Regulation of hepatic ApoC3 expression by PGC-1β mediates hypolipidemic effect of nicotinic acid

Abstract

Peroxisome proliferator-activated receptor (PPAR) γ coactivator-1β (PGC-1β) is a transcriptional coactivator that induces hypertriglyceridemia in response to dietary fats through activating hepatic lipogenesis and lipoprotein secretion. The expression of PGC-1β is regulated by free fatty acids. Here we show that PGC-1β regulates plasma triglyceride metabolism through stimulating apolipoprotein C3 (APOC3) expression and elevating APOC3 levels in circulation. Remarkably, liver-specific knockdown of APOC3 significantly ameliorates PGC-1β-induced hypertriglyceridemia in mice. Hepatic expression of PGC-1β and APOC3 is reduced in response to acute and chronic treatments with nicotinic acid, a widely prescribed drug for lowering plasma triglycerides. Adenoviral-mediated knockdown of PGC-1β or APOC3 in the liver recapitulates the hypolipidemic effect of nicotinic acid. Proteomic analysis of hepatic PGC-1β transcriptional complex indicates that it stimulates APOC3 expression through coactivating orphan nuclear receptor ERRα and recruiting chromatin-remodeling cofactors. Together, these studies identify PGC-1β as an important regulator of the APOC3 gene cluster and reveal a mechanism through which nicotinic acid achieves its therapeutic effects.

Figure 1. Induction of APOC3 expression by PGC-1β in the liver and cultured primary hepatocytes

(A) Plasma triglyceride and non-sterified fatty acids (NEFA) levels and liver triglyceride content in mice transduced with Ad-GFP (open) or Ad-PGC-1β (filled). * p<0.01. (B) Realtime PCR analysis of total liver RNA from mice transduced with Ad-GFP or Ad-PGC-1β. Shown is fold-induction versus Ad-GFP group. Data represent mean ± s.e.m. (n=4). * p<0.05; **p<0.01; ***p<0.001. (C) Immunoblot analysis of APOC3 and APOB proteins in serum samples and PGC-1β in liver nuclear extracts from mice transduced with Ad-GFP or Ad-PGC-1β (n=4). Ponceau S staining and Lamin immunoblot were shown as loading control for serum and nuclear extracts, respectively. (D) Realtime PCR analysis of total RNA from primary hepatocytes transduced with Ad-GFP (open) or Ad-PGC-1β (filled). Shown is fold-induction versus Ad-GFP group. Data represent mean ± s.d. from a representative experiment. * p<0.05; *** p<0.001.

Figure 2. Requirement of APOC3 in PGC-1β-induced hypertriglyceridemia

(A) Plasma triglyceride and NEFA concentrations and liver triglyceride content in mice transduced with Ad-GFP or Ad-PGC-1β in combination with control (open) or siAPOC3 (filled) adenoviruses, as indicated. Four mice were included for each treatment. * p<0.01; ** p<0.01. See also Figure S1. (B) Immunoblots of serum samples and liver nuclear extracts from mice transduced with indicated adenoviruses. (C) Lipoprotein profile analysis. Plasma was fractionated by FLPC and triglycerides and cholesterol content in each fraction was measured. (D) Plasma triglyceride and NEFA concentrations and liver triglyceride content in mice transduced with scrb+GFP (open), siPGC-1β+GFP (filled), or siPGC-1β+APOC3 (blue) adenoviral mixtures, as indicated. (E) qPCR analysis of liver gene expression in transduced mice. Data in D-E represent mean ± s.e.m. (n=5). * p<0.05 scrb vs. siPGC-1β; ** p<0.01 siPGC-1β vs. siPGC-1β+APOC3. (F) Immunoblots of APOC3 in serum samples and PGC-1β liver nuclear extracts from transduced mice.

Figure 3. Role of PGC-1β and APOC3 in mediating triglyceride-lowering effects of nicotinic acid

(A) Realtime PCR analysis of total liver RNA from mice treated with vehicle (open, saline) or nicotinic acid (filled, 100 mg/kg, i.p.) for 2 or 4 hrs. * p<0.01. (B) Immunoblotting analysis of liver nuclear extracts from mice treated with vehicle (Veh) or nicotinic acid (NA) for 4 hrs. (C) Plasma triglyceride and NEFA concentrations in mice fed a high-fat diet containing no (open) or 1% (filled) nicotinic acid for three months (n=5). ** p<0.01. (D) Immunoblotting analysis of APOC3 and PGC-1β in serum samples and liver nuclear extracts, respectively. (E) Realtime PCR analysis of liver gene expression in mice in (C). * p<0.05; **p<0.01; ***p<0.001. (F) Plasma triglyceride concentrations. C57/Bl6J male mice were fed control (filled) or chow containing 1% nicotinic acid (open) for two weeks and transduced with scrb or siPGC-1β adenoviruses for five days, as indicated. (G) Realtime PCR analysis of liver gene expression. Data in F–G represent mean ± s.e.m. (n=4–5). * p<0.05 Veh vs. NA. (H) Immunoblots of PGC-1β and Lamin in liver nuclear extracts from mice in (F). Note that PGC-1β is reduced in response to NA treatment and PGC-1β shRNA. (I) Plasma triglyceride and NEFA concentrations and liver triglyceride content. C57/Bl6J male mice were fed control (filled) or chow containing 1% nicotinic acid (open) for two weeks and transduced with scrb or siAPOC3 adenoviruses for five days. Data represent mean ± s.e.m. (n=7–8). * p<0.05 Veh vs. NA; **p<0.01 scrb vs. siAPOC3. See also Figure S2.

Figure 4. Regulation of APOC3 gene expression by the PGC-1β transcriptional complex

(A) Identification of proteins in the PGC-1β transcriptional complex. Affinity-purified complexes from livers of mice transduced with Ad-GFP or Ad-PGC-1β were separated by SDS-PAGE. Individual bands were excised and proteins identified by mass spectrometry (left panel). Immunoblotting analysis of the PGC-1β complex using indicated antibodies (right panel). (B) Coactivation assays of mouse APOC3 promoter with different amounts of PGC-1β plasmid in the presence or absence of ERRα in transiently transfected McArdle RH7777 hepatoma cells. (C) Coactivation analysis of the mouse APOC3 gene promoter by PGC-1β, HCF1, and CBP. For B–C, shown are representative experiments from at least three independent experiments. Data represent mean ± s.d. (D) ChIP analysis. Chromatin extracts from transduced livers were immunoprecipitated with anti-PGC-1β antibody or IgG. The precipitated genomic DNA were PCR-amplified using primers flanking ERRE located in the enhancer and proximal promoter regions. (E) Coactivation assays of truncated APOC3 promoter reporters. RH7777 cells were transiently transfected with reporter constructs spanning 1.5 kb, 750 bp (enhancer and promoter regions, E/P) or 350 bp (proximal promoter only, P) in combination with ERRα in the presence or absence of PGC-1β. (F) Coactivation assays of APOC3 reporter constructs with mutant ERRE. For E–F, shown are representative experiments from at least three independent experiments. Data represent mean ± s.d. (G) Model illustrating the PGC-1β/APOC3 pathway in mediating the hypolipidemic effects of nicotinic acid.