Identification of atrial fibrillation-related genes through transcriptome data analysis and Mendelian randomization

Atrial fibrillation (AF), the most common persistent arrhythmia, is associated with heightened risks of stroke, heart failure, sudden death, and cardiogenic embolism, leading to increased hospitalizations and mortality rates (1). Presently, around 34 million people worldwide are affected by AF, a number projected to significantly increase owing to an aging global population and a rise in risk factors like obesity, hypertension, and diabetes mellitus. This escalation could boost the economic strain on individuals and societies by as much as 60% by the year 2050 (2). The sinoatrial node, the heart's primary pacemaker, is encased in a thin layer of epicardial adipose tissue. Its interior connects to the atrial myocardial layer, which primarily consists of pacemaker cells, transitional cells, and a few Purkinje fibers. It generates impulses that are transmitted to the atrial muscle and the atrioventricular node through the internodal pathways (3). Research has indicated that atrial fibrillation characterized by long RR intervals often results from a second-degree or higher atrioventricular conduction block. In patients experiencing atrial fibrillation and long RR intervals (Tachy-Brady syndrome), radiofrequency ablation can mitigate the effects of the vagus nerve, reduce concealed conduction in the atrioventricular node, restore sinoatrial node functionality, and successfully sustain sinus rhythm (4). This treatment eliminates the long intervals, obviating the need for pacemaker therapy. Numerous factors contribute to the development of atrial fibrillation (AF), such as cigarette smoking, alcohol consumption, hypertension, obesity, diabetes mellitus, heart attacks, heart failure, and other risk factors (5). Evidence from epidemiology shows that AF tends to run in families; having a parent with AF doubles the likelihood that their children will also develop this condition (6). The connections between specific genetic markers and the development of AF, however, remain poorly understood. Additionally, current drugs for treating AF often fall short in efficacy and carry potential for adverse effects (7). Compared to pharmacological approaches, ablation generally offers greater effectiveness but involves invasive techniques, possible complications, and a significant chance of recurrence over the long term (8). Thus, gaining a clearer understanding of AF's molecular mechanisms is essential for developing novel diagnostic biomarkers and therapeutic targets.

Recent progress in high-throughput gene chip and transcriptome sequencing technologies has enhanced their application in cardiovascular disease research. These technologies range from detecting copy number variations at the genomic level to exploring gene expression at the transcriptome level, with integrated bioinformatics emerging as a key approach for pinpointing potential biomarkers. For instance, Zhang et al. (9) identified CXCR4, IGFBP2, IGFBP3, and FHL2 as genes linked to the risk of AF using bioinformatics analysis. CXCR4, functioning as a receptor for CXCL12/SDF-1, significantly influences cellular growth, differentiation, stress responses, and inflammatory reactions. Studies have shown that myocardial cells in atrial fibrillation display elevated CXCR4 expression and heightened inflammatory responses compared to controls. This may promote inflammation via the PI3K/AKT signaling pathway, potentially leading to AF (10). IGFBP2, part of the insulin-like growth factor binding protein (IGFBP) family, is involved in cell migration, tumor invasion, cell proliferation, and tumor angiogenesis. Research indicates IGFBP2 inhibits PTEN and enhances PTEN tyrosine phosphorylation through dimerization with RPTPβ, thus fostering vascular smooth muscle cell growth (11). Given the role of the PTEN/AKT/mTOR pathway in cardiac hypertrophy and fibrosis (12), IGFBP2's impact on PTEN could affect AF development. It activates integrin β1 and downstream pathways, necessitating ILK for cell motility induction and NF-κB activation (13). NF-κB participates in the regulation of inflammatory cytokines, thrombosis, and fibrosis genes (14), suggesting the IGFBP2/integrin/ILK/NF-κB axis might play a role in the development and progression of AF. IGFBP3, the principal binding target for IGF-1, modulates cellular responses to IGF-1 either by forming specific complexes with or without IGF-1 signaling pathways, thus moderating free IGF-1 levels in the body and restricting IGF-1 signaling activation in target cells. Research by Liakouli et al. shows that IGFBP-3 may facilitate myocardial fibrosis by promoting mesenchymal cell proliferation and extracellular matrix deposition (15). Busch et al. found in the Study of Health in Pomerania (SHIP) that a low IGF-1/IGFBP-3 ratio correlates with a higher incidence of AF (16). FHL2 protein, as a functional partner of the α or β subunit of the delayed rectifier potassium current, regulates potassium ion channels and is essential for the delayed rectifier potassium current formation (17). Pape et al. have observed that expression of Kv1.5 (encoded by KCNA5), a major component of Ikur, is reduced in the atrial muscle of AF patients. Since Ikur is critical for the atrial muscle cell action potential repolarization phase (AERP) and highly selective to atrial tissue, its downregulation reduces APD frequency adaptability and the ERP/APD ratio, thus fostering the onset and persistence of atrial fibrillation (18, 19). In a similar vein, Wang et al. (20) discovered ZBTB20, ERBB4, CREB1, and BNIP3l as potential indicators for the early diagnosis and prospective treatment of AF. eQTLs, which are genetic variants that affect gene expression, play a pivotal role in elucidating biological pathways. They do so by linking genetic variations to gene expression changes and identifying pertinent biomarkers within GWAS (21). Recent GWAS on AF (22) have pinpointed around 140 genetic variant loci linked to the disorder. Yet, establishing a direct causal link between these markers and the pathophysiology of AF still requires more definitive evidence.

MR serves as an analytic tool to determine causal relationships between exposures and outcomes, using genetic variants as instrumental variables (IVs) (23). This method leverages the random distribution of alleles during meiosis, thus minimizing the influence of confounding factors that frequently affect traditional epidemiological studies (24). Additionally, the allocation of genotypes precedes disease onset, thereby precluding any possibility of reverse causation (25). In our research, we merged bioinformatics with MR to pinpoint differentially expressed eQTLs as potential biomarkers for diagnosing and treating AF. Our study further explored the mechanisms of AF pathogenesis through analyses of enrichment and immune cell infiltration.

AF is a prevalent cardiovascular condition with an incidence that escalates with age. Recent advancements have led to the identification of numerous novel biomarkers for the early diagnosis, mechanistic exploration, and drug target identification in AF. Nonetheless, our comprehension of AF's genetic underpinnings remains markedly insufficient. Consequently, there is a critical necessity to unravel AF's pathogenesis and identify viable diagnostic and therapeutic targets to formulate an effective framework for its prevention, early detection, and management. In this research, we employed a holistic approach, integrating bioinformatics analysis with GWAS data mining, to identify 8 genes: AMFR, LAP3, RABAC1, TPSB2, G3BP2, ITGB2, QPCT, and TRIM22—implicated in AF's etiology. Through functional enrichment analysis, we discovered these genes predominantly engage in protein polyubiquitination and neutrophil degranulation, with notable enrichment in pathways such as the SREBP signaling pathway. Furthermore, our investigation into immune cell infiltration revealed that plasma cells and NK cells were abundantly expressed in AF patients. Notably, among the 8 genes, QPCT, LAP3, ITGB2, G3BP2, and RABAC1 exhibited correlations with diverse immune cells.

The AMFR, also known as gp78, is a receptor that binds to AMF ligands, facilitating signal transduction through the activation of G protein-coupled receptors and tyrosine kinases. It plays a pivotal role in the recognition of misfolded proteins, marking them for ubiquitination and subsequent degradation by the proteasome through the endoplasmic reticulum-associated degradation (ERAD) pathway. This process prevents the accumulation of misfolded proteins within the endoplasmic reticulum lumen, thereby maintaining the integrity of the secretory pathway, an essential protective mechanism in vivo (47). In this study, we present the first evidence that AMFR may lower the incidence of AF and explore the expression profile of AMFR to elucidate its mechanism of action in AF further. Leucine aminopeptidases (LAPs), a category of zinc-dependent metalloenzymes, play a crucial role in protein metabolism. LAP3, a significant LAP member, specializes in the cleavage of N-terminal amino acid polypeptides, facilitating peptide bond hydrolysis (48). Its function involves cleaving proteins into smaller peptide molecules via the ubiquitin-proteasome pathway, which is critical in the antigen-presenting intracellular terminal phase of protein degradation. To our knowledge, the regulatory role of LAP3 in AF development remains unreported. RABAC1 encodes the membrane protein PRA1, which is instrumental in protein transport and has been shown to interact with various Rab GTPases (49). Rab proteins are known to bind tightly to GDP Dissociation inhibitor (GDI), forming complexes. This observation has led to the hypothesis that an effector, known as the GDI displacement factor (GDF), exists at the donor membrane. This factor is believed to play a crucial role in the dissociation of Rab-isoprenoid base tails and their reinsertion into the donor membrane (50). In vivo studies suggests that PRA1 knockout results in abnormal ER and Golgi apparatus phenotypes, indicating a more structural role in the early secretory pathway (51). However, the precise function of PRA1 in AF remains to be elucidated. In this study, by screening genes causally associated with AF through eQTL data and intersecting them with AF-associated DEGs from bioinformatics analysis, we confirmed that RABAC1 is overexpressed in AF patients and may serve as a protective factor in reducing morbidity. G3BP2, belonging to the RNA-binding protein family, is a critical component of stress granule assembly. It regulates gene expression in response to various environmental stimuli, affecting cell proliferation, protein degradation, and RNA stability (52). G3BP family proteins play roles in various diseases, including cancer, viral infections, Alzheimer's disease, and cardiovascular diseases, exhibiting diverse biological functions such as RNA metabolism, stress granule formation, signaling, cell cycle regulation, and protein ubiquitination degradation (53), however, the precise mechanisms underlying these processes require further exploration. Recent studies have indicated that G3BPs exhibit abnormal expression levels in cardiovascular diseases, with their mechanism of action primarily linked to stress granule formation. Recent studies have demonstrated that the protein encoded by the G3BP2 gene interacts with the NF-κB pathway (54). NF-κB, a crucial regulator of inflammatory responses, remains inactive when bound to IκB proteins. Upon stimulation, it dissociates from IκB and translocates to the nucleus to regulate the expression of target genes for inflammatory cytokines such as TNF-α, IL-1β, and IL-6, thereby initiating an inflammatory cascade. Ye et al. reported that inhibiting the NF-κB pathway with 5,7-dihydroxyflavone could reduce levels of pro-inflammatory cytokines, potentially decreasing susceptibility to AF (55). Additional research has shown that colchicine can lower rat susceptibility to AF by reducing inflammation-mediated atrial fibrosis (56). NF-κB is pivotal in gene transcription, noted for its rapid transcriptional activation (57). Previous studies have also demonstrated that AF in diabetic rats induced by thiazolidinediones was associated with activation of the NF-κB pathway, leading to a pro-inflammatory state, myocardial hypertrophy, and fibrosis (58). Therefore, it is hypothesized that G3BP2 may promote the onset of AF by interacting with the NF-κB pathway and triggering an inflammatory response. ITGB2, a complex heterodimer composed of α and β chains, functions as an integral cell membrane receptor, facilitating a multifaceted role in extracellular matrix (ECM)-cytoskeletal linkage and biochemical and mechanical signal transduction between the cell and its surroundings (59). Friedrichs et al. (60) demonstrated that ITGB2-mediated infiltration of polymorphonuclear neutrophils contributes to atrial fibrosis, thereby heightening susceptibility to AF in angiotensin II treated mice. Previous research has illustrated that ITGB2 enhances the cross-cell migration of white blood cells in both animal and cell culture experiments, impairing endothelial barrier integrity and participating in atherosclerosis development (61). As a risk factor, ITGB2 has been implicated in accelerating myocardial infarction and arterial thrombotic cerebral infarction via cell adhesion molecular pathways (62). This analysis further posits that ITGB2 could act as a risk factor, increasing AF incidence. The QPCT gene encodes glutaminyl cyclase, an enzyme responsible for the post-translationally modification of proteins by converting N-terminal glutamate into pyroglutamate. This modification increases protein resistance to protease degradation, enhances hydrophobicity and neurotoxicity, and predisposes the proteins to aggregation (63). Yamada et al. (64) demonstrated a correlation between the expression of the QPCT gene and hypertension in Japanese men. To date, research on QPCT in AF has been sparse, with its underlying molecular mechanisms yet to be fully elucidated. The TRIM protein family, prevalent across multicellular animals, functions as immunomodulatory proteins and E3 ubiquitin ligases (65). TRIM22, a member of this family, plays a critical role in modulating the body's immune-inflammatory response. Several studies have highlighted the significance of the inflammatory mechanism in the pathogenesis and progression of AF (66). Clinical and experimental research has identified elevated levels of serum inflammatory biomarkers and cardiac tissue inflammatory markers or cytokines/growth factors in individuals with AF (67). Inflammatory responses in the atrial wall can lead to increased oxidative stress, cardiomyocyte apoptosis, fibrosis, and abnormalities in gap junction regulation and intracellular calcium handling, potentially contributing to the development of arrhythmias and AF (68). TRIM22 has been implicated in the development of various cancers, including gastric (69), ovarian (70), and cervical cancer (71). Numerous epidemiological studies have established a strong link between malignant tumors and atrial fibrillation (AF), noting a higher prevalence of AF among cancer patients compared to those without cancer. A case-control study reported an AF incidence of 7.4% among patients diagnosed with tumors, which is higher than the 6.8% observed in the non-tumor group (72). Additionally, a long-term prospective study lasting 16.3 years showed an increased risk of AF in cancer patients, with a hazard ratio (HR) of 2.47 (73). Similarly, a large meta-analysis involving 5,889,234 patients found a 47% increased risk of AF in patients with malignant tumors within 90 days of diagnosis, compared to those without malignant tumors (74). Therefore, early prevention, recognition, and management of AF should be prioritized in populations with malignant tumors.

Our further exploration into the biological functions and key pathways of the 8 CGs revealed significant insights through GO enrichment analysis. These genes are predominantly associated with protein polyubiquitination and neutrophil-related pathways, including neutrophil degranulation, regulation of neutrophil activation, and neutrophil activation within the immune response, all of which were notably enriched. Protein polyubiquitination, a prevalent post-translational modification, involves a complex enzyme-linked reaction sequence utilizing ubiquitin-activating enzyme E1, ubiquitin-binding enzyme E2, and ubiquitin ligase E3 (75). Several E3 ubiquitin ligases, such as WWP2, SMURF1, and SMURF2, have been identified as key regulators in myocardial fibrosis (76). Steinle et al. (77) observed that both protein modification and neutrophil activation play roles in systemic inflammation and oxidative stress. The KEGG pathway analysis highlighte the SREBP signaling pathway, cytoplasmic sequestering of transcription factors, endoplasmic reticulum mannose trimming, and the cellular response to sterol depletion as main components.

We have also determined the effects of mRNA and miRNA gene interactions on disease. miRNAs have been shown to modulate a wide range of signaling pathways and cellular processes and to participate in intercellular communication for their biological functions (78). Recent studies have shown that mi RNAs control the function of various cells in the heart, such as cardiac myocytes, endothelial cells, smooth muscle cells, and fibroblasts, and play an important role in cardiovascular diseases, such as myocardial infarction, myocardial hypertrophy, fibrosis, cardiac failure, cardiac arrhythmia, inflammation, and atherosclerosis (79). Some miRNAs, including miRNA-21, miRNA-26, miRNA-29, and miRNA-133, have been reported to contribute to the process of atrial fibrosis by participating in the regulation of different signaling pathways (80), which suggests that miRNAs may play an important role in promoting the maintenance of AF.

Inflammation has been identified as a key risk factor for AF, with the immune response it triggers playing a crucial role in the initiation and progression of AF. Litviňuková et al. (81) discovered a substantial presence of immune cells in the atrial tissues of 14 donated hearts, with immune cells constituting 10.4% of the atrial samples. This finding underscores the significance of local immune cells in cardiac pathophysiological alterations. Wu et al. (82) found that CD8+ T cells were significantly more prevalent in AF patients than in individuals with normal rhythm. Zhou et al. (83) observed that macrophage M2, activated NK cells, and neutrophils might be implicated in the pathogenesis of AF. Our findings indicate an increased expression of plasma cells and NK cells in AF patients. Plasma cells, or effector B cells, are involved in synthesizing and storing antibodies and play a part in humoral immune responses, appearing in various diseases such as lung adenocarcinoma, myeloma, and SLE (84–86). NK cells, located in lymphoid organs and peripheral tissues, can influence disease progression at inflammatory sites through cytokine and chemokine production, cell-to-cell interactions, and engagements with other immune cells. However, the precise mechanisms through which inflammation impacts AF remain elusive. Current research suggests that inflammation contributes to atrial remodeling via oxidative stress. Structural and electrophysiological remodeling of the atrium is widely regarded as pivotal in the pathophysiology of AF, thereby influencing the condition's persistence and recurrence.

We investigated the interactions between AMFR, LAP3, RABAC1, TPSB2, G3BP2, ITGB2, QPCT, TRIM22, and immune cell infiltration in greater depth. QPCT was found to inversely correlate with Tregs and activated NK cells. Similarly, LAP3 was inversely related to CD8+ T cells and activated Mast cells, while ITGB2 displayed a positive relationship with Plasma cells and a negative one with Neutrophils. G3BP2 correlated positively with CD4+ memory resting T cells and negatively with activated Mast cells. RABAC1 exhibited positive correlations with Tregs and activated Mast cells, yet it had an inverse relationship with gamma delta T cells. In contrast, TRIM22, AMFR, and TPSB2 showed no discernible correlations with immune cells. These results highlight the complex interactions between AF-related genes and immune cells, emphasizing the importance of further exploration into these dynamics. Notably, QPCT, LAP3, ITGB2, G3BP2, and RABAC1 are identified as crucial in AF pathogenesis through their modulation of immune infiltration.

This study offers several advantages. Primarily, this study is the first, to our knowledge, to integrate bioinformatics with MR for the analysis of AF biomarkers, revealing a causal link between 8 CGs and AF. Furthermore, the utilization of MR Analysis ensures that our results are robust against reverse causality. Additionally, we conducted multiple sensitivity analyses to verify the consistency of the causal estimates, affirming the reliability of our conclusions. However, there are limitations to our study: (1) The microarray analysis involved a small sample size, which could introduce bias and underscores the necessity for more extensive future studies. (2) Although we merged two GEO datasets and identified eight CGs by intersecting genes significant in both DEGs and MR Analysis, the exact roles of these CGs in AF require further elucidation through in vitro and in vivo studies. (3) Since our sample solely comprised individuals from a European background, this may limit the generalizability of our findings to other ethnic groups.

In summary, our study leveraged bioinformatics and MR Analysis to investigate genes associated with AF, identifying 8 genes with causal links to the onset and progression of AF. AMFR, LAP3, RABAC1, and TPSB2 were identified as protective factors that could potentially reduce the incidence of AF, while G3BP2, ITGB2, QPCT, and TRIM22 were deemed risk factors that could promote the disease's development. Furthermore, GO and KEGG enrichment analyses revealed that these 8 GCs were predominantly involved in protein ubiquitination and the SREBP signaling pathway. Analysis of the association between gene expression and immune cell infiltration highlighted that LAP3, RABAC1, G3BP2, ITGB2, and QPCT might influence AF through the regulation of immune cell infiltration. These findings suggest that the genes identified could serve as novel biomarkers and potential therapeutic targets for AF, offering insights into the molecular mechanisms underlying AF pathology.

Microarray datasets related to AF were extracted from GEO, and summary level data of SNPs associated with eQTL and AF were obtained from GWAS. The authors thank all the investigators for sharing these data.

The author(s) declare financial support was received for the research, authorship, and/or publication of this article.

This research was funded by the key research and development Plan of Shaanxi Province (grant numbers: 2023-YBSF-636 and 2021SF-396), and the Shaanxi Province Traditional Chinese Medicine Research Project (grant number: 2021-ZZ-JC042).

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/. 10

Ethical approval was not required for the studies involving humans because this study was conducted using a public database for research purposes.

YZ: Writing – original draft. QL: Methodology, Writing – review & editing. YN: Formal Analysis, Writing – original draft. WZ: Software, Writing – review & editing.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fcvm.2024.1414974/full#supplementary-material↗