Antidiabetic effect of ligustilide-rich total lactones derived from Shunaoxin dropping pills on mice with type 2 diabetes induced by a high-fat diet and streptozotocin

Abstract

Shunaoxin dropping pill, a well-known Traditional Chinese Medicine formula, has been used to treat cerebrovascular diseases in China since 2005. It is composed of two herbs named Chuanxiong (Ligusticum chuanxiong Hort, Umbelliferae) and Danggui (Angelica sinensis radix, Umbelliferae). It has been reported that these two herbs have anti-diabetic activities. Phenolic acids and lactones are the main active components in Chuanxiong and Danggui. Ferulic acid and ligustilide are the representative compounds of phenolic acids and lactones, respectively. Previous study has documented the hypoglycemic activity of ferulic acid. However, the anti-diabetic activities of total lactones from Chuanxiong and Danggui or ligustilide have not been well studied. Therefore, the present study was designed to demonstrate the antihyperglycemic and antihyperlipidemic activities of ligustilide-rich total lactones (LT) from Shunaoxin dropping pills in vivo, thereby elucidating its probable antidiabetic mechanism. The present results showed that after 3 weeks of treating with LT, the blood glucose levels were significantly reduced in diabetic mice, and the oral glucose tolerance test showed that LT could improve glucose tolerance. The serum insulin concentration was also dramatically reduced. In addition, there was a marked decline in the serum levels of TG, TC, and LDL-C in mice treated with LT. In addition, the serum level of HDL-C was enhanced to a certain degree in diabetic mice treated with LT. In the histopathological examination of the mouse pancreas, LT showed significant protection of the pancreas in diabetic mice. The results provide a sound rationale for future clinical trials of the oral administration of LT for the primary prevention of diabetes mellitus.

---

Introduction

Type 2 diabetes mellitus (T2DM) is a chronic metabolic disorder characterized by impaired homeostasis of lipid and carbohydrate metabolism, and which ultimately results in insulin resistance and subsequent hyperglycemia. A lipid metabolic disorder is one of the major pathogeneses in patients with T2DM, characterized by higher serum triglycerides (TG), less high density lipoprotein (HDL-C) and more low density lipoprotein (LDL-C).¹ Over the past few decades, T2DM has precipitated into a worldwide epidemic, now spanning both developed and developing societies.² There are currently 382 million diabetes mellitus patients worldwide and the number of patients is predicted to reach 592 million by the year 2035.³ The prevalence of T2DM is becoming a health burden that is reaching global epidemic proportions.⁴

Five classes of oral antidiabetic agents are available: aglucosidase inhibitors, sulfonylurea, meglitinides, biguanides, and thiazolidinediones.⁵ However, the use of these medicines has serious side effects and causes secondary failure, including bloating, flatulence, and diarrhea.⁶ In ancient China, diabetes mellitus is known as “Xiao Ke Zheng”, which is closely related to lung heat, excessive fire in the stomach, deficiency of kidney Yin, or lack of both Yin and Yang. Traditional Chinese Medicines (TCMs) have been used to counteract similar syndromes diabetes mellitus for thousands of years based on unique theory system with few side effects, and has been attracting more and more attention for its use as a dialectical therapy.⁷

Shunaoxin dropping pill, a well-known Traditional Chinese Medicine formula, has been used to treat cerebrovascular diseases in China since 2005 (approval number Z20050041). It is composed of Chuanxiong (Ligusticum chuanxiong Hort, Umbelliferae) and Danggui (Angelica sinensis radix, Umbelliferae). Both Chuanxiong and Danggui have been widely used in China, Japan, Korea and other Asian countries for centuries. Chuanxiong and Danggui have been used to treat many cardiovascular and cerebrovascular diseases such as atherosclerosis and hypertension.⁸ It has been documented that Danggui polysaccharide could regulate glucose and lipid metabolism disorder in prediabetic and streptozotocin (STZ)-induced diabetic mice through the elevation of glycogen levels and reduction of inflammatory factors.⁹ Administration of tetramethylpyrazine, a compound derived from Chuanxiong, may reduce kidney damage caused by diabetes.¹⁰ Our previous studies have indicated that there are several phenolic acids and lactones in the extract of Shunaoxin dropping pills. The active components include ferulic acid, ligustilide, senkyunolide A, and butylidenephthalide.⁸ Previous studies have reported hypoglycemic activity of ferulic acid in STZ-induced diabetic mice and KK-Ay mice.¹¹ Ligustilide has a variety of biological activities in Traditional Chinese Medicine, including anti-inflammatory and neuroprotective effects. The present study was designed to demonstrate the antihyperglycemic and antihyperlipidemic activities of ligustilide-rich total lactones (LT) from Shunaoxin dropping pills in vivo, thereby elucidating its probable antidiabetic mechanism.

Materials and methods

Materials

STZ and glucose were produced by Sigma-Aldrich Company (St. Louis, MO, USA). Shunaoxin dropping pills were donated by Zhongxin Pharmaceuticals (Tianjin, China). Melbine (DMBG) was purchased from Heowns Biochem Technologies LLC (Tianjin, China). Xiao-ke-wan (XKW) was purchased from Tianjin He-ping drug store (Tianjin, China). Ligustilide was obtained from the National Institute for Control of Pharmaceutical and Biological Products (Beijing, China). Insulin enzyme-linked immunosorbent assay (ELISA) kits were obtained from Nanjing Jiancheng Bioengineer Company (Nanjing, China). Mouse anti-caspase-3, anti-Bax, and anti-Bcl-2 antibodies were purchased from Boster Biological Engineering Co., Ltd (Wuhan, China). Biotinylated goat antirabbit secondary antibody and 3,3-diaminobenzidine tetrahydrochloride (DAB) were purchased from ZSGBBIO (Beijing, China). High-fat feed (5.24 kcal g⁻¹) containing 26.2% of protein, 26.3% of carbohydrate, and 34.9% of fat was purchased from Beijing HFK Bioscience CO., Ltd (Beijing, China). In-house deionized water further purified with a Milli-Q Reagent Water System (Bedford, MA, USA). All the other reagents used in this study were of analytical grade.

Preparation of ligustilide-rich total lactones

Shunaoxin dropping pills were refluxed twice with 75% aqueous EtOH. The 75% aqueous EtOH extract was subjected to D101 macroporous resin CC and eluted with 50% aqueous EtOH followed by 95% aqueous EtOH, then collecting 95% aqueous EtOH. 95% aqueous EtOH extract is the ligustilide-rich total lactones (LT).

Animals

C57BL/6J mice weighing 18–22 g were obtained from Beijing HFK Bioscience CO., Ltd (Beijing, China). All animals were housed as 10 mice per cage and were fed with standard diet and were allowed free access to water, maintained at a constant temperature (25 ± 1 °C) and 45–55% relative humidity with a 12 h light–dark cycle. All experimental protocols were approved by the Animal Ethics Committees of the Faculty of Medicine, Tianjin Medical University, Tianjin, China, and carried out in accordance with “Principles of Laboratory Animal Care and Use in Research” (State Council of China, 1988).

Experimental protocol

Analysis of the chemical compounds in LT by HPLC

The chemical compounds analysis of samples was carried out using Waters 1525-2998 HPLC system attached to a photodiode array detector. A reversed phase C18 column, (250 mm × 4.6 mm i.d, 5 μm, Kromasil C18), preceded by a guard column (4 mm × 3.0 mm, Phenomenex) of the same stationary phase, was used for separation. Determination of content of ligustilide in LT was achieved under 280 nm. The mobile phase was consisted of 0.5% acetic acid in water (A) and acetonitrile (B) using a gradient elution of flow rate with 0.8 mL min⁻¹ at 30 °C. Linear gradient of 20% B to 80% B during first 20 min, followed by 80% B to 20% B during 20–21 min. After that, the organic phase was brought back to 20% (B) lasting for 5 min till column equilibration.

Establishment of diabetic mice

All animals were randomly divided into 2 groups. Ten mice were fed with common diets as the control group, and the other group mice were used to build type 2 diabetic mice model using high fat diet (Beijing HFK Bioscience Co., Ltd, Beijing, China) as the test group. After the exposure to respective diets for 4 weeks, test mice were intraperitoneally injected with either STZ dissolved in citrate buffer (pH = 4.5) at 60 mg kg⁻¹ body weight, or vehicle (0.05 M citric acid, pH = 4.5) consecutive 3 days. One week later, fasting blood glucose (FBG) levels were detected. Mice with FBG levels over 11.1 mM and accompanied with manifestations of polydipsia, polyuria and polyphagia were considered to be diabetic mice accepting following experiments.

Treatment schedule

STZ-induced type 2 diabetes mice were divided into 4 groups in 10 mice scale according to FBG levels and body weight and subsequently orally administered as follows: high-fat diet (HFD, STZ-induced diabetic control mice-distilled water 10 mL kg⁻¹ b.w.), DMBG (STZ-induced diabetic mice-DMBG 480 mg kg⁻¹ b.w.), XKW (STZ-induced diabetic mice-XKW 1250 mg kg⁻¹ b.w.), and LT (STZ-induced diabetic mice-LT 100 mg kg⁻¹ b.w.). The control group orally administered with equal volume of distilled water. Then the mice with type 2 diabetes were kept on the same diet for the following 3 weeks. FBG levels were obtained from the tail vein from overnight fasted mice, and measured using glucose meter (Sannuo, Beijing, China). Body weight was measured 30 min before administration every day. After 4 weeks of treatment, an oral glucose tolerance test (OGTT) was implemented, and fasting plasma was collected for further measurement of fasting insulin and other related biochemical markers. At the end of the study, the mice were killed and the brain and pancreas were isolated and immediately stored at −80 °C until further analysis. The experimental protocol is shown in Fig. 1.

Fig. 1 The sequence of study procedures. After 1 week acclimatization, all C57BL/6J mice were feed with high-fat diet 4 weeks. Then mice were injected subcutaneously STZ 60 mg kg⁻¹ d⁻¹ consecutively for 3 days. After being successfully established T2DM mice model, C57BL/6J diabetic mice were treated with the test sample consecutively for 3 weeks. FBG was determined at the end of every week. Morris water maze test were performed at 2nd week after mice treated with test sample. OGTT was performed at the last day of test sample treatment. After OGTT tests, mice were sacrificed and blood, brain, pancreas tissues were collected for biochemical analysis or histological examination.

Morris water maze test

After LT treatment 2 weeks, mice (n = 10 each group) were subjected to Morris water maze test as described previously.¹² Briefly, the experimental apparatus consisted of a circular pool filled with water (23 ± 1 °C). A platform was located 1 cm below the water surface in the center of the target quadrant, providing the only escape from the water. During 6 days of training, the mice underwent 4 trials a day, alternating among 4 pseudorandom starting points. If a mouse failed to find the platform within 90 s, it was guided to the platform by the researcher and kept there for 15 s. Probe trials were conducted 24 h after the last training trial. During the probe trials, the platform was removed and mice were free to swim in the tank for 90 s. The training and probe trials were recorded by a video camera mounted on the ceiling, and data were analyzed using the behavior analyzing system (Anhui Zhenghua Biological Equipment Co., Ltd, Anhui, China). This system was used to record the swimming trace and calculate the latency to the platform and platform crossing times. The times that the animal crossed the position where the platform was placed during the learning session (crossing times) and time required to locate the hidden escape platform (escape latency) were recorded using the behavior analyzing system.

Assessment of water and food intake, fasting blood glucose, and oral glucose tolerance test

Water and food intake were daily recorded. Fasting blood glucose (FBG) levels were measured weekly by a blood glucose meter (Sannuo, Beijing, China). OGTT was also performed after the 3 weeks of treatment. After 12 h of fasting, the animals were orally given glucose dissolved in water at 2 g kg⁻¹ body weight. Blood glucose taken from the tail tip at 0, 30, 60, 120, and 180 min after glucose administration was measured using Sannuo blood glucose meter.

Blood and tissue sample collection

The mice were weighed every week. On the final day of the experiment, blood samples were collected from the orbital venous plexus of 12 h fasted and anesthetized animals. The blood samples were centrifuged at 4000g for 10 min, and serum was obtained to analyze levels of insulin, TG, total cholesterol (TC), HDL-C, and LDL-C. Then mice were sacrificed by cervical dislocation. Pancreatic and brain tissues were immediately removed and rinsed with cold 0.9% saline, then fixed in 4% paraformaldehyde solution.

Determination of blood lipids and serum insulin

The concentrations of the plasma lipids, including TC, TG, HDL-C, and LDL-C were measured using a fully Glamour 3000 Automatic Biochemical Analyzer. Plasma insulin content was analyzed by a multi-well plate reader using the mice insulin ELISA kit (Nanjing Jiancheng Bioengineer Company, Nanjing, China).

Immunohistochemical observation

Immunohistochemical staining performed on 4 μm thick paraffin pancreatic sections were deparaffinized with xylene and rehydrated with graded concentrations of isopropyl alcohol. Separated sections were processed. Slides were incubated overnight with mouse anti-caspase-3, anti-Bax, or anti-Bcl-2 antibody (1 : 200 dilutions). The slides were rinsed well with phosphate buffer and incubated with super enhancer reagent for 30 min. After rinsing with phosphate buffer, incubation was done with supersensitive polymer-horseradish peroxidase immunohistochemistry detection system. Sections were washed with buffer and incubated with DAB solution for 5 min and then observed under the light microscope.

Histopathological analysis

Pancreatic tissues and brain tissues from each group were fixed in 4% paraformaldehyde solution for 24 h. After routine tissue processing, the tissues were embedded in paraffin. Then, 4 μm thick sections obtained from each paraffin block were stained with hematoxylin and eosin (H&E) staining for histopathological evaluation. Images were taken with a digital camera (ECLIPSE TS 100, Nikon). Images analyzed by using IPP 6.0.

Statistical analysis

All values are expressed as means ± standard errors (s.e.). Data were analyzed by a one-way analysis of variance (ANOVA), and differences among means were analyzed using Dunnett's test or Fisher's protected LSD multiple comparison test. Tests were performed using SPSS 20.0 system (Chicago, IL); P value less than or equal to 0.05 was considered to be statistically significant.

Results

Contents of ligustilide in LT

The content of ligustilide in LT was finally quantified using corresponding calibration curves y = 1 × 10⁹x − 255 363 with good correlation coefficients (r² = 0.9998). The content of ligustilide in LT was 690 mg g⁻¹. The representative chromatograms were shown in Fig. 2.

Fig. 2 Representative HPLC chromatogram of LT at the wavelength 280 nm. The content of ligustilide in LT was 690 mg g⁻¹.

Body weight

The body weights changes of the mice in different groups during the experiment are summarized in Table 1. The final body weights of mice in each group decreased after 3 week treatments. However, as expected, the body weights of mice in control group were significantly higher than that of T2DM mice during the 3 week experiment. Body weights in HFD mice had the significant change comparison with control mice, showing that the percentage of body weight loss was 7.96 ± 1.55%. However, administration with LT could attenuate the increase of body weight loss induced by T2DM, showing a significant decrease in body weight loss compared to the HFD group (2.71 ± 1.32% versus 7.96 ± 1.71%, P < 0.05). LT group also showed similar body weight with control group, the percentage of body weight loss were 2.71 ± 1.32% and 2.70 ± 1.11%, respectively.

Table 1 Effect of LT on body weight of mice during the experimental periods^a

Group	Diet	Body weight (g)				Percentage of weight loss (%)
		After 1 week establishment of T2DM (g)	After administration to mice (g)
		0	1 week	2 week	3 week
a Data are expressed as mean ± SEM, n = 10.* significantly different from the HFD group; # significantly different from control group. Significance = *P < 0.05, **P < 0.01, #P < 0.05, ##P < 0.01. The data were analyzed by a one-way ANOVA.
Control	Common diet	28.44 ± 3.04	26.11 ± 1.27	27.67 ± 2.50	27.67 ± 2.96	2.70**
HFD	High-fat diet	25.11 ± 1.69	23.44 ± 1.81	23.67 ± 2.12	23.11 ± 1.76	7.96##
DMBG	High-fat diet	24.71 ± 4.15	22.86 ± 3.72	22.83 ± 1.94	24.50 ± 1.76	0.80**#
XKW	High-fat diet	24.86 ± 1.77	23.83 ± 3.82	21.17 ± 1.17	23.50 ± 1.87	5.47*#
LT	High-fat diet	24.33 ± 2.73	22.00 ± 1.41	22.00 ± 1.10	23.67 ± 1.37	2.71**

---

Feed and water intake

The intake of feed and water is presented in Table 2. Regardless of the treatments, the intake of feed and water were higher in diabetic groups when compared with the control mice, especially during 2 week and 3 week (P < 0.05). The water intake in DMBG, XKW, and LT groups were decreased significantly compared to HFD group after 3 week treatment, and remained higher than that of control group during the experimental period. However, the feed intake in DMBG, XKW, and LT groups were decreased after 3 week treatment, and there were no significant differences compared to control group.

Table 2 Effect of LT on the average feed and water intake of normal and diabetic mice during the experimental periods^a

Group	Diet	Average feed and water intake
		Feed intake (g)			Water intake (g)
		1 week	2 week	3 week	1 week	2 week	3 week
a Data are expressed as mean ± SEM, n = 10.* significantly different from the HFD group; # significantly different from control group. Significance = *P < 0.05, **P < 0.01, ***P < 0.001, #P < 0.05, ##P < 0.01, ###P < 0.001. The data were analyzed by a one-way ANOVA.
Control	Common diet	4.96 ± 0.57***	5.84 ± 1.49***	7.80 ± 5.13***	3.50 ± 0.86	3.76 ± 1.12*	3.85 ± 0.26**
HFD	High-fat diet	11.61 ± 2.11###	13.43 ± 0.96###	18.42 ± 5.38###	3.68 ± 1.39	4.19 ± 0.50#	5.04 ± 0.22##
DMBG	High-fat diet	8.10 ± 1.09*##	9.90 ± 2.55*##	13.57 ± 8.37**##	3.08 ± 1.21*	3.17 ± 1.05**	4.42 ± 0.38*
XKW	High-fat diet	10.89 ± 2.20##	10.60 ± 1.02*##	18.97 ± 9.92###	3.12 ± 1.75*	2.89 ± 0.61**	4.58 ± 0.63*
LT	High-fat diet	10.44 ± 0.63*##	9.00 ± 0.89*##	14.43 ± 3.40**##	2.61 ± 1.37**#	2.83 ± 0.64***	4.28 ± 0.35*

---

Morris water maze test

The Morris water maze test is one of the most widely used behavioral tasks designed to study the psychological processes and neural mechanisms of spatial learning and memory. To determine the effect of LT on diabetes-induced cognition impaired, Morris water maze test was conducted in which the crossing times and escape latency were measured. The results of the present study showed that diabetic mice in HFD group exhibited a significant cognitive decline as evident from increased escape latency (35.34 ± 5.49 s, P < 0.01, Fig. 3A) compared with the control during the last day of test hidden sessions. HFD mice performed poorer during the probe trial following the last day of hidden platform training, with fewer crossings 4.60 ± 0.41 times over the target, suggesting deficits in spatial memory retention (Fig. 3B). LT treatment could significantly increase the number of times of crossing to 6.51 ± 0.52 over the platform site compared to HFD group (Fig. 3B, P < 0.05). No marked difference was found between LT and DMBG treated groups. The representative traces on the last day of hidden platform training showed in Fig. 3C.

Fig. 3 Effect of LT on cognitive decline caused by STZ administration (n = 10 each group). The Morris water maze test was carried out to test the spatial learning and memory ability of mice. (A) Mean escape latency time to reach hidden platform on the last day of hidden platform training. (B) The number of crossing over the exact location of the previous platform during the probe test. (C) The representative traces on the last day of hidden platform training. Control: the mice with normal blood glucose; HFD: STZ-induced diabetic mice; DMBG: STZ-induced diabetic mice treated melbine; XKW: STZ-induced diabetic mice treated Xiao-ke-wan; LT: STZ-induced diabetic mice treated LT. The results are presented as mean ± SEM, (n = 10). * significantly different from the HFD group; # significantly different from control group. Significance = *P < 0.05, ***P < 0.001, ##P < 0.01, ###P < 0.001. The data were analyzed by a one-way ANOVA.

Fasting blood glucose and oral glucose tolerance test

To further assess the anti-diabetic impact of LT, the effect of LT on glycometabolism was evaluated with FBG values. As shown in Table 3, administration of STZ led to significant increase in FBG levels in the diabetic group as compared to the normal control group. STZ-diabetic mice treated with LT demonstrated a significant decrease in the FBG levels at the end of 3 weeks (Table 3). In addition, LT treatment showed the same therapeutic effect on FBG levels as the DMBG.

Table 3 Effect of LT on FBG of normal and diabetic mice during the experimental periods^a

Group	Diet	FBG (mM)
		Before administration to mice		After administration to mice
		0 week	1 week	2 week	3 week
a Data are expressed as mean ± SEM, n = 10.* significantly different from the HFD group; # significantly different from control group. Significance = *P < 0.05, **P < 0.01, ***P < 0.001, #P < 0.05, ##P < 0.01, ###P < 0.001. The data were analyzed by a one-way ANOVA.
Control	Common diet	6.4 ± 1.0***	4.22 ± 0.47***	5.50 ± 1.34*	5.60 ± 0.69***
HFD	High-fat diet	16.2 ± 2.5###	16.46 ± 1.53###	8.98 ± 2.63#	20.68 ± 3.67###
DMBG	High-fat diet	15.8 ± 3.9###	8.7 ± 3.81***#	6.42 ± 2.12	15.38 ± 4.27*###
XKW	High-fat diet	16.1 ± 0.8###	10.07 ± 3.69**##	6.48 ± 2.20	14.73 ± 2.65***###
LT	High-fat diet	15.8 ± 0.4###	8.47 ± 3.97***#	6.97 ± 1.66	15.17 ± 4.97**###

---

OGTT were carried out at the last day of the experiment. The OGTT glucose curve demonstrated that the blood glucose levels in diabetic mice were higher than the levels in control group as measured during 0–180 min after oral intake of glucose (Fig. 4). However, these phenomena could be reversed by LT, DMBG, and XKW, as evidence by the lower levels of blood glucose measured at 30, 60, 120, and 180 min (Fig. 4A) and smaller areas under the blood glucose curve during 0–180 min (Fig. 4B). In the LT, DMBG, and XKW groups, the 180 min areas under the curve (AUC) of blood glucose decreased by approximately 14.76%, 23.07%, and 11.43%, respectively, when compared with value in HFD group. These results showed that LT treatment could significantly improve the glucose tolerance of diabetic mice (Fig. 4).

Fig. 4 Effect of LT on (A) OGTT and (B) AUC in normal and diabetic mice. Control: the mice with normal blood glucose; HFD: STZ-induced diabetic mice; DMBG: STZ-induced diabetic mice treated melbine; XKW: STZ-induced diabetic mice treated Xiao-ke-wan; LT: STZ-induced diabetic mice treated LT. The results are presented as mean ± SEM, (n = 10). * significantly different from the HFD group; # significantly different from control group. Significance = *P < 0.05, **P < 0.01, #P < 0.05, ##P < 0.01. The data were analyzed by a one-way ANOVA.

Effect of LT on serum insulin in diabetic mice

Diabetic mice showed a substantial increase in insulin level compared with the control group (P < 0.05). Treatment with LT significantly attenuated the increase in insulin level compared with the HFD group (0.54 ± 0.12 mM versus 0.82 ± 0.20 mM, P < 0.01). Additionally, LT group exhibited the similar effect as the positive control group DMBG (Fig. 5).

Fig. 5 Effect of LT on serum insulin levels (mIU L⁻¹) in normal and diabetic mice. Control: the mice with normal blood glucose; HFD: STZ-induced diabetic mice; DMBG: STZ-induced diabetic mice treated melbine; XKW: STZ-induced diabetic mice treated Xiao-ke-wan; LT: STZ-induced diabetic mice treated LT. The results are presented as mean ± SEM, (n = 6). * significantly different from the HFD group; # significantly different from control group. Significance = *P < 0.05, **P < 0.01, #P < 0.05. The data were analyzed by a one-way ANOVA.

Effect of LT on blood lipids profiles in diabetic mice

Fig. 6 presents the values of the lipid metabolic parameters after 3 week of LT administration. The higher levels of TC (5.61 ± 0.66 mM), TG (2.47 ± 0.48 mM), and LDL-C (1.58 ± 0.55 mM) in STZ-diabetic mice were observed. Meanwhile, the level of HDL-C (1.85 ± 0.28 mM) was significantly lowed in STZ-diabetic mice. At the termination of 3 weeks of LT treatment, the elevated levels of TC, TG, LDL-C and decreased level of HDL-C in STZ-diabetic mice were significantly reversed, showing the values were 4.18 ± 0.23 mM (P < 0.05), 1.43 ± 0.28 mM (P < 0.01), 1.21 ± 0.12 mM (P < 0.05), and 3.39 ± 0.24 mM (P < 0.01), respectively.

Fig. 6 Effect of LT on serum (A) TG, (B) TC, (C) HDL-C, and (D) LDL-C levels (mM) in normal and diabetic mice. Control: the mice with normal blood glucose; HFD: STZ-induced diabetic mice; DMBG: STZ-induced diabetic mice treated melbine; XKW: STZ-induced diabetic mice treated Xiao-ke-wan; LT: STZ-induced diabetic mice treated LT. The results are presented as mean ± SEM, (n = 6). * significantly different from the HFD group; # significantly different from control group. Significance = *P < 0.05, **P < 0.01, ***P < 0.001, #P < 0.05, ##P < 0.01, ###P < 0.001. The data were analyzed by a one-way ANOVA.

Effect of LT on protein expression in pancreas

The expressions of Bax, Bcl-2, and caspase-3 in pancreas of control and experimental mice by immunohistochemistry are illustrated in Fig. 7. The pancreas showed upregulated expression of Bax in HFD induced mice as compared to normal mice. Diabetic mice treated with LT showed markedly decrease of the Bax immunostaining when compared with diabetic mice. However, there were no obvious changes of Bcl-2 and caspase-3 expression in control, HFD, and LT treated mice pancreas.

Fig. 7 Immunohistochemistry results for Bcl-2, Bax, and caspase-3 staining. Pancreas cross-sectional tissue slices were stained with Bcl-2, Bax and caspase-3 specific primary antibody, then corresponding secondary respectively and diaminobenzidine (DAB) (brown). Control: the mice with normal blood glucose; HFD: STZ-induced diabetic mice; DMBG: STZ-induced diabetic mice treated melbine; XKW: STZ-induced diabetic mice treated Xiao-ke-wan; LT: STZ-induced diabetic mice treated LT.

Mouse pancreas histological changes

In the study, we performed histological examinations to determine whether LT was capable of protecting pancreatic β-cells in diabetic mice. The histopathology of pancreatic tissue was shown in Fig. 8. Normal pancreatic architecture was noted in control group, showing round or oval, dispersed in pancreatic acinar, and had clear boundaries of islet cell clusters, whereas HFD group showed atrophy of the islets. Administration of LT to diabetic mice could alleviate the abnormality caused by diabetes, showing remarkable expansion of islets and significantly reduced injuries in pancreas.

Fig. 8 Representative changes of the mice pancreas and photomicrograph of H&E staining after 3 weeks of treatment. Normal pancreatic architecture was noted in control group, whereas HFD group showed atrophy of the islets. Administration of LT to diabetic mice could alleviate the abnormality caused by diabetes. Control: the mice with normal blood glucose; HFD: STZ-induced diabetic mice; DMBG: STZ-induced diabetic mice treated melbine; XKW: STZ-induced diabetic mice treated Xiao-ke-wan; LT: STZ-induced diabetic mice treated LT.

Mouse brain histological changes

Finally, we examined the density of neurons in the CA1 region of the hippocampus by counting the number of Nissl bodies, a marker for mature neurons. The Nissl staining analysis showed that the Nissl-stained slices of mouse brain showed intense, rich Nissl bodies in the control group, with no obvious abnormalities in neurons. HFD caused typical neuropathological changes in the CA1 region of hippocampus, including neuron loss and nucleus shrinkage or disappearance. Furthermore, the Nissl-positive cell numbers per 0.01 mm² of the CA1 region of the hippocampus significantly decreased in HFD group compared to control group. Treatment with LT remarkably decreased the neuropathological changes and increased the density of healthy neurons in the CA1 regions of the hippocampus. No differences in neuronal density were observed between LT and DMBG groups in the CA1 region. These findings indicated that LT protected against neuronal apoptosis in the hippocampus of mice (Fig. 9).

Fig. 9 Representative changes of the mice brain and photomicrograph of Nissl staining in the CA1 region of hippocampus. Nissl-positive cell densities were determined by the Nissl staining in the hippocampus CA1 region of mice brain. Extensively damaged neurons in CA1 were observed in the HFD group and the number of surviving neurons was reduced compared with the control group. Control: the mice with normal blood glucose; HFD: STZ-induced diabetic mice; DMBG: STZ-induced diabetic mice treated melbine; XKW: STZ-induced diabetic mice treated Xiao-ke-wan; LT: STZ-induced diabetic mice treated LT. The results are presented as mean ± SEM, (n = 6). * significantly different from the HFD group; # significantly different from control group. Significance = *P < 0.05, ***P < 0.001, #P < 0.05, ###P < 0.001. The data were analyzed by a one-way ANOVA.

Discussion

HFD feeding and STZ-injection have been documented to induce insulin resistance, hyperglycemia and hyperinsulinemia in mice.¹³ The present work was designed to testify the potential anti-diabetic activity of LT by using HFD feeding and STZ-induced diabetic mouse model. To the best of our knowledge, this is the first study to provide direct evidence to show that LT improves glucose homeostasis in diabetic mice. The present results indicated that administration of LT reduced the FBG level in diabetic mice induced by HFD feeding and STZ-injection. The results also displayed that LT could lower blood glucose in diabetic mice and the glucose-lowering properties of LT were partly mediated by improvements of metabolic function.

Diabetes adversely impacts the brain, resulting in increased risk for cognitive decline. Recent findings demonstrate that diabetes impairs learning memory, synaptic plasticity, and adult neurogenesis in both insulin-deficient and insulin-resistant rodents.^14,15 It has been reported that peripheral administration of STZ induces a selective destruction of pancreatic beta cells. The selective toxicity of STZ for the beta cells results from its similar chemical structure with glucose molecule, which enables STZ to enter into the cells through GLUT-2.¹⁶ In fact, GLUT-2 is distributed in the brain of mice and is involved in neurotransmitter release.¹⁷ It has also been shown that icv STZ injection induces brain atrophy, which primarily is related to neural, oligodendroglial cell death triggered by apoptosis, the mitochondria inefficiency, inflammation and oxidative stress.¹⁸ The present study demonstrates that LT can alleviate cognitive decline induced by STZ. This observation has been further strengthened by the decrease of escape latency and increase of platform crossing times in Morris water maze test and the reduction of brain cell damage in the hippocampus by Nissl staining. These results suggest that LT can alleviate the cognition decline triggered by diabetes.

Diabetes is associated with weight loss. Elevated muscle wasting and decrement of tissue proteins may have given rise to body weight loss in STZ-treated mice.¹⁹ Therefore, weight loss is a very common condition in diabetes. In our study, the weight loss was observed in the HFD group. Administration with LT could attenuate the increase of body weight loss induced by T2DM, showing a significant decrease in body weight loss compared to the HFD group. The reversal of weight loss in LT-treated diabetic group indicates that the restorative effect of LT may be by the reversal of gluconeogenesis and glycogenolysis. It has been documented that a higher intake of feed and water, increase in body weight, and larger quantity of urine excretion are clear manifestations of metabolic changes in diabetic mice.²⁰ The present results indicated that administration with LT significantly lowered the feed and water intake of STZ-induced diabetic mice. In addition, a decreasing trend was found in urine secretion in the LT group (data did not show in this paper). These results suggest a certain role of LT in the management of diabetic state.

Impaired OGTT is an important criterion for T2DM.²¹ During the period of experiment, the FBG values in LT group (30–180 min) were significantly decreased, further supporting the hypoglycaemic effect of LT. This hypoglycaemic effect of LT was also confirmed by the OGTT test, showing that LT improved glucose tolerance. In addition, the serum insulin concentration was also dramatically reduced after LT treated. These results indicate that LT may have better hypoglycaemic effect in use for a longer term.

Dyslipidemia is a common characteristic of T2DM and is the primary cause of cardiovascular disease in people with diabetes.²² It has been demonstrated that the abnormality of insulin in diabetes leads to a variety of disorders in metabolic and regulatory processes, which in turn leads to accumulation of lipids such as TC and TG in diabetic patients. Abnormalities in lipid profile are one of the most common complications in diabetes mellitus found in 40% of diabetic subjects.²³ Diabetic dyslipidemia is featured with elevated TG, TC, LDL-C and decreased HDL-C. These changes impose an increased risk for coronary heart disease in patients with diabetes mellitus.²⁴ It has been documented that LDL-C positively and HDL-C negatively correlates with risk of cardiovascular disease.²⁵ In our present study, TC, TG, and LDL-C levels increased and HDL-C level decreased in the HFD group. Administration of the LT to the STZ-diabetic mice significantly improved TG, TC, HDL-C, and LDL-C towards normalcy. The observed antihyperlipidemic effect may be due to decreased cholesterogenesis and fatty acid synthesis, and this may be also attributed to the enhanced glucose utilization.

Bax and Bcl-2 participate in apoptosis regulation, where Bax promotes apoptosis and Bcl-2 has an anti-apoptotic function.²⁶ Immunohistological staining of Bax, Bcl-2, and caspase-3 in pancreas were performed to further clarify the effect of LT on islet cell apoptosis. In our experimental diabetic mouse model, Bax was strikingly accumulated in the pancreas of STZ-diabetic mice but was expressed at lower levels in the pancreas of STZ-diabetic mice treated with 3 weeks of LT. However, there is almost no effect on the expression of Bcl-2 and caspase-3 after LT treatment. The results of histopathology suggested that LT protected brain and pancreas against damages related to STZ induced hyperglycemia in mice.

Conclusion

In conclusion, this study demonstrated the antidiabetic effect of LT, which significantly improved glucose intolerance, dyslipidemia in HFD and STZ-induced type 2 diabetic mice. Based on these results, we hypothesize that the mechanism of antidiabetic effect of LT was inhibiting the apoptosis of islet cell and regulating the lipid metabolic disorder. The results provide a sound rationale for future clinical trials of the oral administration of LT for the primary prevention of T2DM.

Conflict of interest statement

We have no conflict of interest in this research.

Abbreviation

AUC	Areas under the curve
DAB	3,3-Diaminobenzidine tetrahydrochloride
DMBG	Melbine
ELISA	Enzyme-linked immunosorbent assay
FBG	Fasting blood glucose
HDL-C	High density lipoprotein
H&E	Hematoxylin and eosin
HFD	High-fat diet
LDL-C	Low density lipoprotein
LT	Ligustilide-rich total lactones
OGTT	Oral glucose tolerance test
STZ	Streptozotocin
TC	Total cholesterol
T2DM	Type 2 diabetes mellitus
TG	Triglycerides
XKW	Xiao-ke-wan

Acknowledgements

The work was supported by Special Financial Grant from the China Postdoctoral Science Foundation (No. 2015T81140), PhD Research Startup Foundation of Logistics University of Chinese People's Armed Police Forces (No. WHB201509) and Science and Technology Support Program Foundation of Tianjin China (No. 15CZDSY01020).

References

1. X. Zhang, X. Fu, D. Zhu, C. Zhang, S. Hou, M. Li and X. Yang, Eur. J. Pharmacol., 2016, 779, 46–52.
2. R. Ambady, C. Snehalatha, A. Samith Shetty and A. Nanditha, Diabetic Med., 2013, 30, 26–34.
3. S. H. Ley, O. Hamdy, V. Mohan and F. B. Hu, Lancet, 2014, 383, 1999–2007.
4. N. M. Maruthur, M. O. Gribble, W. L. Bennett, S. Bolen, L. M. Wilson, P. Balakrishnan, A. Sahu, E. Bass, W. L. Kao and J. M. Clark, Diabetes Care, 2014, 37, 876–886.
5. O. J. Phung, W. L. Baker, V. Tongbram, A. Bhardwaj and C. I. Coleman, Ann. Pharmacother., 2012, 46, 469–476.
6. B. Luna and M. Feinglos, Am. Fam. Physician, 2001, 63, 1747–1756.
7. W. Li, H. Zheng, J. Bukuru and N. De Kimpe, J. Ethnopharmacol., 2004, 92, 1–21.
8. L. Huo, J. Zhang, Z. Qu, H. Chen, Y. Li and W. Gao, J. Ethnopharmacol., 2015, 173, 352–360.
9. K. Wang, P. Cao, W. Shui, Q. Yang, Z. Tang and Y. Zhang, Food Funct., 2015, 6, 902–909.
10. Q. H. Yang, Y. Liang, Q. Xu, Y. Zhang, L. Xiao and L. Y. Si, Phytomedicine, 2011, 18, 1148–1152.
11. M. Ohnishi, T. Matuo, T. Tsuno, A. Hosoda, E. Nomura, H. Taniguchi, H. Sasaki and H. Morishita, BioFactors, 2004, 21, 315–319.
12. Z. Qu, H. Yang, J. Zhang, L. Huo, H. Chen, Y. Li, C. Liu and W. Gao, Neurochem. Res., 2016, 41, 2199–2214.
13. S. Holowacz, C. Guigné, G. Chêne, S. Mouysset, A. Guilbot, C. Seyrig and M. Dubourdeau, PharmaNutrition, 2015, 3, 101–107.
14. X. Li, S. Aou, Y. Oomura, N. Hori, K. Fukunaga and T. Hori, Neurosci., 2002, 113, 607–615.
15. A. M. Stranahan, T. V. Arumugam, R. G. Cutler, K. Lee, J. M. Egan and M. P. Mattson, Nat. Neurosci., 2008, 11, 309–317.
16. H. A. Elsner, G. Schmitz, M. Ballas, V. Lenhard and R. Blasczyk, Tissue Antigens, 2000, 56, 463–466.
17. M. Arluison, M. Quignon, P. Nguyen, B. Thorens, C. Leloup and L. Penicaud, J. Chem. Neuroanat., 2004, 28, 117–136.
18. A. Bahramian, K. Rastegar, M. R. Namavar and M. Moosavi, Behav. Brain Res., 2016, 311, 247–254.
19. A. Dogan, I. Celik and M. S. Kaya, J. Ethnopharmacol., 2015, 176, 243–251.
20. Y. Xue, Q. Miao, A. Zhao, Y. Zheng, Y. Zhang, P. Wang, H. Kallio and B. Yang, J. Funct. Foods, 2015, 16, 223–233.
21. P. Chen, Q. Zhang, H. Dang, X. Liu, F. Tian, J. Zhao, Y. Chen, H. Zhang and W. Chen, Nutrition, 2014, 30, 1061–1068.
22. A. G. Bertoni, W. G. Hundley, M. W. Massing, D. E. Bonds, G. L. Burke and D. C. Goff, Diabetes Care, 2004, 27, 699–703.
23. S. Sirasanagandla, R. B. Kasetti, A. N. Shaik, R. Natava and V. P. Surtineni, Eur. J. Med. Chem., 2013, 66, 400–406.
24. M. N. Woo, S. H. Bok, M. K. Lee, H. J. Kim, S. M. Jeon, G. M. Do, S. K. Shin, T. Y. Ha and M. S. Choi, J. Med. Food, 2008, 11, 169–178.
25. Y. Liu, J. Sun, S. Rao, Y. Su, J. Li, C. Li, S. Xu and Y. Yang, Food Chem. Toxicol., 2013, 62, 285–291.
26. M. Fuentes, M. Andrews and M. Arredondo-Olguín, Biol. Trace Elem. Res., 2013, 153, 390–395.

---

Footnote

† These two authors contributed equally to this work.

---