Male Sprague Dawley (SD) rats were obtained from the Animal Center of Nanchang University (Nanchang, China). The experiments were performed in compliance with the ARRIVE Guidelines on Animal Research (15). All the procedures were approved by the Institutional Animal Care and Use Committee of Nanchang University.

Neonatal SD rats (70–80 g) were anesthetized and the hearts were removed and retrogradely perfused with Dulbecco’s modified Eagle’s medium (DMEM; Gibco, Carlsbad, CA, USA) for 5 min through the ascending aorta to remove blood cells. Following the removal of connective tissue, the remaining left ventricles were separately finely minced and digested with 0.2% collagenase I (Sigma, St. Louis, MO, USA) in Hank’s balanced salt solution for 30 min at 37°C in a shaking water bath. Trypsin (0.02%; Gibco) was then added and the mixture was incubated for an additional 5 min. The digested solution was filtered through an 100-μm mesh filter. The filtrate was collected and the cells were plated on culture dishes coated with human fibronectin (Invitrogen, Carlsbad, CA, USA) and maintained in DMEM supplemented with 20% fetal bovine serum (FBS; HyClone, Logan, UT, USA), human vascular endothelial growth factor (Invitrogen), human fibro-blast growth factor (Invitrogen) and human EGF (Invitrogen). After a 3-day culture, unattached cells were removed and fresh medium was added to the adherent cells. The medium was replaced once every 3 days. The cells were cultured to 80% confluence before being released with EDTA (Sigma) and subcultured. Some of the adherent cells at passage 3 were collected for immunofluorescence analyses, which was performed by staining using factor VIII (Solarbio, Beijing, China).

A recombinant lentivirus containing human NRG-1 (pLV-hNRG-1) was constructed. Briefly, a full-length hNRG-1 gene cDNA was cloned into the lentivirus shuttle plasmid vector pGC-FU, which contains a cytomegalovirus promoter and a polyadenylation signal of bovine growth hormone. For the construction of lentivirus containing green fluorescent protein (GFP), a shuttle vector containing human phosphoglycerate kinase gene promoter was used. The control virus lacking the hNGR-1 gene was separately prepared. Recombinant lentivirus was generated by homologous recombination and propagated in 293T cells (Genechem, Shanghai, China). At 48 h after transduction, the supernatant from the 293T cells was collected and purified by cesium chloride density gradient centrifugation and stored in 10 mmol/l Tris-HCl (pH 7.4), 1 mmol/l MgCl₂, and 10% (v/v) glycerol at −70°C. Virus titers were determined by a plaque assay on 293T cell monolayers.

For transduction, CMECs at passage 3 were plated at a density of 1×10⁵/well into 6-well dishes with DMEM/20% FBS. The cells at each well were incubated with 20 MOI NRG-1 lentivirus (NRG-1-CMEC) or 20 MOI GFP lentivirus (GFP-CMEC) for 72 h. Lentiviral infection was validated by visualization of enhanced GFP under a fluorescence microscope (Nikon, Tokyo, Japan). Subsequently, the medium was replaced with fresh DMEM/20% FBS and the cells were cultured for an additional 48 h. At the end of the incubation period, the media and cells in each well were collected and analyzed.

The levels of hNRG-1 in the media were analyzed by enzyme immunoassay using a human neuregulin-1 ELISA kit according to the manufacturer’s instructions (PlantSelect Biotechnology Systems Ltd., Dartmouth, NS, Canada). Data were expressed as the mean ± SEM.

Rat hearts were surgically removed from 1- to 3-day-old SD rats, washed instantly with phosphate-buffered saline (PBS) solution, and then minced into 1- to 3-mm³ pieces. The minced tissue was subjected to 6–8 cycles of proteolytic dissociation by magnetic stirring (10 min, 37°C) in 0.06% trypsin solution. The supernatants from each cycle were pooled and centrifuged. The cell pellet was resuspended in DMEM supplemented with 20% FBS. Selective adhesion was achieved by incubation at 37°C for 1.5 h in a humidified atmosphere (5% CO₂ and 95% air) in order to obtain a high purity of cardiomyocytes. Subsequently, 0.1 mM bromodeoxyuridine (Sigma) was added to the medium for the first 48 h of culture to inhibit the growth of fibroblasts.

Cardiomyocytes were plated into 12-well dishes at a cell density of 2×10⁴/well and randomly allocated into three groups. The cells in group A were cultured in fresh DMEM/20% FBS (control); the cells in group B were cultured in 50% fresh DMEM/20% FBS and 50% the medium previously harvested from GFP-CMECs; and the cells in group C were cultured in 50% fresh DMEM/20% FBS and 50% the medium previously harvested from hNRG-1-CMECs. After 3 days of culture, some wells of cardiomyocytes in each group were counted. The number of cardiomyocytes was independently determined by investigators blinded to the type of cell culture using a hemacytometer.

The remaining wells of cardiomyocytes of the three groups were incubated with 20 ng/ml tumor necrosis factor-α (TNF-α) (Peprotech, Rocky Hill, NJ, USA). The cardiomyocytes were harvested 24 h later, stained with Annexin V-APC/propidium iodide (PI; KeyGen Biotech, Nanjing, China) and subjected to flow cytometric analysis to assess apoptosis following the manufacturer’s instructions.

SD rats at a postnatal age of 6 weeks (body weight, 200–220 g) were allocated to the control (n=8) and diabetic groups (n=30). Diabetes was induced by the intraperitoneal (i.p.) injection of STZ (50 mg/kg; Sigma, L’Isle d’Abeaux, France) (16). Tail vein blood glucose was measured every 3 days during the first week; the rats with plasma glucose levels ≥16.7 mmol/l were considered to be diabetic. Concurrently, control rats were injected i.p. with 1 ml/kg body weight 20 mmol/l citrate buffer (pH 4.5) vehicle. The control and diabetic rats both raised on standard food and water for the whole experimental period. Twelve weeks after the induction of diabetes, 24 diabetic rats were used for further analysis; the remaining six rats died or were excluded due to unsuccessful induction of diabetes. The 24 diabetic rats were randomly allocated into three groups: the hNRG-1, GFP and DCM groups (n=8 rats per group). Subsequently, all the diabetic rats were anesthetized with 4% chloral hydrate solution (1 ml/100 g) by i.p. injection and put on an animal ventilator. Thoracotomy was performed. Approximately 50 μl/heart (5×10⁷ TU/ml) of hNRG-1-lentivirus (hNRG-1 group) or GFP-lentivirus (GFP group) or an equivalent volume of PBS alone (DCM group) was injected at five sites in the left ventricles of the rats using a 30-gauge needle. Following the injection of lentiviral vectors, the rats continued to be raised on standard food and water for 4 weeks. All analyses were performed 16 weeks after the induction of diabetes.

To evaluate the cardiac function, cardiac catheterization was performed as previously described (17). Briefly, after the induction of light general anesthesia [4% chloral hydrate solution (1 ml/100 g) by i.p. injection], a catheter was inserted into the right carotid artery and advanced into the left ventricle. Ventricular pressure signals were measured with a transducer and conditioner (MLT0830; ADInstruments, Bella Vista, Australia) and digitally recorded with a data acquisition system (PowerLab; ADInstruments). The following indices were obtained: heart rate (HR), left ventricular systolic pressure (LVSP), left ventricular enddiastolic pressure (LVEDP), as well as the maximum rates of left ventricular pressure rise and fall (+dp/dt max and −dp/dt max, respectively). During this process, the animals were placed on controlled heating pads. Core temperature was measured via a rectal probe and was maintained at 37°C. The rats were sacrificed after analysis of myocardial function and the hearts were harvested for subsequent experiments.

The samples were fixed in 10% formalin and were paraffin-embedded in the Surgical Pathology Facility of Nanchang University. TUNEL analysis was performed with a commercially available kit (Dead End Colorimetric TUNEL System) according to the manufacturer’s instructions (Promega, Madison, WI, USA). The slides were counterstained with hematoxylin (blue). Three midventricular sections (from the apex to the base) of each heart tissue were analyzed. Cardiomyocyte nuclei were quantified by randomly counting 10 fields/section. The apoptotic index (percentage of apoptotic nuclei) was calculated as apoptotic nuclei/total nuclei counted × 100.

Sections of left ventricles were stained with Masson’s trichrome to measure interstitial fibrosis. Interstitial collagen was quantified at a final magnification of ×200 using a microscope (BX51; Olympus, Center Valley, PA, USA) connected to a video camera (DS-Fi1; Nikon, Tokyo, Japan). The images captured under the microscope were used to calculate the collagen volume fraction of the myocardial interstitium using a Computer Imaging Analysis System (MPIAS-500; Nikon, Tokyo, Japan). The content of interstitial collagen, which was expressed as the fractional area of the entire cross-section where the perivascular collagen was excluded, was averaged on nine fields selected across the wall thickness in the septum and free wall.

Tissue samples obtained from the left ventricular free wall were minced. Total RNA was extracted from the samples with TRIzol reagent according to the manufacture’s instructions (Invitrogen). For qPCR, cDNA was synthesized in a 20-μl reaction volume containing 4 μg total RNA and SuperScript™ II Reverse Transcriptase (Fermentas, Burlington, ON, Canada) according to the manufacturer’s instructions. qPCR was carried out with a 7500 Real-Time PCR system (Applied Biosystems, Carlsbad, CA, USA) using SYBR-Green I (Applied Biosystems) as a fluorescent dye according to the manufacturer’s instructions. Relative quantitation of the mRNA expression of the gene of interest was calculated using the comparative threshold cycle number for each sample. To control the variation in the amount of DNA, gene expression of the target sequence was normalized in relation to the expression of an internal control, β-actin. The PCR products of hNRG-1 were size-fractioned by electrophoresis on 2% agarose gels. Primers for hNRG-1, bcl-2, bax, collagen type I and III, as well as β-actin were the following: hNRG-1, forward: 5′-TCACCATGGTGGCGACCGGTTCAGGCAGAGACAGAAAG-3′ and reverse: 5′-TCACCATGGTGGCGACCGGTTCAGGCAGAGACAGAAAG-3′; bcl-2, forward: 5′-CGGGAGATCGTGATGAAGT-3′ and reverse: 5′-CCACCGAACTCAAAGAAGG-3′; bax, forward: 5′-GCAGGGAGGATGGCTGGGGAGA-3′ and reverse: 5′-TCCAGACAAGCAGCCGCTCACG-3′; collagen type I, forward: 5′-GTTCGTGGTTCTCAGGGTAG-3′ and reverse: 5′-TTGTCGTAGCAGGGTTCTTT-3′; collagen type III, forward: 5′-TGCCCACAGCCTTCTACACCCT-3′ and reverse: 5′-CAGCCATTCCTCCCACTCCAG-3′; β-actin, forward: 5′-TGTGCTATGTTGCCCTAGACTTC-3′ and reverse: 5′-CGGACTCATCGTACTCCTGCT-3′.

Samples of left ventricle myocardium were homogenized in tissue protein extraction reagent (Beyotime, Beijing, China) supplemented with protease inhibitors. After centrifugation at 12,000 × g for 10 min at 4°C, the supernatants were collected according to the manufacturer’s instructions. Protein concentration was measured using the BCA Protein Assay kit (Pierce Biotechnology, Rockford, IL, USA). Equal amounts of 20 μg protein were loaded onto 10% sodium dodecyl sulfate-polyacrylamide gels. hNRG-1, total Akt, phospho-Akt (p-Akt; phospho-Ser 473), total eNOS and phospho-eNOS (p-eNOS; phospho-Ser 1177) were detected with mouse anti-hNRG-1, rabbit anti-Akt-1/-2/-3, rabbit anti-phospho-Akt-1/-2/-3 (Ser 473), mouse anti-eNOS and rabbit anti-phospho-eNOS antibodies (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA). After probing with these antibodies, the membranes were stripped of bound immunoglobulins. The immunoblots were developed by the enhanced chemiluminofluorescence method (Thermo Fisher Scientific, Hudson, NH, USA). The signals were quantified by densitometric analysis using a chemiluminescence imaging system (General Electric Company, Fairfield, CT, USA) and normalized to those of β-actin, an endogenous control protein.

All the data were expressed as the mean ± SEM. Comparisons between groups were made using one-way analysis of variance (ANOVA) with Fisher’s protected least significant difference post hoc comparison test. P<0.05 was considered to indicate a statistically significant difference.

CMECs were characterized by positive staining for factor VIII endothelial marker. The cells at passage 3 were transfected with hNGR-1-lentivirus or GFP-lentivirus and the expression of GFP was observed by fluorescence analysis.

Cytokines in the supernatant were measured by ELISA. The level of hNRG-1 in the group transfected with hNGR-1-lentivirus was detected to be 18±5.3 ng/ml, while no hNRG-1 was detected in the group transfected with GFP-lentivirus.

Since the proliferation and survival of cardiomyocytes is an important aspect of DCM, the effects of CMEC-conditioned media on cardiomyocyte growth and survival were determined. The number of cardiomyocytes cultured for 3 days in hNRG-1-CMEC conditioned basal medium was significantly higher compared with the number of cardiomyocytes cultured in GFP-CMEC conditioned basal medium (P<0.05). No significant difference was detected in the number of cardiomyocytes between the control and GFP groups (P>0.05, Fig. 1).

After co-culture with TNF-α for 24 h, the apoptotic rates of cardiomyocytes cultured in hNRG-1-CMEC conditioned basal medium was significantly lower compared with the number of cardiomyocytes cultured in GFP-CMEC conditioned basal medium (P<0.05). No significant difference was detected in the apoptotic rates of cardiomyocytes between the control and GFP groups (P>0.05, Fig. 1).

The expression of hNRG-1 in heart tissues was confirmed by relative quantification of hNRG-1 mRNA and western blot analysis. hNRG-1 mRNA and the fusion proteins of hNRG-1 and GFP were detected in the cardiac samples extracted from the cardiac tissues of the rats in the hNRG-1 group but not in those from the rats in the control, DCM and GFP groups (Fig. 2).

Sixteen weeks after the induction of diabetes, cardiac function was evaluated by invasive hemodynamic measurements. No significant difference was detected in HR among the four groups (P>0.05). A lower LVSP and higher LVEDP were observed in the DCM group compared with those in the control group (P<0.05). Resting maximum rates of rise (+dp/dt max) and fall (−dp/dt max) in left ventricular pressure were also impaired after the induction of diabetes (P<0.05), indicating that systolic and diastolic functions were significantly impaired in the diabetic rats. Following the injection of hNRG-1-lentivirus, these hemodynamic abnormalities were markedly attenuated (P<0.05). However, the hemodynamic abnormalities in the GFP group were comparable with those in the DCM group (P>0.05, Table I).

A TUNEL assay was performed to assess apoptosis in vivo. The number of positively stained cells in the DCM group was higher than that in the control group (P<0.05). Transfection with hNRG-1-lentivirus significantly reduced the number of apoptotic cells compared with those in the GFP and DCM groups (P<0.05, Fig. 3). qPCR was used to determine the mRNA expression levels of bcl-2 and bax, which are known to be markers of apoptosis. Following qPCR, bcl-2 was shown to be downregulated and bax was upregulated in the DCM group, while these alterations were attenuated following transfection with hNRG-1-lentivirus (P<0.05). Since bcl-2 is known to be an anti-apoptotic and bax a pro-apoptotic protein, these results indicated that transfection with hNRG-1-lentivirus protects cardiomyocytes against apoptosis (Fig. 3).

The collagen volume fraction, which is an indicator of interstitial fibrosis, was higher in the DCM group than in the control group (P<0.05, Fig. 4). Similarly, the collagen type I and III mRNA expression levels were also significantly upregulated in the DCM group compared with those in the control group (P<0.05, Fig. 4). The levels of myocardial fibrosis and of type I and III pro-collagen mRNA in the myocardium were markedly inhibited following cardiac transfection with rhNRG-1 rather than GFP, suggesting that hNRG-1 treatment attenuates the myocardial interstitial fibrosis caused by diabetes.

Western blot analysis showed that STZ treatment reduced the level of Akt phosphorylation compared with that in the control group, while hNRG-1 gene transfer increased the level of phospho-Akt (Fig. 5). Total Akt levels were not altered among the four groups. Similarly, hNRG-1 gene delivery significantly increased the level of the phosphorylated form of eNOS compared with the levels in the GFP and control groups. Total eNOS levels remained unaltered (Fig. 5). These results indicate that hNRG-1 gene transfer resulted in the activation of Akt and eNOS pathways by phosphorylation.