Differential expression of glycolysis-related genes and their potential immune mechanisms in acute myocardial infarction

Abstract

Background

Acute myocardial infarction (AMI) remains a leading cause of global morbidity and mortality, necessitating deeper insights into its molecular mechanisms. Glycolysis, the metabolic pathway that converts glucose into pyruvate, plays a crucial role in cardiovascular diseases. This study aimed to identify glycolysis-related genes associated with AMI through integrative bioinformatics analyses.

Methods

Gene expression profiles from GSE48060 and GSE29532 datasets were merged and analyzed to identify differentially expressed genes (DEGs) related to glycolysis in AMI. Functional enrichment, protein-protein interaction, and immune infiltration analyses were performed using various bioinformatics tools.

Results

Eleven glycolysis-related DEGs were identified, with PYGL, HK3, PGAM2, PFKFB3, and PRKACB emerging as hub genes, which were further verified by quantitative real-time PCR. These genes were enriched in pathways related to muscle contraction, hexose metabolism, and fructose/mannose metabolism. Immune infiltration analysis revealed significant differences in several immune cell populations between AMI and control samples.

Conclusions

Our findings provide new perspectives on the molecular underpinnings of AMI, highlighting potential therapeutic targets and biomarkers for further investigation. Further experimental validation is necessary to confirm the functional roles of these genes in the pathogenesis of AMI.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13019-025-03636-y.

Introduction

AMI commonly referred to as a heart attack, continues to be a predominant contributor to global morbidity and mortality. The World Health Organization reports that cardiovascular diseases, including AMI, account for approximately 31% of global deaths annually [1]. Despite significant advancements in medical treatments and early intervention strategies, the prognosis for AMI patients still remains unclear, with high rates of subsequent heart failure and recurrent infarctions [2].Current diagnostic methods for AMI, such as electrocardiograms and cardiac biomarkers like troponins, exhibit limitations in both sensitivity and specificity, particularly during the initial stages of the disease [3]. This situation accentuates the pressing requirement for innovative biomarkers and therapeutic targets that can enhance early stage diagnosis and treatment of AMI [4].

One of the critical challenges in managing AMI is the early and accurate identification of the disease. Traditional diagnostic methods, such as electrocardiograms and cardiac biomarkers, often fail to detect subtle yet significant changes in myocardial tissue, leading to delayed or missed diagnoses [5]. Furthermore, while reperfusion therapies like percutaneous coronary intervention (PCI) and thrombolytics have improved outcomes, they are not universally effective and come with risks of complications [6]. Therefore, there is a pressing need for novel biomarkers and molecular targets that can enhance early detection, provide prognostic information, and offer new therapeutic avenues.

Glycolysis, the metabolic process responsible for the conversion of glucose to pyruvate, is essential for cellular energy and has been associated with a range of pathological conditions, notably cardiovascular diseases [7]. In ischemic heart tissues, glycolysis has been shown to be upregulated, which might act as a potential adaptive mechanism to maintain ATP production under hypoxic conditions [8]. However, the specific roles of GRGs in AMI remain largely unexplored. Recent studies have presented conflicting views on the role of glycolysis in AMI. Some researchers proposed that enhanced glycolysis might provide cardioprotective effects by maintaining energy production in oxygen-deprived cardiomyocytes [9]. Conversely, others argued that excessive glycolysis might exacerbate injury through increased lactate production and intracellular acidosis [10]. This controversy highlights the need for a comprehensive analysis of GRGs in the context of AMI.

The objective of our research is to address this knowledge deficiency by investigating the differential expression of GRGs in AMI and assessing their potential diagnostic value and biological functions. By integrating multiple AMI datasets and employing various bioinformatics approaches, we sought to identify GRDEGs in AMI, subsequently explored their functional implications through enrichment analysis and network constructionand ultimately investigate potential relationships between these GRDEGs and immune cell infiltration in AMI. This stepwise approach aims to offer an in-depth comprehension of the function of GRGs in the pathophysiology mechanisms underlying AMI. We have identified several GRDEGs significantly associated with AMI and demonstrated their involvement in critical biological processes and pathways. Our findings suggest that some of these genes might have high diagnostic value and play important roles in the complex regulatory networks underlying AMI pathophysiology.

Although some studies have indicated the role of glycolysis in ischemic heart disease, there is still a lack of comprehensive research systematically analyzing the expression differences of GRGs in AMI and their association with immune mechanisms. This study systematically reveals the relationship between GRGs and immune infiltration, providing new perspectives and theoretical basis for the research on the molecular mechanisms of AMI.

Materials and methods

Data acquisition

The AMI datasets GSE48060 [11] and GSE29532 [12] were downloaded from the GEO database [13] (http://www.ncbi.nlm.nih.gov/geo/), utilizing the R package GEOquery (Version 2.70.0) [14] (Figure S1). Both GSE48060 and GSE29532 datasets comprised samples derived from Homo sapiens, with the tissue source being peripheral blood. The chip platform utilized for GSE48060 and GSE29532 were GPL570 and GPL5175, respectively. Detailed information regarding these datasets is presented in Table S1. Specifically, the GSE48060 dataset included 31 AMI samples and 21 Control samples, while the GSE29532 dataset contained 49 AMI samples and 6 Control samples. All AMI and Control samples were incorporated into the current study.

GRGs were collected from the GeneCards database (https://www.genecards.org/) [15] and MSigDB database (https://www.gsea-msigdb.org/gsea/msigdb) [16]. The GeneCards database offers extensive data regarding human genes. After using “Glycolysis” as the search term and filtering for GRGs categorized as “Protein Coding” with a “Relevance Score >2”. A total of 208 GRGs were obtained. In addition, utilizing “Glycolysis” as a keywords on the PubMed website (https://pubmed.ncbi.nlm.nih.gov/) yielded an additional 291 GRGs from the published literature [17]. Ultimately, after conducting a thorough deduplication process, a cumulative total of 425 GRGs were acquired.

To eliminate batch effects from the datasets GSE48060 and GSE29532 using the R package sva (Version 3.50.0) [18], resulting in the integrated GEO dataset (AMI_Datasets). The integrated GEO datasets (AMI_Datasets) comprised a total of 80 samples from individuals with AMI and 27 Control samples. Subsequently, these integrated GEO datasets underwent standardization using the R package limma (Version 3.58.1) [19], during which the annotation probes were also standardized and normalized. Conduct principal component analysis (PCA) on the expression matrices obtained prior to and following the removal of batch effects to assess the efficacy of the batch effects correction [20]. PCA serves as a method for dimensionality reduction technique that extracts feature vectors from high-dimensional data into low-dimensional data. This transformation facilitates the visualization of these features in a two-dimensional plot.

Differentially expressed gene analysis

Based on the classification of samples derived from the integrated GEO datasets (AMI_Datasets), the samples were categorized into two distinct groups: the AMI group and control group. Differential gene expression analysis between these groups was conducted utilizing the R package limma (Version 3.58.1) [19]. The threshold was established, where |logFC| >0.25 and p value < 0.05 were set as criteria for identifying differentially expressed genes (DEGs) as previously recommended [21, 22]. Specifically, genes exhibiting logFC >0.25 and p value < 0.05 were classified as up-regulated DEGs, whereas those with logFC < −0.25 and p value < 0.05 were designated as down-regulated DEGs. The outcomes of the differential analysis were illustrated through a volcano plot, which was generated using the R package ggplot2 (Version 3.4.4) [23]

.To identify GRDEGs that were associated with AMI, the GEO dataset (AMI_Datasets) would be utilized. This dataset underwent a variance analysis, focusing on DEGs that meet the criteria of |logFC| >0.25 and p value < 0.05. The intersection of these DEGs with the genes related to GRGs were determined, and a Venn diagram was constructed to illustrate this overlap. Furthermore, the relationship between DEGs and GRDEGs was visualized using heat maps generated by the R package heatmap (Version 1.0.12). Additionally, the chromosomal locations of these genes were mapped using the R package RCircos [24]. The R codes are listed in Supplementary materials.

Gene ontology (GO) and pathway (KEGG) enrichment analysis

GO analysis [25] is a prevalent approach utilized in extensive functional enrichment investigations, encompassing Biological Process (BP), Cell Component (CC), and Molecular Function (MF). The Kyoto Encyclopedia of Genes and Genomes (KEGG) [26] serves as a widely recognized database that houses information pertaining to genomes, biological pathways, diseases, and pharmacological agents. The enrichment analyses for GO and KEGG pathway concerning GRDEGs were conducted employing the R package clusterProfiler (Version 4.16.0) [27]. The criteria for item selection were established as p value < 0.05 and FDR value (q value) < 0.25, which were deemed statistically significant. subsequently, the R package Pathview (Version 1.42.0) [28] was utilized to illustrate the pathway maps associated with the KEGG pathway.

Gene set enrichment analysis (GSEA)

GSEA was considered as a method for assessing the distribution patterns of genes within a predefined gene set, utilizing a gene table that is ranked according to its correlation with specific phenotype. This approach facilitates the evaluation of the genes’ contributions to the observed phenotypes [29]. In the present study, the genes from the integrated GEO datasets, referred to as AMI_Datasets, were initially ranked based on their logFC values. Subsequently, the R package clusterProfiler was employed to conduct GSEA on all genes encompassed within these integrated GEO datasets. The parameters established for this GSEA included a seed value of 2020, a total of 1000 computations, a minimum of 10 genes per gene set, and a maximum of 500 genes per gene set. Access to the Molecular Signatures Database (MSigDB) Database [16] was utilized retrieve the c2 gene sets, specifically the Cp. All. V2022.1. Hs. Symbols. The GMT file, which includes all canonical pathways, was utilized for the GSEA analysis. The criteria for screening in the GSEA were set at an adjusted p value < 0.05 and FDR value (q value) < 0.25. The p value correction method applied was the Benjamini-Hochberg (BH) procedure.

Verification of differential expressions and analysis of ROC curves for GRDEGs

To conduct a comprehensive analysis of the GRDEGs between the AMI and control group within the GEO datasets, comparative maps were generated based on the expression profiles of the identified GRDEGs. Then, the R package pROC (Version 1.18.5) [30] was employed to construct the ROC Curve for GRDEGs, facilitating the calculation of the Area Under the Curve (AUC) to assess the diagnostic utility of GRDEGs expression levels in relation to the incidence of AMI. Typically, the AUC values for the ROC curve range from 0.5 to 1. A value approaching 1 signifies a superior diagnostic performance. Specifically, an AUC between 0.5 and 0.7 reflects low diagnostic accuracy, an AUC from 0.7 to 0.9 indicates moderate diagnostic accuracy, while an AUC exceeding 0.9 denotes high diagnostic accuracy.

Protein-protein interaction (PPI) network and hub gene screening

The PPI Network is composed of proteins that engage in interactions with one another. These interactions are crucial for the transmission of biological signals and the modulation of gene expression. Furthermore, they are integral to vital life processes, including energy metabolism, metabolic activities, and the regulation of the cell cycle. Conducting a systematic analysis of protein interactions within biological systems is paramount for elucidating the operational mechanisms of proteins, comprehending the reaction pathways involved in biological signaling, as well as energy and substance metabolism, particularly under specific physiological conditions such as diseases. Additionally, it aids in deciphering the function interrelationships among proteins. A PPI network associated with GRDEGs was established utilizing the STRING database(https://string-db.org/) [31], employing a minimum interaction score threshold of 0.400.

The interconnected local areas within the PPI Network may signify molecular complexes that fulfill distinct biological roles. For subsequent analysis, we identified genes that engage with other genes within the PPI interaction framework as hub genes. The visualization of this network was accomplished using Cytoscape software [32].

The semantic comparison of GO [25] annotations offers a quantitative approach to assess the similarity among genes and genomes, establishing a crucial foundation for various bioinformatics analytical techniques. The R package GOSemSim (Version 2.30.0) [33] was employed to evaluate the functional correlation of the hub genes, and the functional correlation among hub genes was analyzed by functional similarity (Friends).

The GeneMANIA database (accessible at https://genemania.org/) [34] serves as a valuable tool for formulating hypotheses regarding gene functionality, evaluating gene lists, and prioritizing genes for subsequent functional investigations. When provided with a set of query genes, GeneMANIA identifies genes with functional similarities by leveraging an extensive array of genomics and proteomics datasets. In this context, it assigns weights to each functional genomic dataset in accordance with the anticipated relevance of the query. Additionally, GeneMANIA is employed for predicting gene functions; it identifies genes that are likely to share functional characteristics with a specified query gene, based on the interaction patterns observed. A PPI Network was established utilizing the functionally analogous genes associated with hub genes sourced from the GeneMANIA online platform.

Construction of mRNA-miRNA and mRNA-TF interaction network

The ENCORI database [35] represents the starBase platform (version 3.0). Within this database, various interactions are cataloged, including those between miRNA and ncRNA, miRNA and mRNA, ncRNA and RNA, RNA and RNA, as well as interactions involving RBP (RNA-binding proteins)-ncRNA and RBP-mRNA. These interactions are derived from data mining of CLIP-seq and degradation group sequencing, particularly for plant species. In our study, we employed the ENCORI database to identify miRNAs that were associated with hub genes, utilizing pancancer Num >7 as the criterion for selecting mRNA-miRNA interaction pairs. Furthermore, we utilized Cytoscape software to facilitate the visualization of the mRNA-miRNA interaction network.

The CHIPBase database (version 3.0) (accessible at https://rna.sysu.edu.cn/chipbase/) [36] utilizes data from ChIP-seq experiments involving DNA-binding proteins to identify a multitude of binding sequence matrices and their corresponding binding sites. Furthermore, it predicts the transcriptional regulatory interactions between millions of transcription factors (TFs) and genes. In our investigation, we exploited the CHIPBase database (version 3.0) to identify TFs that interact with hub genes. We established a screening criterion based on the total number of samples identified both upstream and downstream, requiring this sum to exceed 6 to filter mRNA-TF interaction pairs. To visualize the results of mRNA-TF interaction network, we employed Cytoscape software.

Immune infiltration analysis using SsGSEA algorithm

Single-Sample Gene-Set Enrichment Analysis (ssGSEA) [37] serves to quantify the relative level of immune cell infiltration within individual samples. First, various types of infiltrating immune cell identified and classified, including Activated CD8 + T cell, Activated dendritic cell, Gamma-delta T cell, Natural killer cell, and several subtypes of human immune cells, such as Regulatory T cells. Secondly, the enrichment scores derived from the ssGSEA analysis were employed to represent the relative abundance of each immune cell type within the samples, thereby generating an immune cell infiltration matrix. Following this, immune cells exhibiting significant differences between the two groups were identified for further examination. The correlation among immune cells was assessed using the Spearman algorithm, and the R package heatmap (Version 1.0.12) was utilized to create a correlation heatmap, illustrating the relationships among the immune cells. Additionally, the correlation between hub genes and immune cells was calculated using the Spearman algorithm, retaining results with p value < 0.05. The R package ggplot2 (Version 3.4.4) was then employed to generate a correlation bubble plot, effectively presenting the correlation analysis results between hub genes and immune cells.

Isolation of peripheral blood mononuclear cells (PBMCs)

To verify the results of bioinformatics analysis, 8 patients with AMI and 8 health controls were enrolled from Sir Run Run Hospital, Nanjing Medical University. The diagnosis of AMI was based on the fourth edition of the universal definition of myocardial infarction [38]. The exclusive criteria included severe hepatic or renal dysfunction, acute infectious diseases, autoimmune diseases, current usage of glucocorticoid, or malignant tumors. This study was performed in accordance with the principles stated in the Declaration of Helsinki [39] and approved by the Ethics Committee of Sir Run Run Hospital, Nanjing Medical University (No. 2023-SR-029). Written informed consent was obtained from each participant. PBMCs were isolated by using Human Peripheral Blood Lymphocyte Separation Medium (C0025, Beyotime, China).

Quantitative real-time PCR

Total RNA was extracted from the PBMCs by using Total RNA Extraction Kit (DP424, TIANGEN, China). RNA was reverse-transcribed into cDNA using PrimeScript™ RT reagent Kit (RR037A, Takara, Japan). Quantitative real-time PCR was performed using Maxima SYBR Green/ROX qPCR Master Mix (K0222, Thermo Scientific, USA). The relative expression levels were calculated by the 2^ΔΔCt method using glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as the internal reference gene. All the primer sequences were listed in Table S2.

Statistical analysis

The data processing and analytical procedures described in this study were conducted utilizing R software (Version 4.2.2). Unless stated otherwise, the statistical significance of normally distributed variables was assessed using the independent Student’s T-Test to compare continuous variables across two groups. For variables that did not conform to a normal distribution, the Mann-Whitney U Test (also known as the Wilcoxon Rank Sum Test) was employed to evaluate differences. The Kruskal-Wallis test was applied for comparisons involving three or more groups. Additionally, Spearman correlation analysis was utilized to determine the correlation coefficients among various molecules. All statistical p values were considered two-sided unless indicated otherwise, with a threshold of p < 0.05 deemed indicative of statistical significance.

Results

Merging of AMI data sets

The distribution boxplots were employed to assess the expression values of the datasets prior (Fig. 1A) and after (Fig. 1C) the removal of batch effects. In addition, PCA plot showed the distribution of low-dimensional features within the dataset before (Fig. 1B) and after (Fig. 1D) the batch effect correction. The results indicated that the batch effects present in the AMI dataset were largely mitigated following the application of the correction process.

Fig. 1.

Batch Effects Removal of GSE48060 and GSE29532. (A). Box plots of GEO datasets (AMI_Datasets) distribution integrated before batch removal. (B). PCA plot of the datasets (AMI_Datasets) before debatching. (C). Boxplots of the integrated GEO datasets (AMI_Datasets) distribution after going to batch. (D). PCA plot of the integrated GEO datasets (AMI_Datasets) after debatched. PCA, Principal Component Analysis; AMI, Acute Myocardial Infarction. The acute myocardial infarction (AMI) dataset GSE48060 is shown in light blue, and the acute myocardial infarction (AMI) dataset GSE29532 is shown in light red

Glycolysis-related differentially expressed genes (GRDEGs) in AMI

The integrated GEO datasets (AMI_Datasets) were categorized into two distinct groups: the AMI group and the control group. The analysis yielded a total of 306 DEGs with 151 upregulated genes and 155 downregulated genes (Fig. 2A). Among these DEGs, 11 GRDEGs, including PYGL, DSC2, CHST12, RRAGD, TGFA, PPBP, SMAD5, PGAM2, HK3, PFKFB3, and PRKACB, were identified by combination with the database of GRGs (Fig. 2A and B). The chromosome mapping showed that more GRDEGs were located on chromosomes 5 and 7, with SMAD5 and HK3 in chromosomes 5 and CHST12 and PGAM2 on chromosome 7, respectively (Fig. 2D). GO and KEGG enrichment analysis (Fig. 3A and B and Table S3) found that these GRDEGs are predominantly associated with BP including striated muscle contraction, hexose metabolic processes, monosaccharide metabolic processes, muscle contraction, and muscle system processes (Fig. 3C), CC including secretory granule lumen, cytoplasmic vesicle lumen, vesicle lumen, cAMP-dependent protein kinase complex, and desmosome (Fig. 3D), and MF including carbohydrate kinase activity, growth factor activity, ubiquitin protein ligase binding, ubiquitin-like protein ligase binding, and receptor-ligand activity (Fig. 3E). Additionally, the GRDEGs are found to be enriched in KEGG biological pathways, notably in fructose and mannose metabolism (Fig. 3F).

Fig. 2.

Differential Gene Expression Analysis. (A). Volcano plot of differentially expressed genes analysis between AMI and Control groups in the integrated GEO datasets (AMI_Datasets). (B). DEGs and GRGs Venn diagram in the integrated GEO datasets (AMI_Datasets). (C). Heat map of GRDEGs related to glycolysis in the integrated GEO datasets (AMI_Datasets). (D). Chromosomal mapping of GRDEGs. AMI, Acute Myocardial Infarction; DEGs, Differentially Expressed Genes; GRGs, Glycolysis Related Genes; GRDEGs, Glycolysis Related Differentially Expressed Genes. Light red is the AMI group, light blue is the Control group. In the heat map, red represents high expression and blue represents low expression

Fig. 3.

GO and KEGG Enrichment Analysis for GRDEGs (A-B). GO and KEGG enrichment analysis results of GRDEGs Bar graph (A) and bubble plot (B) show: BP, CC, MF and biological pathway. GO terms and KEGG terms are shown on the ordinate. (C-E). The GO enrichment analysis results of GRDEGs network diagram showing: BP (C), CC (D), MF (E). Pink nodes represent items, blue nodes represent molecules, and lines represent the relationship between items and molecules. (F). GRDEGs pathway enrichment analysis results pathway map visualization display: Fructose and mannose metabolism. GRDEGs, Glycolysis Related Differentially Expressed Genes; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; BP, Biological Process; CC, Cellular Component; MF, Molecular Function

Gene set enrichment analysis (GSEA)

To assess the influence of gene expression levels across the integrated AMI Datasets, GSEA was used to investigate the relationship between the expression of all genes and the BP, CC, and MF (Fig. 4A and Table S4). The results revealed significant enrichment of genes in Neutrophil Degranulation (Fig. 4B), Activation of IRF3 and IRF7 Mediated By TBK1/IKK/Epsilon pathway (Fig. 4C), NABA Ecm affiliated (Fig. 4D), Metabolism of RNA (Fig. 4E) and other relevant functions.

Fig. 4.

GSEA for Acute Myocardial Infarction. (A). Mountain plot presentation of 4 biological functions from GSEA of integrated GEO datasets (AMI_Datasets). (B-E). GSEA showed that all genes were significantly enriched in Neutrophil Degranulation (B), Activation of IRF3 IRF7 Mediated by TBK1 IKK Epsilon (C), and activation of IRF3 IRF7 mediated by TBK1 IKK epsilon (C). NABA Ecm affiliated (D), Metabolism of RNA (E). AMI, Acute Myocardial Infarction; GSEA, Gene Set Enrichment Analysis

ROC curve analysis of GRDEGs

There were significant differences of the 11 GRDEGs (PYGL, DSC2, CHST12, RRAGD, TGFA, PPBP, SMAD5, PGAM2, HK3, PFKFB3, PRKACB) between AMI group and control group (Fig. 5A). The ROC analysis (Fig. 5B and E) indicated that the levels of PYGL, CHST12, RRAGD, TGFA, PGAM2, HK3 and PFKFB3 exhibited moderate accuracy in distinguishing the AMI patients (0.7 < AUC < 0.9), while DSC2, PPBP, SMAD5, and PRKACB demonstrated lower accuracy (0.5 < AUC < 0.7).

Fig. 5.

Differential Expression Validation and ROC Curve Analysis. (A). Group comparison plots of GRDEGs in the AMI group and Control group of integrated GEO datasets (AMI_Datasets). (B-E). ROC curves of GRDEGs in integrated GEO datasets (AMI_Datasets), PYGL, DSC2 and CHST12 (B); RRAGD, TGFA and PPBP (C); SMAD5, PGAM2 and HK3 (D); PFKFB3 and PRKACB (E). *, p < 0.05; **, p < 0.01; ***, p < 0.001. AMI, Acute Myocardial Infarction; GRDEGs, Glycolysis Related Differentially Expressed Genes; ROC, Receiver Operating Characteristic; AUC, Area Under the Curve. TPR, True Positive Rate; FPR, False Positive Rate. Light blue represents the Control group and light red represents the AMI group

Protein-protein interaction (PPI) network

PPI Network analysis showed that five GRDEGs, namely PYGL, HK3, PGAM2, PFKFB3 and PRKACB, were related to each other and were then selected as hub genes for further analysis (Figure S2A). Functional similarity scores (Friends) showed that the gene closest to the cut-off value (0.60) was HK3, indicating that HK3 might contribute significantly to AMI (Figure S2B). Further interaction network involving the five central hub genes identified 20 proteins with similar functions (Figure S2C).

Interaction network between mRNA-miRNA and mRNA-TF

The miRNAs related to the 5 hub genes (PYGL, HK3, PGAM2, PFKFB3, PRKACB) were obtained from ENCORI database. The mRNA-miRNA Interaction Network identified 3 hub genes (PYGL, PFKFB3, PRKACB) with 59 interacted miRNAs (Figure S3A and Table S5). The transcription factors (TFs) lined to the hub genes were obtained from ChIPBase database and 5 hub genes (PYGL, HK3, PGAM2, PFKFB3, PRKACB) with 43 interacted TFs were identified (Figure S3B and Table S6).

Immune infiltration analysis

The integrated AMI Datasets were analyzed to assess the immune cell infiltration abundance of 28 immune cell types by using the ssGSEA algorithm. Nine different immune cell types, including Effector memory CD4 + T cell, Regulatory T cell (Treg), T follicular helper cell (Tfh), Type 1 T helper cell, Type 2 T helper cell, Macrophage, MDSC, Monocyte, and Neutrophil, were found between the AMI group and control group (Fig. 6A). A correlation heatmap demonstrated predominantly positive correlations among most of the nine types of identified different immune cells (Fig. 6B). Further correlation analyses were performed between the five hub genes (PYGL, HK3, PGAM2, PFKFB3, PRKACB) and nine immune cell types (Fig. 6C). Our results showed that the expression of PYGL was positively correlated with Neutrophils (r = 0.81, p < 0.05), while the expression of PRKACB was negative correlated with Macrophages (r = −0.72, p < 0.05).

Fig. 6.

Immune Infiltration Analysis by ssGSEA Algorithm. (A). Group comparison plots of immune cells in the AMI and Control groups of the integrated GEO datasets (AMI_Datasets). (B). Correlation heatmap of immune cell infiltration abundance in the integrated GEO datasets (AMI_Datasets). (C). Bubble plot of hub genes correlation with immune cell infiltration abundance in the integrated GEO datasets (AMI_Datasets). (D) The expression levels of the five hub genes (PYGL, HK3, PGAM2, PFKFB3, PRKACB) in the PBMCs isolated from patients with AMI and healthy controls ssGSEA, single-sample Gene-Set Enrichment Analysis; AMI, Acute Myocardial Infarction. ns, no statistical significance; *, p < 0.05; **, p < 0.01; ***, p < 0.001. Light blue is the Control group and light red is the AMI group. Red shows positive correlation, blue shows negative correlation. The depth of the color represents the strength of the correlation

Validation of hub genes

To verify the bioinformatics analysis results of the GRDEGs, we further examined the expression levels of the five hub genes (PYGL, HK3, PGAM2, PFKFB3, PRKACB) in the PBMCs isolated from patients with AMI and healthy controls. Consistently, our quantitative real-time PCR showed that the mRNA expression levels of PYGL, HK3, PGAM2, PFKFB3, and PRKACB were significantly higher in patients with AMI than in those controls (Fig. 6D).

Discussion

AMI, widely recognized as a heart attack, represents a critical medical emergency characterized by the abrupt cessation of blood flow to the cardiac muscle, resulting in tissue injury [40]. The urgency of addressing AMI is underscored by its potential to cause severe complications, including heart failure, arrhythmias, and sudden cardiac death [41]. Glycolysis, a vital metabolic pathway, is instrumental in cellular energy production, particularly under the ischemic conditions during AMI [42]. In this study, by integrating comprehensive bioinformatics approaches, we identified 11 GRDEGs, which may have the potential to function as diagnostic biomarkers and therapeutic targets for AMI.

Among the identified GRDEGs, PYGL (glycogen phosphorylase L) stands out as a crucial enzyme in glycogen metabolism, catalyzing the conversion of glycogen to glucose-1-phosphate [43]. Furthermore, the role of PYGL in glycogenolysis aligns with the metabolic shifts observed in ischemic heart disease, where rapid glucose mobilization is essential for sustaining cardiac function under hypoxic conditions [43]. Previous research has underscored the importance of glycogen metabolism in cardiac stress responses, further validating the relevance of PYGL in AMI pathology [44]. The elevated expression of PYGL in AMI patients may suggest its involvement in meeting the increased metabolic demands of ischemic cardiac tissue.

The involvement of AMI-related GRDEGs in essential biological processes such as muscle contraction, hexose metabolism, and fructose/mannose metabolism suggest that alterations in glycolysis-related genes might not only affect energy metabolism but also influence cardiac contractility during AMI. This is consistent with recent studies showing that metabolic remodeling plays a crucial role in heart failure progression following AMI [45].

Diagnostic potential of GRDEGs was further supported by ROC curve analysis, which indicated that a panel of GRDEGs, particularly PYGL, HK3, and PGAM2, might serve as promising biomarkers for AMI diagnosis. However, further validation in larger and more diverse patient cohorts is essential to confirm their clinical applicability.

Our construction of mRNA-miRNA and mRNA-TF interaction networks revealed intricate regulatory relationships involving the identified hub genes. This observation suggests that the expression of these GRGs in AMI is tightly regulated by various mechanisms. Future investigations should focus on experimentally validating these interactions and exploring their functional consequences in the context of AMI.

Our ssGSEA analysis revealed significant differences in immune cell infiltration between AMI and control subjects, with some hub genes showing notable correlations with specific immune cell populations. This finding highlights the potential interplay between metabolic alterations and immune responses in AMI, a concept that has gained increasing attention in recent years [46–48]. The immune response following AMI is a critical determinant of disease progression and outcomes [49]. Neutrophils play a pivotal role in the early phase of AMI by infiltrating the myocardium and releasing matrix-degrading enzymes and reactive oxygen species, which contribute to tissue damage and post-infarction remodeling [50]. Additionally, the IL-1 signaling pathway has been identified as a significant mediator of inflammation in AMI, while anti-IL-1 therapy shows promising effects on reducing inflammation and improving cardiac function [51]. The infiltration of immune cells, including neutrophils and macrophages, is regulated by various cytokines and chemokines, which further amplify the inflammatory response [52]. Our study identified significant differences in immune cell infiltration between AMI patients and controls, with certain GRDEGs showing strong associations with immune cell types. For instance, genes such as PYGL and HK3 were found to have high diagnostic potential, as indicated by their ROC curve AUC values. These genes were also implicated in the regulation of immune cell infiltration, suggesting their role in the inflammatory response during AMI [53]. Further investigation into how glycolysis-related genes modulate immune cell function in the infarcted myocardium could provide new insights into AMI pathogenesis and potential therapeutic strategies [54].

Although our study provides valuable insights, there are still several limitations that point to important future research directions. Firstly, our analysis was based on publicly available microarray data, which might not capture the full complexity of gene expression changes in AMI. Future studies using RNA-sequencing data could provide more comprehensive view. Additionally, our findings were only associative without proven causality. Therefore, future research should prioritize functional studies to elucidate the specific roles of the identified GRDEGs in AMI pathophysiology, explore their potential as therapeutic targets, and investigate their relationship with other metabolic pathways. The clinical relevance of these findings also needs to be established through prospective studies in diverse patient populations. These efforts will be critical in translating our findings into clinically relevant applications for improved diagnosis and treatment of AMI.

Conclusions

Our study provides valuable insights of GRDEGs in AMI and protrude potential biomarkers and therapeutic targets. These results enhance our comprehension of the intricate relationship between metabolic processes and cardiac performance in AMI, potentially guiding the creation of innovative diagnostic and treatment approaches for this significant health issue.

Author contributions

W.G. and B.J. Z. contributed to the conception and design of the study. Y.M. L., K.Z. Z., Y.R. Q., and J.W. L. contributed to data analysis. Y.M. L., Q.H. L. and K.Z. Z. drafted the manuscript. W.G. and B.J. Z. revised the manuscript. All authors read and approved the final submission.

Funding

This work was supported by grants from the National Natural Science Foundation of China (No.81970217 to Wei Gao and No. 82300400 to Kang-Zhen Zhang), the Natural Science Foundation of Jiangsu Province (No. BK20220829 to Qian-Hui Li), the Jiangsu Commission of Health (No.LKZ2023005 to Wei Gao), the project of Zhongda Hospital Affiliated to Southeast University for cultivating academic talent (No. CZXM-GSP-RC22 to Wei Gao), the Zhongda Hospital Affiliated to Southeast University, Jiangsu Province High-Level Hospital Pairing Assistance Construction Funds (No. zdlyg10 to Wei Gao), the Nanjing Commission of Health (No. YKK24257 to Yi-Min Li),.

Data availability

All data and materials used for this study are displayed or can be displayed upon request.

Declarations

Ethics approval

This study was performed in accordance with the principles stated in the Declaration of Helsinki [39] and approved by the Ethics Committee of Sir Run Run Hospital, Nanjing Medical University (No. 2023-SR-029).

Human ethics and consent to participate declarations

Written informed consent was obtained from each participant.

Competing interests

The authors declare no competing interests.