A total of 30 keloid samples were obtained from patients (19 men and 11 women; age, 22-46; mean age: 32.7±10.1 years) who underwent surgery at the First Affiliated Hospital of Jinzhou Medical University (Jinzhou, China), between December 2017 and July 2018. Patients who received radiotherapy/chemotherapy or laser treatment prior to surgery were excluded. Patients with malignancy and immune disorder of kidney dysfunction were also excluded from the present study. Biopsies of keloids and normal (≥10 cm from keloid) tissue samples were collected during surgery. All skin biopsies were snap-frozen in liquid nitrogen and subsequently stored at -80˚C until further experimentation. The present study was approved by the Ethics Committee of the First Affiliated Hospital of Jinzhou Medical University (Jinzhou, China; approval no. JYD160920). Written informed consent was provided by all patients prior to the start of the study.

Human keloid fibroblasts and normal fibroblasts were purchased from the Bank of Type Culture Collection of the Chinese Academy of Sciences. Samples were maintained in DMEM (Thermo Fisher Scientific, Inc.) supplemented with 10% FBS, 100 U/ml penicillin and 100 µg/ml streptomycin (HyClone; GE Healthcare Life Science), at 37˚C in a humidified atmosphere with 5% CO₂.

In order to establish the circ_101238 and CDK6 knockdown model, short hairpin (sh)RNA sequences targeting circ_101238 (sh-circ_101238; cat. no. C02097) and a control (sh-NC; cat. no. C02098) and small interfering (si)RNA against CDK6 (si-CDK6; cat. no. A01015-297) and the control (si-NC; cat. no. A01015-298) were synthesized by Shanghai GenePharma Co. Ltd. miR-138-5p mimics and inhibitors, together with the negative control (miR-NC) were also purchased from Shanghai GenePharma Co. Ltd. In order to establish the circ_101238 and CDK6 overexpression model, wild-type (WT; LV-circ_101238 and LV-CDK6; cat. no. C02015-383 and C02015-385) or mutant (MUT; LV-NC; cat. no. C02015-384) fragments were inserted into the PLCDH-cir vector (Guangzhou Ribobio, Co., Ltd.). The quantity of each shRNA, siRNA or miRNA mimics/inhibitors was 50 pg/ul for each transfection. The lentiviral vector was generated by Hanbio Biotechnology Co., Ltd. A total of 40 nM vector were transfected into the fibroblasts using Lipofectamine^® 2000 (Takara Biotechnology Co., Ltd.), according to the manufacturer's protocol. Fibroblasts were then selected using 0.5 µg/ml puromycin (Sigma-Aldrich; Merck KGaA) for 2 weeks. The culture medium was replaced with fresh DMEM supplemented with FBS, 8 h post-transfection. Transfection efficiency was confirmed via reverse transcription-quantitative (RT-q)PCR analysis. The subsequent experiments were carried out 24 h after transfection.

miRNA was extracted using the miRNeasy Mini kit (Qiagen China Co., Ltd.). TaqMan microRNA assay (Applied Biosystems; Thermo Fisher Scientific, Inc.) was performed to assess miR-138-5p expression, and qPCR was subsequently performed using Applied Biosystems^® 7500 (Thermo Fisher Scientific Inc.).

Total RNA was extracted from tissues/cells using TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc.), according to the manufacturer's protocol, and protein concentration was determined using a NanoDrop 1000 spectrophotometer (Thermo Fisher Scientific, Inc.). The quality of extracted RNA was assessed using the Agilent 2100 Bioanalyzer (Agilent Technologies GmbH), and the 28S:18S ratios were >1.0. cDNA was synthesized using the PrimeScript^™ RT kit (Takara Biotechnology Co., Ltd.), the sample was incubated at room temperature for 30 min, 42˚C for 45 mins, 99˚C for 5 mins and 5˚C for 5 mins in a PCR cycler. qPCR was subsequently performed using the SYBR Green PCR Master Mix (Takara Biotechnology Co., Ltd.), according to the manufacturer's protocols. The following primer sequences were used for qPCR: circ_101238 forward, 5'-GCTGTGAGCCGATAGCTAAC-3' and reverse, 5'-ACTCCAACACCATAGCGGAG-3'; miR-138-5p forward, 5'-AGCTGGTGTTGTGAATCAGGCCG-3' and reverse, 5'-CTCGACAACCCGCAGAGGCG-3'; CDK6 forward, 5'-TGCACAGTGTCACGAACAGAC-3' and reverse, 5'-TGAATGAAAAGCCTGCCTGGG-3'; GAPDH forward, 5'-AGAAGGCTGGGGCTCATTTG-3' and reverse, 5'-AGGGGCCATCCACAGTCTTC-3'; and U6 forward, 5'-CTCGCTTCGGCAGCACATA-3' and reverse, 5'-AACGATTCACGAATTTGCGT-3'. The following thermocycling conditions were used for qPCR: Initial denaturation at 95˚C for 5 min, followed by 45 cycles of denaturation at 95˚C for 15 sec, annealing at 60˚C for 20 sec and elongation at 72˚C for 10 sec, final extension was at 72˚C for 10 min Relative expression levels were calculated using the 2^-∆∆Cq method (13) and normalized to the internal reference genes GAPDH (mRNA) and U6 (miRNA).

Cells were harvested 24 h post-transfection. A total of 2x10⁴ cells were seeded into 96-well plates and the cell counting kit-8 (CCK-8) assay (Beyotime Institute of Biotechnology) was performed at day 1, 2, 3 and 4 following cell inoculation. Briefly, 10 µl of CCK-8 solution was added into each well at the corresponding time points. Following incubation at 37˚C for 2 h, cell proliferative activity was analyzed at a wavelength of 450 nm, using a microplate reader (Bio-Rad Laboratories, Inc.).

Transfected cells were placed into 6-well plates at a density of 2x10⁵ cells/well and centrifugated at 5,000 x g for 5 min at room temperature. Cell pellets were rinsed with PBS and re-suspended into the wells. Cells were incubated with 5 µl propodium iodide (Thermo Fisher Scientific) in the dark for 30 min at 4˚C and subsequently stained with 5 µl Annexin V-FITC for 15 min at room temperature (cat. no. ab1408; Abcam). Apoptotic cells were subsequently analyzed using a flow cytometer (BD Biosciences) and FlowJo software (version 7.6; FlowJo LLC).

The Targetscan (www.targetscan.org) and miRanda (www.microrna.org/microrna) databases were used to predict the potential targets of circ_101238 and miR-138-5p.

The WT fragments of the 3'-untranslated region (UTR) of circ_101238/CDK6, with the predicted binding sites of miR-138-5p, were purchased from Shanghai GenePharma Co., Ltd. The fragments were cloned into the pmirGLO Dual-Luciferase miRNA Target Expression vector (Promega Corporation), according to the manufacturer's protocol. The circ_101238/CDK6-3'-UTR-MUT plasmid containing the mutant miR-138-5p binding site was generated using the QuikChange Multi Site-Directed Mutagenesis kit (Agilent Technologies, Inc.). The corresponding vectors were subsequently used to co-transfect 293 cells (Bank of Type Culture Collection of the Chinese Academy of Sciences) with miR-138-5p mimics or the control using Lipofectamine^® 2000 (Takara Biotechnology Co., Ltd.). Following incubation for 48 h at 37˚C with 5% CO₂, firefly and Renilla luciferase activities were detected using the Dual-Luciferase Reporter Assay System (Promega Corporation), according to the manufacturer's protocol. Firefly luciferase activity was normalized to Renilla luciferase activity.

Total protein was extracted from fibroblasts using radioimmunoprecipitation assay buffer (Beyotime Institute of Biotechnology). Protein concentration was measured using a bicinchoninic assay (Beyotime Institute of Biotechnology) and 30 µg protein/lane was separated via SDS-PAGE using 10% gels. The separated proteins were subsequently transferred onto nitrocellulose membranes (EMD Millipore) and blocked with TBS containing 5% skimmed milk at room temperature for 1 h. The membranes were incubated with primary antibodies against: CDK6 (dilution 1:3,000; cat. no. ab151247; Abcam), caspase-3 (dilution 1:500; cat. no. ab4051; Abcam) and GAPDH (dilution 1:10,000; cat. no. ab8245; Abcam) overnight at 4˚C. Following the primary incubation, membranes were incubated with horseradish peroxidase-conjugated anti-rabbit (dilution 1:5,000; cat. no. sc-2357; Santa Cruz Biotechnology, Inc.) or anti-mouse IgG (dilution 1:5,000; cat. no. sc-2371; Santa Cruz Biotechnology, Inc.) secondary antibodies at room temperature for 1 h. Protein bands were visualized using the enhanced chemiluminescence protein detection kit (Pierce; Thermo Fisher Scientific, Inc.), and the signal was analyzed using Image J software (version 1.48; National Institutes of Health).

Statistical analysis was performed using SPSS software 21.0 (IBM Corp.) and data are presented as the mean ± standard deviation. Paired Student's t-test was used to compare differences between two groups, while ANOVA, followed by Dunnett's post-hoc test were used to compare differences between multiple groups against the control. Pearson's correlation analysis was performed to determine the correlation between RNA expression levels. All experiments were performed in triplicate. P<0.05 was considered to indicate a statistically significant difference.

The expression levels of circ_101238 were determined in 30 keloid and matched adjacent normal tissue samples via RT-qPCR analysis. The results demonstrated that circ_101238 expression was significantly upregulated in keloid samples compared with the control group (P<0.05; Fig. 1A). Similarly, upregulated circ_101238 expression was also detected in keloid fibroblasts compared with the control group (P<0.05; Fig. 1B). In order to perform the gain- and loss-of-function experiments, circ_101238 knockdown and overexpression models were established by transfecting fibroblasts with sh-circ_101238 and LV-circ_101238, respectively. Transfection efficiencies were confirmed via RT-qPCR analysis (P<0.05; Fig. 1C and D).

In order to investigate the regulatory functions of circ_101238 on the proliferation and apoptosis of keloid fibroblasts, circ_101238 was knocked down in fibroblasts using lentiviral vectors. The CCK-8 assay demonstrated that the proliferative ability of keloid fibroblasts transfected with sh-circ_101238 significantly decreased compared with the control (sh-NC) group (P<0.05; Fig. 1E). Furthermore, mRNA and protein expression levels of the proliferation-associated molecule, CDK6 were significantly decreased following transfection with sh-circ_101238 (P<0.05; Fig. 1F and G). Apoptosis was also assessed in transfected cells via flow cytometric analysis, and the results indicated that the apoptotic rate was significantly increased in keloid fibroblasts transfected with sh-circ_101238 compared with the sh-NC group (P<0.05; Fig. 1H and I). Furthermore, mRNA and protein expression levels of the apoptosis marker, caspase-3 were significantly increased following transfection with sh-circ_101238 (P<0.05; Fig. 1J-L). Taken together, these results suggest that the growth of keloid fibroblasts may be inhibited by knockdown of circ_101238 expression.

In order to confirm whether circ_101238 is a novel oncogenic factor in keloids, and determine its potential roles through targeting downstream miRNAs, the complementary binding sites between miR-138-5p and circ_101238 were predicted via bioinformatics analysis (Fig. 2A). The association between circ_101238 and miR-138-5p was verified via the dual-luciferase reporter assay, in which luciferase reporters containing WT (circ_101238-WT) or MUT (circ_101238-MUT) fragments of potential miR-138-5p binding sites were generated. The results demonstrated that miR-138-5p mimics significantly suppressed luciferase activity of the circ_101238-WT vector compared with the control (P<0.05; Fig. 2B). The expression level of miR-138-5p was significantly increased in keloid fibroblasts transfected with sh-circ_101238 and the levels of miR-138-5p were reduced in cells transfected with LV-circ_101238 (P<0.05; Fig. 2C and D). Furthermore, expression levels of miR-138-5p were significantly decreased in keloid samples compared with the control group, where the levels of circ_101238 and miR-138-5p were inversely correlated (P<0.05; Fig. 2E and F).

Cells were transfected with miR-138-5p mimics or inhibitors and the transfection efficiencies were confirmed via RT-qPCR analysis (P<0.05; Fig. 2G). Transfection with miR-138-5p inhibitors significantly induced proliferation compared with the control group (P<0.05; Fig. 2H). Furthermore, mRNA and protein expression levels of CDK6 were significantly increased in keloid fibroblasts transfected with miR-138-5p inhibitors compared with the control group (P<0.05; Fig. 2I and J). Apoptosis was also assessed in the transfected cells, and the results demonstrated that the apoptotic rate was significantly decreased following transfection with miR-138-5p inhibitors (P<0.05; Fig. 2K and L), which was confirmed by downregulated expression levels of caspase-3 (P<0.05; Fig. 2M-O).

In order to identify the promising targets of miR-138-5p, the binding sites on miR-138-5p and CDK6 transcripts were predicted (Fig. 3A). The association between miR-138-5p and CDK6 was verified via the dual-luciferase reporter assay, in which luciferase vectors containing WT (CDK6-WT) or MUT (CDK6-MUT) fragments of predicted miR-138-5p binding sites were generated. The results demonstrated that miR-138-5p mimics significantly suppressed luciferase activity of the plasmids containing CDK6-WT, but not those containing CDK6-MUT (P<0.05; Fig. 3B). Furthermore, RT-qPCR and western blot analyses indicated that CDK6 expression levels were downregulated in keloid fibroblasts transfected with miR-138-5p mimics compared with the control group (P<0.05; Fig. 3C and D). Notably, the expression levels of miR-138-5p were significantly increased in the keloid tissues compared with the control group, where the levels of miR-138-5p and CDK6 were inversely correlated (P<0.05; Fig. 3E and F).

In order to further identify the effects of miR-138-5p and CDK6 in circ_101238-regulated biological behavior changes within keloid fibroblasts, cells were co-transfected with LV-circ_101238 and miR-138-5p mimics or miR-138-5p mimics/inhibitors and si-CDK6/LV-CDK6, respectively. The results indicated that the enhanced cell proliferation and inhibition of apoptosis induced by LV-circ_101238 were significantly counteracted following transfection with miR-138-5p mimics (P<0.05; Fig. 4A-C). Furthermore, enhanced cell proliferation induced by miR-138-5p inhibitors was significantly reversed following CDK6 knockdown, whereas the effects on cell proliferation caused by miR-138-5p mimics were significantly attenuated following overexpression of CDK6 (P<0.05; Fig. 4D-I). Taken together, these results suggest that circ_101238 regulates the growth of keloid fibroblasts via miR-138-5p/CDK6 signaling.