ABCC4 impairs the clearance of plasma LDL cholesterol through suppressing LDLR expression in the liver

Abstract

Low expression level of low-density lipoprotein receptor (LDLR) in hepatocytes leads to hypercholesterolemia and eventually contributes to atherosclerotic cardiovascular disease (ASCVD). Here, we report that inhibition of hepatocyte ABCC4, identified as a top hit from large-scale CRISPR/Cas9 screens, significantly increases hepatic LDLR abundance and enhances LDL cholesterol clearance. As a hepatic transporter for cAMP efflux, ABCC4 silencing alters its intracellular distribution and activates the downstream Epac2/Rap1a signaling pathway, which ultimately blocks PCSK9 protein expression, thereby preventing lysosomal degradation of LDLR. Furthermore, in both male mice and cell models, we demonstrate that liver-specific disruption and pharmacological inhibition of ABCC4 elevate hepatic plasma membrane LDLR levels and reduce plasma LDL cholesterol through ABCC4-cAMP-PCSK9 pathway. Collectively, our genome-wide CRISPR screening offers a valuable resource for identifying LDLR modifiers, providing potential insights for therapeutic strategies in hypercholesterolemia and atherosclerosis.

Fig. 1. CRISPR screen identifies ABCC4 as a negative regulator of hepatic LDL receptor.

a Schematic workflow of genome-scale CRISPR screening process. b, c MAGeCK gene enrichment scores comparing LDLR^high subpopulation and unsorted cells from two independent screens (M1 and M2). d Venn diagram showing genes identified in two independent biological replicates (M1 and M2).

Fig. 2. ABCC4 depletion increases the cell surface LDLR level independent of its gene expression.

a, b CRISPR/Cas9–mediated knockout of Abcc4 in AML12 cells (a) and LO2 cells (b) using two independent sgRNAs. Immunoblotting analysis of ABCC4 protein expression and Vinculin in control-KO cells and Abcc4-KO cells (Rosa26 sgRNA, Abcc4 sgRNA #1, Abcc4 sgRNA #2). b Immunoblotting experiments of ABCC4 and Vinculin in control-KO cells and ABCC4-KO cells (AAVS1 sgRNA, ABCC4 sgRNA #1, #2). c Flow cytometry data showing that knockout of Abcc4 increased the abundance of LDLR on the AML12 cell surface (Representative data from n = 3 independent experiments with similar results). d Plot showing the relative Mean Fluorescence Intensity (MFI) of PE-LDLR from three independent experiments. e Immunoblotting analysis of plasma membrane fractions demonstrating that Abcc4 knockout dramatically increased the amount of LDLR on PM. f Quantification of band intensity of LDLR protein expression relative to Na/K-ATPase from three independent experiments. g Cultured AML12 cells were precooled to 4°C for 30 min and incubated with Dil-LDL for binding at 4°C. Then the cells were washed 3 times with ice-chilled PBS. The cells were substantially switched to 37 °C for uptake. h Representative immunofluorescence microscopy images showing that Abcc4 knockout in AML12 cells had potentiated influence on DiI-LDL binding (0 h) and uptake (1 h). Blue: DAPI; Red: Dil-LDL. Scale bar: 50 μm. i Dil-LDL uptake assay implying that LDL uptake was significantly promoted in Abcc4-deficient cells by flow cytometry analysis (Representative data from n = 3 independent experiments with similar results). j The relative MFI of Dil-LDL quantification were from 3 independent experiments. k FC data showing that knockout of ABCC4 promotes the cell surface LDLR accessibility in LO2 cells (Representative data from n = 3 independent experiments with similar results). l The relative MFI of LDLR-PE quantification was from three independent experiments. m Immunoblotting analysis of LDLR quantification located on the plasma membrane fractions relative to Na/K-ATPase in LO2 cells. n Flow cytometry analysis showing that LDL uptake by LDLR was significantly promoted in LO2 cells lacking ABCC4 (Representative data from n = 3 independent experiments with similar results). o The relative MFI of Dil-LDL quantification were from three independent experiments. Statistical analysis was performed by a ordinary one-way ANOVA followed by Bonferroni’s multiple comparison test in (d), (f), (j), (l), (o). *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001, ns: no significance. Data are the mean ± SEM. Source data are provided as a Source Data file.

Fig. 3. Liver-specific disruption of ABCC4 promotes hepatic plasma membrane LDLR protein expression and lowers plasma LDL cholesterol levels in vivo.

a WT C57BL/6 male mice were treated with control AAV_GFP_RNAi or AAV_Abcc4_RNAi by tail vein injection (2×10¹¹ viral genomes per mouse, n = 6 per group) for 3 weeks. b Hepatic Abcc4 mRNA expression level in WT mice between two groups. c Representative immunoblotting data of hepatic ABCC4 protein expression level in WT mice between two groups. d Quantification of band intensity of ABCC4 protein level relative to Vinculin in liver tissue from two groups. e Representative immunoblotting data of membrane LDLR protein expression from liver tissue membrane fractions in mice. f Quantification of band intensity of membrane LDLR protein level relative to Na/K-ATPase in liver tissue from two groups. g Serum LDL-C level in WT mice between two groups. h Serum TC level in in WT mice between two groups. i Serum TG level in WT mice between two groups. Statistical analysis was performed by an unpaired two-tailed Student’s t-test in (b), (d), (f), (g), (h), (i). **P ≤ 0.01, ***P ≤ 0.001, ns: no significance. Data are the mean ± SEM. Source data are provided as a Source Data file.

Fig. 4. Pharmacologic inhibition of ABCC4 by Ceefourin-1 facilitates plasma LDL cholesterol clearance by enhancing surface LDLR protein expression on hepatocytes.

a Flow cytometry data showing that ABCC4 inhibitor treatment potentiates surface LDLR availability (Representative data from n = 5 independent experiments with similar results). b Plot showing the relative Mean Fluorescence Intensity (MFI) of PE-LDLR from five independent experiments. c Immunoblotting experiments of the plasma membrane fractions demonstrating that ABCC4 inhibitor treatment up-regulates surface LDLR protein expression in AML12 cells. d Quantification of band intensity of surface LDLR protein expression relative to Na/K-ATPase in AML12 cells (Data from n = 3 independent experiments). e Wild-type male mice (n = 6 per group) were injected intraperitoneally three times a week with a dose of Ceefourin-1 (10 mg/kg) or vehicle control (DMSO and corn oil) for 4 weeks under normal-chow diet (NCD) or high-fat diet (HFD) conditions. f Plot displaying changes of body weight in mice intraperitoneally with vehicle control and ABCC4 inhibitor Ceefourin-1 under a NCD or HFD condition. g Serum LDL-C level in WT mice treated as in (e). h Serum TC level in WT mice treated as in (e). i Serum TG level in WT mice treated as in (e). j Liver TC level in WT mice treated as in (e). k Liver TG level in WT mice treated as in (e). l Hematoxylin and eosin (H&E) and Oil Red O staining analysis revealing that ABCC4 inhibitor treatment improved lipid accumulation, especially under a HFD condition. Scale bar: 50 μm. m Immunoblotting data of LDLR protein expression from liver plasma membrane fractions in NDC-fed mice treated with Ceefourin-1 or vehicle. n Representative immunoblotting data of LDLR expression from liver plasma membrane fractions in HFD-fed mice treated with Ceefourin-1 or vehicle. o Quantification of band intensity of LDLR protein level relative to Na/K-ATPase in liver plasma membrane fractions from the mice treated as in (e). Statistical analysis was performed by an unpaired two-tailed Student’s t-test in (b), (d); a ordinary one-way ANOVA followed by Bonferroni’s multiple comparison test in (g), (h), (i), (j), (k), (o). *P ≤ 0.5, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001, ns: no significance. Data are the mean ± SEM. Source data are provided as a Source Data file.

Fig. 5. Altered intracellular distribution of cAMP by ABCC4 blocks hepatic PCSK9 expression.

a Principal component analysis showing distinct clustering of transcriptomes of the samples between Abcc4-knockout (Abcc4 sgRNA) and control-knockout (Rosa26 sgRNA) AML12 cells. b Hierarchically clustered circos heatmap of top 43 differentially expressed genes (DEGs) between two groups (P < 0.05; fold change>2.0). c KEGG pathway enrichment analysis of DEGs. d, e GSEA showing activated cAMP (d) and Rap1 (e) signaling pathways for DEGs between two groups. f Heatmap showing the enriched pathways related to lipid metabolism process, insulin secretion signaling, cAMP signaling, cGMP-PKG signaling, and Rap1 signaling for DEGs. g ELISA measurements showing increased intracellular and reduced extracellular cAMP levels in Abcc4-deficient cells. h ELISA measurements showing increased intracellular and reduced extracellular cAMP levels in AML 12 cell treated with the ABCC4 inhibitor Ceefourin-1. i Immunoblotting experiments showing PCSK9 protein levels in Abcc4-deficient cells. j Quantification of band intensity of PCSK9 protein expression relative to Vinculin in Abcc4-deficient cells from three independent experiments. k Secreted PCSK9 levels in Abcc4-deficient cells. l RT-qPCR results assessing Pcsk9 mRNA level in Abcc4-deficient cells from three independent experiments. m Immunoblotting experiments showing PCSK9 protein level in AML12 cell treated with DMSO or Ceefourin-1. n Quantification of band intensity of PCSK9 protein level relative to Vinculin between two groups from three independent experiments. o Secreted PCSK9 levels in AML12 cells treated with DMSO or Ceefourin-1. p Pcsk9 relative mRNA expression level by RT-qPCR between two groups from three independent experiments. q Immunoblotting analysis of PCSK9 protein levels in Abcc4-deficient cells treated with cycloheximide (cyclo: 4 μg/mL) for 30 min (+) and 90 min (++). r Flow cytometry analysis of Dil-LDL uptake assay (1 h) in Abcc4-deficient cells treated with Vehicle or rhPCSK9 protein. The relative MFI of Dil-LDL quantification were from three independent experiments. s Immunoblotting analysis of LDLR expression from the plasma membrane fractions in Abcc4-deficient cells. Statistical analysis was performed by a Welch ANOVA test followed by a post hoc analysis using the Tamhane T2 method in (g); an unpaired two-tailed Student’s t-test in (h), (n), (o), (p); a ordinary one-way ANOVA followed by Bonferroni’s multiple comparison test in (j), (k), (l), (r). *P ≤ 0.5, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001, ns: no significance. Data are the mean ± SEM. Source data are provided as a Source Data file.

Fig. 6. ABCC4-cAMP signaling regulating cell surface LDLR availability is dependent on essential roles of Epac2/Rap1a.

a Generation of Abcc4/Epac2 double-knockout (DKO) AML12 cells using two sgRNAs targeting Epac2 gene. Immunoblotting analysis of ABCC4, EPAC2 and Vinculin protein expression in AML12 cells (Rosa26 sgRNA, Abcc4 sgRNA #1, Abcc4 sgRNA #2, Abcc4/Epac2 sgRNA #1, Abcc4/Epac2 gRNA #2). b Immunoblotting analysis of LDLR protein expression from the plasma membrane fractions in AML12 cells (Rosa26 sgRNA, Abcc4 sgRNA #1, Abcc4 sgRNA #2, Abcc4/Epac2 sgRNA #1, Abcc4/Epac2 gRNA #2). c Flow cytometry analysis of Dil-LDL uptake assay (1 h) in AML12 cells (Rosa26 sgRNA, Abcc4 sgRNA, Abcc4/Epac2 sgRNA). d The relative MFI of Dil-LDL quantification were from 3 independent experiments. e Immunoblotting analysis of PCSK9 protein expression in AML12 cells (Rosa26 sgRNA, Abcc4 sgRNA #1, Abcc4 sgRNA #2, Abcc4/Epac2 sgRNA #1, Abcc4/Epac2 gRNA #2). f Secreted PCSK9 levels in AML12 cells (Rosa26 sgRNA, Abcc4 sgRNA, Abcc4/Epac2 sgRNA) from four independent experiments. g Relative Pcsk9 mRNA expression level in AML12 cells (Rosa26 sgRNA, Abcc4 sgRNA, Abcc4/Epac2 sgRNA) from four independent experiments. h Generation of Abcc4/Rap1a DKO AML12 cells using two independent sgRNAs targeting Rap1a gene. Immunoblotting analysis of ABCC4, RAP1A and Vinculin protein expression in AML12 cells (Rosa26 sgRNA, Abcc4 sgRNA #1, Abcc4 sgRNA #2, Abcc4/Rap1a sgRNA #1, Abcc4/Rap1a sgRNA #2). i Immunoblotting analysis of LDLR protein expression from the plasma membrane fractions in AML12 cells (Rosa26 sgRNA, Abcc4 sgRNA #1, Abcc4 sgRNA #2, Abcc4/Rap1a sgRNA #1, Abcc4/Rap1a sgRNA #2). j Flow cytometry analysis of Dil-LDL uptake assay (1 h) in AML12 cells (Rosa26 sgRNA, Abcc4 sgRNA, Abcc4/Rap1a sgRNA). k The relative MFI of Dil-LDL quantification were from 3 independent experiments. l Immunoblotting analysis of PCSK9 protein expression in AML12 cells (Rosa26 sgRNA, Abcc4 sgRNA #1, Abcc4 sgRNA #2, Abcc4/Rap1a sgRNA #1, Abcc4/Rap1a sgRNA #2). m Secreted PCSK9 levels in AML12 cells (Rosa26 sgRNA, Abcc4 sgRNA, Abcc4/Rap1a sgRNA) from four independent experiments. n Relative Pcsk9 mRNA expression level in AML12 cells (Rosa26 sgRNA, Abcc4 sgRNA, Abcc4/Rap1a sgRNA) from four independent experiments. Statistical analysis was performed by a Welch ANOVA test followed by a post hoc analysis using the Tamhane T2 method in (d), (k); a ordinary one-way ANOVA followed by Bonferroni’s multiple comparison test in (f), (g), (m), (n). *P ≤ 0.5, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001, ns: no significance. Data are the mean ± SEM. Source data are provided as a Source Data file.