RNA editing blood biomarkers for predicting mood alterations in HCV patients

Abstract

Treatment-emergent depression is a common complication in patients with chronic hepatitis C virus (HCV) infection undergoing antiviral combination therapy with IFN-α and ribavirin. It has recently been shown that changes in A-to-I RNA editing rates are associated with various pathologies such as inflammatory disorders, depression and suicide. Interestingly, IFN-α induces gene expression of the RNA editing enzyme ADAR1-1 (ADAR1a-p150) and alters overall RNA editing activity. In this study, we took advantage of the high prevalence of pharmacologically induced depression in patients treated with IFN-α and ribavirin to test the interest of RNA editing-related biomarkers in white blood cells of patients. In this 16-week longitudinal study, a small cohort of patients was clinically evaluated using standard assessment methods prior to and during antiviral therapy and blood samples were collected to analyse RNA editing modifications. A-I RNA editing activity on the phosphodiesterase 8A (PDE8A) gene, a previously identified RNA editing hotspot in the context of lupus erythematosus, was quantified by using an ultra-deep next-generation sequencing approach. We also monitored gene expression levels of the ADAR enzymes and the PDE8A gene during treatment by qPCR. As expected, psychiatric evaluation could track treatment-emergent depression, which occurred in 30% of HCV patients. We show that PDE8A RNA editing is increased in all patients following interferon treatment, but differently in 30% of patients. This effect was mimicked in a cellular model using SHSY-5Y neuroblastoma cells. By combining the data of A-I RNA editing and gene expression, we generated an algorithm that allowed discrimination between the group of patients who developed a treatment-emergent depression and those who did not. The current model of drug-induced depression identified A-I RNA editing biomarkers as useful tools for the identification of individuals at risk of developing depression in an objective, quantifiable biological blood test.

Keywords: A-to-I RNA editing; ADAR; Chronic hepatitis C virus; Depression; Epitranscriptomic biomarkers; Inflammation; Interferon alpha; PDE8A.

Conflict of interest statement

The authors declare that they have no competing interests.

Figures

Fig. 1

IFN-α treatment induced alterations of the relative proportion of PDE8A RNA editing sites in SH-SY5Y cell line. a NGS-based quantification of RNA editing on intron 9 of the PDE8A gene in SH-SY5Y cell culture samples for sites A–G. b Correlation, in SH-SY5Y cells, between the relative quantities (RQ) of ADAR1-1 (ADAR1a, p150) transcript level induced by increased concentrations of IFN-α and the relative increase in the proportions of isoform B in the editing profile of PDE8A RNA. The isoform B is defined as the isoform in which the edited site B is alone under the edited form. c Relationship between applied IFN-α concentrations and the mean Δ of variation of the B and non-edited (NE) isoforms in the SH-SY5Y cells. Each point represents the mean ± SEM (n = 8) of the individual values measured 48 h after administration in the incubating medium of 0, 1, 10, 100, 1000 and 10,000 IU of IFN-α

Fig. 2

Listing of RNA editing sites in PDE8A region of interest in SH-SY5Y cell line and human blood cells. a Comparison of the RNA editing sites identified in the human neuroblastoma cell line SH-SY5Y and in human blood cells. X indicates when the editing site is measured at least in one experiment/patient with a cutoff > 0.1%. b Relative proportion of PDE8A mRNA editing at all sites that display more than 0.1% editing. Histograms show mean values ± SEM (n = 5)

Fig. 3

Longitudinal analysis of ADAR enzymes gene expression and PDE8A gene expression and editing in HCV patients over the course of IFN treatment. a Longitudinal analysis of ADAR1-1 (ADAR1a, p150), ADAR1-5 (ADAR1b, p110), ADAR2 and PDE8A gene expression of all HCV patients (n = 10). b Longitudinal analysis of RNA editing on site B of the PDE8A gene of all HCV patients (n = 10). c Longitudinal analysis of RNA editing on the other identified main sites of the PDE8A gene of all HCV patients (n = 10). The mean values are displayed ± SEM (n = 10)

Fig. 4

Diagnostic performances of a combination of biomarkers. a Longitudinal analysis of an mROC combination of biomarkers composed of ADAR gene expression in the group of patients that showed clinical depression (n = 3) and the group that did not show clinical depression (n = 7). b Longitudinal analysis of an mROC combination of ADAR gene expression and RNA editing biomarkers in the group of patients that showed clinical depression (n = 3) and the group that did not show clinical depression (n = 7). c Principal component analysis of ADARs, PDE8A gene expression and PDE8A mRNA editing sites in the group of patients that showed clinical depression (n = 3) and the group that did not show clinical depression (n = 7). The scatter plot visualises the first, second and third principal components and respective variance percentages on the x-, y- and z-axis of ratio (fold change) of biomarkers before and after treatment in longitudinal analysis. Red triangle: patients with psychiatric events. Blue circle: patients without psychiatric events. d ROC curve of a combination of biomarkers to separate the population with and without psychiatric disorders (depression). Diagnostic performances of the algorithm are indicated