RNF130 Regulates LDLR Availability and Plasma LDL Cholesterol Levels

Abstract

Background: Removal of circulating plasma low-density lipoprotein cholesterol (LDL-C) by the liver relies on efficient endocytosis and intracellular vesicle trafficking. Increasing the availability of hepatic LDL receptors (LDLRs) remains a major clinical target for reducing LDL-C levels. Here, we describe a novel role for RNF130 (ring finger containing protein 130) in regulating plasma membrane availability of LDLR.

Methods: We performed a combination of gain-of-function and loss-of-function experiments to determine the effect of RNF130 on LDL-C and LDLR recycling. We overexpressed RNF130 and a nonfunctional mutant RNF130 in vivo and measured plasma LDL-C and hepatic LDLR protein levels. We performed in vitro ubiquitination assays and immunohistochemical staining to measure levels and cellular distribution of LDLR. We supplement these experiments with 3 separate in vivo models of RNF130 loss-of-function where we disrupted Rnf130 using either ASO (antisense oligonucleotides), germline deletion, or AAV CRISPR (adeno-associated virus clustered regularly interspaced short palindromic repeats) and measured hepatic LDLR and plasma LDL-C.

Results: We demonstrate that RNF130 is an E3 ubiquitin ligase that ubiquitinates LDLR resulting in redistribution of the receptor away from the plasma membrane. Overexpression of RNF130 decreases hepatic LDLR and increases plasma LDL-C levels. Further, in vitro ubiquitination assays demonstrate RNF130-dependent regulation of LDLR abundance at the plasma membrane. Finally, in vivo disruption of Rnf130 using ASO, germline deletion, or AAV CRISPR results in increased hepatic LDLR abundance and availability and decreased plasma LDL-C levels.

Conclusions: Our studies identify RNF130 as a novel posttranslational regulator of LDL-C levels via modulation of LDLR availability, thus providing important insight into the complex regulation of hepatic LDLR protein levels.

Keywords: cardiovascular diseases; endocytosis; lipids; lipoproteins; ubiquitin.

Figure 1.. RNF130 expression reduces hepatic LDLR and increases plasma LDL-C level.

(A) Immunofluorescence images of HEK293 and HepG2 cells co-transfected with WT-FLAG-RNF130 and either control plasmid or Rab27a-GFP (pan-endosomal marker). Original magnification 63X. Arrows indicate co-localization. Scale bar indicates 50mm. (B) Experimental design: WT mice were treated once on day 0 with control adenovirus (Ad-Ctr) or adenovirus overexpressing human RNF130 (Ad-RNF130) and tissues harvested 7 days later. (C) Hepatic protein expression and accompanying densitometry of RNF130 in mice treated as in (B) (Ad-Ctr n=5 and Ad-RNF130 n=5). (D) Hepatic protein expression of endocytosed receptors in mice treated as in (B) (Ad-Ctr n=5 and Ad-RNF130 n=5; individual dots represent individual animals). (E) Quantification of hepatic LDLR protein from (D) (Ad-Ctr n=5 and Ad-RNF130 n=5). (F) Hepatic mRNA expression of LDLR from mice treated as in (B) (Ad-Ctr n=8 and Ad-RNF130 n=7; individual dots represent individual animals). (G-H) Plasma total cholesterol (G) and LDL (LDL/VLDL) cholesterol (H) in WT mice treated as in (B) (Ad-Ctr n=8 and Ad-RNF130 n=6; individual dots represent individual animals). (I) FPLC lipoprotein profiles in WT mice treated as in (B). Data are expressed as mean ±SEM. FPLC lipoprotein profiles are plotted as mean absorbance unit (AU) ±SEM (Ad-Ctr n=8 and Ad-RNF130 n=6). P values were determined by (C, E-H) Student’s t-test. ns, not significant. Ad, adenovirus; AU, absorbance unit; Ctr, Control; EGFR, epidermal growth factor receptor; HDL, high-density lipoprotein; IB, immunoblot; LDL, low-density lipoprotein; LDLR, LDL receptor; LRP-1, LDLR-related protein 1; PDI, protein disulfide isomerase; TfR, transferrin receptor; VLDL, very low-density lipoprotein.

Figure 2.. RNF130 is a RING-dependent E3 ligase that ubiquitinates LDLR and redistributes LDLR away from the plasma membrane.

(A) HEK293 cells were co-transfected with LDLR-GFP, FLAG-RNF130, and HA-Ubiquitin expression plasmids. After 36 hours, lysates were subjected to immunoprecipitation (IP) with anti-GFP and immunoblotting (IB) with anti-HA to detect ubiquitinated proteins (upper panel). Total cell lysates (input; lower panels) were processed for western blotting with antibodies to anti-GFP, anti-FLAG, or anti-PDI. (B) HEK293 cells were co-transfected with wildtype (WT) or mutant (MUT) LDLR-GFP (K811/816/830RC839A), FLAG-RNF130, and HA-Ubiquitin. After 36 hours, lysates were subjected to IP and IB as in (A). (C) HEK293 cells were co-transfected with LDLR-GFP, wildtype (WT) FLAG-RNF130 or mutant FLAG-RNF130 (C304A) and HA-Ubiquitin expression plasmids. After 36 hours, lysates were subjected to IP and IB as in (A). (D) HEK293 cells were co-transfected with LDLR-GFP, wildtype (WT) FLAG-RNF130 or mutant FLAG-RNF130 (C304A) and HA-Ubiquitin expression plasmids. After 36 hours, cells were incubated with biotin at 4°C to label cell surface proteins. Total cell lysates (input; lanes 1–4), biotinylated proteins (membrane; lanes 5–8), and unmodified proteins (intracellular; lanes 9–12) were processed for western blotting with antibodies to anti-GFP, anti-FLAG, anti-PDI (intracellular) and anti-Na⁺/K⁺ ATPase (plasma membrane). (E) Immunofluorescence images of HEK293 cells co-transfected with LDLR-GFP and either wildtype (WT) FLAG-RNF130 or mutant FLAG-RNF130 (C304A) Original magnification 63X. White arrows indicate plasma membrane. Scale bars indicate 50mm. (F) Mean fluorescence intensity (MFI) of LDLR-GFP quantification of confocal images in (E). n=3 images from 3 independent transfections. P values were determined by Kruskal-Wallis test with Dunn correction. GFP, green fluorescent protein; HA, haemagglutinin; IB, immunoblot; IP, immunoprecipitation; MFI, mean fluorescence intensity; PDI, protein disulfide isomerase; Ub, ubiquitin.

Figure 3.. RNF130 increases plasma LDL-C levels in vivo in a process that is dependent on LDLR and RNF130 E3 ligase activity.

(A) Experiment design: Wildtype (WT) mice were treated once on day 0 with control adenovirus (Ad-Ctr), or adenovirus overexpressing human RNF130 with an N-terminal 3xFLAG epitope tag (Ad-3F-RNF130) or 3F-RNF130 with a single point mutation in the RING domain (Ad-3F-C304A) and tissues harvested 7 days later. (B) Plasma total cholesterol in WT mice treated as in (A) (Ad-Ctr n=10, Ad-3F-RNF130 n=8, Ad-3F-C304A n=9; individual dots represent individual animals). (C) FPLC lipoprotein profiles in WT mice treated as in (A). FPLC lipoprotein profiles are plotted as mean absorbance unit (AU) ±SEM (Ad-Ctr n=10, Ad-3F-RNF130 n=8, Ad-3F-C304A n=9). (D) Experimental design: Ldlr^–/– mice were treated once on day 0 with control adenovirus (Ad-Ctr) or adenovirus overexpressing human RNF130 (Ad-RNF130) and tissues harvested 7 days later. (E) Plasma total cholesterol in WT mice treated as in (D) (Ad-Ctr n=6, Ad-RNF130 n=6; individual dots represent individual animals). (F) FPLC lipoprotein profiles in WT mice treated as in (D). FPLC lipoprotein profiles are plotted as mean absorbance unit (AU) ±SEM (Ad-Ctr n=6, Ad-RNF130 n=6). Data are expressed as mean ±SEM. P values were determined by Kruskal-Wallis test with Dunn correction (B) or Student’s t test (E). ns, not significant. Ad, adenovirus; AU, absorbance unit; Ctr, Control; HDL, high-density lipoprotein; LDL, low-density lipoprotein; LDLR, LDL receptor; VLDL, very low-density lipoprotein; WT, wildtype; 3F, 3-FLAG.

Figure 4.. RNF130-mediated regulation of plasma LDL-C depends on hepatic LDLR abundance.

(A) Experiment design: Pcsk9^–/– mice were treated once on day 0 with control adenovirus (Ad-Ctr) or adenovirus overexpressing human RNF130 (Ad-RNF130) and tissues harvested 7 days later. (B) Hepatic protein expression of representative animals and accompanying densitometry of RNF130 in Pcsk9^–/– mice treated with Ad-Ctr or Ad-RNF130 (Ad-Ctr n=5, Ad-RNF130 n=7; individual dots represent quantification of individual animals analyzed by western blot). (C-D) Plasma total cholesterol (C) and LDL (LDL/VLDL) cholesterol (D) in Pcsk9^–/– mice treated with Ad-Ctr (n=7) or Ad-RNF130 (n=8). Individual dots represent individual animals. (E) FPLC lipoprotein profiles in Pcsk9^–/– mice treated with Ad-Ctr (n=7) or Ad-RNF130 (n=8). FPLC lipoprotein profiles are plotted as mean absorbance unit (AU) ±SEM. (F) Hepatic Ldlr mRNA expression in Pcsk9^–/– mice treated with Ad-Ctr (n=6) or Ad-RNF130 (n=8). Individual dots represent individual animals. (G) Hepatic LDLR protein expression from representative animals and accompanying densitometry in Pcsk9^–/– mice treated with Ad-Ctr (n=6) or Ad-RNF130 (n=7). Individual dots represent quantification of individual animals analyzed by western blot. Data are expressed as mean ±SEM. P values were determined by Mann-Whitney U-test (B) or Student’s t test (C-D, F-G). ns, not significant. Ad, adenovirus; AU, absorbance unit; Ctr, Control; HDL, high-density lipoprotein; IB, immunoblot; LDL, low-density lipoprotein; LDLR, LDL receptor; Pcsk9, proprotein convertase subtilisin kinase 9; PDI, protein disulfide isomerase; VLDL, very low-density lipoprotein.

Figure 5.. Partial knockout of Rnf130 results in decreased plasma LDL-C.

(A) Total number of pups obtained for each genotype from heterozygous breedings showing that homozygous nulls (–/–) are obtained at less than the expected mendelian ratio (n=35 litters, 204 pups total, mean litter size 5.8). (B) Hepatic mRNA expression of Rnf130 in 10-week old Rnf130^+/+ (n=11), Rnf130^+/– (n=8), and Rnf130^–/– (n=5) mice. Individual dots represent individual animals. (C) Total plasma cholesterol in 10-week old Rnf130^+/+ (n=8) and Rnf130^–/– (n=5) mice. Individual dots represent individual animals. (D) FPLC lipoprotein profiles of 10-week old Rnf130^+/+ (n=8) and Rnf130^–/– (n=5) mice. Data are expressed as mean ±SEM. FPLC lipoprotein profiles are plotted as mean absorbance unit (AU) ±SEM. (E) Representative western blot and accompanying densitometry of hepatic protein expression of LDLR in Rnf130^+/+ (n=6) and Rnf130^–/– (n=6) mice. Individual dots represent quantification of individual animals analyzed by western blot. P values were determined by Kruskal-Wallis with Dunn’s correction (B) or Mann-Whitney U-test (C,E). ns, not significant. AU, absorbance unit; HDL, high-density lipoprotein; LDL, low-density lipoprotein; VLDL, very low-density lipoprotein; +/+, wildtype; –/–, knockout.

Figure 6.. Liver-specific disruption of Rnf130 reduces plasma cholesterol levels.

(A) C57BL/6 wildtype (WT) mice were treated with 5 × 10¹¹ genome copies of Control AAV-CRISPR (n=10) or Rnf130 AAV-CRISPR (n=10) for 2 weeks. (B) Hepatic Rnf130 mRNA expression in WT mice treated as in (A). Individual dots represent individual animals. (C) Plasma total cholesterol in WT mice treated as in (A). Individual dots represent individual animals. (D, E) Hepatic LDLR protein expression and accompanying densitometry of mice treated as in (A). (D) Individual dots represent quantification of individual animals analyzed by western blot. Data are expressed as mean ±SEM. P values were determined by Student’s t-test (B-D). AAV, adeno-associated virus; IB, immunoblot; LDLR, low-density lipoprotein receptor; PDI, protein disulfide isomerase.

Figure 7.. Antisense oligonucleotide treatment to silence hepatic Rnf130 increases hepatic LDLR and decreases plasma cholesterol levels.

(A) C57BL/6 wildtype (WT) mice were treated with Control or RNF130 antisense oligonucleotide (ASO) for 4 weeks. (B) Hepatic mRNA expression of Rnf130 in WT mice treated with Control ASO (n=10) or RNF130 ASO (n=10) as in (A). Individual dots represent individual animals. (C-E) Plasma total cholesterol (C), LDL (LDL/VLDL) cholesterol (D), and FPLC lipoprotein profiles (E) in WT mice treated as in (A). (C-D) Individual dots represent individual animals. (F) Hepatic LDLR protein expression and accompanying densitometry in WT mice treated as in (A). (G) Ldlr^–/– mice were treated with RNF130 ASO for 4 weeks. (H) Hepatic Rnf130 mRNA expression in Ldlr^–/– mice treated with Control ASO (n=7) or RNF130 ASO (n=6) as in (G). Individual dots represent individual animals. (I-J) Plasma total cholesterol (I) and LDL (LDL/VLDL) cholesterol (J) in Ldlr^–/– mice treated as in (G). Individual dots represent individual animals. Data are expressed as mean ±SEM. FPLC lipoprotein profiles are plotted as mean absorbance unit (AU) ±SEM (Control ASO n=10 and RNF130 ASO n=10). P values were determined by Student’s t-test (B-D,F,H-J). ns, not significant. ASO, antisense oligonucleotide; AU, absorbance unit; CON, Control; HDL, high-density lipoprotein; IB, immunoblot; LDL, low-density lipoprotein; LDLR, LDL receptor; PDI, protein disulfide isomerase; VLDL, very low-density lipoprotein.