Topical application of autophagy-activating peptide improved skin barrier function and reduced acne symptoms in acne- prone skin
1 | INTRODUC TION
Acne vulgaris is one of the most common skin diseases, mainly af- fecting the skin regions with dense pilosebaceous units.1 While com- monly occurs in early adolescence, adult acne is also an important health and cosmetic problem.2 In addition to pain, disfiguring lesions, and scarring, significant impact on quality of life and negative effects on self-esteem on patients makes acne one of the highest life-bur- dening diseases. While various factors have been suggested to be involved in acne pathogenesis, including follicular hyperkeratosis, proliferation of Cutibacterium acnes (C acnes), and inflammation, excessive formation and secretion of sebum are considered as an initiating ones.3 Sebaceous lipogenesis is known to be modulated by various agents, including androgens, peroxisome proliferator-acti- vated receptor-γ (PPARγ), aryl hydrocarbon receptor (AhR), growth factors, and their respective receptors.4 Recently, potential involve- ment of cannabinoid receptor 2 (CB2R) in sebaceous lipogenesis was also reported.5 Moreover, a potential role of exposome in acne development and severity has been suggested, including nutrition, pollutants, climate condition, and lifestyle factors,6 at least partially through the modulation of sebum production.7 Based on the poten- tial involvement of each factor, various kinds of anti-acne ingredi- ents have been developed.
Recently, we reported a potential involvement of autophagy signaling in sebaceous lipogenesis. Autophagy is a starvation-in- duced survival pathway through which dysfunctional or super- fluous cellular material is recycled to basic nutrients. As a major cellular protective machinery, autophagy mediates removal of toxic protein aggregates, damaged organelles, and invading patho- gens by guiding them to lysosomal degradation. In skin, autophagy plays an important role in maintaining homeostasis, such as epi- dermal differentiation and immune responses. Changes of autoph- agy signaling in skin diseases, including infections, photodamage and senescence, vitiligo, and skin cancers, are also repeatedly re- ported.8 In acne lesions, decreased expression of microtubule-as- sociated protein 1 light chain 3 (LC3), as an autophagy marker protein, was observed and in vitro studies using human immortal- ized SZ95 sebocytes also confirmed that stimulation of sebaceous lipogenesis by testosterone/linoleic acid (T/LA) treatment reduced LC3 expression. Stimulation of autophagy response by the mTOR1 agonist rapamycin downregulated the T/LA-induced sebaceous li- pogenesis suggesting potential benefits of autophagy activation in controlling sebaceous lipogenesis and probably initial forms of acne.9 Considering the beneficial effects of autophagy activation in epidermal homeostasis, including epidermal differentiation and skin barrier function, it can be postulated that autophagy activa- tion can be a potential approach for acne prevention and initial treatment. Further derivatization of the initially reported autoph- agy activator resulted in several candidates of peptide derivatives,and their autophagy-stimulating activity and physicochemical properties were examined. For the application in acne, a lipophilic derivative was chosen for higher penetration potential through the pilosebaceous unit.
In this study, we investigated the effects of the newly devel- oped autophagy activator on both sebocyte and keratinocyte func- tions, in terms of their involvement in acne pathogenesis. In vitro studies using human immortalized SZ95 sebocytes, normal human epidermal keratinocytes, and a reconstituted 3D skin model were supportive toward a clinical application. At last, the clinical efficacy of a topical formulation of the test peptide was evaluated in a ran- domized, double-blind, vehicle-controlled clinical study leading to promising results.
2 | MATERIAL S AND METHODS
2.1 | Materials
Test peptide (pentasodium tetracarboxymethyl heptadecanoyl dipeptide-12) was synthesized in Incospharm Incorporation as previously reported.10 Trehalose, testosterone, linoleic acid, and Nile red were purchased from Sigma-Aldrich (St. Louis). Anti-LC3B (Catalogue No. L7543) and anti-β-actin (Catalogue No. LF-PA0207) antibodies were purchased from Sigma-Aldrich and AbFrontier, re- spectively. Anti-mouse IgG-horseradish antibody and anti-rabbit IgG-horseradish antibody were from Sigma-Aldrich. All of the other agents used for the experiments were of analytical grade or higher.
2.2 | Cell culture
Normal human epidermal keratinocytes (NHEKs), EpiLife culture medium, and Human Keratinocyte Growth Supplement (HKGS) were purchased from Thermo Fisher Scientific. NHEKs were maintained in EpiLife culture medium supplemented with HKGS and 1% of an- tibiotics mixture under standard cell culture conditions of 37°C and 5% CO2 in a humidified atmosphere. The immortalized human SZ95 sebaceous gland cell line11 was cultured in DMEM/F-12 culture me- dium supplemented with 2 mM glutamax I, 10 μg ml−1 gentamicin, 50 ng ml−1 human EGF, 10% fetal bovine serum (FBS), and 10 mM HEPES (all from Gibco-BRL). To stimulate lipogenesis, sebocytes were treated with a combination of 2 × 10-8 M testosterone and 10-4 M linoleic acid (testosterone/LA) for 48 hours. Cells were main- tained at 37°C and 5% CO2 up to the third passage, and the medium was replaced every 2 days. All compounds were diluted in dimethyl sulphoxide (DMSO) and then diluted with the culture medium (the final concentrations of DMSO was 0.1%).
2.3 | Western blot
NHEKs were seeded in 12-well plates at a density of 5×104 cell/ well and incubated for 24 hours, and then, the cells were treated for 24 hours with either trehalose (10 mM; Sigma-Aldrich) or test peptide (5 µM). After 24 hours, cells were lysed using sample buffer (Laemmli 2x) (Elpis Biotech Inc) and cell lysates were sepa- rated by electrophoresis on 15% Tris-HCl Protein Gel (Bio-Rad). Separated bands were blotted onto PVDF blotting membranes (Roche Diagnostics GmbH), probed with appropriate antibodies, and detected using enhanced chemiluminescence (GE Healthcare). The intensity of analyzed bands was quantitated with Alliance Mini HD6 system (UVITEC).
2.4 | 3D skin model
The EpiDermFT™ (EFT-400) human skin model was obtained from MatTek Corporation. Upon arrival at the laboratories, skin model specimens were acclimatized for 16 hours in 6-well plates contain- ing 2 mL of maintenance medium (EFT-400-ASY) in a humidified atmosphere at 37°C and 5% CO2. Before the application of vehi- cle or test peptide, skin model specimens were placed in 2 mL of fresh maintenance medium and any moisture on the skin surface was gently removed with a cotton tip. Either 50μl of test peptide solution with 5 μM concentration or PBS solution as a vehicle was directly applied on the stratum corneum surface and incubated for 24 hours. During the treatment period, skin model specimens were maintained in a humidified atmosphere at 37°C and 5% CO2, with the maintenance medium being renewed daily. Each treatment ex- periment was performed in triplicate.
2.5 | Immunohistochemical analysis
After treatment, skin model specimens were fixed in 4% neutral buffered formalin (NBF) solution overnight at 4°C, dehydrated in a graded ethanol series, and embedded in paraffin. For immunohisto- chemical staining, 5-μm-thick tissue sections were deparaffinized and rehydrated. Heat-induced epitope retrieval treatment was per- formed in an antigen retrieval solution (ab937, Abcam) at pH 6.0 for 10 minutes at 97°C, followed by cooling for 10 minutes in a cooling chamber. After blocking the nonspecific antibody binding using a blocking reagent (X0909, Dako, Santa Clara, CA, USA), the rabbit anti-human LC3B antibody (ab48394, Abcam) was added to the sections in a humidified chamber overnight at 4°C. A horserad- ish peroxidase–linked secondary antibody (K4003, Dako) was ap- plied to the tissue samples for 1 hour at room temperature and was detected with a 3,3′-diaminobenzidine (DAB) chromogen system (ab64238, Abcam). The tissue sections were counterstained using hematoxylin nuclear staining.
2.6 | Clinical evaluation
The clinical study was performed on 49 acne patients (32 fe- males and 17 males; aged 24.8 ± 4.13) with Investigator’s Global Assessment (IGA) grade 2 or 3 and without any other skin or systemic diseases. The study complied with the World Medical Association’s Declaration of Helsinki (2013) concerning biomedical research involving human subjects and was approved by the Institutional Review Board of DERMAPRO, Ltd. (approval No. 1-220777-A-N- 02-DICN19062), and all subjects provided an informed consent. Participants, who used any anti-acne cosmetic products in the last 2 weeks or oral antibiotics, spreading agents, peeling, scaling for acne in the last 4 weeks, took oral contraceptives within 3 months, or used retinoids or laser therapy within 6 months, were excluded. At the first visit, the participants were asked to complete study-related medical record questionnaires and randomly assigned to either test (n = 24) or placebo group (n = 25). The participants were instructed to apply the topical product on their entire face twice a day for eight weeks and also asked to neither use any cosmetics containing ingre- dients that could potentially interfere with their acne status during the study period nor to have any exposure to direct sunlight. Lesion counting, trans-epidermal water loss (TEWL), casual skin surface li- pids, and photographic documentation were evaluated at baseline, 2, 4, 6, and 8 weeks after treatment. Self-questionnaires concerning efficacy and usability were filled out by the participants, and skin adverse reaction was assessed at 2, 4, 6, and 8 weeks. Before instru- mental measurements, participants were asked to rest for at least 30 min in a humidity (40-60% relative humidity)- and temperature (20 ~ 25°C)-controlled room.
2.7 | Cholesterol and squalene analysis
Stratum corneum (SC) samples were obtained using D-Squame tape (3.8 cm2, CuDerm, Dallas, TX, USA) from the central forehead area. Corneocytes in D-Squame tape were harvested and lysed in RIPA buffer, and then, lipids were extracted using chloroform/methanol (2:1, v/v). The extracted lipids were dried using a speed vacuum sys- tem (Hanil Scientific, Seoul, Korea). The remaining lipids were solu- bilized using 0.15 M potassium hydroxide in methanol at 90°C for 1 hour and were further extracted with n-hexane. The extract was evaporated under speed vacuum system and redissolved in isopropyl alcohol/acetonitrile (7:3, v/v). For the quantitation of cholesterol and squalene, 1260 Infinity Liquid Chromatography (Agilent Technologies, Waldbronn, Germany) combined with a photodiode array detector and a sample injector was used. Octadesyl benzene was used as an in- ternal standard sample and Eclipse XDB-C18 Column (4.6 × 150 mm, 5 μm particle size) from Agilent Technologies was used under isocratic elution solution (acetonitrile/tetrahydrofuran, 80:20, v/v) for 20 min- utes at a flow rate of 1 mL/min. Analytes were monitored with a pho- todiode array detector at a wavelength of 205 nm. Data were acquired using the Chemstation B.04.02 software (Agilent Technologies). The levels of cholesterol and squalene were quantitated using calibration curves with the various concentrations of analytes and internal stand- ard ratio and expressed as pmol/mg protein.
2.8 | Statistical analysis
Parametric, two-tailed, paired t tests or nonparametric, Wilcoxon signed ranked tests were performed to compare the differences be- fore and after treatment. The values are expressed as the arithmetic mean +/− standard deviation (SD). p values less than 0.05 were con- sidered significant.
3 | RESULTS AND DISCUSSION
3.1 | Autophagy activation in human keratinocytes and SZ95 sebocytes
Previously, we reported that sirtuin 1 (SIRT1)-activating peptide deriv- ative (heptasodium hexacarboxymethyl dipeptide-12) induced stimu- lation of autophagy flux in cultured human dermal fibroblast through the deacetylation of forkhead box class O (FOXO) 1.10 Further clini- cal evaluation showed significant improvements in skin elasticity and antioxidant system by topical application of autophagy activator.12 In order to develop new compounds for anti-acne, series of derivatives from heptasodium hexacarboxymethyl dipeptide-12 were further synthesized and tested for their autophagy-modulating signaling. Among the tested ones, pentasodium tetracarboxymethyl heptade- canoyl dipeptide-12 showed a significant autophagy-activating ef- ficacy in cultured normal human epidermal keratinocytes (NHEKs) after 24 hours of treatment, which was assessed by the change in microtubule-associated protein 1 light chain 3 (LC3)-II protein expres- sion. In order to address whether the test compound also stimulated autophagy signaling in sebocytes, changes of autophagy marker pro- teins in cultured immortalized human sebocytes SZ95 were observed. Similar to NHEKs, increase in LC3-II protein was also observed in SZ95 cells after 4 hours of treatment (Figure 1A, B). Preliminary ex- periments resulted in the absence of cytotoxicity by tested compound in both NHEK and SZ95 cells up to 5 μM after 48 hours of treatment (data not shown), and following experiments were performed at 5 μM concentration. As a positive control, trehalose also induced transition of LC3-II from LC3-1 in both NHEKs and SZ95 cells, but at the higher concentration (10mM) than test peptide (5 μM). Further experiments using 3D reconstituted skin model, increased expression of LC3 pro- tein expressions in suprabasal layer was observed (Figure 1C), which confirmed the autophagy-activating effects of test peptide in skin.
3.2 | Effects on sebaceous lipogenesis
In addition to the physiological roles of autophagy in skin homeo- stasis, including regulating epidermal differentiation and melano- genesis, potential involvement of autophagy signaling in various skin diseases has been repeatedly reported.13 In a previous report,Megyeri et al suggested that noninvasive Cutibacterium acnes strains, at the low-level colonization status, can increase au- tophagy activity in keratinocytes through their cell wall compo- nents and propionic acid metabolites.14 However, little has been studied about the roles of autophagy in sebaceous lipogenesis and acne. Recently, we reported that the autophagy-related proteins are constitutively expressed in sebaceous gland of normal human skin, and significant reduction of LC3-II and Atg7 proteins is ob- served in acne lesions. Starvation induced autophagy signaling in SZ95 cells, which was inhibited by sustained lipogenic stimuli treatment (J Dermatol Sci). Here, we have further investigated the effect of autophagy activation on sebocyte lipogenesis using SZ95 cells. Co-treatment of testosterone (2 × 10−8 M) and linoleic acid (10−4 M) (T/LA) on SZ95 sebocytes for 48 hours induced sig- nificant increases in sebaceous lipogenesis, assessed by BODIPY staining (red fluorescence), which was blocked by test peptide treatment. In addition, formation of LC3 puncta (red fluorescence), a morphogenic marker for autophagy signaling stimulation, was also observed in test peptide-treated SZ95 cells (Figure 2A). From the quantitative Nile Red staining with fluorescence measure- ment, significant reduction of both polar lipids and neutral lipids by test peptide was observed in SZ95 cells. As a positive control, epigallocatechin-3-gallate (EGC), a green tea catechin recognized as a potential sebostatic agent,15 also showed inhibitory effects on sebaceous lipogenesis at the similar concentration (10 μM) to that of test peptide (5 μM) (Figure 2B).
3.3 | Stimulation of epidermal differentiation marker protein expressions by autophagy activation
While the excessive production of sebum and consequent clogging of sebaceous gland is one of the major etiological factors for acne, recent studies also suggest that the impaired skin barrier function by sebum components is also important for acne, especially dur- ing the early phase.16 Increased production of sebum and their se- cretion onto skin surface result in changes of composition of skin surface lipids, impairment of skin barrier functions, and inducing inflammatory responses. Previously, we reported that topical ap- plication of ceramide-dominant physiologic lipid mixture with lin- oleic acid prevented oleic acid-induced comedogenesis in rabbit ear model.17 Stimulation of epidermal terminal differentiation can also improve epidermal permeability barrier function by increasing the formation and secretion of stratum corneum intercellular lipids and their precursors. Based on previous studies reporting that au- tophagy signaling is profoundly involved in keratinocyte differen- tiation process and autophagy activation stimulates the epidermal differentiation,18 we tried to evaluate whether the test peptide can also upregulate the expression of differentiation marker proteins in cultured epidermal keratinocytes. As results, protein expressions of keratin 1, keratin 10, and loricrin were significantly upregulated by test peptide and high calcium (1.8 mM) treatment (Figure 3). While keratin 1, as an early differentiation marker, showed dose-depend- ent changes by test peptide, expressions of keratin 10 and loricrin were higher in 5 μM condition. While there are a few plausible rea- sons, such as insufficient culture time (2 days after compound treat- ment), further investigations should be followed to elucidate the reason of the optimum concentration for keratinocytes differentia- tion for the test peptides.
3.4 | Clinical efficacy of autophagy-activating peptide on acne-prone skin
In order to assess the clinical efficacy of autophagy-activat- ing peptide in acne-prone skin, double-blind, randomized,placebo-controlled clinical test was performed. Topical formu- lation containing 100 ppm of test peptide and placebo product was prepared by Cha Bio F&C (Seongnam) and used for the study. After receiving the written consent for the participation of the study, subjects aged 19-40 years old with acne in level 2 or 3 of IGA (Investigator’s Global Assessment, Korea Ministry of Food and Drug Safety guideline for efficacy test of anti-acne products), were randomly assigned into either test group or control group. As a primary endpoint, lesion counting was performed on whole face except the nose and chin. As a result, number of whiteheads (defined as closed comedones smaller than 1 mm in diameter at the initial inflammatory stage) was reduced from 29.1 at baseline to 24.3 at 4 weeks after usage, and 20.1 at 8 weeks of usage in the test product-treated group, and statistical analysis resulted in a significant improvement in both measurements. In contrast, the control product-treated group did not show statistically significant differences during the application period (Figure 4A). Similarly, statistically significant reduction of the skin surface lipids, evalu- ated by Sebumeter SM810 (Courage + Khazaka, Koln, Germany), was only observed in test product-treated site after 8 weeks, which supports the sebosuppressive effects of test peptide (Figure 4B). Interestingly, reduction of trans-epidermal water loss (TEWL), as a representation of epidermal permeability barrier function, was observed in both test product and control product-treated sites, which suggests a skin barrier-improving efficacy of common oil-in- water (O/W)-type emulsion in acne-prone skin (Figure 4C). Notably, no adverse events were reported from the participants or investiga- tors during and after the application periods.
3.5 | Changes of skin surface lipids by test peptide
Further investigation about the clinical effects of test peptide on se- baceous lipid formation and epidermal lipids was performed by com- positional analysis of skin surface lipids collected during the clinical study. Since the endogenous source of squalene in skin surface lipids is mainly from sebocytes, increased amount of squalene in casual skin surface lipids can be a plausible lipid marker for acne-prone
skin.19 Skin surface lipids collected by D-Squame tape were analyzed using LC/MS/MS, and significant reduction of squalene was only ob- served in test product-treated site (Figure 5A), which was consistent with the in vitro sebosuppressive activity of test peptide. Recently, we reported that SIRT1 activation peptide treatment stimulated the epidermal keratinocyte differentiation through ceramide synthase 2- and 3-mediated signaling.20 Higher resemblance in structural and biological properties of test peptide to those of previously reported one20 suggests similar differentiation promoting activity of test pep- tide in skin. As a marker lipid for epidermal lipids constituting stratum corneum intercellular lamellar structure for skin barrier function, sig- nificant increase in cholesterol concentration was observed in test product-treated site after 8 weeks of usage (Figure 5B). While these results suggest that the in vitro activities of test peptide are fur- ther supported by clinical data, further investigation should be per- formed to elucidate the exact mechanisms underlying the increase in cholesterol by test peptide.
4 | CONCLUSION
While there are several limitations in the study design, such as relatively small number of participants, differences in baseline pa- rameters between test group and control group, and lack of clinical dose-response data, these results provide that topical application of autophagy-activating peptide results in clinical improvements in acne- prone skin. Considering that the activation of autophagy signaling in sebocytes does not share the mode of actions with currently used anti-acne ingredients, combination of autophagy-activating peptides with other ingredients can improve the clinical outcomes of topical formulation or alleviate the adverse effects of topical formulation by lowering the concentration QX77 of potentially irritating active ingredients.