This article is mentioned in:

Introduction

The dermal papilla (DP) describes the component of the hair follicle that is involved in hair growth and formation. During development of the hair follicle, the DP is generated by condensation of dermal mesenchymal cells. The DP exists at the base of hair follicles and regulates the hair cycle by providing key signals that control the timing and phase of hair follicle growth and formation (1,2). Anagen is the active growth phase of hair follicles, where the root divides rapidly and adds material to the hair follicle for rapid hair growth. DP stimulates the initiation of the anagen growth phase by secreting Fgf7/10 and TGFβ (3,4). DP-induced β-catenin/Wnt signaling sustains the anagen phase and DP-induced Notch/Wnt5a signaling induces hair follicle differentiation (5,6). The production of reactive oxygen species (ROS) in hair follicles results in hair developmental disorders, including graying and hair loss (7–9). One such example is alopecia, which is induced by a ROS-mediated reduction of hair growth (9,10).

Oxidative stress that results in crucial damage of DNA, proteins and lipids is implicated in several hair follicle disorders, including graying and hair loss (7–10). Oxidative stress is increased by the accumulation of ROS, including hydrogen peroxide (H2O2), hydroxyl radical (OH•), superoxide anion (O2−) and the accumulation of reactive nitrogen species (RNS), including peroxynitrite (ONOO−). The generation of ROS and RNS is triggered by an imbalance between pro-oxidants and antioxidants. It has been identified that high levels of oxidative stress induces growth arrest, apoptosis and necrosis in cells (11). Mammalian cells have developed defense mechanisms to neutralize ROS, including antioxidant enzymes and non-enzymatic antioxidants (12,13). In hair follicles, low levels of ROS are generated in the mitochondria, and these act as crucial signaling molecules for hair follicle differentiation and morphogenesis (7). However, high levels of ROS directly damages cellular membranes, lipids, proteins and DNA (9). ROS levels increase with aging, which causes a decrease in the function and number of functional melanocyte cells in hair follicles (8). ROS-responsive microRNA (miRNA) expression regulates the cell cycle and apoptosis in a variety of cells (14–16).

miRNAs are short oligonucleotides, consisting of ~19–24 nucleotides (17). miRNAs repress the translation of their target genes by binding to partly complementary sequences in the 3′ untranslated region of the target mRNA (18). miRNAs are involved in the control of diverse cellular processes, including cell growth, apoptosis, development, metabolism, stress adaptation, hormone signaling and differentiation (18–22). In hair follicles, a deficiency of miRNAs induced by the knockout of Dicer and Drosha blocks the anagen developmental phase by repression of the catagen phase (23). It has been hypothesized that the miRNA miR-31 is involved in hair follicle growth and hair fiber formation because it targets Krt16, Krt17, Dlx3 and Fgf10 (24). A recent study compared the miRNA expression profiles in balding and non-balding dermal papilla (25).

To the best of our knowledge, no previous studies have reported the miRNA expression profile in H2O2-treated DP. Therefore, in the present study, we analyzed changes in the miRNA expression profiles in DP alone and in DP that had been treated with H2O2. The target genes of significant miRNAs (those with >2-fold changes in expression) were predicted by an in silico prediction algorithm. Based on these data, we derive a model indicating that H2O2-specific miRNAs regulate ROS-responsive cellular functions.

Materials and methods

Cell culture

Human dermal papilla (HDP) cells were purchased from Cellbio Inc. (Seoul, Korea). HDPs were maintained as a monolayer culture in Dulbecco’s modified Eagle’s medium (DMEM; Gibco-Invitrogen Life Technologies, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS; Sigma-Aldrich, St. Louis, MO, USA) and 1% penicillin/streptomycin. HDPs were cultured in a humidified chamber with 5% CO2 at 37°C.

Cell viability

HDPs (5×103 cells) were plated in 96-well culture plates and treated with H2O2 for 24 h under the growth conditions described above. The 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) colorimetric assay was performed by adding 0.5 μg/ml MTT (Sigma-Aldrich) to the culture medium. Following 1 h of incubation under normal cell growth conditions, the growth medium was removed and 200 μl of dimethyl sulfoxide was added to each well. The absorbance was measured at 490 nm using a microplate reader (iMark; Bio-Rad, Hercules, CA, USA).

Cell cycle analysis

The cell cycle was determined by flow cytometry via propidium iodide (PI) staining. Cell cycle analysis was performed as described previously (26).

RNA preparation and miRNA microarray

Total RNAs were extracted by RiboEX (GeneAll Biotechnology Co., Ltd., Seoul, Korea) and quantified by measuring the optical density ratio. The miRNA microarray (SurePrint G3 Human v16.0 miRNA 8×60K; Agilent Technologies, Santa Clara, CA, USA) was performed according to the manufacturer’s instructions. miRNA was stained by pCp-Cy3 (Agilent Technologies), combined with T4 ligase (Agilent Technologies) and hybridized to a probe on the microarray. The microarray was imaged using the Agilent microarray scanner and digitized by Feature extraction. The digitized data were analyzed for fold change, miRNA potential target and gene ontology (GO) using Genespring GX version 11.5 (Agilent Technologies).

Prediction of miRNA target genes and GO analysis

The putative target genes of significantly up and downregulated miRNAs were identified by the web tool TargetScan. Target gene prediction was performed on significant miRNAs with 50 context score percentile using the conserved and non-conserved database. The putative target genes were identified and sorted by GO of each gene.

Statistical analysis

Statistical significance was determined by Student’s t-test. P<0.05 was considered to indicate a statistically significant difference.

Results

H2O2 treatment decreases the viability of HDPs via cell cycle arrest or apoptosis

To determine the cytotoxicity of H2O2, we initially examined the viability and status of the G1 or G2/M population in HDPs. HDPs were treated with H2O2 (0–1,000 μM) and the resulting viability was measured using the MTT assay (Fig. 1A). Following 24 h of H2O2 treatment, a decrease in cell viability occurred in a dose-dependent manner. In the presence of 750 and 1,000 μM H2O2, a significant (P<0.05) decrease in cell viability to 71.23 and 50.60% of that in control cells, respectively, was observed (Fig. 1A). In the presence of 750 μM H2O2, HDPs exhibited a 21.07% reduction in G1 phase and an 11.66% increase in G2/M phase, compared with those of the control (Fig. 1B). These data suggest that H2O2 decreases cell viability of HDPs by inducing cell cycle arrest at the G2/M phase and cell death.

Figure 1 H2O2 repression of cell growth via cell cycle arrest and apoptosis in HDP cells. (A) Growth properties of H2O2-treated HDPs. HDPs (5×103) were seeded in 96-well plates and treated with the indicated H2O2 concentrations, incubated for 24 h and growth properties were measured using the MTT assay. Cell viability (average ± SD) was determined from triplicate experiments (*P<0.05). (B) HDPs (2×106) were seeded in a 60-mm culture dish and treated with 0–750 μM H2O2, incubated for 24 h and processed for flow cytometry by PI staining. Graphical representation of FACS data, bar graph represents percentage of total cell population in each cell cycle phase (± SD). H2O2, hydrogen peroxide; HDP, human dermal papilla; PI, propidium iodide.

Identification of H2O2-induced changes in miRNA expression in HDPs

H2O2-induced changes in the miRNA expression profiles were analyzed using the Agilent miRNA microarray, which contains 1,368 probes that are able to detect 1,205 human miRNAs. The fluorescence intensity data for each sample was normalized by global normalization. To eliminate disorderly data, miRNA expression data were selected by requiring a present-flag for at least one of all samples. Using this method, 155 miRNAs were selected out of 1,205 human miRNAs detected by present-flag selection. To identify H2O2-regulated miRNAs, the refined data were compared for control HDPs grown under normal conditions and HDPs treated with 750 μM H2O2 for 24 h. The results demonstrated that the expression levels of 68 miRNAs were altered at least 1.5-fold in response to treatment with H2O2. The 68 miRNAs are presented in Fig. 2. Fold-change analysis revealed that 62 miRNAs are upregulated and 6 miRNAs are downregulated at levels of 1.5-fold or greater in control HDPs grown under normal conditions and HDPs treated with 750 μM H2O2 for 24 h (Table I).

Figure 2 Changes in miRNA expression levels in H2O2-treated HDP cells. (A) Total RNA was extracted from control HDPs and HDPs treated with 750 μM H2O2 for 24 h. miRNA microarray was performed as described in Materials and methods. To obtain the fold change in expression levels, the fluorescence intensity of each miRNA was compared between normal HDPs and H2O2-treated HDPs. Changes in miRNA expression >2-fold are illustrated by heat map. Blue indicates a relatively low expression and red indicates a relatively high expression. (B) Up and downregulated miRNAs in response to H2O2 treatment were counted and represented by the bar graph. miRNA, microRNA; H2O2, hydrogen peroxide; HDP, human dermal papilla.

Table I Up and downregulated miRNAs in H2O2-treated HDPs.

Table I

Up and downregulated miRNAs in H2O2-treated HDPs.

A, Upregulated
miRNA	Fold change	Chr	miRNA	Fold change	Chr	miRNA	Fold change	Chr
hsa-let-7i-5p	1.52	chr12	hsa-miR-19a-3p	1.52	chr13	hsa-miR-361-5p	1.86	chrX
hsa-miR-101-3p	1.91	chr1	hsa-miR-20b-5p	1.56	chrX	hsa-miR-3663-3p	2.73	chr10
hsa-miR-106b-5p	1.59	chr7	hsa-miR-27a-3p	2.13	chr19	hsa-miR-376a-3p	1.65	chr14
hsa-miR-107	1.55	chr10	hsa-miR-27b-3p	1.56	chr9	hsa-miR-376b-3p	2.39	chr14
hsa-miR-1207-5p	2.08	chr8	hsa-miR-28-5p	2.17	chr3	hsa-miR-4281	2.15	chr5
hsa-miR-1225-5p	1.62	chr16	hsa-miR-2861	2.29	chr9	hsa-miR-4306	1.65	chr13
hsa-miR-1246	1.60	chr2	hsa-miR-29a-3p	1.60	chr7	hsa-miR-487b	1.51	chr14
hsa-miR-1275	1.75	chr6	hsa-miR-29b-1-5p	2.14	chr7	hsa-miR-503-5p	1.66	chrX
hsa-miR-128	1.54	chr2	hsa-miR-29b-3p	3.26	chr1	hsa-miR-574-3p	2.01	chr4
hsa-miR-130a-3p	1.63	chr11	hsa-miR-30a-5p	3.01	chr6	hsa-miR-630	1.66	chr15
hsa-miR-138-5p	2.84	chr3	hsa-miR-31-3p	1.58	chr9	hsa-miR-638	2.33	chr19
hsa-miR-140-5p	1.55	chr16	hsa-miR-3162-5p	2.00	chr11	hsa-miR-654-3p	1.63	chr14
hsa-miR-150-3p	3.95	chr19	hsa-miR-3195	1.81	chr20	hsa-miR-758-3p	3.00	chr14
hsa-miR-151a-5p	1.52	chr8	hsa-miR-3196	1.67	chr20	hsa-miR-762	2.34	chr16
hsa-miR-15a-5p	1.76	chr13	hsa-miR-320c	1.52	chr18	hsa-miR-92a-3p	1.51	chr13
hsa-miR-17-5p	1.56	chr13	hsa-miR-320d	1.52	chr13	hsa-miR-99b-5p	2.01	chr19
hsa-miR-185-5p	1.64	chr22	hsa-miR-324-3p	2.38	chr17	hsa-miR-377-3p	1.55	chr14
hsa-miR-1915-3p	2.81	chr10	hsa-miR-331-3p	1.51	chr12	hsa-miR-381-3p	2.03	chr14
hsa-miR-193a-3p	3.65	chr17	hsa-miR-337-5p	2.52	chr14	hsa-miR-411-5p	2.47	chr14
hsa-miR-193b-3p	1.56	chr16	hsa-miR-34a-5p	2.04	chr1	hsa-miR-424-5p	1.72	chrX
hsa-miR-199a-5p	1.61	chr1	hsa-miR-34b-5p	2.19	chr11
B, Downregulated
miRNA	Fold change	Chr	miRNA	Fold change	Chr	miRNA	Fold change	Chr
hsa-miR-29c-3p	−1.51	chr1	hsa-miR-137	−1.50	chr1	hsa-miR-376c-3p	−1.56	chr14
hsa-miR-20a-5p	−1.51	chr13	hsa-miR-423-5p	−1.57	chr17	hsa-miR-3665	−2.36	chr13

[i] miRNAs, microRNAs; H₂O₂, hydrogen peroxide; HDPs, human dermal papilla cells.

Identification of H2O2-specific miRNA putative target genes and GO analysis

Our study identified 68 novel miRNAs that were significantly up or downregulated in response to H2O2 treatment. As miRNA functions as RNA interference during mRNA translation, we predicted that H2O2-specific miRNAs may regulate all or a number of the H2O2-response genes. Therefore, the target genes of H2O2-specific miRNAs were analyzed using the bioinformatics target gene prediction program TargetScan. For target gene analysis, default parameters were utilized, 50 context score percentile in the conserved and nonconserved database. H2O2-induced miRNAs targeted 14,046 genes; H2O2-repressed miRNAs targeted 6,019 genes. To identify the cellular functions of the putative target genes, GO analysis was performed, which is a method that categorizes the genes according to the cellular function classified for a standard for each gene. As illustrated in Fig. 3, GO analysis identified the following cellular functional activities: for molecular function, catalytic (17.01%), nucleic acid binding transcription factor (6.67%), enzyme regulator (2.62%), molecular transducer (2.61%), binding (67.63%) and transporter (3.45%); for biological process, signaling (7.29%), biological adhesion (1.65%), multicellular organismal (4.11%), cellular (18.52%), metabolic (13.96%), cellular component organization or biogenesis (0.63%), immune system (0.01%), biological regulation (14.89%), establishment of localization (6.16%), localization (6.21%), response to stimulus (7.15%), single organism (15.35%) and developmental (4.06%); for cellular component part, cell (25.45%), membrane (10.69%), organelle (1.93%), extracellular region (0.75%), extracellular matrix (0.2%), organelle (15.01%), membrane-enclosed lumen (1.05%), cell junction (0.96%), extracellular matrix (0.77%), membrane (14.3%), cell (25.45%), extracellular region (2.82%) and synapse (0.61%). All of these were implicated in UVB-mediated responses in HDPs. These GO annotations provided comprehensive information on the function of H2O2-regulated transcripts in HDPs. TargetScan was used to predict the gene targets of the top five miRNAs that demonstrated the greatest increase or decrease in expression levels. The putative miRNA target genes were sorted into cell cycle, apoptosis and cell growth, and proliferation-related GO (Tables II and III). Cell cycle-related GO included cell cycle (GO:0007049), cell cycle arrest (GO:0007049), negative regulation of cell cycle (GO:0045786) and regulation of cell cycle (GO:0051726). Apoptosis-related GO included apoptotic process (GO:0006915), apoptotic signaling pathway (GO:0097190), cell death (GO:0008219), death (GO:0016265), programmed cell death (GO:0012501), regulation of apoptotic process (GO:0042981), regulation of cell death (GO:0010941), regulation of execution phase of apoptosis (GO:1900117) and regulation of programmed cell death (GO:0043067). Cell growth-related and cell proliferation-related GO included positive regulation of cell proliferation (GO:0008284), regulation of cell growth (GO:0001558), regulation of cell proliferation (GO:0042127) and regulation of growth (GO:0040008).

Figure 3 GO enrichment analysis of the target genes of up and downregulated miRNAs. Putative target genes of H2O2-specific miRNAs were identified by TargetScan and the cellular functions were predicted by GO analysis. The charts represent the GO analysis performed by Genespring GX. GO, gene ontology; miRNAs, microRNAs; H2O2, hydrogen peroxide.

Table II Predicted target genes of the top five most upregulated miRNAs in H2O2-treated HDPs.

Table II

Predicted target genes of the top five most upregulated miRNAs in H2O2-treated HDPs.

	Target genes and functions
miRNA	Cell cycle	Apoptosis	Cell growth and proliferation
hsa-miR-150-3p	-	-	-
hsa-miR-193a-3p	PSRC1, JUB, CCNT2, CCPG1, DLGAP5, PAPD5, SIAH1, TSC1, CABLES2, SMC1A, ERBB2IP, SEP3, KLK10, PTEN, CCND1, NEDD9, NF1, NF2, SUFU, CTF8, SASS6, E2F6, SENP5, VASH1, SPECC1L, CADM1, APPL1, GAS1, ING1, IRF1, FOXO4, ATM	BCL2L10, DIDO1, WDR92, RFFL, ZNF346, CADM1, RNF144B, GAS1, TNFRSF21, IL1A, JAK2, MCL1, MLL, OSM, SH3GLB1, PDCD2, DRAM, PSEN1, BAG1, BCL2L2, RTKN, ATXN1, ELMO2, SIAH1, TIAL1, TNFAIP3, TNFRSF1B, TRAF1, PHLDA2, DYRK2, AIFM2, TNFSF9, TNFRSF10B, SQSTM1, EBAG9, TP53INP1, ARHGEF6, PPM1F	BMPR2, TDGF1, KAZALD1, PPP1R9B, WISP1, EBAG9, SOCS6, CIAO1, NET1, ENOX2, TNFSF13B, FLT1, FLT3, HOXD13, IGFBP5, KIT, KRAS, NEDD9, NODAL, OSM, CCDC88A, PRKCQ
hsa-miR-29b-3p	PPM1D, CDC23, CCND2, CCNF, UBA3, CCNT2, AURKB, HDAC4, DCLRE1A, SUPT5H, TACC1, PTP4A1, CDC7, GSG2, SMPD3, RCC2, PTEN, NEDD9, ZAK, XRN1, CDK2, CDK6, HTATIP2, CGREF1, STAG2, SEP9, DBF4, MAPRE2, E2F7, SLC5A8, AHR, VASH1, MAPRE1, CLASP2, SEP6	FEM1B, HTATIP2, IL24, SLC5A8, AHR, FOXO3, PPP1R13B, SIRT1, ZNF346, CECR2, BIRC2, FAS, IL2RA,, MCL1, SH3GLB1, ZAK, CYCS, DIABLO, DUSP22, BIRC6, BAK1, ATXN1, CIDEC, SGK1, ISG20L1, TNFAIP3, TNFRSF1A, TRAF5, FAM130A1, CASP7, AIFM2, BMF, RABEP1, TP53INP1, TRAF4, SLK	BIRC6, PURA, TRAF5, VEGFA, EPC1, CDC7, CREG1, WISP1, CCND2, MORF4L2, CDK2, MORF4L1, CHRNA7, IFNG, IGF1, LIF, LIFR, NDN, NEDD9, PDGFRB, ING3, PPP2CA
hsa-miR-30a-5p	TBRG1, CCNK, CCNA1, BCL10, LMLN, UBA3, CCNT2, CCNE2, SEP7, NEK4, TFDP1, TSC1, UBE2I, PTP4A1, EVI5, MIS12, SUV39H2, CHAF1B, CDC7, RECK, RGS2, RCBTB1, NEK1, NF1, PNN, RBM5, STAG2, DBF4, ESCO1, ESCO2, DDX11, E2F3, EPHB2, CLASP2, SEP6, CD2AP, NSL1, MTBP, INCENP, JAG2, KIF11, RHOB	BCL2L11, EDAR, NLRP3, RFFL, TICAM1, PRUNE2, TRIM35, SIRT1, RNF144B, TCTN3, GJA1, CECR2, SH3KBP1, HTT, HIP1, IL1A, IL2RA, IL17A, MAP3K5, MLL, MNT, BCL2L15, NAIP, DDIT4,. C8orf4, AVEN, PTGER3, BIRC6, TRIB3, ATXN1, SOX9, ACTC1, TFDP1, TIA1, CASP3, UNC5C, TNFSF9, TNFRSF10D, TNFRSF10B, BCL10, ATG12, EBAG9, DLG5, ARHGEF6, ATG5	C19orf10, BIRC6, BNC1, SOX9, VIPR1, CAMK2D, CDC7, CDCA7, CUL3, SOCS1, CREG1, OCS3, EBAG9, CFDP1, ENOX2, ADRA1D, ADRA2A, ADRB2, DDX11, TLX1, IL7, JAG2, KRAS, LIFR, LYN, MAFG, NOV, PDGFRB, PPP2CA
hsa-miR-758-3p	PARD6G, CCNG1, AURKB, MPHOSPH1, DLGAP5, DMTF1, BMP7, NEK4, TACC1, TP53BP2, VHL, WEE1, PPAPDC1B, RBL1, PPP1CB, NUSAP1, WWOX, ZAK, CDKN1B, KHDRBS1, STAG2, FOXN3, SGOL1, E2F6, CLASP2, PDS5A, CD2AP, CADM1, EID1, APPL1, LIN9, STRN3, ING1, LIG4, MCM3, MCM6	MALT1, UNC5D, CTSB, RYBP, CADM1, APAF1, PAK1, ZAK, KRT20, APTX, IFT57, TNFRSF19, BCL2L1, ATXN1, PERP, UBE2Z, BNIP2, TIA1, TIAL1, TP53BP2, C10orf97, ACTN4, RABEP1, EBAG9, BAG2, BRE	SLAMF1, STAT5B, KLF5, BLZF1, CDC2L5, EBAG9, SOCS6, NAMPT, CDKN1B, CFDP1, ENOX2, TCFL5, FGF7, ID4, IGF1, ILK, LIFR, NDN, CRIM1, POU3F2

[i] miRNAs, microRNAs; H₂O₂, hydrogen peroxide; HDPs, human dermal papilla cells.

Table III Predicted target genes of the top five most downregulated miRNAs in H2O2-treated HDPs.

Table III

Predicted target genes of the top five most downregulated miRNAs in H2O2-treated HDPs.

	Target genes and functions
miRNA	Cell cycle	Apoptosis	Cell growth and proliferation
hsa-miR-376c-3p	JUB, CCNB3, CCNT2, NOLC1, MTSS1, SIAH2, AURKA, VHL, HMGA2, CHAF1B, ERBB2IP, PTEN, NPAT, PAFAH1B1, GMNN, XRN1, MPHOSPH8, SPIN1, POLS, LZTS1, CHEK2, SASS6, GADD45A, MAPRE1, KANK1, SASH1, APPL1, ANXA1, HPGD, SEP14, MN1	EGLN3, RFFL, ACVR1C, PARP4, CNTN4, GADD45A, DAPK2, TNFAIP8, SGMS1, ARF6, MCL1, APIP, GULP1, RHOT1, DUSP22, ROCK1, ATXN1, SHB, SIAH2, ACTC1, PHLDA2, FASTKD3, CASP3, CASP7, AIFM1, ATG12, CD5L	MARK4, CXCL5, BMPR2, SSR1, STAT5A, TGFBR1, TNFSF4, HMGA2, SOCS2, SOCS6, CD86, CD47, MORF4L2, DNAJA2, CFDP1, TNFSF13B, TBC1D8, SOCS4, FGF7, FLT1, FLT3, RBM9, ANXA1, IL2, CXCL10, KRAS, NOV, IL17RB
hsa-miR-423-5p	JUB, CCNF, PKMYT1, MPHOSPH1, DMTF1, SUV39H1, TACC1, DBF4B, BIN3, MAPK3, MAPK4, PTCH1, RAP1A, UPF1, TSPYL2, NF2, PAFAH1B1, CINP, SUFU, PFDN1, RBM5, CDKN1A, HMG20B, TXNIP, SPIN1, RCC1, FOXN3, LZTS1, PMF1, E2F7, KCTD11, SNF1LK, DAB2IP, CYLD, DDX11, E2F2, EP300, PDS5A, NSL1, SGSM3, BIRC5, INCENP, IRF1, MCM7, MN1, NBL1, NBN, SEP3, SEP8	BCL2L11, IL24, DIDO1, RFFL, CLU, ACVR1C, DAP, DEDD2, DOCK1, E2F2, EIF5A, EP300, ERN1, PTK2B, FOSL2, CYFIP2, CIDEB, HIPK2, SAP30BP, CARD10, HTT, KCNIP3, HIP1, BIRC5, IAPP, IL17A, LTBR, MDM4, MLL, NGFR, NDUFA13, PDCD1, RNF216, PPARD, PRKCG, BAG1, BCL2L1, RTKN, NOD2, BNIP1, BOK, PRDX2, CACNA1A, BCL2L14, BIRC7, CASP2, PLA2G6, TRAF7, AIFM2, LGALS12, RIPK1, TNFSF12, TNFRSF14, TNFRSF10D, SQSTM1, NOL3, BMF, DNAJA3, TAOK2, NTN1, BRE	C19orf10, RAC2, SHC1, STAT5A, STAT5B, TGFBR2, VEGFB, PROK1, CUL3, CDC2L5, FGF18, NRP1, SOCS3, SCGB3A1, S1PR2, TAOK2, CIAO1, HTRA3, NTN1, CLU, ADM, CSF1, ADRA1D, KCTD11, CTF1, ADRA2A, DDX11, DHPS, S1PR3, PTK2B, FOXO1, FLT3LG, RBM9, HOXC10, VSX2, IGF1, IGFBP6, KRT6A, MAFG, ODC1, PDGFRB, C9orf127, IL17RB, CCDC88A, PRKCQ
hsa-miR-3665	-	-	-
hsa-miR-20a-5p	PARD6B, JUB, RUNX3, CDC23, PCAF, BCL10, CCND2, CCNG2, RNF8, CCNT2, PKMYT1, ZNF830, NEK9, KIF23, RASSF2, RB1CC1, SEP7, CDC25A, DLEC1, DMTF1, PAPD5, BMP2, STK11, TACC1, BUB1, VHL, WEE1, PTP4A1, SUV39H2, MAP9, PPP6C, ERBB2IP, MAPK1, MAPK4, PCNP, PTEN, RAD17, RAP1A, RB1, RBL1, RBL2, CCND1, TIPIN, SEP6, NPAT, PAFAH1B1, FZR1, ZAK, XRN1, CHAF1A, CDKN1A, TXNIP, CETN2, SEP2, CIT, TUSC2, SNF1LK, CYLD, E2F1, E2F3, FANCD2, C11orf82, HEPACAM, MAPRE1, MAPRE3, SEP9, PDS5A, SASH1, CLASP1, RABGAP1, GAK, NSL1, APPL1, FBXO5, CKAP2, PDCD4, C13orf15, GRLF1, RACGAP1, BIRC5, TMPRSS11A, IRF1, MCC, MCM3, NBL1	NOD1, FASTK, EGLN3, NLRP3, E2F1, EIF4G2, C11orf82, CARD8, ACIN1, FAIM2, SULF1, RYBP, SGMS1, MAGI3, CKAP2, GJA1, TNFRSF21, PDCD4, BIRC5, APP, MCL1, MDM4, MAP3K5, BCL2L15, OSM, SH3GLB1, PDE1B, ZAK, RHOT1, BCAP29, PTGER3, BCL2, PERP, ISG20L1, BNIP2, TNFAIP3, SLTM, C10orf97, RNF34, FXR1, CASP2, PLA2G6, CASP7, CASP8, CUL1, AIFM2, TNFRSF10D, TNFRSF10B, IER3, SQSTM1, SGPL1, TAX1BP1, BCL10, RABEP1, ATG12, DEDD, CD5L, STK17B, TP53INP1, ATG5, LITAF, TRAF4, SLK	PURA, BCL2, CXCL5, BMPR2, SSR1, TBX3, TGFBR2, TSG101, CAMK2D, ITCH, CDCA7, BRMS1L, CUL3, CCND2, S1PR2, SOCS6, ARHGEF11, DNAJA2, CFDP1, HOXB13, MORF4L1, TBC1D8, CHRNA7, S1PR3, ERBB3, HEPACAM, FGF4, FGF7, LRP12, IGFBP7, LIF, LIFR, OSM, DERL2, CRIM1, ING3, TIPIN, PPP2CA
hsa-miR-29c-3p	PPM1D, CDC23, CCND2, CCNF, UBA3, CCNT2, AURKB, HDAC4, DCLRE1A, SUPT5H, TACC1, PTP4A1, CDC7, SMPD3, RCC2, PTEN, NEDD9, ZAK, XRN1, CDK2, CDK6, HTATIP2, GREF1, STAG2, SEP9, DBF4, MAPRE2, FOXN3, E2F7, SLC5A8, AHR, VASH1, MAPRE1, CLASP2, SEP6, PPP1R13B, MCC, MLF1, NASP	FEM1B, HTATIP2, IL24, SLC5A8, AHR, FOXO3, PPP1R13B, SIRT1, ZNF346, CECR2, BIRC2, FAS, IL2RA, MCL1, SH3GLB1, ZAK, CYCS, DIABLO, DUSP22, BIRC6, BAK1, ATXN1, CIDEC, SGK1, ISG20L1, TNFAIP3, TNFRSF1A, TRAF5, FAM130A1, CASP7, AIFM2, BMF, RABEP1, TP53INP1, TRAF4, SLK	BIRC6, PURA, TRAF5, VEGFA, EPC1, CDC7, CREG1, CCND2, MORF4L2, CDK2, MORF4L1, CHRNA7, RBM9, IFNG, IGF1, LIF, LIFR, NDN, NEDD9, PDGFRB, ING3, PPP2CA

[i] miRNAs, microRNAs; H₂O₂, hydrogen peroxide; HDPs, human dermal papilla cells.

Discussion

ROS, such as H2O2, are generated as reactive byproducts of cellular metabolism in the mitochondria. The intracellular level of ROS is distinctly regulated by the cellular antioxidant system, including non-enzymatic and enzymatic antioxidants (27). The ROS level is increased in response to environmental stresses, including UV irradiation, toxic chemicals, heat and even high glucose concentrations (27,28). High levels of ROS induce cell cycle arrest, senescence and apoptosis due to ROS damage of cellular membranes, lipids, proteins and DNA (7–10). In the present study, and in others previously, it has been identified that H2O2 induces growth arrest in HDPs (Fig. 1) (29–31). The H2O2-induced growth arrest occurs later within the cascade of events activated in response to H2O2 treatment. Previous investigations identified H2O2-responsive miRNAs, including miR-34 and miR-145, which also are implicated in the ROS-responsive pathway (32,33). Therefore, miRNA appears to be required for H2O2-dependent growth arrest.

The present study identified 68 miRNAs that were regulated by H2O2 in HDPs (Fig. 2). miR-193-3p and miR-29b increased in H2O2-treated HDPs and induced apoptosis by targeting MCL-1 (34,35). MCL-1 is a BCL-2 family member that is involved in mitochondria-dependent intrinsic apoptosis (36). MCL-1 represses apoptosis by preventing the formation of mitochondrial membrane potential (36). Our data, together with those of Lin et al (15), demonstrated that miR-193a-3p increased in response to ROS and subsequently induced cell death. A previous study revealed that miR-30a-5p was upregulated by H2O2, it repressed autophagy by targeting beclin-1 and eventually induced apoptosis (37). In the present study, it was demonstrated that miR-20a-5p and miR-423-5p were repressed by H2O2 in HDPs (Table I). miR-20a-5p had multiple target genes, including BNIP2, APP, ASK1 and TNKS2. Therefore, miR-20a-5p regulated proliferation, migration, invasion and inflammation by the regulation of its target genes (38–41). miR-424-5p targeted p21Cip1/Waf1, which functions in proliferation and G1 phase transition (42).

Numerous studies have demonstrated that ROS, including H2O2, induces intrinsic apoptosis (43–45). H2O2-induced apoptosis is regulated by mitochondrial membrane permeability, which is regulated by the BCL-2 family (36). The anti-apoptotic BCL-2 family, including BCL2L10, BCL2L11, BCL2L2 and BCL10 were predicted as targets of miRNAs that were upregulated by H2O2 (Table II). The pre-apoptotic BCL-2 family members BOK and BAK1 were predicted as targets of miRNAs that were downregulated by H2O2 (Table III). Cell cycle regulating proteins, such as cyclins and CDKs, function during each phase of the cell cycle (G1, S, G2 and M) (46). CCNA1 and CDK2 are required for the regulation of the G2/M phase (47,48). The results of the present study predict that hsa-miR-30a-5p and hsa-miR-29b-3p target CCNA1 and CDK2. These data suggest that H2O2-mediated growth arrest and cell death in HDPs is associated with the changes in expression of specific miRNAs.

Bioinformatics analysis of miRNA expression profiles, miRNA target genes and the GO of target genes provided a more holistic view of the underlying cellular mechanisms that occur in response to H2O2-induced growth arrest and apoptosis. The identification of miRNAs and their putative targets may offer new therapeutic strategies for H2O2-induced hair follicle disorders, such as hair loss.

Acknowledgements

The authors are grateful to all the members of our research group for their support and advice regarding this study. This study was supported by the KU Research Professor Program of Konkuk University.

References