The present study was designed to detect the gene expression profiles associtaed with Dioscorea bulbifera L. (DB)-induced liver injury in mice by RNA-sequenceing (RNA-Seq), a next-generation sequencing technology, and analyzed its molecular mechanisms.

Forty SPF Kunming female mice (4-6 weeks old, 20±2 g)were purchased from Shanghai Slack Laboratory Animal Co. Ltd (animal certification number: 2015000533474). Animals were randomly divided into the control group, high, moderate and low dose groups (n=10 in each group). Mice in the latter three groups were administered intragastrically (i.g.) with DB decoction (24, 12 and 6 g/kg/d, respectively) for 21 consecutive days, while the control group were given equal volume of pure water.After experiment the mice were killed, and their livers were collected. The optimum dose to induce hepatotoxicity was chosed by compared the treated groups and the control group in terms of mice weight gain, liver index, serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels and liver tissue pathology after H&E staining. Two digital gene expression (DGE) libraries of mice liver in the control and the DB group were created by RNA-seq, following with data analysis and differential gene expression analysis. Finally, Gene ontology (GO) enrichment and kyoto encyclopedia of genes and genomes (KEGG) pathway analyses were implemented on all differentially expressed genes to elucidate their biological functions and processes. The analyses were carried out using the SPSS 20.0 software package, and the results were expressed as ±s. Comparison between two groups was done with independent sample T-test, with P<0.05 considered as statistically significant. Moreover, padj<0.05 was set as the threshold for significantly differential expression in DGE library analysis.

Firstly, obvious liver injury was observed only in the high-dose group, which mainly manifested as decreased weight gain, elevated liver index and serum ALT and AST levels (P<0.05). Histopathological changes were apparent in liver tissue, with destruction of structure, liver cell swelling and cord derangement and infiltration of neutrophilic granulocytes. A total of 13, 214, 693 and 11, 124, 617 raw reads were generated from the control and DB exposure DGE libraries respectively, with 12, 819, 933 and 10, 786, 300 clean reads left for assembly after removing adaptor related, containing N and low quality reads. According to the statistical result of the two DGE libraries, 302 genes were detected differentially expressed with 148 up-regulated and 164 down-regulated. And 9537 high-quality sequences were observed co-expression among two DGE libraries, with 702 in the DB group only and 539 in the normal group only, after clustering analysis. GO annotation there mainly had molecular function and biological process. All differentially expressed genes were mapped to 140 pathways of KEGG database, and the top 8 significantly enriched KEGG pathways were retinol metabolism, arachidonic acid metabolism, PPAR signaling pathway, drug metabolism-cytochrome P450, metabolism of xenobiotics by cytochrome P450, steroid hormone biosynthesis, glutathione metabolism and biosynthesis of unsaturated fatty acids.

The results showed that DB may induce hepatotoxicity by seriously damaging structures and functions of liver and influence related gene expressions. The DGE sequencing technology provides a better way to revealed that the molecular mechanisms of DB induced hepatotoxicity may include generation of oxygen free radical after DB metabolized by the CYP450 enzyme in liver, which resulted in lipid peroxidation of cellular membrane and organelle and decrease of anti-oxidation capacity, or DB may cause liver injury by increasing intrahepatic lipid metabolism and fatty acid oxidation.