N-Terminal Acetylation and C-Terminal Amidation of Spirulina platensis-Derived Hexapeptide: Anti-Photoaging Activity and Proteomic Analysis

Ultraviolet (UV) irradiation is a potent inducer for skin photoaging. This paper investigated the anti-photoaging effects of the acetylated and amidated hexapeptide (AAH), originally identified from Spirulina platensis, in (Ultraviolet B) UVB-irradiated Human immortalized keratinocytes (Hacats) and mice. The results demonstrated that AAH had much lower toxicity on Hacats than the positive matrixyl (81.52% vs. 5.32%). Moreover, AAH reduced MDA content by 49%; increased SOD, CAT, and GSH-Px activities by 103%, 49%, and 116%, respectively; decreased MMP-1 and MMP-3 expressions by 27% and 29%, respectively, compared to UVB-irradiated mice. Employing isobaric tags for relative and absolute quantitation (iTRAQ)-based proteomics, 60 differential proteins were identified, and major metabolic pathways were determined. Network analysis indicated that these differential proteins were mapped into an interaction network composed of two core sub-networks. Collectively, AAH is protective against UVB-induced skin photoaging and has potential application in skin care cosmetics.


Introduction
Solar ultraviolet (UV) light is composed of UVC (200-280 nm), UVB (280-315 nm), and UVA (315-400 nm). UVB can penetrate the epidermis and is particularly damaging to skin. It contributes predominantly to skin photoaging. Clinically, photoaging is characterized by coarse solar scars, roughness, dryness, wrinkles, laxity, and pigmentation [1]. UVB-induced photoaging not only damages biological macromolecules such as deoxyribonucleic acid (DNA), carbohydrates, lipids and proteins, but also decrease the activities of antioxidant enzymes in the skin such as superoxide dismutase (SOD) and glutathione peroxidase (GSH-Px) [2]. In addition, UVB-induced photoaging causes damage to the extracellular matrix (ECM) integrity in skin tissues by stimulating the production of various matrix metalloproteinases (MMP) such as MMP-1 and MMP-3 [3].
As early as the year of 2006, the micro alga Spirulina platensis extracts was revealed to possess anti-photoaging activity [4]. In addition, Neyrinck et al. showed that the oral administration of a Spirulina is able to modulate the gut microbiota and to activate the immune system in the gut, which is a mechanism that may be involved in the improvement of the hepatic inflammation in aged mice [5]. Meanwhile, numerous studies showed that Spirulina and Spirulina-derived peptides possess multiple activities such as prevention of hyperglycemia [6], antioxidant, immunomodulatory, and anti-inflammatory effects after dietary supplementation [7]. Furthermore, Souza and his research group payed much attention to the benefits of the topical application of Spirulina. Their results demonstrated

UV Irradiation and Peptide Treatment
UVB has been thought to be responsible for the damaging effects in the skin. Therefore, we used a UVB lamp apparatus (peak at 302 nm) for construction of a photoaged skin animal model in this study. The dorsal skin of mice was exposed to UVB light for 10 weeks of three times per week at an intensity of 60 mJ/m 2 , which was closed to four minimally erythemogenic doses (MED). During the experiment, once erythema, blisters, and erosion occurred in the dorsal skin of mice, the irradiation was stopped for 2-3 days, which could continue until the symptoms disappeared.
Hair in dorsal skin of mice was removed within an area of 2.5 × 3 cm 2 using a lady shaver, and the mice were acclimatized for two days before the experiment. A total of 100 µL of AAH (10 mg/mL), Matrixyl (10 mg/mL), and vehicle (no UVB exposure group and UVB irradiation group) were applied on the shaved dorsal skin. In this study, we used a solvent mixture of ethanol:water:propylene glycol = 3:3:4 (v:v:v) as a vehicle. This solvent mixture is the proper vehicle to fully solve the peptide. We topically applied all samples (100 µL/each mouse) on the dorsal skin of mice, and then gave enough time to be absorbed into the skin (about 2 h). The skin care product Matrixyl was used as the positive control. The mice were exposed to UVB radiation three times per week at 60 mJ/cm 2 for 10 weeks. We euthanized the mice in a humane way at the end of the study. Mice were anesthetized with ether and then executed cervical dislocation without pain. Then, skin specimens from the central dorsum of the mice were obtained.

Histological Examination and Moisture Content Test of Skin
The dorsal skin of mice was photographed under anesthesia by diethyl ether inhalation at the end of the study. An area of 1.0 × 1.0 cm 2 was harvested freshly and fixed in 10% neutral buffered formalin. The degree of skin structure alteration and elastosis were assessed microscopically using Haematoxylin-eosin (H&E) staining. To quantify epidermal thickness following UV exposure, Mar. Drugs 2019, 17, 520 4 of 22 measurement was conducted at 10 randomly selected locations per slide using an optical microscope with 200× magnification. Histological alterations were evaluated and quantified through the image analysis program Image Pro Plus 6.0 (Silver Spring, MD, USA). Skin (0.2 g) was quickly cut and precisely weighted (w 1 ), then it was moved into the oven, and dried at 80 • C to constant weight (w 2 ). Therefore, the skin moisture percentage could be determined by the following formula: the skin moisture percentage (%) = (w 1 − w 2 )/w 1 × 100.
2.4.4. Determination of SOD, CAT, GSH-Px, and MDA in Skin Tissue The harvested skin tissue (0.4 g) was homogenized (10,000 rpm, 20 s) in nine volumes of 0.9% saline (4 • C) to obtain the 10% skin tissue homogenate. The total supernatant was used for protein content, SOD, CAT, GSH-Px and MDA determination according to the manufacturer's protocols.

Determination of MMP-1 and MMP-3 in Skin Tissue
Skin tissue (0.4 g) was homogenized (10,000 rpm, 20 s) in nine volumes of PBS (4 • C) to obtain the 10% skin tissue homogenate. The undissolved pellet was removed by centrifugation at 3000× g for 20 min at 4 • C, and the total supernatant was saved for the subsequent assays. Secreted MMP-1 and MMP-3 were estimated using ELISA kits (Thermo scientific, Waltham, MA, USA) and protein content was determined according to the manufacturer's instructions.

Sample Preparation and Labeling
A total of four samples were collected as the target tissue for iTRAQ analysis at Shanghai Majorbio BioMedical Technology Co., Ltd. The samples were lysed to enrich the phosphorylated proteins with phosphoprotein enrichment kit according to the manufacturer's instructions. The protein concentration was determined by the Bradford method. Afterwards, 150 µg of protein for each sample was mixed with Dithitol (DTT, final concentration of 10 mM), reacted for 30 min at 56 • C, then, Iodoacetamide (IAA, final concentration of 20 mM) was added keeping in darkness for 30 min. Then, five volumes of cold acetone were precipitated for 2 h at −20 • C and then centrifuged at 12,000 rpm for 20 min at 4 • C. The precipitation was dissolved by 20 µL TEAB buffer with 1 M urea. Next, Trypsin (1/50 protein) was added and reacted for 15 h at 37 • C. Trichloroacetic acid (TFA, final concentration of 0.5%) was added to stop the hydrolysis reaction. The deposit was collected and dried by vacuum freeze dryer. Then, the sample (100 µg) was dissolved in dissolution buffer of the iTRAQ kit and reacted with 40 µL reducing reagent and 70 µL Isopropyl alcohol, and then 40 µL Milli-Q water was added after 2 h incubation. All the labeled samples from different treatment groups were pooled and freeze-dried.

Sample Fractionation and LC-MS/MS Analysis
Dried peptides were resuspended with buffer A: water (formic acid and ammonia water was used to adjust the pH to 10). Buffer B was 100% acetonitrile (ACN). Detection wavelengths were set as 214 and 280 nm, meanwhile, flow rate was 200 µL/min. The 60 min gradient comprised of 0-5% buffer B for 5 min, 5-25% buffer B for 35 min, 25-80% buffer B for 5 min, 80% buffer B for 5 min, 80-100% buffer B for 1 min, and finally 100% buffer B for 9 min. The combined samples were separated into 20 fractions.
Subsequently, the collected fractions were pooled according to the chromatogram profile based on the peak intensity and the products dried in a vacuum for LC-MS/MS analysis.

Data Analysis
The MS/MS spectra were extracted and searched against the database with Uniform Resource Locator (URL) of http://www.uniprot.org/proteomes/UP000000589 using Mascot 2. The cellular component, molecular function, and biological process were annotated by GO (Gene Ontology) database (http://www.geneontology.org/). The signaling pathways of proteins were elucidated by searching against the Kyoto Encyclopedia of Genes and Genomes database (http://www.genome.jp/kegg/pathway.html). The protein-protein interaction network was analyzed by Search Tool for the Retrieval of Interacting Genes/Proteins (STRING) software (http://string.embl.de/).

Western Blot
The extraction of total protein was carried out according to the extraction method in the experiment of iTRAQ. The quantification of the protein was carried out in strict accordance with the BCA kits. Subsequently, 30 µg of the sample was taken slowly from the micro-injector and added to the sample wells, meanwhile, 5 µL of protein markers were loaded and followed by electrophoresis, transfer, blocking, primary incubation, secondary incubation, and dyeing process. Among them, slightly different conditions were used in the transfer process, the conditions for most antibodies were 300 mA and 40 min, but 300 mA and 60 min for HSP60, 300 mA and 20 min for Cytochrome c. All antibody concentrations during the incubation were 1:1000 dilution. After the dyeing wass finished, pictures were taken or scanned, recording the analysis results.

Statistical Analysis
All assays were carried out in triplicate, and the experimental results were expressed as means ± standard deviations. Statistical analysis was performed by SPSS 16 (SPSS Inc., Chicago, IL, USA). Data were analyzed using the least significant difference (LSD) method by analysis of variance, and the value differences were considered to be significant when p < 0.05. The spectrum data from iTRAQ results were submitted for protein identification, and a database search was carried out using ProteinPilot Software 4.5 (AB SCIEX, Seattle, WA, USA) to perform database searches. The database used was the SwissProt_2013_09 (total sequence 540958). The search parameters used were as follows: Cysteine alkylation with MMTS; Trypsin Digestion; Triple TOF 5600; ID focus with Biological modifications; Search effort with thorough ID. A decoy database search strategy was used to determine the false discovery rate (FDR) for protein identification. The criteria for protein identification was set to FDR < 0.1%.

The Toxicity on Hacats and the Protective Effect on the Damaged Hacats
By MTT assay, the surviving rates of Hacats treated by Spirulina platensis-derived hexapeptide GMCCSR, AAH and matrixyl were 99.18% ± 6.97%, 81.52% ± 6.26%, and 5.32% ± 1.83%, respectively. Then, the Hacats were irradiated by UVB, and the damaged Hacats were treated with the peptides again. The surviving rates corresponding to Spirulina platensis-derived hexapeptide GMCCSR, AAH and matrixyl treatment were 112.97% ± 10.73%, 137.14% ± 8.42%, and 121.56% ± 1.42%, respectively, and the UVB control was set as 100%. Comparing with the unmodified hexapeptide, the surviving rates of Hacats treated by AAH reduced from 99.18% ± 6.97% to 81.52% ± 6.26%, which was still above 80%. Thus, the follow-up step was to conduct further experiments to explore its protection effects for the UVB-damaged Hacats. In addition, the modified hexapeptide exhibited significantly lower toxicity (p < 0.05) and significantly stronger protective effect (p < 0.05) on Hacats than the positive control matrixyl.

Morphology, Thickness, and Moisture Content of the Dorsal Skin on Mice
The morphological observation indicated that the dorsal skin of UVB-irradiated mice had distinct features: dark-brown color, deep wrinkle, relaxable surface, and local aging and necrosis appeared on the surface cortex ( Figure 1B). After matrixyl treatment, the appearance of skin was improved, but was still a little bit brown, with clear signs of UVB irradiation ( Figure 1C). However, the dorsal skin of mice treated with AAH ( Figure 1D) was greatly improved, which was better than the positive control matrixyl group, and was most similar to normal group ( Figure 1A): smooth surface, rosy color, satiation, no signs of relaxation and wrinkle. Notably, Figure 1(D1) also showed irregular epidermal thickening, and it is possible that there still exists some inflammation. Anyway, the extent of thickening was much smaller than those of the model group (Figure 1(B1)) and positive drug group (Figure 1(C1)), although it was slightly greater than that of the normal group (Figure 1(A1)). Furthermore, the histological examination by haematoxylin-eosin (H&E) staining displayed that the mice skin in the normal group had thin cortex, regularly arranged collagen fibrils, plump subcutaneous hair follicles and sebaceous glands ( Figure 1A). The skin in the UVB-irradiation group had irregularly thickened cortex, broken nuclei, vacuolar degeneration of epidermal basal cells, and inflammatory cells (lymphocytes and monocytes) infiltration of the dermis (Figure 1(B1)). In the positive control matrixyl group, although the cortex became thinner, there was still slight vacuolar degeneration and inflammatory cells infiltration ( Figure 1(C1)). While in the AAH treatment group, the cortex was further thinned, and followed by disappearance of vacuolar degeneration and inflammatory cells infiltration (Figure 1(D1)). In a word, these results suggest that AAH possessed better effects in improving mice skin than the positive control matrixyl.

Morphology, Thickness, and Moisture Content of the Dorsal Skin on Mice
The morphological observation indicated that the dorsal skin of UVB-irradiated mice had distinct features: dark-brown color, deep wrinkle, relaxable surface, and local aging and necrosis appeared on the surface cortex ( Figure 1B). After matrixyl treatment, the appearance of skin was improved, but was still a little bit brown, with clear signs of UVB irradiation ( Figure 1C). However, the dorsal skin of mice treated with AAH ( Figure 1D) was greatly improved, which was better than the positive control matrixyl group, and was most similar to normal group ( Figure 1A): smooth surface, rosy color, satiation, no signs of relaxation and wrinkle. Notably, Figure 1(D1) also showed irregular epidermal thickening, and it is possible that there still exists some inflammation. Anyway, the extent of thickening was much smaller than those of the model group (Figure 1(B1)) and positive drug group (Figure 1(C1)), although it was slightly greater than that of the normal group ( Figure 1(A1)). Furthermore, the histological examination by haematoxylin-eosin (H&E) staining displayed that the mice skin in the normal group had thin cortex, regularly arranged collagen fibrils, plump subcutaneous hair follicles and sebaceous glands ( Figure 1A). The skin in the UVB-irradiation group had irregularly thickened cortex, broken nuclei, vacuolar degeneration of epidermal basal cells, and inflammatory cells (lymphocytes and monocytes) infiltration of the dermis (Figure 1(B1)). In the positive control matrixyl group, although the cortex became thinner, there was still slight vacuolar degeneration and inflammatory cells infiltration (Figure 1(C1)). While in the AAH treatment group, the cortex was further thinned, and followed by disappearance of vacuolar degeneration and inflammatory cells infiltration (Figure 1(D1)). In a word, these results suggest that AAH possessed better effects in improving mice skin than the positive control matrixyl. By measuring the thickness of the dorsal skin (Figure 2), the results showed that UVB irradiation significantly (p < 0.05) increased the thickness of skin from 13.1 μm in the normal group to 40.2 μm; the positive control matrixyl decreased 40.2 μm of thickness in the UVB irradiation group to 23.8 μm, but still significantly (p < 0.05) greater than 13.1 μm in the normal group; and the thickness of skin in AAH group (13.5 μm) was close to 13.1 μm in the normal group. Further evaluation of moisture content indicated that UVB irradiation significantly (p < 0.05) decreased the By measuring the thickness of the dorsal skin (Figure 2), the results showed that UVB irradiation significantly (p < 0.05) increased the thickness of skin from 13.1 µm in the normal group to 40.2 µm; the positive control matrixyl decreased 40.2 µm of thickness in the UVB irradiation group to 23.8 µm, but still significantly (p < 0.05) greater than 13.1 µm in the normal group; and the thickness of skin in AAH group (13.5 µm) was close to 13.1 µm in the normal group. Further evaluation of moisture content indicated that UVB irradiation significantly (p < 0.05) decreased the moisture content (58%), compared to normal group (63.8%); the positive control matrixyl increased the moisture content (62.4%), especially, AAH significantly (p < 0.05) elevated the moisture content to 63.2%, compared to UVB irradiation group. This suggests that AAH was also better than the positive control matrixyl in terms of skin thickness and moisture content.
Mar. Drugs 2019, 17, x 7 of 22 moisture content (58%), compared to normal group (63.8%); the positive control matrixyl increased the moisture content (62.4%), especially, AAH significantly (p < 0.05) elevated the moisture content to 63.2%, compared to UVB irradiation group. This suggests that AAH was also better than the positive control matrixyl in terms of skin thickness and moisture content.

Effects on MDA, SOD, CAT, GSH-Px, MMP-1, and MMP-3 in Skin Tissue of Mice
As shown in Figure 3A, UVB irradiation caused the significant (p < 0.05) increase of MDA from 5.8 nmol/mg in the normal mice to 9.8 nmol/mg in UVB-irradiated mice. Matrixyl treatment decreased the increased MDA from 9.8 to 7.9 nmol/mg, while AAH significantly (p < 0.05)  As shown in Figure 3A, UVB irradiation caused the significant (p < 0.05) increase of MDA from 5.8 nmol/mg in the normal mice to 9.8 nmol/mg in UVB-irradiated mice. Matrixyl treatment decreased the increased MDA from 9.8 to 7.9 nmol/mg, while AAH significantly (p < 0.05) decreased the increase from 9.8 to 4.0 nmol/mg, up to 49%, even lower than 5.8 nmol/mg in the normal group.
For antioxidant enzymes SOD, CAT, and GSH-Px, there was an identical trend in their alterations: UVB irradiation decreased the activities of SOD, CAT, and GSH-Px, but they were increased in the group of matrixyl treatment and in the group of treatment with AAH. Specifically, the activities of SOD in normal, UVB-irradiation, matrixyl, and AAH treatment groups were 57.5, 19.1, 4.7, and 38.8 U/mg, respectively ( Figure 3B); the activities of CAT in normal, UVB-irradiation, matrixyl, and AAH treatment groups were 13.4, 7.1, 8.2, and 10.6 U/mg, respectively ( Figure 3C); the activities of GSH-Px in normal, UVB-irradiation, matrixyl, and AAH treatment groups were 130.3, 48.8, 105.1, and 105.2 U/mg, respectively ( Figure 3D).
Moreover, Figure 3E shows that the expression of MMP-1 in UVB irradiation group (150.2 ng/mg) was increased compared to normal group (90 ng/mg), but such an increase was slowed down in the matrixyl treatment group (120.1 ng/mg) and the modified hexapeptide treatment group (109.8 ng/mg). Similarly, Figure 3F displayed that the expression of MMP-3 in the UVB irradiation group (82.6 ng/mg) was also increased compared to normal group (71.2 ng/mg), while both matrixyl and the modified hexapeptide decreased the expression of MMP-3 to 63.4 and 58.8 ng/mg, respectively. decreased the increase from 9.8 to 4.0 nmol/mg, up to 49%, even lower than 5.8 nmol/mg in the normal group. For antioxidant enzymes SOD, CAT, and GSH-Px, there was an identical trend in their alterations: UVB irradiation decreased the activities of SOD, CAT, and GSH-Px, but they were increased in the group of matrixyl treatment and in the group of treatment with AAH. Specifically, the activities of SOD in normal, UVB-irradiation, matrixyl, and AAH treatment groups were 57.5, 19.1, 4.7, and 38.8 U/mg, respectively ( Figure 3B); the activities of CAT in normal, UVB-irradiation, matrixyl, and AAH treatment groups were 13.4, 7.1, 8.2, and 10.6 U/mg, respectively ( Figure 3C); the activities of GSH-Px in normal, UVB-irradiation, matrixyl, and AAH treatment groups were 130.3, 48.8, 105.1, and 105.2 U/mg, respectively ( Figure 3D).
Moreover, Figure 3E shows that the expression of MMP-1 in UVB irradiation group (150.2 ng/mg) was increased compared to normal group (90 ng/mg), but such an increase was slowed down in the matrixyl treatment group (120.1 ng/mg) and the modified hexapeptide treatment group (109.8 ng/mg). Similarly, Figure 3F

Proteomic Analysis of Skin Tissue in Mice
In order to explore molecular mechanism of protective effects on mice skin exerted by AAH, the iTRAQ-based proteomics was performed on the skin tissue from mice treated with UVB irradiation (model group) and with UVB + AAH (experimental group). The results showed that 60 differential proteins were identified, 33 proteins were up-regulated, and 27 proteins were down-regulated proteins (Table 1)

Proteomic Analysis of Skin Tissue in Mice
In order to explore molecular mechanism of protective effects on mice skin exerted by AAH, the iTRAQ-based proteomics was performed on the skin tissue from mice treated with UVB irradiation (model group) and with UVB + AAH (experimental group). The results showed that 60 differential proteins were identified, 33 proteins were up-regulated, and 27 proteins were down-regulated proteins (Table 1)  The volcano diagram of differential proteins is shown in Figure 4A. GO enrichment analysis ( Figure 4B) indicated that 13 functions were extremely significantly enriched (p < 0.001), for example, keratinization, carbohydrate catabolic and metabolic processes, intermediate filament cytoskeleton organization, hair follicle morphologenesis, lysosomal membrane, etc.; 27 functions were very significantly enriched (p < 0.01), for example, glucose metabolic process, hair cycle process, regulation of bone remodeling, mitochondrial electron transport cytochrome c to oxygen, vacuolar membrane, basal plasma membrane, misfolded protein binding, response to heat, etc.; others were significantly enriched (p < 0.05), such as positive regulation of apoptotic signaling pathway via death domain receptor, negative regulation of hydrogen peroxide metabolic process, regulation of interleukin-10 and -12 productions, etc.

Western Blot Verification
As mentioned above, proteomics identified 60 differential proteins between the UVB irradiation group and AAH treatment group. From them, six proteins were selected for verification by Western blot: Haptoglobin, Cytochrome c, Nucleophosmin, HSP60, CA3, PDHA1, and two parallels per sample. Expression intensities were determined by the digital gel image analysis system (LG2000, Hangzhou LongGene Scientific Instrument Co., Ltd. Hangzhou, China). Results revealed that all six proteins displayed consistent alterations between iTRAQ and Western blot experiments ( Figure 5 and Table 2). significantly enriched functions include (p < 0.001) keratinization, intermediate filament cytoskeleton organization, hair follicle morphologenesis and lysosomal membrane; but only three pathways were significantly (p < 0.05) regulated: HIF-1 signaling pathway, glycolysis/ gluconeogenesis, and methane metabolism. The network analysis of differential proteins demonstrated the existence of two core sub-networks: one was the sub-network centered by Pgam2; another was the sub-network that consisted of four proteins (Rps11, Rps3a, Rps21, and Mrpl11).

Western Blot Verification
As mentioned above, proteomics identified 60 differential proteins between the UVB irradiation group and AAH treatment group. From them, six proteins were selected for verification by Western blot: Haptoglobin, Cytochrome c, Nucleophosmin, HSP60, CA3, PDHA1, and two parallels per sample. Expression intensities were determined by the digital gel image analysis system (LG2000, Hangzhou LongGene Scientific Instrument Co., Ltd. Hangzhou, China). Results revealed that all six proteins displayed consistent alterations between iTRAQ and Western blot experiments ( Figure 5 and Table 2).

Discussion
Except for intrinsic aging, photoaging is a major process of skin aging. Keratinocytes are the outermost layer of the skin, constituting 95% of the cells in the epidermis. UV radiation induces apoptosis of keratinocytes to form sunburn cells, showing premature and abnormal keratinization [14]. Tsoyi et al. [15] proved that anthocyanins from black soybean seed coats protected Hacats from UVB-induced apoptosis. Lee et al. [16] reported that the processed Panax ginseng, Sun Ginseng, has an anti-apoptotic effect on UVB-irradiated Hacats. The present study indicated that AAH had much lower toxicity on Hacats than the positive control matrixyl (81.52% vs. 5.32%). Additionally, AAH possessed significantly stronger ability (p < 0.05) to proliferate UVB-damaged Hacats than the positive control matrixyl (137.14% vs. 121.56%). This suggests better potential of AAH as a skin care agent in terms of toxicity and protective effect on keratinocytes, compared to the positive control matrixyl.
Following keratinocytes experiment, the protective effect of AAH on mice skin was tested. The specific characteristics of photoaging included epidermal thickening and inflammatory infiltration [17]. Indeed, Feng et al. [18] reported that UV irradiation increased, by over 2-fold, the epidermal thickness of mice, while patchouli alcohol, the major active sesquiterpene found in Pogostemonis Herba, decreased by over 20% of the epidermal thickness. Meanwhile, UV exposure led to severe wrinkling with deep furrows, laxity and erythema in the skin of mice, but patchouli alcohol inhibited the formation of UV-induced erythema, roughness, and deep wrinkles. Similarly, Wu et al. [19] found that UVB resulted in an over 3-fold increase in the skin thickness compared with that of normal mice, and induced erythema and inflammation of the mice skin, Coffea arabica extract decreased the epidermal thickness by 30%, and ameliorated UVB-induced inflammation and erythema. Our data also indicated that the thickness of mice skin after UVB irradiation (40.2 µm) was dramatically increased, over 3-fold compared to the thickness of normal mice skin (13.1 µm), while AAH reversed the increased thickness of mice skin (13.5 µm). On the other hand, UVB-irradiated mice skin displayed vacuolar degeneration of epidermal basal cells and inflammatory cells infiltration of the dermis, while AAH removed the vacuolar degeneration and inflammatory cells infiltration.
It is well known that UV irradiation induces oxidative damage by inhibiting the activities of endogenous antioxidant enzymes such as SOD and GSH-Px, elevating the production of MDA (a well-known biomarker for lipid peroxidation) [20]. In addition, UV irradiation alters the connective tissues of the skin by up-regulating the expression of MMPs, which play an important role in skin photoaging. For example, by cleaving type I and type III collagen, MMP-1 initiates collagen breakdown; after being activated by MMP-3, MMP-1, -2, and -9 derived from dermal fibroblasts or keratinocytes initiate degradation of type I and III collagens [21]; MMP-2 and MMP-9 are important mediators in UV-irradiated skin damage and in the formation of wrinkles [22]. Lee et al. [23] showed that by suppressing the NF-kappaB pathway, cordycepin down-regulated MMP-1 and -3 gene expression in UVB-irradiated human dermalfibroblasts. Feng et al. [18] indicated that patchouli alcohol markedly reversed the decreased activities of SOD and GSH-Px, reduced MDA production by 30%, and inhibited the increase of MMP-1 and MMP-3 expression by about 24.6% and 39.3% respectively, in UVB-treated mice. Kim et al. [24] revealed that youngiaside increased the expression of SOD and suppressed MMP-1 production via Nrf2 and AMPK pathways in Hacats and Human Dermal Fibroblasts. Wu et al. [25] reported that Coffea arabica extract attenuated UVB-induced MMP-1 expression in the hairless mouse skin. The present study demonstrated that compared with the UVB-irradiated mice group, AAH reduced MDA content by 49%; increased SOD, CAT, and GSH-Px activities by 103%, 49%, and 116%, respectively; decreased MMP-1 and MMP-3 expressions by 27% and 29%, respectively; which exhibited better effects than the positive control matrixyl.
Although too many studies of UVB damage were reported, the molecular mechanisms causing this damage requires further elucidation. Over the past few years, proteomic tools have been used to investigate the biological effects of UVB exposure on human skin. For example, Bertrand-Vallery et al. [26] performed a 2D-DIGE proteomic profiling of human keratinocytes undergoing UVB-induced alternative differentiation, 69 differentially abundant proteins were identified by mass spectrometry, especially, confirming TRIpartite Motif Protein 29 as a survival factor. Wu et al. [25] employed lysine-and cysteine-labeling 2D-DIGE and MALDI-TOF mass spectrometry to conduct proteomic analysis of UVB-induced protein expression in skin fibroblasts. The results showed that 89 significantly changed proteins were identified, these UVB-modulated proteins were involved in many cellular responses including photoaging, melanogenesis, anti-apoptosis, tumorigenesis, and cell migration. Fang et al. [17] applied a high-throughput 2DE analysis coupled with MALDI-TOF MS to profile the global proteins involved in chronologically aged and photoaged skin in nude mice, and 15 differential proteins were identified. The most striking characteristic was that 14-3-3 sigma was down-regulated by 3.41-fold in the chronological aging skin, and proliferating cell nuclear antigen was up-regulated by 1.5-fold in UVB-induced aging skin.
In the present study, a novel iTRAQ-based proteomics tool was applied to obtain new insights into the protein profiles involved in photoaged mice skin. Compared with UVB treatment group, 60 differential proteins were identified in AAH treatment group. These proteins formed a complex acting network consisting of two core sub-networks and were involved in three major pathways: HIF-1 signaling pathway, glycolysis/gluconeogenesis, and methane metabolism.
Among the significantly up-regulated proteins, the proteins with a little link to aging include ODPA (Pyruvate dehydrogenase E1 component subunit alpha) (1.6 fold) and AT2A1 (Sarcoplasmic/endoplasmic reticulum calcium ATPase 1, SERCA1) (1.6 fold). It was reported that pyruvate dehydrogenase was decreased in senescent skin fibroblasts [27]; increased expression of SERCA can reduce hydroxyl radical injury in murine myocardium [28]. The most remarkedly down-regulated protein was HPT (Haptoglobin), which was 3.4-fold reduced in the modified hexapeptide group compared to the UVB group. Haptoglobin is an acute-phase protein secreted by white adipose tissue or liver cells. HPT is induced in pro-oxidative conditions such as systemic inflammation or obesity. In both inflammation and obesity HPT is up-regulated [29]. K2C6A (Keratin 6A) was also a remarkedly altered protein with 3.2-fold reduction. Similarly, other keratins were also decreased, including K1C17 (keratin 17) (−1.8 fold), K1C28 (keratin 28) (−1.8 fold), K1C27 (keratin 27) (−1.7 fold), K1C16 (keratin 16) (−1.6 fold), and K2C71 (keratin 71) (−2.1 fold). It was reported that keratin-6 was significantly higher (p < 0.05) in elderly skin [30]. The aging of skin is associated with decreased barrier function and gradual deterioration of the epidermal immune response, leading to chronic inflammation. Depianto et al. [31] pointed out that absence of keratin 17 attenuated hyperplasia and inflammation in models of acute dermatitis. High expression of keratin 16 in mice skin could cause skin lesions and alterations in keratin filament organization and in cell adhesion [32]. Thus, the above-mentioned disappearance of inflammatory cells infiltration could be related to attenuation of skin inflammation caused by significant down-regulation of HPT, K2C6A, and K1C17.
In conclusion, this is the first study to demonstrate the anti-photoaging effects of an acetylated and amidated peptide in UVB-irradiated keratinocytes and mice. The results showed that AAH reduced the content of MDA, increased the activities of SOD, CAT, and GSH-Px, and decreased the expression of MMP-1 and MMP-3. By employing a novel iTRAQ-based proteomic analysis, 60 differential proteins were identified, which were mapped into an interaction network composed of two core sub-networks, and key pathways were determined. In a word, the present data suggests that AAH was superior to the positive control matrixyl and has strong potential to be developed into a cosmetic product against skin photoaging.