Abstract
Gastric cancer (GC) remains a significant global health burden with complex etiology. While lipid species are known to be associated with GC risk, the underlying biological mechanisms remain poorly understood. This study utilizes a multi-omics approach, combining two-step Mendelian randomization (MR), summary-data-based MR (SMR), and macromolecular docking to identify plasma proteins that mediate the causal effect of lipids on GC. Our analysis identified CCDC80 as a key mediator, accounting for approximately 31% of the causal effect of specific diacylglycerols on GC risk. Furthermore, computational docking identified several potential small-molecule ligands for CCDC80, suggesting its potential as a novel therapeutic target.
Introduction
Gastric cancer (GC) is a leading cause of cancer-related mortality worldwide. Observational studies have long suggested a link between blood lipid levels and cancer risk, but establishing causality is difficult due to confounding factors and reverse causation. Identifying the specific plasma proteins that translate lipid signals into oncogenic processes is essential for developing targeted interventions. This study aims to bridge this gap by using genetic causal inference to map the pathways between the lipidome, the proteome, and GC risk.
Methods
Researchers employed a sophisticated "multi-omics" workflow to identify causal pathways:
- Two-Step Mendelian Randomization (MR): Genetic variants were used as instrumental variables to determine if lipids (exposure) cause GC (outcome) through specific plasma proteins (mediators).
- SMR and Colocalization: These techniques were used to ensure that the genetic signals driving protein expression and GC risk were identical, reducing the risk of pleiotropy.
- Mediation Analysis: The propagation of error method was used to quantify the proportion of the lipid-to-cancer effect mediated by specific proteins.
- Macromolecular Docking: Once a target protein (CCDC80) was identified, AutoDock Vina was used to simulate the binding stability of potential small-molecule ligands.
The study leveraged large-scale datasets, including the decode2021 and UKBppp proteomics datasets (over 89,000 combined subjects), lipidomics data from the GeneRISK cohort, and GC summary statistics from the IEU catalog (7,921 cases).
Results
Causal Associations
The MR analysis identified significant causal links between 12 lipid species and GC risk. Additionally, three proteins—CCDC80, PDCD1LG2, and IGFALS—showed causal associations with GC. CCDC80 (Coiled-Coil Domain Containing 80) was identified as a protective factor, where higher genetically predicted levels were associated with a lower risk of GC.
Mediation Effect
A critical finding was that CCDC80 mediates 30.8% (95% CI: 6.4%–64.0%) of the causal effect of Diacylglycerol (16:1_18:1) on gastric cancer. This suggests that certain lipid species influence cancer risk specifically by modulating the levels of this protective secreted protein.
Potential Drug Targets
To explore the therapeutic potential of CCDC80, macromolecular docking was performed. Four ligands showed stable binding affinities (all < -4.9 kcal/mol):
- 2,3,7,8-Tetrachlorodibenzo-P-dioxin (-5.51 kcal/mol)
- Benzo[a]pyrene
- Bisphenol A
- Valproic Acid
Discussion
CCDC80 is a secreted plasma protein involved in cell adhesion and tumor suppression. Its role as a mediator suggests that maintaining or increasing CCDC80 levels could mitigate the pro-carcinogenic effects of specific lipid profiles. While the computational docking results are promising, particularly for established compounds like Valproic Acid, these findings require rigorous in vitro and in vivo validation.
Conclusion
This study provides genetic evidence that CCDC80 is a key mediator in the lipid-gastric cancer axis. By identifying the specific lipid species and proteins involved, this research offers a clearer roadmap for the development of targeted therapies and risk-stratification strategies for gastric cancer.
Summary updated