Vitisin B, a resveratrol tetramer from Vitis thunbergii var. taiwaniana, ameliorates impaired glucose regulations in nicotinamide/streptozotocin-induced type 2 diabetic mice

Abstract

Background and aim

In Taiwan, Vitis thunbergii var. taiwaniana (VTT) is used in traditional medicine and as a local tea. VTT rich in resveratrol and resveratrol oligomers have been reported to exhibit anti-obesity and anti-hypertensive activities in animal models; however, no studies have investigated type 2 diabetes mellitus (T2DM) treatments. This study aimed to investigate the anti-T2DM effects of resveratrol tetramers isolated from the VTT in nicotinamide/streptozotocin (STZ)-induced Institute of Cancer Research (ICR) mice.

Experimental procedure

The oral glucose tolerance test (OGTT) was used to imitate postprandial blood glucose (BG) regulations in mice by pre-treatment with VTT extracts, resveratrol tetramers of vitisin A, vitisin B, and hopeaphenol 30 min before glucose loads. Vitisin B (50 mg/kg) was administered to treat T2DM-ICR mice once daily for 28 days to investigate its hypoglycemic activity.

Results and conclusion

Mice pre-treated with VTT-S-95EE, or vitisin B (100 mg/kg) 30-min before glucose loading showed significant reductions (P < 0.001) in the area under the curve at 120-min (BG-AUC_0-120) than those without pre-treatment with VTT-S-95 E E or vitisin B. Vitisin B-treated T2DM mice showed hypoglycemic activities via a reduction in plasma dipeptidyl peptidase (DPP)-IV activities to maintain insulin actions and differed significantly than those of untreated T2DM mice (P < 0.05), and also reduced BG-AUC_0-120 and insulin-AUC_0-120 in the OGTT.

These in vivo results showed that VTT containing vitisin B would be beneficial for developing nutraceuticals and/or functional foods for glycemic control in patients with T2DM, which should be investigated further.

Graphical abstract

Highlights

• •OGTT was used in the healthy ICR mice to imitate postprandial BG regulations by VTT extracts.
• •The nicotinamide/STZ-induced T2DM mice were used to evaluate the impaired glucose regulations.
• •The 28-day treatments of vitisin B improved fasting BG and postprandial BG in OGTT of T2DM mice.

Abbreviations

ALT

alanine transaminase

AST

aspartate transaminase

AUC

area under curve

BG

blood glucose

BG-AUC_0-120

BG in area under curve during 120-min

BMI

body mass index

BUN

bilirubin

CREA

creatinine

DM

diabetes mellitus

DMSO

dimethyl sulfoxide

DPP-IV

dipeptidyl peptidase-IV

GIP

glucose-dependent insulinotropic polypeptide

GLP-1

glucagon-like peptide-1

HDL

high-density lipoprotein cholesterol

LDL

low-density lipoprotein cholesterol

OGTT

oral glucose tolerance test

STZ

streptozotocin

T-CHO

total cholesterol

T2DM

type 2 diabetes mellitus

VTT

Vitis thunbergii var. taiwaniana

VTT-S-95 E E

95% ethanol extracts of VTT stem parts

1. Introduction

Diabetes Atlas 2021, 10th edition has been published by the International Diabetes Federation.¹ They estimated that the number of global diabetic adults aged 20–79 years was approximately 537 million in 2021, accounting for one-tenth of the global population, and will increase to 643 million by 2030 and 783 million by 2045.¹ The WHO released the top ten global causes of death in 2019 and diabetes mellitus (DM) occupied the ninth position.² In 2019 and 2020, DM was the seventh and the eighth cause of death, respectively, according to the statistical data reported by The National Center for Health Statistics of CDC in the United States.³ The prevalence of DM, especially type 2 (T2)DM, is increasing globally due to sedentary lifestyles, fatty and unhealthy diets, alcoholic drinks, obesity, and population growth.^4,5

Insulin regulates glucose metabolism to reduce blood glucose (BG) levels rapidly after consumption of starchy foods. Incretin hormones, such as glucagon-like peptide-1 (GLP-1) and/or glucose-dependent insulinotropic polypeptide (GIP) increase insulin secretion.⁶ Dipeptidyl peptidase-IV (DPP-IV) hydrolyzes and inactivates these incretin hormones to bind to the receptor and stop insulin secretions.⁶ Therefore, prolonging the action of GLP-1 and/or GIP by inhibiting DPP-IV activities is one of the promising methods for stimulating insulin secretion in T2DM patients with impaired glucose regulation and reducing hyperglycemia. The chemically developed DPP-IV inhibitors, including sitagliptin, saxagliptin, anagliptin, linagliptin, gemigliptin, vildagliptin, and teneligliptin are commercially available.⁷ Several studies have focused on potential DPP-IV inhibitory compounds from natural products or protein hydrolysates for their potential applications in impaired glucose controls, These compounds include steroidal saponin and diosgenin from Trillium govanianum Wall. ex D. Don,⁸ gallic acid and ellagic acid from Terminalia arjuna (Roxb. ex DC.) Wight & Arn.,⁹ quercetin, diosmetin, methyl rosmarinate, kaempferol, and acacetin from Lobelia chinensis Lour.,¹⁰ yam (Dioscorea alata L. cv. Tainong No. 1) tuber protein hydrolysates and their synthetic peptides,^11,12 and peptides LLAP, ILAP, and MAGVDHI were identified from the proteinase hydrolysates of the macroalga Palmaria palmata (Linnaeus) F. Weber et D. Mohr.¹³

The Vitis thunbergii var. Taiwaniana (VTT) is steroidal saponin and an endemic variety of V. thunbergii Sieb. & Zucc. in Taiwan has smaller fruits than grapes (V. vinifera). The dried fragments of whole VTT are used as tea-substituted materials and recognized as folk medicines for treating hepatitis, jaundice, diarrhea, and arthritis. One class of the main active components is resveratrol and its oligomers.¹⁴

Resveratrol oligomers isolated from VTT have shown several biological activities in vitro and in vivo. Resveratrol tetramers with the same molecular mass, including hopeaphenol, vitisin A, and vitisin B, show dose-dependent inhibitory activities against acetylcholinesterase and monoamine oxidase-B, and the oral administration of vitisin A showed improved learning and memory function assessed by passive avoidance tests in scopolamine-induced amnesiac mice.¹⁴ Resveratrol tetramer of vitisin A, a trimer of ampelopsin C, and dimer of (+)-ε-viniferin exhibited angiotensin converting enzyme inhibitory activities and vasorelaxant effects in phenylephrine-induced endothelium-intact aortic ring tensions. Thus, VTT extracts may be beneficial for regulating blood pressure.¹⁵ Resveratrol tetramer of vitisin A and dimer of (+)-ε-viniferin exhibited anti-obesity activities to reduce weight gains and improve cardiovascular risk parameters in high-fat diet-induced obese mice.^16,17 Resveratrol tetramers of vitisin A, vitisin B, and hopeaphenol had proved to show in vitro dose-dependent inhibitions against α-glucosidase and DPP-IV, implying the possible roles in regulations of glucose metabolisms, such as postprandial BG regulations and T2DM treatments, and needed further investigations using animal models.¹⁸ This study aimed to perform a proof-of-concept investigation of isolated resveratrol tetramers from VTT as anti-T2DM in nicotinamide/streptozotocin (STZ)-induced animal models. These in vivo animal experimental results revealed that VTT would be beneficial in developing nutraceuticals and/or functional foods for BG controls in T2DM patients, and this should be investigated further.

2. Materials and methods

2.1. Materials

Glucose, Gly-Pro-p-nitroanilide, nicotinamide, dimethyl sulfoxide (DMSO), STZ, and sitagliptin phosphate were purchased from Sigma Chemical Co. (St. Louis, MO, USA). Standard mouse/rat chow (Prolab® RMH2500, 5P14 diet) was purchased from PMI Nutrition International (MO, USA). Other chemicals were purchased from Sigma-Aldrich (St. Louis, MO). Proteins were quantified using the BCA protein assay kit (Pierce Biotechnology Inc., Rockford, IL, USA). The blood glucose content was determined using a glucose assay kit (Randox Laboratories, Ltd., Kearneysville, WV, USA). The blood insulin contents were determined by insulin ELISA kits (measurement range: 0.2–6.5 μg/L; 10–1247, Mercodia AB, Uppsala, Sweden).

2.2. Preparations of ethanol extracts of Vitis thunbergii var. taiwaniana and isolations of resveratrol tetramers

A voucher specimen (no. BT305) was deposited at the Department of Biological Science and Technology, National Pingtung University of Science and Technology, Pingtung, Taiwan. The preparation of VTT-S-95 E E and isolation of three resveratrol tetramers were as per previously published methods.^14,16,18 The dried, powdered VTT-S was immersed and extracted in 95% ethanol (W/V = 1:5) at room temperature for 1 week. After filtration, the residues were extracted for 1 week using the same procedure. These extracts were combined, concentrated, and lyophilized to produce VTT-S-95 E E (the extraction rate was approximately 5%) which was stored at −20 °C for further use. To isolate and purify the three resveratrol tetramers of the same molecular mass,¹⁸ the ethyl acetate fraction of VTT-S-95 E E was loaded onto a silica gel column (90 × 5 cm) for chromatography and eluted into 24 fractions using n-hexane-ethyl acetate mixing gradients. The last fraction (the 24th fraction) was purified using a reversed-phase C-18 column (60 × 3 cm) and eluted into seven fractions using water-methanol mixing gradients. The semi-preparative high-performance liquid chromatography (HPLC; Thermo Betasil C-18 column; flow rate, 2 mL/min) was used to prepare (+)-hopeaphenol (1) and (+)-vitisin A (2) from the 4th sub-fraction using water/acetonitrile (9:1 to 1:1) gradient, and to prepare (−)-vitisin B (3) from the 5th sub-fraction using water/acetonitrile (9:1 to 5:5) gradient. The purities of hopephenol, vitisin A, and vitisin B were greater than 98%, as determined by HPLC-UV analysis. For HPLC fingerprinting analysis of VTT-S-95EE, the analytical C-18 column (Phenomenex Luna, 5 μm, 250 × 46 mm) equipped with Hitachi L-7000 chromatography system was used, and gradient elution was performed applying the following programs of solvent mixtures containing distilled water and acetonitrile: water/acetonitrile, 95/5, at 0–5 min; 76/24, at 35 min; 60/40, at 70 min.¹⁴ The identified resveratrol-related compounds included resveratrol (37.05 min), (+)-hopeaphenol (43.51 min), (+)-ε-viniferin (46.95 min), (+)-vitisin A (52.77 min), and (−)-vitisin B (61.22 min).

2.3. Simulation of postprandial blood glucose regulations in ICR mice by oral glucose tolerance test

We obtained 8-week-old male ICR mice (N = 40) from the Laboratory Animal Center at National Taiwan University. They were housed in stainless steel cages in a temperature- (22 °C) and humidity-controlled room with a 12-h light and dark cycle, and unfettered access to food and water. The Institutional Animal Care and Use Committee of Taipei Medical University (LAC-98-0075) reviewed and approved the protocols for the animal experiments (Fig. 1A). After 1 week of acclimation, the mice were categorized randomly into seven groups (N = 5 for each group) for three oral experiments. The first oral experiment involved three groups: blank, hopephenol (100 mg/kg), and vitisin A (100 mg/kg). The second experiment involved two groups: blank and vitisin B (100 mg/kg). The third experiment involved three groups: blank, VTT-S-95 E E (100 mg/kg), and vitisin B (50 mg/kg). Sample pre-treatments and oral glucose tolerance test (OGTT) to ICR mice followed the previous reports.^12,19 Each sample was sonically suspended and categorized into three groups: control, sitagliptin-treated (positive control), and distilled water, as previously reported.^14,15,17 After being fasted for about 16–18 h, mice were orally administered samples (VTT-S-95EE, hopeaphenol, vitisin A, and vitisin B) or distilled water (the blank) by oral gavage using a feeding tube 30 min before the oral glucose administration (2.5 g/kg of body weight).^18,20 For mice in the blank group, an equal volume of distilled water was used by oral gavage using a feeding tube 30 min before the oral glucose administration (2.5 g/kg of body weight). Blood was drawn from the tail vein at −30, 0, 15, 30, 60, 90, and 120 min to determine BG levels using a glucose assay kit. The area under the BG curve from 0 to 120 min (BG-AUC_0-120) was calculated by adding each assay value at 0, 15, 30, 60, 90, and 120 min for each blank and sample. Animals were euthanized using carbon dioxide.

Fig. 1.

Effects of VTT-S-95 E E on the changes of blood glucose. (A) Schematic protocols of animal experiments on simulations of postprandial blood glucose regulations in healthy ICR mice by oral glucose tolerance test (OGTT). Sample was administered orally 30 min before the glucose loads (2.5 g/kg of body weight, red arrow). (B) Effects of oral administration of VTT-S-95 E E (100 mg/kg, blue arrow) on the changes of blood glucose. The distilled water was used instead of sample additions as the blank. (C) The HPLC chromatograms of VTT-S-95EE, and the structures of hopeaphenol (1), vitisin A (2), and vitisin B (3). The identified compounds included resveratrol (37.05 min), (1) (43.51 min), (+)-ε-viniferin (resveratrol dimer, 46.95 min), (2) (52.77 min), and (3) (61.22 min). For OGTT in the healthy ICR mice, the difference between the test sample and the blank group at the same time point was statistically compared by the Student t-test when P < 0.05 (∗), or P < 0.01 (∗∗), or P < 0.001 (∗∗∗).

2.4. Nicotinamide/streptozotocin-induced mice with type 2 diabetes mellitus

The use of nicotinamide/STZ to induce T2DM-ICR mice was performed as per previous reports with modifications.^21,22 After a 2-week acclimation period, the mice were randomly divided into two groups: blank (N = 6) and T2DM-induced (N = 18 for three subgroups: control, sitagliptin-treated, and vitisin B-treated). After fasting for approximately 16 h, blood was drawn from the tail vein of each ICR mouse to determine the basal BG level, and hyperglycemia was induced by nicotinamide/STZ for 2 weeks. For mice in the T2DM-induced group in the first stage (day 0–14), nicotinamide (120 mg/kg in 0.9% sodium chloride solution) was injected intraperitoneally on day 14, followed by the freshly prepared STZ (100 mg/kg in the 50 mM sodium citrate buffer, pH 4.5) in an ice bath 15 min later. On day −12, the same injection protocol was used to induce T2DM in each ICR mouse. For mice in the blank group, 0.9% sodium chloride solution was administered instead of nicotinamide solution. The 50 mM sodium citrate buffer (pH 4.5) was administered instead of STZ in the first induction stage. Each mouse with a BG level >230 mg/dL was identified as having T2DM successful induced. In the second stage, T2DM-induced ICR mice were randomly categorized into control, sitagliptin-treated (positive control), and vitisin B-treated. Each sample was sonicated in distilled water, as previously reported.^14,15,17 The control group received a single administration of distilled water daily by oral gavage using a feeding tube. However, the two intervention groups received either 10 mg/kg sitagliptin (sitagliptin-treated group) or 50 mg/kg vitisin B (vitisin B-treated group) for additional 28 days (days 0–28) to evaluate hypoglycemic activities. In the second stage, mice in the blank group were administered an equal volume of distilled water once daily via oral gavage using a feeding tube. Each mouse was weighed weekly, and the accumulated feed, water intake, and fasting BG levels from the tail vein were monitored every 7 days until the end of the experiments (days 14–28).

2.5. Effects of vitisin B treatments on the hypoglycemic activities of mice with type 2 diabetes mellitus

On day 28, the untreated (the control) and treated-T2DM mice in each group were examined for impaired glucose regulations by OGTT.^18,20 Mice were fasted for 16–18 h, the blood was drawn from the tail vein at time 0 to determine baseline BG and insulin levels, and the glucose (2.5 g/kg of body weight) was orally administered as described in the previous section. Tail vein blood was drawn at 15, 30, 60, 90, and 120 min, and BG and insulin levels were determined using glucose and mouse insulin enzyme-linked immunosorbent assay (ELISA) kits, respectively. (BG-AUC_0-120 and insulin-AUC_0-120 during the OGTT were calculated for each group. The animals were treated with carbon dioxide and blood samples from each group were assayed for aspartate transaminase (AST), alanine transaminase (ALT), total cholesterol (T-CHO), triglycerides (TG), bilirubin (BUN), creatinine (CREA), low-density lipoprotein cholesterol (LDL), and high-density lipoprotein cholesterol (at the National Laboratory Animal Center (Nangang, Taipei) using an automatic biochemistry analyzer (Hitachi 7080, Hitachi HiTech Co., Tokyo, Japan). The pancreases of mice in each group were harvested, sectioned, and separately fixated in 10% buffered formalin. The required tissues were embedded in paraffin and sectioned at 3–5 μm thickness. Hematoxylin and eosin (H&E) stained sections were viewed under a microscope (Eclipse TS-100, Nikon) equipped with a DS camera (DS-L2).

The plasma DPP-IV activity of mice in each group was measured according to a previous report.¹² On an ELISA reader, the absorbance was measured at 405 nm every 5 min for 30 min using 150 μL of reaction solution containing 5 or 10 μL of plasma sample, 100 mM Tris buffer (pH 8.0), and 1 mM Gly-Pro-p-nitroanilide. As a sample blank, sitagliptin phosphate (200 nM) in Tris buffer (pH 8.0) was premixed with the plasma sample for 10 min to inhibit DPP-IV activity in the plasma, Following the addition of 1 mM Gly-Pro-p-nitroanilide, absorbance was measured at 405 nm every 5 min for 30 min. The plasma DPP-IV activity of the mice in each group was calculated as A405 (in plasma) − A405 (in plasma premixed with sitagliptin phosphate). The plasma DPP-IV activity of mice in the blank group was considered to be 100%, and the changes in plasma DPP-IV activity of each group were expressed as relative DPP-IV activity (% of the blank). The pancreatic insulin content was determined using a mouse insulin ELISA kit. The extracted protein was quantified using the BCA protein assay kit with bovine serum albumin as a standard. The hepatic glycogen content was determined using the anthrone-sulfuric acid method.²³ Each group's mouse liver tissue (5 mg) was extracted with 1 mL of 30% potassium hydroxide in a water bath at boiling temperature for 20 min, cooled, precipitated with 95% ethanol, and centrifuged at 750×g for 15 min. The precipitate was redissolved in distilled water, precipitated with 95% ethanol, and centrifuged again. The recovered precipitate was re-dissolved in distilled water. Under ice-cold conditions, A 1-fold volume of the sample was aliquoted and mixed with a 2-fold volume of 0.2% anthrone in 95% sulfuric acid. After 10 min of heating the mixture in boiling water, the absorbance was measured at 680 nm after cooling. Glucose was used to plot a standard curve, and the amount of glycogen was expressed as mg/g of liver tissue.

2.6. Statistical analyses

Data are expressed as mean ± SD. Prism software (version 5.0; San Diego, CA, USA) was used for statistical analyses. For OGTT in ICR mice, the difference between the test sample and the blank group at the same time point, or BG-AUC_0-120 between the test sample and the blank group, was statistically compared using Student's t-test when P < 0.05 (∗), P < 0.01 (∗∗), or P < 0.001 (∗∗∗). For multiple group comparisons in nicotinamide/STZ-induced T2DM mice, one-way analysis of variance (ANOVA) followed by post hoc Tukey's test was used to compare differences among groups, including plasma DPP-IV activity, pancreatic insulin content, hepatic glycogen content, OGTT of T2DM mice at the same treatment time, BG-AUC_0-120, and insulin-AUC_0-120. Different lowercase letters in each bar or the same time interval among the groups indicate statistically significant differences (P < 0.05).

3. Results

3.1. Effects of VTT-S-95 E E pre-treatments on blood glucose levels using oral glucose tolerance test in ICR mice

Fig. 1A shows the protocol of OGTT experiments in healthy ICR mice. Postprandial BG regulation was imitated by OGTT, and sample pre-treatment once 30 min before the glucose loads was designed to evaluate improved postprandial BG activity. Fig. 1B shows the effects of VTT-S-95 E E pre-treatment on BG levels over 120 min using OGTT. A single oral administration of VTT-S-95 E E (100 mg/kg) in healthy ICR mice 30 min before glucose (2.5 g/kg of body weight) oral administration reduced the increased BG at time intervals of 15, 30, 60, 90, and 120 min. It differed significantly from those in the blank without sample pre-treatment (P < 0.05 × or P < 0.01∗∗). The average BG levels of mice pre-treated with VTT-S-95 E E (100 mg/kg) in the OGTT returned to normal levels faster than those in untreated mice, indicating the potential use of VTT-S-95 E E to reduce postprandial BG levels after meals. The identified resveratrol-related compounds of VTT-S-95 E E in the HPLC profile (Fig. 1C) included resveratrol, hopeaphenol (1), ε-viniferin, vitisin A (2), and vitisin B (3), among which hopeaphenol (1), vitisin A (2), and vitisin B (3) were resveratrol tetramers with the same molecular mass.

3.2. Effects of hopeaphenol, vitisin A, and vitisin B pre-treatments on blood glucose levels by oral glucose tolerance test in ICR mice

The molecular masses of hopeaphenol, vitisin A, and vitisin B were 906 Da. Each was used for OGTT experiments to evaluate BG regulation and identify the possible active components of VTT-S-95 E E (Fig. 2A–D). A single oral pre-treatment of vitisin B (100 mg/kg, Fig. 2C), but not hopeaphenol (100 mg/kg, Fig. 2A) or vitisin A (100 mg/kg, Figs. 2B), 30 min before glucose loading significantly reduced the BG of treated mice at 15, 30, 60, and 120 min. The difference was statistically significant than that in the blank (P < 0.05∗). Vitisin B pre-treatment (50 mg/kg, Fig. 2D) also decreased the BG of treated mice at 60, 90, and 120 min, and the difference was statistically significant than that in the blank (P < 0.05 × or P < 0.01∗∗). Fig. 2E shows the calculated BG-AUC_0-120 from the OGTT experiments. VTT-S-95 E E (100 mg/kg) and vitisin B (50 and 100 mg/kg) pre-treatments, but not hopeaphenol (100 mg/kg) and vitisin A (100 mg/kg), significantly decreased BG-AUC_0-120 than those in the blank group (P < 0.001). Fig. 1, Fig. 2 show that VTT-S-95 E E or vitisin B pre-treatment 30-min before glucose loading in the OGTT experiments of the healthy ICR mice could reduce the increased BG levels, and the decrease to the normal levels occurred more rapidly than that in the blank without sample pre-treatments. Therefore, vitisin B was further selected to investigate proof-of-concept to treat impaired glucose metabolisms in nicotinamide/STZ-induced T2DM mice.

Fig. 2.

Effects of resveratrol tetramers on the changes of blood glucose. The oral administration of (A) hopeaphenol (100 mg/kg), (B) vitisin A (100 mg/kg), (C) vitisin B (100 mg/kg), and (D) vitisin B (50 mg/kg) on the changes of blood glucose. The distilled water was used instead of sample additions as the blank. (E) The area under curve during 0–120 min of blood glucose (AUC_0-120) of the blank and the sample preload. For OGTT in the healthy ICR mice, the difference between the test sample and the blank group at the same time point, or AUC_0-120 between the test sample and the blank group was statistically compared by the Student t-test when P < 0.05 (∗), or P < 0.01 (∗∗), or P < 0.001 (∗∗∗).

3.3. Effects of vitisin B on nicotinamide/streptozotocin-induced mice with type 2 diabetes mellitus

Fig. 3A shows the protocol for T2DM induction in mice using nicotinamide/STZ. The induction period for each T2DM mouse was 2 weeks (day -14–0). Herein, each mouse with a fasting BG level >230 mg/dL (13 mM) was identified as having successful T2DM induction. The established T2DM-ICR mice were grouped and orally administered either vitisin B (50 mg/kg), sitagliptin (10 mg/kg), or the same volume of distilled water (control) using oral gavage. The mice in the T2DM-induced group consumed more food and water than those in the non-induced mice in the blank group (Supplementary Figs. S1A and S1B). The mice's body weights did not differ significantly among the groups during the experimental period (Supplementary Fig. S1C). Fig. 3B shows the BG levels of each group during the experiments. The mice's average BG (mg/dL) in the blank were 137.57 ± 7.39 (day −14), 145.74 ± 8.51 (day −7), 145.74 ± 14.20 (day 0), 133.56 ± 3.48 (day 7), 126.19 ± 5.05 (day 14), and 128.22 ± 3.54 (day 21), which were in the ranges of 130–140 mg/dL. After a 2-week induction, the OGTT from 0 to 120 min in mice's average BG in control on day 0 was 241.47 ± 1.16 mg/dL, which was higher than 230 mg/dL and increased gradually during experiments of 237.14 ± 1.45 mg/dL (day 7), 285.42 ± 0.35 mg/dL (day 14), and 288.34 ± 3.05 mg/dL (day 21), and were much higher, with significant differences (P < 0.05) than those in the same time intervals of the blank. On day 0, the sitagliptin-treated (277.50 ± 6.12 mg/dL) and vitisin B-treated (263.16 ± 9.26 mg/dL) T2DM mice showed BG levels higher than 230 mg/dL, and both were much higher, with significant differences (P < 0.05) than those in the blank; therefore, these mice were identified as successful T2DM inductions, and the success rate in the chemical induction of this study was 100%.

Fig. 3.

Effects of 28-day vitisin B treatments on impaired glucose regulations of nicotinamide/streptozotocin (STZ)-induced type 2 diabetes mellitus (T2DM)-ICR mice. (A) Schematic protocols of T2DM animal experiments included a two-week T2DM induction (day −14 to day 0) and a four-week sample treatment (day 0 to day 28), included vitisin B (50 mg/kg) and sitagliptin (10 mg/kg) daily treatments. The T2DM mice received distilled water were the control group, and healthy ICR mice (without T2DM induction) received distilled water were the blank group. (B) The fasting blood glucose levels of mice in healthy mice (the blank) and the three T2DM mice during induction periods (day −14 to day 0) and treatment periods (day 0 to day 21). (C) At day 28, the simulation of postprandial blood glucose regulations of healthy mice (the blank) and the three groups of T2DM-ICR mice by OGTT during 0–120 min. The AUC_0-120 of blood glucose was calculated for blank and the sample preload during 0–120 min. (D) The changes of blood insulin levels of healthy mice (the blank) and the three groups of T2DM-ICR mice by OGTT during 0–120 min. The AUC_0-120 of blood insulin was calculated during 0–120 min. For multiple group comparisons in the healthy ICR mice and nicotinamide/STZ-induced T2DM-ICR mice, the one-way analysis of variance (ANOVA) followed by the post hoc Tukey's test was used to compare differences among groups. The different lowercase alphabet in each bar or the same time interval among groups were significantly different (P < 0.05).

On day 0, mice in the sitagliptin-and vitisin B-treated groups showed significantly higher average BG levels than those in the control group (P < 0.05). After daily treatment with sitagliptin (positive control), the fasting BG levels of T2DM mice were reduced from 277.50 ± 6.12 mg/dL (day 0) to 212.78 ± 9.57 mg/dL (day 7), 215.16 ± 10.14 mg/dL (day 14), and 188.82 ± 2.21 mg/dL (day 21), demonstrating an excellent improvement in the fasting BG levels of T2DM mice, which is significantly different than those in the control at each time interval (P < 0.05). Furthermore, after daily treatment with vitisin B, the fasting BG levels of T2DM mice also reduced from 263.16 ± 9.26 mg/dL (day 0) to 211.84 ± 1.81 mg/dL (day 7), 215.65 ± 4.59 mg/dL (day 14), and 215.34 ± 9.66 mg/dL (day 21), which showed excellent hypoglycemic activities similar to sitagliptin on improving the fasting BG levels and the differences compared to those in the control were significant (P < 0.05) at each time intervals. This study is the first to report that daily treatment with vitisin B has anti-diabetic effects in vivo by improving impaired glucose metabolism in the fasting BG of T2DM-ICR mice.

Fig. 3C shows the changes in simulated postprandial BG in the OGTT from 0 to 120 min in glucose-intolerant T2DM mice and healthy mice on day 28. The baseline levels (time 0) of fasting BG of blank, control, sitagliptin- or vitisin B-treated mice were 131.32 ± 4.34 mg/dL, 254.24 ± 3.28 mg/dL, 223.34 ± 10.08 mg/dL, and 217.46 ± 3.64 mg/dL, respectively. Sitagliptin- or vitisin B-treated T2DM mice still showed a fasting BG less than 230 mg/dL, and this effect was significantly different from that of the control (P < 0.05). After glucose administration, the mice BG levels in the blank group at 15, 30, 60, 90, and 120 min were 292.41 ± 9.47 mg/dL, 275.94 ± 16.05 mg/dL, 207.30 ± 15.94 mg/dL, 182.49 ± 6.31 mg/dL, and 158.84 ± 7.50 mg/dL, respectively, demonstrating well-controlled BG at each time point, which was the lowest of the four groups. The changes in BG levels after glucose loading in the three T2DM-ICR mice were as follows: The BG levels (mg/dL) in the control group at 15, 30, 60, 90, and 120 min were 533.33 ± 34.84, 489.74 ± 1.21, 433.39 ± 5.43, 388.18 ± 0.98, and 371.83 ± 9.35, respectively; the BG levels (mg/dL) in the sitagliptin-treated group at 15, 30, 60, 90, and 120 min were 464.29 ± 18.42, 524.83 ± 6.38, 338.21 ± 16.55, 368.98 ± 16.89, and 320.95 ± 15.02, respectively; the BG levels (mg/dL) in the vitisin B-treated group at 15, 30, 60, 90, and 120 min were 471.04 ± 13.95, 434.77 ± 14.67, 430.52 ± 4.13, 385.99 ± 43.07, and 345.22 ± 33.62, respectively. Among the three groups of T2DM-ICR mice in the OGTT, vitisin B-treated T2DM mice had the lowest BG levels at 15-min and 30-min (P < 0.05), but the differences were not significant (P > 0.05) at 90-min and 120-min (P > 0.05). Among the three groups of T2DM-ICR mice, OGTT showed that the sitagliptin-treated T2DM mice showed similar BG levels to those of vitisin B-treated ones at 15-min; the BG levels peaked at 30-min and were at their lowest ones at 60-min (P < 0.05). The calculated BG-AUC_0-120 in the control T2DM mice increased 2-fold than that of healthy mice in the blank group (P < 0.05). Both vitisin B and sitagliptin treatment significantly decreased BG-AUC_0-120 than the control (P < 0.05). Fig. 3D shows the blood insulin concentration during the OGTT from 0 to 120 min in glucose-intolerant T2DM mice and healthy mice on day 28. Healthy ICR mice in the blank quickly responded to the sudden increase in BG by secreting the highest amounts of insulin at 15-min (approximately 1.31 μg/L) to regulate BG. However, the dysfunctional pancreatic β-cells in three nicotinamide/STZ-induced T2DM-ICR mice showed lowered insulin secretion at 15-min (0.65, 0.56, and 0.52 μg/L, respectively, for mice in the control, sitagliptin-treated, and vitisin B-treated group), thus, lowered insulin secretion could not properly regulate the sudden increase in BG in OGTT. The calculated insulin-AUC_0-120 in the blank group showed a 1.24-fold increase compared to that of T2DM mice in the control group (P < 0.05). Both vitisin B-treated and sitagliptin-treated T2DM mice showed similar values and had significant reductions in insulin-AUC_0-120 compared to the control (P < 0.05), which matched the BG-AUC_0-120 (Fig. 2C).

3.4. Effects of vitisin B treatments on dipeptidyl peptidase-IV activities, insulin levels, and plasma biochemical indices of nicotinamide/streptozotocin-induced mice with type 2 diabetes mellitus

After 28 days of treatment, healthy mice in the blank group and three nicotinamide/STZ-induced T2DM-ICR mice were sacrificed for further investigation. We estimated changes in plasma biochemical indices, including AST and ALT (Supplementary Fig. S1D), T-CHO and TG levels (Supplementary Fig. S1E), BUN and CREA levels (Supplementary Fig. S1F), and LDL/HDL ratio (Supplementary Fig. S1G). Generally, nicotinamide/STZ-induced T2DM mice had elevated plasma biochemical indices compared to healthy mice (blank) (P < 0.05). The 28-day sitagliptin-treated T2DM-ICR mice had decreased T-CHO and TG levels but increased levels of AST, ALT, BUN, and the LDL/HDL ratio, with statistically significant differences than those in the control (P < 0.05). The side effects of sitagliptin treatment on the plasma biochemical indices of T2DM-ICR mice are unkwon and could be due to the peculiarities of ICR mice, which require further investigation. Vitisin B-treated T2DM mice for 28 days showed decreased plasma biochemical indices, including AST, ALT, T-CHO, and TG, and the LDL/HDL ratio, and the differences were statistically significant (P < 0.05) than those in untreated T2DM-ICR mice (control). These plasma biochemical parameters (except CREA in Supplementary Fig. S1F) of vitisin B-treated T2DM mice were also significantly lower (P < 0.05) than that of sitagliptin-treated T2DM-ICR mice. Fig. 4A shows a significant increased plasma DPP-IV activity in the untreated T2DM-ICR mice (control group) than that in the healthy ICR mice in the blank group (P < 0.05). Both vitisin B-treated and sitagliptin-treated T2DM-ICR mice showed significant decrease in plasma DPP-IV activity (P < 0.05) than control mice. Determination and comparison of the insulin content in the pancreatic extracts (Fig. 4B) showed that the insulin content in the healthy mice and 28-day-treated T2DM mice was 3.30 ± 0.007 (μg/mg pancreatic protein) and 0.29 ± 0.007 (μg/mg pancreatic protein), respectively. The latter maintained only approximately 8.78% of the level in the healthy mice. However, the insulin content in the pancreatic extracts of the sitagliptin-treated mice and vitisin B-treated T2DM-ICR mice was 0.34 ± 0.005 (μg/mg pancreatic protein) and 0.44 ± 0.002 (μg/mg pancreatic protein), respectively, showing a significant increase in insulin levels (P < 0.05) of 17% and 51.7% than those of the untreated T2DM-ICR mice in the control. Glycogen content in the liver extracts of untreated T2DM mice and treated T2DM-ICR mice showed no significant differences among the groups (P > 0.05, Fig. 4C). Based on these results, the improvement in impaired glucose regulation in nicotinamide/STZ-induced T2DM mice after 28-day treatment with the resveratrol tetramer of vitisin B might be partly due to DPP-IV inhibition, which the very low level of insulin could maintain and continue to regulate abnormal BG levels. H&E staining of paraffin-embedded sections of pancreatic islets from mice (Fig. 5) in blank (A), untreated T2DM control mice (B), sitagliptin-treated T2DM mice (C), and vitisin B-treated T2DM mice (D). The shrinkage of pancreatic islets in nicotinamide/STZ-induced T2DM mice was examined, and sitagliptin- and vitisin B-treated T2DM-ICR mice had more intact and compact pancreatic islets tissues than untreated T2DM-ICR mice (control).

Fig. 4.

The changes of DPP-IV activities, insulin and glycogen levels in T2DM-ICR mice. After being sacrificed at day 28, (A) the plasma DPP-IV activity (% of the blank), (B) the insulin contents (μg/mg pancreatic protein) in the pancreatic extracts, and (C) the glycogen contents (μg/mg liver tissue) in the hepatic extracts of healthy mice (the blank) and the three groups of T2DM-ICR mice. Data were expressed as mean ± SD. For multiple group comparisons in the healthy ICR mice and nicotinamide/STZ-induced T2DM-ICR mice, the one-way analysis of variance (ANOVA) followed by the post hoc Tukey's test was used to compare differences among groups. The different lowercase alphabet in each bar or the same time interval among groups were significantly different (P < 0.05).

Fig. 5.

The hematoxylin and eosin (H&E) stains of the fixed pancreatic tissues. (A) The healthy ICR mice, (B) the T2DM-ICR mice in the control, (C) the sitagliptin-treated T2DM-ICR mice, and (D) the vitisin B-treated T2DM-ICR mice.

4. Discussion

In this study, daily treatment with vitisin B for approximately 1 month showed anti-diabetic activities in vivo by improving the impaired glucose metabolism in fasting BG and the AUC of postprandial BG (measured by OGTT over 120 min) in T2DM-ICR mice. Vitisin B pre-treatment in a simulated postprandial BG change in healthy mice decreased the elevated BG levels. The decrease to normal levels was more rapid than that in the blank without sample pre-treatment. Vitisin B ranked highest in inhibiting α-glucosidase and DPP-IV among the same molecular mass of resveratrol tetramers of hopeaphenol, vitisin A, and vitisin B.¹⁸ Therefore, vitisin B was selected to investigate the proof-of-concept for treating impaired glucose metabolism in nicotinamide/STZ-induced T2DM mice. The initial 7-day treatment with sitagliptin or vitisin B achieved the highest reduction. BG level decrease to normal levels of 50 mg/dL. It could maintain fasting BG levels at around 210 mg/dL until day 21, below the diabetic threshold of 220–240 mg/dL in nicotinamide/STZ-induced mice.²¹ The 28-day vitisin B treatment improved fasting glucose levels and glucose intolerance of the nicotinamide/STZ-induced T2DM mice, and these effects were similar to those of sitagliptin in this study. The doses administered in the animal experiments and human clinical trials are not multiplied directly by body weight.²⁴ In this study, the human equivalent dose was calculated to be 4.05 mg vitisin B/kg of human body weight from vitisin B-treated T2DM-ICR mice (50 mg/kg) based on body surface area normalization for dose translation from animal to human studies suggested by Reagan-Shaw et al.²⁴ An adult (60 kg body weight) might have to administer approximately 250 mg of vitisin B daily for 28 days to improve fasting BG in patients with mild T2DM, which should be investigated further.

It was reported that fasting BG levels and postprandial BG showed improvements in high-fat diet-induced obese rats after 33 days VTT-S-95 E E treatment. The three isolated resveratrol tetramers from VTT-S-95EE, including hopeaphenol, vitisin A, and vitisin B, showed dose-dependent inhibitory activities in vitro against α-glucosidase and DPP-IV.¹⁸ Therefore, animal experiments were performed to evaluate the BG regulation of VTT-S-95 E E and/or isolated resveratrol tetramers. The identified resveratrol-related compounds in the HPLC profile of VTT-S-95 E E (Fig. 1C) included resveratrol, hopeaphenol (1), ε-viniferin, vitisin A (2), and vitisin B (3),²⁵ and these resveratrol oligomers were also identified in the 95  E E of VTT root parts.¹⁴ The ε-viniferin showed dose-dependent inhibitions against α-glucosidase and DPP-IV activities,²⁶ and resveratrol has been reported to exhibit α-glucosidase and DPP-IV inhibitory activities.²⁷ Hopeaphenol and vitisin B could be biosynthesized enzymatically from ε-viniferin by peroxidase catalysis in the presence of hydrogen peroxide. Vitisin A could be generated by the acid-catalyzed reaction of vitisin B.^28,29 Vitisin B could also be chemically synthesized from ε-viniferin using silver acetate as the oxidant.²⁹

The oral administration of VTT-S-95 E E seemed to show better reductions in BG-AUC_0-120 (from 100% to 79.1%) than those treated with vitisin B (from 100% to 89.8%) under the same concentration of 100 mg/kg pre-treatment; thus, active compounds, including vitisin B, in the mixtures of VTT-S-95 E E might contribute to synergetic effects in BG regulation. VTT-S-95EE, associated with DPP-IV inhibition, can prolong GLP-1 activity and have preventive functions to ameliorate postprandial BG levels after ingesting starchy foods, thus lowering the possibility of an increase in the incidence of impaired glucose tolerance and T2DM, which needs to be investigated further. Among extracts of different parts of VTT (including stem, root, leaf, and branch), VTT-S-95 E E exhibited the highest inhibitory activities against angiotensin-converting enzyme. However, oral administration of VTT-S-95 E E exhibited anti-hypertensive activities in spontaneously hypertensive rats.¹⁵ Methanol extracts of different parts of VTT (stem, root, leaf, and branch) at the same concentrations showed the highest inhibition against lipopolysaccharide-induced and IL-1β-induced prostaglandin E₂ production in human chondrocytes.²⁵

STZ is frequently used to induce DM through selective pancreatic islet cell toxicity. Sequential treatments with nicotinamide and STZ reduced STZ toxicity in cells^22,30 and the functions of-cells in nicotinamide/STZ-induced T2DM mice were lowered but not completely lost. The fasting glucose concentrations of STZ-induced mice for mild hyperglycemia should be > 150 mg/dL; the BG level should be > 250 mg/dL in severe type 1 DM (T1DM) rats, and >300 mg/dL in severe T1DM mice.^22,31 The BG levels of DM animal models have been based on mice or rat strain^21,22; thus, the STZ injection frequency, doses, and animal body weight were accordingly adjusted.³² The 2-day successive nicotinamide/STZ treatments for T2DM inductions for 2-week periods suggested that the mouse with non-fasting BG level of 220–240 mg/dL suffered a diabetic condition.²¹ Herein, each mouse with fasting BG levels >230 mg/dL (13 mM) was recognized as a successful induction, and T2DM mice consumed more food and water which matched the T2DM symptoms of polyphagia, polydipsia, and polyuria.³³ The side effects of sitagliptin treatments on these plasma biochemical indices of T2DM-ICR mice were unknown and could be from the characteristics of ICR mice, which will need further investigation. In one case report, after sitagliptin treatment, two clinical patients with acute liver injuries showed increased AST and ALT levels, one with a 5-fold increase and the other with a >10-fold increase after 6 months.³⁴

Resveratrol demonstrated hypoglycemic activities in nicotinamide/STZ-induced T2DM rats.^35,36 A novel formulation containing resveratrol, ferulic acid, and epigallocatechin-3-O-gallate improved insulin-resistant smooth muscle cells by enhancing glucose transporter-4 translocation.³⁷ The combination of gallic acid and andrographolide stimulates insulin secretions in STZ-induced diabetic rats.³⁸ Osthole showed hypoglycemic effects in diabetic db/db mice via dual PPARα/PPARγ activations.³⁹ Similarly, isoquercitrin inhibited DPP-IV in mice with type 2 diabetes, enhancing and maintaining GLP-1 and insulin activities.⁴⁰ The anti-T2DM mechanisms of vitisin B exclusing DPP-IV inhibitory activities, require further investigations.

5. Conclusion

In conclusion, this animal study confirmed that the 28-day treatment with vitisin B, one of the active compounds in VTT-S-95EE, could improve fasting BG and postprandial BG in the OGTT, in part by DPP-IV inhibition and the elevation of insulin secretion, and thus showed similar anti-diabetic effects as sitagliptin in nicotinamide-STZ-induced T2DM mice. Moreover, the OGTT demonstrated that VTT-S-95 E E could ameliorate postprandial BG in healthy ICR mice, similar to what was observed in diet-induced obese rats. In Taiwan, dried whole VTT is used as a tea-substitute. Therefore, these results suggest that vitisin B may be useful in developing functional foods for glycemic control in patients with T2DM and will require further investigation.

Author contributions

Y.-H. L., Conceptualization, Resources, Supervision, Funding acquisition; Y.-S. L., Investigation, Data curation, Formal analysis; Y.-Y. S., Investigation, Data curation, Formal analysis; C.-C. W., Conceptualization, Resources, Supervision; C.-I C., Conceptualization, Resources, Supervision; W.-C. H., Conceptualization, Resources, Supervision, Funding acquisition, Writing – original draft, Writing - review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article.