotrexate transport into the Franz cell receptor. The systemic use of methotrexate is known to elicit some adverse effects such as hepatotoxicity, nausea, anemia, and thrombocytopenia (Schneider et al., 2013). Methotrexate for topical application should be used carefully on lesional skin. The side effects of systemic 8-methoxypsoralen are gastrointestinal and central nervous system disturbances (Der-Petrossian et al., 2000). Although the AD lesion increased only 8-methoxypsoralen flux by a factor of 2.4, this lipid-soluble drug should still be used with caution. Tacrolimus showed a neglected flux in AD-like skin.
Y.-P. Fang et al. / International Journal of Pharmaceutics 497 (2016) 277–286
This suggests a low threat of tacrolimus systemic absorption. However, a 2.7-fold increase in tacrolimus skin deposition by atopic skin may generate the local side effects of tacrolimus such as cutaneous burning, itching, pruritus, and pain. Although currently there is no evidence of the risk of anti-AD drug overabsorption, monitoring after long-term use and the modulation of the applied dose may be greatly needed. Topical administration on large areas of lesional skin should be avoided. 5. Conclusion The incidence of AD and the substantial burden of this disorder are increasing worldwide. Drug therapy for AD is restricted due to the development based on chance observation and the insufficient understanding of drug delivery/targeting. This report introduces an animal model with AD-like lesions, opening up chances to investigate the cutaneous permeation of anti-AD drugs. This model was proved to be predictable, reproducible, and low cost. The representations of Th2 cytokines and differentiation proteins in this AD model were similar to those in AD patients. The experimental result manifests the barrier dysfunction and the increase of drug and macromolecule absorption by OVA sensitization. It was noticeable that 8-methoxypsoralen in the skin reservoir was decreased in AD-like skin more so than in intact skin. This may be due to the lesser amount of partitioning that occurs with a lipid-deficient SC. This suggests that the lesional skin was not necessary to elicit permeation enhancement. There remains a need to develop new anti-AD drugs with low adverse effects. Most studies have employed intact skin to assess drug absorption for developing new drugs. The results from such investigation are not suitable for simulating the actual condition. Application of the AD model was feasible to measure drug absorption for approximating clinically diseased status. This model may also expand its application to other lesional skins. This platform allows us to test large numbers of treatments that cannot be performed in patients due to ethical considerations. Acknowledgment The authors are grateful for the financial support from Chang Gung Memorial Hospital (CMRPD1D0432-3). References Agrawal, R., Woodfolk, J.A., 2014. Skin barrier defects in atopic dermatitis. Curr. Allergy Asthma Rep. 14, 433. Andersen, L.K., Davis, M.D.P., 2013. The epidemiology of skin and skin-related diseases: a review of population-based studies performed by using the Rochester Epidemiology Project. Mayo Clin. Proc. 88, 1462–1467. Bieber, T., 2010. Atopic dermatitis. Ann. Dermatol. 22, 125–137. Braun-Falco, O., Korting, H.C., 1986. Normal pH value of human skin. Hautarzt 37, 126–129. Cork, M.J., Danby, S.G., Vasilopoulos, Y., Hadgraft, J., Lane, M.E., Moustafa, M., Guy, R. H., MacGowan, A.L., Tazi-Ahnini, R., Ward, S.J., 2009. Epidermal barrier dysfunction in atopic dermatitis. J. Invest. Dermatol. 129, 1892–1908. Danso, M.O., van Drongelen, V., Mulder, A., van Esch, J., Scott, H., van Smeden, J., El Ghalbzouri, A., Bouwstra, J.A., 2014. TNF-a and Th2 cytokines induce atopic dermatitis-like features on epidermal differentiation proteins and stratum corneum lipids in human skin equivalents. J. Invest. Dermatol. 134, 1941–1950. Deo, M., Yung, A., Hill, S., Rademaker, M., 2014. Methotrexate for treatment of atopic dermatitis in children and adolescents. Int. J. Dermatol. 53, 1037–1041. Der-Petrossian, M., Seeber, A., Hönigsmann, H., Tanew, A., 2000. Half-side comparison study on the efficacy of 8-methoxypsoralen bath-PUVA versus narrow-band ultraviolet B phototherapy in patients with severe chronic atopic dermatitis. Br. J. Dermatol. 142, 39–43. Fang, J.Y., Fang, C.L., Liu, C.H., Su, Y.H., 2008. Lipid nanoparticles as vehicles for topical psoralen delivery: solid lipid nanoparticles (SLN) versus nanostructured lipid carriers (NLC). Eur. J. Pharm. Biopharm. 70, 633–640. Fang, J.Y., Lee, W.R., Shen, S.C., Wang, H.Y., Fang, C.L., Hu, C.H., 2004. Transdermal delivery of macromolecules by erbium:YAG laser. J. Control. Release 100, 75–85.
285
Guttman-Yassky, E., Nograles, K.E., Krueger, J.G., 2011. Contrasting pathogenesis of atopic dermatitis and psoriasis—part I: clinical and pathologic concepts. J. Allergy Clin. Immunol. 127, 1110–1118. Hadgraft, J., 2001. Modulation of the barrier function of the skin. Skin Pharmacol. Physiol. 14 (Suppl. 1), 72–81. Hänel, K.H., Cornelissen, C., Lüscher, B., Baron, J.M., 2013. Cytokines and the skin barrier. Int. J. Mol. Sci. 14, 6720–6745. Has, C., Spartà, G., Kiritsi, D., Weibel, L., Moeller, A., Vega-Warner, V., Waters, A., He, Y., Anikster, Y., Esser, P., Straub, B.K., Hausser, I., Bockenhauer, D., Dekel, B., Hildebrandt, F., Bruckner-Tuderman, L., Laube, G.F., 2012. Integrin a3 mutations with kidney, lung, and skin disease. N. Engl. J. Med. 366, 1508–1514. Hay, R.J., Johns, N.E., Williams, H.C., Bolliger, I.W., Dellavalle, R.P., Margolis, D.J., Marks, R., Naldi, L., Weinstock, M.A., Wulf, S.K., Michaud, C., Murray, C.J.L., Naghavi, M., 2014. The global burden of skin disease in 2010: an analysis of the prevalence and impact of skin conditions. J. Invest. Dermatol. 134, 1527–1534. Hung, C.F., Chen, W.Y., Aljuffali, I.A., Lin, Y.K., Shih, H.C., Fang, J.Y., 2015. Skin aging modulates percutaneous drug absorption: the impact of ultraviolet irradiation and ovariectomy. Age 37, 21. Jin, H., He, R., Oyoshi, M., Geha, R.S., 2009. Animal models of atopic dermatitis. J. Invest. Dermatol. 129, 31–40. Kawasaki, H., Nagao, K., Kubo, A., Hata, T., Shimizu, A., Mizuno, H., Yamada, T., Amagai, M., 2012. Altered stratum corneum barrier and enhanced percutaneous immune responses in filaggrin-null mice. J. Allergy Clin. Immunol. 129, 1538–1546. Kezic, S., Kemperman, P.M., Koster, E.S., de Jongh, C.M., Thio, H.B., Campbell, L.E., Irvine, A.D., McLean, W.H., Puppels, G.J., Caspers, P.J., 2008. Loss-of-function mutations in the filaggrin gene lead to reduced level of natural moisturizing factor in the stratum corneum. J. Invest. Dermatol. 128, 2117–2119. Kim, S.R., Choi, H.S., Seo, H.S., Choi, Y.K., Shin, Y.C., Ko, S.G., 2012. Topical application of herbal mixture extract inhibits ovalbumin- or 2,4-dinitrochlorobenzeneinduced atopic dermatitis. Evid. Based Complement. Altern. Med. 545497. Lee, W.R., Shen, S.C., Fang, C.L., Zhuo, R.Z., Fang, J.Y., 2008. Topical delivery of methotrexate via skin pretreated with physical enhancement techniques: lowfluence erbium:YAG laser and electroporation. Lasers Surg. Med. 40, 468–476. Lehman, P.A., Franz, T.J., 2014. Effect of induced acute diabetes and insulin therapy on stratum corneum barrier function in rat skin. Skin Pharmacol. Physiol. 27, 249–253. Levin, J., Friedlander, S.F., Del Rosso, J.Q., 2013. Atopic dermatitis and the stratum corneum. Part I: the role of filaggrin in the stratum corneum barrier and atopic skin. J. Clin. Aesthet. Dermatol. 6, 16–22. Lin, Y.K., Huang, Z.R., Zhuo, R.Z., Fang, J.Y., 2010. Combination of calcipotriol and methotrexate in nanostructured lipid carriers for topical delivery. Int. J. Nanomed. 5, 117–128. Lin, Y.K., Yang, S.H., Chen, C.C., Kao, H.C., Fang, J.Y., 2015. Using imiquimod-induced psoriasis-like skin as a model to measure the skin penetration of anti-psoriatic drugs. PLoS One 10, e0137890. Man, M.Q., Hatano, Y., Lee, S.H., Man, M., Chang, S., Feingold, K.R., Leung, D.Y.M., Holleran, W., Uchida, Y., Elias, P.M., 2008. Characterization of a hapten-induced, murine model with multiple features of atopic dermatitis: structural, immunologic, and biochemical changes following single versus multiple oxazolone challenges. J. Invest. Dermatol. 128, 79–86. Mathes, S.H., Ruffner, H., Graf-Hausner, U., 2014. The use of skin models in drug development. Adv. Drug Deliv. Rev. 69–70, 81–102. Mildner, M., Jin, J., Eckhart, L., Kezic, S., Gruber, F., Barresi, C., Stremnitzer, C., Buchberger, M., Mlitz, V., Ballaun, C., Sterniczky, B., Fodinger, D., Tschachler, E., 2010. Knockdown of filaggrin impairs diffusion barrier function and increases UV sensitivity in a human skin model. J. Invest. Dermatol. 130, 2286–2294. Miyagaki, T., Sugaya, M., 2015. Recent advances in atopic dermatitis and psoriasis: genetic background, barrier function, and therapeutic targets. J. Dermatol. Sci. 78, 89–94. Nielsen, J.B., 2005. Percutaneous penetration through slightly damaged skin. Arch. Dermatol. Res. 296, 560–567. Norlén, L., 2013. Current understanding of skin barrier morphology. Skin Pharmacol. Physiol. 26, 213–216. Oyoshi, M.K., He, R., Li, Y., Mondal, S., Yoon, J., Afshar, R., Chen, M., Lee, D.M., Luo, H.R., Luster, A.D., Cho, J.S., Miller, L.S., Larson, A., Murphy, G.F., Geha, R.S., 2012. Leukotriene B4-driven neutrophil recruitment to the skin is essential for allergic skin inflammation. Immunity 37, 747–758. Pople, P.V., Singh, K.K., 2013. Development and evaluation of colloidal modified nanolipid carrier: application to topical delivery of tacrolimus, part II—in vivo assessment, drug targeting, efficacy, and safety in treatment for atopic dermatitis. Eur. J. Pharm. Biopharm. 84, 72–83. Sahle, F.F., Gebre-Mariam, T., Dobner, B., Wohlrab, J., Neubert, R.H.H., 2015. Skin diseases associated with the depletion of stratum corneum lipids and stratum corneum lipid substitution therapy. Skin Pharmacol. Physiol. 28, 42–55. Savinko, T., Matikainen, S., Saarialho-Kere, U., Lehto, M., Wang, G., Lehtimäki, S., Karisola, P., Reunala, T., Wolff, H., Lauerma, A., Alenius, H., 2012. IL-33 and ST2 in atopic dermatitis: expression profiles and modulation by triggering factors. J. Invest. Dermatol. 132, 1392–1400. Schneider, L., Tilles, S., Lio, P., Boguniewicz, M., Beck, L., LeBovidge, J., Novak, N., 2013. Atopic dermatitis: a practice parameter update 2012. J. Allergy Clin. Immunol. 131, 295–299. Schreml, S., Meier, R.J., Albert, M.G., Seidl, U., Zeller, V., Behm, B., Landthaler, M., Abels, C., Babilas, P., 2012. The impact of 10% a-hydroxy acid emulsion on skin pH. Skin Pharmacol. Physiol. 25, 34–38.
286
Y.-P. Fang et al. / International Journal of Pharmaceutics 497 (2016) 277–286
Shiohara, T., Hayakawa, J., Mizukawa, Y., 2004. Animal models for atopic dermatitis: are they relevant to human diseases? J. Dermatol. Sci. 36, 1–9. Svensson, A., Chambers, C., Gånemo, A., Mitchell, S.A., 2011. A systematic review of tacrolimus ointment compared with corticosteroids in the treatment of atopic dermatitis. Curr. Med. Res. Opin. 27, 1395–1406. Vemulapalli, V., Yang, Y., Friden, P.M., Banga, A.K., 2008. Synergistic effect of iontophoresis and soluble microneedles for transdermal delivery of methotrexate. J. Pharm. Pharmacol. 60, 27–33.
Wollenberg, A., Wetzel, S., Burgdorf, W.H., Haas, J., 2003. Viral infections in atopic dermatitis: pathogenic aspects and clinical management. J. Allergy Clin. Immunol. 112, 667–674. Yanaba, K., Kamata, M., Asano, Y., Tada, Y., Sugaya, M., Kadono, T., Tedder, T.F., Sato, S., 2013. CD19 expression in B cells regulates atopic dermatitis in a mouse model. Am. J. Pathol. 182, 2214–2222.
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
What is the discrepancy between drug permeation into/across intact and diseased skins? Atopic dermatitis as a model Yi-Ping Fanga , Sien-Hung Yangb,c , Chih-Hung Leed,e, Ibrahim A. Aljuffalif , Hsiao-Ching Kaog, Jia-You Fangg,h,i,* a
School of Pharmacy, Kaohsiung Medical University, Kaohsiung, Taiwan School of Traditional Chinese Medicine, Chang Gung University, Kweishan, Taoyuan, Taiwan Department of Traditional Chinese Medicine, Chang Gung Memorial Hospital at Linkou, Taoyuan, Taiwan d Department of Dermatology, College of Medicine, Chang Gung University, Taoyuan, Taiwan e Department of Dermatology, Chang Gung Memorial Hospital at Kaohsiung, Kaohsiung, Taiwan f Department of Pharmaceutics, College of Pharmacy, King Saud University, Riyadh, Saudi Arabia g Pharmaceutics Laboratory, Graduate Institute of Natural Products, Chang Gung University, Kweishan, Taoyuan, Taiwan h Chinese Herbal Medicine Research Team, Healthy Aging Research Center, Chang Gung University, Kweishan, Taoyuan, Taiwan i Research Center for Industry of Human Ecology, Chang Gung University of Science and Technology, Kweishan, Taoyuan, Taiwan b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 16 September 2015 Received in revised form 12 November 2015 Accepted 3 December 2015 Available online 4 December 2015
The discrepancy in drug absorption between healthy and diseased skins is an issue that needs to be elucidated. The present study attempted to explore the percutaneous absorption of drugs via lesional skin by using atopic dermatitis (AD) as a model. Tape-stripping and ovalbumin (OVA) sensitization induced AD-like skin. The lesions were evaluated by physiological parameters, histology, cytokines, and differentiation proteins. The permeants of tacrolimus, 8-methoxypsoralen, methotrexate, and dextran were used to examine in vitro and in vivo cutaneous permeation. Transepidermal water loss (TEWL) increased from 5.2 to 27.4 g/m2/h by OVA treatment. AD-like lesions were characterized by hyperplasia, skin redness, desquamation, and infiltration of inflammatory cells. Repeated OVA challenge produced a Thelper 2 (Th2) hypersensitivity accompanied by downregulation of filaggrin, involucrin, and integrin b. Tacrolimus, the most lipophilic permeant, revealed an increase of cutaneous deposition by 2.7-fold in ADlike skin compared to intact skin. The transdermal flux of methotrexate and dextran, the hydrophilic permeants, across AD-like skin increased about 18 times compared to the control skin. Surprisingly, ADlike skin showed less skin deposition of 8-methoxypsoralen than intact skin. This may be because the deficient lipids in the atopic-affected stratum corneum (SC) diminished drug partitioning into the superficial skin layer. The fluorescence and confocal microscopic images demonstrated a broad and deep passage of small-molecular and macromolecular dyes into AD-like skin. The results obtained from this report were advantageous for showing how the lesional skin influenced percutaneous absorption. ã 2015 Elsevier B.V. All rights reserved.
Keywords: Atopic dermatitis Diseased skin Percutaneous absorption Stratum corneum Tight junction
1. Introduction Skin diseases are the most common human disorders, affecting 30–70% of populations (Hay et al., 2014). The epidemiologic data report an increased incidence of skin diseases over the decades (Andersen and Davis, 2013). The barrier function is usually compromised in diseased conditions. Atopic dermatitis (AD) is a relapsing chronic inflammatory skin disorder with signs of eczema,
* Corresponding author at: Pharmaceutics Laboratory, Graduate Institute of Natural Products, Chang Gung University, 259 Wen-Hwa 1st Road, Kweishan, Taoyuan 333, Taiwan. Fax: +886 3 2118236. E-mail address: [email protected] (J.-Y. Fang). http://dx.doi.org/10.1016/j.ijpharm.2015.12.006 0378-5173/ ã 2015 Elsevier B.V. All rights reserved.
erythema, pruritus, and lichenification. The T-helper 2 (Th2) immune response plays a key role in the pathogenesis of AD (Schneider et al., 2013). The prevalence of AD is measured at 15– 30% for children and 2–10% for adults (Bieber, 2010). The increase of AD incidence by 2–3-fold during the last 3 decades results from increased exposure to environmental allergens, life-style change, and reduced immune tolerance. Skin-barrier deterioration is always accompanied by the occurrence of AD. Previous studies (Wollenberg et al., 2003; Miyagaki and Sugaya, 2015) demonstrated the facile entrance of viruses and bacteria into AD skin due to the defective barrier. There is no information available pertaining to the impact of AD on cutaneous absorption of drugs and actives. It is expected that drug permeation via disrupted skin is greater as compared to that
278
Y.-P. Fang et al. / International Journal of Pharmaceutics 497 (2016) 277–286
viaintact skin. However, the level of this discrepancy between AD and healthy skins is not known. Whether the overabsorption of the drugs induces the adverse risk is also unrevealed. The aim of this study was to evaluate the cutaneous penetration of drugs and macromolecules into/across AD skin in contrast to normal skin. Balb/c mice were used as the animal model by repeated epicutaneous sensitization with ovalalbumin (OVA) to induce AD-like skin. This method has proven to be successful for eliciting the Th2 cell-infiltrating immune response (Oyoshi et al., 2012). Three anti-AD drugs were selected as the permeants in this work for comparing the permeation between diseased and intact skins. These included tacrolimus, 8-methoxypsoralen, and methotrexate. These drugs have various molecular sizes and polarities that are beneficial to exploring the penetration mechanisms of ADlike skin. A macromolecule of fluorescein isothiocyanate (FITC)conjugated dextran with a molecular weight (MW) of 4 kDa (FD4) was also used as a permeant. The skin after OVA treatment was examined by macroscopic and microscopic observations, physiological parameters, and levels of Th2 cytokines. The proteins involved in barrier function such as filaggrin, involucrin, and integrin b were determined by quantitative real-time polymerase chain reaction (qRT-PCR). The Franz diffusion cell was the in vitro skin permeation platform. The in vivo permeant distribution in the skin was monitored by fluorescence and confocal microscopies for vertical and horizontal views, respectively. The present study’s findings may have critical implications for dissecting how the cutaneous diseases influenced skin permeability and the subsequent drug absorption.
2.4. Macroscopic visualization After OVA challenge or vehicle treatment, the macroscopic images of mouse skin were observed using a digital camera (Nikon, Tokyo, Japan) and a handheld digital magnifier (Mini Scope-V, M&T Optics, Taipei, Taiwan). A magnification of 100 was used to capture the images for the magnifier. 2.5. Physiological parameters After OVA or vehicle application, the mouse skin was examined for physiological parameters, including transepidermal water loss (TEWL), erythema (a*), and cutaneous surface pH. A Tewameter (TM300, Courage and Khazaka, Köln, Germany) was employed for estimating TEWL (g/m2/h). A spectrocolorimeter (CD100, Yokogawa, Tokyo, Japan) was used to quantify cutaneous erythema. The pH was determined by Skin-pH-Meter PH905 (Courage and Khazaka). 2.6. Microscopic visualization The dorsal skin was excised from the mouse after sacrifice. The excised skin was immersed in a 10% buffered formaldehyde using ethanol, embedded in paraffin wax, and sliced at a thickness of 5 mm. The specimens were stained with hemoxylin and eosin (H&E) and viewed under light microscopy (IX81, Olympus, Tokyo, Japan). 2.7. Cytokine measurements
2. Materials and methods 2.1. Materials 8-Methoxypsoralen, methotrexate, FITC, FD4, polyethylene glycol 400 (PEG400), and OVA from chicken egg white were purchased from Sigma–Aldrich (St. Louis, MO, USA). Tacrolimus was supplied by United States Pharmacopeia (Rockville, MD, USA). 2.2. Animals Seven-week-old Balb/c mice were supplied by National Laboratory Animal Center (Taipei, Taiwan). This study was carried out in strict accordance with the recommendation in the Guidelines for the Care and Use of Laboratory Animals of Chang Gung University of Science and Technology. All efforts were made to minimize suffering. 2.3. OVA sensitization The procedures of epicutaneous sensitization for the mice were modified from previous studies (Kim et al., 2012; Yanaba et al., 2013). At first, the mice were sensitized with 1 mg/ml OVA (0.1 ml) by intraperitoneal administration every other day for 10 days. At day 8, the dorsal hair of the mice was shaved using an electric clipper and tape-stripped 6 times, mimicking cutaneous injury produced by scratching in AD patients. Next, 100 mg of OVA in 100 ml normal saline was pipetted onto a 1 1 cm sterile gauze, which was topically applied to the dorsal skin. Normal saline without OVA was employed as the vehicle control. The gauze was fixed with Tegaderm adhesive dressing (3M, Maplewood, MN, USA) and a Silkypore stretch bandage (Alcare, Tokyo, Japan). After 24 h, the gauze was removed and new gauze was applied. This procedure was repeated for 7 consecutive days.
The mouse skin was removed 24 h after the last challenge of OVA. The excised skin was suspended in PBS and homogenized by MagNA Lyser (Roche, Indianapolis, IN, USA). After centrifugation, levels of interleukin (IL)-4, IL-5, IL-10, and IL-13 in the tissue supernatant were measured using an enzyme immunoassay kit according to the manufacturer’s instructions (BioLegend). 2.8. Quantitative real-time PCR (qRT-PCR) The levels of filaggrin, involucrin, and integrin b were determined for their mRNA expression. Total RNA was extracted with Trizol (Invitrogen), and cDNA was synthesized by reverse transcription using an iScript cDNA Synthesis Kit (Bio-Rad). The qRT-PCR was performed using iQ SYBR Green Supermix (Bio-Rad) based on the manufacturer’s instructions. The housekeeping gene GAPDH was used as an internal control, and gene expression was calculated using the comparative threshold cycle method. 2.9. In vitro cutaneous permeation The in vitro drug permeation into/across the skin was conducted with the Franz cell. The excised OVA- or vehicletreated skin was mounted between the donor and receptor compartments with the stratum corneum (SC) facing upward toward the donor. The receptor medium was 30% ethanol in pH 7.4 buffer for tacrolimus, 8-methoxypsoralen, and methotrexate to maintain the sink condition. The pH 7.4 buffer was the receptor medium for FD4. The donor (0.5 ml) was loaded with tacrolimus (4% w/v in 30% PEG400/water), 8-methoxypsoralen (0.8% in 40% PEG400/pH 7.4 buffer), methotrexate (1% in water), or FD4 (0.56% in pH 7.4 buffer). The effective diffusion region for the permeants was 0.785 cm2. The receptor temperature and stirring rate of the stirrer were kept at 37 C and 600 rpm, respectively. A 300-ml medium was withdrawn from the receptor at determined intervals. After a 24-h application, the skin was removed from the Franz cell. The skin was washed by double-distilled water for three times to
Y.-P. Fang et al. / International Journal of Pharmaceutics 497 (2016) 277–286
remove the residual drug on skin surface. The drug deposition within the skin was extracted by methanol for tacrolimus, and 8methoxypsoralen and 0.1 N HCl for methotrexate and FD4, respectively. A glass homogenizer (Chin-Fa, Hsinchu, Taiwan) was used to extract the permeants from skin. The high-performance liquid chromatography (HPLC) methods for analyzing the three small-molecule drugs were described previously (Fang et al., 2008; Lin et al., 2010, 2015). FD4 was analyzed by fluorescence spectrophotometer as described by Fang et al. (2004). 2.10. In vivo cutaneous permeation A glass cylinder with a hollow area of 0.785 cm2 was attached to the mouse back by cyanoacrylate superglue. An aliquot of 0.2 ml FITC or FD4 in the same condition as in the in vitro experiment was pipetted into the cylinder. The administration period was 6 h. The animal was then sacrificed, and the treated skin area was excised. The skin species were monitored by fluorescence and confocal microscopies. The fluorescence microscopy was used to observe the vertical section of the skin. The biopsies were cut vertically, embedded in OCT, and frozen at 70 C. The samples were sectioned in a cryostat microtome and mounted with glycerin and gelatin. The slices were monitored by inverted microscope (IX81, Olympus) using a filter set at 450–490 and 515–565 nm for excitation and emission. A confocal laser scanning microscopy (TCS, SP2, Leica, Wetzlar, Germany) was utilized to observe the horizontal section of the skin. The thickness of the skin was scanned at 5 mm increments from the surface via the z-axis. Images were taken by summing 10 fragments at different depths.
279
2.11. Statistical analysis Statistical analysis of differences between the OVA and the vehicle control groups was performed using the Kruskal–Wallis test. The post hoc test for checking individual differences was Dunn’s test. All data were analyzed using WINKS statistical software (Texasoft, Cedar Hill, TX, USA). 3. Results 3.1. Macroscopic visualization and physiological parameters OVA was used to induce AD-like symptoms in the mouse skin. As shown in the left panel of Fig. 1A, the gross appearance of the vehicle-treated mouse skin exhibited an intact surface without disruption. After the repeated challenge by OVA, dryness, redness, edema, hemorrhage and excoriation could be detected on the skin surface as illustrated in the right panel of Fig. 1A. Fig. 1B shows the close-up pictures. As compared with the control skin, the OVAchallenged mouse skin displayed dry, scaly skin with a shiny and erythema tone. Some deep furrows were also observed on the ADlike skin. We next examined whether OVA treatment was associated with alterations in cutaneous physiology, including TEWL, erythema, and pH. As shown in Fig. 1C, the vehicle-treatment group revealed a constant water evaporation rate of 5.2 g/m2/h. The OVA-treated group had a significantly increased TEWL level (27.4 g/m2/h) compared to the control. OVA application resulted in a tendency toward an increase in a* as shown in Fig. 1D. After a treatment of OVA, the skin surface pH increased by about 1 unit as compared to the control (6.3 vs. 5.4) as depicted in Fig. 1E.
Fig. 1. Gross observation and physiological parameters of mouse skin treated with and without OVA: (A) the gross imaging by digital camera; (B) the close-up imaging by handheld digital microscope; (C) transepidermal water loss (TEWL); (D) erythema; and (E) skin surface pH value. The control group corresponds to the mice treated with the vehicle. The data of physiological assessment are presented as the mean of 12 experiments S.D. *, p < 0.05 between the control and AD-like groups.
280
Y.-P. Fang et al. / International Journal of Pharmaceutics 497 (2016) 277–286
Fig. 2. Histological examination and the levels of cytokines and differentiation proteins in mouse skin treated with and without OVA: (A) hemoxylin and eosin (H&E) staining; (B) expression of IL-4, IL-5, IL-10, and IL-13; AND (C) mRNA levels of filaggrin, involucrin, and integrin b. The data of cytokines are presented as the mean of 10– 14 experiments S.D. The data of differentiation proteins are presented as the mean of 4–5 experiments S.D.
Y.-P. Fang et al. / International Journal of Pharmaceutics 497 (2016) 277–286
3.2. Microscopic visualization and skin protein measurements The skin sections stained by H&E were examined to clarify whether OVA influenced cutaneous architecture. As demonstrated in the left panel of Fig. 2A, no pathological changes were found in the vehicle-treated group. OVA sensitization had thickened both the epidermal and dermal layers (the right panel of Fig. 2A). Some detachments of SC were observed in the AD-like lesion. The epidermis deepened into the upper layer of the dermis. There was a visible granular layer in the upper epidermis. We found an excessive infiltration of inflammatory cells into the dermis. These included lymphocytes, neutrophils, macrophages, and mast cells. IL-4, IL-5, IL-10, and IL-13 were analyzed to check the effect of OVA on the production of Th2 cytokines. The OVA-treated skin demonstrated a significant elevation (p < 0.05) of the cytokines compared with the normal skin as shown in Fig. 2B. OVA generally increased these cytokines by 2-fold. This result suggested the involvement of the Th2 immune response with the epicutaneous sensitization of OVA. Some skin proteins related to epidermal junctions were determined as shown in Fig. 2C. The result of qRTPCR demonstrated that mRNA levels of these proteins were significantly downregulated (p < 0.05) in the OVA group. OVA treatment lessened filaggrin, involucrin, and integrin b by about 3, 9-, and 4-fold, respectively. 3.3. In vitro cutaneous permeation We compared the in vitro skin delivery of tacrolimus, 8methoxypsoralen, methotrexate, and FD4 between intact and ADlike skins. Table 1 summarizes the partition coefficient (log P) and MW of the permeants employed in this study. Tacrolimus revealed the greatest lipophilicity, followed by 8-methoxypsoralen, methotrexate, and FD4. Fig. 3 depicts the skin uptake and flux across the skin of the four permeants. The flux was computed from the slope of the cumulative amount in the receptor–time curve. The flux predicted transdermal penetration to the systemic circulation, whereas the skin deposition suggested the drug uptake into the cutaneous reservoir. When tacrolimus was applied to the healthy and diseased skins, the diffusion amount into the receptor was below the detection limit of the analytical method. Fig. 3A shows the tacrolimus deposition within the skin. Tacrolimus accumulation increased significantly (p < 0.05) from 0.25 to 0.66 mg/mg by OVA sensitization. The ratio between the skin deposition of AD-like and normal skins is displayed above the column for AD-like skin in Fig. 3A (2.65). Contrary to the results of tacrolimus, AD-like skin deposition of 8-methoxypsoralen was lower by a factor of 0.6 as compared to that in intact skin as shown in Fig. 3B. On the other hand, increased 8-methoxypsoralen flux was achieved with the OVA treatment. The AD-like lesion significantly promoted (p < 0.05) the flux by 2.4-fold. Fig. 3C shows that methotrexate deposition into AD-like skin was about two folds higher (p < 0.05) than that into vehicle-treated skin. A great increment of methotrexate flux by 17.5-fold was detected for AD-like skin compared to normal skin. No deposition was formed for FD4 delivery in the control group. OVA application Table 1 Physicochemical properties of the permeants tested in this study. Permeant
Partition coefficient (log P)
Molecular weight (Da)
Tacrolimus 8-Methoxypsoralen Methotrexate FD4 FITC
3.3 1.7 0.9 2.0 2.0
822.0 216.2 454.4 4000 389.4
The physicochemical properties of all drugs were obtained from DrugBank website (www.drugbank.ca/).
281
could facilitate FD4 deposition from 0 to 0.46 mg/mg (Fig. 3D). OVA also largely enhanced the transdermal flux of FD4 compared to the control. The increment ratio of FD4 flux (17.8) was comparable to that of methotrexate flux. 3.4. In vivo cutaneous permeation Imaging techniques were employed to examine in vivo cutaneous distribution of fluorescent dyes in normal and lesional skins. FITC was classified as a small-molecule and hydrophilic model permeant (Table 1). No fluorescence was found in the control and AD-like skins without dye administration (data not shown). Fig. 4A presents the fluorescence images of the sectioned skin with or without OVA challenge. A faint green signal was observed in the superficial epidermis of the vehicle-treated skin with a discontinuous distribution. Some FITC molecules accumulated in hair follicles (arrow in the left panel of Fig. 4A), which acted as a reservoir. This indicates the penetration pathway of appendages for topically applied FITC. The fluorescence distribution from FITC was also limited to epidermis and hair follicles of AD-like skin. A deeper and stronger FITC signal was observed in AD-like skin than in the control skin (arrow in the right panel of Fig. 4A). As shown in the left panel of Fig. 4B, FD4 distribution was constrained to the follicles with very few signals in the epidermis. AD-like skin showed a broad distribution of FD4 in the upper epidermis (arrows in the right panel of Fig. 4B). The fluorescence in hair follicles was comparable in intact and diseased skins. Fig. 5 illustrates that the separate x,y-sections gained at each zaxis of the skin by confocal imaging. Each photograph is a planar image taken at every 5-mm depth from the skin surface. The last photograph is the summary of ten separate sections. Increased FITC fluorescence in AD-like skin compared to vehicle-treated skin can clearly be detected as shown in Fig. 5A. An extensive distribution of FITC was observed in lesional skin. A similar result was found in the case of topically applied FD4 (Fig. 5B). The FD4 signal, more pronounced with AD-like skin than with normal skin, could be seen in all skin layers examined. A deeper fluorescence signal was detected in the horizontal image but not in the vertical fluorescence image. This may be due to the higher resolution and sensitivity of confocal microscopy. 4. Discussion AD can be categorized into two forms: an extrinsic type associated with elevated immunoglobulin E (IgE) by environmental allergens and an intrinsic type without IgE sensitization. Approximately 70–80% of AD patients present the extrinsic type (Jin et al., 2009). It is assumed that Balb/c mice exposed to allergens such as OVA reveal the extrinsic form (Shiohara et al., 2004), reflecting most of AD patients. The OVA-challenged model exhibits the advantages of good reproducibility, a short time for AD-like lesion induction, and ease of quantitative evaluation. It is important to assess how the change of barrier function by AD influences skin permeability. Previous studies (Mildner et al., 2010; Mathes et al., 2014) have utilized an organotypic skin culture model to examine the permeation profiles of lucifer yellow and biotin for atopic skin. The dermatitis model developed in this study provided a platform for allowing the assessment of atopic skin permeability. This is the first investigation reporting the effect of AD-like lesion on percutaneous absorption of anti-AD drugs in animals. The classic findings to characterize AD are dryness, erythema, hemorrhage, and scaling. These symptoms were observed in our OVA-induced AD model. The scratching behavior was also revealed in our model. Scaling is a result of premature detachment of a cornified layer in the SC, which can produce a fragile skin structure
282
Y.-P. Fang et al. / International Journal of Pharmaceutics 497 (2016) 277–286
(A)
(C)
(B)
(D)
Fig. 3. Skin deposition and transdermal flux of the permeants into/across mouse skin treated with and without OVA: (A) tacrolimus; (B) 8-methoxypsoralen; (C) methotrexate; and (D) FD4. The data are presented as the mean of 4–5 experiments S.D.
(Kawasaki et al., 2012). TEWL can be a reflection of skin dryness and barrier function. The experimental result indicated an elevated TEWL by OVA challenge, suggesting the disruption of the
cutaneous barrier. The pH of SC is essential for controlling skin homeostasis and barrier function (Schreml et al., 2012). The skin surface pH in healthy humans ranges between 5.4 and 5.9 (Braun-
Y.-P. Fang et al. / International Journal of Pharmaceutics 497 (2016) 277–286
283
Fig. 4. Fluorescence microscopic images of mouse skin treated with and without OVA: (A) the skin topically applied with FITC and (b) the skin topically applied with FD4. Scale bar = 50 mm.
Fig. 5. Confocal microscopic images of mouse skin treated with and without OVA: (A) the skin topically applied with FITC and (b) the skin topically applied with FD4. Both fragments from the skin surface to the depth of 45 mm and a summary of 10 fragments at various skin depths are displayed in this figure. Scale bar = 300 mm.
Falco and Korting, 1986), approximating the skin pH of a normal mouse (5.4) in the present study. The pH value increased in AD-like lesions. This is a common phenomenon observed in AD patients (Levin et al., 2013). The elevation of pH is associated with the
enhanced protease activity and reduced SC lipid lamellae synthesis (Cork et al., 2009), causing the barrier dysfunction. The increase of SC pH and protease activity provides a mechanistic basis of desquamation, as observed in our H&E-stained histology. The
284
Y.-P. Fang et al. / International Journal of Pharmaceutics 497 (2016) 277–286
hyperplasia, keratosis, and influx of neutrophils, macrophages, and mast cells detected in microscopic visualization in response to combined tape stripping and OVA challenge also confirmed the successful dermatitis induction (Oyoshi et al., 2012). The overexpression of IL-4, IL-5, IL-10, and IL-13 was in line with the findings observed in AD. Th2 cytokines are dominant in the acute inflammation phase of AD (Savinko et al., 2012). Our model can be regarded as the acute skin lesion of AD. Th2 cytokines are proven to affect epidermal morphology and barrier characteristics. Overexpressed IL-4 and IL-13 lead to the reduction of filaggrin and involucrin (Hänel et al., 2013). Filaggrin and involucrin synthesized by the differentiating keratinocytes in the basal layer are associated with the cornified envelope in the SC (Danso et al., 2014). Filaggrin is a predominant structural protein in the SC, contributing to a critical role for SC barrier formation. It is reported that 50% of AD patients have the inherited genetic mutation that encodes filaggrin (Guttman-Yassky et al., 2011). The deficiency of filaggrin decreases natural moisture factors (NMFs) in the SC, which results in the increased TEWL and SC pH (Kezic et al., 2008). Our result was similar to this demonstration. Involucrin is a major precursor protein for the cornified envelope. The consequences of involucrin abnormality are barrier deterioration and SC lipid loss (Agrawal and Woodfolk, 2014). Integrin b in the basal layer mediates cell–cell and cell-extracellular matrix interactions, especially the epidermal–dermal adhesion (Has et al., 2012). The suppressed expression of these proteins caused the impairment of barrier function in the AD-like lesion in our model. Both the SC layer and tight junctions (TJs) in the epidermis contribute to the main barrier for drug penetration (Lehman and Franz, 2014). The elevation of TEWL and pH in AD-like skin implied SC damage. The SC abnormality in atopic skin relates to the diminished ceramide amount and impaired lipid lamellar secretion (Man et al., 2008; Sahle et al., 2015). The ceramides and fatty acids in atopic SC have a shorter alkyl chain length than healthy skin (Agrawal and Woodfolk, 2014), suggesting a larger space for permeant penetration. An effortless entrance of allergen and irritant into AD skin via the creation of gaps between the corneocytes has been previously reported (Cork et al., 2009). The facile penetration of the drugs into/across atopic skin can be anticipated through similar pathways. Our permeation profiles confirmed this supposition. The in vivo topical application examined by vertical and horizontal views demonstrated the extensive and deep permeation of small molecules and macromolecules in AD-like skin. Local immune response occurs mainly in the epidermal and dermal layers. It is desirable to target the antiAD drugs to deeper skin strata where inflammation occurs (Pople and Singh, 2013). The follicular route is an important pathway for hydrophilic permeants and macromolecules. According to fluorescence microscopic images, the AD-like lesion did not seem to affect follicular entrance. 8-Methoxypsoralen combined with ultraviolet A irradiation is effective in the treatment of severe chronic AD (Der-Petrossian et al., 2000). Although an AD lesion is expected to accelerate drug diffusion into the skin, it is surprising that 8-methoxypsoralen deposition was decreased with AD-like skin. The basic drug transport into the skin can be controlled by partitioning, solubility, and MW. Hadgraft (2001) has indicated that the ability of drug partitioning into the SC is proportional to the drug amount within the skin. The first principal step in the cutaneous delivery is the partitioning into the SC for the drugs with high lipophilicity such as 8-methoxypsoralen. Although the SC disruption could reduce the inherent barrier effect for promoting drug penetration, the partitioning to the disrupted SC should be decreased because of the deficient ceramides and fatty acids in atopic SC to reduce SC lipophilicity. Another possibility was that the hyperplasia of ADlike skin created a long-distance path for 8-methoxypsoralen
diffusion. Our previous study (Fang et al., 2008) had demonstrated that the pathway length largely influenced 8-methoxypsoralen transport. Further study is needed to confirm this hypothesis about the drug absorption decrease by lesional skin. Tacrolimus shows a higher log P than 8-methoxypsoralen; however, the OVA challenge increased skin uptake of tacrolimus. The partitioning cannot simply explain the permeation of lipophilic permeants through atopic skin. Tacrolimus is a macrolide immunosuppressant for the first-line drug therapy of AD (Svensson et al., 2011). The large molecular volume of tacrolimus (MW = 822 Da) may hinder the passage into/across intact skin. The expansive space for permeation due to ceramide and fatty acid deficiency was beneficial for macrolide penetration. Nevertheless, the magnification of the tacrolimus skin deposition increment by OVA treatment was only 2.7. This indicates that the difficulty of partitioning into AD-like skin still governed tacrolimus permeation. The magnitude of the flux enhancement by AD-like skin of hydrophilic permeants such as methotrexate and FD4 was much higher than that of lipophilic drugs. Oral methotrexate has been demonstrated to be efficient for treating moderate-to-severe AD and psoriasis (Deo et al., 2014). Topical delivery of methotrexate was already developed for treating psoriasis (Vemulapalli et al., 2008). The rate-limiting step for methotrexate transport is at the level of the SC due to its polar and ionic characteristics. SC and TJ damage in AD-like skin led to a great augmentation of methotrexate flux. Methotrexate penetration via AD-like skin displayed a higher enhancement in flux than skin deposition because the hydrophilic permeants can be forced into lipophilic SC compartments, and then quickly continue to the hydrophilic receptor medium (Nielsen, 2005). The lengthened path produced by hyperproliferative skin did not influence the methotrexate passage as shown in the previous report (Lee et al., 2008). FD4 penetration into/across vehicle-treated skin was negligible because of the high hydrophilicity and MW. Both the SC and TJs manifest the obstruction for macromolecule delivery (Hung et al., 2015). Once the skin was disturbed by AD, the macromolecule penetration could be largely increased. This is reasonable since the lesional skin of AD favors the entrance of high-molecular size structures, including bacteria and allergens (Bieber, 2010). It is a general concept that the transport pathways for lipid-soluble molecules through the SC are mainly the extracellular lipid bilayers, whereas the intracellular keratin and interfacial corneocyte cell envelope are the primary routes for water-soluble molecule permeation (Norlén, 2013). Our result suggests that OVA affected both pathways to change the cutaneous drug absorption. The more hydrophilic permeants exhibited higher passage enhancement by AD-like skin. The SC is a skin layer presenting a lipophilic nature. Percutaneous absorption of more hydrophilic drugs would be more affected by skin with a compromised SC barrier. Our result indicates that the SC damage was a principal factor influencing drug permeation via AD-like skin. TJ disruption in viable skin may be the secondary factor. The increased absorption should help in reaching effective drug concentration at the target site; however, it also results in the increased possibility of drug penetration across the skin and into the bloodstream. This risk may be severe for methotrexate since AD-like skin largely facilitated methotrexate transport into the Franz cell receptor. The systemic use of methotrexate is known to elicit some adverse effects such as hepatotoxicity, nausea, anemia, and thrombocytopenia (Schneider et al., 2013). Methotrexate for topical application should be used carefully on lesional skin. The side effects of systemic 8-methoxypsoralen are gastrointestinal and central nervous system disturbances (Der-Petrossian et al., 2000). Although the AD lesion increased only 8-methoxypsoralen flux by a factor of 2.4, this lipid-soluble drug should still be used with caution. Tacrolimus showed a neglected flux in AD-like skin.
Y.-P. Fang et al. / International Journal of Pharmaceutics 497 (2016) 277–286
This suggests a low threat of tacrolimus systemic absorption. However, a 2.7-fold increase in tacrolimus skin deposition by atopic skin may generate the local side effects of tacrolimus such as cutaneous burning, itching, pruritus, and pain. Although currently there is no evidence of the risk of anti-AD drug overabsorption, monitoring after long-term use and the modulation of the applied dose may be greatly needed. Topical administration on large areas of lesional skin should be avoided. 5. Conclusion The incidence of AD and the substantial burden of this disorder are increasing worldwide. Drug therapy for AD is restricted due to the development based on chance observation and the insufficient understanding of drug delivery/targeting. This report introduces an animal model with AD-like lesions, opening up chances to investigate the cutaneous permeation of anti-AD drugs. This model was proved to be predictable, reproducible, and low cost. The representations of Th2 cytokines and differentiation proteins in this AD model were similar to those in AD patients. The experimental result manifests the barrier dysfunction and the increase of drug and macromolecule absorption by OVA sensitization. It was noticeable that 8-methoxypsoralen in the skin reservoir was decreased in AD-like skin more so than in intact skin. This may be due to the lesser amount of partitioning that occurs with a lipid-deficient SC. This suggests that the lesional skin was not necessary to elicit permeation enhancement. There remains a need to develop new anti-AD drugs with low adverse effects. Most studies have employed intact skin to assess drug absorption for developing new drugs. The results from such investigation are not suitable for simulating the actual condition. Application of the AD model was feasible to measure drug absorption for approximating clinically diseased status. This model may also expand its application to other lesional skins. This platform allows us to test large numbers of treatments that cannot be performed in patients due to ethical considerations. Acknowledgment The authors are grateful for the financial support from Chang Gung Memorial Hospital (CMRPD1D0432-3). References Agrawal, R., Woodfolk, J.A., 2014. Skin barrier defects in atopic dermatitis. Curr. Allergy Asthma Rep. 14, 433. Andersen, L.K., Davis, M.D.P., 2013. The epidemiology of skin and skin-related diseases: a review of population-based studies performed by using the Rochester Epidemiology Project. Mayo Clin. Proc. 88, 1462–1467. Bieber, T., 2010. Atopic dermatitis. Ann. Dermatol. 22, 125–137. Braun-Falco, O., Korting, H.C., 1986. Normal pH value of human skin. Hautarzt 37, 126–129. Cork, M.J., Danby, S.G., Vasilopoulos, Y., Hadgraft, J., Lane, M.E., Moustafa, M., Guy, R. H., MacGowan, A.L., Tazi-Ahnini, R., Ward, S.J., 2009. Epidermal barrier dysfunction in atopic dermatitis. J. Invest. Dermatol. 129, 1892–1908. Danso, M.O., van Drongelen, V., Mulder, A., van Esch, J., Scott, H., van Smeden, J., El Ghalbzouri, A., Bouwstra, J.A., 2014. TNF-a and Th2 cytokines induce atopic dermatitis-like features on epidermal differentiation proteins and stratum corneum lipids in human skin equivalents. J. Invest. Dermatol. 134, 1941–1950. Deo, M., Yung, A., Hill, S., Rademaker, M., 2014. Methotrexate for treatment of atopic dermatitis in children and adolescents. Int. J. Dermatol. 53, 1037–1041. Der-Petrossian, M., Seeber, A., Hönigsmann, H., Tanew, A., 2000. Half-side comparison study on the efficacy of 8-methoxypsoralen bath-PUVA versus narrow-band ultraviolet B phototherapy in patients with severe chronic atopic dermatitis. Br. J. Dermatol. 142, 39–43. Fang, J.Y., Fang, C.L., Liu, C.H., Su, Y.H., 2008. Lipid nanoparticles as vehicles for topical psoralen delivery: solid lipid nanoparticles (SLN) versus nanostructured lipid carriers (NLC). Eur. J. Pharm. Biopharm. 70, 633–640. Fang, J.Y., Lee, W.R., Shen, S.C., Wang, H.Y., Fang, C.L., Hu, C.H., 2004. Transdermal delivery of macromolecules by erbium:YAG laser. J. Control. Release 100, 75–85.
285
Guttman-Yassky, E., Nograles, K.E., Krueger, J.G., 2011. Contrasting pathogenesis of atopic dermatitis and psoriasis—part I: clinical and pathologic concepts. J. Allergy Clin. Immunol. 127, 1110–1118. Hadgraft, J., 2001. Modulation of the barrier function of the skin. Skin Pharmacol. Physiol. 14 (Suppl. 1), 72–81. Hänel, K.H., Cornelissen, C., Lüscher, B., Baron, J.M., 2013. Cytokines and the skin barrier. Int. J. Mol. Sci. 14, 6720–6745. Has, C., Spartà, G., Kiritsi, D., Weibel, L., Moeller, A., Vega-Warner, V., Waters, A., He, Y., Anikster, Y., Esser, P., Straub, B.K., Hausser, I., Bockenhauer, D., Dekel, B., Hildebrandt, F., Bruckner-Tuderman, L., Laube, G.F., 2012. Integrin a3 mutations with kidney, lung, and skin disease. N. Engl. J. Med. 366, 1508–1514. Hay, R.J., Johns, N.E., Williams, H.C., Bolliger, I.W., Dellavalle, R.P., Margolis, D.J., Marks, R., Naldi, L., Weinstock, M.A., Wulf, S.K., Michaud, C., Murray, C.J.L., Naghavi, M., 2014. The global burden of skin disease in 2010: an analysis of the prevalence and impact of skin conditions. J. Invest. Dermatol. 134, 1527–1534. Hung, C.F., Chen, W.Y., Aljuffali, I.A., Lin, Y.K., Shih, H.C., Fang, J.Y., 2015. Skin aging modulates percutaneous drug absorption: the impact of ultraviolet irradiation and ovariectomy. Age 37, 21. Jin, H., He, R., Oyoshi, M., Geha, R.S., 2009. Animal models of atopic dermatitis. J. Invest. Dermatol. 129, 31–40. Kawasaki, H., Nagao, K., Kubo, A., Hata, T., Shimizu, A., Mizuno, H., Yamada, T., Amagai, M., 2012. Altered stratum corneum barrier and enhanced percutaneous immune responses in filaggrin-null mice. J. Allergy Clin. Immunol. 129, 1538–1546. Kezic, S., Kemperman, P.M., Koster, E.S., de Jongh, C.M., Thio, H.B., Campbell, L.E., Irvine, A.D., McLean, W.H., Puppels, G.J., Caspers, P.J., 2008. Loss-of-function mutations in the filaggrin gene lead to reduced level of natural moisturizing factor in the stratum corneum. J. Invest. Dermatol. 128, 2117–2119. Kim, S.R., Choi, H.S., Seo, H.S., Choi, Y.K., Shin, Y.C., Ko, S.G., 2012. Topical application of herbal mixture extract inhibits ovalbumin- or 2,4-dinitrochlorobenzeneinduced atopic dermatitis. Evid. Based Complement. Altern. Med. 545497. Lee, W.R., Shen, S.C., Fang, C.L., Zhuo, R.Z., Fang, J.Y., 2008. Topical delivery of methotrexate via skin pretreated with physical enhancement techniques: lowfluence erbium:YAG laser and electroporation. Lasers Surg. Med. 40, 468–476. Lehman, P.A., Franz, T.J., 2014. Effect of induced acute diabetes and insulin therapy on stratum corneum barrier function in rat skin. Skin Pharmacol. Physiol. 27, 249–253. Levin, J., Friedlander, S.F., Del Rosso, J.Q., 2013. Atopic dermatitis and the stratum corneum. Part I: the role of filaggrin in the stratum corneum barrier and atopic skin. J. Clin. Aesthet. Dermatol. 6, 16–22. Lin, Y.K., Huang, Z.R., Zhuo, R.Z., Fang, J.Y., 2010. Combination of calcipotriol and methotrexate in nanostructured lipid carriers for topical delivery. Int. J. Nanomed. 5, 117–128. Lin, Y.K., Yang, S.H., Chen, C.C., Kao, H.C., Fang, J.Y., 2015. Using imiquimod-induced psoriasis-like skin as a model to measure the skin penetration of anti-psoriatic drugs. PLoS One 10, e0137890. Man, M.Q., Hatano, Y., Lee, S.H., Man, M., Chang, S., Feingold, K.R., Leung, D.Y.M., Holleran, W., Uchida, Y., Elias, P.M., 2008. Characterization of a hapten-induced, murine model with multiple features of atopic dermatitis: structural, immunologic, and biochemical changes following single versus multiple oxazolone challenges. J. Invest. Dermatol. 128, 79–86. Mathes, S.H., Ruffner, H., Graf-Hausner, U., 2014. The use of skin models in drug development. Adv. Drug Deliv. Rev. 69–70, 81–102. Mildner, M., Jin, J., Eckhart, L., Kezic, S., Gruber, F., Barresi, C., Stremnitzer, C., Buchberger, M., Mlitz, V., Ballaun, C., Sterniczky, B., Fodinger, D., Tschachler, E., 2010. Knockdown of filaggrin impairs diffusion barrier function and increases UV sensitivity in a human skin model. J. Invest. Dermatol. 130, 2286–2294. Miyagaki, T., Sugaya, M., 2015. Recent advances in atopic dermatitis and psoriasis: genetic background, barrier function, and therapeutic targets. J. Dermatol. Sci. 78, 89–94. Nielsen, J.B., 2005. Percutaneous penetration through slightly damaged skin. Arch. Dermatol. Res. 296, 560–567. Norlén, L., 2013. Current understanding of skin barrier morphology. Skin Pharmacol. Physiol. 26, 213–216. Oyoshi, M.K., He, R., Li, Y., Mondal, S., Yoon, J., Afshar, R., Chen, M., Lee, D.M., Luo, H.R., Luster, A.D., Cho, J.S., Miller, L.S., Larson, A., Murphy, G.F., Geha, R.S., 2012. Leukotriene B4-driven neutrophil recruitment to the skin is essential for allergic skin inflammation. Immunity 37, 747–758. Pople, P.V., Singh, K.K., 2013. Development and evaluation of colloidal modified nanolipid carrier: application to topical delivery of tacrolimus, part II—in vivo assessment, drug targeting, efficacy, and safety in treatment for atopic dermatitis. Eur. J. Pharm. Biopharm. 84, 72–83. Sahle, F.F., Gebre-Mariam, T., Dobner, B., Wohlrab, J., Neubert, R.H.H., 2015. Skin diseases associated with the depletion of stratum corneum lipids and stratum corneum lipid substitution therapy. Skin Pharmacol. Physiol. 28, 42–55. Savinko, T., Matikainen, S., Saarialho-Kere, U., Lehto, M., Wang, G., Lehtimäki, S., Karisola, P., Reunala, T., Wolff, H., Lauerma, A., Alenius, H., 2012. IL-33 and ST2 in atopic dermatitis: expression profiles and modulation by triggering factors. J. Invest. Dermatol. 132, 1392–1400. Schneider, L., Tilles, S., Lio, P., Boguniewicz, M., Beck, L., LeBovidge, J., Novak, N., 2013. Atopic dermatitis: a practice parameter update 2012. J. Allergy Clin. Immunol. 131, 295–299. Schreml, S., Meier, R.J., Albert, M.G., Seidl, U., Zeller, V., Behm, B., Landthaler, M., Abels, C., Babilas, P., 2012. The impact of 10% a-hydroxy acid emulsion on skin pH. Skin Pharmacol. Physiol. 25, 34–38.
286
Y.-P. Fang et al. / International Journal of Pharmaceutics 497 (2016) 277–286
Shiohara, T., Hayakawa, J., Mizukawa, Y., 2004. Animal models for atopic dermatitis: are they relevant to human diseases? J. Dermatol. Sci. 36, 1–9. Svensson, A., Chambers, C., Gånemo, A., Mitchell, S.A., 2011. A systematic review of tacrolimus ointment compared with corticosteroids in the treatment of atopic dermatitis. Curr. Med. Res. Opin. 27, 1395–1406. Vemulapalli, V., Yang, Y., Friden, P.M., Banga, A.K., 2008. Synergistic effect of iontophoresis and soluble microneedles for transdermal delivery of methotrexate. J. Pharm. Pharmacol. 60, 27–33.
Wollenberg, A., Wetzel, S., Burgdorf, W.H., Haas, J., 2003. Viral infections in atopic dermatitis: pathogenic aspects and clinical management. J. Allergy Clin. Immunol. 112, 667–674. Yanaba, K., Kamata, M., Asano, Y., Tada, Y., Sugaya, M., Kadono, T., Tedder, T.F., Sato, S., 2013. CD19 expression in B cells regulates atopic dermatitis in a mouse model. Am. J. Pathol. 182, 2214–2222.