Induction of phospholipase A2 group 4C by HCV infection regulates lipid droplet formation

Abstract

Background & aims: Hepatic steatosis, characterized by lipid accumulation in hepatocytes, is a key diagnostic feature in patients with chronic hepatitis C virus (HCV) infection. This study aimed to clarify the involvement of phospholipid metabolic pathways in the pathogenesis of HCV-induced steatosis.

Methods: The expression and distribution of lipid species in the livers of human liver chimeric mice were analyzed using imaging mass spectrometry. Triglyceride accumulation and lipid droplet formation were studied in phospholipase A2 group 4C (PLA2G4C) knockout or overexpressing cells.

Results: Imaging mass spectrometry of the infected mouse model revealed increased lysophosphatidylcholine levels and decreased phosphatidylcholine levels in HCV-positive regions of the liver. Among the transcripts associated with phosphatidylcholine biosynthesis, upregulation of PLA2G4C mRNA was most pronounced following HCV infection. Activation of the transcription factor NF-κB and upregulation of c-Myc were important for activation of PLA2G4C transcription by HCV infection and expression of the viral proteins Core-NS2. The amount and size of lipid droplets were reduced in PLA2G4C-knockout cells. Inhibition of NF-κB or c-Myc activity suppressed lipid droplet formation in HCV-infected cells. HCV infection promoted the stabilization of lipid droplets, but this stability was reduced in PLA2G4C-knockout cells. Overexpression of PLA2G4C decreased the levels of phosphatidylcholine species in the lipid droplet fraction and led to lower levels of key factors involved in lipolysis (breakdown of triglycerides into glycerol and free fatty acids), such as ATGL, PLIN1 and ABHD5 on the lipid droplets.

Conclusions: HCV infection markedly increases PLA2G4C expression. This may alter the phospholipid composition of the lipid droplet membrane, leading to stabilization and enlargement of the droplets.

Impact and implications: The involvement of phospholipid metabolism pathways in the pathogenesis of hepatitis C virus (HCV)-related liver diseases remains unclear. We found that PLA2G4C expression is upregulated through NF-κB and c-Myc activation upon HCV infection, and this upregulation is associated with a decrease in phosphatidylcholine species. The increased expression of PLA2G4C resulted in changes in the phospholipid composition of lipid droplets, led to the dissociation of lipolysis-related factors from the lipid droplet surface and the accumulation of lipid content within the droplets. These findings suggest that the disruption of the phospholipid metabolism pathway caused by HCV infection may contribute to the development of HCV-associated fatty liver. It would be interesting to determine whether alcohol- and/or metabolic dysfunction-associated steatohepatitis are also associated with increased PLA2 activity, altered phospholipid composition and decreased levels of ATGL and its cofactors in lipid droplet membranes.

Keywords: HCV; PLA2G4C; human hepatocyte chimeric mice; lipidome analysis; liver steatosis.

Graphical abstract

Fig. 1

IMS of HCV-infected humanized mouse liver samples and quantitative analyses of signal intensities of PC and LPC species. (A) Liver samples from HCV-infected and non-infected (control) chimeric (PXB) mice were analyzed. Mouse CYP3A expression was assessed by immunohistochemical staining (first column). IMS was performed at m/z 534.3 and 772.5 (second and third columns). HCV core expression within the red boxes was detected by immunofluorescence (fourth column). Scale bar: 100 μm. (B) LPC and PC species signal intensities were measured from IMS data in the liver of HCV-infected PXB mice at 14 and 35 dpi, and in uninfected PXB mice (control). Relative levels of mean signal intensities for LPC and PC species are shown. Data are represented as mean ± SEM (n = 4 mice/group, LPC: 6 species, PC: 10 species). The relative levels of LPC and PC species signal intensities are shown in C and D, respectively. Data were analyzed using two-tailed one-way ANOVA with post hoc Tukey tests for (B) and T-test with Bonferroni correction (C and D). ∗p <0.05 vs. 14 dpi. Dpi, days post infection; HCV, hepatitis C virus; IMS, imaging mass spectrometry; LPC, lysophosphatidylcholine; PC, phosphatidylcholine.

Fig. 2

Effect of HCV infection on the expression of genes involved in the phosphatidylcholine synthesis and PLA2 activity. (A) Huh7.5.1 cells were infected with HCV (J6/JFH-1) at MOIs of 0.1 and 1. At 3 dpi, mRNA expression of enzymes involved in phosphatidylcholine synthesis (see Fig. S4) was determined by RT-qPCR, normalized to GAPDH, and compared to mock-infected cells (MOI = 0) using a two-tailed unpaired Student's t test (∗p <0.05). n.d., not determined. (B) RNA from the livers of PXB mice, with or without HCV infection at 14 dpi, was analyzed. (C) Total PLA2 activities were measured in lysates of PLA2G4C-KO cells (PLA2G4C-KO-1, PLA2G4C-KO-2), rescued KO cells (PLA2G4C-KO-1/rescued), and parental Huh7.5.1 cells (WT). (D) PLA2G4C-KO-1 cells were transfected with EV, normal PLA2G4C (G4C)-, or catalytically inactive PLA2G4C (G4C/S82A) expressing constructs. Total PLA2 activities were determined at 3 days post-transfection. (E) Total PLA2 activities were measured in lysates of Huh7.5.1 cells infected with HCV at MOIs of 0.1 and 1, and mock-infected cells, at 2.5 dpi. Data (mean ± SEM, n = 3) were analyzed using two-tailed one-way ANOVA with post hoc Tukey tests. ∗p <0.05, ∗∗p <0.01 vs. uninfected-, WT-, EV-, and mock infection groups in B-E, respectively. Dpi, days post infection; EV, empty vector; HCV, hepatitis C virus; KO, knockout; MOI, multiplicity of infection; RT-qPCR, quantitative reverse-transcription PCR; WT, wild-type.

Fig. 3

Induction of PLA2G4C mRNA by HCV proteins, and identification of the PLA2G4C promoter region responsible for HCV-mediated transcriptional activation. (A) Huh7.5.1 cells were transfected with HCV protein-expression plasmids or EV. PLA2G4C mRNA levels were determined by RT-qPCR at 2 days post-transfection, normalized to GAPDH, and compared to EV-transected cells. (B) Data (mean ± SEM, n=3) were analyzed using two-tailed one-way ANOVA with post hoc Tukey tests. ∗p < 0.05, ∗∗p < 0.01 vs. EV-, mock infection- and PLA2G4C promoter (nt -1182/+332) groups in A-C, respectively. (C) Partial deletions or mutations were introduced within the PLA2G4C promoter reporter plasmid. (Left) Huh7.5.1 cells transfected with PLA2G4C promoter-firefly luciferase reporter plasmids were infected with HCV (MOI = 0.5). Luciferase activities were measured at 2 dpi. (Right) Huh7.5.1 cells were co-transfected with core-p7 expression plasmid or EV, PLA2G4C promoter-firefly luciferase reporter plasmid, and CMV promoter-Renilla luciferase plasmid. Luciferase activities were measured at 3 days post-transfection. Data (mean ± SEM, n = 3) were analyzed using two-tailed one-way ANOVA with post hoc Tukey tests. ∗p <0.05, ∗∗p <0.01 vs. EV-, mock infection- and PLA2G4C promoter (nt -1182/+332) groups in A-C, respectively. CMV, cytomegalovirus; dpi, days post infection; EV, empty vector; HCV, hepatitis C virus; KO, knockout; MOI, multiplicity of infection; RT-qPCR, quantitative reverse-transcription PCR.

Fig. 4

Involvement of NF-κB and c-Myc in the transcriptional regulation of PLA2G4C. (A) Huh7.5.1 cells were mock-infected (-) or infected with HCV (+) and harvested at 2.5 dpi. Nuclear fractions were used for chromatin immunoprecipitation with antibodies against NF-κB, c-Myc, or rabbit IgG (control). qPCR targeted the proximal and distal PLA2G4C transcription start site. (B) Huh7.5.1 cells were harvested at 2 dpi with or without HCV infection (MOI = 1) for immunoblotting. Protein levels were quantified by ImageJ. The intensity of each band in mock infected cells was set to 1, with relative values for HCV-infected cells. (C) Huh7.5.1 cells were cultured with imiquimod (a NF-κB activator) at different concentrations for 2 days. PLA2G4C mRNA levels were assessed by RT-qPCR (upper panel), and NF-κB activation was examined by immunoblotting (lower panel). (D) Huh7.5.1 cells infected or uninfected with HCV were cultured with Bay 11-7082 (a NF-κB inhibitor) at different concentrations for 2 days. PLA2G4C mRNA levels were assessed by RT-qPCR. (E) Huh7.5.1 cells were transfected with a c-Myc expression plasmid or EV at different concentrations for 2 days. PLA2G4C mRNA levels were assessed by RT-qPCR (upper panel), and c-Myc levels were examined by immunoblotting (lower panel). (F) Huh7.5.1 cells infected with or without HCV at the indicated MOIs were cultured for 2.5 dpi. c-Myc mRNA levels were assessed by RT-qPCR. Relative RNA expression values were normalized to those of GAPDH. c-Myc induction was examined by immunoblotting (lower panel). Data (mean ± SEM, n = 3) were analyzed using two-tailed one-way ANOVA with post hoc Tukey’s tests. ∗p <0.05, ∗∗p <0.01 vs. mock infection (A,B and F), no-drug control (C and D) or 0.5 EV (E). Dpi, days post infection; EV, empty vector; HCV, hepatitis C virus; MOI, multiplicity of infection; RT-qPCR, quantitative reverse-transcription PCR.

Fig. 5

Increase in the fluorescence intensity of LD and TG accumulation following PLA2G4C overproduction. (A) PLA2G4C-KO cells (PLA2G4C-KO-1, PLA2G4C-KO-2), rescued KO cells (PLA2G4C-KO-1/rescued), and parental Huh7.5.1 cells (WT) were infected with HCV (MOI = 1). Intracellular viral RNA was quantified by RT-qPCR at 2.5 dpi. (B) Uninfected PLA2G4C-KO-1, PLA2G4C-KO-2, rescued KO, and WT cells were cultured, fixed, and stained for nuclei and LDs at 1 day. MFIs of LDs were analyzed using an IN Cell analyzer (upper panel). TG levels were determined at 3 days (lower panel). (C) PLA2G4C-KO-1, rescued KO, and WT cells were starved for 12 h, then cultured with (+) or without (-) 94 mg/ml oleic acid-albumin for 12 h. MFI of LDs were analyzed. (D) HCV-infected WT cells, PLA2G4C-KO cells rescued KO were cultured with 10058-F4 (a c-Myc inhibitor) or Bay 11-7082 (a NF-κB inhibitor) for 2 days. Total RNA was assessed for HCV RNA levels. (E) WT cells, PLA2G4C-KO cells, and rescued KO cells were cultured with a c-Myc inhibitor or an NF-κB inhibitor, and the next day, these cells were either infected (+) or uninfected (-) with HCV and harvested at 2 dpi. MFI of LDs were analyzed. Data (mean ± SEM, n = 3) were analyzed using one-way ANOVA with post hoc Tukey’s tests. ∗p <0.05, ∗∗p <0.01 for comparisons as indicated (A, B and C), or vs. the HCV-infected no-drug control (D and E). Dpi, days post infection; HCV, hepatitis C virus; KO, knockout; LD, lipid droplet; MFI, mean fluorescence intensity; MOI, multiplicity of infection; RT-qPCR, quantitative reverse-transcription PCR; WT, wild-type.

Fig. 6

LD parameter changes in PLA2G4C KO cells or HCV-infected cells. (A) Uninfected Huh7.5.1 (WT)- and PLA2G4C-KO cells (PLA2G4C-KO-1) were cultured for 1 day, fixed, and stained for nuclei and LDs. LD parameters were analyzed using an IN Cell Analyzer, including 3,693 WT cells and 4,853 PLA2G4C-KO cells. (B) WT cells were infected with HCV at various MOIs. At 2.5 dpi, cells were fixed, stained, and analyzed for LDs. (C) WT and PLA2G4C-KO cells were transfected with EV, pCAG-Core-NS2, pHH/SGR-Gluc, or both. Replicative SGR RNA (genotype 2a) was expressed from pHH/SGR-Gluc under Pol I induction. At 3 dpi, cells were fixed, stained, and analyzed for LDs. Data are presented as mean ± SEM (n = 3). Statistical analysis was performed using two-tailed one-way ANOVA with post hoc Bonferroni tests. ∗p <0.05, ∗∗p <0.01 compared as indicated. Dpi, days post infection; EV, empty vector; HCV, hepatitis C virus; KO, knockout; LD, lipid droplet; MOI, multiplicity of infection; SGR, subgenomic replicon; WT, wild-type.

Fig. 7

Decreases in lipolysis-related proteins associated with LDs following PLA2G4C overproduction. (A) Huh7.5.1 cells and PLA2G4C-KO cells, with (+) and without (-) HCV infection, were cultured with triacsin C for 9 h. Samples were collected every 3 h, stained for nuclei and LDs, and the MFI of LDs were analyzed using an IN Cell Analyzer. (B) Huh7.5.1 cells were transfected with pCAG-Myc-ATGL, pCAG-HA-PLIN1, or pCAG-ABHD5-His, along with pCAG-FLAG-PLA2G4C or EV. At 2 days post-transfection, cells were fixed, immunostained with antibodies against Myc, HA, His, and FLAG tags, and BODIPY™ 493/503 for LDs, and imaged by confocal laser scanning microscopy; representative images are shown for a sample from each group. Scale bar: 5 μm (Left panel). Fluorescence intensities of Myc-ATGL, HA-PLIN1, and ABHD5-His associated with LDs were analyzed (Right panel). Values in parentheses indicate the numbers of LDs examined. (C) Huh7.5.1 cells were transfected with pCAG-PLA2G4C (G4C) or EV and infected with HCV at MOIs of 0 (mock) or 1 (HCV). At 2 days post-transfection or infection, cells were harvested for immunoblotting of whole cell lysates and LD fractions. Equal amounts of triglycerides (4 μg/lane) were loaded for LD fraction (see Table S4). Data are presented as mean ± SEM (n = 3). Statistical analysis was conducted using two-tailed one-way ANOVA with post hoc Tukey’s tests. ∗p <0.05 vs. 0 h (A) or EV (B). EV, empty vector; HCV, hepatitis C virus; KO, knockout; LD, lipid droplet; MFI, mean fluorescence intensity; MOI, multiplicity of infection; RT-qPCR, quantitative reverse-transcription PCR; WT, wild-type.

Fig. 8

Schematic model of PLA2G4C-mediated stabilization and enlargement of LDs in HCV-infected cells. This model illustrates the mechanism for TG accumulation and LD enlargement in HCV-infected liver tissues. PLA2G4C is significantly upregulated (>100-fold) in response to HCV infection, with NF-κB and c-Myc activation as key factors. Increased PLA2G4C leads to enhanced phospholipase activity, depleting membrane PC, altering LD surface hydrophobicity, and reducing the localization of TG degradation-related factors such as ATGL, PLIN1, and ABHD5 to the LD membrane. Consequently, TG degradation in LDs is attenuated, leading to increased LD size. HCV, hepatitis C virus; LD, lipid droplet; PC, phosphatidylcholine; TG, triglyceride.