Construction Of AnimalTFDB Database And Network Analysis On Carcinogenic Mechanism Of K562 Microvesicle

Transcription factors (TFs) are proteins that bind to specific DNA sequences to activate or repress gene expression. They play critical roles in the process of transcription regulation. Identifying and annotating all TFs are crucial steps for illustrating their functions and understanding the transcriptional regulation. TFs and miRNAs,as key regulators of gene expression, are able to co-regulate the expression of target genes in forms of feed-forward loops (FFLs) or feedback loops (FBLs). These regulatory motifs have been applied successfully into the researches of breast cancer and glioblastoma multiforme. Our cooperators in their previous work demonstrated that MVs derived from K562 chronic myeloid leukemia (CML) cells can transform mononuclear cells (MNCs)from normal hematopoietic transplants to acute leukemia-like cancer cells. During the transformation, a new group of leukemia-like cells could be observed after 14 d of consecutive incubation with MVs, and most of them were transformed into leukemia-like cells after 21 d. This provides a model to generate leukemia-like cells and study the leukemogenesis of CML transformation, but little is understood about the mechanism. In this study, we firstly predicted the TFs in all of the sequenced animal genomes and provided abundant annotation for each gene. Then, we analyzed the regulatory mechanisms of K562-MVs transforming normal MNCs through constructing the TF-miRNA co-regulatory network and got the following results.Firstly, we collected and curated a comprehensive animal TF family list from literatures and constructed the Hidden Markov Models (HMMs) for each of them with their DNA binding domains. We utilized these HMMs to predict TFs in all of the sequenced animal genomes and organized the predicted results into AnimalTFDB database.In AnimalTFDB, there are 72,336 TF genes in 70 TF families, 21,053 transcription co-factor genes and 6,502 chromatin remodeling factor genes from 65 species covering main animal lineages. To better serving the community, we provided abundant annotations.In the gene level, the annotations includ basic gene information, functional domain, Gene Ontology, KEGG and Biocarta pathways, protein-protein interaction, gene expression,gene phenotype, paralogs, orthologs and TF targets. In the TF family level, we provided detailed descriptions for the structures and functions of TF families. For each TF family,we also did multiple sequence alignment with TF DNA-binding domains and built phylogenetic trees based on the alignment. In addition, a TF prediction sever and a BLAST server were available, which are helpful for users to identify TFs from their own protein sequences.Secondly,we thoroughly analyzed the regulatory mechanism of K562-MVs transforming MNCs process. RNA sequencing and small RNA sequencing for samples of K562-MVs, MNCs, one week (1W), two weeks (2W) and three weeks (3W) cells after MVs incubation were performed. Firstly, we utilized multiple methods to investigate the change of gene expression in the transformation process. Through the differential expression analysis, we identified 3,717 differentially expressed genes and 143 differentially expressed miRNAs in the three stages of transformation. The GO enrichment result indicated most of the differentially expressed genes in each stage are enriched in immune related pathways. However, there were also some genes enriched in cell cycle and cell death pathways in 2W-3W process. The gene clustering result displayed genes in cell cycle, DNA replication and repair, oxidative phosphorylation and metabolism process were up-regulated in 3W sample. The genes related to apoptosis, immune system and signal transmission were down-regulated in 3W sample. In addition, most of the oncomiRs were up-regulated in 3W, such as clusters miR-17-92 and miR-183/-96/-182,and most of the miRNAs that down-regulated in 3W are tumor suppressed miRNAs, such as let-7 and miR-181 families and miR-15a/16-5p. Next, we studied the dysregulatory mechanisms of genes and miRNAs through constructing TF-miRNA co-regulatory network. Expression and regulatory network analyses revealed that a number of TFs and miRNAs were responsible for the dysregulation of these pathways and were required for the transformation, especially regulators highly expressed in MVs, such as TFs YBX1,STAT5A, MYC, GATA2, and miRNAs miR-146b-5p, miR-17-92 and miR-92b-3p. They also could up-regulate the expression of oncogenic TFs and miRNAs, and down-regulate the anti-oncogenic TFs and miRNAs. Our experiments revealed that up-regulation of miR-146b-5p in K562 could accelerate the transformation process via targeting NUMB and NOTCH2.Our study predicted the TFs in all of the sequenced animal genomes systematically and organized the results into AnimalTFDB, which provides a solid resource for researching the function of TFs and the mechanism of transcriptional regulation. We also identified several TFs and miRNAs that highly expressed in K562 MV could promote the transformation of MNCs into leukemia-like cells and verified the up-regulation of miR-146b-5p in K562 cells do accelerate the transformation process. This study provided important clues for understanding the leukemogenesis of CML and new biomarkers for the diagnosis and treatment of CML.