Fifteen male 10-week-old C57BL/6 mice (Chengdu DOSSY Experimental Animals Co., Ltd., China) from the same batch were divided into a sham-control group (n = 5) and an AMI model group (n = 10). As described by Gao et al. [13], the mice were anesthetized with an intraperitoneal injection of ketamine (50 mg/kg) and pentobarbital sodium (50 mg/kg) without ventilation support. A small incision was made left of the sternum, and a purse-string suture was prepared. The thorax was punctured and dilated using a mosquito clamp at the left 4th intercostal space. The beating heart was gently “popped out” and an 8-0 silk suture was quickly applied at a site on the left free wall approximately 3 mm below the atrium to block the left anterior descending coronary artery. Ligation was confirmed when the anterior left ventricular wall became pale. Following ligation, the heart was returned to the intrathoracic space and intrathoracic air was evacuated manually. The previous purse-string suture was tied and the mouse monitored during the recovery period. Mice in the sham-control group underwent the same surgical procedure and the beating hearts “popped out” without any ligation. All animal experiments were strictly performed following the National Institutes of Health Guide for the Care and Use of Laboratory Animals (Publication no. 8023, revised in 1978) and registered with the Animal Ethics Committee of West China Hospital, Sichuan University (no. 20220221020).

All serum samples were collected at 24 h postoperative. As in our previous study [11], each serum sample was precipitated using methanol spiked with two internal standards. The supernatant was then collected. The dried supernatant was reconstituted with 10 mM ammonium acetate in 30% water/70% acetonitrile + 0.2% acetic acid containing 12.6 µM ${}^3$C${}_5$–${}^{15}$N-l-tyrosine and 34.68 µM ${}^{13}$C${}_1$-l-lactate.

The final sample reconstitution was added to the ultra-high-performance liquid chromatography with tandem mass spectrometry under positive and negative ion modes. The quality control (QC) sample was set as the pooled sample with a mixture of all serum samples. The targeted metabolomics analysis was conducted using an LC-40A Ultra-Performance Liquid Chromatography (UHPLC) system coupled with an AB Sciex triple quadrupole 5500 mass spectrometer (Nexera, Framingham, MA, USA). Chromatographic separation was performed on a BEH Amide column (2.1 $\times$ 100, 1.7 µm; Waters, Milford, MA, USA) at 35 °C. The mobile phase and gradient conditions, as well as the parameters for the mass spectrometer, were described previously [11]. Multiple reaction monitoring mode was used to detect the metabolites of interest. The targeted analysis included 238 metabolites selected from over 40 metabolic pathways. The data were acquired using Analyst 1.6.3${}^{\text{TM}}$ software (AB Sciex, Framingham, MA, USA) and analyzed using Multiquant 3.0.2 software (AB Sciex, Framingham, MA, USA).

Myocardiocytes are isolated rat cardiomyocytes, and the cell type was further verified with IF staining of α-actinin (Supplementary Fig. 1). Neonatal cardiomyocytes were prepared as described previously [14]. In brief, the hearts were collected from neonatal mice within 24–48 h and cut into pieces in D-Hank’s balanced salt solution. The cardiac tissues were digested in 22.5 µg/mL liberase blendzyme IV (Roche, Mannheim, Germany) at 37 °C for 40 min. Isolated cardiomyocytes in the pellet after centrifugation were resuspended in M199 medium with 10% fetal bovine serum and cultured in pre-coated plates. After 48 h, the cardiomyocytes were subjected to various treatments and measurements after 48 h of culture.

The oxygen–glucose deprivation (OGD) model was defined as a glucose-free medium under a deoxygenated atmosphere. In the present study, the OGD model was set with no glucose and 0% oxygen for 6 h of treatment before the sample collection.

NMDA (HY-17551; MedChemExpress, NJ, USA) was used as an agonist of NMDAR. NMDA was administered 30 min before the ODG or normoxic condition. In the preliminary study, NMDA concentrations were set at 10 µM, 50 µM, and 100 µM, and the final working dose was 10 µM. MK-801 (HY-15084B; MedChemExpress), which was used as a selective NMDAR antagonist, was pre-administered for 30 min at a concentration of 10 µM.

Cardiomyocyte lysates were subjected to western blotting analysis. Specific target proteins (Caspase-3, NMDAR1A, Bax, and Bcl-2) were separated from the cell lysates, transferred to polyvinylidene difluoride membranes, and visualized by chemiluminescence. The primary antibodies included rabbit anti-$\alpha$-tubulin (bs-0159R; Bioss, Bejing, China), rabbit anti-cleaved-caspase 3 and caspase 3 (ab214430 and ab184787; Abcam, Cambridge, England), rabbit anti-Bax (ab182734; Abcam), rabbit anti-Bcl-2 (ab18258; Abcam), and mouse anti-NMDAR1A (32-0500; Thermo Fisher Scientific, Cleveland, OH, USA).

Total RNA from cultured cardiomyocytes was extracted using TRIzol reagent (Sigma, Poole, Dorset, UK). Circular DNA was synthesized using M-MLV reverse transcriptase (Invitrogen, Burlington, IA, USA). SYBR Green PCR master mix was used according to the manufacturer’s instructions (Eurogentec, San Diego, CA, USA), while 28S rRNA was used as a loading control. The primer sequences were as follows: nmdar1a 5′-ACT TGA AGG GCT TGG AGA-3′ (forward) and 5′-GGA GTG GAA CGG AAT GAT-3′ (reverse); nmdar2b 5′-GCT GGA GGC GTT GGA TGT-3′ (forward) and 5′-AAT GTG GAT TGG GAG GAC-3′ (reverse); 28S rRNA 5′-TTG AAA ATC CGG GGG AGA G-3′ (forward) and 5′-ACA TTG TTC CAA CAT GCC AG-3′ (reverse). A total of 34 amplification cycles were conducted on an Eppendorf Real-Time PCR machine. The expression of nmdar1a/nmdar2b in relation to the 28S rRNA control was determined.

The samples were fixed with 3% glutaraldehyde for 24 h, fixed in 1% osmium tetroxide, dehydrated in a series of acetone, infiltrated in Epon 812 for 30–60 min each time, and finally embedded. The semi-thin sections were stained with methylene blue, and ultrathin sections were cut with a diamond knife and stained with uranyl acetate and lead citrate. The sections were examined using a transmission electron microscopy (TEM, JEM-1400 Flash, JEOL Ltd., Tokyo, Japan).

A cell counting kit-8 (CCK-8) assay kit was purchased from Dojindo (Dojindo Molecular Technologies, Inc., Kumamoto, Japan). Cardiomyocytes were cultured in 96-well plates according to the manufacturer’s instructions. Following the different treatments, 10 µL of CCK-8 solution was added to each well. After 2 h of incubation, the absorbance was measured at 450 nm by a microplate reader.

Calcium concentration (ab142780; Abcam), mitochondrial ROS (M36008; Thermo Fisher Scientific), and mitochondrial membrane potential (M34152; Thermo Fisher Scientific) were measured using specific assay kits according to the manufacturer’s instructions.

The statistical analysis was performed using SPSS software version 19 (SPSS Inc., Chicago, IL, USA). The results are presented as mean $\pm$ standard error of the mean (SEM). The data were analyzed using one-way analysis of variance, followed by the Bonferroni post-hoc test. An unpaired Student’s t-test was used to compare groups. Statistical significance was set at p 0.05.

To identify the altered profiles of metabolites in myocardial infarction, a targeted liquid chromatography with tandem mass spectrometry comprehensive analysis of 238 metabolites in positive and negative modes was performed on serum samples. A principal component analysis indicated that the metabolic profiles of mice in the AMI groups deviated from those of the sham-control group, showing alterations in metabolism profiles in AMI (Fig. 1).

Fig. 1.

PCA Scores plot (A) and 3D scores plot (B) for control and model groups. Abbreviations: PC1, principle component 1; PC2, principle component 2.

A heat map is shown in Fig. 2A. The most affected pathways were identified according to the Kyoto Encyclopedia of Genes and Genomes signaling pathway analysis based on the metabolites (impact values $>$0.2), including: (1) d-glutamine and d-glutamate metabolism; (2) alanine, aspartate, and glutamate metabolism; and (3) taurine and hypotaurine metabolism (Fig. 2B). Several metabolites involved in these signaling pathways are shown in Fig. 3 (p 0.0001).

Fig. 2.

Heatmap and impact pathways. (A) Heatmap of 128 metabolites in control and model groups. The colour bar from blue to red represents low to high mean intensity after normalization using the Z-score. (B) KEGG pathway analysis on the metabolites of two groups. The bubble diagram displayed selected pathways with an impact value larger than 0.2. The size of the bubbles represented impact values ranging from 0.2 to 1.0, and the colour of the bubbles means the p value after the negative Log transform.

Fig. 3.

Box plots showing integrated intensities for 5 selected LC-MS-identified biomarkers in two groups. The black dots represent the concentrations of the selected feature from all samples. t-test was used to identify the difference.

According to the metabolomics results, glutamate was selected as an essential excitatory neurotransmitter, which reportedly exerts multiple regulatory effects through its specific ligand (N-methyl-d-aspartic acid receptor [NMDAR]) [15]. Thus, to confirm the role of NMDAR in ischemic injury–induced cardiomyocyte apoptosis, we performed further studies in cell culture. NMDA is a highly selective agonist of NMDAR. Different doses (10 µM, 50 µM, and 100 µM) of NMDA were used for pre-treatment for 30 min. As shown in Fig. 4A, NMDA did not exert any adverse effects under normal conditions (p $>$ 0.05). Hypoxic injury–induced cell death and NMDA aggravated hypoxic injury in a dose-dependent manner (p 0.05). We also detected significantly elevated expression of cleaved caspase 3 under hypoxic injury (p 0.05). After NMDA treatment, cleaved caspase 3 levels were further increased (p 0.05; Fig. 4B). Surprisingly, the mRNA levels of NMDAR were similar under normal and hypoxic conditions (p $>$ 0.05; Fig. 4C). Therefore, mitochondrial proteins were extracted, and we found enriched NMDAR expression in the mitochondria under hypoxic injury (p 0.05; Fig. 4D).

Fig. 4.

NMDAR participates in the ischemic injury-induced cardiomyocyte apoptosis. (A) CCK-8 assay showed NMDA could aggravate the hypoxic injury in a dose-dependent manner (*p 0.05 vs. Blank; **p 0.01 vs. Blank; ***p 0.001 vs. Blank; ****p 0.0001 vs. Blank; n = 5) while NMDA did not exert any adverse effects under normoxia (ns p $>$ 0.05). (B) NMDA significantly increased the level of cleaved caspase-3 under hypoxia (p 0.05 vs. Blank; n = 5). (C) mRNA level of NMDAR showed no difference (p $>$ 0.05 vs. Blank; n = 5). (D) The protein level of NMDAR in mitochondria showed significant elevation under hypoxic injury (p 0.05 vs. Blank; n = 5). ns, not significant.

Given the translocation of NMDAR into the mitochondria under hypoxic injury, we further explored NMDAR-mediated mitochondrial damage in hypoxic injury–induced apoptosis. MK-801, a selective NMDAR antagonist, was administered. With NMDAR inhibition, apoptosis under hypoxic injury was significantly blocked, and the CCK-8 assay showed improved cell viability (p 0.05; Fig. 5A,B). Bax and Bcl-2 are the two key factors involved in mitochondrial damage–induced apoptosis. Similar to the expression of cleaved caspase 3, the inhibition of NMDAR significantly decreased Bax expression and increased Bcl-2 expression under hypoxia and hypoxia with NMDA (p 0.05; Fig. 5C). Transmission electron microscopy images showed normal mitochondrial morphology under normal conditions, disrupted vesicular mitochondria under hypoxia, and significant vesicular and swollen mitochondria under hypoxia and NMDA treatment. With MK-801 treatment, less vesicular and less swollen mitochondria were observed (Fig. 5D).

Fig. 5.

NMDAR-mediated mitochondrial damage induces cardiomyocyte apoptosis. (A,B) Inhibition of NMDAR with MK-801 significantly limited apoptosis and improved survival (*p 0.05; **p 0.01, n = 5). (C) The protein level of Bax was inhibited with treatment of MK-801 and the Bcl-2 level was increased (p 0.05 vs. Hypoxia, n = 5). (D) The transmission electron microscope showed mitochondria in different conditions. vesicular, disrupted mitochondria under hypoxia; significant vesicular and swollen mitochondria under hypoxia and NMDA treatment; less vesicular and less swollen mitochondria under hypoxia and MK-801.

Calcium accumulation or overload is an essential phenomenon in ischemic injuries. The Rhod-2 calcium indicator was used to detect calcium levels in mitochondria under different circumstances. Fluorescent images showed remarkable calcium accumulation under hypoxic and non-hypoxic conditions with NMDA. Treatment with MK-801 significantly reduced mitochondrial calcium accumulation in cardiomyocytes (p 0.05; Fig. 6A,B). In addition, loss of mitochondrial membrane potential was observed in hypoxia and hypoxia with NMDA conditions, while the inhibition of NMDAR restored the mitochondrial membrane potential (p 0.05; Fig. 6C). Mitochondrial ROS (mitoROS) levels were measured (Fig. 6D–F); MK-801 significantly inhibited mitoROS formation (p 0.05).

Fig. 6.

NMDAR promotes calcium and mitoROS accumulation in the mitochondria. (A,B) Representative images of calcium indicator and quantification showed less calcium accumulation after MK-801 treatment (white bar =500 µm; *p 0.05, n = 5). (C) Loss of mitochondrial membrane potential was identified in both hypoxia and hypoxia with NMDA, and inhibition of NMDAR could restore the mitochondrial membrane potential (*p 0.05; **p 0.01; ***p 0.001). (D–F) Representative images of mitochondrial ROS accumulation and quantification showed inhibition of mitoROS accumulation after MK-801 treatment (white bar =500 µm; *p 0.05; **p 0.01; ***p 0.001; ****p 0.0001; ns p $>$ 0.05; n = 5).