Capsaicin, capsazepine, transforming growth factor beta 1 (TGF-β1), N-acetyl-l-cysteine (NAC), 2′-7′ dichlorofluorescin diacetate (DCFH-DA) and PD98058 were all purchased from Sigma-Aldrich (St Louis, MO). PAF C-16 was purchased from Santa Cruz Biotechnology (Dallas, TX).

Murine dermal fibroblasts were isolated from C57BL/6 mice, as described previously (Seluanov et al., 2010). Briefly, the dorsal skin was harvested from day 2 mouse, after sacrificed. The collected skin was incubated at 4 mg/ml dispase overnight at 4°C. Afterwards, the dermis was separated from the epidermis and then digested in 3 mg/ml collagenase for 1 h at 37°C. After filtration through 40 µm cell strainer, the dermal fibroblasts were harvested by centrifugation at 180 × g for 3 min. Cells were cultured in Dulbecco’s modified eagle medium (DMEM) with 20% fetal bovine serum (FBS). Primary murine epidermal keratinocytes were obtained, as described previously (Li et al., 2017) and cultured in EpiLife™ medium with 60 μM calcium. The medium was supplemented with 1% (v/v) penicillin/streptomycin (10,000 U and 10,000 μg/ml), and the cells were incubated in a humidified atmosphere with 5% CO² at 37°C. The culture reagents were purchased from Thermo Fisher Scientific (Waltham, MA).

MTT [3-(4,5-dimethylthiazolyl-2)-2,5-diphenyltetrazolium bromide] assay was performed to determine cell viability with drug treatments. Cells were seeded onto 96-well plates for 24 h and treated with various concentrations of capsaicin for further 24 h. Afterwards, the medium was removed, and 100 μl of MTT (0.5 mg/ml in complete growth medium) solution was added and incubated at 37°C for 2 h. The medium was replaced with 100 μl of DMSO in each well and shaken for 15 min. Absorbance was measured using a Multiskan FC Microplate Photometer (Thermo Fisher Scientific) at a wavelength of 570 nm.

Animals were obtained from Animal and Plant Care Facility of Hong Kong University of Science and Technology (HKUST), and the experiments were performed according to the guidelines of Department of Health, The Government of Hong Kong SAR. The experimental procedures were approved by Animal Ethics Committee at the University (Reference No (15–50) in DH/SHS/8/2/2 Pt.2). The female C57BL/6 mice at 8 weeks old were divided into four groups. The hair on dorsal skin was shaved. The mice in ultraviolet B (UVB) and capsaicin + UVB groups were pre-treated with saline and capsaicin, respectively, on dorsal skin each day, followed by exposed to UVB irradiation (300 mW/m² for 30 min each day) generated by Philips UVB narrowband lamp (Philips, Amsterdam, Netherlands) for 5 days. The control and capsaicin group were treated with saline and capsaicin, respectively, without exposure to UVB irradiation. The dorsal skins were collected after the mice were sacrificed by cervical dislocation, for haematoxylin and eosin staining, and immuno-histochemistry.

The dorsal skins were excised from C57BL/6 mice (8 weeks old) after sacrificed by cervical dislocation with the hair shaved, and then maintained in DMEM supplemented with 10% (v/v) FBS and 1% (v/v) penicillin/streptomycin (10,000 U and 10,000 g/ml) in 5% CO₂ at 37°C. The skin was treated with drugs followed by exposure to UVB irradiation (300 mW/m² for 30 min each day) for 3 days. The protein and total RNA were extracted from the ex vivo skins.

Total RNA was extracted using RNAzol^@RT reagent (Molecular Research Center, Cincinnati, OH). Briefly, the homogenized mouse skins, or cells, were incubated in RNAzol^@RT at room temperature. Then, total RNA was precipitated in 75% ethanol (v/v) by centrifugation at 12,000 × g for 10 min. The RNA pellet was washed by 75% ethanol and dissolved in RNAase-free water. The RNA quality was determined according to a ratio (∼2.0) of absorbance at 260–280 nm by NanoDrop™ (Thermo Fisher Scientific). One μg RNA of each sample was applied for reverse transcription using First Strand cDNA Synthesis Kit (Thermo Fisher Scientific), in accord with the manufacturer’s protocol. The sequences of specific primers were shown in Table 1. Amplification was performed for 45 cycles. Each cycle consisted of denaturation at 95°C for 30 s, annealing at 55°C for 30 s, and extension at 72°C for 20 s, performed on Roche Lightcycler 480 System (Roche, Basel, Switzerland).

The dorsal skin was collected from mice after sacrificed. Skin tissues were embedded in optimal cutting temperature (OCT) compound by snap freezing in liquid nitrogen and then stored at −80°C for further experiments. The skin section at 10 µm-thickness was made using Thermo CryoStar X 70 Cryostat (Thermo Fisher Scientific), followed by fixation with 4% paraformaldehyde for 10 min at room temperature. After sectioning, skin section was subjected to haematoxylin and eosin staining, according to the protocol of a staining kit (Yuan Ye Biology, Shanghai, China) with minor modification. Briefly, skin section was incubated in haematoxylin solution for 5 min followed by 1% (v/v) acid alcohol. Then, the skin section was incubated with eosin for 3 s before dehydration with 95 and 100% ethanol, then incubated with xylene for 15 s for three times.

The pre-fixed skin sections were incubated with 30% hydrogen peroxide for 10 min and then blocked at room temperature for 1 h. Afterwards, skin sections were incubated with primary antibody anti-collagen type I (Abcam Ltd., Cambridge, UK) at 4°C overnight then with horseradish peroxidase (HRP)-conjugated secondary antibody (Sigma-Aldrich) at room temperature for 2 h. The signal was detected using DAB substrate kit (Abcam Ltd.) and further stained with haematoxylin for the nuclear followed by mounted.

The cultured cells were fixed with 4% paraformaldehyde for 10 min. Samples were incubated with 1% BSA with 0.2% Triton X-100 for 1 h. Cultures were incubated with primary antibodies, anti-cytokeratin at 1:100 (Abcam Ltd.), anti-collagen type I at 1:100 (Abcam Ltd.), at 4°C overnight, followed by incubation with Alexa 488/647-conjugated antibodies (Abcam Ltd.). Samples were mounted with ProLong™ Gold Antifade Mountant with DAPI (Thermo Fisher Scientific) and then subjected to visualised by Zeiss laser scanning confocal microscope.

The amounts of collagen in mouse skin and cultured dermal fibroblasts were measured using a collagen ELISA kit (Shanghai Fankel Industrial Co., Shanghai, China), according to the protocol. Briefly, after exposure to UVB or treatment with various drugs, the protein lysates were extracted from mice dorsal skin and cultured fibroblasts, and incubated with reaction reagents at room temperature for 1 h. The amount of collagen was determined by the absorbance at 450 nm according to the standard curve of human collagen.

The protein lysates were extracted from murine dermal fibroblasts with RIPA lysis buffer, and 40 μg of each was subjected to 8% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) transferred to nitrocellulose membranes. The membranes were blocked with 5% skim milk powder in Tris-buffered saline with 0.1% Tween-20 (TBST) for 2 h. After blocking, the membranes were incubated at 4°C overnight with specific primary antibodies, including anti-collagen type I at 1:1,000 (Abcam Ltd.), anti-Erk 1/2 (Cell Signaling Technology, Danvers, MA) at 1:1,000, anti-phospho-Erk 1/2 (Cell Signaling Technology), at 1:1,000, and anti-α tubulin at 1:2,000 (Sigma-Aldrich), followed by incubated with HRP secondary antibodies at room temperature for 1 h. The immune-reactive proteins were detected using enhanced chemiluminescence (ECL) western blotting detection kit (Thermo Fisher Scientific). The intensities of bands were quantified using ChemiDoc Imaging System (Bio-Rad).

Mouse dermal fibroblasts were loaded with calcium indicator by incubation with 2 μM fluo-4 AM (Thermo Fisher Scientific) in culture medium at 37°C for 30 min. Then, the cells were observed with drug treatment and the image was captured by Zeiss confocal microscope. The intensity of fluorescent signal was quantified using ImageJ software (NIH Image).

The cultured dermal fibroblasts were pre-treated with capsaicin for 30 min, followed by exposure to UVB irradiation at 300 mW/m² for 5 min. After 6 h, cells were washed with PBS twice and then incubated with DCFH-DA (20 µM) in serum-free medium at 37°C for 20 min. For image captured by confocal microscope, cells were fixed with 4% paraformaldehyde for 10 min at room temperature and then mounted with ProLong™ Gold Antifade Mountant with DAPI (Thermo Fisher Scientific) followed by visualizing through a Zeiss laser scanning confocal microscope. The intensities of the fluorescent signals were quantified using ImageJ software.

The DNA construct pARE-Luc containing four antioxidant response elements (ARE; 5′-TGA nnn GCA-3′) and a downstream luciferase-reporter gene was used. Cultured fibroblasts were transfected using jetPRIME reagent (Polyplus Transfection, New York, NY), as described (Lai et al., 2021). The pARE-Luc transfected cells were pre-treated with capsaicin for 30 min, followed by exposure to UVB irradiation. The promoter-driven luciferase assay was performed using a commercial kit (Thermo Fisher Scientific). In brief, cells were lysed by 100 mM potassium phosphate buffer (pH 7.8), 0.2% Triton X-100 and 1 mM DTT and agitated for 30 min. Afterwards, cells were centrifuged at 16,100 × g for 10 min at 4°C. Twenty micrograms of cell lysates were used for assay. The luminescent reaction was quantified in a GloMax^® 96 Microplate Luminometer and normalized by protein amount.

Data are expressed as the mean ± SD. Statistical comparisons of the means between different treatments were analysed by using one-way ANOVA followed by a Bonferroni post-hoc test. Significant values are presented as *p < 0.05, **p < 0.01, and ***p < 0.001.

The shaved mouse dorsal skins were topically treated with, or without, capsaicin for 30 min before stimulated by UVB irradiation (300 mW/m² for 30 min) every day. After 5 days of irradiation, the skin condition was assessed by staining with haematoxylin and eosin, as well as immunostaining using antibody against collagen type I. In the UVB-irradiated group, the skin was drier and showed significant pachylosis, compared with that of control group (no UVB) (Figure 1A). The dryness and pachylosis of skin were considerably alleviated by pre-treatment with capsaicin before the irradiation. Capsaicin itself did not affect the skin condition (Figure 1A). The skin histological result illustrated that the inflammatory responses, e.g., increased thickness and nuclear staining of skin, were triggered markedly under UVB exposure. However, the damages were significantly relieved by applied capsaicin (Figures 1A,B). Furthermore, capsaicin promoted collagen type I (COL1) content in the UVB-irradiated skin back to the control level, as shown by the staining of collagen (Figure 1C) and ELISA assay (Figure 1D). Capsaicin alone increased the level of collagen type I of unexposed skin by ∼25%. These results therefore implied that capsaicin was able to protect the skin from UVB irradiation, especially by promoting the amount of collagen.

To confirm the effect of capsaicin on collagen production, various concentrations (1–10 μM) of capsaicin were applied onto an ex vivo mouse skin culture model. The isolated skin was treated with applied capsaicin. The mRNA levels of different forms of collagen, i.e., COL1A1, COL1A2, and COL3A1, were determined. The result showed that UVB irradiation reduced the transcriptional levels of these genes by ∼60%, dramatically, as compared to the control (no UVB). However, the pre-treatment of capsaicin reversed the UVB-induced decline on the mRNA levels of COL1A1, COL1A2, and COL3A1, indicating a dose-dependent effect in promoting the gene transcriptions (Figure 2A). The maximal concentration of capsaicin used here was 10 μM, which restored the mRNA levels to over 20% of increase, as compared to the control. On the other hand, the treatment of capsaicin only promoted the mRNA levels of these genes in the non-UVB exposed mouse skins, again in a dose-dependent manner (Figure 2A). The mRNA expression of collagen could be maximized at ∼100% increase of the control under the challenge of 10 μM capsaicin. In parallel, the protein level of COL1 showed similar tendency, in which capsaicin induced collagen content in skin with or without UVB irradiation (Figure 2B). This inductive effect of capsaicin was in a dose-dependent manner as well. TGF-β1, a well-known promoter on collagen synthesis in fibroblast, was applied as a positive control.

In order to reveal the roles of capsaicin in skin cell models, two cell types, keratinocytes and dermal fibroblasts, were isolated from mouse skin and cultured in vitro. The cell identities were validated by morphology and antibodies against collagen type I for fibroblast, or cytokeratin for keratinocyte, in immunofluorescent staining (Supplementary Figure S1A). By MTT assays, the toxicity dose of capsaicin in cell viabilities of both fibroblasts and keratinocytes was determined: the concentration from 1 to 10 μM showed no cytotoxicity to the cultures (Supplementary Figure S1B). Thus, capsaicin did not show toxicity at our working concentrations to the major cell types of skin.

The irradiation of UVB suppressed the transcriptional levels of COL1A1, COL1A2, and COL3A1 in cultured fibroblasts by over 50%, similar to that in animal tests (Figure 3A, left panel). The mRNA levels of these genes were increased significantly by pre-treatment of capsaicin in cultured fibroblasts being exposed to UVB. Meanwhile, the treatment of capsaicin by itself increased these mRNAs in cultures by 100% to over 200%, as compared to the control group, in a dose-dependent manner (Figure 3A, right panel). The expression of COL1A1 mRNA was highly sensitive to the challenge of capsaicin of having ∼200% increase of control under 10 μM capsaicin. By using antibody against COL1 in western blot, the expressions of COL1 α1 (∼138 kDa) and α2 (∼129 kDa) were both up-regulated by capsaicin in UVB- (Figure 3B left panel) or non-UVB-exposed conditions (Figure 3B right panel). TGF-β1 was a control here.

Matrix metalloproteinases (MMPs) are known as collagenase and the UVB-induced expressions of MMP1, 3, and 9 are considered as the dominant cause of decreased collagen in skin (Kang et al., 2020). In cultured dermal fibroblasts, the transcriptional levels of different MMP isoforms, i.e., MMP1a, MMP3, and MMP9, were increased markedly by few folds under UVB irradiation (Figure 4). The increase in MMP could be accounted for the cause of decreased collagen in fibroblasts. The mRNAs of MMPs, as induced by UVB exposure, were markedly reduced by pre-treatment of capsaicin: the MMP expression was restored to normal level under a treatment of 10 μM capsaicin (Figure 4). Capsaicin alone showed a reduction of MMP in cultured fibroblasts, in a dose-dependent manner (Figure 4 ). These results implied that capsaicin increased the collagen protein content through suppressing the collagen degradation as triggered by MMPs.

The UV-induced phosphorylated Erk has been reported to increase the expressions of MMP1, 3, and 9 in dermal fibroblasts (Kim et al., 2005; Alge-Priglinger et al., 2009). To determine the role of capsaicin in affecting MMP expression through the Erk phosphorylation, Western blot was performed from cultured fibroblasts being treated with capsaicin with or without UVB irradiation. Exposure of UVB induced Erk phosphorylation (∼40 and ∼42 kDa) by ∼200% (Figure 5A left panel). Capsaicin notably inhibited the UVB-induced phosphorylation of Erk in a dose-dependent manner (Figure 5A, left panel). The Erk phosphorylation was almost back to control level after treatment of 10 μM capsaicin. In the absence of UVB irradiation, the treatment of capsaicin reduced the level of phosphorylated Erk in a dose-dependent manner (Figure 5A, right panel). PD98059, an inhibitor of Erk phosphorylation, was applied here as a positive control. Total level of Erk was not changed in all cases.

The phosphorylated c-Jun is part of AP-1 complexes (Huang et al., 1991), which is known as an activator for the expression of MMP gene (Poulalhon et al., 2006). In cultured fibroblasts, UVB exposure caused an increase of ∼100% of phosphorylated c-Jun, as well as the total c-Jun (Figure 5B left panel). Capsaicin reduced the expression of both phosphorylated and total c-Jun with ( Figure 5B left panel) or without (Figure 5B right panel) UVB irradiation in a dose-dependent manner. In addition, PAF C-16, an activator of MAP kinase inducing Erk phosphorylation, was added to the culture before capsaicin and UVB treatment. PAF C16 reversed the capsaicin-induced increase of collagen after UVB damage. Meanwhile, the inhibition of UV-induced phosphorylations of Erk and c-Jun by capsaicin was dramatically attenuated by the applied PAF C-16 (Figure 5C). Hence, our results indicated that capsaicin increased the collagen content could be through inhibiting Erk/c-Jun-mediated MMP signaling.

TRPV1 is known as the capsaicin receptor detecting the sensation of heat and pain (Szallasi and Sheta, 2012). Capsaicin activates TRPV1 and changed Ca²⁺ permeability, which results in an intracellular Ca²⁺ influx to cells in triggering multiple cellular responses (Zhang et al., 2020). Our previous work has revealed that capsaicin activates TRPV1 in skin melanocytes to boost PKC activity, resulted in the inhibition on melanin synthesis (Wu et al., 2020). To verify the role of TRPV1 in capsaicin’s effect on skin fibroblasts, we examined the Ca²⁺ influx in capsaicin-treated fibroblasts. The results indicated that capsaicin significantly induced Ca²⁺ influx, which could be blocked fully by pre-treatment of capsazepine, a TRPV1 antagonist (Figure 6). This result was consistent with that in our previous report on melanocyte. A23187, a calcium ionophore, was applied as a positive control here. In contrast, the transcriptional levels of MMPs were not affected by blocking TRPV1 with capsazepine in cultures (Figure 7A). In addition, capsazepine itself from 1 to 10 μM had no effect on the transcriptions of those MMP genes (Figure 7A). In parallel, the expressions of COL1A1, COL1A2, and COL3A1, as induced by capsaicin, were not affected by capsazepine at all (Figure 7B), suggesting that the induction of collagen and MMP by capsaicin should not be mediated by TRPV1.

Exposure to UVB is known to stimulate ROS generation in skin cells, i.e., keratinocytes and fibroblasts (Marionnet et al., 2010). In dermal fibroblasts, ROS decreased the production of collagen and glucosaminoglycan (Tanaka et al., 1993). The formation of ROS in cultured dermal fibroblasts was measured after UVB irradiation: the intracellular ROS was detected by incubation of the cells with DCFH-DA probe. UVB obviously stimulated ROS generation (green) in cultured fibroblasts (Figure 8A). The pre-treatment of capsaicin before UVB irradiation significantly decreased the amount of ROS to ∼1.5 folds of control; while N-acetyl-l-cysteine (NAC) was used as a positive control to reduce ROS formation (Figure 8A,B). The transcriptional activity of antioxidant response element (ARE) in fibroblasts was determined using a luciferase reporter system, i.e., pARE-Luc. The result was similar with that of ROS generation, in which the treatment of capsaicin dramatically inhibited the transcription of pARE-Luc, as reflected by the luciferase activity in pARE-Luc-transfected fibroblasts (Figure 8C). In addition, the collagen content was determined by using ELISA. The result indicated that capsaicin and NAC could attenuate the UVB-induced decrease on collagen content (Figure 8D), suggesting that ROS could be the cause of capsaicin-mediated collagen/MMP regulation.