Treatment-Induced Viral Cure of Hepatitis C Virus-Infected Patients Involves a Dynamic Interplay among three Important Molecular Players in Lipid Homeostasis: Circulating microRNA (miR)-24, miR-223, and Proprotein Convertase Subtilisin/Kexin Type 9

Abstract

In patients with chronic hepatitis C virus (HCV) infection, viral hijacking of the host-cell biosynthetic pathways is associated with altered lipid metabolism, which contributes to disease progression and may influence antiviral response. We investigated the molecular interplay among four key regulators of lipid homeostasis [microRNA (miR)-122, miR-24, miR-223, and proprotein convertase subtilisin/kexin type 9 (PCSK9)] in HCV-infected patients (n=72) who achieved a treatment-based viral cure after interferon-based therapy with first-generation direct-acting antivirals. Real-time PCR was used to quantify microRNA plasma levels, and ELISA assays were used to determine plasma concentrations of PCSK9. We report that levels of miR-24 and miR-223 significantly increased in patients achieving sustained virologic response (SVR), whereas the levels of miR-122, a liver-specific cofactor for HCV infection, decreased in these patients. PCSK9 concentrations were significantly increased in SVRs, suggesting that PCSK9 may help impede viral infection. The modulatory effect of PCSK9 on HCV infection was also demonstrated in the context of HCV-infected Huh-7.5.1 cells employing recombinant human PCSK9 mutants. Together, these results provide insights into a novel coordinated interplay among three important molecular players in lipid homeostasis - circulating miR-24, miR-223 and PCSK9 - whose regulation is affected by HCV infection and treatment-based viral cure.

Keywords: Antiviral therapy; Hepatitis C virus; Proprotein convertase subtilisin/kexin type 9; miR-122; miR-223; miR-24.

Fig. 1

Cohort overview and study design for miRNA analysis. a) Plasma samples were collected from a cohort of 94 patients chronically infected with HCV. All patients were treated with interferon-based antiviral therapy that included pegylated interferon-alpha (PEG-IFN) and ribavirin (RBV) with/without an oral HCV protease inhibitor [boceprevir (BOC)/telaprevir (TPV)] for 24–48 weeks [Treatment Week (TW) 24-TW48]. Plasma samples were collected before [Week 0 (W0)], during treatment (TW4-TW24/TW48), and after treatment [Follow-Up Week: (FUW)] at the indicated time points (FUW12/FUW24). b) Circulating levels of plasma miR-122, miR-24, and miR-223 were quantified by quantitative real-time PCR after total RNA isolation. MicroRNA levels were normalized to a non-human spike-in control [Caenorhabditis elegans miR-39 (cel-miR-39)]. Clinical relevance of circulating levels of plasma miR-122, miR-24, and miR-223 in CHC infection and their impacts on treatment outcome were evaluated using various statistical models (see Materials and methods).

Fig. 2

Circulating miR-24 and miR-223 levels in plasma show a strong positive linear relationship in all CHC patients and negatively correlate with liver injury and liver fibrosis. (a–c) Graphs representing the negative linear relationship between miR-24 and miR-223 levels and liver injury scores [APRI (a) and FIB-4 (b)], and the robust, positive linear relationship between miR-122 and APRI scores (a), and miR-122 and HCV virus load (c). The slopes and ordinates values (see Supplementary Tables 2–4) of the plots presented in panels (a–c) were estimated by applying a mixed model analysis to our clinical data sets as described in the Materials and methods section. (d–e) Spearman's correlation plot of circulating miR-223 and miR-24 levels (d), miR-24 and miR-122 levels (e), and miR-223 and miR-122 levels (f) in all CHC patients. The rho_s value observed (rho_s = 0.91, p-value < 0.0001) is indicative of a very strong, positive monotonic correlation between circulating levels of miR-24 and miR-223 (d). In contrast, both miR-24 or miR-223 weakly correlated with miR-122 levels (rho_s = 0.33, p-value < 0.0001) (e) and (rho_s = 0.31, p-value < 0.0001) (f), respectively. Note: There is an inverse correlation between ΔCt and miRNA expression level: Lower ΔCt values are associated with increased miRNA expression.

Fig. 3

Circulating miR-24 and miR-223 plasma levels, not miR-122, significantly increase in patients who have achieved SVR. Two-way analysis of variance (ANOVA) was used to determine if plasma levels of miR-24, miR-223, and miR-122 were specifically dysregulated for the SVRs and relapsers after antiviral treatment [follow-up: (FU)] when normalized to individual baseline levels before treatment (W0). The results revealed a specific dysregulation of plasma miR-24 and miR-223 levels with SVRs, characterized by a significant 6- and 7-fold increase in circulating miR-24 and miR-223 levels, respectively (a-b) after antiviral treatment. In contrast, results of our analyses revealed that plasma miR-122 levels were specifically dysregulated for the relapsers, with a significant 8-fold increase in its circulating levels after antiviral treatment (c). Abundance of each circulating miRNA was measured in paired samples of HCV patients before (baseline: W0) and after antiviral treatment (FU). Data were normalized to individual baseline levels before treatment, expressed as the fold change and analyzed based on treatment outcomes. Data are shown as means ± SEM and compared with the two-way ANOVA. ns = non-significant.

Fig. 4

Plasma PCSK9 concentrations significantly increase in HCV-infected patients who achieved SVR. A paired dependent t-test was used to determine differences associated with plasma PCSK9 concentrations before and after treatment for SVRs (a) and relapsers (b). PCSK9 levels were significantly higher (p = 0.002) in plasma samples from SVRs (a) but not from relapsers (b). No significant differences were observed in baseline plasma PCSK9 concentrations between SVRs, relapsers, or non-responders (NRs) (c). The straight line in the data represents mean, and data were compared using a two-tailed paired t-test [(a) and (b)] and with a one-way ANOVA (c). Results are reported as the average of two technical replicates.

Fig. 5

Extracellularly applied recombinant PCSK9 and the gain-of-function mutant PCSK9-D374Y but not the loss-of-function PCSK9-R194A variant inhibit HCV infection in Huh-7.5.1 cells. a) PCSK9 is a secreted protein highly expressed in the liver where it plays an important role as a post-translational regulator of LDLR levels. It is biosynthesized as a preproprotein that contains a prodomain, catalytic domain, and cysteine-histidine-rich domain (CHRD). The PCSK9 catalytic triad contains aspartate 186 (D186), histidine 226 (H226), and serine 386 (S386). b) Gain-of-function mutations such as D374Y in PCSK9 have been associated with hypercholesterolemia due to lower levels of LDLR (10 × higher affinity for LDLR) and reduced clearance of plasma LDL. Loss-of-function PCSK9 mutations (R194A) are conversely associated with abnormally low circulating cholesterol levels due to increased LDLR abundance (lower affinity for LDLR) on the surface of liver cells. (c–d) Huh-7.5.1 cells grown in LPDS-supplemented media were treated with varying concentrations of recombinant (r) PCSK9 (r-PCSK9) or PCSK9-D374Y (r-D374Y), or (e) 25 μg/mL PCSK9-R194A (r-R194A), for 8 h. Treated cells were infected with HCV [multiplicity of infection [(MOI) = 0.5] and cells were fixed 48 h post-infection. Cells probed with HCV core-specific antibodies and stained with Hoechst dye to visualize cell nuclei were counted using Cellomics HCS to determine the percentage of total HCV-infected cells (Olmstead et al., 2012). Representative images are shown acquired at 10 × magnification objective (Cellomics HCS) (c). The EC₅₀ values for recombinant PCSK9 and PCSK9-D374Y mutant were calculated based on the dose response (d). Values are expressed as relative HCV infection in treated cells compared to untreated cells, which are set to 1. Results (mean ± SEM) from 3 independent experiments are shown (c–e).