Wave patterns in lambskin hair follicles are an important factor determining the quality of sheep’s wool. Hair follicles in lambskin from Hu sheep, a breed unique to China, have 3 types of waves, designated as large, medium, and small. The quality of wool from small wave follicles is excellent, while the quality of large waves is considered poor. Because no molecular and biological studies on hair follicles of these sheep have been conducted to date, the molecular mechanisms underlying the formation of different wave patterns is currently unknown. The aim of this article was to screen the candidate microRNAs (miRNA) and genes for the development of hair follicles in Hu sheep. Two-day-old Hu lambs were selected from full-sib individuals that showed large, medium, and small waves. Integrated analysis of microRNA and mRNA expression profiles employed high-throughout sequencing technology. Approximately 13, 24, and 18 differentially expressed miRNAs were found between small and large waves, small and medium waves, and medium and large waves, respectively. A total of 54, 190, and 81 differentially expressed genes were found between small and large waves, small and medium waves, and medium and large waves, respectively, by RNA sequencing (RNA-seq) analysis. Differentially expressed genes were classified using gene ontology and pathway analyses. They were found to be mainly involved in cell differentiation, proliferation, apoptosis, growth, immune response, and ion transport, and were associated with MAPK and the Notch signaling pathway. Reverse transcription-polymerase chain reaction (RT-PCR) analyses of differentially-expressed miRNA and genes were consistent with sequencing results. Integrated analysis of miRNA and mRNA expression indicated that, compared to small waves, large waves included 4 downregulated miRNAs that had regulatory effects on 8 upregulated genes and 3 upregulated miRNAs, which in turn influenced 13 downregulated genes. Compared to small waves, medium waves included 13 downregulated miRNAs that had regulatory effects on 64 upregulated genes and 4 upregulated miRNAs, which in turn had regulatory effects on 22 downregulated genes. Compared to medium waves, large waves consisted of 13 upregulated miRNAs that had regulatory effects on 48 downregulated genes. These differentially expressed miRNAs and genes may play a significant role in forming different patterns, and provide evidence for the molecular mechanisms underlying the formation of hair follicles of varying patterns.

Machine learning (ML) methods have shown promising results in identifying genes when applied to large transcriptome datasets. However, no attempt has been made to compare the performance of combining different ML methods together in the prediction of high feed efficiency (HFE) and low feed efficiency (LFE) animals. In this study, using RNA sequencing data of five tissues (adrenal gland, hypothalamus, liver, skeletal muscle, and pituitary) from nine HFE and nine LFE Nellore bulls, we evaluated the prediction accuracies of five analytical methods in classifying FE animals. These included two conventional methods for differential gene expression (DGE) analysis (t-test and edgeR) as benchmarks, and three ML methods: Random Forests (RFs), Extreme Gradient Boosting (XGBoost), and combination of both RF and XGBoost (RX). Utility of a subset of candidate genes selected from each method for classification of FE animals was assessed by support vector machine (SVM). Among all methods, the smallest subsets of genes (117) identified by RX outperformed those chosen by t-test, edgeR, RF, or XGBoost in classification accuracy of animals. Gene co-expression network analysis confirmed the interactivity existing among these genes and their relevance within the network related to their prediction ranking based on ML. The results demonstrate a great potential for applying a combination of ML methods to large transcriptome datasets to identify biologically important genes for accurately classifying FE animals.

Wool is the critical textile raw material which is produced by the hair follicle of sheep. Therefore, it has important implications to investigate the molecular mechanism governing hair follicle development. Due to high cellular heterogeneity as well as the insufficient cellular, molecular, and spatial characterization of hair follicles on sheep, the molecular mechanisms involved in hair follicle development and wool curvature of sheep remains largely unknown. Single-cell RNA sequencing (scRNA-seq) technologies have made it possible to comprehensively dissect the cellular composition of complex skin tissues and unveil the differentiation and spatial signatures of epidermal and hair follicle development. However, such studies are lacking so far in sheep. Here, single-cell suspensions from the curly wool and straight wool lambskins were prepared for unbiased scRNA-seq. Based on UAMP dimension reduction analysis, we identified 19 distinct cell populations from 15,830 single-cell transcriptomes and characterized their cellular identity according to specific gene expression profiles. Furthermore, novel marker gene was applied in identifying dermal papilla cells isolated in vitro. By using pseudotime ordering analysis, we constructed the matrix cell lineage differentiation trajectory and revealed the dynamic gene expression profiles of matrix progenitors' commitment to the hair shaft and inner root sheath (IRS) cells. Meanwhile, intercellular communication between mesenchymal and epithelial cells was inferred based on CellChat and the prior knowledge of ligand–receptor pairs. As a result, strong intercellular communication and associated signaling pathways were revealed. Besides, to clarify the molecular mechanism of wool curvature, differentially expressed genes in specific cells between straight wool and curly wool were identified and analyzed. Our findings here provided an unbiased and systematic view of the molecular anatomy of sheep hair follicle comprising 19 clusters; revealed the differentiation, spatial signatures, and intercellular communication underlying sheep hair follicle development; and at the same time revealed the potential molecular mechanism of wool curvature, which will give important new insights into the biology of the sheep hair follicle and has implications for sheep breeding.

Hu sheep lambskin is a unique white lambskin from China that exhibits three types of flower patterns, including small waves, medium waves, and large waves, with small waves considered the best quality. However, our understanding of the molecular mechanism underlying flower pattern formation in Hu sheep lambskin is limited. The aim of the present study was to further explore the relevance between candidate microRNAs (miRNAs) and developmental characteristics of hair follicles and screen miRNAs for later functional validation. Herein, we employed Illumina Hiseq 2500 to identify differentially expressed miRNAs in hair follicles of different flower patterns with small, medium, and large waves to construct a comprehensive sequence database on the mechanism of hair follicle development. Paraffin sections of lambskin tissue were prepared to assess the structure of different hair follicles. Expression levels of candidate miRNAs in different flower patterns were analyzed by relative quantitation using real-time PCR, combined with histological observation and micro-observation technologies, and the correlation between expression levels of candidate miRNAs and histological properties of hair follicles was analyzed by using SPSS 17.0. A total of 522 differentially expressed miRNAs were identified, and RNA-seq analysis detected 7,266 target genes in different groups of flower patterns. Gene ontological analysis indicated these target genes were mainly involved in cell proliferation, differentiation, growth, apoptosis, and ion transport, and 14 miRNAs, including miR-143, miR-10a, and let-7 were screened as candidate miRNAs in Hu sheep hair follicle growth and development. In the same field of vision, variance analysis showed that the number of secondary follicles in small waves was significantly larger than that in large and medium waves (P<0.01); the diameter of the primary and secondary follicles in large waves was respectively larger than those in medium and small waves (P<0.01). Combined with correlation analysis between miRNA expression and histological properties of hair follicles, highly significant differences in miRNA-143 expression levels between large and small waves were observed (P<0.01), and significant differences in the miRNA-10a expression levels between large and small waves (P<0.05) and in let-7i expression levels between large and medium waves were observed (P<0.05). Significant differences in the expression of novel miRNAs of NW_004080184.1_6326 between medium and large waves were detected (P<0.05), and highly significant differences between medium and small waves were observed (P<0.01). Highly significant differences in the expression level of NW_004080165.1_8572 between medium and large and small waves (P<0.01), in that of NW_004080181.1_3961 between medium and small waves (P<0.01), and in that of NW_004080190.1_13733 between medium and large waves were observed, whereas no significant differences in the other miRNAs among large, medium, and small waves were detected. Overall, the present study showed that miRNA-143, miRNA-10a, let-7i, NW_004080184.1_6326, NW_004080165.1_8572, NW_004080181.1_3961, and NW_004080190.1_13733 could be considered as important candidate genes, indicating these seven miRNAs may play significant roles in hair follicle growth and development in Hu sheep lambskin.

Dorper and Hu sheep exhibit different characteristics in terms of reproduction, growth, and meat quality. Comparison of the genomes of two breeds help to reveal important genomic information. In this study, whole genome resequencing of 30 individuals (Dorper, DB and Hu sheep, HY) identified 15,108,125 single nucleotide polymorphisms (SNPs). Population differentiation (Fst) and cross population extended haplotype homozygosity (XP-EHH) were performed for selective signal analysis. In total, 106 and 515 overlapped genes were present in both the Fst results and XP-EHH results in HY vs DB and in DB vs HY, respectively. In HY vs DB, 106 genes were enriched in 12 GO terms and 83 KEGG pathways, such as ATP binding (GO:0005524) and PI3K-Akt signaling pathway (oas04151). In DB vs HY, 515 genes were enriched in 109 GO terms and 215 KEGG pathways, such as skeletal muscle cell differentiation (GO:0035914) and MAPK signaling pathway (oas04010). According to the annotation results, we identified a series of candidate genes associated with reproduction (UNC5C, BMPR1B, and GLIS1), meat quality (MECOM, MEF2C, and MYF6), and immunity (GMDS, GALK1, and ITGB4). Our investigation has uncovered genomic information for important traits in sheep and provided a basis for subsequent studies of related traits.

Lambskin of the Hu sheep exhibits high economic value due to its water-wave pattern. Wool curvature is the key factor of the pattern types and quality of lambskin, and it is formed by the interaction between dermal papilla cells and hair matrix cells in the hair follicle, which is regulated by various genes and signaling pathways. Herein, three full-sibling pairs of two-day-old healthy lambs (n = 6) were divided into a straight wool group (ST) and small waves group (SM) with three repetitions. RNA-seq was applied to determine the expression profile of mRNAs and lncRNAs in Hu sheep hair follicles. 25 differentially expressed mRNAs and 75 differentially expressed lncRNAs were found between SM and ST. FGF12, ATP1B4, and TCONS_00279168 were probably associated with hair follicle development. Then, Gene Ontology (GO) and KEGG enrichment analysis were implemented for the functional annotation of target genes of differentially expressed lncRNAs. The results showed that many genes, such as FGF12 and ATP1B4, were found enriched in PI3K-Akt signaling, MAPK signaling, and Ras signaling pathway associated with hair follicle growth and development. In addition, the interaction network of differentially expressed lncRNAs and mRNAs showed that a total of 6 differentially expressed lncRNAs were associated with 12 differentially expressed mRNAs, which may be as candidate mRNAs and lncRNAs. TCONS_00279168 may target ATP1B4 and FGF12 to regulate MAPK, PI3K-Akt, Ras signaling pathways involved in the sheep hair follicle development process. These results will provide the basis for exploring hair follicle development.