miR-1 Regulates Differentiation and Proliferation of Goat Hair Follicle Stem Cells by Targeting IGF1R and LEF1 Genes

Abstract

Hair follicle stem cells (HFSCs) play a significant role in hair development. miR-1 has been reported as an important regulatory factor that affects hair follicle growth and development, but its regulatory mechanism on HFSC development remains unknown. In this study, the molecular mechanism of miR-1 in regulating HFSC proliferation and differentiation was investigated. High-throughput RNA-seq and integrated analysis were performed to identify differentially transcribed mRNAs and microRNAs (miRNAs) in HFSCs co-cultured with dermal papilla cells (named dHFSCs) and control HFSCs. We then determined the molecular function of miR-1 in HFSCs. Compared with HFSCs, 13 differentially transcribed miRNAs were identified in dHFSCs. The in vitro results indicated that the overtranscription of miR-1 inhibited HFSC proliferation, but enhanced HFSC differentiation by targeting IGF1R and LEF1 genes. This study provides new insights into the molecular mechanisms of HFSC development. Approval ID (2014ZX08008-002).

Introduction

Hair follicle stem cells (HFSCs) (Yang and Peng, 2010; Ojeh et al., 2015; Du et al., 2018a) are a type of adult stem cells located in the bulge region of HFs, and they have ability to self-renew and differentiate multi-directionally (Hoffman, 2006; Elaine, 2007; Cédric and Elaine, 2014; Shen et al., 2017). Dermal papilla cells (DPCs) are located at the base of HFs, and they play a leading role in regulating HF development and cyclical growth (Driskell et al., 2011; Madaan et al., 2018). During the telogen (resting) phase, when the number of matrix cells decreases owing to apoptosis, DPCs release molecular signals to activate the differentiation of HFSCs and promote the regeneration of HFs (Reynolds et al., 1991; Zhang et al., 2016).

Previous studies have reported a co-culture system of HFSCs with DPCs (Zhang et al., 2013; Yan et al., 2019), but the molecular mechanisms of HFSC development and the related signaling pathways remain to be fully understood.

MicroRNAs (miRNAs) are small molecules that regulate cell organogenesis, apoptosis, proliferation, and differentiation (Mardaryev et al., 2010; Ahmed et al., 2014; Li et al., 2019). In previous studies, miRNAs have been reported in regulating the proliferation and differentiation of HFSCs (Xuejuan et al., 2013; Du et al., 2018a); for example, the differentiation of HFSCs was promoted by miR-21 (Cai et al., 2018) and miR-203a-3p (Luo et al., 2020). miR-1 has been considered as an important regulatory factor that affects the growth and development of HF (Yuan et al., 2013; Zhou et al., 2018; Yan et al., 2019). However, the mechanism of miR-1 in regulating the development of HFSCs requires further clarification.

Thus, we aimed to elucidate the molecular mechanism of miR-1 in regulating the proliferation and differentiation of HFSC. In this study, HFSCs co-cultured with DPCs (dHFSCs) and control HFSCs were used as experimental materials and subjected to high-throughput RNA-seq. Integrative analysis was then carried out to determine the differentially transcribed miRNAs and their target genes. We also identified the target genes of miR-1. Our findings will provide a deeper insight into the molecular mechanism underlying the development of HFSCs.

Materials and Methods

Animals and cell culture

All studies were performed in accordance with the guidelines of the Northwest A&F University, Yangling, China (approval ID: 2014ZX08008-002).

The HFSC line was established and cultured as described previously (Yan et al., 2019). HFSCs were cultured in Dulbecco's modified Eagle's medium (DMEM)/F12 medium supplemented with 10% fetal bovine serum, 1% penicillin–streptomycin, 10 ng/mL insulin, 10 ng/mL epidermal growth factor, and 0.4 μg/mL hydrocortisone in 5% CO₂ at 37°C. The medium was replaced every 2 days.

HFSCs and DPCs were co-cultured in DMEM/F12 medium in Transwell chambers (Yan et al., 2019). DPCs were isolated from Shaanbei White Cashmere goats in our laboratory (He et al., 2016; Zhou et al., 2017; Yan et al., 2019).

RNA-seq analysis

For mRNA and small RNA libraries construction, total RNA was extracted from dHFSCs and HFSCs (each group had three biological replicates) using an Eastep^® Super Total RNA Extraction Kit (Promega, Shanghai, China) following the manufacturer's instructions. RNA was analyzed using a 2100 Bioanalyzer (Agilent Technologies) and sequenced on a HiSeq 2500 platform (Illumina).

mRNA libraries were constructed according to previous studies (Gao et al., 2016). After sequencing, the raw data contained a number of connectors, low-quality reads, so Cutadapt (Martin, 2011) was used to filter the raw reads to get clean data for further analysis. Then the clean data were mapped to the latest goat reference genome (GCF_001704415.1_ARS1_genomic.fna) using TopHat2 (Kim et al., 2013). The mismatch was no more than 2.

Small RNA library were constructed as previous studies (Du et al., 2018b); after sequencing, the data containing poly-N, 5′ adapter contaminants, poly A, T, G or C, reads without 3′ adapter or insert tag, low-quality reads were cleaned using Cutadapt (Martin, 2011; Yan et al., 2019), then the sequence of 20–25 nt would be reserved, which covered the main length of mature miRNA. After this, the data were mapped to the latest goat reference genome (GCF_001704415.1_ARS1_genomic.fna) using Bowtie (Langmead et al., 2009). The mapped reads were obtained using miRBase (Griffiths-Jones et al., 2008) as a reference. miREvo (Wen et al., 2012) and mirdeep2 (Friedländer et al., 2012) were used to predict novel miRNAs based on their secondary structures and Dicer cleavage sites.

mRNA transcription was estimated using fragments per kilobase of exon model per million mapped reads (Marioni et al., 2008). For miRNAs, the data were normalized as transcripts per million (Zhou et al., 2010).

DESeq R package (Love et al., 2014) was used to analyze the differential transcriptions of mRNAs and miRNAs in dHFSCs and HFSCs. A |log2 (fold change)| >1 and adjusted p < 0.05 were used as the threshold cutoffs to screen significantly differentially transcribed genes and miRNAs.

We used GOseq (Young et al., 2010) to perform Gene Ontology (GO) analysis. KO-Based Annotation System (Mao et al., 2005) was used to assess the enrichment of DE genes or miRNA target genes in the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. A process or pathway with p-value <0.05 was considered to be significantly enriched.

Integrated analysis of differentially transcribed mRNA–miRNAs networking between HFSCs and dHFSCs

Integrated analysis of differentially transcribed mRNA (DE mRNA)–miRNA networking was performed as previously described (Zhou et al., 2018). Based on the mRNA and miRNA libraries analysis results, we screened the target genes of differentially transcribed miRNAs (DE miRNAs). The target genes that were differently transcribed in mRNA libraries were selected for further research.

Cell transfection

miR-1 mimics and negative control (NC) were synthesized by GenePharma (Shanghai, China). The sequence of miR-1 mimic was as follows: uggaauguaaagaaguauguau. miR-1 mimics or NC were transfected into HFSCs using Lipofectamine 3000 (Invitrogen, Carlsbad, CA), following the manufacturer's instructions. After 48 h, transfected HFSCs were collected for further analysis.

Cell proliferation assay

HFSCs transfected with miR-1 mimics or NC were collected for cell proliferation assays. Cell proliferation was examined by using a cell counting kit-8 (Dojindo Molecular Technologies, Inc., Kumamoto, Japan) and EdU cell proliferation kit (Thermo Fisher Scientific), following the manufacturers' protocols.

For cell cycle analysis, HFSCs transfected with miR-1 mimics or NC were fixed with ice-cold 70% ethanol and stained with propidium iodide and RNase solution (550825; BD Pharmingen). The cells were analyzed by using a FACScan flow cytometer (Becton Dickinson) equipped with a 488-nm excitation laser using the CellQuest Pro software (Becton Dickinson).

Quantitative real-time PCR

To validate the accuracy of RNA-seq analysis results, total RNA was extracted from dHFSCs and HFSCs as described previously, and cDNA was synthesized by using a cDNA Synthesis Kit (Thermo Fisher Scientific). Six DE genes, CREB5, KLF4, PTGS2, ID1, SPP1, and DLX5, were selected randomly, with the housekeeping gene β-actin used as an endogenous control. Six DE miRNAs (miR-1, miR-504, miR-455-5p, miR-660, miR-100-5p, and miR-182) were also selected randomly for quantitative real-time (RT) PCR. U6 was used as an endogenous control.

To confirm the effect of miR-1 on HFSCs, total RNA was isolated from HFSCs transfected with miR-1 mimics and NC, and cDNA was synthesized. The relative transcriptions of the cell proliferation marker genes (KI67, PCNA, AKT, and mTOR), the differentiation marker genes (K6, S100A3, Notch1, and β-catenin) were detected.

The qRT-PCR was performed on a CEX96 Touch Real-time PCR Detection System (Bio-Rad) in a 20-μL reaction volume containing 10 μL of SYBR Premix Ex Taq II (Takara, China), 7 μL of water, 2 μL of cDNA samples, 0.5 μL of forward primer, and 0.5 μL of reverse primer. All primers used in this assay are listed in Supplementary Table S7.

Western blotting

HFSCs transfected with miR-1 mimics or NC were collected and lysed in 200 μL radio immunoprecipitation assay buffer containing 1 mM phenylmethylsulfonyl fluoride (Solarbio, China). The lysed cells were subjected to 12% sodium dodecyl sulfate–polyacrylamide gel electrophoresis gel in a running buffer. Next, the proteins were electrotransferred to polyvinylidene fluoride membranes (Roche, Switzerland). The membranes were blocked for 1 h at room temperature with 5% skim milk and incubated at 4°C overnight with the following primary antibodies: anti-K6, anti-S100A3, anti-Notch1, anti-β-catenin, and anti-PCNA rabbit polyclonal antibodies (Proteintech, China); anti-AKT rabbit polyclonal antibody (CST); anti-mTOR rabbit monoclonal antibody (BOSTER, China); and anti-β-actin (BOSTER).

The membranes were then incubated at room temperature for 1 h with secondary horseradish peroxidase-conjugated antibodies. After that, the membranes were detected using a chemiluminescence reagent (ECL, Millipore) and visualized with a fluorescence imaging system (Sagecreation). Subsequently, the images were analyzed by using the ImageJ software (Gallo-Oller et al., 2018).

Target gene prediction

RNAhybrid (Krüger and Rehmsmeier, 2006) and Pita prediction tools (version 7.0) (Kertesz et al., 2007) were used to predict the target genes of miR-1. FOXP1, IGF1R, SOX9, MSI2, LEF1, and FOXC1 were selected as miR-1 target genes.

Dual luciferase vector construction

The 3′UTR of FOXP1, IGF1R, SOX9, MSI2, LEF1, and FOXC1 genes were amplified from Shaanbei White Cashmere goat cDNA (cDNA synthesis method was described previously) and verified by Sanger sequencing. Fragments were cloned into the psiCheck2 vector, which was digested with XhoI and NotI. A site-directed mutagenesis kit (TaKaRa, Dalian, China) was used to generate the mutant vectors, following the manufacturer's instructions. All mutants were verified by sequencing. All primers used in this assay are given in Supplementary Tables S8 and S9.

Dual luciferase reporter assays

HFSCs transfected with FOXP1, IGF1R, SOX9, MSI2, LEF1, and FOXC1 dual luciferase reporter vectors and mutant vectors were lysed using cell lysis buffer. After 48 h, firefly and Renilla luciferase activities were detected by using a Dual Luciferase Reporter Gene Assay Kit (Promega, Madison) according to the manufacturer's protocol. Relative luciferase activity indicated the ratio of firefly luciferase to Renilla luciferase activities.

Statistical analysis

Data were presented as the mean ± standard error of the mean. All results were analyzed using Microsoft Excel. The means were compared using t-tests, and statistical significance was set at p < 0.05.

Results

Identification of mRNAs and miRNAs in HFSCs and dHFSCs

For mRNA libraries, an average of 25,245,466.83 raw reads were produced per sample. After filtration, 24,123,618 clean reads were obtained, which formed 3.62 G clean bases per sample. For miRNA libraries, an average of 18,427,949.5 raw reads were produced per sample, and 17,654,735.7 clean reads were obtained after filtration, which formed 0.921 G clean bases for each sample. The clean reads were classified and mapped to the latest goat reference genome (GCF_001704415.1_ARS1_genomic.fna). The sequencing data for each sample are summarized in Supplementary Tables S1 and S2.

More than 94% of the clean reads were mapped to the reference genome in both the mRNA and miRNA datasets (Supplementary Tables S3 and S4). The correlation of triple biological replicates was up to 0.98 (r ² ≥ 0.982) (Supplementary Fig. S1). All these data showed that the samples had sufficient reproducibility and rationality; therefore, all of them were used for further analysis of mRNA and miRNA transcription profiles.

Transcriptional profiles of mRNAs and miRNAs HFSCs and dHFSCs

We identified 357 known miRNAs and 203 novel miRNAs (Supplementary Table S3). There were 12,802 co-transcribed genes and 656 DE genes (342 upregulated genes and 314 downregulated genes in dHFSCs compared with HFSCs) (Supplementary Table S5). The top 20 most significantly DE genes are given in Table 1 and Figure 1. A total of 403 co-transcribed miRNAs (Supplementary Table S6) and 13 DE miRNAs (6 upregulated and 7 downregulated in dHFSCs) (Table 2, Fig. 2) were detected.

FIG. 1.

Transcriptional profiles in HFSCs. (A) Volcano plot indicates up- and downregulated genes in HFSCs compared with dHFSCs. (B) Heat maps and hierarchical clustering of differentially transcribed genes. (Red indicates higher transcription in treatment samples, whereas blue indicates higher transcription in control samples). dHFSCs, HFSCs co-cultured with DPCs; HFSCs, hair follicle stem cells. Color images are available online.

FIG. 2.

miRNA transcriptional profiles in HFSCs. (A) Volcano plot indicates up- and downregulated miRNA in HFSCs compared with dHFSCs. (B) Heat maps of miRNAs show hierarchical clustering of altered miRNAs (Red indicates higher transcription in treatment samples, whereas blue indicates higher transcription in control samples). miRNAs, microRNAs. Color images are available online.

Table 1.

Top 20 Most Significantly Differentially Transcribed Genes

Gene ID	Read counts of dHFSCs	Read counts of HFSCs	Log2 (fold change)	p	Adjusted p
ENSCHIG00000000023	1685.718975	1395.279887	0.27281	4.90E-05	0.0030653
ENSCHIG00000000043	135.0466079	196.4365325	−0.54061	0.00010968	0.0059051
ENSCHIG00000000101	7955.535454	9265.906457	−0.21997	0.00025681	0.011977
ENSCHIG00000000147	181.0765982	83.65430756	1.1141	2.53E-12	6.33E-10
ENSCHIG00000000282	1123.014524	1381.059095	−0.2984	1.36E-05	0.0010083
ENSCHIG00000000350	880.4602365	635.4229017	0.47054	1.20E-08	1.71E-06
ENSCHIG00000000386	667.2215085	812.9420501	−0.28499	0.00035758	0.015473
ENSCHIG00000000407	4982.798383	5717.088706	−0.19832	0.00082221	0.030174
ENSCHIG00000000416	33132.6369	29624.46723	0.16146	0.0013902	0.045213
ENSCHIG00000000480	9308.907836	7680.822116	0.27735	6.94E-07	6.97E-05
ENSCHIG00000000495	28044.08771	21185.33337	0.40463	9.22E-09	1.34E-06
ENSCHIG00000000703	742.6756116	602.4234481	0.30195	0.00031588	0.013966
ENSCHIG00000000812	328.5203675	445.9834532	−0.44101	7.05E-06	0.00056985
ENSCHIG00000000920	2938.264539	2237.739909	0.39292	5.08E-10	8.81E-08
ENSCHIG00000001086	15.45046959	37.49576112	−1.2791	0.00030857	0.013793
ENSCHIG00000001148	38748.22166	33660.92666	0.20306	8.06E-05	0.0046082
ENSCHIG00000001297	113.9035833	62.46708286	0.86665	2.82E-05	0.0019119
ENSCHIG00000001308	1354.953235	1150.846066	0.23555	0.00094056	0.033605
ENSCHIG00000001374	1502.903571	1241.485311	0.27569	0.00059183	0.023322
ENSCHIG00000001557	1199.728548	942.4664346	0.34819	2.10E-06	0.0001908

dHFSCs, co-cultured HFSCs with DPCs; HFSCs, hair follicle stem cells.

Table 2.

Top 13 Most Significantly Differentially Transcribed MicroRNAs

miRNAs	Read counts of dHFSCs	Read counts of HFSCs	Log2 (fold change)	p	Adjusted p
novel_2	525700.6188	1297205.338	−1.2278	1.90E-09	3.67E-07
chi-miR-455-5p	23770.16142	14008.44684	0.73308	1.55E-05	0.0014943
chi-miR-151-3p	275929.5315	137225.2251	0.93664	2.98E-05	0.0017152
chi-miR-182	8845.368034	16479.42136	−0.84573	3.55E-05	0.0017152
novel_686	1301.483931	2091.655372	−0.65907	6.12E-05	0.0023622
chi-miR-143-3p	341.1949085	187.7248791	0.8094	8.28E-05	0.0026644
chi-miR-1	145.4998904	66.49171855	1.015	0.00012721	0.0035072
chi-miR-99b-5p	12484.49993	21238.9037	−0.7218	0.00043967	0.010607
novel_89	393.6877592	594.4414921	−0.57272	0.00058127	0.012465
chi-miR-148a-5p	771.725237	505.8720508	0.58458	0.00066435	0.012822
chi-miR-660	2813.104779	3922.873634	−0.4666	0.0010179	0.01786
chi-miR-504	938.1001945	666.0179532	0.47775	0.002069	0.033277
chi-miR-100-5p	25689.17058	42323.44467	−0.67049	0.0025543	0.037922

To validate the accuracy of the RNA-seq results, six DE genes (CREB5, KLF4, PTGS2, ID1, SPP1, and DLX5) and six DE miRNAs (miR-1, miR-504, miR-455-5p, miR-660, miR-100-5p, and miR-182) were subjected to qRT-PCR. The results were consistent with the RNA-seq data (Fig. 3).

FIG. 3.

qPCR validation of RNA-seq. qPCR results of six differently transcribed genes and six differently transcribed miRNAs. *p < 0.05, **p < 0.01. qPCR, quantitative PCR. Color images are available online.

GO and KEGG pathway enrichment analyses of DE mRNAs and miRNAs

GO and KEGG enrichment analyses were performed to further validate the functions of the DE mRNAs and miRNAs. GO enrichment was divided into three categories: biological process (BP), cellular component (CC), and molecular function (MF). The most significantly enriched BP term was proteolysis. Among CC terms, the cation transport extracellular region was significantly enriched. Finally, among MF terms, insulin-like growth factor binding and growth factor binding were enriched significantly (Fig. 4A).

FIG. 4.

Integrated co-transcription and co-localization network analysis of DE mRNAs and miRNAs. (A) GO analysis of the conjoint analysis for DE mRNAs and miRNAs and KEGG enrichment of conjoint analysis. (B) The network regulation between candidate miRNAs and their target genes (Red represents upregulated, whereas green represents downregulated). (C) The network regulation of gene involved in four significant pathways (Yellow circles represents pathways, whereas green circles represents genes participated in the pathways). DE mRNA, differentially transcribed mRNA; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes. Color images are available online.

KEGG enrichment analysis was based on the richness factor of filtration, Q-values, and the number of genes. The most enriched pathways were the TNF signaling pathway, FoxO signaling pathway, amphetamine addiction, and apoptosis (Fig. 4A, Table 3). A total of 138 DE genes involved in the top 20 signaling pathways were identified. Among these, 30 DE genes (22 upregulated and 8 downregulated genes in dHFSCs), which were the most enriched genes in the top three signaling pathways, were selected to determine the most important pathways that affect the proliferation and differentiation of HFSCs.

Table 3.

Top 20 Most Enriched Pathways of Conjoint Analysis

Pathway term	Rich factor	q	Gene no.
TNF signaling pathway	0.1	0.058213664	11
FoxO signaling pathway	0.091603053	0.058213664	12
Amphetamine addiction	0.123076923	0.058213664	8
Apoptosis	0.074324324	0.227230915	11
Mineral absorption	0.108695652	0.534620085	5
Rheumatoid arthritis	0.073684211	0.675955451	7
p53 signaling pathway	0.081081081	0.675955451	6
Longevity regulating pathway—multiple species	0.083333333	0.733249516	5
Cell cycle	0.0625	0.733249516	8
Bladder cancer	0.097560976	0.733249516	4
Small cell lung cancer	0.068181818	0.854208573	6
Salmonella infection	0.064516129	0.904633194	6
B cell receptor signaling pathway	0.069444444	0.904633194	5
Circadian rhythm	0.096774194	0.904633194	3
Oxytocin signaling pathway	0.050955414	0.904633194	8
Pertussis	0.0625	0.904633194	5
TGF-beta signaling pathway	0.06097561	0.904633194	5
Insulin secretion	0.060240964	0.904633194	5
Glycosphingolipid biosynthesis—globo series	0.117647059	0.904633194	2
Pathways in cancer	0.04040404	0.904633194	16

TGF, transforming growth factor; TNF, tumor necrosis factor.

The 22 upregulated genes were NOD2, CREB5, TNFAIP3, CSF1, FAS, JUN, CXCL1 (LOC102182115), FOS, ICAM1, PTGS2, COX2, PLK2, SIRT1, GADD45B, GADD45G, IRS1, CDKN1A, GADD45A, ARC, EIF2S1, LOC108638455, and ACTG1. The eight downregulated genes were CEBPA, GABARAPL2, CDKN2B, FBX032, GABARAPL1, LOC102175311, CTSD, and CTSH.

In the TNF signaling pathway, the transcriptions of NOD2, TNFAIP3, and CXCL1 was significantly increased. It has been reported (Williams et al., 2017) that skin healing rate was affected in NOD2 knockout rats. Moreover, a significant increase in the transcription of CXCL1 was observed in injured rat and human skin tissue (He et al., 2018), whereas knockout of TNFAIP3 exacerbated inflammatory skin in mice (Devos et al., 2018). These previous studies indicate that the CXCL1 and TNFAIP3 genes are closely related to skin growth and development.

The FoxO signaling pathway participates in cell apoptosis, cell cycle regulation, intracellular glucose metabolism, oxidative stress, and various cellular physiological processes. Integrated analysis revealed that the transcription levels of SIRT1 and PLK2 were significantly increased. The network regulation of genes involved in these pathways is given in Figure 4C.

Networking between DE mRNAs and miRNAs in HFSCs and dHFSCs

To explore which DE miRNAs were involved in the proliferation and differentiation of HFSCs, we performed an analysis of DE mRNA–miRNA networking between HFSCs and dHFSCs. We found that in the presence of DPCs, miR-1, miR-151-3p, and miR-143-3p were upregulated, whereas miR-182 was downregulated. Therefore, we constructed miRNA–mRNA networks with these four miRNAs as the centers and corresponding mRNAs as the targets (Fig. 4B). We divided the mRNA–miRNA networks based on the expression patterns of upregulated and downregulated genes (Fig. 4B). The results showed that miR-1, miR-151-3p, and miR-143-3p could downregulate the target genes, whereas miR-182 could upregulate its target genes to regulate HFSC development.

miR-1 inhibited the proliferation but enhanced the differentiation of HFSCs by targeting IGF1R and LEF1 genes

It was clear that miR-1 transcription was significantly increased in dHFSCs compared with that in HFSCs. This is consistent with a previous report that miR-1 participates in HF development (Zhou et al., 2018). Cell proliferation assay results showed that after transfection with miR-1 mimics, the proliferation of HFSCs was significantly reduced (p < 0.01) compared with that of the cells transfected with NC (Fig. 5A, C). Cell cycle assay results showed that the overtranscription of miR-1 induced cell cycle arrest at the G0–G1 phase in HFSCs (Fig. 5B), suggesting the inhibitory effect of miR-1 on the proliferation of HFSCs.

FIG. 5.

miR-1 inhibited the proliferation but enhanced the differentiation of HFSCs. (A) The statistical results and representative figures of miR-1 on the proliferation of HFSCs as detected by EDU, **p < 0.01. (B) The effect of overtranscription of miR-1 on the HFSCs cycle, *p < 0.05, **p < 0.01. (C) CCK-8 was used to detected the effect of miR-1 on HFSCs proliferation, **p < 0.01. (D) mRNAs and protein expressions of proliferation-related genes, *p < 0.05. (E) mRNAs and protein expressions of differentiation-related genes, *p < 0.05, **p < 0.01. CCK-8, cell counting kit-8; EDU, 5-ethynyl-2′-deoxyuridine. Color images are available online.

After being transfected by miR-1 mimics, the mRNA transcriptional and protein expressional levels of KI67, PCNA, and AKT were significantly lower in HFSCs (p < 0.01) than in NC (Fig. 5D). These results further showed that the overtranscription of miR-1 inhibited the expression of cell proliferation marker genes (KI67 and PCNA) by regulating the AKT signaling pathway. Moreover, after the overtranscription of miR-1 mimics, the mRNA transcriptional and protein expressional levels of K6, S100A3, and Notch1 were significantly higher in HFSCs than in NC (Fig. 5E). These results indicated that the overtranscription of miR-1 enhanced the expression of differentiation marker genes (K6 and S100A3) by activating the Notch1 signaling pathway to enhance the differentiation of HFSCs.

Based on the prediction of miR-1 target genes, we screened FOXP1, FOXC1, IGF1R, SOX9, MSI2, and LEF1 and further explored their molecular mechanisms of miR-1. A dual luciferase reporter assay was conducted to analyze their regulatory relationships. The luciferase activities of IGF1R and LEF1 were inhibited after transfection with miR-1 mimics. After the target gene binding sites were mutated (Supplementary Fig. S2), the luciferase activities of IGF1R and LEF1 were recovered. However, there was no interaction between miR-1 and FOXP1, FOXC1, SOX9, and MSI2 (Fig. 6). These results indicated that miR-1 regulated IGF1R and LEF1. Taken together, the results showed that miR-1 inhibited the proliferation but enhanced the differentiation of HFSCs by targeting IGF1R and LEF1 genes.

FIG. 6.

Dual fluorescein reporter assay. **p < 0.01.

Discussion

As the basis for the growth and development of animal coats, HFSCs serve as the best cell model for studying the mechanism and promotion of cashmere fiber growth (Tudorita et al., 2004; Paus and Foitzik, 2010; Kandyba and Kobielak, 2014). In this study, we investigated the regulatory mechanism of HFSC proliferation and differentiation in dHFSCs and control HFSCs. Based on the bioinformatics analysis, 13 DE miRNAs were identified. These 13 DE miRNAs showed differential transcriptions in the skin tissue samples of Shaanbei White Cashmere goats at different phases (anagen, catagen, and telogen) of follicle growth (Yuan et al., 2013; Zhou et al., 2018).

A previous study found that miRNA-143 was abundantly transcribed in ear tissues, whereas miR-1 and miR-455-3p were only transcribed in the skin tissues of Inner Mongolia Cashmere goats (Williams et al., 2017). In another study, mesenchymal stem cells were transplanted to tight-skin mice, and miR-151-5p was identified as a therapeutic target for systemic sclerosis (Chen et al., 2017). Moreover, miR-182-5p was highly transcribed in goat skin tissue, whereas miR-143-3p showed differential transcription in goats with different coat colors (Wu et al., 2014).

We found that miR-1 inhibited the proliferation but enhanced the differentiation of HFSCs by targeting LEF1 and IGF1R genes.

LEF1 is an important transcriptional factor that regulates HF patterning and stem cell fate through the Wnt signaling pathway (Genderen et al., 1994; Zhou et al., 1995). LEF1 knockout led to a loss of vibrissal HFs and an arrest of dorsal HFs, whereas LEF1 activation induced de novo HF formation (Merrill et al., 2001; Zhang et al., 2013). The IGF signaling pathway is involved in cell proliferation, apoptosis, differentiation, survival, metabolism, and migration (Stewart and Rotwein, 1996). It also plays a critical role in HF development (Little et al., 2010). IGF1 and IGF1R are important transcriptional factors in the IGF signaling pathway, and IGF1R knockout mice show growth retardation (Liu et al., 1993), tissue and organ dysplasia, infertility, and even immediate death after birth (Pandini et al., 2007).

Therefore, according to the findings of this study, LEF1 and IGF1R genes determine HFSC-directed hair fate by regulating the Wnt and IGF signaling pathways, respectively. Our study revealed that the miR-1-LEF1 and miR-1-IGF1R axes may constitute new signaling pathways for regulating HFSC proliferation and differentiation.

Conclusions

This study provided new insights into the molecular mechanisms driving the development of HFSCs. We performed integrated analysis of mRNAs and miRNAs and investigated the potential regulatory mechanisms underlying HFSC proliferation and differentiation. Thirteen miRNAs showed differential transcriptions in dHFSCs compared with HFSCs. We further showed that miR-1 inhibited the proliferation but enhanced the differentiation of HFSCs by targeting IGF1R and LEF1 genes. However, the relationship between miR-1 and the candidate target genes (IGF1R and/or LEF1) in regulating HFSC proliferation and differentiation requires further investigation.

Footnotes

Authors' Contributions

H.Y., M.J., W.Z., and Y.C. designed the research. H.Y., M.J., Y.L., and Q.D. performed the experiments. H.Y., M.J., Y.L., Y.G., and X.W. analyzed the data and wrote the article.

Acknowledgment

All authors acknowledge and thank the respective institutes and universities.

Disclosure Statement

No competing financial interests exist.

Funding Information

This work was supported by the Natural Science Foundation of China (31872332); Datong Municipal Science and Technology Bureau (2020061 to H.Y.); PhD Initiation Grant of Datong University (2019-B-07 to H.Y.); Science and Technology Innovation Project of Universities, Shanxi Province of China (2020L0493 to H.Y.)

Supplementary Material

Supplementary Figure S1

Supplementary Figure S2

Supplementary Table S1

Supplementary Table S2

Supplementary Table S3

Supplementary Table S4

Supplementary Table S5

Supplementary Table S6

Supplementary Table S7

Supplementary Table S8

Supplementary Table S9

References

Ahmed

M.I.

, Majid

, Emelianov

V.U.

, Krzysztof

, Ankit

, Sharov

A.A.

, et al. (2014). MicroRNA-214 controls skin and hair follicle development by modulating the activity of the Wnt pathway. J Cell Biol, 207, 549–567.

Cai

, Zheng

, Ma

, Xing

, Wang

, Yang

, et al. (2018). Long non-coding RNA regulates hair follicle stem cell proliferation and differentiation through PI3K/AKT signal pathway. Mol Med Rep, 17, 5477–5483.

Cédric

, and Elaine

(2014). Stem cell plasticity. Plasticity of epithelial stem cells in tissue regeneration. Science, 344, 1242281.

Chen

, Wang

, Moshaverinia

, Liu

, Kou

, Yu

, et al. (2017). Mesenchymal stem cell transplantation in tight-skin mice identifies miR-151-5p as a therapeutic target for systemic sclerosis. Cell Res, 27, 559–577.

Devos

, Mogilenko

D.A.

, Fleury

, Gilbert

, Becquart

, Quemener

, et al. (2019). Keratinocyte expression of A20/TNFAIP3 controls skin inflammation associated with atopic dermatitis and psoriasis. J Invest Dermatol, 139, 135–145.

Driskell

R.R.

, Carlos

, Michael

, and Watt

F.M.

(2011). Hair follicle dermal papilla cells at a glance. J Cell Sci, 124, 1179–1182.

K.T.

, Deng

J.Q.

, He

X.G.

, Liu

Z.P.

, Peng

, and Zhang

M.S.

(2018a). MiR-214 regulates the human hair follicle stem cell proliferation and differentiation by targeting EZH2 and Wnt/β-catenin signaling way in vitro . Tissue Eng Regen Med, 15, 341–350.

, Pirouz

, Choi

, Huebner

A.J.

, Clement

, Meissner

, et al. (2018b). An intermediate pluripotent state controlled by microRNAs is required for the naive-to-primed stem cell transition. Cell Stem Cell, 22, 851.e855–864.e855.

Elaine

(2007). Scratching the surface of skin development. Nature, 445, 834.

10.

Friedländer

M.R.

, Mackowiak

S.D.

, Li

, Chen

, and Rajewsky

(2012). miRDeep2 accurately identifies known and hundreds of novel microRNA genes in seven animal clades. Nucleic Acids Res, 40, 37–52.

11.

Gallo-Oller

, Ordoñez

, and Dotor

(2018). A new background subtraction method for Western blot densitometry band quantification through image analysis software. J Immunol Methods, 457, 1–5.

12.

Gao

, Wang

, Yan

, Zeng

, Ma

, Niu

, et al. (2016). Comparative transcriptome analysis of fetal skin reveals key genes related to hair follicle morphogenesis in cashmere goats. PLoS One, 11, e0151118.

13.

Griffiths-Jones

, Saini

H.K.

, van Dongen

, and Enright

A.J.

(2008). miRBase: tools for microRNA genomics. Nucleic Acids Res, 36, D154–D158.

14.

J.T.

, Huang

H.Y.

, Qu

, Xue

, Zhang

K.K.

, Xie

X.L.

, et al. (2018). CXCL1 and CXCR2 as potential markers for vital reactions in skin contusions. Forensic Sci Med Pathol, 14, 1–6.

15.

, Chao

, Zhou

, and Chen

(2016). Fibroblast growth factor 5-short (FGF5s) inhibits the activity of FGF5 in primary and secondary hair follicle dermal papilla cells of cashmere goats. Gene, 575, 393–398.

16.

Hoffman

R.M.

(2006). The pluripotency of hair follicle stem cells. Cell Cycle, 5, 232–233.

17.

Kandyba

, and Kobielak

(2014). Wnt7b is an important intrinsic regulator of hair follicle stem cell homeostasis and hair follicle cycling. Stem Cells, 32, 886–901.

18.

Kertesz

, Iovino

, Unnerstall

, Gaul

, and Segal

(2007). The role of site accessibility in microRNA target recognition. Nat Genet, 39, 1278–1284.

19.

Kim

, Pertea

, Trapnell

, Pimentel

, Kelley

, and Salzberg

S.L.

(2013). TopHat2: accurate alignment of transcriptomes in the presence of insertions, deletions and gene fusions. Genome Biol, 14, R36.

20.

Krüger

, and Rehmsmeier

(2006). RNAhybrid: microRNA target prediction easy, fast and flexible. Nucleic Acids Res, 34, W451–W454.

21.

Langmead

, Trapnell

, Pop

, and Salzberg

S.L.

(2009). Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol, 10, R25.

22.

, Jia

, Mallik

, Fei

, Yoshioka

, Suzuki

, et al. (2019). Critical microRNAs and regulatory motifs in cleft palate identified by a conserved miRNA–TF–gene network approach in humans and mice. Brief Bioinform, 21, 1465–1478.

23.

Little

J.C.

, Redwood

K.L.

, Granger

S.P.

, and Jenkins

(2010). In vivo cytokine and receptor gene expression during the rat hair growth cycle. Analysis by semi-quantitative RT-PCR. Exp Dermatol, 5, 202–212.

24.

Liu

J.P.

, Baker

, Perkins

A.S.

, Robertson

E.J.

, and Efstratiadis

(1993). Mice carrying null mutations of the genes encoding insulin-like growth factor I (Igf-1) and type 1 IGF receptor (Igf1r). Cell, 75, 59–72.

25.

Love

M.I.

, Huber

, and Anders

(2014). Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol, 15, 550.

26.

Luo

, Dou

, Xie

, Lu

, Han

, Zhou

, et al. (2021). miR-203a-3p promotes loureirin A-induced hair follicle stem cells differentiation by targeting Smad1. Anat Rec, 304, 531–540.

27.

Madaan

, Verma

, Singh

A.T.

, and Jaggi

(2018). Review of hair follicle dermal papilla cells as in vitro screening model for hair growth. Int J Cosmet Sci, 40, 429–450.

28.

Mao

, Cai

, Olyarchuk

J.G.

, and Wei

(2005). Automated genome annotation and pathway identification using the KEGG Orthology (KO) as a controlled vocabulary. Bioinformatics, 21, 3787–3793.

29.

Mardaryev

A.N.

, Ahmed

M.I.

, Vlahov

N.V.

, Fessing

M.Y.

, Gill

J.H.

, Sharov

A.A.

, et al. (2010). Micro-RNA-31 controls hair cycle-associated changes in gene expression programs of the skin and hair follicle. FASEB J, 24, 3869–3881.

30.

Marioni

J.C.

, Mason

C.E.

, Mane

S.M.

, Stephens

, and Gilad

(2008). RNA-seq: an assessment of technical reproducibility and comparison with gene expression arrays. Genome Res, 18, 1509–1517.

31.

Martin

(2011). Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet J, 17, 10.

32.

Merrill

B.J.

, Gat

, DasGupta

, and Fuchs

(2001). Tcf3 and Lef1 regulate lineage differentiation of multipotent stem cells in skin. Genes Dev, 15, 1688–1705.

33.

Ojeh

, Pastar

, Tomic-Canic

, and Stojadinovic

(2015). Stem cells in skin regeneration, wound healing, and their clinical applications. Int J Mol Sci, 16, 25476–25501.

34.

Pandini

, Wurch

, and Akla

(2007). Functional responses and in vivo anti-tumour activity of h7C10: a humanised monoclonal antibody with neutralising activity against the insulin-like growth factor-1 (IGF-1) receptor and insulin/IGF-1 hybrid receptors. Eur J Cancer, 43, 1318–1327.

35.

Paus

, and Foitzik

(2010). In search of the “hair cycle clock”: a guided tour. Differentiation, 72, 489–511.

36.

Reynolds

A.J.

, Oliver

R.F.

, and Jahoda

C.A.

(1991). Dermal cell populations show variable competence in epidermal cell support: stimulatory effects of hair papilla cells. J Cell Sci, 98 (Pt 1), 75–83.

37.

Shen

, Yu

, Fang

, Yao

, and Yang

(2017). Beta-catenin can induce hair follicle stem cell differentiation into transit-amplifying cells through c-myc activation. Tissue Cell, 49, 28.

38.

Stewart

C.E.

, and Rotwein

(1996). Growth, differentiation, and survival: multiple physiological functions for insulin-like growth factors. Physiol Rev, 76, 1005–1026.

39.

Tudorita

, Geraldine

, Valentina

, Cedric

, Lowry

W.E.

, Michael

, et al. (2004). Defining the epithelial stem cell niche in skin. Science, 303, 359–363.

40.

Van Genderen

, Okamura

R.M.

, Fariñas

, Quo

R.G.

, Parslow

T.G.

, Bruhn

, et al. (1994). Development of several organs that require inductive epithelial-mesenchymal interactions is impaired in LEF-1-deficient mice. Genes Dev, 8, 2691–2703.

41.

Wen

, Shen

, Shi

, and Tang

(2012). miREvo: an integrative microRNA evolutionary analysis platform for next-generation sequencing experiments. BMC Bioinform, 13, 140.

42.

Williams

, Crompton

R.A.

, Thomason

H.A.

, Campbell

, Singh

, Mcbain

A.J.

, et al. (2017). Cutaneous Nod2 expression regulates the skin microbiome and wound healing in a murine model. J Invest Dermatol, 137, S0022202X17316615.

43.

, Fu

, Cao

, Yu

, Tang

, and Zhao

(2014). Identification of differentially expressed miRNAs between white and black hair follicles by RNA-sequencing in the goat (Capra hircus). Int J Mol Sci, 15, 9531–9545.

44.

Xuejuan

, Lei

, Jinyu

, Shichao

, Xiaohua

, Xiaoguang

, et al. (2013). miR-18b inhibits TGF-β1-induced differentiation of hair follicle stem cells into smooth muscle cells by targeting SMAD2. Biochem Biophys Res Commun, 438, 551–556.

45.

Yan

, Gao

, Ding

, Liu

, Li

, Jin

, et al. (2019). Exosomal micro RNAs derived from dermal papilla cells mediate hair follicle stem cell proliferation and differentiation. Int J Biol Sci, 15, 1368–1382.

46.

Yang

, and Peng

(2010). Unveiling hair follicle stem cells. Stem Cell Rev Rep, 6, 658–664.

47.

Young

M.D.

, Wakefield

M.J.

, Smyth

G.K.

, and Oshlack

(2010). Gene ontology analysis for RNA-seq: accounting for selection bias. Genome Biol, 11, R14.

48.

Yuan

, Wang

, Geng

, He

, Qu

, and Chen

(2013). Discovery of cashmere goat (Capra hircus) microRNAs in skin and hair follicles by Solexa sequencing. BMC Genom, 14, 511.

49.

Zhang

, Tsai

P.C.

, Gonzalez-Celeiro

, Chung

, Boumard

, Perdigoto

C.N.

, et al. (2016). Hair follicles' transit-amplifying cells govern concurrent dermal adipocyte production through Sonic Hedgehog. Genes Dev, 30, 2325.

50.

Zhang

, Yu

, Shi

, Huang

, Wang

, Yang

, et al. (2013). Lef1 contributes to the differentiation of bulge stem cells by nuclear translocation and cross-talk with the Notch signaling pathway. Int J Med Sci, 10, 738–746.

51.

Zhou

, Kang

, Ma

, Wang

, Gao

, Yang

, et al. (2018). Integrative analysis reveals ncRNA-mediated molecular regulatory network driving secondary hair follicle regression in cashmere goats. BMC Genom, 19, 222.

52.

Zhou

G.X.

, Yuan

, He

X.L.

, Kang

D.J.

, Wang

X.L.

, and Chen

Y.L.

(2017). Effect of miR-125b on dermal papilla cells of goat secondary hair follicle. Electron J Biotechnol, 25, 64–69.

53.

Zhou

, Chen

, Li

, Hu

, Huang

, et al. (2010). Integrated profiling of microRNAs and mRNAs: microRNAs located on Xq27.3 associate with clear cell renal cell carcinoma. PLoS One, 5, e15224.

54.

Zhou

, Byrne

, Jacobs

, and Fuchs

(1995). Lymphoid enhancer factor 1 directs hair follicle patterning and epithelial cell fate. Genes Dev, 9, 700–713.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.77 MB

0.01 MB

3.76 MB

0.02 MB

0.01 MB

0.10 MB

0.06 MB

0.01 MB