Anthocyanins and their physiologically relevant metabolites alter the expression of IL-6 and VCAM-1 in CD40L and oxidized LDL challenged vascular endothelial cells

Abstract

Scope: In vitro and in vivo studies suggest that dietary anthocyanins modulate cardiovascular disease risk; however, given anthocyanins extensive metabolism, it is likely that their degradation products and conjugated metabolites are responsible for this reported bioactivity.

Methods and results: Human vascular endothelial cells were stimulated with either oxidized LDL (oxLDL) or cluster of differentiation 40 ligand (CD40L) and cotreated with cyanidin-3-glucoside and 11 of its recently identified metabolites, at 0.1, 1, and 10 μM concentrations. Protein and gene expression of IL-6 and VCAM-1 was quantified by ELISA and RT-qPCR. In oxLDL-stimulated cells the parent anthocyanin had no effect on IL-6 production, whereas numerous anthocyanin metabolites significantly reduced IL-6 protein levels; phase II conjugates of protocatechuic acid produced the greatest effects (>75% reduction, p ≤ 0.05). In CD40L-stimulated cells the anthocyanin and its phase II metabolites reduced IL-6 protein production, where protocatechuic acid-4-sulfate induced the greatest reduction (>96% reduction, p ≤ 0.03). Similarly, the anthocyanin and its metabolites reduced VCAM-1 protein production, with ferulic acid producing the greatest effect (>65% reduction, p ≤ 0.04).

Conclusion: These novel data provide evidence to suggest that anthocyanin metabolites are bioactive at physiologically relevant concentrations and have the potential to modulate cardiovascular disease progression by altering the expression of inflammatory mediators.

Keywords: Adhesion molecule; Cyanidin-3-glucoside; Cytokine; Inflammation; Metabolites.

Figure 1

Recently identified metabolites of C3G¹ screened for their activity against IL‐6 and VCAM‐1 expression in oxLDL‐ and CD40L‐stimulated HUVECs. C3G, cyanidin‐3‐glucoside; PCA, protocatechuic acid, PGA, phloroglucinaldehyde; VA, vanillic acid; IVA, isovanillic acid; PCA‐4‐Gluc, PCA‐4‐glucuronide; PCA‐3‐Gluc, PCA‐3‐glucuronide; PCA‐4‐Sulf, PCA‐4‐sulfate; PCA‐3‐Sulf, PCA‐3‐sulfate; VA‐4‐Sulf, VA‐4‐sulfate; IVA‐3‐Sulf, IVA‐3‐sulfate; FA, ferulic acid. ¹Metabolites recently identified and reported by de Ferrars 25.

Figure 2

OxLDL‐stimulated IL‐6 production in HUVECs co‐incubated with C3G, phenolic metabolites or LDL controls for 24 h. (A) cells incubated with or without LDL (5 μg/mL); (B–G) cells incubated with C3G or phenolic metabolites at 0.1, 1 and 10 μM and oxLDL. All data expressed as mean percentage (± SD, n = 3) of oxLDL‐induced controls (expressed as 100%). ***p ≤ 0.001, **p ≤ 0.01, *p ≤ 0.05 (analysis of variance (ANOVA) with Tukey post‐hoc) relative to oxLDL‐stimulated control cells. C3G, cyanidin‐3‐glucoside; PCA, protocatechuic acid; PGA, phloroglucinaldehyde; VA, vanillic acid; IVA, isovanillic acid; PCA‐4‐Gluc, PCA‐4‐glucuronide; PCA‐3‐Gluc, PCA‐3‐glucuronide; PCA‐3‐Sulf, PCA‐3‐sulfate; PCA‐4‐Sulf, PCA‐4‐sulfate; VA‐4‐Sulf, VA‐4‐sulfate; IVA‐3‐Sulf, IVA‐3‐sulfate; FA, ferulic acid.

Figure 3

Change in OxLDL‐stimulated IL‐6 mRNA in HUVECs co‐incubated with C3G and phenolic metabolites at 10 μM or oxLDL controls for 24 h. All graphical data expressed as mean fold change (± SD, n = 3) of un‐stimulated (basal) IL‐6 mRNA levels. *p ≤ 0.05 (ANOVA with Tukey post‐hoc) relative to oxLDL‐stimulated control. PCA, protocatechuic acid; VA, vanillic acid; PCA‐3‐Sulf, PCA‐3‐sulfate; PCA‐4‐Sulf, PCA‐4‐sulfate; VA‐4‐Sulf, VA‐4‐sulfate; IVA‐3‐Sulf, IVA‐3‐sulfate; FA, ferulic acid.

Figure 4

CD40L‐stimulated IL‐6 protein production in HUVECs co‐incubated with C3G, phenolic metabolites or CD40L controls for 24 h. (A) Cells incubated with or without D1.1 cells (1 × 10⁶ cells/well); (B–G) cells incubated with C3G or phenolic metabolites at 0.1, 1, 10 μM and D1.1 cells (1 × 10⁶ cells/well). All data expressed as mean percentage (± SD, n = 3) of CD40L‐induced controls. ***p ≤ 0.001, **p ≤ 0.01, *p ≤ 0.05 (ANOVA with Tukey post‐hoc) relative to CD40L‐stimulated control. C3G, cyanidin‐3‐glucoside; PCA, protocatechuic acid; PGA, phloroglucinaldehyde; VA, vanillic acid; IVA, isovanillic acid; PCA‐4‐Gluc, PCA‐4‐glucuronide; PCA‐3‐Gluc, PCA‐3‐glucuronide; PCA‐3‐Sulf, PCA‐3‐sulfate; PCA‐4‐Sulf, PCA‐4‐sulfate; VA‐4‐Sulf, VA‐4‐sulfate; IVA‐3‐Sulf, IVA‐3‐sulfate; FA, ferulic acid.

Figure 5

CD40L‐stimulated VCAM‐1 protein production in HUVECs co‐incubated with C3G, phenolic metabolites or CD40L controls for 24 h. (A) Cells incubated with or without D1.1 cells (1 × 10⁶ cells/well); (B–G) cells incubated with C3G or phenolic metabolites at 0.1, 1, 10 μM and D1.1 cells (1 × 10⁶ cells/well). All data expressed as mean percentage (± SD, n = 3) of CD40L‐induced controls. ***p ≤ 0.001, **p ≤ 0.01, *p ≤ 0.05 (ANOVA with Tukey post‐hoc) relative to CD40L‐stimulated control. C3G, cyanidin‐3‐glucoside; PCA, protocatechuic acid; PGA, phloroglucinaldehyde; VA, vanillic acid; IVA, isovanillic acid; PCA‐4‐Gluc, PCA‐4‐glucuronide; PCA‐3‐Gluc, PCA‐3‐glucuronide; PCA‐3‐Sulf, PCA‐3‐sulfate; PCA‐4‐Sulf, PCA‐4‐sulfate; VA‐4‐Sulf, VA‐4‐sulfate; IVA‐3‐Sulf, IVA‐3‐sulfate; FA, ferulic acid.

Figure 6

Change in CD40L‐stimulated IL‐6 and VCAM‐1 mRNA in HUVECs co‐incubated with C3G and phenolic metabolites or CD40L controls for 24 h. (A) IL‐6 mRNA fold change in cells co‐incubated C3G or phenolic metabolites (10 μM) with or without D1.1 cells (1 × 10⁶ cells/well). (B) VCAM‐1 mRNA fold change in cells co‐incubated with C3G or phenolic metabolites (10 μM) with or without D1.1 cells (1 × 10⁶ cells/well). All data graphically expressed as mean fold change (± SD, n = 3) relative to un‐stimulated (basal) IL‐6 and VCAM‐1 mRNA levels. *p ≤ 0.05 (ANOVA with Tukey post‐hoc) relative to CD40L‐stimulated control. C3G, cyanidin‐3‐glucoside; PCA, protocatechuic acid; PGA, phloroglucinaldehyde; VA, vanillic acid; IVA, isovanillic acid; PCA‐3‐Sulf, PCA‐3‐sulfate; IVA‐3‐Sulf, IVA‐3‐sulfate; FA, ferulic acid.