Daily Cardiology Research Analysis
Analyzed 61 papers and selected 3 impactful papers.
Summary
Multi-ancestry and multitrait genetic studies substantially advance understanding of aortic stenosis by identifying hundreds of loci, valve-specific TWAS genes, and clinically promising polygenic risk scores. Complementing this, a mechanistic study uncovers IDH2 lactylation as a driver of angiogenesis in diabetic myocardial infarction via Cav1–eNOS regulation, highlighting a modifiable pathway with therapeutic potential.
Research Themes
- Genetic architecture and risk prediction of aortic stenosis
- Mechanistic angiogenesis pathways in diabetic myocardial infarction
- AI-enabled imaging phenotypes integrated with genomics
Selected Articles
1. Genomic and transcriptomic analyses of aortic stenosis enhance therapeutic target discovery and disease prediction.
A multi-ancestry GWAS meta-analysis of 2.85 million individuals identified 244 risk loci (including X chromosome) for aortic stenosis, with valve-specific TWAS implicating 54 genes. Functional silencing of CMKLR1 and LTBP4 reduced mineralization in human valve cells, and a new polygenic risk score was developed, linking fatty acid and TGF-β pathways to disease biology and risk prediction.
Impact: This work maps the genetic architecture of AS at scale, prioritizes causal pathways via valve TWAS and functional assays, and provides a clinically actionable polygenic risk score.
Clinical Implications: Findings support risk stratification using polygenic scores and nominate therapeutic targets (for example, CMKLR1, TGF-β signaling) for drug development in AS, a disease lacking pharmacotherapy.
Key Findings
- Multi-ancestry GWAS meta-analysis (86,864 AS cases among 2,853,408 individuals) identified 241 autosomal and 3 X-chromosome risk loci.
- Valve TWAS discovered 54 genes whose genetically predicted expression influences AS risk.
- Silencing CMKLR1 and LTBP4 in human valvular interstitial cells substantially decreased mineralization, implicating PUFA and TGF-β pathways.
- A new polygenic risk score for AS was constructed.
Methodological Strengths
- Very large, multi-ancestry GWAS meta-analysis with sex- and ancestry-stratified analyses
- Integration of valve tissue eQTL-based TWAS and functional gene silencing to support causality
Limitations
- Observational genetic associations limit causal inference beyond tested genes
- Clinical utility of the polygenic score requires external validation across diverse populations and care settings
Future Directions: Prospective validation of the AS polygenic risk score, mechanistic dissection of prioritized genes (for example, CMKLR1, LTBP4), and target-based drug discovery and preclinical testing.
2. Multitrait analyses identify genetic variants associated with aortic valve function and aortic stenosis risk.
Deep learning-derived MRI valve metrics in 59,571 UK Biobank participants enabled GWAS that, when integrated with AS GWAS via MTAG, revealed 166 loci including lipid genes (PCSK9, LDLR). A multitrait polygenic score strongly stratified AS risk in the All of Us cohort (HR 3.32 for top 5% vs others), demonstrating translational potential for prediction.
Impact: By fusing AI-derived imaging phenotypes with genetics, this study uncovers loci influencing valve function and delivers a high-performing AS polygenic score validated across cohorts.
Clinical Implications: Multitrait polygenic risk scores could inform earlier identification and monitoring of individuals at high risk for aortic stenosis, complementing imaging-based assessment and preventive strategies.
Key Findings
- Deep learning quantified aortic valve peak velocity, mean gradient, and valve area from MRI in 59,571 participants for GWAS.
- MTAG integrating valve trait GWAS with AS GWAS identified 166 distinct loci, including PCSK9 and LDLR.
- The MTAG-derived AS polygenic score predicted AS in All of Us (HR 3.32 for top 5% vs others).
Methodological Strengths
- Integration of AI-derived imaging phenotypes with large-scale genomics
- External validation of polygenic scoring in an independent national cohort
Limitations
- Discovery dataset primarily from UK Biobank, potentially limiting ancestry diversity
- Imaging-derived phenotypes may introduce measurement biases; PRS calibration is needed across health systems
Future Directions: Expand validation in non-European ancestries, test clinical impact of AS PRS-guided surveillance, and dissect lipid and valve biology at implicated loci.
3. IDH2 lactylation promotes angiogenesis in murine diabetic myocardial infarction via blocking Cav1-eNOS interaction.
Using LC–MS/MS, the study identifies IDH2 K272 lactylation in diabetic MI, which enhances Cav1 binding, disrupts Cav1–eNOS interaction, activates eNOS, and promotes endothelial angiogenesis under high glucose and hypoxia. Endothelial K272R knock-in mice exhibit impaired angiogenesis and worse remodeling; ACAT1 and HDAC1 modulate IDH2 lactylation with lactate via MCT1. Empagliflozin enhanced IDH2 lactylation and mitigated injury.
Impact: This mechanistic discovery links metabolic lactylation to endothelial eNOS signaling and angiogenesis in diabetic MI, nominating a tractable enzymatic pathway for therapeutic intervention.
Clinical Implications: Targeting the IDH2 lactylation axis (for example, ACAT1/HDAC1 or lactate transport) may enhance post-MI angiogenesis in diabetes; findings also suggest a potential mechanistic contributor to benefits of SGLT2 inhibitors.
Key Findings
- IDH2 is lactylated at K272 in diabetic MI, enhancing Cav1 binding and disrupting Cav1–eNOS interaction to activate eNOS and promote angiogenesis.
- Endothelial cell-specific IDH2-K272R knock-in mice show impaired angiogenesis, worse cardiac function, and adverse remodeling.
- ACAT1 and HDAC1 act as lactyltransferase and delactylase using lactate delivered by MCT1; empagliflozin enhanced IDH2 lactylation and mitigated injury.
Methodological Strengths
- Proteome-wide lactylation mapping with LC–MS/MS and targeted mechanistic assays
- In vivo endothelial-specific knock-in model establishing causality for IDH2 K272 lactylation
Limitations
- Preclinical study in male mice; human translatability and sex differences remain to be established
- Pharmacologic modulation (for example, empagliflozin) may have pleiotropic effects beyond lactylation
Future Directions: Validate IDH2 lactylation signatures in human diabetic MI tissues, test selective ACAT1/HDAC1 modulators, and evaluate therapeutic efficacy in large-animal diabetic MI models.