Differential Gene Expression Analysis Of Capra Hircus Skin Between Anagen And Telogen Of Hair Growth Cycle

The next-generation DNA sequencing technology together with bioinformatics tools have been a potent approach to obtain gene expression profiles of eukaryote or tissue under specific physiological conditions during certain developmental periods. In the present study, using Iuulmina/Solexa sequencing technology, we deeply sequenced Inner Mongolia Cashmere goat skin transcriptomes which are derived from three different skin sites corresponding anagen and telogen of the hair cycle. After de novo assembly, we then performed differential gene expression analysis so as to better understand the molecular mechanisms of hair growth and provide potential candidates for further characterizing gene functions associated cashmere fiber growth.1. The belly, back and side of the body of the skin tissues from two female cashmere goats during telogen and anagen of the hair cycle, respectively, were sampled for this study. Six RNA-Seq libraries were then constructed and subsequently sequenced by using Illumina/Solexa sequencing technology. A total of280million short reads was obtained. De novo assembly short reads generated57,568transcriptome sequences greater300bp with the average length of928bp and N50length of l,397bp. The accumulative size of the transcriptome sequences reached53Mb. Annotation of57,568transcriptome sequences to NR database revealed that24,640sequences have homologs. Ab initio CDS prediction of57,568transcriptome sequences revealed23,602contigs contained putative protein coding sequences representing a total of24,058CDS. The number of putative CDS is similar to that of the blastx search in the public database. We randomly selected ten sequences which predicted to have a putative full-length CDS but without harboring homologs in the NR database to perform RT-PCR and Sanger sequencing so as to validate transcript accuracy after de novo assembly. Sequence alignment analysis suggested that greater than98%sequence similarities were present in each comparison between Sanger sequencing result and putative CDS except one is RT-PCR negative. In addition, mapping short reads back to57,568contigs revealed that88.1%of the total reads could be uniquely mapped to the reference,92.4%of which were paired-end reads. These results indicated that our assembly has generated high quality sequences relecting the overall transcriptomes of goat skin tissues during anagen and telogen of the hair cycle.2. Based on the de novo assembled transcriptome sequences, we paired the libs coming from the same skin region but different in hair cycle. We then constructed gene expression spectrums for each pairs. Diffenertial gene expression profiling between each comparison pairs obtained5,501consistently differentially expressed genes. We subsequently carried out KOG, KEGG and GO enrichment analysis of these consistently differentially expressed genes compared with the transcriptome background. We found that genes associated with cytoskeleton and extracllualr matrix were over-represented in the differentially expressed genes dataset and participated into the ECM-receptor interaction, focal adhesion and gap junction of KEGG pathway. These results indicate the interaction between cell and extracelluar matrix is required for active hair growth. In addition, many signal molecules and receptors were identified as differentially epressed gene between telogen and anagen and most of them showed up-regulated expression during anagen compared with telogen in all three pairs. Some of these molecules have been demonstrated to be essential for hair follicle patterning and development. In addition, we select eight differentially expressed genes by performing qPCR experiment and compared with RNA-Seq result to validate sequencing accuracy The results showed that high consistency between two techniques which would indicate our DEG dataset is confident. 3. Based on the Yunnan Black Goat genome sequence, we analyzed gene expression differences of the Cashmere goat between telogen and telogen of hair cycle. A total of2,339consistently differentially expressed genes were identified. Functional enrichment analysis of these differentially expressed genes suggested that genes associated with signal transduction molecules, extracllualr matrix, cytoskeleton and cell adhesion molecules were over-representation, which are similar to the transcriptome sequence-based anslysis. These results indicate that profiling of differentially expressed genes by using these two different models have higher consistency between each other.