Eighy female Institute of Cancer Research (ICR) mice, 6–7 weeks of age, were purchased from the Shanghai Experimental Animal Center (Shanghai, China) and kept in climate-controlled quarters with a 12-h light and dark cycle with food and water in cages under germ-free conditions. All experimental procedures were carried out following the Guide for the Care and Use of Laboratory Animals of Huai'an First People's Hospital, Nanjing Medical University (Nanjing, China) and before the animal experiments were performed, the procedures were approved by the Research Ethics Committee of Huai'an First People's Hospital, Nanjing Medical University (Nanjing, China).

ICR mice were randomly divided into four groups, 20 animals per group. The experimental design of the in vivo study is exhibited in Fig. 1A. All mice were shaved ahead of our study. In brief, the groups receiving DMBA/TPA, and DMBA/TPA+SR were first administered 60 µg DMBA dissolved in 0.2 ml to the naked backs. DMBA was administered to mice for two weeks, from week 1 to week 3. The first two weeks after skin tumor initiation with DMBA, animals in the DMBA/TPA and DMBA/TPA+SR groups were further exposed to 4 µg TPA twice a week for a total of 20 weeks ranging from week 3 to week 23. In addition, the mice treated with SR (20 and 40 mg/kg) were topically treated 30 min before each DMBA/TPA treatment five times a week until the sacrifice of the animals at week 22 (32–35). DMBA and TPA were purchased from Sigma-Aldrich (St. Louis, MO, USA). SR (>98% purity, molecular formula: C₁₄H₂₀O₇, CAS 10338-51-9) was purchased from Shanghai Ronghe Medicine Science and Technology Development Co, Ltd. (Shanghai, China). Sizes of the skin tumors >1 mm in diameter were measured every week using calipers. The dorsal skin of mice derived from different experiments was excised. After the fat was removed from the dorsal skin on ice, the skin tissue samples were placed in liquid nitrogen immediately for further research. The eye blood was collected for pro-inflammatory cytokine determination.

Normal human epidermal keratinocytes, HaCaT, were purchased from Combioer Biosciences Co., Ltd. (Nanjing, China). Human hypertrophic scar fibroblasts (HSFs) were purchased from Bioleaf Corp. (Shanghai, China). Human normal liver cell line L02 was obtained from the Cell Bank of the Type Culture Collection of the Chinese Academy of Science (Shanghai, China). All cells were cultured and maintained in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum (FBS) (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) and 1% penicillin/streptomycin (Gibco; Thermo Fisher Scientific, Inc.) in a humidified 5% CO₂ atmosphere at 37°C. SR used for the treatment of skin cancer was dissolved in DMSO (KeyGen Biotech Co., Ltd., Nanjing, China) and stored at −20°C, and then it was diluted in DMEM at the indicated concentrations for experimental treatment. The final DMSO concentration was no more than 0.1% (v/v) in each treatment.

To calculate the growth inhibitory role of SR in different cell lines, ~1×10³ cells/well were planted in 96-well plates (Corning Inc., Corning, NY, USA) with complete growth medium. On the following day, the cells were treated with different concentrations of SR ranging from 0 to 160 µM and incubated at 37°C for 24 h. Then, the cell viability was determined by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) assay at 570 nm.

The Annexin V-FITC/propidium iodide (PI) apoptosis detection kit was purchased from KeyGen Biotech Co., Ltd., to measure the apoptotic cell levels. All cells after the different treatments as described were harvested and washed with ice-cold PBS for twice, then incubated in a darkroom with Annexin V-FITC and PI for 15 min. Subsequently, the cells were analyzed by flow cytometry (BD Biosciences, San Jose, CA, USA). The percentage of cells undergoing apoptosis was quantified.

The eye blood was subjected to centrifugation at 12,000 × g for 10 min to carefully collect the supernatant. Then, TNF-α, IL-1β, IL-18, IL-6, COX2 and TGF-β1 levels in serum were determined using the Mouse TNF-α Quantikine ELISA kit (R&D Systems, Inc., Minneapolis, MN, USA), Mouse IL-1β ELISA kit (IL-1β) (Abcam, Cambridge, UK), Mouse IL-18 ELISA (R&D Systems, Inc.), Mouse IL-6 Quantikine ELISA kit (R&D Systems, Inc.), Mouse COX2 ELISA kit (Abcam) and Mouse TGF-β1 ELISA kit (Abcam) following the manufacturer's instructions.

Fixed skin and tumor tissues obtained from mice were embedded in paraffin blocks and 3-µm thick sections were cut. Skin and tumor sections were then deparaffinized and stained with hematoxylin and eosin (H&E) staining. The thickness of the skin epidermis was measured using Magnus Analytics Magnuspro software. Epidermal thickness of H&E staining sections was further assessed by ImageJ software (National Institutes of Health, Bethesda, MD, USA). For immunohistochemical images, the skin tissue sections were then exposed to HCl (3.5 M) for 20 min at room temperature and washed using PBS for 3 times. Subsequently, the skin tissue sections were treated with peroxidase (0.3%) to diminish endogenous peroxidase activity. Then, tissue sections were incubated with normal goat serum (5%) for 30 min followed by incubation with primary antibodies (anti-p53 antibody, ab131442; Abcam) at 1:100 dilution for 2 h at room temperature. The section was then incubated with HRP-conjugated compact polymer systems. Diaminobenzidine (DAB; ChemService, West Chester, PA, USA) was used as the chromogen according to the manufacturer's instructions. Apoptosis assay of tissue samples was determined by TUNEL using an In Situ Cell Death Detection kit, Fluorescein (Roche Applied Science, USA) following the manufacturer's protocol. Tumor tissue sections were counterstained with hematoxylin. Then, the number of TUNEL-positive cells was evaluated under a microscope. The ratio of apoptotic cells was determined by the ratio of the apoptotic cells to total cells.

As for the fluorescence assays, the cells were carefully harvested after various treatments and then fixed in 4% paraformaldehyde for 30 min. Then, the cells were incubated with primary antibody (p-NF-κB; Abcam) at 4°C overnight. Fluorophore-conjugated secondary antibodies were treated for 1 h at 25°C. The skin tissue sections were dried for 10 min at room temperature, fixed with chilled acetone for 10 min at −20°C, and washed with PBS for three times (5 min each). The pre-incubation was conducted with 5% normal rabbit serum at room temperature for 1 h, and sections were incubated with specific antibody: polyclonal rabbit anti-p-NF-κB (1:50; Abcam, Cambridge, MA, USA) at 4°C overnight. The Alexa Fluor 488 and 594 labeled anti-rabbit secondary antibodies (Invitrogen) were used in this part. Sections were then subjected to immunofluorescence staining via epifluorescence microscopy (Sunny Co.).

Total RNA was isolated from the skin tissue samples and HaCaT cells after various treatments using TRIzol reagent (KeyGen Biotech Co., Ltd.). Reverse transcription PCR was conducted using the PrimeScript RT Reagent kit (Takara Biotechnology Co., Ltd., Dalian, China) following the manufacturer's instructions and cDNA was then used as the template for the subsequent reactions. qPCR was conducted with SYBR Premix Ex Taq II obtained from Takara. The ABI-PRISM 7500 Sequence Detection System (Applied Biosystems Life Technologies, Foster City, CA, USA) was used according to the manufacturer's instructions. The primer sequences used in our study were commercially synthesized and are as follows. The mRNA level of GAPDH was used as the loading control. The 2^−∆∆Cq analyzing method was applied to evaluate the fold changes in mRNA levels in each group. The primers were as follows: TNF-α forward, 5′-CGAAAGGGAGTAGAAGTGCG-3′ and reverse, 5′-AAACATACAGAGCCGGCTAGCC-3′; IL-1β forward, 5′-ACATAGAGAGGGAGTACAC-3′ and reverse, 5′-CAGCGTAGATTACTAGTTCG-3′; IL-6 forward, 5′-GAGAGACGGAGTGGCCAC and reverse, 5′-CTCAAGTGAGAAGAGGCAACGGTAGT-3′; IL-18 forward, 5′-CTGATGAGCGGTCACAAGAAC-3′ and reverse, 5′-TTCTCTAACGCGTTAAGAGGAC-3′; TGF-β1 forward, 5′-TCGTGGAGCTCGAAGAACAC-3′ and reverse, 5′-TGGCTGACTTCACAACAGCGTA-3′; GAPDH forward, 5′-AACGGTGTCACAGACAGGCTCA-3′ and reverse, 5′-TCCACCTGACACGACACAACA-3′.

The western blot analysis was performed as previously described (36). Briefly, after treatments under different conditions, the skin cells were colllected and the medium was removed. Then, cells were washed with ice-cold PBS three times and lysed in ice-cold lysis buffer (pH 7.4, 50 mM Tris-HCl, 150 mM NaCl, 1 mM NaF, 1 mM ethyleneglycol-bis(aminoethylether)-tetraacetic acid, 1% NP-40, 1 mM phenylmethane-sulfonyl fluoride, and 10 µg/ml leupeptin) in the presence of fresh protease inhibitor cocktail. Frozen dorsal skins and epidermal and tumor of mice were obtained from the experimental mice treated under various conditions. Approximately 100 mg tissue sample was lysed with 1 ml lysis buffer. The cell lysates were centrifuged at 15,000 × g for 15 min at 4°C to collect the supernatant. BSA protein assay kit (Thermo Fisher Scientific, Inc.) was used to calculate the protein concentrations following the manufacturer's instructions. Protein extracts (40 ng) were separated by 10% SDS-PAGE and were then transferred to polyvinylidene fluoride membrane (PVDF; Millipore, Billerica, MA, USA). The PVDF membranes with proteins were blocked with 5% skim fat dry milk in 0.1% Tween-20 in Tris-buffered saline (TBS) for 2 h to block the non-specific sites on the blots. The primary antibodies dissolved in blocking buffer were used to detect the target protein blots at 4°C overnight for incubation. The bands on the PVDF membranes were visualized using chemiluminescence with Pierce ECL Western Blotting Substrate reagents (Thermo Fisher Scientific, Inc.). All experiments were performed in triplicate and carried out three times independently. The primary antibodies used in our study included: anti-p21 (1:1,000, ab86696), anti-PUMA (1:1,000, ab9643), anti-Bax (1:1,000, ab32503), anti-p53 (1:1,000, ab131442), anti-caspase-3 (1:1,000, ab90437), anti-IκBα (1:1,000, ab32518), anti-p-IκBα (1:1,000, ab133462), anti-NF-κB (1:1,000, ab16502), anti-p-NF-κB (1:1,000, ab86299) and GAPDH (1:1,000, ab8245) all from Abcam.

Data are expressed as the mean ± standard error of the mean (SEM). Statistical analyses were carried out by GraphPad Prism (version 6.0; GraphPad software) by ANOVA with Dunnet's least significant difference post hoc tests. P<0.05 was considered to indicate a statistically significant result.

In order to explore the chemopreventive effect of SR, DMBA-initiated and TPA-promoted mouse skin carcinogenesis in mice was first established in vivo. Fisrt, the body weight of mice was measured, and no significant difference was observed among the different groups, although reduced body weight was exerted in the DMBA/TPA-treated group (Fig. 1B). Compared to the Con group, DMBA/TPA-induced mice showed a significantly high incidence of papillomas, which was reduced by SR (Fig. 1C). In addition, DMBA/TPA exposure triggered a higher multiplicity of skin papilloma formation that was suppressed in the mice treated with SR, exhibiting a reduced number of tumors per mouse (Fig. 1D). Furthermore, the suppressive effect of SR on tumorigenesis was supported by the papilloma size distribution (Fig. 1E). Consistently, tumor area was increased by DMBA/TPA exposure, which was reduced after SR administration with an increase in time (Fig. 1F). Consistently, the tumor area was elevated by DMBA/TPA treatment, which was reduced after SR administration with the increase of time (Fig. 1F). In addition, the histological analysis further revealed that SR markedly ameliorated the increase in epidermal thickness (hyperplasia) in mice with DMBA/TPA induction (Fig. 2). In conclusion, the animal study above strongly indicated that SR efficiently prevented skin carcinogenesis induced by DMBA/TPA and TPA in mice.

Pro-inflammatory cytokines were also determined to assess the role of SR in DMBA/TPA-induced mice with skin cancer. As shown in Fig. 3A, serum TNF-α was higher in the DMBA/TPA-treated mice, which was enhanced with increasing time. SR also significantly suppressed TNF-α expression. Next, cytokines IL-1β (Fig. 3B), IL-18 (Fig. 3C), IL-6 (Fig. 3D), COX2 (Fig. 3E) and TGF-β1 (Fig. 3F) were all observed to have elevated expression in the DMBA/TPA-treated mice, which were suppressed by SR during the treatment procedure. Taken together, the findings above indicated that SR has a potential role in blocking the secretion of pro-inflammatory cytokines.

Next, we attempted to ascertain whether the NF-κB signaling pathway is also involved in SR-ameliorated skin cancer induced by DMBA/TPA. As shown in Fig. 4A, RT-qPCR analysis further indicated that pro-inflammatory cytokines, including TNF-α, IL-1β, IL-18, IL-6, and COX2, as well as anti-inflammatory factor TGF-β1, were increased in the tissue samples of DMBA/TPA-induced mice, which were all reduced by SR administration in a dose-dependent manner. The IκB/NF-κB signaling pathway was also explored. The data indicated that IκBα phosphorylation was upregulated in the DMBA/TPA-treated mice, while IκBα was downregulated. Phosphorylated NF-κB levels were also elevated due to DMBA/TPA induction in mice (Fig. 4B). Of note, SR exerted an inhibitory effect on IκBα and NF-κB phosphorylation, indicating its anti-inflammatory property. Furthermore, the immunofluorescence analysis revealed that enhanced NF-κB phosphorylated levels by DMBA/TPA were reduced after SR administration (Fig. 4C). Together, the results above demonstrated that SR inhibited skin carcinogenesis by impeding inflammation, linked to the suppression of the IκBα/NF-κB signaling pathway.

As it was described above, we found that inflammation blockage by SR might be a possible molecular mechanism by which to prevent skin carcinogenesis progression in DMBA/TPA-induced mice in vivo. In order to further confirm our findings above, the in vitro study was conducted. First MTT assays were used to calculate the safety of SR to normal cells. The results indicated that compared to the group in the absence of any treatment, no significant difference was observed among the various groups of treated cells as indicated (Fig. 5A). Thus, we supposed that SR might be safe for application with little cytotoxicity to normal cells. Normal human epidermal keratinocytes, HaCaT, were treated with 20 µg/ml DMBA for 48 h in the absence or presence of SR at 40, 80 and 160 µM. As shown in Fig. 5B, we found that cells exposed to DMBA displayed accelerated IκBα phosphorylation and reduced IκBα expression, which were reversed by SR administration. NF-κB was also activated by DMBA exposure. Significantly, SR treatment dose-dependently reduced NF-κB activity in the cells. RT-qPCR analysis indicated that SR suppressed high levels of pro-inflammatory cytokines, including TNF-α, IL-1β, IL-18, IL-6, and COX2, and TGF-β1 was also found to be reduced, indicating the attenuated inflammatory response, which were in line with the in vivo results (Fig. 5C). Moreover, the immunofluorescence assays also showed that SR reduced NF-κB phosphorylation caused by DMBA in normal human epidermal keratinocytes (Fig. 5D). Next, to further illustrate the role of SR in suppressing inflammation, the cells were pre-treated with SR (80 µM) for different times, ranging from 0 to 48 h. Then, they were exposed to 50 ng/ml TNF-α for another 1 h to induce inflammation. The western blot analysis indicated that IκBα and NF-κB were activated, which were markedly reduced by SR administration in a time-dependent manner (Fig. 6A, B and D). Oppositely, the downregulated level of IκBα due to TNF-α exposure was reversed by SR treatment (Fig. 6A and C). In conclusion, the data above indicate that SR, indeed, suppressed the inflammatory response in DMBA-treated cells or skin tissue samples, exhibiting its preventive effects on skin carcinogenesis.

Immunohistochemical analysis indicated that in the skin tissue samples of the DMBA/TPA-induced mice reduced TUNEL levels were noted when compared to the control group, revealing that the apoptotic response might be disrupted following DMBA/TPA treatment. In contrast, SR treatment significantly enhanced the percentage of TUNEL-positive cells, suggesting cell death during skin carcinogenesis (Fig. 7A and B). p53, an essential tumor suppressor, was found to be downregulated by DMBA/TPA, and in agreement with TUNEL alterations, SR administration reversed the p53 reduction (Fig. 7A and C). Additionally, western blot analysis illustrated that DMBA/TPA reduced p53 expression, while SR augmented p53 protein expression levels (Fig. 7D). In summary, the data above indicated that apoptosis might be involved in SR-regulated skin carcinogenesis caused by DMBA/TPA.

p53 can induce apoptosis, which is linked to Bax or other pro-apoptotic molecules (37). Therefore, we analyzed the expression of these proteins in the skin tissues of mice after various treatments. The immunoblot analysis showed a decrease in p21, PUMA, Bax, and cleaved caspase-3 in the DMBA/TPA-treated mice, which were significantly reversed by SR (Fig. 8). Next, the in vitro study was conducted to further confirm our data above. Human normal epidermal keratinocytes, HaCaT, exposed to DMBA were treated with or without SR at the indicated concentrations. Then, flow cytometric analysis indicated that DMBA caused reduced percentages of apoptotic cells, which was in line with the TUNEL assays in vivo. Notably, SR treatment enhanced apoptosis in the DMBA-treated cells (Fig. 9A and B). Finally, pro-apoptotic signals of p21, PUMA, Bax and cleaved caspase-3 protein levels were also reduced by DMBA in vitro, which were elevated after SR administration, indicating the role of SR in inducing apoptosis to avoid skin carcinogenesis in vitro (Fig. 9C).