Effects of D-Chiro-Inositol on Glucose Metabolism in db/db Mice and the Associated Underlying Mechanisms

In this study, we observed the effect of D-chiro-inositol (DCI) on glucose consumption in type 2 diabetic db/db mice, and investigated the relevant mechanism. We discovered that the stability of 24-h blood glucose under the nonfasting condition and decreased glucose tolerance were both alleviated after treatment with DCI. Moreover, the content of glycosylated protein and advanced glycation end products in the serum was reduced, the damage in the liver tissue was alleviated, and the synthesis of liver glycogen was significantly promoted. In addition, DCI increased the expression of insulin receptor substrate 2 (IRS2), phosphatidylinositol 3-kinase (PI3K), protein kinase B (AKT), glucose transporters 4 (GLUT4), and phospho-AKT (S473) protein. In contrast, DCI decreased the expression level of glycogen synthase kinase 3β (GSK3β) protein in liver tissue to various degrees, as shown by immunohistochemistry and western blotting. Furthermore, DCI increased the mRNA expression of IRS2, PI3K, AKT, and GLUT4, and reduced that of GSK3β in liver tissue, as demonstrated by polymerase chain reaction. Finally, DCI promoted glucose consumption in high glucose-stimulating HepG2 cells and increased the expression of IRS2 protein in HepG2 cells, as revealed by fluorescence staining and flow cytometry. Our results indicate that DCI can significantly improve glucose metabolism in diabetic mice and HepG2 cells. This effect may be associated with the upregulation of IRS2, PI3K, AKT, and GLUT4 and downregulation of GSK3β.


INTRODUCTION
Diabetes mellitus is a lifelong metabolic disease characterized by hyperglycemia, which is caused by insufficient secretion or dysfunction of insulin (Hummel et al., 1966;Rutter et al., 1999;Tian et al., 2017). At present, the incidence of diabetes is increasing annually. The prevalence of diabetes in China is approximately 11%, ranking first worldwide (Cheung et al., 2018). As one of the independent risk factors of type 2 diabetes mellitus (T2DM), insulin resistance (IR) is closely associated with various metabolic diseases (Cordain et al., 2003;Mehanna et al., 2018). In addition, IR induces various diabetic complications (e.g., diabetic liver injury) (Ji et al., 2009), which occur throughout the entire developmental process of T2DM and cause harm to diabetic patients. Therefore, it is of great importance to identify effective drugs for the prevention and reduction of IR.
D-chiro-inositol (DCI) is one of the nine isomers of inositol, which has optical rotation and is present in plants (e.g., buckwheat and soybean) (Steadman et al., 2000;Song et al., 2016). This type of inositol is frequently used as a dietary health supplement in the clinical treatment of T2DM (Gao et al., 2016b), promoting liver lipid metabolism and possessing other physiological functions (Ken-Ichi et al., 2006;Whiting et al., 2013). It has been reported that, as an insulin sensitizer, the structure of DCI is similar to that of pH2.0 insulin regulatory mediators. The latter is an important component of insulin's second messenger phosphoinositol glucan, which may promote lipid metabolism in liver and insulin signal transduction (Larner et al., 2010). Furthermore, DCI may significantly increase glucose consumption in IR-HepG2 cells and significantly reduce their IR (Gao et al., 2016a).
However, now the molecular mechanism isn't very clear, still needs to be improved. Therefore, in this study, we further investigated the molecular mechanism through the entire IRS2/ PI3K/AKT/GSK3b/GLUT4 signaling pathway. In the present study, an intervention with DCI in db/db mice and high glucose-stimulating HepG2 cells was used to observe its effect on glucose metabolism in vivo and in vitro. Moreover, the possible mechanism from the insulin signal transduction pathway was investigated, which may provide an experimental basis for further research regarding the clinical applications of DCI.
Following adaptive feeding for 1 week, blood was drawn from the tails of the mice to determine the level of blood glucose using a blood glucose meter (Abbott, USA). Ten male db/m mice were selected to form the normal control group (NCG), while 30 db/ db mice were selected based on the levels of random blood glucose (RBG) and blood glucose after 2 h of oral glucose load. Subsequently, the 30 db/db mice were randomly divided into three groups receiving different treatments: high-dose DCI group (HDCIG), low-dose DCI group (LDCIG) and model control group (MCG). The HDCIG and LDCIG mice received DCI (purity: > 98.0%, Lot number: I0632, TCI, Japan) at a dose of 70 mg·kg -1 ·d -1 and 35 mg·kg -1 ·d -1 by gavage (gavage volume is 0.1 ml/10g), respectively. Meanwhile, the NCG and MCG mice received an equal volume of pure water. All treatments were administered at 9 a.m. every day, and the duration of the entire experiment was 6 weeks.

Measurement of Weight and Concentration of Blood Glucose
Daily water intake and feed intake were measured every three days, and the weight of each group was measured at 9:00 in every week. After 3 weeks of treatment, the concentration of glucose was measured after 10 h of fasting using the oral glucose tolerance test (OGTT). The dose of glucose used for the OGTT was 2g·kg -1 . The area under the curve (AUC) was calculated according to the level of blood glucose and the trapezoidal area summation method, wherein G0, G30, G60, and G180 are blood glucose levels at different time points (Bao et al., 2016). AUC=15 (G0+G30) + 15(G30+G60) + 30(G60+G120). After 6 weeks of administration, changes in the 24-h blood glucose under nonfasting conditions were measured at 4:00, 8:00, 12:00, 16:00, 20:00, and 24:00 (0:00).

Measurement of Serum Biochemical Indices
Twenty-four hours after the last administration, blood samples were collected from the eye socket after anesthetized with isoflurane inhalation, and the serum was separated. Insulin (INS) and levels of advanced glycation end products (AGEs) in the serum were measured using the enzyme-linked immunosorbent assay (ELISA) kit (Cat number: JYM0351Mo and JYM0171Mo, Wuhan Colorful Gene Biological Technology Co. Ltd., China) according to the instructions provided by the manufacturer. The level of glycosylated serum protein (GSP) was measured using the GSP assay kit (Cat number: A037-2, Nanjing Jiancheng Bioengineering Institute, China).

Liver Sampling and Microscopical Examination
The liver tissue was removed immediately after the mice were sacrificed, and rinsed with cool saline water. Weigh the liver and calculate the liver index (LI, LI = liver weight/body weight×100%).
In addition, the left lobe of the liver was divided into two portions and stored for Western blotting and polymerase chain reaction (PCR) analyses. Following the partial cutting of the intact liver tissue, it was soaked in 4% polyoxymethylene solution, and subsequently subjected to conventional paraffin embedding. Finally, the liver tissue was cut into 3.5-mm-thick sections. After periodic acid-Schiff (PAS) staining using a glycogen staining kit (Cat number: D004, Nanjing Jiancheng Bioengineering Institute, China), the histological structure of liver tissue was observed under a BX50 microscope (Olympus, Japan). The level of hepatic glycogen was measured using Liver/ Muscle glycogen assay kit (Cat number: A043-1-1, Nanjing Jiancheng Bioengineering Institute, China).

PCR Analysis
The mRNA expression levels of IRS2, AKT, PI3K, GLUT4, and GSK3b were determined through PCR. Total RNA was extracted from the liver of mice using Trizol (Cat number: 15596026, Thermo Fisher Scientific, USA), and the quality of the extracted RNA was confirmed using a NanoPhotometer-N50 ultramicrospectrophotometer (Implen, Germany) (Liang et al., 2018). Total RNA (5 mg) was retro-transcribed into cDNA using the RevertAid first strand cDNA synthesis kit (Lot number: 00422714, Thermo Fisher Scientific, USA). PCR was performed using 1.5 ml cDNA, 12.5 ml 2×Es Taq MasterMix (Dye), and 1.0 ml primers, in a total volume of 25 ml. PCR amplification was performed under the following conditions: initial activation of the hot-start DNA polymerase for 5 min at 95°C, followed by 34 cycles of step 2 (i.e., 95°C for 30 s, 56°C for 30 s, and 72°C for 40 s), and subsequently 72°C for 8 min before a constant temperature of 4°C. The amplified cDNA was separated through agarose gel electrophoresis, and quantification of the signals was performed by the automatic gel imaging analysis system (Cat number: 1708270, BIO-RAD, USA). The PCR primer sequences used are shown in Table 1.

Investigation of the Effect of DCI on the Viability of HepG2 Cells
DCI was diluted using Dulbecco's Modified Eagle Medium (Cat number: 31600, Solarbio, China) to a concentration of 4 mg/ml and stored at −80°C. HepG2 cells were supplied by the Shanghai Fudan University Cell Library (Shanghai, China). Subsequently, HepG2 cells were placed into 96-well plates (0.15 ml per well) and cultured for 24 h in a CB150 5% carbon dioxide culture box (BINDER, Germany) at 37°C. After discarding the culture medium, complete medium containing DCI in different concentrations (i.e., 0, 25, 50, 100, 200, 400, and 800 mg/ml) was added (each concentration was assayed in six wells). After cultivation for 24 h, 10-ml Cell Counting Kit-8 (CCK-8) was added into each well, and the culture plate was incubated in the incubator for 1 h at 37°C. The optical density (OD) value at 450 nm was determined using a multimode reader, and the OD value was proportional to the number of living cells. Cell viability in the control group was 100%, and cell viability in each group was TABLE 1 | Primers of polymerase chain reaction (PCR) analysis for genes.

Primer
Primer sequence Product length (bp) Under the aforementioned cell culture conditions, high-glucose medium (33.3 mmol/L and 55.5 mmol/L) was added for 48 h, and the supernatant was discarded. The high-glucose medium containing the corresponding DCI was added to each well for 48 h. After addition of CCK-8 (Dojindo Laboratories, Japan) for 1 h, the OD value was determined. At the same time, eight wells of normal cells containing normal culture Medium (glucose concentration: 22.2 mmol/L) were prepared. According to the results of this investigation, the optimal concentrations of DCI ( i . e . , 8 , 1 6 , a n d 32 m g / m l ) w e r e s e l e c t ed a s t h e experimental concentrations.

Glucose Oxidase Method for the Detection of Glucose Consumption in HepG2 Cells
Under the aforementioned cell culture procedures, the highglucose medium (55.5 mmol/L) was added to the 96-well plate containing HepG2 cells for 48 h. After the supernatant was discarded, high-glucose medium containing different concentrations of DCI was added to each well. There were five groups of cells in this experiment: normal cells in normal medium (glucose concentration: 22.2 mmol/L), model cells in high-glucose medium (glucose concentration: 55.5 mmol/L), and cells in high-glucose medium containing different concentrations of DCI (i.e., 32, 16, and 8 mg/ml). Each group of cells was assayed in eight wells and cultured for 48 h. Subsequently, the concentration of glucose was determined using the glucose oxidase method, which required to 5 ml of medium from each well. Lastly, CCK-8 was added to each hole for 1 h and the OD value was measured. The formula used for the calculation of glucose consumption was as follows: Glucose consumption = ðglucose concentration in the medium at the time of administration − glucose concentration in the medium 48 h after administration) :

Fluorescence Staining
HepG2 cells in the logarithmic growth phase were collected, and their concentration was adjusted to 10 4 /ml using normal medium. The cells were placed in six-well plates at 5 ml per well (5×10 4 cells). After the cells were attached, the medium was aspirated, and 5 ml of high-glucose medium (55.5 mmol/L) were added to the wells for 72 h. Subsequently, they were replaced with high-glucose medium (5 ml per well) containing DCI (32 mg/ml, 16 mg/ml) for 96 h. There were four groups in this experiment: normal cells in normal medium, model cells in high-glucose medium, and cells in high-glucose medium containing different concentrations of DCI (i.e., 32 or 16 mg/ ml). Each group was assayed in two wells. The cells were soaked in 4% paraformaldehyde at 4°C for 10 min. Subsequently, the cells were incubated with primary antibody (anti-IRS2, 1:500) at 4°C overnight. In addition, a fluorescent secondary antibody (Cat number: ab150073, abcam, USA, 1:500) was added to the cells. Then, 4′,6diamidino-2-phenylindole was added to the cells, and the nuclei were stained for 5 min (Shridas et al., 2018). Finally, the cells were observed under a microscope (OLYMPUS BX5, Japan). The negative control group received PBS instead of primary antibody.

Flow Cytometry
Under the cell culture condition and treatment administration described in Fluorescence Staining of this article, the cells were trypsinized. The trypsin reaction was terminated using PBS, and the cells were transferred to 5 ml of Eppendorf tubes. The cells were dispersed into single cells, centrifuged at 1,500 rpm for 5 min, and the supernatant was removed. After adding 2% paraformaldehyde for 10 min, the cells were centrifuged at 1,500 rpm for another 5 min and washed once with PBS. The cells of each group were incubated with primary antibody (anti-IRS2, diluted with 1% bovine serum albumin, 1:150) at 37°C for 45 min, and washed once with PBS. Subsequently, the cells of each group were incubated with a fluorescent secondary antibody (Cat number: ab150073, abcam, USA, diluted with PBS, 1:200, 37°C, 20 min), and resuspended in PBS, to reach a concentration of 10 6 /ml cells in each tube. The expression rate of positive cells was determined through flow cytometry (Becton Dickinson, USA). Cells in the negative control group were incubated with PBS instead of primary antibod.

Statistical Analysis
Statistical analysis was performed using the SPSS 19.0 software, and the experimental data were expressed as means ± standard deviation (SD) or means ± standard error of the mean (SEM) (Soll et al., 2018). Statistical significance was determined using one-way analysis of variance, followed by a least significant difference test for multiple comparisons. P < 0.05 denoted statistical significance.

In Vivo Study
Results of Changes in Daily Water Intake, Daily Feed Intake, Weight, and LI In Figure 1, the levels of daily water intake and daily feed intake of db/db mice might be higher than those in db/m mice at each point in time (P < 0.01). Compared with MCG, daily water intake in administration groups gradually decreased with unvisible difference in daily feed intake. After 6 weeks of feeding, each group of mice gained weight, and there was no significant

Effects of DCI on the Level of Blood Glucose in db/ db Mice
The levels of blood glucose in the different groups after 3 weeks of DCI administration were determined using the OGTT (Figure 2A). The AUC derived from the OGTT is shown in Figure 2B. The AUC in the MCG was significantly higher than that observed in the NCG (P < 0.01).
In addition, the AUC in the HDCIG and LDCIG was significantly decreased compared with that reported in the MCG (P < 0.01). Therefore, DCI decreased the levels of blood glucose in db/db mice. Dynamic changes in the nonfasting 24-h blood glucose in each group were determined after 6 weeks of administration ( Figure 2C). The levels of blood glucose in the NCG reached their highest value around 24:00 (P < 0.01). In the MCG, HDCIG, and LDCIG groups, the blood glucose was relatively higher around 20:00, and there was no significant difference compared with the blood glucose result at 24:00, which may be related to abnormal glucose metabolism in db/db mice. Notably, the levels of blood glucose in these groups were significantly higher than those reported in the NCG at each time point (P < 0.01). The levels in the HDCIG and LDCIG were all reduced to different degrees compared with those reported in the MCG.

Effects of DCI on the Levels of INS, GSP, and AGEs in the Serum of db/db Mice
There was no significant difference in the levels of INS between the groups (P > 0.05, Figure 3A). Of note, the levels of GSP in the NCG were lower than those observed in the other three groups (P < 0.05, Figure 3B). In addition, the highest level of GSP was noted in the MCG (P < 0.05). Moreover, the levels of AGEs in the MCG were higher than those reported in the other three groups (P < 0.05, Figure 3C).

Effects of DCI on the Synthesis of Hepatic Glycogen in db/db Mice
The levels of hepatic glycogen were determined through PAS staining, in which the glycogen and other PAS-positive components in the liver tissue stained red ( Figure 4A). The hepatocellular morphology of NCG was normal, the content of glycogen was higher, and the distribution of glycogen near the blood vessel was more. The structure of liver cells in the MCG was disordered that the cells were obviously swollen and deformed, and decreased levels of glycogen were observed (P < 0.01, Figure 4B). Compared with the MCG, the distribution of glycogen of HDCIG mice was relatively increased (P < 0.05).

Determination of the Effects of DCI on the Expression Levels of Insulin IRS2/PI3K/AKT/GSK3b/ GLUT4 Signaling Pathway-Related Proteins in Liver Tissues Through Immunohistochemistry
The results obtained for the different groups were normalized to those reported in the NCG ( Figure 5). It was shown that representative pictures of paraffin-embedded liver tissue sections processed by immunohistochemical staining regarding IRS2, PI3K, AKT, p-AKT (S473), GLUT4, and GSK3-b proteins(200×). The expression levels of IRS2, PI3K, AKT, P-AKT, and GLUT4 protein in the NCG were significantly higher than those measured in the MCG (all P < 0.01). In the groups treated with DCI, these levels were significantly higher than those observed in the MCG (all P < 0.01). The expression level of GSK3b protein in the NCG was significantly lower than that reported in the MCG (P < 0.01). Moreover, this level was significantly lower in the DCI-treated groups versus the MCG (P < 0.01).  The results obtained for the different groups were normalized to those reported in the NCG. As shown in Figure 6, the expression levels of PI3K, P-AKT, and GLUT4 proteins in the diabetic model mice were lower than those noted in the normal mice. In contrast, the expression level of GSK3b protein was higher in the liver tissue of diabetic model mice (P < 0.01). Following treatment with DCI, the expression levels of PI3K, P-AKT, and GLUT4 proteins in the liver tissue of db/db mice were upregulated, whereas the expression level of GSK3b protein was downregulated to various degrees (P < 0.05). Moreover, there was no statistical difference in the expression level of AKT protein between the groups. Determination of the Effects of DCI on the Expression Levels of Insulin IRS2/PI3K/AKT/GSK3b/ GLUT4 Signaling Pathway-Related mRNA in Liver Tissues Through PCR The internal reference was b-actin. The results obtained for the different groups were also normalized to those reported in the NCG. As shown in Figure 7, the expression levels of IRS2, PI3K, AKT, and GLUT4 mRNA in liver tissues of NCG mice were higher than those observed in the MCG, which of HDCIG and LDCIG were upregulated (P < 0.05). Moreover, the expression level of GSK3b mRNA in the liver tissue of db/db mice was higher than that reported in the NCG, and this expression level was downregulated after treatment with DCI (P < 0.05).

Effects of Different Concentrations of DCI on the Viability of HepG2 Cells
The OD values and cell viability results of HepG2 cells treated with different concentrations of DCI are shown in Figures 8A, B.
When the concentration of DCI ranged from 0 to 50 mg/ml, the cell viability increased to its highest value. Notably, the cell viability decreased as the concentration of DCI increased.

Effects of DCI on the Viability of High Glucose-Stimulating HepG2 Cells
The OD values and cell viability results of high glucosestimulating (55.5 mmol/L) and non-high glucose-stimulating (33.3 mmol/L) HepG2 cells are shown in Figures 8C, D. The cell viability of high glucose-stimulating HepG2 cells was higher than that observed in non-high glucose-stimulating cells; however, the difference was not statistically significant.

Effects of DCI on Glucose Consumption in High
Glucose-Stimulating HepG2 Cells Table 3 shows that the glucose consumption in high glucosestimulated HepG2 cells was significantly higher than that reported in non-high glucose-stimulated cells (normal cells). The glucose consumption of the group of high glucose stimulated by DCI was significantly higher than that observed in the high-glucose control group. In particular, the ratio of glucose consumption to cell proliferation activity (CCK-8) was greater in DCI intervention with a certain concentration effect.

Effects of DCI on the Expression Level of IRS2 Protein in HepG2 Cells
The expression level of IRS2 protein in HepG2 cells was detected through staining with green fluorescence ( Figure 9A). In normal control cells, the expression level of IRS2 protein was higher than that reported in high-glucose control cells. After treatment with DCI, the expression level of IRS2 protein was increased comparing with that observed in high-glucose control cells. The flow cytometry results showed a significant difference between these groups ( Figures 9B, C).

DISCUSSION
T2DM accounts for > 90% of diabetes cases, and is characterized by hyperglycemia (Chadha and Morris, 2015;Li et al., 2018). It has been reported that IR is a crucial factor in the development of T2DM, playing an important role in many diseases, such as cardiovascular disease, lipid metabolism, and obesity (Bjornshave and Hermansen, 2014). Furthermore, inhibition or attenuation of the insulin signal is the main cause of IR . The liver is one of the main targets of insulin action, and plays an important role in the development of T2DM (Cai et al., 2018). Insulin signal transduction is associated with IRS2, PI3K, AKT, GSK3b, GLUT4, and other pathway proteins (Schultze et al., 2012). Congenital and acquired factors (e.g., genetic defects, gene mutations, obesity, and the environment) affect the IRS2/PI3K/AKT/GSK3b/GLUT4 pathway, causing IR (Maull et al., 2012). The db/db mice are spontaneous T2DM mice exhibiting very similar symptoms to those observed in clinical T2DM cases (Yin et al., 2018). IR-HepG2 cells are common hepatocyte models for the investigation of the hypoglycemic effects of drugs (Chen et al., 2018). In the present study, we investigated the effect of DCI on glucose metabolism, as well as the mechanism involved in this process. Previous studies reported that, at a dose range of 35-75 mg/ kg, DCI exerts a hypoglycemic effect on diabetic mice (Fan et al.,  2018). Therefore, we selected 35 mg/kg and 70 mg/kg as the experimental doses in this study. The duration of the entire intervention was 6 weeks. Our results showed that, in db/db mice, the levels of blood glucose after glucose loading were significantly increased, accompanied by the occurrence of IR. In addition, the level of nonfasting blood glucose in db/db mice fluctuated greatly within 24 h. Notably, the level of blood glucose was higher between 20:00-24:00, whereas it was relatively low during the other time points. However, the overall level of blood glucose was high. Collectively, these results displayed the characteristics of T2DM, which were consistent with those previously reported in the literature (Xie et al., 2002;Michael et al., 2006). The hypoglycemic effect of high-and low-dose DCI on the level of fasting blood glucose was not obvious. However, all concentrations of DCI significantly improved glucose tolerance, stabilized the level of blood glucose within 24 h, and significantly lowered the level of blood glucose versus that observed in the MCG. These observations indicated that DCI exerted a stable hypoglycemic effect. In addition, DCI significantly promoted glucose consumption in HepG2 cells stimulated by high glucose, and increased the ratio of glucose consumption to cell proliferation activity (removing the interference of cell proliferation). The observed increase of glucose consumption was more pronounced in cells treated with DCI, showing a concentration-dependent relationship in the range of 8-32 mg/ml. This result further proved that DCI can promote glucose metabolism. INS, GSP, and AGEs in the serum are important indicators for assessing the body's abnormal glucose metabolism. We measured these factors to assess the prevalence of diabetes in db/db mice, and then to evaluate the hypoglycemic effect of DCI. The results showed that there was no significant difference in the level of INS between the groups, suggesting that DCI had no effect on the release of INS. However, DCI decreased the levels of GSP, AGEs in the serum of db/db mice, and the effect of highdose DCI was more pronounced. Furthermore, the level of GSP can reflect the level of blood glucose within 1-3 weeks, which is an important indicator for monitoring the blood glucose (Morris et al., 1986). Sustained high concentrations of blood glucose and various proteins in the body produce nonenzymatic glycation reactions to form AGEs (Gawandi et al., 2018). This process plays an important role in the pathogenesis of chronic complications of diabetes. This study revealed that DCI can reduce the levels of GSP, AGEs in the serum of db/db mice, which may be related to its hypoglycemic effect. Further research is warranted to identify other factors involved in the process. Hepatic glycogen is the stored form of sugar; its synthesis increases when the blood glucose level is high (Zahra et al., 2013). On the contrary, when the level of blood glucose is low, hepatic glycogen is converted into glucose to supplement blood glucose. Therefore, the synthesis and decomposition of hepatic glycogen is very important in maintaining the relative stability of blood glucose levels (Martin et al., 2004). Figure 4 showed that the content of glycogen in diabetic model mice was lower than that observed in normal mice. Moreover, the level of hepatic glycogen in mice treated with high-and low-dose DCI was increased compared with that measured in the model group mice. These findings revealed that DCI promotes the synthesis of hepatic glycogen, which may also be one of the mechanisms related to its hypoglycemic effect. In previous studies, obvious liver damage (i.e., steatosis and fibrosis) was found in db/db mice, which may explain the reduced glycogen synthesis.
The role of DCI in the body may be multifaceted, multichanneled, and multitargeted. According to the available evidence, the hypoglycemic mechanism of DCI had been preliminarily explored from the perspective of the insulin signaling pathway (Gao et al., 2016b), suggesting that DCI may upregulate the expression levels of PI3K, P-AKT, and GLUT4, and downregulate the expression of GSK3b through the PI3K/AKT signaling pathway. Our findings are basically consistent with the current findings. According to the classic insulin signal transduction pathway, the mRNA and protein expression levels of several relevant signal factors were detected using PCR and western blotting. The results showed that DCI increased the mRNA and protein expression of IRS2, PI3K, AKT, and GLUT4, upregulated the level of P-AKT protein, and downregulated the level of GSK3b protein in the liver tissues of db/db mice. Furthermore, the expression of fluorescently labeled IRS2 protein in HepG2 cells was detected through flow cytometry. The results showed that DCI upregulated the expression level of IRS2 protein. It is suggested that DCI may increase the expression level of relevant proteins of the insulin PI3K/AKT/GSK3b/GLUT4 signaling pathway to enhance its signal transduction, and improve the functional level of GLUT4   protein. Consequently, this enhances the ability to transport glucose in cells and lower the level of blood glucose.
As an important signaling factor in the insulin signaling pathway, IRS2 is involved in liver metabolism, as well as the development and survival of pancreatic parenchyma cells for hepatic metabolism and pancreatic beta-cells (Joana Moitinho et al., 2014;Saltiel, 2016). Studies have shown that IRS2 may inhibit hyperglycemia and hyperinsulinemia, while promoting endogenous glucose production and glycogen synthesis (Previs et al., 2000;Bertinat et al., 2018).
Recent studies have shown that IR is a key factor for the occurrence and development of T2DM (Nuthalapati and Indukuri, 2016), and a major feature of T2DM (Zahra et al., 2013). As the target organ of insulin action, IR in the liver is a main component of the mechanism of T2DM (Polyzos et al., 2012). The mechanism of IR mainly involves IRS2, PI3K, GLUT4, and other pathway proteins. Therefore, in this study, the effect of DCI on the IR in the liver of db/db diabetic mice and HepG2 cells was investigated through the insulin IRS2/PI3K/GLUT4 signaling pathway. Furthermore, the mechanism of IR reduction by DCI was investigated at the cellular and molecular levels, providing the basis for the development into hypoglycemic drugs and clinical application of DCI.

CONCLUSIONS
Overall, DCI may significantly decrease and stabilize the level of blood glucose, improve glucose tolerance, and reduce IR in spontaneous type 2 diabetic db/db mice. The hypoglycemic mechanism of DCI involves the following: increase in the mRNA and protein expression levels of IRS2, PI3K, AKT, and GLUT4 signal factors; promotion of phosphorylation of AKT proteins, downregulation of the expression level of GSK3b, increase of glycogen synthesis, and promotion of the signal transduction of insulin PI3K/AKT/GSK3b/GLUT4 pathway. However, the action mechanism of DCI may involve multiple targets and pathways. Therefore, further research is warranted to elucidate this mechanism.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservations, to any qualified researcher.

FUNDING
This work was supported by the Natural Science Foundation of Hebei Province (grant number: H2014209194).