Antihyperglycaemic and organic protective effects on pancreas, liver and kidney by polysaccharides from Hericium erinaceus SG-02 in streptozotocin-induced diabetic mice

The present work was designed to investigate the antihyperglycaemic and protective effects of two Hericium erinaceus intracellular polysaccharide (HIPS) purified fractions (HIPS1 and HIPS2) from mycelia of H. erinaceus SG-02 on pancreas, liver and kidney in streptozotocin (STZ)-induced diabetic mice. The supplementation of HIPS1 and HIPS2 significantly decreased the blood glucose (GLU) levels; suppressed the abnormal elevations of alkaline phosphatase (ALP), alanine aminotransferase (ALT), aspartate aminotransferase (AST), urea nitrogen (BUN) and creatinine (CRE) levels in serum; improved the antioxidant enzymatic (superoxide dismutase (SOD), glutathione peroxidase (GSH-Px) and catalase (CAT)) activities; and attenuated the pathological damage to these organs. The HIPS1 showed superior effects in antihyperglycaemia and organic protection than HIPS2 possible owing to the abundant functional groups (-NH2, -COOH and S=O) in HIPS1, indicating that H. erinaceus SG-02 could be used as a functional food and natural drug for the prevention of diabetes and its complications.


Ultraviolet (UV) spectroscopy, Gas chromatography (GC) and Fourier transform infrared (FT-IR)
analysis of polysaccharides. The UV scanning spectrums of HIPS1 and HIPS2 showed no absorption at 280 and 260 nm, indicating the polysaccharides were purified and free of nucleic acids and proteins.
The monosaccharide compositions of HIPS1 and HIPS2 are shown in Fig. 2. HIPS1 was composed of xylose (Xyl) (3.02%), mannose (Man) (4.36%), galactose (Gal) (9.41%) and glucose (Glc) (83.21%), with molar ratios of 1:1.2:2.6:23.1 (Fig. 2B), while HIPS2 contained Man, Gal and Glc, with mass percentages of 23.40%, 50.34% and 26.27%, and molar ratios of 1:2.15:1.12 (Fig. 2C), respectively. FT-IR spectra of HIPS1 and HIPS2 are shown in Fig. 2. It is obvious that both HIPS1 and HIPS2 exhibited typical major -OH broad stretching bands for carbohydrates at approximately 3500-3400 cm −1 . The absorption peaks at 1637.85 and 1637.37 cm −1 are attributed to the stretching vibration of C=O. The bands at approximately 1618 cm −1 indicate the presence of -NH 2 in the two samples. The absorption peak for HIPS1 at 2030.49 cm −1 is a characteristic absorption peak for C≡C 26 , and the bands at approximately 1450 cm −1 indicate the presence of -COOH groups in HIPS1 27 . The band at approximately 1153 cm −1 suggests valent vibrations of the C-O-C bond and a glycosidic bridge in HIPS1 23 , while α-glucan 28 had an additional band at 1026 cm −1 . The characteristic absorption peak at 866.03 cm −1 indicates the presence of a furan ring in HIPS1 13 , while the additional weak peak at approximately 620.83 cm −1 describes S=O vibrations in HIPS1 26, 29 . Inhibitory effects on α-amylase and α-glucosidase activities. The inhibitory effects of HIPS1 and HIPS2 on α-amylase and α-glucosidase activities were investigated in the present work. As shown in Fig. 3, the inhibitory activities of HIPS2 on α-amylase and α-glucosidase were lower than those of HIPS1. The α-amylase inhibition rate of HIPS1 (53.27 ± 2.16%) was maximal at a concentration of 5 mg/mL, which was 32.09% higher than that for HIPS2. The 50% inhibitory concentration (IC50) value for HIPS1 on α-amylase activity was 3.42 ± 0.29 mg/mL, and the inhibition rate declined at concentrations greater than 5 mg/mL. At a concentration of 6 mg/mL, the α-glucosidase inhibition rate of HIPS1 was 36.64 ± 1.83%, which was 14.21% higher than that for HIPS2.
Effects of HIPS1 and HIPS2 on blood glucose (GLU) levels, body weights, and pancreas, liver and kidney indices. The results for GLU levels, body weight, and pancreas, liver and kidney indices in STZ-induced diabetic mice are listed in Table 1. Before treatment, the GLU levels of the diabetic mice were markedly increased than those of normal mice (Table 1, P < 0.05). After treatment, the GLU levels in the six dose groups were significant decreased than that in the MC groups, indicating that the pathological increases could be suppressed by HIPS1 and HIPS2 administration at different doses (P < 0.05). Specifically, compared with the MC group, the final GLU levels in the H-HIPS1, M-HIPS1, L-HIPS1, H-HIPS2, M-HIPS2 and L-HIPS2 groups were reduced by 51.65 ± 9.11, 38.77 ± 5.48, 34.21 ± 6.54, 37.05 ± 6.80, 32.56 ± 9.67 and 28.93 ± 5.94%, respectively. Additionally, both HIPS1 and HIPS2 showed effects in reducing GLU levels but the values remained different in comparison with that in NC groups among the test doses.
The body weights of pre-treatment and post-treatment diabetic mice were also determined. At the onset of the experiments, there was no significant difference (P < 0.05) between the body weights of normal and diabetic mice, while the diabetic mice without any treatment had emaciated bodies compared with the seven treatment groups at the end of the study. Specifically, the body weights of mice in the NC groups were significantly (P < 0.05) increased during the treatment. However, the diabetic mice in the MC group exhibited significant weight loss compared with the NC group (Table 1, P < 0.05). When compared with that in the MC groups, the mice in H-HIPS1, M-HIPS1, L-HIPS1, H-HIPS2 and M-HIPS2 groups showed significant improvements on the body weights (P < 0.05). And the results also indicated that HIPS1 showed stronger effects in improving the body weights, which was reflected by the paralleled values in comparison with that in the NC groups.
Furthermore, the indices for the pancreas, liver and kidneys in the mice were also investigated, and the results are listed in Table 1. Significant increases in the pancreas, liver and kidney indices were observed in the diabetic mice (MC) group compared with those in the NC group (P < 0.05). However, the three indices were decreased after treatment with HIPS1 and a high dose of HIPS2. Especially in the H-HIPS1 group, the pancreas, liver and kidney indices of the mice were reduced by 32.01 ± 2.16, 22.71 ± 2.94 and 39.11 ± 3.32%, respectively, compared with those in the MC group. However, HIPS2 was much less effective than HIPS1 in restoring the organ indexes, especially at the low dose.
Effects of HIPS1 and HIPS2 on serum biochemistry. The activities and levels of albumin (ALB), alkaline phosphatase (ALP), alanine aminotransferase (ALT), aspartate aminotransferase (AST), urea nitrogen (BUN) and creatinine (CRE) in serum are clinically used as biochemical markers for early liver and kidney damage. As shown in Table 2, mice exhibited liver and kidney damage after injection with STZ, as evidenced by significant increases in the serum levels of ALP, ALT, AST, BUN and CRE (P < 0.05) and decreased ALB levels. Dramatically, when compared with that in the MC groups, the activities of ALT, AST and ALP, as well as the levels of BUN and CRE were significantly (P < 0.05) decreased, while the ALB levels were increased by treatment with HIPS1 at the three tested dosages. However, only the high dose of HIPS2 (600 mg/kg) was effective in maintaining the above effectiveness. Particularly, the serum biochemical parameters for mice in the H-HIPS1 groups were similar to that in normal mice, indicating the high dose of HIPS1 was much more effective than the other doses of HIPS1 and HIPS2. In the present work, HIPS1 could suppress the abnormal elevations of ALP, ALT, AST, BUN and CRE levels and the decline in ALB levels in diabetic mice, indicating that the polysaccharides extracted from H. erinaceus SG-02 had potential protective effects against STZ-induced liver and kidney damage. Additionally, supplementation with glibenclamide could achieve effects similar to supplementation with HIPS1 ( Table 2).
Effects of HIPS1 and HIPS2 on GSH peroxide (GSH-Px), superoxide dismutase (SOD), catalase (CAT), and malondialdehyde (MDA). To study the effects of HIPS1 and HIPS2 administration on enzyme activities and free radical production, GSH-Px, SOD and CAT activities and MDA levels were measured Group GLU levels (mmol/L) Body weight (g) Pancreas index (g/100 g) Liver index (g/100 g)  Table 1. Effects of HIPS1 and HIPS2 on the GLU levels, body weights, as well as index of pancreas, liver and kidney. The values are reported as the means ± SD (n = 10 for each group). Means with the same letter are not significantly different (P < 0.05). # Significant difference compared to Pre-treatment (P < 0.05).   Table 2. Effects of HIPS1 and HIPS2 on serum biochemistry in different groups. The values are reported as the means ± SD (n = 10 for each group). Means with the same letter are not significantly different (P < 0.05).
in the three organs ( Fig. 4). At the end of the experiment, the GSH-Px, SOD and CAT activities in STZ-induced diabetic mice (MC groups) were significantly reduced (P < 0.05), while MDA levels were markedly increased (P < 0.05) compared with those in the NC groups, indicating that the pancreas, liver and kidney all suffered oxidative stress. As shown in Fig. 4, the abnormal physiological and biochemical changes in the three organs of diabetic mice were restored by HIPS1 and HIPS2 administration. The GSH-Px, SOD and CAT activities in the liver reached maximum values of 72.75 ± 4.38, 158.68 ± 16.42 and 223.14 ± 26.12 U/mg protein, respectively, in the H-HIPS1 group, values that were significantly higher than those in the MC group (20.98 ± 3.26, 65.85 ± 5.31 and 74.96 ± 9.41 U/mg protein), respectively, (P < 0.05 for all), and with no significant difference compared to the NC group ( Fig. 4A-C). The hepatic MDA levels reached 5.98 ± 0.23 μmol/L in the H-HIPS1 group, which was reduced by 49.14 ± 6.72% compared with those in the MC groups (Fig. 4D). Similar results were also observed for GSH-Px, SOD and CAT activities in the kidney (Fig. 4E-G) and pancreas ( Fig. 4I-K). When compared with the MC group, significant (P < 0.05) increases were observed after the administration of HIPS1 (including the three doses) and HIPS2 (including the high dose and middle dose). The MDA levels for the kidney (Fig. 4H) and pancreas (Fig. 4L) also declined after treatment with the polysaccharides. Glibenclamide-treated mice also manifested significant increases in SOD, GSH-Px, and CAT activities, and decreases in the MDA levels in the three organs compared with diabetic mice (P < 0.05).
The relevant indices in the H-HIPS1 group were approximately equal to those in the NC group. This conclusion demonstrated that the high dose of HIPS1 exhibited strong potential antioxidant effects against oxidative stress in the liver, kidney and pancreas.
Histopathological observations of the pancreas, liver and kidney. Histopathological observations of the pancreas, liver and kidney were performed using haematoxylin and eosin (HE) staining (Figs 5-7).
The effects of HIPS1 and HIPS2 on pancreatic tissues are shown in Fig. 5. Severe degeneration in the pancreas appeared in the STZ-induced diabetic mice, including lymphocyte infiltration, hypochromatosis and the disappearance of cell borders in the pancreatic islets (Fig. 5B). After treatment with HIPS1 at all doses and HIPS2 at the high and middle doses, improvement in the pancreatic tissues could be observed, which could be evidenced by decreased infiltration and more integrated cell structure in the pancreatic islets compared with the mice in the MC group (Fig. 5D-I). These results indicated that treatment with polysaccharides from H. erinaceus SG-02 could repair islet damage and improve the structural integrity of pancreatic islet beta-cells and tissues.
As shown in Fig. 6A, the hepatocyte morphology was normal, and the hepatic cells were arranged in a regular manner with abundant cytoplasm, distinct cell borders and normal central nuclei. In the MC group, liver damage was observed in the mice treated with STZ. The changes in hepatocyte morphology were characterized by perivascular inflammatory cell infiltration, fatty degeneration, hypochromatosis and the disappearance of cell borders (Fig. 6B). In the dosage groups, treatment with HIPS1 (600 and 400 mg/kg) and HIPS2 (600 mg/kg) showed effects in ameliorating the liver damage (Fig. 6D,E and G). Especially for mice in the H-HIPS1 groups, the liver structures were similar to normal liver architectures (Fig. 6D).
The changes in kidney cortex histology are shown in Fig. 67. Compared with the mice in the NC groups with normal architecture (Fig. 7A), glomerular augmentation, glomerular swelling, mesangial matrix proliferation and loss of interstitial space were observed in mice in the MC group after STZ injection (Fig. 7B), indicating that kidney damage occurred in the diabetic mice. After the treatments with HIPS1 at high and middle doses, these pathological changes were attenuated, which were evidenced by the normal morphology of glomerulus in Fig. 7D and C.
Interestingly, the histopathological analysis of the pancreas and liver was consistent with the results of enzyme activities and MDA levels as biochemical indicators of their function (Table 2 and Fig. 4). Histological studies of the kidney cortices of mice in the H-HIPS1 and PC groups demonstrated similarities to those in the NC group, which echoed the results of the biochemical assays (Table 2 and Fig. 4).
Acute toxicity. In the acute toxicity assays, the mice treated with HIPS1 and HIPS2 did not exhibit any gross behavioural changes (including irritation, restlessness, respiratory distress, abnormal locomotion and catalepsy) or toxic symptoms (such as decreased food intake or shaggy hair) either immediately or during the post-treatment period compared with the control group, indicating that these polysaccharides were essentially non-toxic substances.

Discussion
Increasing data indicate enhanced oxidative stress and changes in antioxidant capacity in both clinical and experimental forms of DM that are key mechanisms in the pathogenesis of diabetic complications 30 . STZ injection has been used to establish an experimental hyperglycaemia model, owing to its high toxicity in stimulating the generation of H 2 O 2 in β-cells, which may induce lipid peroxidation and the depletion of antioxidant enzymes, and even worse, organic damage [31][32][33] . Previous publications have reported decreases in antioxidant enzyme activities, including GSH-Px, SOD and CAT, in the pancreas, liver and kidney of diabetic mice, in agreement with our present conclusions 34 . Interestingly, the reductions in enzyme activities and the increase of MDA levels were markedly suppressed after treatment with HIPS1 or HIPS2 (Fig. 4), demonstrating that the polysaccharides may inhibit oxidative damage to pancreatic, hepatic and nephritic tissue. Pancreas, one of the most vital organs of polysaccharide metabolism, maintains homeostasis of blood glucose through insulin synthesis. STZ exhibits toxicity on pancreatic β-cells, leading to pancreatic damage, and results in disordered insulin secretion 32,33 . As sensitive organs, the liver and kidneys have a great capacity to detoxify toxic substances, and they play pivotal roles in glucose metabolism in vivo 31 . Therefore, the pathological damages that are inflicted on the liver and kidneys by hepatotoxic and nephrotoxic agents appear to be fatal in diabetes 35,36 . Furthermore, in diabetic animals, the oxidative stress that is induced by free radicals can affect the functions of the pancreas, liver and kidneys 37 . In the present work, as shown in Figs 5-7, the STZ-induced diabetic mice exhibited serious damage to the pancreas, liver and kidneys, including cell necrosis, hypochromatosis, fatty degeneration and glomerular proliferation. Interestingly, this damage was considerably recovered by HIPS1 and HIPS2 administration, indicating that the damage to the pancreas, liver and kidneys in STZ-induced diabetic mice can be protected and repaired by intervention with HIPS1 and HIPS2.
High blood glucose levels are the main characteristic of DM 1 . In STZ-induced diabetic mice, high GLU levels and increased uptake of food and water indicate the successful establishment of the model 38 . One therapeutic approach for decreasing hyperglycaemia is to retard the absorption of glucose by inhibition of carbohydrate hydrolysis enzymes, such as α-amylase and α-glucosidase. Hence, it is important to investigate the inhibition of α-amylase and α-glucosidase to analyse the antihyperglycaemic effects 19,39 . In the present study, HIPS1 and HIPS2 produced inhibitory effects against the two enzymes at different concentrations, and HIPS1 was more effective than HIPS2, apparently due to the different structures of the polysaccharides. Nevertheless, Chiba noted that the inhibitory effects may be associated with the individual structures of the enzymes 40 .
Blood has direct contact with most of the tissues in the animal body, and pathological changes are likely to be reflected in levels of some proteins and enzymes in blood samples 21 . Enhanced ALB levels and decreased activities of ALP, ALT and AST in serum have been used clinically as biochemical markers for liver damage 41,42 . BUN and CRE, endogenous by-products that are released into body fluids and excreted by glomerular filtration, have been widely used in the clinic to reflect the physical status of the kidneys 43 . In the present study, changes in the activities of ALB, ALP, ALT, and AST and in serum levels of BUN and CRE were suppressed by HIPS1 and HIPS2 administration, indicating the stabilization of plasma membranes and repair of STZ-induced pancreatic, hepatic and nephritic damages. These results are also consistent with our previous studies 8 .
The present findings suggest that HIPS1 and HIPS2 from H. erinaceus SG-02 are novel bioactive compounds with potential anti-diabetic and organic protection effects. Moreover, HIPS1 was much more efficient than HIPS2 in preventing the diabetes-induced organs damages, with the possible mechanism might be due to the monosaccharide composition and bond types. Figure 2 shows that HIPS1 contains Xly and a high percentage of Glc in the monosaccharide composition compared with HIPS2. Furthermore, HIPS1 contains more active radicals and functional groups, such as -NH 2 , -COOH and S=O, than HIPS2. These structural characteristics might contribute to the scavenging capacities of ROS. Moreover, HIPS1 and HIPS2, especially HIPS1, could indirectly relieve oxidative damage to the pancreas, liver and kidneys by improving the antioxidant enzyme activities and reducing MDA levels (Fig. 4), possibly resulting in the amelioration of tissue necrosis and inflammatory damage in these organs. Otherwise, the repair capacities of HIPS1 and HIPS2 on the organs may involve a reduction of insulin secretion or stimulation of insulin release 44 and may indirectly participate in inflammatory pathways 45 , possibilities that require further study.

Methods
Ethics statement. All experiments were performed in accordance with the guidelines and regulations of the ethics committee of the Shandong Agricultural University and the Animals (Scientific Procedures) Act of 1986 (amended 2013).
All experimental protocols were submitted to and approved by the ethics committee of the Shandong Agricultural University in accordance with the Animals (Scientific Procedures) Act of 1986 (amended 2013).

Materials. The H. erinaceus SG-02 strain used in this experiment was preserved in the Fungi and Application
Laboratory of Shandong Agriculture University (Taian, China) and was maintained on a potato dextrose agar (PDA) slant. The diagnostic kits for analysing SOD, GSH-Px, and CAT activities and MDA levels were purchased from the Nanjing Jiancheng Bioengineering Institute (Nanjing, China). The standard monosaccharide samples, including rhamnose (Rha), ribose (Rib), arabinose (Ara), Xyl, Glc, Man and Gal, were provided by the Merck Company (Darmstadt, Germany) and the Sigma Chemical Company (St. Louis, MO, USA). All other reagents and chemicals used in the present work were analytical reagent grade and were supplied by local chemical suppliers.

Isolation and purification of polysaccharides.
H. erinaceus SG-02 was initially maintained on PDA slants and then inoculated into 500 mL of liquid medium containing glucose (10 g), peptone (1.5 g), yeast extract, (2 g), KH 2 PO 4 (1 g) and MgSO 4 (0.5 g). After 25 days of fermentation, the mycelia of H. erinaceus SG-02 were collected and cleaned three times with distilled water. The dry mycelia were shattered and extracted with distilled water at 80 °C for 3 h and centrifuged at 3000 rpm for 10 min, after which the precipitate was discarded. The supernatant was mixed with anhydrous ethanol (4 °C overnight) to obtain the crude polysaccharide, which was deproteinized using Sevag reagent (chloroform/n-butanol, 5:1, v/v) and labelled as HIPS.
HIPS (1 g) was dissolved in 15 mL of distilled water and purified on a DEAE-cellulose column (26 × 400 mm) at a flow rate of 2.0 mL/min and eluted with sodium chloride solutions at concentrations of 0, 0.2, 0.3 and 0.5 mol/L. The major eluate was then collected separately and purified by gel permeation chromatography (Sephadex G-100 column AQ3, 1.3 × 50 cm). The above steps were repeated twenty times, and the main fractions were lyophilized by vacuum freeze-drying (Labconco, USA) for further study.
UV spectroscopy, GC and FT-IR analysis of polysaccharides. The UV scanning spectrum was determined by microplate spectrophotometer (Dynex Spectra MR, Dynex Technologies, USA) The monosaccharide composition was determined by GC (GC-2010, Shimadzu, Japan) according to the method of Sheng, et al. 46 . The monosaccharide compositions were identified by comparing the retention times with the standards, and the monosaccharide contents were calculated using area normalization methods.
Infrared spectra of the samples were recorded using an infrared spectrometer (Nicolet 6700, Thermo Fisher Scientific, USA) with a range of 4000-400 cm −1 47 .
Inhibition assay on α-amylase and α-glucosidase activities. The HIPS fractions were stored at −80 °C prior to the assay of α-amylase inhibition. The reaction mixture, including 0.2 mL of α-amylase (6 U/ mL) and 0.2 mL of polysaccharide solution (1.0 to 6.0 mg/mL, dissolved in 0.2 mol/L phosphate buffer, pH 6.6), was activated by adding the starch substrate solution (0.4 mL, 1%, w/w) and was processed at 37 °C for 10 min, followed by termination with 2 mL of DNS reagent (1% 3,5-dinitrosalicylic acid and 12% sodium potassium tartrate in 0.4 mol/L sodium hydroxide) 48 . The sample was maintained in boiling water for 10 min and then diluted with 15 mL of distilled water in an ice bath. The α-amylase activity was determined by measuring the absorbance at 540 nm. The IC50 value was defined as the concentration of polysaccharide that produced an inhibition rate of 50% under the assay conditions 19 . Inhibition of α-glucosidase activity was measured according to reported methods with some modifications 19 . The α-glucosidase (0.2 mL, 3.75 U/mL) was premixed with the samples (dissolved in 0.2 mol/L phosphate buffer, pH 6.8) at various concentrations (0.5 to 6.0 mg/mL), and 6 mmol/L p-nitrophenyl-α-D-glucopyranoside (in phosphate buffer) as a substrate was added to the mixture to initiate the reaction. The reaction was incubated at 37 °C for 30 min and stopped by adding 6 mL of 0.1 mol/L sodium carbonate. The α-glucosidase activity was determined by measuring the absorbance at 400 nm. Animal experiments. Ninety Kunming strain mice (4 weeks old, weight: 20 ± 2 g, purchased from Taibang Biological Products Ltd. Co. (Taian, China)) were acclimated for one week under the following conditions: temperature (23 ± 1 °C), humidity (55 ± 5%) and a 12-h light-dark cycle, during which time they had free access to food and water ad libitum. After the acclimation, the mice were induced to become diabetic by an intraperitoneal injection with 120 mg/kg STZ (freshly dissolved in sodium citrate buffer, pH 4.5), while the NC group was injected intraperitoneally with citrate buffer 49 . The GLU levels in the blood samples obtained from the tail of each mouse were measured 48 h after the last STZ injection using an ACCU-CHEK blood glucose metre (Roche, Basel, Switzerland). Mice with GLU levels greater than 13.3 mmol/L were considered successful models of diabetes 50  The treatments lasted for two weeks. At the end of the experiment, all mice were weighed and sacrificed using a diethyl ether anaesthetic treatment after fasting overnight 51 .
The organic index for the pancreas, liver and kidneys was calculated as follows: (organ weight)/(body weight) (g/100 g) 29 .
ALT, AST and ALP activities, along with the levels of ALB, BUN and CRE, were measured using an automatic biochemical analyser (ACE, USA).
The pancreas, liver and kidneys were removed rapidly, weighed, and homogenized (1:9, w/v) in phosphate-buffered solution (PBS, 0.2 mol/L, pH 7.4). After centrifugation at 4 °C for 20 min (5000 rpm), the supernatants were collected and assayed for the hepatic/nephritic activities of GSH-Px, SOD, and CAT and MDA levels according to the commercial kit instructions 7 .
Fresh pancreas, liver and kidney tissue masses were fixed in 4% formaldehyde solution (pH 7.4) for more than 24 h, embedded in paraffin and then cut into slices using a microtome. The slices were stained with HE and photographed using an optical microscope (400 × magnification) 7 .
Acute toxicity. Acute toxicity was determined by the method of Chao et al. 52 . Briefly, fifteen Kunming strain mice (4 weeks old, weight: 20 ± 2 g) were randomly divided into one control group, which received the normal saline solution (0.9%), and two dosage groups, which received isometric HIPS1 or HIPS2 solutions at the final concentration of 2000 mg/kg by gavage. The mice were observed continuously for gross behavioural changes, toxic symptoms and mortality during the 14-day feeding period.
Statistical analysis. Data were statistically analysed using two-way ANOVA and the T-test (SAS, USA).
Data were expressed as the means ± SD (standard deviation). Differences were considered statistically significant if P < 0.05.

Conclusions
The present study demonstrated that HIPS1 and HIPS2 from H. erinaceus SG-02 exerted protective effects on the pancreas, liver and kidney in STZ-induced diabetic mice based on the results of biochemical and histopathological analyses. Especially, the HIPS1 was much more efficient in preventing the diabetes-induced organ damages owing to the functional groups (-NH 2 , -COOH and S=O). These results may provide a mechanistic basis for the use of H. erinaceus SG-02 as potentially natural and functional foods and drugs for the prevention and alleviation of diabetes and its complications.