Site-Specific Glycoprofiles of HDL-Associated ApoE are Correlated with HDL Functional Capacity and Unaffected by Short-Term Diet

Abstract

Since high-density lipoprotein (HDL) glycoprofiles are associated with HDL functional capacity, we set out to determine whether diet can alter the glycoprofiles of key HDL-associated proteins, including ApoE, a potent driver of chronic disease risk. Ten healthy subjects consumed a fast food (FF) and a Mediterranean (Med) diet for 4 days in randomized order, with a 4-day wash-out between treatments. A multiple reaction monitoring method was used to characterize the site-specific glycoprofiles of HDL proteins, and HDL functional capacity was analyzed. We describe for the first time that ApoE has 7 mucin-type O-glycosylation sites, which were not affected by short-term diet. The glycoprofiles of other HDL-associated proteins were also unaffected, except that a disialylated ApoC-III glycan was enriched after Med diet, and a nonsialylated ApoC-III glycan was enriched after FF diet. Twenty-five individual glycopeptides were significantly correlated with cholesterol efflux capacity and 21 glycopeptides were correlated with immunomodulatory capacity. Results from this study indicate that the glycoprofiles of HDL-associated proteins including ApoE are correlated with HDL functional capacity but generally unaffected by diet in the short term, except ApoC-III sialylation. These results suggest that HDL protein glycoprofiles are affected by both acute and long-term factors and may be useful for biomarker discovery.

Keywords: ApoC-III; ApoE; Mediterranean diet; cholesterol efflux; fast food diet; glycomics; glycoproteomics; high-density lipoprotein.

Figures

Figure 1:

A-C: Box plots of HDL ApoAI (mg/dL) (A), % HDL cholesterol efflux (A), and % macrophage TNF-α suppression (B) for subjects before and after FF and Med diets, with unadjusted P-values. Lines with the same color represent the same subjects. D: Scatter plot of % HDL cholesterol efflux and % TNF-α suppression. E&F: Box plots of HDL glycans ApoC3_74_O_1102 and ApoC3_74_O_2230 before and after FF and Med diets, with unadjusted P-values

Figure 2:

Glycan site-heterogeneity of HDL associated ApoE. Sites of attachment identified include Ser63, Ser76/Thr83/Thr89, Ser129/Thr130, Thr194, Ser197, Thr289/Ser290 and Ser296. The 4-digit glycan composition number corresponds to the number of hexose (Hex), N-acetylhexosamine (HexNAc) (which could be either N-acetylglucosamine (GlcNAc) or N-acetyl-galactosamine (GalNAc)), fucose (Fuc), and N-acetylneuraminic acid (Neu5Ac) or sialic acid. A glycan containing 1 Hex, 1 HexNAc, 0 Fuc and 2 Neu5Ac is designated as 1102. Yellow square: GalNAc; blue square: GlcNAc; yellow circle: Gal; red triangle: Fuc; purple diamond: Neu5Ac. *: multiple possible O-glycosylation sites for this glycan.

Figure 3:

A: Heatmap of Spearman’s rho values between glycoforms and HDL functions. Glycans with p < 0.05 were selected. B-E: Scatterplot of glycoforms A2HSG_319_O_1102 (B), A1AT_271_N_5402 (C), A2HSG_319_O_1101 (D), and ApoE_197_O_2201 (E) correlated with HDL cholesterol efflux capacity. F-I: Scatterplot of glycoforms A2HSG_176_N_6501 (F), ApoE_197_O_1100 (G), ApoC3_74_O_1102 (H), and A2HSG_346_O_1101 (I) correlated with % TNF-α suppression.