A total of 60 male weaning Sprague-Dawley rats (3-weeks old; specific-pathogen-free class; body weight, 75±10 g) were raised under controlled-environmental conditions (12-h light-dark cycle; temperature, 25±1°C; humidity, 65±4%). The standard diet (containing 0.2 mg selenium/kg food) was produced by the Animal Experimental Center of Xi’an Jiaotong University (Xi’an, China) and the low-selenium diet (<0.02 mg selenium/kg food) was produced by Trophic Animal Feed High-tech Co. (Jiangsu, China) according to the AIN-93 M formula. The food was stored at 4°C and fresh tap water was allocated continuously. The study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Committee on the Ethics of Animal Experiments of Xi’an Jiaotong University. All the surgeries were performed under chloral hydrate anesthesia, and all efforts were made to minimize suffering.

All the animals were randomly assigned into three groups: Control (n=20), low selenium (LS) (n=20) and selenium supplementation (SS) (n=20). In the control group, the animals were fed with the standard diet for 14 weeks and were treated by intraperitoneal injection of physiological saline every day for 21 days. In the LS group, the animals were fed with the low-selenium diet for 14 weeks and were treated by intraperitoneal injection of physiological saline every day for 21 days. In the SS group, the animals were fed with the low-selenium diet for 14 weeks and were treated by intraperitoneal injection of sodium selenite (0.05 mg/kg bodyweight; Sigma-Aldrich, St. Louis, MO, USA) as selenium supplementation every day for 21 days (21). All the animals were monitored every two days by observing their mental status and activities. Two animals succumbed as a result of Se deficiency. Following this, these two animals were immediately quarantined and sent to the Animal Center for unified treatment.

Samples of whole blood and hearts harvested from rats were wet-washed by mixed acids (nitric acid and perchloric acid) in borosilicate reaction tubes. Following removal of the excess acids, the samples were titrated to 10 ml by ultra-pure water (Aqulix 5 water purification system; Merck Millipore, Billerica, MA, USA). The selenium concentration was subsequently determined by the flameless atomic absorption spectrophotometry method using a Z-5000 spectrophotometer (Hitachi, Ltd., Tokyo, Japan) with a cathode lamp of Se (resonance line, 196.0 nm; Photron, Victoria, Australia). Standard selenium solutions were used to calibrate the results.

Cardiac glutathione peroxidase (GPx) activity in heart extract was assessed by a spectrophotometry method using the Glutathione Peroxidase Cellular Activity Assay kit (Sigma-Aldrich), following the manufacturer’s instructions. The assay is based on the reaction in which oxidation of glutathione (GSH) to oxidized glutathione (GSSG) is catalyzed by GPx. As NADPH is consumed when GSSG is recycled back to GSH, the decrease in NADPH absorbance at 340 nm (Tecan Sunrise Absorbance Reader, Tecan, Austria) can be utilized to calculate the activity of GPx indirectly.

Following femoral artery puncture, whole blood samples were collected using EDTA-Na₂ vacuum blood collection tubes. The supernatant was collected after centrifugation at 377 × g for 20 min. The samples were processed by the Triage BNP assay (Biosite, San Diego, CA, USA) within 1 h after collection at room temperature, according to the manufacturer’s instructions.

ECG in lead II was recorded (PowerLab 4/25; ADInstrument, New South Wales, Australia) during the experiment time. Prior to electrocardiography, all the rats were anesthetized by intraperitoneal injection of chloral hydrate (10%, 0.03 ml/kg bodyweight). The left upper limb, right upper limb and right lower limb electrodes were placed for leads I. Subsequently, the ECG was analyzed for changes in the ST segment, T wave, AV block and arrhythmia (premature ventricular contraction, ventricular tachycardia and fibrillation and atrial fibrillation). Cardiac conduction of rats was evaluated by the number of ventricular arrhythmic events (VAEs) within 15 min.

Echocardiographic tests were performed according to the instructions described in previous studies (22,23). Animals were anesthetized with chloral hydrate (10%, 0.03 ml/kg bodyweight) intraperitoneally 10 min before imaging. In order to maintain an optimal image quality, the hair of the anterior chest wall was removed by 7% sodium sulfite solution and the rats were placed left laterally. Vivid 7 dimension (GE Healthcare, Pittsburgh, PA, USA) with a probe (12 L) working at 10 MHz was utilized. The probe was placed parallel to the left margin of the sternum and adjusting the image depth, ranging 2.0–4.0 cm. Regurgitant jets were assessed by two-dimensional color and continuous Doppler. M-mode tracings were applied to determine the left ventricular end-systole diameter (LVESD) and left ventricular end-diastole diameter (LVEDD). Finally, the functional parameters, including left ventricular fractional shortening (LVFS%), which is the percentage of blood ejected by the left ventricle for each heart beat, and left ventricular ejection fraction (LVEF%), which measures the change in the left ventricular diameter from each diastole to systole were calculated respectively: LVFS% = ((LVEDD − LVESD)/LVEDD)%; and LVEF% = [(LVEDV − LVESV)/LVEDV]%.

Hemodynamic determination was conducted according to methods described in a previous study (24). The animals were anesthetized with chloral hydrate (10%, 0.03 ml/kg bodyweight) intraperitoneally. A catheter connected to the Powerlab 4/25 biological analysis system was intubated from the right carotid artery into the left ventricle. Left ventricular systolic pressure (LVSP), left ventricular end-diastolic pressure (LVEDP), maximum rising rate of left ventricular pressure (LVdp/dt_max) and maximum dropping rate of left ventricular pressure (LVdp/dt_min) were measured.

The animals were anesthetized with chloral hydrate (10%, 0.03 ml/kg bodyweight) intraperitoneally. Subsequently, they were sacrificed by removing the hearts following complete anesthesia. The harvested hearts were rinsed in phosphate-buffered saline, fixed in 4% paraformaldehyde for 24 h, embedded in paraffin and cross-sectioned into 10-μm slices. The sections were stained with hematoxylin and eosin (HE) for cells alignment according to the general procedure. The morphological structures of the heart were observed by light microscopy.

Total RNA was isolated using TRIzol (Invitrogen) and the miRNeasy mini kit (Qiagen) according to the manufacturer’s instructions, which efficiently recovered all RNA species, including miRNAs. RNA quality and quantity was measured by using nanodrop spectrophotometer (ND-1000, Nanodrop Technologies, Wilmington, DE, USA) and RNA integrity was determined by gel electrophoresis. Following RNA isolation from the samples, the miRCURY™ Hy3™/Hy5™ Power labeling kit (Exiqon, Vedbaek, Denmark) was used according to the manufacturer’s guideline for miRNA labeling. Each sample (1 μg) was 3′-end-labeled with the Hy3TM fluorescent label, using T4 RNA ligase as follows: RNA in 2.0 μl water was combined with 1.0 μl calf intestinal alkaline phosphatase (CIP) buffer and CIP (Exiqon). The mixture was incubated for 30 min at 37°C, and was terminated by incubation for 5 min at 95°C. Subsequently, 3.0 μl labeling buffer, 1.5 μl fluorescent label (Hy3TM), 2.0 μl dimethyl sulfoxide and 2.0 μl labeling enzyme were added into the mixture. The labeling reaction was incubated for 1 h at 16°C, and terminated by incubation for 15 min at 65°C. Subsequent to stopping the labeling procedure, the Hy3TM-labeled samples were hybridized on the miRCURYTM LNA Array (v.16.0) (Exiqon) according to the array instructions. The total mixture (25 μl) from Hy3TM-labeled samples with 25 μl hybridization buffer were first denatured for 2 min at 95°C, incubated on ice for 2 min and hybridized to the microarray for 16–20 h at 56°C in a 12-Bay Hybridization System (Hybridization System-Nimblegen Systems, Inc., Madison, WI, USA), which provides an active mixing action and constant incubation temperature to improve hybridization uniformity and to enhance the signal. Following hybridization, the slides were obtained, washed several times using Wash buffer kit (Exiqon), and dried by centrifugation for 5 min at 100 × g. Subsequently, the slides were scanned using the Axon GenePix 4000B microarray scanner (Axon Instruments, Foster City, CA). The scanned images were imported into the GenePix Pro 6.0 software (Axon Instruments) for grid alignment and data extraction. Replicated miRNAs were averaged and miRNAs with intensities >50 in all the samples were chosen for calculating normalization factor. Expressed data were normalized using the Median normalization (25). Following normalization, differentially expressed miRNAs were identified through fold change filtering. Hierarchical clustering was performed using the MEV software (Dana-Farber Cancer Institute, Boston, MA, USA).

GO analysis was applied in order to organize genes into hierarchical categories and uncover the miR-gene regulatory network on the basis of biological process and molecular function. In detail, two-side Fisher’s exact test was used to classify the GO category, and the false discovery rate (FDR) was calculated to correct the P-value. Only GOs that had a P-value of <0.001 and an FDR of <0.05 were chosen. Within the significant category, the enrichment rare earth (Re) was: Re=(n_f/n)/(N_f)/N), where n_f is the number of flagged genes within the particular category, n is the total number of genes within the same category, N_f is the number of flagged genes in the entire microarray and N is the total number of genes in the microarray. Subsequently, the apoptosis-related network of miRNA-mRNA interaction, representing the critical miRNAs and their targets, was established according to the degree of miRNA (26).

Differentially expressed miRNAs of miR-374, miR-16, miR-199a-5p, miR-195, miR-30e^*, miR-3571, miR-675 and miR-450a^* were selected for verification. Total RNA was extracted from harvested heart using the MiniBEST Universal RNA Extraction kit (Takara, Otsu, Japan). cDNA was synthesized from total RNA with the SYBR PrimeScript™ miRNA RT-PCR kit (Takara). The expression of miRNA was analyzed by qPCR using SYBR Premix Ex TaqTMII (Takara). PCR was performed in triplicate for each sample. The relative amount of miRNAs was normalized against U6 small nuclear RNA. Detection of mRNA was performed as described previously (27). The sequences of the primers are shown in Table I.

Frozen cardiac tissue was homogenized in radioimmunoprecipitation assay lysis buffer system (Santa Cruz Biotechnology, Inc., Dallax, TX, USA) with phenylmethylsulfonyl fluoride (Santa Cruz Biotechnology, Inc.). All the procedures followed manufacturer’s instructions. The sample protein concentration was detected by using bicinchoninic acid protein assay kit (Santa Cruz Biotechnology, Inc.). The sample protein was boiled in 1x SDS-PAGE loading buffer, separated by electrophoresis in 10% SDS-polyacrylamide gel and subsequently transferred to a polyvinylidene fluoride membrane. Antibodies against Wnt (ab15251) and β-catenin (ab6302) (Abcam, Cambridge, MA, USA) were applied to incubate the bolts at 4°C overnight. Tris-buffered saline (containing 0.02% Tween 20) was used to wash the membranes, which were subsequently incubated with goat polyclonal secondary antibody to rabbit immunoglobulin G conjugated to horseradish peroxidase (Abcam). The membranes were developed using Super Signal West Pico chemiluminescence reagent (Thermo Scientific, Waltham, MA, USA) and were visualized on X-ray films.

All the results were expressed as mean ± standard deviation. Statistical analysis was performed with one-way analysis of variance for multiple comparisons. P<0.05 was considered to indicate a statistically significant difference.

The Se concentration in blood is shown in Fig. 1. Significant changes of Se concentration were observed in the LS and SS groups. Compared to the control group, the Se concentration in blood decreased significantly in the LS and SS groups (P<0.05). Se intraperitoneal injection was proved to increase the Se concentration in blood in the SS group compared to the LS group (P<0.05).

As shown in Fig. 2, the activity of GPx in the LS and SS groups decreased significantly compared to the control group (P<0.05). The activity of GPx was observed to reduce more significantly in the LS group. However, an evident recovery of GPx activity in the SS group was confirmed by the GPx activity assay when compared to the LS group (P<0.05).

The Triage BNP assay showed that the plasma BNP level increased significantly in the LS and SS groups compared to the control group (P<0.05). A clearer increase of plasma BNP was found in the LS group. Following Se supplementation by intra-peritoneal injections, a reduction in the plasma BNP level in the SS group was identified compared to the LS group (P<0.05) (Fig. 3).

ECG of rats in the control group was normal (Fig. 4A). However, ECG of rats in the LS group showed significant differences. In the LS group, a few rats exhibited normal ECG, but the majority had abnormal ECG, which manifested as premature ventricular contraction and paroxysmal supraventricular tachycardia (Fig. 4B and C). The majority of ECG in the SS group were normal, but there remained a few abnormal ECG (Fig. 4D). VAEs in the LS group were higher than those of the control group (P<0.05), but following Se supplementation the VAEs significantly decreased compared to the LS group (P<0.05) (Fig. 4E).

The echocardiographic parameters, including LVEDD, LVESD, LVEF and LVFS, were detected and calculated in the present study as shown in Table II. An overall increase of LVEDD and LVESD accompanied by an overall decrease of LVEF and LVFS were found in the LS and SS groups. However, there was a significant increase of LVEDD and LVESD and a significant decrease of LVEF and LVFS in the LS group compared to the SS group. A significant recovery of cardiac function marked by echocardiographic parameters was found in the SS group compared to the LS group, which was evidenced by restoration of LVEDD, LVESD, LVEF and LVFS following Se supplementation.

The hemodynamic parameters, including LVEDP, LVSP, LVdp/dt_max and LVdp/dt_min, were detected and calculated in the study as shown in Table III. An overall increase of LVEDP and LVSP accompanied by an overall decrease of LVdp/dt_max and LVdp/dt_min were found in the LS and SS groups. However, there was a significant increase of LVEDP and LVSP and decrease of LVdp/dt_max and LVdp/dt_min in the LS group compared to the SS group. A significant recovery of cardiac function marked by hemodynamic parameters was found in the SS group compared to the LS group, which was evidenced by restoration of LVEDP, LVSP, LVdp/dt_max and LVdp/dt_min following Se supplementation.

HE staining showed that the structure of myocardial tissue in the control group was clear, including well-connected intercalated discs and normally arranged filaments (Fig. 5A). In the LS group, the muscle fibers of the rats were swelling and disorganized (Fig. 5B). Myocardial necrosis, nuclear condensation and partially intermittent muscle fibers were observed in the LS group (Fig. 5C and D). In addition, the small myocardial necrosis focus accompanied by inflammatory infiltration can be observed in certain areas (Fig. 5E). Following Se supplementation, the myocardial fibers were mildly swelling compared to the control group, but their arrangements were orderly (Fig. 5F).

Using the miRCURY LNA Array platform, the miRNAs expression profiles were assessed in Se-deficient rats. The expression profiles of 116 miRNAs were regulated between the control, LS and SS groups, and the samples were separated into biologically interpretable groups. Among these, 5 miRNAs were identified to be upregulated >5-fold in the LS group compared to the SS group, whereas 3 miRNAs were less than the threshold level (3-fold) set during the progression of Se deficiency, but following selenium supplementation these miRNAs were >1.5-fold compared to Se deficiency. These 8 miRNAs were validated to be significantly different (P<0.05). As shown in Fig. 6, the levels of miR-374, miR-16, miR-199a-5p, miR-195 and miR-30e^* were upregulated in rats with selenium deficiency, while miR-3571, miR-675 and miR-450a^* showed an opposite expression pattern, in agreement with the results of microarray hybridization.

According to the threshold of GOs that were significantly regulated by miRNAs, the P-value and FDR was <0.001 and <0.05, respectively. The high-enrichment GOs targeted by upregulated miRNAs included signal transduction, transport, cell differentiation and response to stress. By contrast, the significant GOs corresponding to downregulated miRNAs appeared to include signal transduction, cell differentiation, transport and cell proliferation. Among these, the maximum-enriched-GO associated with signal transduction, together with the numerous miRNAs that interacted with signal transduction-related genes, suggested them to have an important role in the activation of selenium deficiency (Fig. 7).

Western blotting was employed to detect the changes of Wnt and β-catenin expression in the present study. As shown in Fig. 8, the protein expression of Wnt and β-catenin in cardiac tissue were highest in the LS group compared to the control group. Increasing expression of Wnt and β-catenin in the LS group was decreased by Se supplementation. However, it appeared that Se supplementation reversed the increasing expression of Wnt and β-catenin in the LS group.