Guan Xin Dan Shen dripping pills (40 mg/pill; cat. no. 20160426) were obtained from Harbin Pharmaceutical Group Co., Ltd. In our previous study, the aforementioned components of GXDSF were determined via HPLC (22). Metformin was provided by Sino-American Shanghai Squibb Pharmaceuticals, Ltd. Valsartan was provided by Novus Biologicals, Ltd.

Mice were obtained from GemPharmatech Co., Ltd. In total, 60 leptin receptor-deficient db/db 8-week-old male mice (weight, 40±2 g; BKS.Cg^+/+ Leprdb NJU) and 10 non-diabetic 8-week-old male mice (weight, 20±2 g; C57BLKS/J) were used in the present study. The mice were maintained at 22–24°C and 40% humidity with 12-h light/dark cycles and ad libitum access to food and water. All animal experiments were performed in accordance with the Guide for the Care and Use of Laboratory Animals (30). The present study was approved by the Laboratory Animal Ethics Committee of the Institute of Medicinal Plant Development, Peking Union Medical College and Chinese Academy of Medical Sciences (license no. SLXD-20190406017).

After 4 weeks of acclimatization, the db/db mice were randomly divided into the following six groups (n=10 per group): i) db/db group (model); ii) db/db + metformin (140 mg/kg/day) group; iii) db/db + valsartan (48 mg/kg/day) group; iv) db/db + GXDSF high dose (180 mg/kg) group; v) db/db + GXDSF moderate dose (90 mg/kg) group; and vi) db/db + GXDSF low dose (45 mg/kg) group. The control group consisted of the C57BLKS/J mice (n=10). The clinic dose of GXDSF for patients with coronary heart disease is 20 mg/kg day. The doses of GXDSF used in the present study were based on the recommended clinical dose of GXDSF (31). Over a period of 10 weeks, the mice were administered drugs via gavage daily (Fig. 1A). Mice in the control and model groups were orally administered distilled water.

The blood glucose of mice fasted overnight was measured using an automatic glucometer (Roche Diagnostics). For assessing insulin tolerance, an ITT assay was performed via intraperitoneal injection of 1 U/kg insulin (Sigma-Aldrich; Merck KGaA). The tail blood glucose concentration was measured every 30 min after injection for a total of 120 min, and is expressed as the area under the curve (AUC).

Following treatment for 10 weeks, the mice were anesthetized with 1.2% avertin (300 mg/kg; intraperitoneal) and echocardiography was performed. The echocardiograms were obtained using a Vevo⁷⁷⁰ high-resolution imaging system (VisualSonics, Inc.). To calculate percentage ejection fraction (% EF) and percentage fractional shortening (% FS), M-mode tracing of the left ventricle (LV) based on the parasternal long-axis view was used. To evaluate left ventricular end diastolic diameter (LVIDd) and left ventricular end systolic diameter (LVIDs), the pulse-wave and tissue Doppler in mice at baseline and after treatment were used. LV end-diastolic volume (LVVd) and LV end-systolic volume (LVVs) were automatically calculated by an ultrasound machine (VisualSonics, Inc.).

After 10 weeks of treatment, the db/db mice were anesthetized using 4% chloral hydrate (400 mg/kg; intraperitoneal). Subsequently, ~0.75 ml blood was collected from the eye socket vein of mice fasted overnight. The anesthetized mice were then sacrificed via cervical dislocation. To obtain the serum samples, the blood was centrifuged at 3,000 × g for 15 min at room temperature. An automatic biochemical analysis system (Beckman Coulter, Inc.) was used to measure serum triglyceride (TG, cat. no. A110-1-1), low-density lipoprotein (LDL, cat. no. A113-1-1) and total cholesterol (TC, cat. no. A111-1-1) levels according to the manufacturer's protocol (Nanjing Jiancheng Bioengineering Institute).

The serum myocardial enzyme activities of aspartate transaminase (AST; cat. no. C010-1-1), lactate dehydrogenase (LDH; cat. no. A020-1-2) and creatine kinase MB (CK-MB; cat. no. E006-1-1) were measured using specific detection kits (Nanjing Jiancheng Bioengineering Institute) according to the manufacturer's protocol.

After the mice were sacrificed, five heart tissue samples in each group were fixed in 4% buffered paraformaldehyde for 48 h at room temperature. Following embedding in paraffin, tissues were cut into 4-µm sections. Subsequently, sections were stained with hematoxylin and eosin (H&E) at room temperature as previously (32). Stained sections were examined by a blinded pathologist using a BX53 light microscope (Olympus Corporation; magnification, ×200).

The aforementioned fixed and embedded heart samples were cut into 4-µm sections and then deparaffinized using xylene. The antigen retrieval was performed by heating the tissues to 80°C. Subsequently, the sections were incubated with Protease K for 15 min at room temperature before pre-incubation with terminal deoxynucleotidyl transferase buffer (Sigma-Aldrich; Merck KGaA) at 37°C for 60 min. After washing with PBS, the sections were incubated with an anti-digoxin and anti-serum alkaline phosphatase complex (Sigma-Aldrich; Merck KGaA) at 37°C overnight. Following washing in Tris buffer, sections were counterstained with hematoxylin at 25°C for 2 min and washed again in Tris buffer. In total, five visual fields were randomly selected in each group to observe the apoptosis under a light microscope (BX53; Olympus Corporation; magnification, ×200). The results were quantified using Image-Pro Plus software (version 6.0; Media Cybernetics, Inc.).

Total RNA was extracted from heart tissues using TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc.). Total RNA purity and concentration were assessed using a Nanodrop 2000 Spectrophotometer (Thermo Fisher Scientific, Inc.). Total RNA (1 µg) was reverse transcribed into cDNA using PrimeScript RT MasterMix (37°C for 15 min; 85°C for 5 sec, held at 4°C; Takara Biotechnology Co., Ltd.). Subsequently, qPCR was performed to measure the mRNA expression levels of atrial natriuretic peptide (ANP), brain natriuretic peptide (BNP), α-myosin heavy polypeptide (MHC), β-MHC, Bcl-2, Bax, caspase-3, caspase-9 and 18s using a CFX-96 Touch Thermocycler PCR system and SYBR Green Master Mix. The thermocycling conditions used for qPCR were as follows: Initial denaturation at 94°C for 30 sec, followed by 45 cycles of 94°C for 5 sec, 58°C for 30 sec and 72°C for 30 sec, then a dissociation stage (95°C for 15 sec, 65°C for 30 sec and 95°C for 15 sec). The sequences of the primers used for qPCR are listed in Table I. mRNA expression levels were quantified using the 2^−ΔΔCq method (33) and normalized to the internal reference gene 18s.

A standard western blotting protocol was used to measure protein expression, as previously described (32). Briefly, cytoplasmic and nuclear proteins were extracted from heart tissues using a protein extraction kit (cat. no. CW0199S; CoWin Bioscience Co., Ltd.). Protein concentrations were measured using a BCA protein quantification kit (CWBio). Proteins (50 µg) were separated via 12% SDS-PAGE and transferred to nitrocellulose membranes. Following blocking in 5% fat-free milk in TBS-0.1% Tween-20 (TBST) for 2 h at room temperature, the membranes were incubated at 4°C overnight with primary antibodies targeted against: Bcl-2 (1:500; cat. no. sc-7382; Santa Cruz Biotechnology, Inc.), Bax (1:500; cat. no. sc-7480; Santa Cruz Biotechnology, Inc.), cleaved caspase-3 (1:1,000; cat. no. ab214430; Abcam), cleaved caspase-9 (1:1,000; cat. no. 9509; Cell Signaling Technology, Inc.), phosphorylated (p)-Akt (1:1,000; cat. no. 4060; Cell Signaling Technology, Inc.), Akt (1:1,000; cat. no. 9272; Cell Signaling Technology, Inc.), Nrf2 (1:500; cat. no. sc-13032; Santa Cruz Biotechnology, Inc.), HO-1 (1:1,000; cat. no. ab68477; Abcam), NAD(P)H quinone oxidoreductase-1 (NQO-1; 1:500; cat. no. sc-32793; Santa Cruz Biotechnology, Inc.), γ-GCS (1:500; cat. no. sc-166603; Santa Cruz Biotechnology, Inc.), lamin B1 (1:500; cat. no. sc-374015; Santa Cruz Biotechnology, Inc.) and β-actin (1:1,000; cat. no. CW0096M; CWBio). Following washing with TBST, the membranes were incubated with a HRP-conjugated secondary antibody (1:10,000; anti-rabbit IgG, cat. no. 31460 or anti-mouse IgG, cat. no. 31430; Thermo Fisher Scientific, Inc.) at room temperature for 2 h. Protein bands were visualized using SuperSignal™ West Pico PLUS chemiluminescent substrate (cat. no. 34580; Thermo Fisher Scientific, Inc.). Protein expression was semi-quantified using Image Lab software (version 3.0; Bio-Rad Laboratories, Inc.). β-actin and lamin B1 were used as the loading controls for cytoplasmic and nuclear proteins, respectively.

All experiments were performed ≥3 times. Data are presented as the mean ± SD. Statistical analyses were performed using SPSS software (version 19.0; IBM Corp.). Comparisons among multiple groups were analyzed using one-way or mixed two-way ANOVA follow by Tukey's post hoc test. P<0.05 was considered to indicate a statistically significant difference.

The body weight of db/db mice was measured over the 10-week treatment period. Compared with the model group, the body weight of the mice in the GXDSF high-dose groups was reduced, but this trend was not significant (Fig. 1B). Compared with the model group, metformin significantly decreased elevated FBG levels in db/db mice (P<0.05; Fig. 1C), whereas the effect of GXDSF on FBG was not significant compared with the model group, which suggested that GXDSF displayed no distinct hypoglycemic effect in db/db mice. Subsequently, insulin sensitivity following GXDSF treatment was assessed. The ITT results demonstrated that both moderate- and high-dose GXDSF enhanced insulin sensitivity, as indicated by significantly decreased AUC values compared with the model group (P<0.01; Fig. 1D and E).

Compared with the control group, db/db mice displayed significantly increased serum lipid levels in the model group (Fig. 2). Administration of metformin or high-dose GXDSF significantly reduced serum TG and TC levels in db/db mice compared with the model group (P<0.05; Fig. 2A and B). Additionally, increased serum LDL levels in db/db mice were significantly reduced by metformin or valsartan administration compared with the model group (both P<0.05; Fig. 2C). Serum LDL levels in the GXDSF groups were slightly reduced, but the differences were not significant compared with the model group.

After 10 weeks of GXDSF treatment, M-mode echocardiography was used to assess cardiac function. Compared with the control group, db/db mice in the model group displayed a significantly larger LVIDd (Fig. 3A). Administration of valsartan or high-dose GXDSF significantly reduced LVIDd compared with the model group (P<0.01). Compared with the model group, the effects of metformin and moderate- and low-dose GXDSF on the LVIDd and LVIDs were not significant (Fig. 3B and C). Additionally, compared with the control group, the LV end-systolic volume and LV end-diastolic volume of the db/db mice in the model group were significantly reduced (P<0.01), whereas valsartan, moderate-dose GXDSF and high-dose GXDSF treatment significantly increased these volumes in db/db mice (P<0.05; Fig. S1A and B). Additionally, compared with the model group, valsartan or GXDSF treatment significantly improved the impaired left ventricular EF and FS in db/db mice (P<0.05; Fig. 3D and E).

After GXDSF treatment, the heart weight was initially measured. Compared with the control group, the heart weight of the model group was significantly increased, which was significantly reduced by treatment with metformin, moderate-dose GXDSF or high-dose GXDSF (P<0.05; Fig. 4A and B). In the model group, the db/db mice developed notable myocardial hypertrophy, as evidenced by H&E staining, which demonstrated disordered cardiac fibers (Fig. 4C). However, compared with the model group, treatment with metformin or GXDSF for 10 weeks markedly reduced myocardial hypertrophy in db/db mice. Moreover, compared with the control group, the mRNA expression levels of cardiac hypertrophy-related genes ANP, BNP and β-MHC/α-MHC were significantly increased in the heart tissues of the model group (P<0.01), which was significantly reversed by treatment with metformin, valsartan or GXDSF (P<0.05; Fig. 4D). Furthermore, treatment with moderate- or high-dose GXDSF significantly decreased the serum levels of CK-MB and LDH compared with the model group (Fig. 4E). Additionally, high-dose GXDSF could reduce the increased serum AST level. Collectively, the aforementioned results indicated that GXDSF prevented diabetes-induced cardiac hypertrophy.

Diabetes-induced cardiomyocyte apoptosis is closely associated with cardiac hypertrophy (34). The TUNEL assay results demonstrated that there was a significantly higher proportion of apoptotic cells in the hearts of the model mice compared with the control group (Fig. 5A and B). After 10 weeks of treatment with metformin or GXDSF, the ratio of apoptotic cells was notably decreased. The mRNA and protein expression levels of apoptosis-related markers were further examined. In line with the results of the TUNEL assay, compared with the model group, GXDSF significantly decreased the mRNA expression levels of Bax, caspase-3 and caspase-9 (Fig. 5D and E), but significantly upregulated the mRNA expression levels of Bcl-2 (Fig. 5C). The western blotting results revealed that the Bcl-2/Bax ratio significantly decreased, whereas the protein expression levels of cleaved caspase-3 and cleaved caspase-9 were significantly upregulated in the heart tissues of db/db mice in the model group compared with the control group (Fig. 5G-I). Diabetes-induced alterations in protein expression were significantly reversed by GXDSF treatment (Fig. 5G-I), indicating that GXDSF protected the diabetic heart against apoptosis.

The PI3K/Akt signaling pathway and Nrf2-mediated antioxidant enzymes serve key anti-apoptotic roles (35), and our previous study demonstrated that GXDSF increased Akt phosphorylation (22). Therefore, it was hypothesized that GXDSF may protect against diabetes-induced cardiomyocyte apoptosis by increasing Akt phosphorylation and Nrf2 signaling. The western blotting results demonstrated that Akt phosphorylation, Nrf2 nuclear translocation and the expression of antioxidant enzymes (γ-GCS, NQO-1 and HO-1) were significantly decreased in the heart tissues of db/db mice in the model group compared with the control group (Fig. 6A-F). After treatment with GXDSF, the expression levels of p-Akt, nuclear Nrf2, HO-1 and NQO-1 were significantly upregulated compared with the model group. The results suggested that GXDSF promoted Akt and Nrf2 signaling, which subsequently upregulated the expression of antioxidant enzymes (Fig. 6G).