Gelatin/Chitosan Bilayer Patches Loaded with Cortex Phellodendron amurense/Centella asiatica Extracts for Anti-Acne Application

Acne is a chronic inflammatory skin disease that often occurs with anaerobic Propionibacterium acnes (P. acnes). Anti-acne patches, made of hydrocolloid or hydrogel, have become a popular way of topical treatment. The outer water-impermeable layer of commercial patches might create hypoxic conditions and promote P. acnes growth. In this study, gelatin/chitosan (GC) bilayer patches were prepared at different temperatures that included room temperature (RT), −20 °C/RT, and −80 °C/RT. The most promising GC bilayer patch (−80 °C /RT) contained a dense upper layer for protection from bacteria and infection and a porous lower layer for absorbing pus and fluids from pimples. The anti-acne bilayer patch was loaded with Cortex Phellodendri amurensis (PA) and Centella asiatica (CA) extracts. PA extract could inhibit the growth of P. acnes and CA extract was reported to improve wound healing and reduce scar formation. Moreover, the water retention rate, weight loss rate, antibacterial activity, and in vitro cytotoxicity of the patches were investigated. The porous structure of the patches promoted water retention and contributed to absorbing the exudate when used on open acne wounds. The GC bilayer patches loaded with PA/CA extracts were demonstrated to inhibit the growth of P. acnes, and accelerate the skin fibroblast cell viability. Based on their activities and characteristics, the GC bilayer patches with PA/CA extract prepared at −80 °C/RT obtain the potential for the application of acne spot treatment.


Introduction
Acne is a chronic inflammatory skin disease associated with comedones, papules, pustules, and nodules. It is a common problem affecting many teenagers. Acne pathogenesis is multifactorial and includes increased sebum and inflammation [1]. Propionibacterium acnes colonizes pilosebaceous follicles and is a major factor in the inflammatory reaction that occurs with acne vulgaris. P. acnes is a Gram-positive human skin commensal microorganism that prefers anaerobic growth conditions and is involved in acne pathogenesis [2]. P. acnes produces lipases, proteases, hyaluronidases, and neutrophil chemotactic factors,

Preparation of Herbal Extracts
PA and CA (~100 g) were air-dried, ground into a powder, and soaked in 90% (v/v) methanol for 4 h at room temperature. The supernatant from the methanol extraction was then filtered through Whatman filter paper. The filtrate was evaporated and lyophilized to obtain the dry plant extract.

Cytotoxicity Assay of PA and CA Extracts
Human skin fibroblasts (CCD-SH68) were used to determine the cytotoxicity of the herbal extracts. The cells were cultured at a density of 1 × 10 4 cells/well in a 96-well, flatbottomed tissue culture plate, with different concentrations ranging from 9.5-300 µg/mL of PA and 75-2400 µg/mL of CA extracts at 37 • C for 48 h. Three replicates were considered for each extract concentration. The cell viability was determined via a WST-1 cell proliferation colorimetric assay (BioVision, Milpitas, CA, USA). 100 µL/well of WST-1 reagent was added and incubate for 37 • C for 30 min, and the absorbance at 440 nm was measured.

Disk Diffusion Analysis of Herbal Extracts
The herbal extracts were initially screened for antibacterial activity by disk diffusion analysis. The bacterial inoculums (Staphylococcus aureus ATCC 25923, Pseudomonas aeruginosa ATCC 27853, and Escherichia coli ATCC 25922) were adjusted to the 0.5 McFarland standard and spread onto a sterile Luria-Bertani (LB) agar plate. A disk containing 60 µL of herbal extract at a concentration of 15 mg/mL was placed on the LB agar-seeded plate and incubated at 37 • C for 16 h, then the inhibition zone diameter was measured. Propionibacterium acnes (BCRC 10723) was incubated on a blood agar plate under anaerobic conditions.

Determination of Minimum Inhibitory Concentrations
The minimum inhibitory concentration (MIC) of the methanolic herbal extract against the bacteria was determined using the twofold serial dilution broth method. Each MIC was determined in triplicate, and the mean values are reported. The 96-well plates were scanned using an ELISA reader at 540 nm. The MIC was determined as the lowest concentration of methanolic extract that caused an optical density reduction of more than 90% when compared with the control.

Preparation of Gelatin/Chitosan Solution Loaded with PA/CA Extracts
The chitosan powder was dissolved in 0.5% v/v acetic acid to prepare a 1% w/w solution. The gelatin solution (12.5% w/w) was prepared by dissolving gelatin powder in deionized water at 40 • C. Next, 5 mL of 1% w/w chitosan solution, 1 mL of 12.5% w/w gelatin solution, 2 mL of PA extract at the concentrations of 0.5 and 1 mg/mL, and 2 mL of CA extract at the concentrations of 1.2 and 2.4 mg/mL were loaded into the solution. the dish after moving the dish to room temperature for 1 h. The patch was dried at room temperature for 48 h before being crosslinked with 10% glutaraldehyde for 1 h at 40 • C. The patch was then cut into a round shape with an 8 mm diameter using a steel mold.

Preparation of
2.6.3. Bilayer Patches Constructed at −80 • C/Room Temperature 5 mL of the herbal extract containing the gelatin/chitosan solution was poured into an 85 mm diameter dish, frozen at −80 • C for 12 h, and lyophilized for 48 h (VirTis, Warminster, PA, USA). Then, the solution was poured into the dish after moving the dish to room temperature for 1 h. The patch was dried at room temperature for 48 h before being crosslinked with 10% glutaraldehyde for 1 h at 40 • C. The patch was cut into a round shape with an 8 mm diameter using a steel mold. Dry patches (monolayer GC patch, bilayer GC patch, −20 • C/RT, bilayer GC patch, −80 • C/RT) were coated with a gold layer and examined via scanning electron microscopy (SEM, Hitachi S3000V, Tokyo, Japan). Histograms of the diameters of 50 individual pores were generated from the SEM images. Three samples were examined per group. The effective sizes of the pores were calculated as the mean patch diameters.

Fourier Transform Infrared Spectroscopy
The composition of the bilayer GC patch was analyzed by using Fourier transform infrared spectroscopy (FTIR, Spectrum Two, PerkinElmer) in the range of 4000-400 cm −1 using KBr pellets.

Water Retention Assay
1 milliliter of distilled H 2 O was added to each weighed patch (monolayer GC patch, bilayer GC patch, −20 • C/RT, bilayer GC patch, −80 • C/RT), and the mixtures were then incubated for 3, 6, 12, 24, and 48 h at room temperature. The samples were blotted dry before being weighed. The water retention (%) was calculated as (Wt-Wi)/Wi × 100%, where Wi is the specimen's initial dry weight and Wt is the specimen's weight after submersion in distilled H 2 O for a set time.

Weight Loss Assay
1 milliliter of distilled H 2 O was added to each weighed patch (monolayer GC patch, bilayer GC patch, −20 • C/RT, bilayer GC patch, −80 • C/RT), and the samples were then incubated for 3, 6, 12, 24, and 48 h. Each sample was removed and gently blotted with filter paper, then lyophilized and weighed. The weight loss (%) was calculated as (Wi-Wd)/Wi × 100%, where Wi is the specimen's initial dry weight and Wd is the specimen's weight after submersion in distilled H 2 O for a set time. 2.7.5. Drug Release Assay 1 mg of bilayer GC patch (−80 • C/RT) which contained PA extract at the concentration of 1 mg/mL, and CA extract at the concentration of 2.4 mg/mL were soaked in 1 mL of distilled water for 0.5, 1, 2, 4, 6, 8, 10, and 12 h respectively. Each sample was then analyzed via UV-Vis spectroscopy at 325 nm. The concentration of drug release was calculated by using the formula of absorbance against concentration. The drug release rate was calculated as followed: Drug release rate (%) = the amount of released drug (mM)/the amount of drug in the GC patch (mM) 2.7.6. In Vitro Cell Viability Assay of GC Patches The bilayer GC patch (−80 • C/RT) with different concentrations of PA and CA extract samples were soaked individually in 1 mL of the culture medium for 3, 6, 12, 24, and 48 h. A GC patch with no herbal extract was used as the control. The cells were cultured in triplicate with the patch-conditioned medium at a density of 1 × 10 4 cells/well in a 96-well, flat-bottomed tissue culture plate. The cell viability was determined via a WST-1 cell proliferation colorimetric assay.

Antibacterial Activity Assay of GC Patches
The antibacterial activity of the bilayer GC patch (−80 • C/RT) and commercial antiacne patches was also determined by using disk diffusion analysis. The patches were placed on the LB agar-seeded plate with S. aureus, P. aeruginosa, and E. coli, respectively, and incubated at 37 • C for 16 h, then the inhibition zone diameter was measured. P. acnes was incubated on a blood agar plate under anaerobic conditions. The inhibition zone diameters were measured after 48 h. Each assay was performed in triplicate and repeated twice. The commercial anti-acne patches were purchased from 3M, MENTHOLATUM, and MAYSKIN.

Skin Irritation Test
A skin irritation test was conducted on New Zealand white rabbits (n = 3) one day after shaving their dorsal. The test patch was applied to the back skin for 4 h. After removing the patch, the remaining substance was wiped away. The subsequent scoring of signs of irritation including redness, swelling, cloudiness, edema, hemorrhage, and discharge was scored at 1, 24, 48, and 72 h and based on the ISO10993-10 [27]. The experimental protocol was approved by the Animal Studies Committee of Central Taiwan University of Science and Technology.

Statistical Analysis
All quantitative data are presented as the mean ± standard deviation. Statistical analyses were performed using Student's t-test or one-way analysis of variance, followed by a post hoc Fisher's least significant difference test for multiple comparisons. Differences were deemed significant at p < 0.05.

Results and Discussion
3.1. Cytotoxicity and Antibacterial Assay of PA/CA Extracts 3.1.1. Cytotoxicity Assay of the PA/CA Extract Human skin fibroblasts (CCD-SH68) were used to determine the cytotoxicity of the herbal extracts. The CCD-SH68 fibroblasts were incubated with various concentrations of CA and PA for 48 h ( Figure 1). With the addition of CA extract, the cell viability of fibroblasts increased by up to 142%. However, at the higher concentration of 1200 µg/mL, the CA extract demonstrated an antiproliferative effect on the cultured human skin fibroblasts ( Figure 1A). Based on the result of the cytotoxicity assay, concentrations of CA extracts in the range of 75-600 µg/mL were used for further experiments.
In the range of 75-150 µg/mL, PA extract obtained no cytotoxicity. Furthermore, the concentration of PA extract between 9.5-38 µg/mL not only showed no cytotoxic effects but also exhibited the effect to promote cell viability up to160%. When the concentration increased to 300 µg/mL, the PA extract demonstrated an antiproliferative effect on the cultured human skin fibroblasts ( Figure 1B).

Antibacterial Assay of the PA/CA Extract
A disk containing 60 μL of CA or PA extract at 15 mg/mL was placed on the LB agarseeded plate and incubated at 37 °C for 16 h, then the inhibition zone diameter was measured. P. acnes was incubated on a blood agar plate under anaerobic conditions. The disc with 60 μL of 15 mg/mL of PA extract showed strong antimicrobial activity against P. acnes; the inhibition zone diameter was 55 ± 2.25 mm. The PA extract inhibited the growth of P. acnes, with a MIC of 150 μg/mL (data are not shown). The PA extract also showed antimicrobial activity against S. aureus. The inhibition zones for the disks with 60 μL of 15 mg/mL of CA were <15 mm against P. acnes, S. aureus, P. aeruginosa, and E. coli. The CA extract showed little antibacterial activity.

Microscopic Morphological Observation
The surface and cross-sectional morphologies of the monolayer patches (RT) showed a smooth and dense structure (Figure 2A, B). Figure 2C shows several cracks on the bilayer patch surface (−20 °C/RT), but the vertical section remained intact ( Figure 2D). Figure 2E-G showed bilayer patches (−80 °/RT) with cracked surfaces on the upper layer and porous structures on the lower layer. The mean pore size in the loose layer was 105.1 ± 2.89 μm ( Figure 2G). Figure 2H shows that the bilayer GC patches are 8 mm in diameter and 0.5 mm high. Because the patches are porous and thin, they were easily penetrated by air. The bilayer's dense structure may serve as a protective barrier to eliminate dirt and microorganisms.

Antibacterial Assay of the PA/CA Extract
A disk containing 60 µL of CA or PA extract at 15 mg/mL was placed on the LB agar-seeded plate and incubated at 37 • C for 16 h, then the inhibition zone diameter was measured. P. acnes was incubated on a blood agar plate under anaerobic conditions. The disc with 60 µL of 15 mg/mL of PA extract showed strong antimicrobial activity against P. acnes; the inhibition zone diameter was 55 ± 2.25 mm. The PA extract inhibited the growth of P. acnes, with a MIC of 150 µg/mL (data are not shown). The PA extract also showed antimicrobial activity against S. aureus. The inhibition zones for the disks with 60 µL of 15 mg/mL of CA were <15 mm against P. acnes, S. aureus, P. aeruginosa, and E. coli. The CA extract showed little antibacterial activity.

Microscopic Morphological Observation
The surface and cross-sectional morphologies of the monolayer patches (RT) showed a smooth and dense structure (Figure 2A,B). Figure 2C shows several cracks on the bilayer patch surface (−20 • C/RT), but the vertical section remained intact ( Figure 2D). Figure 2E-G showed bilayer patches (−80 • /RT) with cracked surfaces on the upper layer and porous structures on the lower layer. The mean pore size in the loose layer was 105.1 ± 2.89 µm ( Figure 2G). Figure 2H shows that the bilayer GC patches are 8 mm in diameter and 0.5 mm high. Because the patches are porous and thin, they were easily penetrated by air. The bilayer's dense structure may serve as a protective barrier to eliminate dirt and microorganisms.  patches not only smoothed the patch's surface but also functioned as a protective b 3.2.3. Weight Loss Assay Figure 3B displays the weight loss of the various GC patches. After soaking i tilled H2O for 3 to 48 h, the weight loss of the 3 kinds of GC patches ranged from 7 74% ( Figure 3B). Crosslinking the GC patches with glutaraldehyde at 40 °C for 60 m only kept the GC patches stable in water for 48 h but also enabled the GC patches t ciently release the drug. When the process of glutaraldehyde crosslinking lasted l than 1 h, the GC patches could not release enough of the drug or produce antiba activity (data not shown).    Figure 3B displays the weight loss of the various GC patches. After soaking in distilled H 2 O for 3 to 48 h, the weight loss of the 3 kinds of GC patches ranged from 70% to 74% ( Figure 3B). Crosslinking the GC patches with glutaraldehyde at 40 • C for 60 min not only kept the GC patches stable in water for 48 h but also enabled the GC patches to efficiently release the drug. When the process of glutaraldehyde crosslinking lasted longer than 1 h, the GC patches could not release enough of the drug or produce antibacterial activity (data not shown).

Drug Release Assay
The drug release rate of bilayer GC patches containing 2.4 mg/mL of CA extract and 1 mg/mL of PA extract, respectively, is shown in Figure 4. The CA release profile of the GC patches accelerated during the first 1 h, which then gradually increased after 1 h of soaking ( Figure 4). The release profile of PA accelerated during the first hour and then became constant over subsequent hours (Figure 4). After 12 h of soaking, the drug release rates for PA and CA were 42% and 21%, respectively. When preparing the herbal extract solution, the PA extract was more soluble in water than the CA extract. It might explain that the drug release efficiency of the PA-containing GC patches is higher than that of CA -containing GC patches.

Drug Release Assay
The drug release rate of bilayer GC patches containing 2.4 mg/mL of CA extract and 1 mg/mL of PA extract, respectively, is shown in Figure 4. The CA release profile of the GC patches accelerated during the first 1 h, which then gradually increased after 1 h of soaking ( Figure 4). The release profile of PA accelerated during the first hour and then became constant over subsequent hours (Figure 4). After 12 h of soaking, the drug release rates for PA and CA were 42% and 21%, respectively. When preparing the herbal extract solution, the PA extract was more soluble in water than the CA extract. It might explain that the drug release efficiency of the PA-containing GC patches is higher than that of CA -containing GC patches.

In Vitro CytotoxicityAssay and Antibacterial Assay of GC Patches Containing Different Concentrations of PA/CA Extracts
Based on the results of cell viability assay and antibacterial assay, the ideal anti-acne patch should release 150 μg/mL of PA extract to inhibit the growth of P. acnes, and 75-600 μg/mL of CA extract to promote fibroblast proliferation. In Figure 5A, the antibacterial assay showed the diameters of inhibition zones against P. acnes. Both of the GC patches containing PA/CA extracts exhibited an antibacterial effect against P. acne.
For the cytotoxicity assay ( Figure 5B), various GC patches were soaked in 1 mL of medium for 6 h. After removing the GC patches, 1 × 10 4 fibroblasts were cultured with the soaking medium for 48 h. The cells cultured in the medium that was soaked with GC patches containing 1 mg/mL of PA and 2.4 mg/mL of CA obtained the best cell proliferation rate and increased to 1.51 × 10 4 . The result revealed that the prepared GC patches

In Vitro Cytotoxicity Assay and Antibacterial Assay of GC Patches Containing Different Concentrations of PA/CA Extracts
Based on the results of cell viability assay and antibacterial assay, the ideal anti-acne patch should release 150 µg/mL of PA extract to inhibit the growth of P. acnes, and 75-600 µg/mL of CA extract to promote fibroblast proliferation. In Figure 5A, the antibacterial assay showed the diameters of inhibition zones against P. acnes. Both of the GC patches containing PA/CA extracts exhibited an antibacterial effect against P. acne.

Comparison of Antibacterial Activity of GC Patches and Commercial Anti-Acne Patches
The antibacterial activity of the GC bilayer patches was compared with that of the commercial anti-acne patches. Besides P. acnes, the most common causative microorganisms associated with wound infections include Staphylococcus aureus (S. aureus), Pseudomonas aeruginosa (P. aeruginosa), and Escherichia. coli (E. coli) were also tested. Figure 6A shows that the GC patches with PA/CA extract had a 26 mm inhibition zone diameter; thus, they demonstrated the most effective inhibition of P. acnes. In Figure 6B, the commercial antiacne patches A and C showed no antibacterial activity against P. acnes, S. aureus, P. aeruginosa, or E. coli. Commercial anti-acne patch B, which contained chlorhexidine diacetate, exhibited inhibition zone diameters of 21.75 ± 0.23, 13.25 ± 0.38, and 11.5 ± 0.27 mm against P. acnes, S. aureus, and E. coli, respectively. In the comparison of antibacterial activity, the For the cytotoxicity assay ( Figure 5B), various GC patches were soaked in 1 mL of medium for 6 h. After removing the GC patches, 1 × 10 4 fibroblasts were cultured with the soaking medium for 48 h. The cells cultured in the medium that was soaked with GC patches containing 1 mg/mL of PA and 2.4 mg/mL of CA obtained the best cell proliferation rate and increased to 1.51 × 10 4 . The result revealed that the prepared GC patches were biocompatible. Moreover, with the addition of PA/CA extracts, the patches could accelerate fibroblast cell viability.
The GC patch with 1 mg/mL of PA and 2.4 mg/mL of CA extracts not only showed strong antimicrobial activity against P. acnes, and also promoted cell viability. It was processed for the following experiments.

Comparison of Antibacterial Activity of GC Patches and Commercial Anti-Acne Patches
The antibacterial activity of the GC bilayer patches was compared with that of the commercial anti-acne patches. Besides P. acnes, the most common causative microorganisms associated with wound infections include Staphylococcus aureus (S. aureus), Pseudomonas aeruginosa (P. aeruginosa), and Escherichia. coli (E. coli) were also tested. Figure 6A shows that the GC patches with PA/CA extract had a 26 mm inhibition zone diameter; thus, they demonstrated the most effective inhibition of P. acnes. In Figure 6B, the commercial anti-acne patches A and C showed no antibacterial activity against P. acnes, S. aureus, P. aeruginosa, or E. coli. Commercial anti-acne patch B, which contained chlorhexidine diacetate, exhibited inhibition zone diameters of 21.75 ± 0.23, 13.25 ± 0.38, and 11.5 ± 0.27 mm against P. acnes, S. aureus, and E. coli, respectively. In the comparison of antibacterial activity, the GC patches with 1 mg/mL of PA and 2.4 mg/mL of CA extracts performed most effectively compared to three other commercial patches.

Skin Irritation Test of the GC Bilayer Patches
The GC patches were applied to the back skin of New Zealand white rabbits. After 24 h, the GC patches were removed, and no signs of irritation responses appeared on the rabbits' back skin (Figure 7).

Skin Irritation Test of the GC Bilayer Patches
The GC patches were applied to the back skin of New Zealand white rabbits. After 24 h, the GC patches were removed, and no signs of irritation responses appeared on the rabbits' back skin (Figure 7).

Discussion
Most anti-acne patches are hydrocolloid or hydrogel stickers with a waterproof surface to protect pimples from secondary infection. The commercial circular hydrocolloid patches can absorb fluid from the pimple and make the pimples flatten out. Some patches contain triclosan, salicylic acid, or chlorhexidine diacetate as anti-acne ingredients which might cause hypersensitivity, including general allergic reactions [19,28,29]. Thus this research aims to develop an anti-acne patch based on natural materials and obtain the antiacne effect without irritation properties.
In this study, the bilayer anti-acne patch made of gelatin and chitosan was crosslinked with glutaraldehyde to form a stable complex. The FTIR analysis by Zhang et al. revealed that the increasing intensity of absorption peak at 1659 cm −1 suggested that glutaraldehyde as a cross-linker could react with the amide group on chitosan/gelatin scaffolds and form a stable complex [30]. The extracts of PA and CA loaded in the GC patches encouraged the interaction of phenolic compounds with NH2 and hydroxyl groups in gelatin, which thus led to the crosslinking of the gelatin matrix [31,32]. It was also proved that the addition of CA extract could form hydrogen bonds with related functional groups of gelatin. [33]. The FTIR analysis of various GC bilayer patches were performed ( Figure  S1A). The difference of various herbal extracts (PA+CA, CA, and PA) loaded in the bilayer GC patches were analyzed ( Figure S1B). Both PA and CA extracts are not pure compounds; multiple peaks showed at 2600-2700 cm −1 . It confirmed the loading of herbal extract in the GC patches.
After soaking in water, the hydrophilic PA and CA extract also could be efficiently released into the water.
The waterproof surface of commercial anti-acne patches might create hypoxic environments and benefit the growth of P. acnes. In the manufacturing process of the GC patches, the freeze-dried process created a porous structure, and the average pore size was 105.1 ± 2.89 nm. The porous structure allows oxygen to penetrate through and create

Discussion
Most anti-acne patches are hydrocolloid or hydrogel stickers with a waterproof surface to protect pimples from secondary infection. The commercial circular hydrocolloid patches can absorb fluid from the pimple and make the pimples flatten out. Some patches contain triclosan, salicylic acid, or chlorhexidine diacetate as anti-acne ingredients which might cause hypersensitivity, including general allergic reactions [19,28,29]. Thus this research aims to develop an anti-acne patch based on natural materials and obtain the anti-acne effect without irritation properties.
In this study, the bilayer anti-acne patch made of gelatin and chitosan was crosslinked with glutaraldehyde to form a stable complex. The FTIR analysis by Zhang et al. revealed that the increasing intensity of absorption peak at 1659 cm −1 suggested that glutaraldehyde as a cross-linker could react with the amide group on chitosan/gelatin scaffolds and form a stable complex [30]. The extracts of PA and CA loaded in the GC patches encouraged the interaction of phenolic compounds with NH2 and hydroxyl groups in gelatin, which thus led to the crosslinking of the gelatin matrix [31,32]. It was also proved that the addition of CA extract could form hydrogen bonds with related functional groups of gelatin. [33]. The FTIR analysis of various GC bilayer patches were performed ( Figure S1A). The difference of various herbal extracts (PA+CA, CA, and PA) loaded in the bilayer GC patches were analyzed ( Figure S1B). Both PA and CA extracts are not pure compounds; multiple peaks showed at 2600-2700 cm −1 . It confirmed the loading of herbal extract in the GC patches.
After soaking in water, the hydrophilic PA and CA extract also could be efficiently released into the water.
The waterproof surface of commercial anti-acne patches might create hypoxic environments and benefit the growth of P. acnes. In the manufacturing process of the GC patches, the freeze-dried process created a porous structure, and the average pore size was 105.1 ± 2.89 nm. The porous structure allows oxygen to penetrate through and create a condition unfavorable for the growth of anaerobic P. acne. The sponge-like lower layer of the GC patches also obtained good water-absorb characteristics and made the GC patches capable of absorbing the exudate from the pimple. The other (upper) layer is also made of chitosan and gelatin but dried at room temperature. Without the freeze-drying process, the architecture of the upper layer is more compact. The compact structure could improve the mechanical property of the bilayer patch, and support the GC patch and prevent it from curling up. The upper layer might also act as a barrier to keep the pimple from dirt and bacteria.
The natural ingredients, PA and CA extracts, were added to the GC patches to perform anti-acne and healing effects. The PA extract exhibited antimicrobial activity against P. acnes, with a MIC of 150 µg/mL. The CA extract ranging from 75 to 600 µg/mL can accelerate fibroblast cell viability and collagen synthesis. Based on different drug release rates, GC patches were incorporated with 1 mg/mL of PA and 2.4 mg/mL of CA. The original volume for one GC patch was 0.25 mL, which means one GC contained 0.25 mg of the PA and 0.6 mg of the CA extracts. Both extracts were in the ranges of effective and non-cytotoxic dosages.
The results of the antibacterial assay and cytotoxicity assay demonstrated the bilayer GC patches containing 1 mg/mL of PA and 2.4 mg/mL of CA could release an effective dosage of PA and CA extracts to inhibit P. acnes growth and promoted cell viability. In Figure 5B, the commercial anti-acne patch B contained chlorhexidine diacetate and exhibited inhibition zone diameters of 21.75 ± 0.23 against P. acnes. Chlorhexidine diacetate is a disinfectant and topical anti-infective agent. Its mechanism of action involves the destabilization of the outer bacterial membrane. Chlorhexidine diacetate contains chlorhexidine, which might induce hypersensitivity, including general allergic reactions [34,35]. To prevent the possible side effects, herbal extract (PA) was added to the GC patches instead of topical disinfectant, and showed the most effective antibacterial activity against P. acnes. The biocompatibility of the GC patches with CA/PA was confirmed by the irritation test performed on the animal models.
Usage of herbal medicine has increased many folds on account of the side effects observed with conventional drugs. In this study, the combination of PA and CA extracts were added to the bilayer patches which were made of chitosan and gelatin. The manufactured bilayer GC patches that obtained effective antibacterial activity, promoted cell viability, high exudate-absorbing ability, and appropriate drug release rate, thus have massive potential in the anti-acne application.

Conclusions
The GC bilayer patch that was constructed at −80 • C/room temperature with the addition of 1 mg/mL of PA and 2.4 mg/mL of CA demonstrated effective antibacterial activity and accelerated skin fibroblast cell viability. The porous lower layer provided the GC bilayer patch with a good water retention rate of up to 150% after 3 h of soaking and may be useful for absorbing exudate from open acne wounds. The dense upper layer thinned the surface, enabling this layer to function as a protective barrier. These properties make the cytocompatible GC bilayer with PA/CA extract an excellent candidate for anti-acne spot treatments.

Patents
A patent issued in Taiwan (patent number I626054) was obtained for this novel product.