Site-Specific Apo AI Glycation Impairs HDL Function and Promotes Atherosclerosis in Diabetes

BACKGROUND: Apo (apolipoprotein) AI glycation has been implicated in HDL (high-density lipoprotein) dysfunction, but its site-resolved landscape, clinical relevance, relationship with diabetic atherosclerosis, and underlying mechanisms remain incompletely defined. METHODS: We performed unbiased site-resolved glycation proteomics in 860 patients with type 2 diabetes and coronary atherosclerosis (CAS) and 294 controls with type 2 diabetes without CAS to define the plasma apo AI glycation landscape. Apo AI glycation signatures associated with diabetic CAS were identified using integrated unsupervised and supervised analyses. An Apo AI Glycation Index was developed using least absolute shrinkage and selection operator regression and evaluated in training, testing, and independent validation cohorts. To assess functional relevance, a glycation-resistant apo AI mutant, cross-linked apo AI (apo AI CL ), was engineered and tested under glycation conditions using surface plasmon resonance, in vitro and in vivo reverse cholesterol transport assays, HDL remodeling analyses, and atherosclerosis models. RNA sequencing, macrophage-specific knockout models, receptor-binding assays, signaling inhibition, and cholesterol efflux rescue experiments were used to investigate the underlying macrophage pathway. RESULTS: Apo AI glycation patterns differed between patients with type 2 diabetes with and without CAS, with K96 and K106/107 emerging as prominent disease-associated glycation sites. The Apo AI Glycation Index was independently associated with CAS and coronary artery disease in patients with type 2 diabetes and was inversely associated with HDL-mediated reverse cholesterol transport and lecithin-cholesterol acyltransferase activity. Compared with native apo AI, apo AI CL showed reduced key-site glycation, preserved structural stability, enhanced binding affinity to lecithin–cholesterol acyltransferase, and improved HDL-mediated reverse cholesterol transport under glycation conditions. In diabetic mouse models, apo AI CL attenuated HDL dysfunction and atherosclerotic lesion formation compared with glycated apo AI. These data support a candidate mechanism in which glycated apo AI enhances RAGE (receptor for advanced glycation end products) interaction and activates ERK1/2 (extracellular signal-regulated kinase 1/2)–NF-κB (nuclear factor κB)/p65 signaling, leading to upregulation of NR2C2 (nuclear receptor subfamily 2 group C member 2). LXRα (liver X receptor α) activation and macrophage NR2C2 deficiency restored cholesterol efflux in glycated apo AI–treated macrophages. CONCLUSIONS: This study provides a comprehensive site-resolved map of apo AI glycation in patients with diabetes and identifies the Apo AI Glycation Index as a glycation signature associated with diabetic CAS and impaired HDL function. Experimental data support a mechanism in which site-specific apo AI glycation contributes to HDL dysfunction and macrophage cholesterol efflux impairment through RAGE-associated NR2C2–LXRα signaling. Glycation-resistant apo AI CL preserves HDL function and attenuates atherosclerosis in preclinical models, supporting site-specific apo AI glycation as a mechanistically relevant feature of diabetic atherosclerosis. REGISTRATION: URL: https://www.clinicaltrials.gov ; Unique identifier: NCT05659043.