Dios reversed sustained transcriptional changes in KK-Ay diabetic mice.
To study the mechanism of Dios on KK-Ay diabetic mice, the liver samples were collected for transcription analysis. We calculated FPKM for all samples and performed PCA analysis (Fig. 3A) and correlation analysis (Fig. 3B), which indicates that our samples have good repeatability. To identify gene sets with a statistically significant difference in hepatocytes, we screened the differentially expressed genes (DEGs) between Control vs Model and Model vs Dios by the following criteria: |logFC| >1, FDR < 0.05 and p value < 0.05. We found 4182 DEGs between model and control (Fig.3C), of those DEGs, 1967 genes were up-regulated and 2215 genes were down-regulated. Besides, a total of 173 DEGs between Model and Dios were revealed. Among them, 101 genes were up-regulated and 72 genes were down-regulated respectively. We used hierarchical cluster analysis (Fig. 3D-3E) to compare the DEGs in three groups, which shows that the similarity of expression patterns of three biological replicates. To identify the biological features of DEGs, we performed significant GO analysis by the clusterProfiler R package, and the biological processes analysis revealed that the DEGs between Model and Dios group (Fig 3F) were associated with glucose metabolism including fructose metabolic process, carbohydrate metabolic process, carboxylic acid biosynthetic process. Among significant GO terms of down-regulated genes (Fig 3G), it was obvious that most GO terms were associated with organic anion transport, carboxylic acid transport, organic acid transport and lipid transport. To further explore the potential mechanism of DEGs between Model and Dios group, KEGG pathway analysis was performed using the clusterProfiler R package. The results of KEGG analysis revealed that DEGs were mainly associated with glucose metabolism including glycolysis/gluconeogenesis, biosynthesis of amino acids, fructose and mannose metabolism (Fig 3H). Interestingly, we found that pathways involved in glucose metabolism were the most enriched up-regulated pathways (Fig 3I), which is consistent with the reactome functional enrichment analysis (Fig 3J). Those results revealed that the potential mechanism of Dios ameliorated T2DM may be through regulating glucose metabolism.
Dios treatment ameliorated HFD-induced glucose metabolism disorder inKK-Ay diabetic mice.
To clarify the potential mechanism of Dios in KK-Ay diabetic mice, the glucose metabolism and insulin resistance-related signaling proteins in liver and skeletal muscle were observed by western blot analysis. Our data showed that a continuous HFD diet can lead to severe insulin resistance in the liver (Fig. 4A-4C) and skeletal muscle (Fig. 4D-4F). However, treatment with Dios and Met could significantly ameliorate insulin resistance by activating the IRS/PI3K/AKT signaling pathway. Once AKT is activated, it participates in glucose metabolism related pathways caused by insulin. We found that Dios promoted glycogen synthesis in liver and skeletal muscle by phosphorylating GSK-3β and activating GS. Besides, the level of phosphorylation AS160 and Glut4 were significantly up-regulated in skeletal muscle by Dios treatment, which suggests that Dios promote glucose transport.
Dios treatment reverses HFD-induced gut dysbiosis in KK-Ay diabetic mice.
To further explore the effects of Dios on gut microbiota composition, the 16S bacterial rRNA genes composing the V4 hypervariable regions in the faeces of KK-Ay mice were performed. After removing low-quality sequences, a total of 72292 tags were generated. The remaining clean tags were clustered into OTUs (Operational Taxonomic Units) based on 97% similarity. UniFrac distance-based principal coordinate analysis (PCoA) revealed a relative clustering of gut microbiota within each group (Fig 5A). Remarkably, the microbes in Dios and Met groups were more closely clustered relative to ND groups, which is an indication that Dios treatment induced similar microbial composition changes. Unweighted Pair-group Method with Arithmetic Mean (UPGMA) indicates a statistically significant separation between Model group and H-Dios, Met groups (Fig 5B). Notably, Firmicutes and Bacteroidetes are the most abundant microbiota at the phylum level. Compared with the CK group, the content of the phylum Firmicutes in the Model group significantly increased and theBacteroidetes abundance significantly increased. However, a remarkable reversal of this pattern was observed following Dios and Met treatment (Fig 5C-5F). An increasedFirmicutes/Bacteroidetes ratio is a commonly used indicator of diabetic mice as reported in a previous study, as shown in Fig 5D, the increased ratio of the phylum Firmicutes /Bacteroideteswere observed in the diabetic model group. On the contrary, Dios and Met treatment remarkably suppressed the ratio of the phylum Firmicutes/Bacteroidetes . At the same time, we interestingly found that the ratio ofActinobacteria phylum was significantly higher in the H-Dios group than in the other groups. Next, we further analyzed the relative abundances of the gut microbiota at the species level, as illustrated in Fig 5G-5K, there were several bacteria decreased in the diabetic Model group, which were markedly increased by Dios treatment, includingAerococcus_viridans (70 fold), C._glu (155 fold), Combined, these data suggested that Dios treatment may effectively reverse gut microbiota dysbiosis in diabetic mice.
C. glu and Dios treatmentreducedHFD-induced body weight and blood glucose gain in KK-Ay diabetic mice.
Recent studies have shown that the ability of the gut microbiota to modulate obesity can be transferred to other animals. To explore whether the gut microbiota of Dios-treated animals may improve the condition of KK-Ay diabetic mice, we selected the highest abundance bacterium(C.glu ) for further study. After treatment with Dios andC.glu for 4 weeks, the related indexes were measured. As shown in Fig. 6A-6B, it is noteworthy that the body weight and blood glucose of Dios, C.glu and Dios+C. glu groups were significantly reduced compared to the Model group at the end of the feeding period, similar to the results shown above (Fig 1A and 1B). In addition, our results further showed that the weight gain (Fig. 6C), liver index (Fig. 6D) and adipose index (Fig. 6E) of Dios,C.glu and Dios+C. glu groups were significantly lower than that of KK-Ay diabetic mice. On the other hand, Expression of TC(Fig 6F), TG (Fig 6G), LDL-C (Fig 6H) was also increased and HDL-C (Fig 6I) and insulin (Fig 6J) decreased in KK-Ay diabetic mice compared with NC group at the end of transplantation. Surprisingly, Dios, C.glu and Dios+C. glu treatment significantly decreased the serum levels of TC, TG, LDL-C and increased the serum level of HDL-C and insulin. Furthermore, the liver histopathology also produced significant changes in less vacuolation and lipid droplets when treatment with Dios,C.glu and Dios+C. glu. compared with KK-Ay diabetic mice (Fig 6K).