Human dermal fibroblasts (HDFs) were purchased from Lonza (Basel, Switzerland) and cultured in Dulbecco's modified Eagle's medium (HyClone; GE Healthcare Life Sciences, Logan, UT, USA) supplemented with 10% fetal bovine serum (Hyclone; GE Healthcare Life Sciences) and 1% penicillin and streptomycin. Cells were transfected with 100 nM miRNA (miR)-378b mimic (5′-ACUGGACUUGGAGGCAGAA-3′; Bioneer Corporation, Dajeon, Korea), 100 nM anti-miR-378b (5′-UUCUGCCUCCAAUCCUGU-3′; Bioneer Corporation) or 100 nM scrambled control (AccuTarget™ Negative Control siRNA; Bioneer Corporation) using Lipofectamine^® RNAiMAX Transfection reagent (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA), according to the manufacturer's protocol. After transfection for 4 h, cells were washed with PBS and incubated for 24 h. Transfected HDFs were subsequently exposed to UVB using a Super light-VI UV illuminator (Boteck, Gunpo, Korea). After incubation for 24 h following UVB exposure, reverse transcription-quantitative polymerase chain reaction (RT-qPCR), western blotting and luciferase assay were performed.

Total RNA was isolated from cells using TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc.). cDNA was synthesized using 2 µg RNA with the miScript II RT kit (Qiagen GmbH, Hilden, Germany), according to the manufacturer's protocol. Following cDNA synthesis, the mRNA expression levels of COL1A1 and the gene encoding β-actin were detected using the StepOnePlus Real-Time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.), using EvaGreen™ premix (Solis BioDyne, Tartu, Estonia) and the following specific primers: Forward, 5′-AGGGCCAAGACATC-3′ and reverse, 5′-AGATCACGTCATCGCACAACA-3′ for human COL1A1; and forward, 5′-GGATTCCTATGTGGGCGACGA-3′ and reverse, 5′-CGCTCGGTGAGGATCTTCATG-3′ for human β-actin. Expression levels of miR-378b were detected with miR-378b specific primers (Qiagen GmbH) using the hsa-mir-378b miScript Primer assay (cat. no. MI0014154) and Hs_RNU6-2_11 miScript Primer assay (cat. no. MS00033740) from Qiagen GmbH, and the miScript SYBR^® Green PCR kit (Qiagen GmbH) with the StepOnePlus Real-Time PCR system. Forward primers were included in the miScript Primer assay kits and reverse primers were included in the miScript SYBR^® Green PCR kit. The expression levels of COL1A1 and miR-378b were normalized to β-actin and U6 respectively, using the 2^−ΔΔCq method (26). All RT-qPCRs were performed as follows: Initialization step at 94°C for 5 min, followed by 40 cycles (denaturing, 94°C for 30 sec; annealing, 60°C for 30 sec; polymerization, 72°C for 30 sec) and a final elongation step at 72°C, for 5 min. All experiments were repeated three times. Data was analyzed with Excel 2016 (Microsoft Corporation, Redmond, WA, USA) and presented as the mean value of viable cells ± standard deviation.

Predicted targets of miR-378b were identified using the bioinformatic analysis tool, microRNA.org (www.microrna.org). A luciferase reporter construct containing the predicted target sequence of miR-378b in the SIRT6 3′UTR was generated by ligating a region (+1,312 to +1,329) of the human SIRT6 gene into the XbaI restriction site, downstream of the luciferase gene in the pGL3 vector (Promega Corporation, Madison, WI, USA). HDFs were subsequently transfected with the 1 µg reporter assay vector and the 0.2 µg pSV-β-galactosidase control plasmid (Promega Corporation), with or without an miR-378b mimic or anti-miR-378b, using Lipofectamine RNAiMAX Transfection reagent (Invitrogen; Thermo Fisher Scientific) for 4 h, and subsequently incubated at 37°C for 24 h. After incubation for 24 h, luciferase and β-galactosidase assays were performed using Luciferase Assay Reagent (Promega Corporation) and the β-galactosidase Detection kit II (Clontech Laboratories, Inc., Mountainview, CA, USA), according to the manufacturer's protocol. Luciferase results were normalized using β-galactosidase activity.

Cells were lysed in radioimmunoprecipitation assay buffer, containing 1% NP-40, 150 mM NaCl, 10 mM Tris-HCl (pH 8.0), 1 mM EDTA and complete protease inhibitor cocktail (Roche Diagnostics, Basel, Switzerland). Extracted proteins (20 µg) were loaded onto 12% gels and separated by electrophoresis. Proteins were subsequently transferred onto nitrocellulose membranes (EMD Millipore, Billerica, MA, USA) and blocked with blocking buffer [(5% skim milk in TBS-Tween-20 buffer (50 mM Tris, 150 mM NaCl, 0.1% Tween 20)] at 25°C for 1 h. Protein expression levels of SIRT6 and β-actin were detected using rabbit anti-SIRT6 (1:2,000; D8D12; cat. no. 12486; Cell Signaling Technology, Inc., Danvers, MA, USA) anti-β-actin (1:10,000; N-21; cat. no. sc-130656; Santa Cruz Biotechnology, Inc., Dallas, TX, USA) primary antibodies, followed by anti-mouse IgG horseradish peroxidase (HRP)-conjugated (1:5,000; cat. no. 7076; Cell Signaling Technology, Inc.) and anti-rabbit IgG HRP-conjugated antibody (1;3,000; cat. no. 7074; Cell Signaling Technology, Inc.) secondary antibodies. Membranes were incubated with primary antibody at 25°C for 4 h, followed by incubation with secondary antibody at 25°C for 1 h. Proteins were visualized using SuperSignal™ West Pico Chemiluminescent Substrate (Thermo Fisher Scientific, Inc.). The intensity of each band was measured using ImageJ software Version 1.50 (National Institutes of Health, Bethesda, MD, USA).

Statistical significance was calculated using one-way analysis of variance with Tukey's post-hoc test. P<0.05 was considered to indicate a statistically significant difference using Excel 2016 (Microsoft Corporation, Redmond, WA, USA). Data are presented as the mean ± standard error.

The expression levels of various miRNAs alter during skin aging and in UVB-exposed cells (27). Therefore, to determine whether miR-378b is modulated by UVB, miR-378b expression was measured by RT-qPCR in UVB-exposed HDFs. miR-378b expression levels were significantly enhanced in cells exposed to 5–25 mJ/cm² UVB in a dose-dependent manner, compared to untreated cells (0 mJ/cm² UVB; P<0.05; Fig. 1A). In addition, in HDFs exposed to the same doses of UVB, expression levels of COL1A1 were reduced in a dose-dependent manner compared with untreated cells, and were inversely associated with miR-378b expression levels (Fig. 1B).

To determine whether the expression of COL1A1 is directly affected by miR-378b, the mRNA expression levels of COL1A1 in HDFs treated with anti-miR-378b or an miR-387b mimic were measured. Transfection with anti-miR-378b significantly up-regulated COL1A1 mRNA expression levels (P<0.05; Fig. 2A), whereas transfection with an miR-378b mimic resulted in a significant reduction in COL1A1 mRNA expression levels compared with cells transfected with a scramble control (P<0.05; Fig. 2B). However, the 3′UTR of COL1A1 does not contain a predicted binding site for miR-378b. In addition, the miR-378b mimic did not directly modulate the luciferase activity of a luciferase-COL1A1 3′UTR fusion construct in HDFs (data not shown). Therefore, a target prediction for miR-378b was performed using microRNA.org, which revealed that SIRT6 contained a binding site for the miR-378b seed sequence in its 3′UTR. SIRT6 has been identified as an anti-aging protein due to its ability to regulate the mRNA expression levels of COL1A1 and COL3A1 in HDFs (16). Therefore, it was hypothesized that miR-378b may directly regulate SIRT6 expression.

The majority of miRNAs regulate target genes by binding to their 3′UTRs (28,29). It was therefore determined whether the miR-378b mimic directly interacted with the SIRT6 3′UTR, by measuring the luciferase activity from a luciferase-SIRT6 3′UTR fusion construct. It was determined, using microRNA.org, that the seed sequence from miR-378b matched a region between +1,312 and +1,324 bp of the SIRT6 3′UTR (Fig. 3A). The SIRT6 3′UTR was subsequently cloned into a luciferase plasmid and transfected into HDFs. Co-transfection of an miR-378b mimic significantly reduced luciferase activity compared with pGL3-SIRT6-3′UTR only (P<0.05; Fig. 3B), whereas co-transfection with an miR-378b mimic and anti-miR-378b significantly enhanced luciferase activity compared with cells transfected with SIRT6 3′UTR and an miR-378b mimic (P<0.05; Fig. 3B). Thus, these results suggested that miR-378b directly binds to, and interferes with, SIRT6 mRNA.

It was subsequently determined whether the miR-378b mimic affected endogenous SIRT6 protein expression by western blot analysis. Overexpression of miR-378b reduced the expression levels of SIRT6 (Fig. 4A). Transfection with anti-miR-378b had the reverse effect and enhanced SIRT6 protein expression levels (Fig. 4B). These results suggested that miR-378b may inhibit COL1A1 mRNA expression via effects on SIRT6. In addition, a previous study demonstrated that SIRT6 regulates the expression of genes associated with stress and aging (30). In HDFs from younger individuals, SIRT6 was highly activated compared with HDFs from older individuals (31). Therefore, miR-378b, which was upregulated by UVB in the present study (Fig. 1A), may regulate aging and collagen expression via interference of translation of SIRT6 mRNA. Investigation of the association between miR-378b and skin aging in human skin samples is required in future studies to indicate whether miR-378b is a marker of photo-aging.