Heimioporus retisporus fruiting bodies were purchased from Kunming, Yunnan Province. Ion exchange resins CM-Sepharose and DEAE-Sepharose were from General Electric Co. (United States). Metformin hydrochloride (MET) was from Beijing Coway Pharmaceutical Co. (China). L-arabinose, D-glucose, D-galactose, D-mannose, D-xylose, glucuronic acid, and galacturonic acid were from Dionex Ltd. (China), Streptozotocin (STZ) and TRIzol reagent were from Sigma-Aldrich (United States). Reverse transcription kit and polymerase chain reaction mix were from Mei5 Biotechnology Co. (Beijing, China). All other reagents used in this study were analytical grade.

Heimioporus retisporus Polysaccharide was extracted and purified as shown schematically in Figure 1. Fruiting bodies were dried at 45°C to constant weight, and ground into powder (mesh #40; particle size ∼420 μm) with a triturator. The powder was evenly dispersed in water at ratio 1:15 (w/v), the mixture was heated in a water bath 4 h at 95°C and centrifuged (10,000 rpm, 10 min), and supernatant was collected. Supernatant was mixed with ethanol at ratio 1:3 (v/v), mixture was kept 10 h at 4°C and centrifuged (10,000 rpm, 10 min), and sediment was collected as crude HRP (CHRP).

CHRP was dissolved in deionized water and deproteinized in n-butanol/chloroform (v/v; 4:1) for 4 h. After stratification, upper fraction was collected and dialyzed thoroughly against phosphate buffer (0.05 mol/L, pH 7.4) at room temperature. Dialyzed CHRP solution was subjected to DEAE-Sepharose chromatography, eluted with phosphate buffer, and unbound fraction (DHRP) was collected. DHRP was dialyzed against acetate buffer (0.05 mol/L, pH 4.6) at room temperature, subjected to CM-Sepharose chromatography, eluted with acetate buffer, and unbound fraction (CDHRP) was collected. CDHRP was subjected to gel filtration (Superdex 75 16/600 column, GE Healthcare, United States) using an AKTA purifier (Amersham Biosciences, Sweden), and eluted with 0.15 mol/L NaCl in 0.05 mol/L phosphate buffer (pH 7.5) at flow rate 0.5 mL/min. Fractions were collected with an automated fraction collector, and the peak with highest polysaccharide content (S2) was collected. Fraction S2 was dialyzed in deionized water for 48 h with a dialysis bag (MW cutoff 3.5 kDa), freeze-dried, and termed HRP (Figure 2).

Polysaccharide content was determined by DuBois’ method (20). 1.5 mL polysaccharide solution was mixed with 5 mL sulfuric acid, and the mixture was kept in boiling water bath for 20 min and then cooled to room temperature. Absorption value was determined by spectrophotometry at wavelength 490 nm, and carbohydrate concentration was estimated based on a standard curve.

Weight-average (M_w) and number-average (M_n) molecular weights of HRP were determined by gel permeation chromatography (GPC) using an Agilent 1200 HPLC system equipped with PL aquagel-OH 50 column (7.7 × 300 mm) and differential refractive index detector. Samples (5.0 mg) were dissolved in 1.0 mL phosphate buffer (0.2 mol/L, pH 7.5) containing 1.0 mL NaCl (0.02 mol/L), and filtered through a membrane (pore size 0.45 μm). For each run, 20 μL solution (0.1 mg HRP) was injected and eluted with phosphate buffer (flow rate 0.5 mL/min, 30°C). M_w and M_n values were estimated using a calibration equation based on PL pullulan polysaccharide standards.

Monosaccharide composition of HRP was analyzed by high-performance anion exchange chromatography (HPAEC) coupled with pulsed amperometric detector (PAD). Neutral sugars and uronic acids were released by hydrolysis (10% H₂SO₄, 2.5 h, 105°C). Acid hydrolysates of HRP were diluted and analyzed using HPAEC system (Dionex ISC 3000; United States) with PAD, AS50 autosampler, CarboPac PA20 column (4 × 250 mm, Dionex), and PA-20 guard column (3 × 30 mm). Standard solutions of L-arabinose, D-glucose, D-xylose, D-glucose, D-mannose, D-galactose, glucuronic acid, and galacturonic acid were used for calibration.

Fourier transform infrared spectroscopy was performed using Optik GmbH Tensor II system (Bruker, Germany). Spectra were recorded from 4,000 to 400 cm^–1, with resolution 4 cm^–1 and maximal source aperture (21).

∼40 mg HRP was dissolved in 0.55 mL chloroform-d (CDCl₃), and solution-state ¹H and ¹³C nuclear magnetic resonance (NMR) spectroscopy (Bruker system) were performed with parameters: spectral width 1,800 Hz for ¹H dimension and 10,000 Hz for ¹³C dimension; delay between transients 2.6 s; delay for polarization transfer corresponding to estimated average ¹H-¹³C coupling constant 150 Hz. Data were processed using Bruker Topspin-NMR software program (22).

Thermogravimetric analysis (TGA) and derivative thermogravimetry (DTG) were performed using simultaneous thermal analyzer (model STA449F3; Netzsch; Germany). ∼8 mg lyophilized HRP powder was placed in a platinum crucible under nitrogen atmosphere, and heated at rate 10°C/min in temperature range of 30–800°C (23). Data were analyzed using software program Origin 8.0.2.8.

Dried HRP samples were gold-coated by sputter-coater (model IB-3, EiKo, Japan), and morphological features were observed by scanning electron microscopy (model S-3400N, Hitachi, Japan) (accelerating voltage 10.0 kV; magnifications ×100, ×500, ×1,000, ×2,000; high vacuum conditions) as described previously (24).

Male SPF Balb/c mice (weight 20 ± 2 g) from Charles River (Beijing) were maintained in the Experimental Animal Public Service Platform at China Agricultural University (25 ± 2°C, humidity 50 ± 10%, 12 h light/12 h dark cycle), and fed normal chow diet ad lib. After 1 wk acclimination period, mice were i.p. injected with 1% STZ (40 mg/kg) for 5 days (25), and fasting blood glucose (FBG) was measured 24 h after the last injection. A successful model was defined as mice with FBG ≥ 11.0 mmol/L, stable for 1 week (26).

Group Blank consisted of five untreated normal mice. 25 diabetic mice were assigned randomly to one control and four experimental groups (intragastric administration; 4-wk feeding period) as follows:

Group Blank: blank, deionized water.

Group CK: deionized water; control.

Group Met: metformin (40 mg/kg); positive control.

Group HRP-20: HRP 20 mg/kg.

Group HRP-40: HRP 40 mg/kg.

Group HRP-80: HRP 80 mg/kg.

Body weight and FBG data were collected for the four experimental groups. At the end of experimental period, mice were sacrificed (cervical dislocation), heart, liver, spleen, and kidney were removed, connective tissue was cleaned and washed with saline to remove blood, and collected organs were weighed for calculation of organ indices, immediately frozen in liquid nitrogen, and stored at -80°C for further analysis. All experiments were approved by the Institutional Ethics Committee of China Agricultural University, and performed in accordance with International Standards and Ethical Guidelines for Animal Welfare.

Mice were fasted for 8 h, blood was extracted from tail vein, and the FBG was measured using express glucose meter (On Call).

For each of various visceral organs, an index was calculated by the formula:

Total RNA extraction from heart tissue was performed using TRIzol reagent. First-strand cDNA synthesis was performed using a commercial reverse transcription kit as per manufacturer’s instructions. Primer sequences used were as follows:

PCR procedure was: initial denaturation at 94°C for 4 min, 30 cycles of denaturation at 94°C for 30 s, annealing at 60°C (β-actin) or 62°C (IL-6, TNF-α) for 30 s, extension at 72°C for 30 s, final extension at 72°C for 10 min. Amplification products were confirmed by electrophoresis (1.0% agarose gels) and visualized by Gel Red staining (27).

Results were expressed as mean ± SE, and data were analyzed using software program SPSS 20.0 (IBM). Means were compared by one-way analysis of variance (ANOVA), with p < 0.05 as criterion for significant difference.

M_w and M_n of HRP were, respectively, 1,949 and 873.34 kDa, and polydispersity index (PDI, calculated as M_w/M_n) was 2.232. PDI reflects distribution of molecular weight in each polymer sample, and the low value indicates that chain lengths of HRP vary over a relatively narrow range of molecular weights.

Monosaccharide composition of HRP, determined by HPAEC/PAD analysis, is summarized in Table 1. The major component was glucose (49.0%), followed by mannose, galactose, glucuronic acid, and xylose (percentages ranging from 17.8 to 8.6%). Arabinose was a minor component (0.7%).

In the FT-IR spectrum of HRP (Figure 3), the absorption bands at 3,407 and 2,923 cm^–1 represent stretching vibrations of O-H and C-H groups of the sugar ring (28). The 1,652 cm^–1 band reflects C=O linkage (29), weak symmetric stretching band near 1,357 cm^–1 corresponds to carboxylate groups (30), 1,159 cm^–1 band reflects stretching of α-(1,4) glycosidic linkages (31), 1,071 cm^–1 band indicates that HRP sugar rings are pyranose rings (32), 855 cm^–1 peak represents α-glycosidic bonds (33), 943 cm^–1 band represents β-glycosidic bonds (34), and 530 cm^–1 band reflects in-plane C=O bending (35).

Structural features of HRP were elucidated by measuring 1-D NMR (¹H) and 2-D NMR (heteronuclear single quantum coherence; HSQC) spectra. In the ¹H NMR spectrum (Figure 4A), H-1 signals representing six residues were seen at 3.11, 3.17, 3.31, 3.51, 3.73, and 4.06 ppm. The first four (strong signals) indicate presence of β-D-glucose (36), and the latter two reflect β-D-mannose configured residues (37).

In the HSQC spectrum (Figure 4B), six cross-peaks (4.48/102.9, 3.36/72.8, 3.51/85.8, 3.34/68.0, 3.38/76.0, and 3.64, 3.58/60.4 ppm) were assigned, respectively, to H-1/C-1, H-2/C-2, H-3/C-3, H-4/C-4, H-5/C-5, and H-6(a), H-6(b)/C-6 of →3)-β-D-Glcp-(1→residues. ¹H/¹³C chemical shifts at 3.51/71.2, 73.1/3.62, 68.8/3.95, and 3.78/70.9 ppm were assigned to H-2/C-2, H-3/C-3, H-4/C-4, and H-6/C-6 of →6)-β-D-Galp-(1→residues, and cross-peaks at 3.43/69.8, 3.88/69.3, and 3.41/62.4 were assigned to H-2/C-2, H-4/C-4, and H-6/C-6 of →3)-β-D-Galp-(1→residues. Cross-peaks at 4.22/102.2, 3.13/73, 3.46/74.2, and 3.62/75.8 ppm were assigned to H-1/C-1, H-2/C-2, H-3/C-3, H-4/C-4 of →4)-β-D-xylan-(1→, and those at 4.86/97.2, 3.71/66.2, 3.58/68.0, 4.08/68.2, and 3.72/68.2 were assigned to H-1/C-1, H-3/C-3, H-4/C-4, H-5/C-5, and H-6/C-6 of →6)-β-D-Manp-(1→residues. Cross-peaks at 4.99/102.1 and 3.50/66.30 were assigned to →5)-α-L-Araf-(1→residues (36, 38–40). These findings are consistent with those for monosaccharide composition and ¹H spectra.

Weight loss (TG) and DTG curves of samples are shown in Figure 5. The TG curve shows two stages (30–130 and 165–540°C) of weight loss. The first stage (∼9%), resulting from vaporization and removal of bound water in HRP, reflects characteristic moisture sorption based on abundance of hydroxyl radical. The second (degrading) stage, involving alteration of functional groups and depolymerization of structure, resulted in substantial loss (∼54%) of sample weight. The two curves indicate onset degradation temperature (T₀) = 200°C and maximum degradation temperature (T_max) = 263°C.

Analysis of surface morphology by SEM is a qualitative tool to characterize polysaccharides. SEM images of HRP (magnifications ×100, ×500, ×1,000, ×2,000) (Figure 6) demonstrate an entanglement structure with irregular sheets and coils. The tangled structure reflects the complex nature of HRP (41). The apparent pores may be artifacts of the freeze-drying process.

Figure 7 illustrates in vivo antidiabetic effect of HRP. FBG levels on day 28 were significantly lower for Groups HRP-40 and Met than for Group CK. Group HRP-20 had a striking reduction in FBG level between days 0 and 7. No such reduction was observed for Group HRP-80, suggesting that the effect of HRP was not dose-dependent (Figure 7A). No notable effects on body weight were observed for the experimental groups (Figure 7B).

Visceral organ indices did not differ significantly for Group CK vs. the three HRP groups, except for heart (data not shown). Heart indices are shown in Figure 7C. Heart index for Group CK was higher than those of all the other groups, and the difference was significant for Groups Met and HRP-40. RT-PCR assays for inflammatory cytokines IL-6 and TNF-α showed that IL-6 expression levels in heart tissue were higher for Group CK than for Group Met or the HRP groups, while TNF-α expression levels were similar for all groups. These findings are consistent with the high heart index for Group CK (Figure 7D).