Compr Psychoneuroendocrinol. 2026 Jun 25;27:100357. doi: 10.1016/j.cpnec.2026.100357. eCollection 2026 Aug.
ABSTRACT
BACKGROUND: Allostatic load (AL) represents the cumulative physiological burden arising from chronic stress across neuroendocrine, immune, metabolic, and cardiovascular systems. Although AL is strongly associated with cardiometabolic and inflammatory diseases, its underlying genetic architecture remains poorly characterized.
METHODS: We integrated genome-wide association summary statistics for five AL-related phenotypes-systolic blood pressure, white blood cell count, C-reactive protein, body mass index, and triglycerides-using Genomic Structural Equation Modeling (Genomic SEM) to derive a latent genetic factor indexing the shared cardiovascular, inflammatory, and metabolic components of AL. A common-factor model was fitted to derive a latent genetic representation of AL, followed by multivariate GWAS of the latent factor. Downstream analyses included functional annotation, Bayesian fine-mapping, transcriptome-wide association analysis (TWAS), pathway enrichment, cell type-specific heritability estimation, and partitioned SNP heritability analyses.
RESULTS: The common-factor model showed good fit (CFI = 0.964; SRMR = 0.037), supporting a shared genetic architecture across the five AL-related traits. Functional annotation indicated that AL-associated variants were predominantly located in non-coding regulatory regions. Pathway enrichment analyses implicated metabolic regulation, lipid processing, neuroendocrine signaling, and immune-related pathways. Integrative fine-mapping and TWAS prioritized genes involved in metabolic homeostasis and neuronal signaling, including HNF4A, MLXIPL, BDNF, and SH2B1. Cell type-specific analyses showed enrichment in neuronal populations and stress-related brain regions, while partitioned heritability analyses demonstrated significant enrichment in conserved regions, promoters, enhancers, and active chromatin marks.
CONCLUSION: This study characterizes the shared polygenic architecture of an AL-related latent factor derived from cardiovascular, inflammatory, and metabolic biomarkers. The results suggest that genetic liability captured by this AL-related latent factor converges on metabolic, immune-inflammatory, neuroendocrine and neural regulatory pathways, providing a system-level perspective on AL-related physiological burden and multisystem disease vulnerability.
PMID:42405204 | PMC:PMC13329553 | DOI:10.1016/j.cpnec.2026.100357
