MATERIALS AND METHODS

Chemicals and reagents: Berberine (BBR), streptozotocin (STZ) and sodium citrate were purchased from Sigma-Aldrich Co. (St. Louis, MO, U.S.A.); Y53 was chemically synthesized in our laboratory as described previously. Superoxide dismutase (SOD) Activity Colorimetric Assay Kit and the Lipid Peroxidation Malondialdehyde (MDA) Assay Kit were from the Beyotime Institute of Biotechnology (Shanghai, China). Pierce™ Bicinchoninic Acid (BCA) Protein Assay Kit was from Thermo Fisher Scientific Inc. (Waltham, MA, U.S.A.). Commercially available kits for determination of serum cholesterol (CHO), LDL-cholesterol (LDL-c), triglyceride (TG), glucose, insulin, alanine aminotranferease (ALT) and aspartate aminotransferase (AST) were purchased from Nanjing Jiancheng Bioengineering institute (Nanjing, China). Isopropanol was from Beijing Baishun Chemical Technology Co., Ltd. (Beijing, China). Tissue Total Cholesterol Assay Kit and Tissue Triglyceride Assay Kit were from the Applygen Technologies Inc. (Beijing, China). SV Total RNA Isolation System, GoScript™ Reverse Transcription System and GoTaq® qPCR Master Mix were from Promega (Madison, WI, U.S.A.).

Animal experiment: The protocol of the animal experiment was reviewed and approved by the Research Committee of the Institute of Medicinal Biotechnology, Chinese Academy of Medical Sciences (CAMS); animals were cared for according to the institutional guidelines of CAMS.

Male C57BL/6J mice (20.0±1.50 g) were purchased from the Vital River Laboratories (Beijing, China) and housed in a room at 20-24°C with 50-60% humidity and 12 h light/dark cycle. After 3 days of accommodation period, some mice were fed with regular rodent diet as control group (n = 8), other mice were fed with a High-Fat Diet (HFD) (D12492, Research Diets, Inc., New Brunswick, NJ, U.S.A.) for 4 weeks.

Then after 12 h fasting, the HFD-fed mice were intraperitoneally injected with single dose of STZ (120 mg kg^–1) which was dissolved in 0.1 mol L^–1 of cold sodium citrate (pH = 4.5) to induce diabetes. Seventy two hours later, about one third of the mice died; the rest of the mice were subjected to tail snipping after fasting, blood glucose levels were determined by a glucometer (Roche Applied Science, Indianapolis, IN, U.S.A.). The mice were considered as diabetic if their fasting blood glucose ≥ 11.1 mmol L^–1.

The diabetic mice were randomly divided into 4 groups with 10 mice each, which received saline (DM group), 50 mg kg^–1 of BBR (DM+BBR50 group), 50 mg kg^–1 of Y53 (DM+Y53-50 group) and 100 mg kg^–1 of BBR (DM+BBR100 group), respectively. The BBR and Y53 were suspended in saline and orally administered to the mice at 2 pm every day for 4 weeks. Body weights and food intakes of the mice were recorded every other day.

At the end of the experiment, mice were fasted for 12 h, blood samples were collected by retro-orbital puncture; serums were isolated for the measurement of CHO, LDL-c, TG, glucose, insulin, ALT and AST levels by commercially available kits. The mice were sacrificed by cervical dislocation; their livers and pancreases were harvested and weighed, liver index was calculated as liver weight/body weight (after experiment)×100. A portion of every liver tissue was fixed in 10% formaldehyde for hematoxylin and eosin (H and E) staining; the remaining liver and pancreas tissues were immediately frozen in liquid nitrogen.

Semiquantitative histopathological evaluation of liver sections: Liver sections were subjected to pathological scoring for steatosis (0~3) and lobular inflammation (0~3) as described in another report (Tang et al., 2015).

Determination of liver CHO and TG: About 0.1 g of liver tissue from each sample was homogenized in 1 mL of isopropanol for total lipid extraction as described before (Kong et al., 2008); CHO and TG contents were determined with commercially available kits and normalized to the weight of liver tissue.

Measurement of SOD activities and MDA levels: At the end of the experiment, SOD activities and MDA levels of serums, livers and pancreases were determined according to the suppliers’ protocols after homogenization (for livers and pancreases). The SOD activities and MDA levels were normalized to protein concentrations in livers and pancreases.

RNA extraction and Real-Time reverse transcriptase-polymerase Chain Reaction (RT-PCR): Total RNAs were isolated from livers and pancreases and reversely transcribed into cDNAs as described previously (Kong et al., 2008). Real-time PCR was performed with gene specific primers (Table 1, Thermo Fisher Scientific Inc.) with glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as an internal control. The reactions were performed in an ABI Prism® 7900 High-Throughput Real-Time PCR System (Applied Biosystems, Foster City, CA, U.S.A.) with a condition same as our previous report (Kong et al., 2008). The comparative threshold cycle (C_T) method was used for relative quantification of target gene expression, which was plotted as fold of control.

Statistical analysis: Values are Mean±S.D. of 8-10 animals in each group. After validation of the test for homogeneity of variance, differences among study groups were examined by one-way ANOVA followed by the Newman-Keuls test for multiple comparisons. p<0.05 was considered to be statistically significant.

RESULTS

The Y53 improves metabolic disorders more effectively than BBR in diabetic C57BL/6J mice with liver steatosis. Diabetes Mellitus (DM) of the C57BL/6J mice was induced by HFD-feeding and STZ injection. The blood glucose levels were almost the same among studying groups of diabetic mice before drug administration.

As shown in Table 2 and 3, compared to those of the control group, serum glucose increased for about 4.09-fold (p<0.001) while serum insulin declined by about 78.8% (p<0.001) and pancreas weight declined by about 25.6% (p<0.05) in the DM group of mice. After 4 weeks of treatment, BBR at 50 mg kg^–1 could reduce serum glucose and restore serum insulin level to some extents (p<0.05 vs. DM group). Notably, Y53 at the same dose manifested stronger efficacies in modulating serum glucose/insulin and restoring pancreas weight (p<0.05 vs. DM+BBR50 group). Four weeks’ administration of Y53 at 50 mg kg^–1 reduced serum glucose by about 52.8% (p<0.01 vs. DM group), increased serum insulin by about 2.97-fold (p<0.01 vs. DM group) and restored pancreas weight to nearly normal (p<0.05 vs. DM group). The efficacies of Y53 were nearly the same as those of BBR at double dose (Table 2 and 3).

The HFD-fed mice ate less food than the control mice which were fed with a regular rodent diet (Table 3), perhaps due to the unpleasant smell and taste of the HFD.

Table 1:	Primers for real-time PCR (5’ to 3’)

Fig. 2:	Effects of Y53 and BBR on hepatic LDLR and InsR mRNA levels. Values are Mean±SD of 8 to 10 animals in each group. *p<0.05 vs. that of control group, #p<0.05, ##p<0.01 vs. that of DM group, $p<0.05 vs. that of DM+BBR50 group

Table 2:	Effects of Y53 and BBR on serum biochemical parameters and liver lipid profiles in diabetic C57BL/6J mice with liver steatosis
Values are Mean±SD of 8 to 10 animals in each group. **p<0.01, ***p<0.001 vs. that of control group, #p<0.05, ##p<0.01 vs. that of DM group, $p<0.05 vs. that of DM+BBR50 group, DM: Diabetes mellitus, BBR: Berberine, Y53: Pseudoberberine, LDLC: Low-density lipoprotein-cholesterol, CHO: Cholesterol, TG: Triglyceride, AST: Aspartate aminotransferase, ALT: Alanine aminotranferease

Table 3:	Effects of Y53 and BBR on food intake, body and organ weights in diabetic C57BL/6J mice with liver steatosis
Values are Mean±SD of 8 to 10 animals in each group. *p<0.05, **p<0.01 vs. that of control group, #p<0.05 vs. that of DM group, $p<0.05 vs. that of DM+BBR50 group, DM: Diabetes mellitus, BBR: Berberine, Y53: Pseudoberberine, STZ: Streptozotocin

The body weights of the mice were not statistically different among groups before the experiment or before STZ injection. However, at the end of the experiment, due to STZ injection, the body weights of the mice with DM significantly declined compared to control mice (p<0.05). The BBR and Y53 had no influence on the body weight of the mice in this experiment (Table 3).

The HFD-feeding resulted in hyperlipidemia in diabetic mice (Table 2), Y53 lowered serum LDL-c, CHO and TG more effectively than BBR at the same dose (p<0.05), which was in agree with our previous results (Li et al., 2009). We also measured the mRNA expression levels of LDLR and Insulin Receptor (InsR) in the liver, which were proved to be upregulated by BBR in our previous studies (Kong et al., 2004, 2009; Li et al., 2009). As shown in Fig. 2, LDLR and InsR mRNA levels declined averagely by about 30% in diabetic mice fed with a HFD (p<0.05 vs. control), which might reflect compromises of the lipid and glucose metabolism in the animals. Y53 at 50 mg kg^–1 increased hepatic LDLR and InsR mRNA levels by 1.79 and 1.87-fold (p<0.01 vs. DM group), respectively, which were close to those of BBR at 100 mg kg^–1.

The mice developed severe fatty liver after HFD-feeding. As shown in Fig. 3, H and E staining of the liver sections illustrated macro- and microvesicular steatosis as well as inflammatory cell infiltration in diabetic mice.

Fig. 3(a-b):

Effects of Y53 and BBR on the pathological changes in the livers of C57BL/6J mice. Livers were harvested and fixed in 10% formaldehyde for H and E staining, (a) Representative images of each group (×400) and (b) Liver sections were subjected to pathological scoring for steatosis and lobular inflammation. Values are Mean±S.D. of 8 to 10 animals in each group. ***p<0.001 vs. that of control group, #p<0.05, ###p<0.001 vs. that of DM group, $$p<0.01 vs. that of DM+BBR50 group

Their liver weight and index (Table 3), liver TG and CHO contents (Table 2) increased greatly as well (p<0.05 or p<0.01 vs. control). As a result, serum ALT level of the diabetic mice elevated dramatically (p<0.001 vs. control). Four weeks’ administration of Y53 significantly ameliorated liver steatosis (Fig. 3a-b), reduced liver TG and CHO contents (Table 2), decreased liver weight and index (Table 3) and restored liver function (Table 2) in diabetic mice. The efficacies of Y53 were close to those of BBR at 100 mg kg^–1 and superior to BBR at 50 mg kg^–1 (p<0.05 or p<0.01). The Y53 ameliorates oxidative stress and proinflammatory response more potently than BBR in diabetic C57BL/6J mice with liver steatosis.

Next, the antioxidant and anti-inflammatory activities of Y53 were determined in the animals. The mice developed obvious oxidative stress after the onset of DM. As shown in Table 4, the SOD activities declined while the MDA levels increased greatly in the serum, liver and pancreas of the diabetic mice (p<0.01 vs. control). Four weeks’ administration of Y53 at 50 mg kg^–1 restored SOD activities and reduced MDA contents in the serum, liver and pancreas to near baseline levels (p<0.01 vs. DM group).

Fig. 4(a-b):

Effects of Y53 and BBR on the mRNA expression levels of proinflammatory cytokines and key genes of the Nrf2 antioxidant pathway. Livers and pancreases were harvested for, (a) Real-time RT-PCR analysis of the mRNA levels of IL-6/TNF-α and (b) Nrf2, HO-1 and NQO-1. The expression levels of target genes were normalized to that of GAPDH and plotted as fold of control, which was designated as 1. Values are Mean±SD of 8 to 10 animals in each group. *p<0.05, **p<0.01, ***p<0.001 vs. that of control group, #p<0.05, ##p<0.01 vs. that of DM group, $p<0.05 vs. that of DM+BBR50 group

Table 4:	Effects of Y53 and BBR on SOD activity and MDA level in the serum, liver and pancreas of diabetic C57BL/6J mice with liver steatosis
Values are Mean± SD of 8 to 10 animals in each group, **p<0.01 vs. that of control group, #p<0.05, ##p<0.01 vs. that of DM group, $p<0.05 vs. that of DM+BBR50 group, DM: Diabetes mellitus, BBR: Berberine, Y53: Pseudoberberine, SOD: Superoxide dismutase, MDA: Malondialdehyde

The antioxidant activity of Y53 was similar to that of BBR at 100 mg kg^–1 and superior to BBR at 50 mg kg^–1 (p<0.05).

The mRNA expression levels of proinflammatory cytokines like interleukin-6 (IL-6) and Tumor Necrosis Factor-α (TNF-α) elevated significantly in the liver and pancreas of the diabetic mice (Fig. 4a) (p<0.01 or p<0.001 vs. control). The Y53 at 50 mg kg^–1 largely suppressed the over-expression of IL-6 and TNF-α in the liver and pancreas (Fig. 4a) and inhibited inflammatory cell infiltration in the liver (Fig. 3a-b).

As BBR was reported to upregulate the expression of nuclear factor erythroid-2-related factor-2 (Nrf2) and stimulate the Nrf2/heme oxygenase-1 (HO-1) antioxidant pathway in fatty liver and DM (Li et al., 2014a; Yuan et al., 2015), we examined the efficacies of Y53 on this pathway in the present study. As shown in Fig. 4b, the expression levels of Nrf2 and its target genes like HO-1 and NADPH quinine oxidoreductase-1 (NQO-1) increased in the DM group of mice (p<0.05 vs. control). Similar results which might due to an adaptive response were observed in another report (Wang et al., 2011). When the diabetic mice were treated with 50 mg kg^–1 of Y53, the levels of Nrf2, HO-1 and NQO-1 mRNAs elevated by about 1.94 to 2.32-fold in the liver and pancreas (p<0.01 vs. DM group). The activities of Y53 in reducing proinflammatory response and upregulating Nrf2/HO-1/NQO-1 were similar to those of BBR at 100 mg kg^–1 and superior to BBR at 50 mg kg^–1 (p<0.05 or p<0.01).

DISCUSSION

Natural product BBR is an ancient drug but may become a promising candidate for the development of novel oral hypoglycemic agents in the near future. Here we study the efficacies of compound Y53, a BBR analogue, in diabetic mice with liver steatosis.

Our results proved that Y53 lowered serum lipids and glucose stronger than BBR at the same dose, which were in agreement with our previous reports (Li et al., 2009; Shan et al., 2013). We also found that in the mice, Y53 could upregulate hepatic expression of LDLR and InsR more potently than BBR. Berberine modulated LDLR and InsR mRNA expression through different mechanisms. In detail, LDLR mRNA was upregulated by BBR through the AMP-activated protein kinase (AMPK)/Raf-1/mitogen-activated protein kinase kinase (MEK)/extracellular signal-regulated kinase (ERK) pathway (Kong et al., 2004; Li et al., 2014b) while InsR expression was increased by BBR in a Protein Kinase D (PKD)-dependent but ERK-independent manner (Kong et al., 2009; Zhang et al., 2010).

Interestingly, in the in vitro experiments, the LDLR-upregulating activity of Y53 was superior to that of BBR (Li et al., 2009) but its activity on InsR expression was equal to that of BBR (Shan et al., 2013), which might suggest that Y53 itself had stronger bioactivity in modulating LDLR but not InsR. Our previous results also demonstrated that the pharmacokinetic profiles of Y53 such as the maximum plasma concentration (C_max) and area under concentration-time curve (AUC_0–24) were superior to those BBR, which might be attributed to the avoidance of P-glycoprotein (P-gp)-mediated cellular efflux of Y53 (Shan et al., 2013). The improved pharmacokinetic profiles of Y53 could explain its enhanced activity on InsR expression in vivo, which was crucial for insulin signaling and glucose homeostasis (Kong et al., 2009).

The stimulating activity of Y53 on the AMPK pathway was similar to that of BBR in cultured cells (Shan et al., 2013). However, Y53 promoted LDLR mRNA expression more effectively than BBR (Li et al., 2009), which implied that some unknown mechanisms might be involved in the activity of Y53 in modulating LDLR. In liver cells, in addition to the ERK pathway (Kong et al., 2004), BBR could upregulate LDLR expression through proprotein convertase subtilis in/kexin type 9 (PCSK9) down regulation (Dong et al., 2015). It is of scientific importance to study the efficacy of Y53 on PCSK9, a post-translational inhibitor of LDLR (Dong et al., 2015), to clarify its improved activity on LDLR expression.

Due to the improved pharmacokinetic profiles, it could be inferred that Y53 might stimulate the AMPK pathway more potently than BBR in vivo. The AMPK is a key molecule that controls energy balance and lipid/glucose metabolism in organisms (Coughlan et al., 2014). The improved efficacies of Y53 on TG metabolism and liver steatosis in the current study were in agreement with our inference.

One of the new findings of this study was that Y53 had improved activities in suppressing oxidative stress and proinflammatory response than BBR in various tissues in diabetic mice. Oxidative stress and proinflammatory response were critical for the onset of DM as excessive Reactive Oxygen Species (ROS) and proinflammatory cytokines might damage the insulin signaling pathway and have negative influence on islet function (Evans et al., 2005; Donath and Shoelson, 2011). Our results indicated that Y53 significantly restored pancreas weight and serum insulin level in diabetic mice. The detailed mechanisms for Y53 to protect the islet function in DM merit further investigation in the future.

In addition to the antioxidant efficacy, the Nrf2 pathway also had anti-inflammatory and cytoprotection activities in tissues (Jimenez-Osorio et al., 2015; Li et al., 2014a). Now Nrf2 is considered as a useful target to treat DM and its complications (Jimenez-Osorio et al., 2015; Li et al., 2014a). Results showed that Y53 had strong activities in upregulating Nrf2 and its target genes, which were in agreement with its potent antioxidant and anti-inflammatory activities.

CONCLUSION

Results of the present study indicate that compound Y53, a BBR analogue, ameliorates oxidative stress and proinflammatory response and improves metabolic disorders more potently than BBR when administered at the same dose in a mice model of DM and fatty liver. Further research work should focus on the activities of Y53 on the islet function as well as complications in DM. The Y53 may be developed as a useful new anti-diabetic agent in the future.

ACKNOWLEDGMENTS

The study was supported by the National Natural Science Foundation of China (81321004) and National Mega-Project for Innovation Drugs (2014ZX09101005-008).