Nanomicelle-generating Microneedles Loaded With Tranilast for Treatment of Hypertrophic Scars in a Rabbit Model

Abstract

Background/Aim: Hypertrophic scars (HS) are the result of pathological wound healing characterized by a red, raised scar formation. The goal of this research was development of a new method for treatment of HS formation.

Materials and Methods: A tranilast-loaded microneedle (TMN) was developed and applied in a rabbit ear model to treat an induced HS. Scar elevation index, the thickness of dorsal skin by hematoxylin and eosin staining, collagen deposition by Masson trichrome staining and expression of myofibroblast biomarker proteins were evaluated.

Results: The 12×12 array of the TMN containing 2.9 μg tranilast per needle released more than 80% of the drug within 30 min. During the procedure, control, non-loaded MN and TMN loaded with three different doses of tranilast (low: 2.5-3, medium: 25-30, and high: 100-150 μg) were applied to the HS in rabbit ears. High-level TMN led to a clear and natural appearance of skin, a decrease in scar elevation index by 47% and decline in the thickness of the epidermis from 69.27 to 15.92 μm when compared to the control group. Moreover, the collagen density also decreased in groups treated with medium- or high-level TMNs, by 10.2% and 9.06%, respectively. Furthermore, the expression of transforming growth factor-β, collagen-1, and α-smooth muscle actin proteins was reduced in TMN-treated HSs compared to the control.

Conclusion: The findings show the overall efficacy of TMNs in inhibiting HS. Thus, use of TMN is a simple and cosmetic remedy for HS, with good protection and reliability.

The management of cutaneous scars has focused primarily on practitioners’ experience, instead of on outcomes from large-scale randomized, controlled, and evidence-based trials. Hypertrophic scars (HSs) can have severe physical, psychological, and social effects (1). In recent decades, increased treatment of acute burn cases has minimized deaths and the percentage of mortality has been reduced (2). Owing to physical and psychosocial deprivation, the quality of life of the patients with HSs declines dramatically. HSs cause not only a variety of health problems such as scratching, discomfort, and functional problems, but also emotional distress due to the deformity (3). The incidence of HS is around 70% following a burn injury and it induces neuropathic pain, surface abnormalities and skin rigidity, which subsequently lead to physiological dysfunction (4). Although the exact etiology of scar formation is not well known, HS causes a permanent and increased level of inflammation at the wound site (5).

A variety of treatments and procedures are currently being suggested to minimize the occurrence of HS by using medications (for example, retinoids, corticosteroids, and 5-fluorouracil injections), operative and noninvasive treatments (cryotherapy and electrosurgery), and laser and radiofrequency therapies (6). Therapeutic drugs have significantly improved HS treatment, however, it difficult to adjust the dose appropriately since higher doses cause cytotoxicity and tissue necrosis (7). Even applying a suitable dose of triamcinolone resulted in adverse effects including subcutaneous and skin fat atrophy, and hypopigmentation. In addition, interlesional injection may lead to lymphogenous uptake of corticosteroid crystals, thereby producing pain. Furthermore, it is suggested that combinational therapy is more suitable than individual drug therapy in HS treatment. Cryotherapy was recently addressed in some studies with corticosteroids for improvement of keloid scar (8-10). However, it has not been examined in control of HS. Dermal radiation therapy is another promising approach for reducing the density of collagen, but it has major complications for pregnant and young patients. In addition, the utilization of high-energy radiation induced cutaneous malignancies in irradiated skin of patients (11).

Recently, tranilast [N-(3,4-dimethoxycinnamoyl) anthranilic acid] has been used as an anti-allergic drug which reduces fibrotic conditions in keloids or HS. Pharmacologically, it inhibits the release of inflammatory cytokines from monocytes and macrophages and has anti-fibrotic effect through suppressing collagen formation and accumulation (12,13). Additionally, it controls pathological fibrosis including pulmonary, hepatic, muscle, and ventricular fibrosis. However, despite its potential effects in the treatment of keloids and HS, the poor aqueous solubility of tranilast limits its pharmaceutical formulation and clinical application (14,15). Moreover, the high dose required for oral administration often raises concerns of systemic adverse effects, including liver toxicity, skin rash and abdominal discomfort (16). Drug delivery through the transdermal route has advantages over the oral administration of tranilast, since transdermal drug delivery may achieve relatively higher concentrations with reduced systemic adverse effects. However, the profile of low drug absorption through the skin is still a crucial factor for application; various approaches have been examined to enhance the permeation of the drug, including iontophoresis and the use of nanoparticles (17).

During the past decade, microneedles (MNs) have attracted much attention due to their capability for effective transdermal drug delivery (18-20). MNs are arrays of needles of several hundred micrometers, suitable for penetrating the stratum corneum of the skin in a minimally invasive manner (21). MNs can also be self-administered by patients with limited supervision, which reduces the need for medical expert participation during treatment (18). MNs are considered particularly useful for various scar treatments in aesthetic medicine (22). It was reported that drug-free MNs exhibited an inhibitory effect on keloid and normal fibroblast proliferation, suggesting that the application of MNs potentially suppresses the formation/growth of abnormal scars and normalizes the architecture of the connective tissues of a mature HS (18,23). Dissolving MNs loaded with bleomycin were used for the treatment of hypertrophic scar, inhibiting the growth of fibroblasts at the applied area (22,24). In our previous study, we developed a dissolving MN which can generate self-assembled nanomicelles upon the application of MNs to the skin (25). A tranilast-loaded MN (TMN) can be readily fabricated by casting in a silicone mold. The matrix of the TMN is composed of a biocompatible amphiphilic triblock copolymer, Pluronic F127, and tranilast, which readily dissolve in the interstitial fluid. The dissolution of the triblock copolymer leads to the formation of nanomicelles by self-assembly, followed by the localization of tranilast in the hydrophobic core of the micelles. In this study, the ability of nanomicelle-generating TNMs for the prevention of scar formation and enhancement of the overall physical appearance of the skin was assessed.

Materials and Methods

Materials. Pluronic F127(F127), polyethyleneglycol (PEG, Mw 6,000), polyvinylpyrrolidone (PVP, Mw 40,000) and polyvinyl alcohol (PVA, Mw 90,000) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Tranilast was purchased from Dalim Biotech Company (Seoul, Republic of Korea). High-quality chemicals were used in this research.

Production of TMNs. TMNs were produced by a micro-molding method as described in a previous article (26). Briefly, Pluronic F127 and tranilast (at a ratio of F127/tranilast=19:1) were dissolved in a cosolvent of methanol and acetone (1:1). The organic solvents were then removed to form a thin film in a round-bottomed flask using a rotary evaporator under reduced pressure. The film was then rehydrated by adding an aqueous solution containing PEG (at a ratio of F127/PEG=7:3) during sonication. After filtration through a 0.8-μm filter, the resulting solution (100 μl) was placed on 0.8×0.8 cm polydimethylsiloxane mold at room temperature (Figure 1A). To form the backing layer, 100 μl of a solution containing PVP and PVA was layered on top of the mold. The mold was then dried at 37˚C for 24 h under reduced pressure. After drying, the TMN was detached from the mold using adhesive tape and stored at room temperature in a desiccator before use. A length of 525 μm was chosen to ensure that TMNs would successfully perforate the scarred stratum corneum for effective transdermal delivery of tranilast. The morphologies and dimensions of the fabricated TMN were observed by scanning electron microscopy (JSM-7600F; JEOL, Tokyo, Japan). The TMN had a square pyramidal shape with a height of 525.0 μm (Figure 1B). The lengths of one side of the rectangular base and the needle pitch were 200 μm and 500 μm, respectively.

Figure 1. A: A schematic representation of the fabrication of a tranilast-loaded microneedle (TMN) array (A). B: Scanning electron microscopy images of a TMN array. MeOH: Methanol, PEG: polyethylene glycol, PVP: poly(vinylpyrrolidone), PVA: polyvinyl alcohol.

The loading content and efficiency of drug delivery were determined by UV-Vis spectrometry as 4.4% and 88.0%, respectively. The amount of tranilast loaded in TMNs was measured using an UV-Vis spectrophotometer at 330 nm and was found to be 2.9 μg tranilast in each MN, so that 417.6 μg of tranilast was loaded in each TMN array patch.

Characterization of tranilast-loaded nanomicelles from TMNs. Nanomicelles containing tranilast were generated by dissolving a TMN in PBS (pH 7.4). A transmission electron microscopy device (JEM-3010; ZEOL, Tokyo, Japan) was used for observing morphology of the nanomicelles. The solution (10 μl) was dropped onto a copper grid and allowed to stand for 10 min. A 1% diluted aqueous solution of uranyl acetate (5 μl) was added to the grid for 10-15 s and dried. Transmission electron microscopy images were acquired at an accelerating voltage of 100 kV. The size distribution of the nanomicelles in an aqueous solution was determined by dynamic light scattering (Malvern Nano ZS, Malvern, UK).

In vitro release of tranilast from TMNs. To determine the in vitro release profile of tranilast from TMNs, a TMN array containing 416 μg of tranilast was immersed in 1 ml of phosphate-buffered saline (PBS) buffer and incubated at 37˚C. Samples of the solution were collected at predetermined time intervals (0, 5, 10, 15, 20, 30, 45, 60, 90, 120, 150, and 180 min) and 1 ml of PBS buffer was replaced with the same volume of new PBS buffer at each collection. The amount of the drug released from TMNs was determined using a UV-Vis spectrophotometer.

In vivo rabbit model of HS and gross examination. The animal study was approved by the Bundang Hospital Animal Experimental Center Animal Care and Use committee of the Seoul National University under the BA-1911-284-086-01. The methodology of National Institutes of Health Animal Care and Use guidelines were followed (27). Four young adult, New Zealand White female rabbits weighing between 3 and 4 kg were used for this study. A successful development of a HS model from Kloeters et al. (28) was established in the current research. Briefly, rabbits were anesthetized using ketamine and xylazine before creating wounds on the ears. The bare cartilage on the ventral surface of the left ear of each rabbit of received five wounds of around 12 mm in diameter. The epidermis, dermis, perichondrium, and cartilage were carefully removed using a dermal punch, surgical needle, and scissors to successfully create a HS. Two weeks were allowed for the formation of HS wounds before treatment with TMN. Five treatment groups were created: one scar served as the control without treatment, one scar for non-loaded MNs and another three scars for TMNs loaded with three different doses of tranilast (low: 2.5-3, medium: 25-30, and high: 100-150 μg), to be applied throughout the treatment period without removal. were used. The MN system was mounted above the scar and fixed by 3M Tegaderm^® translucent tape. Rabbits were under regular surveillance during the post-surgical period to record signs of microbial contamination and epithelialization. After 3 weeks of treatment, the rabbits were sacrificed to harvest the ears. Figure 2 presents the time schedule of HS formation and TMN treatment of induced HSs.

Figure 2. The time and workflow of an in vivo model in ears of rabbits of induced hypertrophic scars (HS) treated with tranilast-loaded microneedles (TMNs). After implantation, the TMN was maintained for 3 weeks at the wound site followed by harvesting.

Quantification of HS by histological assessment. The scars from three rabbits were harvested, fixed in formalin, followed by sequential dehydration in alcohol series and xylene, eventually paraffin embedded. The samples were sectioned at approximately 5 μm thickness and analysed using hematoxylin and eosin staining. By calculating the Scar Elevation Index (SEI) under 10× magnification, the scar elevation was quantified. The SEI is the ratio between the height of tissue in the overall area of injury and the height of tissue next to the scar (28). Subsequently, the epidermal thickness was also evaluated. Eight different locations were selected on the images from each slide and the mean average values were recorded.

Organization of collagen of the dermis. The paraffin-embedded tissue block was cut about 5 μm thickness and stained with Masson trichrome. The Masson trichrome staining procedure was followed to examine the pattern of collagen organization in a semi-quantitative manner by monitoring alignment and thickness (29). Collagen fibers stained blue in Masson trichrome staining, while cell cytoplasm was stained red. Muscle fibers and cell nuclei were marked in pink and dark brown, respectively. Random photographs of selected regions were taken and the relative percentage of collagen in eight different locations of the same lesions was measured using ImageJ software (National Institute of Health, Bethesda, MD, USA). The data represent the mean average value with standard deviation.

Western blot analysis. The western blot experiment was conducted for α-smooth muscle actin (α-SMA), collagen I (COL1), and transforming growth factor β1 (TGF-β1) expression. The tissues obtained were lysed with a cell lysis buffer and the standard protocol stated by Mahmood and Yang (30) was followed. A standard bicinchoninic acid method was used to conduct the protein assay, and subsequently, the protein samples were prepared for immunoblotting. Mouse anti-α-SMA (cat. No. ab7817, 1:300; Abcam, Cambridge, UK), anti-GAPDH (cat. no. sc32233, 1:1,000; Santa Cruz Biotechnology, Dallas, TX, USA), mouse anti-COL1 (cat. no. ab90395, 1:1,000; Abcam), mouse anti-TGF-β1 (cat. no. ab190503, 1:1,000; Abcam) were used as primary antibodies, while as a secondary antibody, goat anti-mouse IGG H&L (cat. no. ab6789, 1:5,000; Abcam) was used. The levels of protein expression were measured by an image analyzer to analyze the signals recorded by the nitrocellulose membranes used in immunoblotting (Las-3000; Photo Fuji, Tokyo, Japan). The images were taken using a GE Healthcare Amersham Imager 600 (GE Healthcare USA, Chicago, IL, USA). Densitometric analysis was performed with the program Image Quant TL (Cytiva, Marlborough, MA, USA). Image J program (National Institute of Health, Bethesda, MD, USA) was used to determine the expression of target proteins and normalize them to the expression of glyceraldehyde 3-phosphate dehydrogenase.

Statistical analysis. Student t-test was used for statistical analysis using SPSS version 18 (SPSS Inc., Chicago, IL, USA) with the statistical significance level of p<0.05. The experiments were performed five times and the values quoted are the mean±standard errors of the means.

Results

For improved transdermal delivery of tranilast, which has limited water solubility, we designed dissolving MNs which generate nanomicelles upon injection into the skin (25). The formation of tranilast-loaded nanomicelles facilitates the dissolution and delivery of the drug in the dermis. PEG was used to improve the mechanical properties of the MNs and to facilitate the formation of the triblock copolymer micelles in the intradermal fluid by reducing the tendency to form hydrogen bonds between F127 and water molecules (31). TNMs were prepared by a two-step process: Film casting and rehydration, followed by molding of the TNM in a polydimethylsiloxane mold.

In vitro release of tranilast from TMNs. The dissolution of a TMN leads to the formation of nanomicelles by self-assembly, resulting in tranilast being localized in the hydrophobic core of the nanomicelles. The nanomicelles exhibited a spherical morphology with an average diameter of 25.2±6.7 nm (Figure 3A). The matrix of TMN, consisting of hydrophilic F127 and PEG, readily dissolves upon contact with water. Owing to the small size and large surface area of the nanomicelles, tranilast should be efficiently released and dissolved in an aqueous medium. The in-vitro release experiment showed that more than 80% of tranilast was released from TMN within 30 min (Figure 3B). Since nanomicelles with a diameter of less than 50 nm can migrate through the interstitial space in the epidermis and dermis (32), the nanomicelles containing tranilast would not remain deposited at the site of injection, but probably be distributed over the region of the scar tissue. This may also minimize the occurrence of undesired adverse effects potentially caused by a high local concentration of the drug.

Figure 3. A: The size and morphology of nanomicelles generated by the dissolution of tranilast-loaded microneedles. The size distribution (plot) and image (inset) of nanomicelles were observed by light scattering method and transmission electron microscopy, respectively. B: In vitro release profile of tranilast from tranilast-loaded microneedles (n=3).

Macroscopic observation of HSs in the rabbit ear model. In the rabbit ear HS model, we tested the effectiveness of MN therapy. Wounds produced in the ears by punch biopsies showed raised tissues at the site of injury. This enabled successful recapitulation of HS in the model. On the 35^th day, rabbits were sacrificed, and the macroscopic scars were observed, sampled and sections photographed. A clear red scar, rough to palpation with a dense region at the center of the induced HS was noted in the control, where no medication was given. The use of non-loaded MNs led to a slightly reduced redness and smoother surface when compared to the control (Figure 4A). Treatment with TMNs led to clear skin with no dense region at the center and a more natural skin color.

Figure 4. Characteristics of hypertrophic scars. A: Appearance. B Scar elevation index (SEI) for each scar under each treatment and controls. C: Scar elevation index experimental groups as a percentage of that of the untreated control. The data shown are means±standard error of the means. **Significantly different from the untreated control at p<0.01.

The SEI represents the degree of hyperplasia of a scar. A reduction in the SEI in ears treated with non-loaded MNs was also noted when compared to the controls, with a drop of 11.6%. Based on the obtained results, it can be seen that while the medium-level (25-30 μg) TMNs reduced the SEI value significantly, the high-level (100-150 μg) TMNs provided the maximum reduction in the SEI (Figure 4B and C).

Assessment of the anti-fibrotic effect of TMN in the rabbit ear model of HS in vivo. The healing process was assessed by estimating the scar width and epidermal and dermis thickness after 3 weeks of treatment. In the current study, the fibers of collagen were observed to be thick, disturbed, and disordered in the controls. (Figure 5A). The epidermis thickness decreased with increasing tranilast loading. The non-loaded MNs reduced thickness by 2.21% when compared to the control. In particular, the high-level TMNs significantly reduced the thickness by 22.98% compared to the control (Figure 5B), indicating the efficiency of this approach in inhibition of tissue growth at the developed wound site.

Figure 5. The effects of treatments of hypertrophic scar in dorsal skin tissue. A: Hematoxylin and eosin staining of tissue sections from scars treated with non-loaded microneedles or with tranilast-loaded microneedles (TNM), and untreated controls. B: Measurement of epidermal thickness among experimental groups. The values represent the mean±standard error of the mean (n=8 replicates). ***Significantly different from the untreated control at p<0.001.

Evaluation of collagen density. In the control under the microscope, Masson trichrome stain indicated very dense and disordered collagen fibers (Figure 6A). Sections from ear HSs treated with TMNs showed organized collagen. However, HSs treated with non-loaded MNs had thick collagen fibers when compared to TMN-treated HSs. The level of collagen expression was measured and compared to that of the control. The result showed that the degree of collagen expression non-significantly decreased compared to the control (Figure 6B). HSs treated with low-level TMNs showed a smooth texture of collagen fibers, in contrast to samples from the other groups.

Figure 6. Reduction of collagen density during treatment of hypertropic scars in dorsal skin tissue. A: Masson trichrome staining of tissue sections from scars treated with non-loaded microneedles (MN) or with tranilast-loaded MNs revealed differences in the pattern of collagen fiber deposition. B: Collagen density as a percentage of the total dermal area. The values represent the mean±standard error of the mean (n=8 replicates).

Western blot evaluated the expression of myofibroblast biomarker proteins α-SMA, COL1, and TGF-β1. This analysis enabled further understanding of the underlying mechanism for the reduction of epidermal and dermis thickness and collagen density in the rabbit HS model by TMN treatment. In the current study, the expression of α-SMA, COL1, and TGF-β1 was greater in the control and progressively decreased in HS treated with non-loaded MNs and different TMNs (Figure 7A). This indicates a decrease in the expression of these biomarker proteins in the TMN-treated rabbit models. In HSs treated with high-level TMNs, the expression of these proteins strongly declined. The relative expression was calculated and is presented in Figure 7B. Little variation in the expression of the marker proteins was noted between the medium- and high-level TMN-treated HSs (p>0.05). In the case of α-SMA protein expression, the non-loaded MN led to a reduction of 57.8%, and low-, medium- and high-level TMNs reduced expression by 72.4%, 85.1%, and 84.9%, respectively, which were statistically significantly different from the control group. A decline in the expression of these marker proteins in the TMN-treated groups indicates the anti-inflammatory activity of the TMNs.

Figure 7. Suppression of myofibroblast biomarker proteins in hypertropic scars generated on rabbit ears and then treated with tranilast-loaded microneedles. A: Western blot analysis of expression of transforming growth factor-β1 (TGF-β1), collagen-1 (COL1), α-smooth muscle actin (SMA) and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) proteins. B: The bar graph illustrates the densiometric expression levels of myofibroblast marker proteins. The data shown are means±standard error of the means. Significantly different from the untreated control at: *p<0.05, **p<0.01 and ***p<0.001.

Discussion

A HS is variable in its intensity and morphology depending on the type of injury (18). HS treatment is usually difficult because its pathogenesis is only partly understood (33). Patients are generally not satisfied regarding the current treatment choices for abnormal scars such as keloid and HS (34). In this research, we generated a scar injury without harming the surrounding cartilage to reveal an observable HS on day 35 of post-intervention. Although the rabbit ear HS model from excision wounds is well known, the approach for HS analysis still lacks a cost-effective, dependable process (5,26). In several animal models, an applied burn and its scar was studied, each having advantages and inconveniences, and thus the model should be carefully selected. Thus, the study selected an HS model in the rabbit ear via excisional wounds to make the cartilage act as a splint, preventing wound contracture, and simulating the tensile strength that can be found in humans (26,35). One of the key advantages of using this model is the ability to generate and treat several wounds/scars per ear, which increases the number of scars examined without increasing the number of animals required (5).

Tranilast is commonly used in the treatment of pulmonary diseases, hepatic fibrosis, and cancer, owing to its anti-inflammatory properties (14,36), suggesting the potential application of the drug for the treatment of HS. It inhibits fibroblast function through suppression o collagens synthesis and accumulation. However, due to its poor aqueous solubility and profile of low skin absorption, the transdermal application of tranilast for HS is often limited. MN loading with drugs for treating dermatological scarring diseases has created much attention in recent times (37). In our previous study, a nanomicelle-generating MN was developed for simple formulation of poorly water-soluble drugs which can be applied through a transdermal route (24).

In this study, tranilast was readily incorporated into the MN matrix, consisting of an amphiphilic triblock copolymer, Pluronic F127, by a simple film casting/rehydration and molding processes (Figure 1A). The application of the TMN to the skin results in the rapid dissolution of the MN matrix, followed by the formation of nanomicelles containing tranilast in the epidermis. The small nano-scale size of the tranilast-loaded nanomicelles allows the efficient migration of micelles through the epidermal tissue, and drug release from the nanomicelles can also be achieved. The high-water content of scarred areas may also facilitate the efficient release of tranilast from these nanomicelles. The result is similar to another study in which bleomycin-loaded MNs had been applied for the treatment of HS (21). In addition to no pain and blood during the injection, the rapid dissolving nature of TMNs enables self-application by a patient without the aid of healthcare professionals and significantly reduces the application period of the patch, which would improve patient compliance and quality of life.

A better understanding of the molecular mechanisms responsible for unexpected wound healing and the formation of the HS might assist in the creation or development of new therapeutic procedures. Thus, in this study, we used different concentrations of tranilast and developed a MN to assess its implications versus a control on the development of HS formation and expression of myofibroblast marker proteins. A previous study showed implantation of MNs significantly reduced the SEI (0.80±0.14) compared to an untreated control (1.62±0.14) since it inhibited tissue growth at the wound site (18). In the present study, use of TMNs considerably reduced the height of the scar, thickness of the epidermis, and dermis compared to the untreated control group since tranilast released from MNs may reduce fibroblast activities at the wound site. The TMN treatment also revealed that, in comparison to control tissues, collagen fibers were thinner and wider and oriented parallel to the surface of the skin. Fibroblasts play a crucial role in collagen synthesis and deposition (38). However, tranilast has the ability to inhibit fibroblast function, reducing collagen deposition at wound sites (39). We believe that MNs with burst release of tranilast caused fibroblast toxicity and reduced collagen formation at wound sites.

TGF-β1, which promotes collagen synthesis and proteoglycan formation, is known to encourage HS formation by influencing the extracellular matrix (40). Tranilast, an antagonist of histamine, was shown to reduce collagen synthesis in keloid fibroblasts by suppressing TGF-β1 release (41). The diverse variety of microcellular actors and processes form an integral part of the pathophysiological foundation of HS development, which consists of stimulation of dysfunctional connective tissue (such as fibroblasts, keratinocytes and myofibroblasts), biosynthesis, growth, and development, as well as growth regulators and cytokines (33). The transformation of fibroblasts to myofibroblasts results in increased risk of scar formation from injury due to high pressure and contraction caused by myofibroblasts (42). The consistency of treated scars that demonstrated a crater in MN-treated wounds in our study may reflect potent inhibition of the migration/proliferation of fibroblasts. Moreover, TGF-β1 was shown to induce production of α-SMA by myofibroblasts, giving them increased collagen synthesizing capability and resistance to apoptotic inducers (43). However, our current results showed non-loaded MNs still had considerable protein expression. Initially, MNs had been applied for skin rejuvenation since they demonstrated the capability for skin regeneration (44). This technique influences the expression of proteins responsible for skin regeneration Hofny et al. showed that MNs containing platelet-rich plasma may favor the increase of TGF-β1. MN therapy showed slightly deposition of collagen and elastin (45). In addition, it enhanced the connection of cell layers in the granular lamina to the epidermis which was probably due to COL1 expression (46). Despite the MN technique favoring protein expression, the drug-loaded MNs led to significantly lower protein expression, making them efficient for HS treatment.

Several reports indicated that tranilast stops the fibroblast cell cycle at the G₀/G₁ phase and then inhibits fibroblast development (41,47,48). Inhibition of scar contraction by tranilast is known to be linked to the effects of TGF-β1 inhibition (49). Furthermore, it is noted that the inhibitory activity of tranilast in the contraction of a scar is also mainly associated with effects on myofibroblasts rather than fibroblasts. The use of tranilast inhibited expression of protein markers of myofibroblasts. The expression of myofibroblast proteins did not significantly differ between MNs with medium (25-30 μg) and high concentrations (100-150 μg) of tranilast in treating in the rabbit ear model of HS in this study. The lower expression of these proteins is expected to correlate with an anti-fibrotic tendency.

Conclusion

In this study, tranilast, a poorly water-soluble anti-inflammatory drug, was successfully formulated into a dissolving MN which had therapeutic effects for 3 weeks. The significant reduction of important biomarkers TGF-β1, COL1, and α-SMA in rabbits with wounds in their ears highlights the therapeutic potential of TMNs for the treatment of HS. While the current findings suggest that the application of MNs is successful, it should be further tested on larger clinical cohorts.

Conflicts of Interest

The Authors declare no conflicts of interest.

Authors’ Contributions

Conceptualization: C. Y. Heo and Ji H. Jeong; methodology: S. Y. Nam, S. Y. Lim. and Jae H. Jeong; validation: P. N. Chien; formal analysis: S. Y. Nam, P. N. Chien, S. Y. Lim and N. V. Long; data curation: Jae H. Jeong and P. N. Chien; writing: S. Y. Nam and Jae H. Jeong, X. R. Zhang; supervision: Ji H. Jeong and C. Y. Heo. All Authors have read and agreed to the published version of the article.

Acknowledgements

This work was supported by Institute of Information & communications Technology Planning & Evaluation (IITP) grant funded by the Korea government (MSIT) (No.2020-0-00990, Platform Development and Proof of High Trust & Low Latency Processing for Heterogeneous·Atypical·Large Scaled Data in 5G-IoT Environment) and by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (2020R1A2C2010449). This work was also supported by the Ministry of Trade, Industry and Energy by the Korea government (Project No. 1415169283).