Anti-diabetic effect of the polyphenol-rich extract from Tadehagi triquetrum in diabetic mice

Purpose: To clarify the diabetes-reducing abilities of the polyphenol-rich extract from Tadehagi triquetrum (HC) in diabetic ob/ob mice. Methods: Aerial parts of T. triquetrum were extracted under reflux and partitioned by n-butanol to generate HC. The effects of HC consumption on blood glucose and lipids, insulin resistance, and liver glucose metabolism were evaluated in vivo. The main compounds of HC were tested for their effects on stimulating glucose consumption and uptake by HepG2 hepatocytes and C2C12 myotubes. Results: After HC treatment, body fat, subcutaneous fat, and epidydimal fat masses decreased (p < 0.05), while mean daily food intake was unaffected. HC (200–400 mg/kg) decreased fasting blood glucose, glycosylated serum protein (GSP), and glycosylated hemoglobin (HbAlc); it also lowered hyperinsulinemia, improved oral glucose tolerance, and reduced hyperlipidemia and liver fat content (p < 0.05). HC treatment markedly elevated liver glycogen content and activity of hepatic glucokinase and pyruvate kinase (p < 0.05). Eight polyphenols were isolated from HC, six of which potently stimulated glucose consumption and uptake in vivo. Conclusion: HC has potent antidiabetic activities. Polyphenols are the main compounds accounting for these effects. Chronic oral administration of HC may be an alternative therapy for managing diabetes, but this has to be subjected first to clinical studies.


INTRODUCTION
Type 2 diabetes (T2DM) is prevailed throughout the world. Hyperglycaemia and insulin resistance (IR) are main characteristics of T2DM [1]. Although multiple antidiabetic medicines are currently available, many are associated with undesirable side effects [2,3]. Consequently, there is a tremendous need for developing effective antidiabetic drugs with fewer side effects. The flowering plant Tadehagi triquetrum is a herbal remedy used for inflammation and liver disease [4,5]. In South China, people usually boil the aerial parts of T. triquetrum in water and drink it as a tea called "Hulu tea." Previous chemical studies identified phenolic compounds and terpenoids as the major constituents of T. triquetrum. Recent investigations have revealed that phenolic compounds, especially phenylpropanoid glucosides, in this medicinal plant exhibit potent lipid-and glucose-decreasing effects in vitro [6][7][8]. However, the antidiabetic effects of T. triquetrum aerial parts have not been heretofore tested in vivo. In this study, the diabetesreducing effects of the polyphenol-rich fraction from T. triquetrum aerial parts (HC) were tested in diabetic ob/ob mice. The antidiabetic ability of HC was evaluated. The main active compounds of HC were also examined to determine their role in the glucose-lowering abilities.

Plant materials
T. triquetrum was collected from Changliu in Haikou, Hainan Province, in July 2018, and authenticated by professor Nian-Kai Zeng. After identification, a sample (no. HC-07-2018) was deposited in herbarium at Hainan Medical University.

Extraction and isolation
Dried stems and leaves of T. triquetrum were extracted using 70:30 ethanol:distilled water as a solvent. The resulting crude extract was dissolved in pure water and partitioned with dichloromethane to eliminate lipophilic compounds. The residue was further extracted using n-butanol to obtain HC, after concentration. HC (150 g) was isolated using a column of silica gel with a gradient solvent of chloroformmethanol to produce 6 fractions (Fractions A-F). Fraction B was processed in an additional step using the gel column, and methanol was used as the eluent to purify the compounds. The fractions obtained in this manner were further separated by high-performance liquid chromatography (HPLC) using a mixed solvent (methanol:water 35:65, v/v) and a semipreparative HPLC column, which yielded compound 6 (71.0 mg). Fraction C was isolated using the same process as for Fraction B and then purified using HPLC (with methanol:water 35:65, v/v) to generate compounds 7 (124.5 mg) and 8 (255.4 mg). Fraction D was prepared in the same manner using a gel column and HPLC (with methanol:water 42:58 v/v), generating compounds 3 (34.0 mg), 4 (124.0 mg), and 5 (55.0 mg). Fraction E was isolated and separated in the same way as Fraction D and purified using HPLC (with methanol:water 30:70, v/v) as above, yielding compounds 1 (20.5 mg) and 2 (30.0 mg).

Animals and experimental design
All animal experiments were performed in accordance with the animal ethics committee of Hainan Medical University (reg. no. 201706017/HMU). A total of 36 male 12-weekold mice, including 30 ob/ob mice and 6 C57BL/6J mice, were purchased from a qualified animal feeding company. The mice were maintained under humidity-controlled conditions with a 12-hour/12-hour light/dark cycle. They were feed ad libitum for 7 days and then randomly separated into six groups. Mice in the normal control (C57BL/6J mice) and negative control (vehicle only, ob/ob mice) were administered equivalent volumes of distilled water orally. Metformin (200 mg/kg) was applied as the positive and given daily via oral gavage, and the HC groups were administered HC daily (100, 200, or 400 mg/kg) via the same route.
The body mass of each mouse was measured weekly using a balance. Feed intake of each mouse was also recorded. After 4 weeks, all mice were fasted overnight. The next morning, their body mass was measured, and the animals were then anesthetized using intraperitoneal chloral hydrate (400 mg/kg). Approximately 0.5 mL whole blood was then withdrawn from the abdominal aorta. The animals were subsequently euthanized, after which we measured their liver, subcutaneous fat, and epidydimal fat masses. Biological indicators (except blood insulin) were evaluated using commercial kits, following the manufacturer's recommendations. Serum insulin levels were quantified using a radioimmunoassay kit.

Oral glucose tolerance test
On the twenty-four days after initiating study, each mouse underwent an oral glucose tolerance (OGTT) using oral gavage of glucose (2 g/kg), as described previously [9]. Blood was obtained from tail-vein at 0, 0.5, 1.0, 1.5, and 2.0 hours after glucose administration. The glucose content was checked immediately after withdrawal using a Roche glucose meter (ACCU-CHEK Active, Roche).

Cell culture, and glucose consumption and uptake assay
The C2C12 myotube and HepG2 hepatocyte cell lines were used for in vitro assessment of the individual HC compounds. The origin, culturing, and handling of these cell lines have been previously described [9,10]]. Briefly, both cells were cultured in DMEM containing 10% fetal bovine serum at a certain temperature and a fixed concentration of carbon dioxide as previous reports [9,10]. To differentiate cells, the culture medium was changed into DMEM possessing 2% horse serum. After incubation for 12 h, cells were cultured with 10 μM of each compound.
Glucose consumption and uptake assay were performed as described previously [10]. Glucose consumption was investigated on HepG2 cells. The cells were cultured with DMEM possessing HC compounds. After incubation for 24 h, the glucose content in the medium was tested according to glucose kit. Glucose uptake assay was conducted on C2C12 myotubes. Cells were cultured with the medium containing 2-Deoxy-2-[(7-nitro-2,1,3-benzoxadiazol-4-yl)amino]-Dglucose (2-NBDG) and HC compounds or insulin (100 nM). After 12 h, the medium was moved away and C2C12 myotubes were cleaned by phosphate-buffered saline (PBS) for two times. After this step, the C2C12 myotubes were processed by scrapping out and then transferring. The contents of 2-NBDG uptake by C2C12 myotubes were tested according to reported method.

Statistical analysis
Data are displayed as mean ± standard error of the mean. Groups were compared using oneway analysis of variance followed by Dunnett's ttests. Statistical analysis was performed using SPSS 22.0 software.

HC reduced mouse body mass
At 4 weeks, all groups of ob/ob mice exhibited a considerably higher body mass than normal control mice (Figure 1 a). Body mass gain from baseline to 4 weeks was also considerably more in ob/ob mice than in normal mice (Figure 1 b). There was a dose-dependent reduction in body mass gain in the HC-treated mice (Figure 1 b). Liver, subcutaneous and epidydimal fat masses were apparently reduced with 400 mg/kg HC (Figure 1 d -f). Feed intake showed no significantly different between any of the groups (Figure 1 c).

HC decreased blood glucose
At 4 weeks, animals exhibited higher fasting blood glucose (FBG) (Figure 2 a), greater increase in FBG (Figure 2 b), HbAlc (Figure 2 c), and GSP (Figure 2 d), compared with normal mice. HC treatment was associated with reductions in FBG, HbAlc, and GSP (Figure 2 ad). The effectiveness of HC (400 mg/kg) for decreasing blood glucose was similar to that observed with metformin. Serum aspartate aminotransferase (AST) and alanine aminotransferase (ALT) displayed unimportant differences between any of the groups of ob/ob mice, suggesting that the beneficial effects of HC were not associated with liver toxicity (Figure 3).

HC ameliorated hyperlipidemia
The lipid-lowering effects of HC were tested by measuring triglyceride (TG), total cholesterol (TC), low-density lipoprotein cholesterol (LDL-c), and high-density lipoprotein cholesterol (HDL-c). At 4 weeks, serum TG, TC, and LDL-c and liver TC and TG were higher in ob/ob mice than in normal mice ( Figure 5). Treatment with HC (200 mg/kg and 400 mg/kg) reduced serum TG, TC, and LDL-c, as well as liver TC and TG ( Figure 5). HC (400 mg/kg) appeared to be slightly more effective that in metformin for lowering serum and liver lipids. Serum HDL-c showed no significantly different between any of the groups (Figure 5 d).

HC enhanced hepatic glycolysis and stimulated glycogen production
Liver abnormal glucose often leads to raised blood glucose levels [11]. Compared with C57BL/6J animals, ob/ob mice had significantly lower hepatic pyruvate kinase and glucokinase activity (Figures 6 b and c), although liver glycogen content did not significantly differ between ob/ob and normal mice (Figure 6 a). ). Liver glycogen, as well as pyruvate kinase and glucokinase activity, increased with HC treatment ( Figure 6).

DISCUSSION
T2DM often require long-term medications to control blood glucose levels. Dietary interventions are a simple, risk-free option for diabetes management. Dietary interventions have gained increasingly more attention from both researchers and patients. In the current study, the polyphenol-rich extract HC possessed potent diabetes-reducing effects in ob/ob mice. Four-week treatment with HC decreased FBG and prevented blood glucose increases dosedependently. The efficacy of HC (400 mg/kg) for decreasing blood glucose was similar with that of metformin (200 mg/kg). Serum GSP and HbAlc levels, which are indicators of diabetes in the previous 1-3 weeks or 2 months, respectively, were also significantly decreased with HC, indicating that treatment with HC could maintain blood glucose levels over a prolonged period of time.
Polyphenol-rich extracts from various medicinal plants have documented beneficial effects in treating T2DM [13,14]. This area has attracted much research interest from both pharmacologists and chemists for many years and has been the subject of excellent reviews [15][16][17]. The hypoglycemic effects of phenolic compounds in HC have been investigated. Many of these have unique structures and obvious biological activities and some have shown significant effects, with potentially important roles in the treatment of diabetes [18,19].
Previously, phenylpropanoid glucosides have been extracted from the whole plants of T. triquetrum, some of which have been demonstrated to improve glucose consumption by HepG2 and C2C12 cells [7,8]. Polyphenols, and especially phenylpropanoid glucosides, may be the principal active components of HC for treating diabetes. To assess this, the glucoselowering activity of the eight main compounds were assessed. Of these, six compounds (3)(4)(5)(6)(7)(8), particularly compounds 5 and 6, exhibited significant ability in glucose consumption and uptake in vitro. These data suggest that these polyphenols have glucose-lowering effects and may be the agents responsible for the antihyperglycemic activity of HC. Thus, phenylpropanoid glucosides isolated from HC could be natural compounds leading to the development of new agents for treating or curing T2DM.

CONCLUSION
HC, the n-butanol extract of T. triquetrum aerial parts, exhibits potent antidiabetic activity in diabetic mice. Polyphenols are the main bioactive compounds accounting for these antidiabetic effects. Chronic oral administration of HC might be an alternative therapy for management diabetes.