Effect of niacin monotherapy on high density lipoprotein composition and function

Abstract

Background: Niacin has modest but overall favorable effects on plasma lipids by increasing high density lipoprotein cholesterol (HDL-C) and lowering triglycerides. Clinical trials, however, evaluating niacin therapy for prevention of cardiovascular outcomes have returned mixed results. Recent evidence suggests that the HDL proteome may be a better indicator of HDL's cardioprotective function than HDL-C. The objective of this study was to evaluate the effect of niacin monotherapy on HDL protein composition and function.

Methods: A 20-week investigational study was performed with 11 participants receiving extended-release niacin (target dose = 2 g/day) for 16-weeks followed by a 4-week washout period. HDL was isolated from participants at weeks: 0, 16, and 20. The HDL proteome was analyzed at each time point by mass spectrometry and relative protein quantification was performed by label-free precursor ion intensity measurement.

Results: In this cohort, niacin therapy had typical effects on routine clinical lipids (HDL-C + 16%, q < 0.01; LDL-C - 20%, q < 0.01; and triglyceride - 15%, q = 0.1). HDL proteomics revealed significant effects of niacin on 5 proteins: serum amyloid A (SAA), angiotensinogen (AGT), apolipoprotein A-II (APOA2), clusterin (CLUS), and apolipoprotein L1 (APOL1). SAA was the most prominently affected protein, increasing 3-fold in response to niacin (q = 0.008). Cholesterol efflux capacity was not significantly affected by niacin compared to baseline, however, stopping niacin resulted in a 9% increase in efflux (q < 0.05). Niacin did not impact HDL's ability to influence endothelial function.

Conclusion: Extended-release niacin therapy, in the absence of other lipid-modifying medications, can increase HDL-associated SAA, an acute phase protein associated with HDL dysfunction.

Keywords: Apolipoproteins; Cholesterol efflux; High density lipoprotein; Niacin; Proteomics; Serum amyloid a; Vitamin B3.

Fig. 1

Study design. Schematic of the time course of the study. After qualifying for the study, eligible participants had a baseline visit followed by a two-week run in period where extended-release niacin was escalated to the target dose of 2000 mg/day. Tolerance was evaluated after week 2. Target dose was maintained until week 16 followed by a 4 week washout period. Research blood collected at baseline, week 16, and week 20 was used for lipoprotein proteome and functional assays

Fig. 2

Effect of niacin therapy on plasma lipids, lipoprotein particles, and apolipoproteins. Lipoprotein profile was measured using nuclear magnetic resonance spectroscopy on a Vantera™ clinical analyzer (LabCorp). Plasma lipids (a-d), lipoprotein particle numbers (e-g), particle sizes (h-j), and apolipoprotein concentrations (k-l) were compared at baseline, on niacin (week 16), and washout (week 20). Comparisons were evaluated using repeated measures one-way ANOVA with false discovery rate (FDR) correction for multiple comparisons. * q < 0.05, ** q < 0.01, n.s. = not significant. If no indicators are present, then none of the comparisons was statistically significant

Fig. 3

Niacin alters the HDL proteome. The HDL proteome was analyzed at baseline and after 16 weeks on niacin (2 g/day, extended release). (a) Changes to the HDL proteome are represented in a volcano plot. Each point indicates one of the 63 detected proteins. Proteins above the horizontal line were considered statistically significant changes. Green colored points indicate proteins increased while taking niacin and red points indicate reduced protein abundance while taking niacin. (b) Functional annotation of proteins affected by niacin was performed using Panther (version 14.1) gene list analysis [17, 18]

Fig. 4

Effects of niacin on HDL protein composition are reversible. Relative protein abundance for the 5 proteins influenced by niacin were analyzed at baseline, on niacin (week 16), and washout (week 20). Changes in serum amyloid a (SAA; a), angiotensinogen (AGT; b), apolipoprotein L1 (APOL1; c), clusterin (CLUS; d), and apolipoprotein A-II (APOA2; e) were evaluated across time points using repeated measures one-way ANOVA with false discovery rate (FDR) correction for multiple comparisons. FDR adjusted probability value: * q < 0.05, ** q < 0.01, n.s. = not significant

Fig. 5

Niacin does not alter HDL-mediated cholesterol efflux or endothelial signaling. HDL function assays were performed on samples collected at baseline, on niacin (week 16), and washout (week 20). a Efflux of radiolabeled cholesterol from macrophage cells to apoB depleted serum. b Activation of endothelial cell signaling pathways eNOS and Akt was evaluated by measurement of phosphorylation after treatment with isolated HDL. Statistical comparisons were made using repeated measures one-way ANOVA with false discovery rate (FDR) correction for multiple comparisons. FDR adjusted probability value: * q < 0.05, n.s. or no indicator = not significant