Cur was purchased from Vientiane Tianjin HengYuan Technology Co., Ltd., calcitriol (Cal) from Roche Diagnostics (Shanghai) Co., Ltd. and streptozotocin (STZ) from Sigma-Aldrich (Merck KGaA). High-sugar and high-fat fodder (57.3% carbohydrate, 20% fructose, 10% lard, 2.5% cholesterol, 10% egg yolk and 0.2% sodium cholate) was purchased from the Animal Experimental Center of Southern Medical University (Guangzhou, China).

TC (cat. no. A111-1-1), TG (cat. no. F001-1-1) and low-density lipoprotein cholesterol (LDL-C; cat. no. A113-2-1) assay kits and serum osteocalcin (OCN; cat. no. H152) and C-terminal type-I peptide (CTX-I; cat. no. H287) ELISA kits were purchased from Nanjing Jiancheng Bioengineering Institute. Smad2/3 (cat. no. TA347074; 1:1,000) and phosphorylated (p)-Smad2/3 (cat. no. TA501728; dilution, 1,000) antibodies were purchased from Beijing Zhongshan Golden Bridge Biotechnology Co., Ltd.

The animal experiments were approved by the Ethics Committee of Zhaoqing Medical College. A total of 50 male Sprague-Dawley rats (age, 8 weeks; weight; 180±20 g) were obtained from the Southern Medical Experimental Animal Center (certification no. SCXK 2017-0012). The animals were allowed to acclimatize to the laboratory conditions for 7 days before experiments and were housed at 25±2˚C and a relative humidity of 60-70% with 12 h light/dark cycles. The animals had access to dry pellet food and water ad libitum.

A total of ten rats were assigned to the control group and were administered a high-sugar, high-fat diet only for 12 weeks. A total of 40 rats were used to establish the T2DM model. According to previous literature (14), these rats were administered a high-sugar, high-fat diet for 4 weeks. After 4 weeks, each rat received an intraperitoneal injection of 3% STZ (40 mg/kg) (15). Rats with fasting blood glucose (FBG) ≥7.0 mmol/l were selected for further study.

The rats continued to receive the high-sugar, high-fat diet throughout the course of the present study. The animals were divided into five groups (n=10/group) and received the following treatments for 8 weeks (Table I): Control, T2DM, T2DM treated with 110 mg/kg/day Cur (T2DM + Cur), T2DM treated with 0.045 µg/kg/day Cal (T2DM + Cal) and T2DM treated with 200 mg/kg/day metformin (T2DM + Met; cat. no. 317240; Sigma-Aldrich; Merck KGaA). As Cur and Cal are insoluble in water (16,17), their suspension was prepared with 1% sodium carboxymethyl cellulose(cat. no. 419273; Sigma-Aldrich; Merck KGaA). The control and T2DM groups were given 1% sodium carboxymethyl cellulose (5 ml/kg) daily by oral gavage and the T2DM + Cur, T2DM + Cal and T2DM + Met groups were given their respective drugs (5 ml/kg) by oral gavage. FBG and body weight were measured once weekly for 8 weeks.

At the end of the experiment, all rats were euthanized via cardiac puncture under anesthesia with intraperitoneal injection of 45 mg/kg sodium pentobarbitone. A total of 2 ml of serum was collected for biomarker assays. The left and right femora were dissected for bone biomechanical analysis and micro-CT analysis, respectively. The left proximal tibial metaphysis (PTM) was cut into decalcified 5 µm sections for immunohistochemistry (IHC) analysis. The right PTM was used for reverse transcription-quantitative PCR (RT-qPCR) analysis.

Blood samples (5 ml) were processed as previously described (12). Briefly, the serum was separated by centrifugation at 1,000 x g at room temperature for 15 min and stored at -80˚C. The serum lipids (TC, TG and LDL-C) and serum bone formation markers (OCN and CTX-I) were determined using ELISA kits (Nanjing Jiangcheng Biological Bioengineering), according to the manufacturer's protocol.

In the biomechanical bone test, the left femora from each group of rats was placed in 70% ethanol for preservation at room temperature. The liquid attached to the femur was blotted with medical gauze, followed by the experiment of bone biomechanical measurement. The Lloyd LR5K Plus Bone Biomechanical Detection system (LLOYD Instruments Ltd.; AMETEK Test & Calibration Instruments) was used to perform a three-point bending test to analyze the biomechanical properties of the femora. The left femur of each group of rats was placed into the machine at a loading speed of 2 mm/min and a span load of 20 mm. Each biomechanical property, including maximum load, breaking load, elastic load and the bone rigidity coefficient, was plotted corresponding to the load-displacement curve and calculated according to the respective equations that reflect bone stiffness (18).

The metaphyses of the right femurs were measured using a micro-CT (CT-40; Scanco Medical). The specific parameters to test micro-CT parameters were as follows: X-line energy 70 kVP, 114 µA, 500 sheet of 700-nm slices and 30 min scan time. Following the scan, 1.0 mm growth plates were selected from left distal femora. The 3D reconstruction of the femora were constructed according to the following conditions: A thickness of 3.0 mm bone was considered as the cancellous bone region of interest to draw a reconstruction line (19) for bone reconstruction and quantitative image analysis was performed to extract information using the software supplied with the micro-CT (version 4; SCANCO Medical AG). The physical parameters examined were as follows: Bone volume fraction (bone volume/total volume; BV/TV), connection density (Conn.D), trabecular number, (Tb.N), trabecular separation (Tb.Sp) and trabecular thickness (Tb.Th).

Gene expression was analyzed using RT-qPCR. Bone tissue from the right tibia (100 mg) was weighed and RNA was extracted using TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc.), according to the manufacturer's protocol. RT was performed to synthesize cDNA using a kit (cat. no. 6110A; Takara Bio, Inc.), according to the manufacturer's protocol. RT-qPCR was conducted using a SYBR green kit (cat. no. RR420A; Takara Bio, Inc.) on a QuantStudio 12K Flex RT PCR System (Thermo Fisher Scientific, Inc.), according to the manufacturer's protocol. The thermocycling conditions were as follows: Initial denaturation at 95˚C for 30 sec; 40 cycles of 95˚C for 5 sec, 60˚C for 45 sec, and the melting curve analysis at 95˚C for 5 sec and 60˚C for 60 sec. The primers used are listed in Table II. β-actin was used as the reference gene. The 2^-ΔΔCq method was used to quantify the mRNA expression. All data are presented as relative to the control group.

Left tibia specimens were fixed in 4% paraformaldehyde for 12 h at room temperature and decalcified for 4 weeks and embedded in paraffin. Sections (5 µm) were placed in a drying oven at 60˚C for 3 h, dewaxed and hydrated, and then washed with 1X PBS solution three times. After 5 min, the tissue specimens were placed in citrate buffer (pH 6.0) at 60˚C for 30 min for antigen retrieval, followed by washing with 1X PBS solution three times for 5 min. Specimens were then incubated with smad2/3 and p-smad2/3 (dilution, 1:200) at 4˚C overnight, followed by incubation with the corresponding secondary antibody (Goat anti-mouse; 1:20,000; cat. no. sc-2005; Santa Cruz Biotechnology, Inc.) at room temperature for 2 h. DAB staining was applied for 30 sec at room temperature and hematoxylin counterstaining for 2 min at room temperature, followed by drying and mounting. The stained sections were observed under a light microscope at a magnification of x200 and three fields of view from each bone tissue section were imaged with a Nikon Eclipse E400 camera (Nikon Corporation) with Picture Frame software (E400; Nikon Corporation). The images were analyzed and measured with Image-ProPlus software (version 6.0; Media Cybernetics, Inc.) to determine the mean optical density of the positive area of each specimen, as previously described (20).

Data are presented as mean ± standard deviation and were analyzed using SPSS software for Windows (version 16.0; SPSS, Inc.). All experiments were performed in triplicate. Statistical differences between groups were analyzed using one-way ANOVA and Tukey's honest significant difference post-hoc test. Mixed ANOVA was used to compare body weight and glucose level. P<0.05 was considered to indicate a statistically significant difference.

After the T2DM rat model was established, weight were measured once weekly every 8 weeks. Success of the model was determined by analyzing the body weight and FBG. According to the results, the weight of the rats in the T2DM group was significantly decreased compared with controls (P<0.05) and Cur, Cal and Met increased the body weight of T2DM rats compared with the T2DM group; however, there was no significance between these groups (Fig. 1A). FBG levels in T2DM rats were increased compared with controls (P<0.05), while Cur and Met significantly decreased FBG level compared with the T2DM group. However, there was no significant difference in the levels of of Cal on FBG compared with the T2DM group (P<0.05; Fig. 1B).

At the end of the experiments, serum lipid levels were determined and the results demonstrated that TC, TG and LDL-C were significantly increased following a high-fat, high-sugar diet in T2DM rats compared with controls (P<0.05; Fig. 2). Following treatment with Cur and Met, lipid levels significantly decreased compared with the T2DM group (P<0.05). Furthermore, no significant difference was reported following Cal treatment. These results indicated that Cur may regulate dysregulated lipid and glucose levels and that Met had similar effects.

Bone formation marker OCN and bone turnover marker CTX-I levels in serum were measured using respective ELISA kits. The results indicated lower bone formation and higher bone absorption activity in T2DM rats compared with controls (P<0.05). However, Cur and Cal reversed this effect by significantly increasing OCN and decreasing CTX-I levels compared with the T2DM group (P<0.05; Fig. 3). There was no significant effect of Met on OCN and CTX-I level compared with the T2DM group.

To assess the role of Cur in T2DM rat bones, the biomechanical properties of the femur were examined. The results demonstrated that elastic load, breaking load, maximum load and the bone rigidity coefficient of the T2DM group were significantly lower compared with controls (P<0.05; Fig. 4). Moreover, Cur- and Cal-treated groups had significantly improved biomechanical properties compared with the T2DM group. There was no significant effect of Met on biomechanical properties compared with the T2DM group.

Subsequently, micro-CT was used to create representative 3D reconstruction images to examine the trabecular microarchitecture of rat femora (Fig. 5A). Microarchitecture parameters (BV/TV, Conn.D, Tb. N, Tb and Th) in T2DM rats were significantly decreased compared with controls (P<0.05; Fig. 5B), indicating that the high-fat, high-sugar diet was detrimental to rat bone and that the T2DM rat model was successfully established. Following treatment with Cur, the parameters BV/TV, Conn.D, Tb.N and Tb.Th were significantly increased, while Tb.Sp was decreased compared with the T2DM group (P<0.05). Cal exerted similar effects to Cur in terms of bone biomechanical properties and bone microarchitecture. However, Met did not exhibit any significance between groups.

mRNA levels of TGFβ1, TβRI, TβRII and Smad2/3 were determined using RT-qPCR to examine the mechanism of action of Cur. The mRNA levels of TGFβ1, TβRI, TβRII and Smad2/3 in T2DM group were significantly decreased compared with controls (P<0.05) and Cur and Cal significantly increased TGFβ1, TβRI, TβRII and Smad2/3 mRNA expression levels in the right PTM compared with the T2DM group (P<0.05; Fig. 6) and Met significantly increased TβRII and Smad2 mRNA expression levels compared with the T2DM group (P<0.05; Fig. 6)

Furthermore, IHC was performed to evaluate Smad2/3 and p-Smad2/3 protein expression levels. The results indicated that Cur significantly increased Smad2/3 expression and promoted Smad2/3 phosphorylation (Fig. 7). These results suggested that may Cur protect bone via TGFβ/Smad2/3 signaling pathway activation.