Non-alcoholic fatty liver disease and heart failure: A comprehensive bioinformatics and Mendelian randomization
Yayun Zhang1, Lu Feng1, Xin Guan2
1Shanxi Bethune Hospital, Shanxi Academy of Medical Sciences, Tongji Shanxi Hospital, Third Hospital of Shanxi Medical University, Taiyuan, China.
Insights
This study reveals shared molecular signatures between heart failure (HF) and non-alcoholic fatty liver disease (NAFLD). Bioinformatics and Mendelian randomization confirm a bidirectional causal relationship, suggesting shared pathways and therapeutic potential.
Area of Science:
- Cardiovascular Medicine
- Hepatology
- Genetics
Background:
- Heart failure (HF) and non-alcoholic fatty liver disease (NAFLD) are prevalent global health issues.
- A complex interrelationship exists between HF and NAFLD, necessitating investigation into shared mechanisms.
Purpose of the Study:
- To identify shared molecular biomarkers between HF and NAFLD.
- To investigate the causal relationship between HF and NAFLD using bioinformatics and Mendelian randomization (MR).
Main Methods:
- Analysis of gene expression datasets (GEO) for differentially expressed genes (DEGs) in HF and NAFLD.
- Construction of protein-protein interaction (PPI) networks and pathway enrichment analyses (GO, KEGG).
- Two-sample Mendelian randomization (MR) analysis using genome-wide association study (GWAS) data for HF and NAFLD.
Main Results:
- Identification of 4032 DEGs in NAFLD and 286 DEGs in HF.
- Top 10 hub genes (e.g., CD163, CXCL10, MRC1) significantly enriched in immune response pathways.
- MR analysis confirmed a bidirectional causal relationship: NAFLD increases HF risk (OR=1.024) and HF increases NAFLD risk (OR=1.117).
Conclusions:
- Novel molecular signatures common to NAFLD and HF were identified.
- A bidirectional causal relationship between NAFLD and HF was confirmed, indicating shared underlying pathophysiology.
- Findings suggest potential for targeted therapeutic interventions for both conditions.
Aims:
Heart failure (HF) and non-alcoholic fatty liver disease (NAFLD) are significant global health issues with a complex interrelationship. This study investigates their shared biomarkers and causal relationships using bioinformatics and Mendelian randomization (MR) approaches.
Methods:
We analysed NAFLD and HF datasets from the Gene Expression Omnibus (GEO). The GSE126848 dataset included 57 liver biopsy samples [14 healthy individuals, 12 obese subjects, 15 NAFL patients and 16 non-alcoholic steatohepatitis (NASH) patients]. The GSE24807 dataset comprised 12 NASH samples and 5 healthy controls. The GSE57338 dataset included 313 cardiac muscle samples [177 HF patients (95 ischaemic heart disease patients and 82 idiopathic dilated cardiomyopathy patients) and 136 healthy controls]. The GSE84796 dataset consisted of 10 end-stage HF patients and 7 healthy hearts procured from organ donors. We identified differentially expressed genes (DEGs) and constructed a protein-protein interaction (PPI) network. Functional pathways were elucidated through enrichment analyses using Gene Ontology (GO), the Kyoto Encyclopedia of Genes and Genomes (KEGG) and GeneMANIA annotation. Single nucleotide polymorphism (SNP) data for HF and NAFLD were sourced from genome-wide association studies (GWAS). The HF dataset included 486 160 samples (14 262 experimental and 471 898 control), and the NAFLD dataset comprised 377 988 samples (4761 experimental and 373 227 control). MR analysis investigates the causal interrelations.
Results:
Our analysis revealed 4032 DEGs from GSE126848 and 286 DEGs from GSE57338. The top 10 hub genes (CD163, VSIG4, CXCL10, FCER1G, FPR1, C1QB, CCR1, C1orf162, MRC1 and CD38) were significantly enriched in immune response, calcium ion concentration regulation and positive regulation of monocyte chemotaxis. CIBERSORT analysis indicated associations between these hub genes and natural killer (NK) cells and macrophages. Transcription factor (TF) target prediction for CD38, CXCL10 and CCR1 highlighted related TFs. A two-sample MR analysis confirmed a bidirectional causal relationship between NAFLD and HF. The main method [inverse variance weighted (IVW)] demonstrated a significant positive causal relationship between NAFLD and HF [P = 0.037; odds ratio (OR) = 1.024; 95% confidence interval (CI): 1.001 to 1.048]. Similarly, HF was associated with an increase in the risk of NAFLD (P < 0.001; OR = 1.117; 95% CI: 1.053 to 1.185).
Conclusions:
Our findings reveal novel molecular signatures common to NAFLD and HF and confirm their bidirectional causality, highlighting the potential for targeted therapeutic interventions and prompting further investigation into their intricate relationship.
