Activation of the oncogenic miR-21-5p promotes HCV replication and steatosis induced by the viral core 3a protein

Article

Reference

Activation of the oncogenic miR-21-5p promotes HCV replication and steatosis induced by the viral core 3a protein

CLEMENT, Sophie, et al.

Abstract

miR-21-5p is a potent oncogenic microRNA targeting many key tumour suppressors including phosphatase and tensin homolog (PTEN). We recently identified PTEN as a key factor modulated by hepatitis C virus (HCV) to promote virion egress. In hepatocytes, expression of HCV-3a core protein was sufficient to downregulate PTEN and to trigger lipid droplet accumulation. Here, we investigated whether HCV controls PTEN expression through miR-21-5p-dependent mechanisms to trigger steatosis in hepatocytes and to promote HCV life cycle.

CLEMENT, Sophie, et al. Activation of the oncogenic miR-21-5p promotes HCV replication and steatosis induced by the viral core 3a protein. Liver International, 2019, vol. 39, no. 7, p.

1226-1236

PMID : 30938910 DOI : 10.1111/liv.14112

Available at:

http://archive-ouverte.unige.ch/unige:124242

Activation of the oncogenic miR-21-5p promotes HCV replication and steatosis induced by the viral core 3a protein

Sophie Clément^1,*, Cyril Sobolewski^3,*, Diana Gomes⁴, Angela Rojas⁴, Nicolas Goossens², Stéphanie Conzelmann⁴, Nicolas Calo³, Francesco Negro^{1, 2, 5§} and Michelangelo Foti^{3, 5,§}

Divisions of Clinical Pathology¹ and of Gastroenterology and Hepatology², University Hospital; Departments of Cell Physiology and Metabolism³, of Pathology and Immunology⁴, and Diabetes Center⁵, Faculty of Medicine, University of Geneva, Geneva.

* and §; both authors have equally contributed to the work

Corresponding author: Michelangelo Foti, CMU, 1 rue Michel-Servet, 1206 Geneva, Switzerland. Tel: 41/22/3795204, Fax: 41/22/3795260, email: Michelangelo.foti@unige.ch Word count for the main body of manuscript: 3990

Number of figures: 6 figures, 3 supplementary figures

List of abbreviations: HCV: hepatitis C virus; HCC: hepatocellular carcinoma; SVR:

sustained virologic response; PTEN: phosphatase and tensin homolog; miRNAs: MicroRNAs;

AAV: adeno-associated virus; PMH: Primary mouse hepatocytes; LD: lipid droplets; CE:

cholesterol ester; TG: triglycerides; GEO: Gene Expression Omnibus; TCID50: 50% Tissue Culture Infectious Dose; IRS-1: Insulin substrate receptor 1

Keywords: microRNA, hepatitis C, lipid metabolism, phosphatase and tensin homolog Conflict of interest: None declared

Financial supports: The Swiss National Science Foundation N° 310030_172862 and N°CRSII3_160717 to M. Foti and N° 314730-166609 to F. Negro. The FLAGS foundation to S. Clément, M. Foti and F. Negro.

Authors’ contributions

S.Cl., C.S.: conception and design of the work; acquisition, analysis and interpretation of data;

review, drafting, and final approval of the manuscript.

D.G., A.R., N.G., S.Co. and N.C.: acquisition, analysis and interpretation of data; review and final approval of the manuscript.

F.N. and M.F.: conception and design of the work; analysis and interpretation of data; review, drafting, and final approval of the manuscript.

Abstract

Background & Aims: miR-21-5p is a potent oncogenic microRNA targeting many key tumor suppressors including PTEN. We recently identified PTEN as a key factor modulated by HCV to promote virion egress. In hepatocytes, expression of HCV-3a core protein was sufficient to downregulate PTEN and to trigger lipid droplet accumulation. Here, we investigated whether HCV controls PTEN expression through miR-21-5p-dependent mechanisms to trigger steatosis in hepatocytes and to promote HCV life cycle.

Methods: MiR-21-5p expression in HCV-infected patients was evaluated by transcriptome meta-analysis. HCV replication and viral particle production were investigated in Jc1-infected Huh-7 cells after miR-21-5p inhibition. PTEN expression and steatosis were assessed in HCV- 3a core protein-expressing Huh-7 cells and in mouse primary hepatocytes having miR-21-5p inhibited or genetically deleted, respectively. HCV-3a core-induced steatosis was assessed in vivo in Mir21a knockout mice.

Results: MiR-21-5p expression was significantly increased in hepatic tissues from HCV- infected patients. Infection by HCV-Jc1, or transduction with HCV-3a core, upregulated miR- 21-5p expression and/or activity in Huh-7 cells. miR-21-5p inhibition decreased HCV replication and release of infectious virions by Huh-7 cells. HCV-3a core-induced PTEN downregulation and steatosis were further prevented in Huh-7 cells following miR-21-5p inhibition or in Mir21a knockout mouse primary hepatocytes. Finally, steatosis induction by AAV8-mediated HCV-3a core expression was reduced in vivo in Mir21a knockout mice.

Conclusion: MiR-21-5p activation by HCV is a key molecular step, promoting both HCV life cycle and HCV-3a core-induced steatosis and may be among the molecular changes induced by HCV-3a to promote carcinogenesis.

Word count for the abstract: 243

Keywords: microRNA, hepatitis C, lipid metabolism, phosphatase and tensin homolog

Lay summary Steatosis and epigenetic alterations associated with HCV infection are important drivers of hepatocarcinogenesis. Here, we show that HCV infection activates miR-21-5p, a potent oncogenic microRNA promoting viral replication and downregulating the tumor suppressor PTEN. Steatosis induction by the HCV-3a core protein is also mediated by miR-21- 5p. miR-21-5p activation and PTEN downregulation are key molecular events in HCV life cycle, likely fostering hepatitis C progression towards liver carcinogenesis.

Introduction

Hepatitis C virus (HCV) infection is a major health issue worldwide. Its end-stage sequelae, i.e cirrhosis and hepatocellular carcinoma (HCC), are responsible for ~400,000 deaths annually. ¹ The advent of potent interferon-free treatments has reduced the incidence of HCV-related decompensated liver disease² and mortality. ³ However, the risk of developing HCC, in spite of treatment-induced viral clearance, remains significant in patients with advanced liver disease or following HCC ablative treatment. ⁴ Diabetes, age and HCV genotype 3 (HCV-3a) are independent risk factors for HCC occurrence after sustained virologic response (SVR). ⁵ Whether the association between HCV-3a and HCC⁶ proceeds via steatosis associated with this viral genotype remains unclear. ^{7, 8} Nevertheless, steatosis is a known risk factor for HCC in several chronic liver disorders and SVR may reduce its severity or even lead to its regression.

7 Thus, the residual risk of HCC occurrence after viral clearance suggests that oncogenic molecular alterations persist after SVR in the absence of non-viral cofactors. In this regard, potential oncogenic modifications of the host epigenome induced by a long-lasting HCV infection are of particular interest. ⁹

Among the molecular mechanisms induced by HCV-3a to trigger steatosis, ⁷ we reported that downregulation of the phosphatase and tensin homolog (PTEN) by a 3’-UTR-dependent post- transcriptional repression, promotes both steatosis and HCV life cycle. ^{8, 10}

MicroRNAs (miRNAs) are highly conserved small non-coding RNAs regulating post- transcriptionally gene expression by binding to the 3′-UTR sequence of target mRNAs. ¹¹ Several miRNAs were reported to inhibit PTEN expression in cancer cells. ¹² Of particular interest is miR- 21-5p, which has been implicated in the pathogenesis of liver steatosis, ^13-15 fibrosis and HCC. ¹⁶ miR-21-5p was reported to be activated by HCV infection in hepatocytes and to promote evasion from the host immune system. ¹⁷

Herein, we provide new evidence that miR-21-5p activation by HCV is a key process promoting viral life cycle and HCV-3a-induced steatosis. Our data suggest that increased miR-21-5p activity induced by HCV may contribute, among other molecular alterations, to liver carcinogenesis, even after viral clearance.

Materials and Methods

Primers, antibodies, plasmids, and reagents

Reagents, primers, antibodies and plasmids are described in Supplementary Methods.

Cells and animals

Huh-7 and Huh-7.5 cells were cultured as described. ⁸ Lentiviral transduction and transfection were performed as described in Supplementary Methods. The generation and phenotyping of control (control, Mir21a^lox/lox mice) and Mir21a knockout mice (miR21KO) was reported previously. ¹⁵ For adenovirus infection, two-month old mice were injected retro-orbitally with 5x10¹¹ genome copies of adeno-associated virus (AAV8, see Supplementary Methods) and sacrificed 10 days later for histopathological analyses of liver tissues. Primary mouse hepatocytes (PMH) were isolated as previously described. ¹⁵

Real-time PCR, immunoblot analyses and lentivirus production

RNA isolation, reverse transcription, real-time PCR, immunoblot analyses and lentivirus production are described in Supplementary Methods.

HCV production and Huh-7 cell infection

HCV infectious particles (Jc1, genotype 2a) were produced and titrated as described in Supplementary Methods. Huh-7 cells were infected with 2-5 MOI for 48-72 hours.

HCV pseudoparticle production

The plasmid phCMV 1b9.9 containing the luciferase as reporter gene was used to produce HCV pseudoparticles. ¹⁸ Control pseudoparticles were generated with the VSV-G glycoprotein.

Luciferase assay was performed 48 h after transduction using the Dual-Luciferase assay (Promega).

Subgenomic replicon assay

Huh-7 cells were transfected with 2.5 µg of pFK_i389LucNS3¹⁹ in vitro transcribed and with 0.2 µg pTK renilla_Luc using Lipofectamine 2000 (Invitrogen AG). Luciferase assay was performed 48h after transfection.

Cells/tissues Immunochemistry and lipid staining

Immunostaining of HCV-3a core and neutral lipid staining were performed on cells or tissues as described in Supplementary Methods. The surface area of individual lipid droplets (LDs) was calculated using MetaMorph® (Molecular Devices).

Lipid measurements

Intracellular triglycerides (TG) and cholesterol esters (CE) were measured using GPO/PAP (Roche Diagnostics AG) and cholesterol/cholesteryl ester quantitation kits (Calbiochem), respectively.

Meta-analysis

Meta-analysis of miR-21-5p expression in the hepatic tissues and sera of HCV-infected patients was performed using NCBI Gene Expression Omnibus (GEO) and EMBL-EBI databases as described in Supplementary Methods.

Statistical analysis

Results were expressed as mean±SEM of at least three independent experiments, and analyzed by either Student t-test or one-way analysis of variance (ANOVA). Values of p<0.001 (***), p<0.01 (**) and p<0.05 (*) were considered statistically significant.

Results

miR-21-5p expression is upregulated in the liver of HCV-infected patients

To comprehensively assess the relevance of miR-21-5p in HCV infection, we performed a meta-analysis of public, microRNA transcriptome datasets comparing miR-21-5p expression in human HCV-positive liver and serum vs. uninfected controls. Our search identified 59 studies assessing differential miRNA expression in HCV vs. controls. Among the six studies meeting the inclusion criteria, four assessed miRNAs expression in the liver^20-23 and two in the serum/plasma (Fig.1A and Fig. S1A, respectively). The meta-analysis of studies assessing miRNA expression in the liver of patients (including a total of 75 HCV patients and 41 controls) found increased miR-21-5p expression in HCV-positive livers (standardized mean difference 0.65 [95% CI 0.19-1.10, p=0.005], Fig.1B), with limited heterogeneity (I² = 39%, p=0.18). In these studies, no information was available on viral genotype and we could therefore not perform a subgroup analysis of genotype 3 patients. In the serum/plasma, however, no difference of miR21-5p according to HCV infection was found (standardized mean difference -0.33 [95% CI -0.72 – 0.06, p=0.0.98], Fig. S1B), although these results were limited by the small number of studies (n=2) included and are in contradiction with one non-systematic candidate gene study.²⁴ Thus, our meta-analysis combining all publicly available whole- genome miRNA datasets in HCV found that miR-21-5p expression is increased in human HCV-

positive livers compared to controls, a finding consistent with other candidate gene studies or non-systematic miRNA transcriptome studies in cell culture or human liver. ^{17, 25, 26}

miR21-5p promotes HCV replication

Consistent with previous studies of in vitro models of HCV infection, ^{17, 27} miR-21-5p was significantly upregulated in Jc1(genotype 2a)-infected Huh-7.5 cells (Fig.2A).

However, since miR-21-5p activity and/or bioavailability can be independent of its expression in the liver, ²⁸ we also investigated miR-21-5p bioavailability/activity in Jc1-infected cells by expressing a construct encoding a luciferase reporter gene under the control of the CMV promoter and coupled to a 3’UTR sequence containing 4 binding site for miR21 [pCMV-luc- miR21(P)] (Fig. S2A).²⁹ In cells expressing this reporter construct, the luciferase mRNA translation is inhibited when miR-21-5p binds to the 3’UTR sequence of this construct. As illustrated in Fig.2B , luciferase activity was decreased in Jc1-infected cells indicating that Jc1- induced miR-21-5p upregulation was functionally relevant since it was accompanied by an increased miR-21-5p bioavailability and binding to its target sequences.²⁹

To examine the potential role of miR-21-5p in HCV life cycle, we modulated its expression with synthetic oligonucleotides either inhibiting (anti-miR21) or mimicking (mimic miR21) miR-21-5p activity (Fig. S2B and C). Viral RNA replication was assessed by transfecting Huh- 7.5 cells with HCV subgenomic replicon (in vitro transcript produced from the pFK_i389LucNS3 construct¹⁹). Luciferase activity, which correlates with HCV polymerase activity, increased upon transfection with mimic miR21 but decreased with anti-miR21 (Fig.2C). Neither mimic miR21, nor anti-miR21, significantly affected HCV pseudoparticle entry (Fig.2D). Production of infectious viral particles (extracellular HCV RNA released) by Jc1-infected Huh-7.5 cells and their infectivity (efficiency of virions in the medium to infect naïve Huh-7.5 cells, expressed as 50% Tissue Culture Infectious Dose, TCID50) were respectively increased by mimic miR21 and decreased by anti-miR21 (Fig.2E-F). Nevertheless

the specific infectivity, calculated as the ratio of infectivity titer (TCID50/ml) to viral load (extracellular RNA), remained unchanged in both conditions (Fig. S3), indicating that the increase of HCV particles released in the medium is most likely due to the increased replication.

miR-21-5p is transiently activated during HCV-3a core-induced steatosis in hepatoma cells

Downregulation of the tumour suppressor PTEN by the core protein of HCV-3a triggers steatosis development, ⁸ but whether HCV-3a core-mediated PTEN downregulation is dependent of miR- 21-5p activity is unknown.

To answer this question, we first performed a time-dependent analysis of metabolic alterations occurring in Huh-7 hepatoma cells following HCV-3a core expression over 5 days (Fig.3).

Expression of HCV-3a core induced an increase in LD size, paralleled by a decrease in their number, suggesting that LDs undergo fusion (Fig.3A-C). While CE significantly accumulated after 2 days, accumulation of TG was detectable only after 3 days following HCV-3a core expression (Fig.3D-E), indicating distinct kinetics for CE and TG induction of biosynthesis.

PTEN downregulation occurs after 2 days of HCV-3a core expression and was maintained over time (Fig.3F), paralleled by downregulation of insulin substrate receptor-1 (IRS-1). Finally, although miR-21-5p expression was unaffected by HCV-3a core in Huh-7 cells, its activity increased significantly from day 2 to 3 post-transfection (Fig.3G-H). Thus, HCV-3a core has the capacity to enhance miR-21-5p bioavailability and binding to specific targets in a time frame compatible with induction of PTEN downregulation and lipid accumulation in Huh-7 cells.

miR-21-5p is required for HCV-3a core-mediated PTEN downregulation and

steatosis development

We further evaluated the functional relevance of miR-21-5p activation in HCV-3a-mediated PTEN downregulation and steatosis development in Huh-7 cells. First, anti-miR21 nucleotides prevented HCV-3a core-induced PTEN downregulation in Huh-7 cells (Fig.4A). Then, we assessed whether PTEN is a direct target of miR-21-5p in HCV-3a core-expressing cells using a construct encoding the luciferase gene coupled to the 3’UTR sequence of PTEN, which contains binding sites for miR-21-5p. ¹³ When cells expressed the HCV-3a core protein, luciferase activity of this reporter construct decreased (Fig.4B, condition HCV-3a core + Control anti-miR as compared to GFP + Control anti-miR). These data confirm our previous results suggesting that PTEN downregulation by the HCV-3a core is triggered by PTEN- 3’UTR-dependent mechanisms ⁸. Furthermore, when HCV-3a core-expressing cells were treated with miR-21-5p inhibitors (condition HCV-3a core + Anti-miR21), luciferase activity of the reporter construct was not decreased as compared to control conditions, indicating that binding of miR-21-5p to the 3’UTR sequence of PTEN is inhibited by Anti-miR21. These data demonstrate that downregulation of PTEN observed by Western blot in Fig.4A is mediated by miR-21-5 binding to the 3’UTR sequence of PTEN upon activation by the HCV-3a core protein.

Finally, inhibition of miR-21-5p by synthetic nucleotides restrained the formation of large LDs and accumulation of CE/TG in Huh-7 cells expressing HCV-3a core (Fig.4C-F). Together, these data indicate that PTEN downregulation and large LDs induced by HCV-3a core in Huh- 7 cells are triggered by miR21-5p activation.

In vivo genetic deletion of miR-21-5p restrains HCV-3a core-induced

accumulation of large lipid droplets in mouse hepatocytes

To confirm in vivo the role of miR-21-5p on HCV-3a-dependent lipid accumulation, control (control, Mir21a ^lox/lox mice) and Mir21a knockout mice (miR21KO mice)¹⁵ were infected with hepatotropic AAV8, ³⁰ encoding for HCV-3a core under the control of the hepatocyte-specific albumin promoter. Mice were sacrificed ten days post-infection and LD size and HCV-3a core

expression were investigated in liver tissue sections by Bodipy^® staining and immunofluorescence, respectively. More than 80% of hepatocytes expressed HCV-3a core 10 days post-infection with AAV8 (84.9±4.3% and 86.8±1.3% in control and miR21KO mice, respectively). The AAV8-mediated expression level of HCV-3a core in hepatocytes were however heterogeneous (Fig.5A) with ~10% of hepatocytes displaying a high expression, as defined by the 4^th quartile of the staining intensity. In control mice, LD size was significantly increased only in hepatocytes expressing high levels of HCV-3a core. In contrast, high expression of HCV-3a core did not trigger enlarged LDs in miR21KO mice (Fig.5B). These data indicate that in vivo high expression of HCV-3a core in mouse hepatocytes induces biosynthesis of large LDs through mechanisms requiring miR-21-5p activity.

Finally, HCV-3a core induced also accumulation of large LDs in PMH isolated from control but not from miR21KO mice (Fig.5C-D), indicating that miR-21-5p-mediated HCV-3a core- dependent accumulation of large LDs is a cell autonomous mechanism.

Together, these data demonstrate that miR-21-5p is a key intracellular mediator of hepatic metabolic alterations triggered by the HCV-3a core in vivo.

Discussion

We previously reported that HCV downregulates the tumor suppressor PTEN in vitro and in vivo, leading to clinically relevant effects on viral life cycle and steatosis development. ^{8, 10} In this study, we now provide evidence that miR-21-5p is a key mediator upstream of PTEN in these processes by showing that (i) miR-21-5p is upregulated in the liver of HCV-infected patients, (ii) miR-21-5p activity promotes viral replication and increased virion production; (iii) HCV-3a core-induced PTEN downregulation and steatosis in human hepatocytes requires miR- 21-5p, and (iv) genetic deletion of miR-21-5p in mice reduces HCV-3a-induced steatosis in vivo (Fig.6). These observations require a thorough discussion in particular since miR-21-5p

activation/upregulation may represent a key event in the pathogenesis of steatosis-associated oncogenesis.

miR-21-5p was previously shown to play a key role in the development of obesity-associated steatosis and inflammation in mice, ^{15, 31, 32} while high levels of this miRNA were also reported in steatotic livers of ethanol-fed animals and in patients with alcoholic steatohepatitis. ^{23, 33} Together with the data presented in this study, miR-21-5p upregulation/activation appears as a key and common mechanism leading to steatosis development with all major disease etiologies.

This is further supported by the downregulation/inactivation of PTEN, observed in hepatic metabolic disorders associated with different etiologies, 8, 31, 34, 35 and likely also occurring through miR-21-5p-dependent mechanisms as reported for diet-induced steatosis³¹ or HBV infection. ³⁶ Hence, it is likely that deregulation of the miR-21-5p/PTEN signaling axis represents a pathological mechanism shared by all major chronic liver disorders and therefore of high relevance for therapeutic targeting.

Based on prediction algorithms, numerous cell factors can be regulated by miR-21-5p and many others, although not predicted by bioinformatic tools, have been validated experimentally. ¹⁵ It is therefore unlikely that miR-21-5p exerts these pleiotropic effects on hepatocyte metabolism through modulation of a single cell protein. Other miR-21-5p specific targets indeed deserve further work, such as INSIG2, a transcription factor able to modulate both TG and CE anabolism in hepatocytes. ^{15, 37} Furthermore, miR-21-5p is known to target also MyD88 and IRAK1, two factors required for activation of IRF7, a master regulator of IFNα signaling. ¹⁷ miR-21-5p was therefore proposed to facilitate HCV replication by counteracting the antiviral activity of IFNα. ¹⁷ With HCV genotype 2 (e.g. Jc1), this mechanism was shown to be triggered by the NS3/4A and NS5A viral proteins whose signaling converge to promote miR-21-5p expression. ¹⁷ It thus appears that different viral genotypes can trigger miR-21-5p upregulation and/or activity through distinct mechanisms leading to pleiotropic effects promoting HCV life

cycle including lipid accumulation in large lipid droplets and/or other processes required for viral replication. Our study identified an alternative mechanism promoting miR-21-5p activity that is mediated by the HCV-3a core. Whether HCV-3a core-induced miR-21-5p activation is sufficient to stimulate HCV-3a replication and, more generally, whether this phenomenon is common to all viral genotypes replication remains to be investigated. Of note, PTEN, which is a target of miR-21-5p and is downregulated by HCV-3a core, was reported to affect MyD88 in leucocytes through the induction of miRNAs specifically targeting this factor. ³⁸ Furthermore, IRF3 activity and IFNβ production are under the control of PTEN. ³⁹ Future studies are required to understand how miR-21-5p and PTEN deregulation synergize through different mechanisms to promote viral replication and whether these mechanisms are HCV genotype dependent.

Expression of HCV-3a core in hepatocytes is sufficient to increase miR-21-5p activity, whereas hepatocyte infection with the full-length Jc1 virus (genotype 2) increases both miR-21-5p expression and activity. We cannot exclude that other viral proteins may increase miR-21-5p expression through activation of specific transcription factors, e.g. AP-1, ¹⁷ STAT-3⁴⁰ or NFkB,

31 as it has been demonstrated for Jc1. ¹⁷ MiR-21-5p is highly expressed in the liver, but it was reported to be mostly inactive, as indicated by its impaired association with polysome- associated mRNAs.²⁸ In the same study, miR-21 was shown to become activated and to interact with polysome-associated mRNAs, when stress conditions occur. miRNA bioavailability for inhibition of a specific target can be restricted by numerous mechanisms. These include at least the differential miRNA editing⁴¹, the competition for the same binding site on a specific target with long-non coding RNA (lncRNA) ⁴² or RNA-binding Proteins (RBPs). ⁴³ For instance, in the specific case of miR-21-5p, the RNA-binding Protein HuR can directly bind to miR-21-5p and block its binding to specific targets. ⁴⁴ Although such mechanisms have not yet been described in the HCV context, these observations indicate that not only the expression of miR- 21-5p, but also its bioavailability/activity needs to be considered.

Interestingly, miR-21-5p activation in hepatocytes infected with Jc1 (genotype 2) does not result in steatosis, although it is required for HCV-3a core-induced large LDs. This suggests that activation of miR-21-5p in hepatocytes is necessary but not sufficient to promote the formation of large LDs. In this regard, our previous work investigating the role of PTEN and IRS-1 in HCV core 3a-mediated steatosis already suggested the requirement of another currently unknown mechanism, in addition to alterations of PTEN/IRS-1 expressions, to develop steatosis. ⁸ Future work is warranted to unravel the effects on lipid metabolism of all 6 major HCV genotypes. More importantly, steatosis is associated with liver oncogenesis, both in HCV transgenic mice and in HCV-infected patients. ⁴⁵ miR-21-5p is a potent oncogenic miRNA overexpressed in most human cancers, ⁴⁶ whereas PTEN is a tumor suppressor frequently mutated, deleted or downregulated in human cancers. ⁴⁷ It is therefore tempting to consider that miR-21-5p upregulation and PTEN downregulation in HCV-infected hepatocytes are strong drivers of oncogenesis, and may do so via steatosis. Alternatively, steatosis may be a simple proxy for another, true oncogenetic pathway, involving or not lipid metabolism.

Besides steatosis, in fact, it is likely that alterations of other miR-21-5p targets may potentially prime hepatic carcinogenesis.

The miR-21-5p/PTEN signaling pathways are potential drug targets. The activity of miR-21- 5p can be blocked by modified synthetic nucleotides (e.g. antagomiRs) complementary to its sequence. However, a current drawback of antagomiRs is the lack of cell specificity, particularly important in the case of miR-21-5p, since this miRNA is also required for an efficient immune response. ⁴⁸ Thus, it cannot be excluded that a non-specific targeting of miR- 21-5p inhibitors may reduce immune responses directed to cancer cells thereby fostering HCC development instead of preventing it. Regarding PTEN, therapeutic approaches aiming at restoring its activity to physiological levels are also pertinent. However, the multitude and high complexity of mechanisms (epigenetic, transcriptional, post-transcriptional and post-

translational) regulating PTEN expression and activity have to date hindered the discovery of clinically relevant and selective PTEN modulators. ^{47, 49}

HCV infection accounts for 167,000 (21%) of the 810,000 liver cancer-related deaths occurred in 2015. ¹ Although treatment-induced viral clearance is associated with a significant reduction of HCC incidence, the risk remains significant in patients with advanced liver disease and prior ablative therapy of HCC. ⁴ This raises the issue of physiopathological modifications induced by the long-standing viral infection and persisting after viral clearance in the liver. Epigenetic changes have been advocated to account for the residual oncogenic risk after SVR. ⁹ There is limited evidence suggesting that mir-21-5p upregulation may have diagnostic and prognostic relevance in HCV-related HCC. ⁵⁰ A single study analyzing intrahepatic miRNA profiling of HCV-related HCC vs. other liver disease stages reported high levels of expression of miR-21- 5p in both non-tumor (cirrhotic) and tumor tissues compared to liver tissues from uninfected controls. ²² Our data identifies miR-21-5p as being activated (upon expression of the single core protein of HCV-3a) and/or upregulated (following infection with the full-length Jc1 of genotype 2) by HCV, and the experimental findings were corroborated by the meta-analysis on human tissue. Whether this upregulation can be exploited for the management of human HCC remains to be proven in prospective trials. Detection of miR21-5p in non-tumoral tissue of patients in whom HCC has been subjected to ablative treatment may provide a rationale for strict surveillance in order to early identify a recurrence. The potential use of antagomiR targeting miR21-5p to treat HCC remains hypothetical, for the reasons outlined above, but warrants further studies.

Ethics approval: Animal care and experimental procedures were performed in accordance with the Swiss guidelines for animal experimentation and ethically approved by the Geneva Health office.

Acknowledgements: We thank Dr S. Startchik and Dr N. Liaudet (Bioimaging Core Facility, Faculty of Medicine, Geneva) for their help and advice in using MetaMorph®, Dr. R.

Bartenschlager, Dr B. Bartosch, Dr. C. Rice for providing the plasmids pFK-J6/C3 (Jc1), phCMV 1b9.9 and the Huh-7.5 cells respectively, Dr B. Cullen for providing the plasmids pCMV-luc-miR21(P) and pCMV-luc-random, and C. Vesin for his help with primary mouse hepatocyte isolation. We are very grateful to Dr Dulce Alfaiate for the critical reading of the manuscript.

References:

1. WHO. Global Hepatitis Report 2017. World Health Organization 2017.

2. Cheung MCM, Walker AJ, Hudson BE, et al. Outcomes after successful direct-acting antiviral therapy for patients with chronic hepatitis C and decompensated cirrhosis. J Hepatol 2016;65:741-747.

3. Backus LI, Belperio PS, Shahoumian TA, Mole LA. Direct-acting antiviral sustained virologic response: Impact on mortality in patients without advanced liver disease. Hepatology 2018;68:827-838.

4. Baumert TF, Juhling F, Ono A, Hoshida Y. Hepatitis C-related hepatocellular carcinoma in the era of new generation antivirals. BMC Med 2017;15:52.

5. El-Serag HB, Kanwal F, Richardson P, Kramer J. Risk of hepatocellular carcinoma after sustained virological response in Veterans with hepatitis C virus infection. Hepatology 2016;64:130-137.

6. Kanwal F, Kramer JR, Ilyas J, Duan Z, El-Serag HB. HCV Genotype 3 is Associated with an Increased Risk of Cirrhosis and Hepatocellular Cancer in a National Sample of U.S.

Veterans with HCV. Hepatology 2014;60:98-105.

7. Bugianesi E, Salamone F, Negro F. The interaction of metabolic factors with HCV infection: does it matter? J Hepatol 2012;56 Suppl 1:S56-65.

8. Clement S, Peyrou M, Sanchez-Pareja A, et al. Down-regulation of phosphatase and tensin homolog by hepatitis C virus core 3a in hepatocytes triggers the formation of large lipid droplets. Hepatology 2011;54:38-49.

9. Jühling F, Hamdane N, Thumann C, et al. Hepatitis C virus-induced epigenetic changes associated with hepatocarcinogenesis persist following viral cure. The 68th Annual Meeting of the American Association for the Study of Liver Diseases: The Liver Meeting 2017:

10. Peyrou M, Clement S, Maier C, et al. PTEN protein phosphatase activity regulates hepatitis C virus secretion through modulation of cholesterol metabolism. J Hepatol 2013;59:420-426.

11. Bartel DP. MicroRNAs: target recognition and regulatory functions. Cell 2009;136:215- 233.

12. Leslie NR, Foti M. Non-genomic loss of PTEN function in cancer: not in my genes.

Trends Pharmacol Sci 2011;32:131-140.

13. Vinciguerra M, Carrozzino F, Peyrou M, et al. Unsaturated fatty acids promote hepatoma proliferation and progression through downregulation of the tumor suppressor PTEN.

J Hepatol 2009;50:1132-1141.

14. Loyer X, Paradis V, Henique C, et al. Liver microRNA-21 is overexpressed in non- alcoholic steatohepatitis and contributes to the disease in experimental models by inhibiting PPARalpha expression. Gut 2016;65:1882-1894.

15. Calo N, Ramadori P, Sobolewski C, et al. Stress-activated miR-21/miR-21* in hepatocytes promotes lipid and glucose metabolic disorders associated with high-fat diet consumption. Gut 2016.

16. Zhang J, Jiao J, Cermelli S, et al. miR-21 Inhibition Reduces Liver Fibrosis and Prevents Tumor Development by Inducing Apoptosis of CD24+ Progenitor Cells. Cancer Res 2015;75:1859-1867.

17. Chen Y, Chen J, Wang H, et al. HCV-induced miR-21 contributes to evasion of host immune system by targeting MyD88 and IRAK1. PLoS Pathog 2013;9:e1003248.

18. Bartosch B, Dubuisson J, Cosset FL. Infectious hepatitis C virus pseudo-particles containing functional E1-E2 envelope protein complexes. J Exp Med 2003;197:633-642.

19. Lohmann V, Hoffmann S, Herian U, Penin F, Bartenschlager R. Viral and cellular determinants of hepatitis C virus RNA replication in cell culture. J Virol 2003;77:3007-3019.

20. Peng X, Li Y, Walters KA, et al. Computational identification of hepatitis C virus associated microRNA-mRNA regulatory modules in human livers. BMC Genomics 2009;10:373.

21. Ramachandran S, Ilias Basha H, Sarma NJ, et al. Hepatitis C virus induced miR200c down modulates FAP-1, a negative regulator of Src signaling and promotes hepatic fibrosis.

PloS one 2013;8:e70744.

22. Diaz G, Melis M, Tice A, et al. Identification of microRNAs specifically expressed in hepatitis C virus-associated hepatocellular carcinoma. International journal of cancer 2013;133:816-824.

23. Blaya D, Coll M, Rodrigo-Torres D, et al. Integrative microRNA profiling in alcoholic hepatitis reveals a role for microRNA-182 in liver injury and inflammation. Gut 2016;65:1535- 1545.

24. Bihrer V, Waidmann O, Friedrich-Rust M, et al. Serum microRNA-21 as marker for necroinflammation in hepatitis C patients with and without hepatocellular carcinoma. PloS one 2011;6:e26971.

25. Marquez RT, Bandyopadhyay S, Wendlandt EB, et al. Correlation between microRNA expression levels and clinical parameters associated with chronic hepatitis C viral infection in humans. Laboratory investigation; a journal of technical methods and pathology 2010;90:1727- 1736.

26. Jiang J, Gusev Y, Aderca I, et al. Association of MicroRNA expression in hepatocellular carcinomas with hepatitis infection, cirrhosis, and patient survival. Clin Cancer Res 2008;14:419-427.

27. Bandiera S, Pernot S, El Saghire H, et al. Hepatitis C Virus-Induced Upregulation of MicroRNA miR-146a-5p in Hepatocytes Promotes Viral Infection and Deregulates Metabolic Pathways Associated with Liver Disease Pathogenesis. J Virol 2016;90:6387-6400.

28. Androsavich JR, Chau BN, Bhat B, Linsley PS, Walter NG. Disease-linked microRNA- 21 exhibits drastically reduced mRNA binding and silencing activity in healthy mouse liver.

RNA 2012;18:1510-1526.

29. Zeng Y, Cullen BR. Sequence requirements for micro RNA processing and function in human cells. RNA 2003;9:112-123.

30. Zincarelli C, Soltys S, Rengo G, Rabinowitz JE. Analysis of AAV serotypes 1-9 mediated gene expression and tropism in mice after systemic injection. Mol Ther 2008;16:1073-1080.

31. Vinciguerra M, Sgroi A, Veyrat-Durebex C, et al. Unsaturated fatty acids inhibit the expression of tumor suppressor phosphatase and tensin homolog (PTEN) via microRNA-21 up- regulation in hepatocytes. Hepatology 2009;49:1176-1184.

32. Wu H, Ng R, Chen X, Steer CJ, Song G. MicroRNA-21 is a potential link between non- alcoholic fatty liver disease and hepatocellular carcinoma via modulation of the HBP1-p53- Srebp1c pathway. Gut 2016;65:1850-1860.

33. Francis H, McDaniel K, Han Y, et al. Regulation of the extrinsic apoptotic pathway by microRNA-21 in alcoholic liver injury. J Biol Chem 2014;289:27526-27539.

34. Sanchez-Pareja A, Clement S, Peyrou M, et al. Phosphatase and tensin homolog is a differential diagnostic marker between nonalcoholic and alcoholic fatty liver disease. World J Gastroenterol 2016;22:3735-3745.

35. Shearn CT, Smathers RL, Backos DS, et al. Increased carbonylation of the lipid phosphatase PTEN contributes to Akt2 activation in a murine model of early alcohol-induced steatosis. Free Radic Biol Med 2013;65:680-692.

36. Damania P, Sen B, Dar SB, et al. Hepatitis B virus induces cell proliferation via HBx- induced microRNA-21 in hepatocellular carcinoma by targeting programmed cell death protein4 (PDCD4) and phosphatase and tensin homologue (PTEN). PLoS One 2014;9:e91745.

37. Sever N, Yang T, Brown MS, Goldstein JL, DeBose-Boyd RA. Accelerated degradation of HMG CoA reductase mediated by binding of insig-1 to its sterol-sensing domain. Mol Cell 2003;11:25-33.

38. Sisti F, Wang S, Brandt SL, et al. Nuclear PTEN enhances the maturation of a microRNA regulon to limit MyD88-dependent susceptibility to sepsis. Sci Signal 2018;11.

39. Li S, Zhu M, Pan R, et al. The tumor suppressor PTEN has a critical role in antiviral innate immunity. Nat Immunol 2016;17:241-249.

40. Li S, Liang Z, Xu L, Zou F. MicroRNA-21: a ubiquitously expressed pro-survival factor in cancer and other diseases. Mol Cell Biochem 2012;360:147-158.

41. Li L, Song Y, Shi X, et al. The landscape of miRNA editing in animals and its impact on miRNA biogenesis and targeting. Genome Res 2018;28:132-143.

42. Abdollahzadeh R, Daraei A, Mansoori Y, et al. Competing endogenous RNA (ceRNA) cross talk and language in ceRNA regulatory networks: A new look at hallmarks of breast cancer. J Cell Physiol 2018.

43. Young LE, Moore AE, Sokol L, Meisner-Kober N, Dixon DA. The mRNA stability factor HuR inhibits microRNA-16 targeting of COX-2. Mol Cancer Res 2012;10:167-180.

44. Poria DK, Guha A, Nandi I, Ray PS. RNA-binding protein HuR sequesters microRNA- 21 to prevent translation repression of proinflammatory tumor suppressor gene programmed cell death 4. Oncogene 2016;35:1703-1715.

45. Pekow JR, Bhan AK, Zheng H, Chung RT. Hepatic steatosis is associated with increased frequency of hepatocellular carcinoma in patients with hepatitis C-related cirrhosis. Cancer 2007;109:2490-2496.

46. Pfeffer SR, Yang CH, Pfeffer LM. The Role of miR-21 in Cancer. Drug Dev Res 2015;76:270-277.

47. Lee YR, Chen M, Pandolfi PP. The functions and regulation of the PTEN tumour suppressor: new modes and prospects. Nat Rev Mol Cell Biol 2018.

48. He W, Wang C, Mu R, et al. MiR-21 is required for anti-tumor immune response in mice: an implication for its bi-directional roles. Oncogene 2017;36:4212-4223.

49. McLoughlin NM, Mueller C, Grossmann TN. The Therapeutic Potential of PTEN Modulation: Targeting Strategies from Gene to Protein. Cell Chem Biol 2018;25:19-29.

50. Shi K-Q, Lin Z, Chen X-J, et al. Hepatocellular carcinoma associated microRNA expression signature: integrated bioinformatics analysis, experimental validation and clinical significance. Oncotarget 2015;6:25093-25108.

Figure legends:

Figure 1: Expression of miR-21 in the liver of HCV-infected patients. (A-B) Meta-analysis of publicly available transcriptomic datasets of miR-21 expression in HCV infected human liver samples compared to controls. Table in A shows characteristics of the 4 studies included in the meta-analysis. (B) Forest plot shows absolute differences in normalized gene expression of individual study and pooled estimate. Size of the square for each individual study reflects weight of the study in the pooled estimate.

CI, confidence interval; SMD, standardised mean difference.

Figure 2: Role of miR-21 in HCV life cycle. (A-B) Effect of HCV Jc1 (genotype 2a) infection on miR-21 expression and activity. Seventy-two hours after infection of Huh-7 cell with the Jc1 full length virus, miR-21 expression was assessed by RT-qPCR (A) and miR-21 activity was measured using a pCMV-Luc containing four miR-21 target sequences at the 3’-UTR of luciferase (B). (C) HCV replication was measured in cells co-transfected with the subgenomic replicon pFK_i389LucNS3 in vitro transcript and either anti-miR21, control anti-miR, miR-21 mimic

or control mimic nucleotides. Luciferase activity was measured at 48h post-transfection (D) HCV entry was assessed on cells transfected with the different nucleotides and further transduced with HCVpp. Luciferase assay was performed 48h post-transduction. (E-F) Cells transfected with the nucleotides were infected with Jc1. Viral particle production was estimated by measuring the level of extracellular HCV RNA by RT-qPCR (E) in the supernatant. In parallel, supernatants were used to infect naïve Huh 7.5 cells to investigate viral infectivity by TCID50/ml calculation (F). Results are means±SEM of at least three independent experiments.

***P < 0.001, **P < 0.01, and *P < 0.05.

Figure 3: Effect of HCV-3a core protein expression over time on lipid droplet morphology (A- C), lipid content (D and E), PTEN and IRS-1 expression (F) and miR-21 expression and activity (G and H). Representative confocal image of ORO (red) and GFP or core (green) staining of Huh-7 cells transduced with GFP (left) or HCV-3a core (right) for 1-5 days (A). Quantification of individual lipid droplet size (B) or number (C) in 20 or more cells using MetaMorph®. (D, E) Cholesterol ester (CE) and triglyceride (TG) levels were measured and values were normalized to the number of cells. (F) Representative immunoblots and quantification of IRS- 1 and PTEN expression in Huh-7 cells transduced with either GFP or HCV-3a core expressing lentiviectors. (G) Expression of miR-21 in Huh-7 cells transduced with either GFP or HCV-3a core expressing lentivectors assessed by RT-PCR. (H) Luciferase activity of a pCMV-Luc containing four miR-21 target sequences at the 3’-UTR of luciferase [pCMV-luc-miR21(P)].

Results are means±SEM of three independent experiments. Only statistically significant differences were indicated in the plots. ***P < 0.001, **P < 0.01, and *P < 0.05.

Figure 4: Effect of anti-miR21 on HCV core3a-induced PTEN downregulation (A-B), lipid droplet size increase (C-D) and lipid content (E-F). Huh-7 cells either untransfected (UT) or previously transfected with anti-miR21 or control anti-miR, were transduced with a lentivector expressing either GFP or HCV-3a core protein. Seventy-two hours later, PTEN expression was

analyzed by immunoblot (A); pLuc-PTEN-3’-UTR luciferase activity was measured (B) and the size of lipid droplets was estimated by immunofluorescence after ORO staining (C) and quantified using MetaMorph® (D). Cholesterol ester (CE) and triglyceride (TG) levels were measured and values were normalized to the number of cells. (E and F). Results are means±SEM of at least three independent experiments. Only statistically significant differences were indicated in the plots. ***P < 0.001, **P < 0.01, and *P < 0.05.

Figure 5: Effect of HCV-3a core on lipid droplet accumulation in miR-21 KO mice. (A and B) Liver cryosections of miR-21 knockout (miR21KO) or control Mir21a ^lox/lox (Control) mice infected with 5x10¹¹ genome copies of adeno-associated virus serotype 8 (AAV-8) carrying either HCV-3a core or GFP genes under the control of albumin promoter were stained with anti-HCV core antibody (red) and bodipy (green). Representative confocal images are shown in A. The mean surface area of lipid droplets (LD) and HCV core staining level were calculated using MetaMorph®. High core expressing cells were defined as the 4^th quartile of the core staining data. Insets correspond to zoomed cutouts of HCV core-high expressing cells. (B) Each dot on the plot represents the mean surface area of lipid droplets for one individual mice.

(C-D) Primary hepatocytes isolated from either miR21KO or control mice were transduced with a lentivector expressing either GFP or the HCV-3a core protein. Seventy-two hours later, the size of lipid droplets was estimated by immunofluorescence after Oil-Red-O staining (C) and quantified using MetaMorph® (D). Results are means±SEM of at least five mice per groups. Only statistically significant differences were indicated in the plots. ***P < 0.001, **P

< 0.01, and *P < 0.05.

Figure 6: Hypothetical model of the role of HCV-mediated activation of miR21-5p in viral replication, steatosis development, potentially leading to oncogenesis.

Activation of the oncogenic miR-21-5p promotes HCV replication and steatosis induced by the viral core 3a protein

Sophie Clément, Cyril Sobolewski, Diana Gomes, Angela Rojas, Nicolas Goossens, Stéphanie Conzelmann, Nicolas Calo, Francesco Negroand Michelangelo Foti

Table of Contents

I- Supplementary materials and methods ... 2

Primers, antibodies, plasmids and reagent ... 2

Primary antibodies... 2

Secondary antibodies ... 2

Plasmids and HCV constructs ... 3

Antagomir and mimic oligonucleotides ... 3

Real time-PCR primers:... 4

Other reagents: ... 4

AAV8 production and infection... 5

Cell transduction and transfection... 5

RNA isolation, reverse transcription and real-time PCR ... 6

Immunoblot analyses... 6

In vitro transcription ... 6

HCV particles production, titration ... 6

Cells/tissues Immunochemistry and lipid staining ... 7

Meta-analysis ... 7

II- Supplementary Figures ... 8

III- Supplementary references ... 10

I- Supplementary materials and methods

Primers, antibodies, plasmids and reagent Primary antibodies

Targeted protein Host Clone Provider Catalogue number

PTEN Mouse A2B1 Santa Cruz Biotechnology,

Inc., Dallas, Texas, USA

SC-7974

IRS1 Rabbit Santa Cruz Biotechnology,

Inc., Dallas, Texas, USA

SC-559

HCV core Mouse C7-50 Covalab (Lyon France) mab51009

β-cytoplasmic actin Mouse C4 A gift from C. Chaponnier, Geneva, Switzerland Secondary antibodies

Host Provider Catalogue

number

HRP-conjugated anti-mouse

Goat Bio-Rad Laboratories AG (Cressier, Switzerland)

170-6516

HRP-conjugated anti-rabbit

Goat Bio-Rad Laboratories AG (Cressier, Switzerland)

170-6515

Alexa Fluor 488 conjugated anti- rabbit IgG

Goat Invitrogen AG, Basel Switzerland A-11008

Alexa Fluor 555- conjugated anti- mouse IgG

Goat Invitrogen AG, Basel Switzerland 21127

HRP conjugated anti-Goat

Rabbit Sigma-Aldrich Chemie GmbH (Buchs, Switzerland)

A-5420

AffiniPure Fab fragment goat anti- mouse IgG (H+L)

Goat Jackson ImmunoResearch Europe Ltd., Cambridgeshire, UK

115-007-003

Plasmids and HCV constructs

Plasmid name Plasmid insert Backbone References/sources pIRES2-EGFP-core

3a

HCV-3a core pIRES2-EGFP (Clontech Laboratories, CA, USA)

(1)

pLenti-GFP / pLenti- HCV-2a core / pLenti-HCV-3a core

GFP / HCV-2a core / HCV-3a core

2K7 (Invitrogen AG, Basel Switzerland)

(2)

pFK_i389LucNS3 HCV replicon I389/NS3-3’UTR (GeneBank:

AJ242654.1)

A gift from R.

Bartenschlager, Heidelberg, Germany (3)

pFK-J6/C3 (Jc1) Full lengh HCV A gift from R.

Bartenschlager, Heidelberg, Germany (4)

phCMV 1b9.9 HCV E1/E2 A gift from B. Bartosch,

Lyon, France (5) pLuc-PTEN3’UTR Luciferase-

PTEN3’UTR

pCMV-luc REF : LR-1001 Signosis, Inc. (CA, USA)

pCMV-luc-miR21(P) mir21(P) targets pCMV-luc A gift of B. Cullen (6) (Addgene plasmid # 20382) pCMV-luc-random random sequence

targets

pCMV-luc A gift from B. Cullen (6) (Addgene plasmid # 20877) pTKrenilla_Luc Renilla Luciferase D. Garcin, Geneva,

Switzerland Antagomir and mimic oligonucleotides

Name Provider Catalogue number

Antagomir 21: miRIDIAN microRNA human HSA-miR- 21-5p - Hairpin Inhibitor

Dharmacon, Inc., Lafayette, CO, USA IH-300492-05-0002

Control Antagomir : - miRIDIAN microRNA Hairpin Inhibitor Negative Control #1

Dharmacon, Inc., Lafayette, CO, USA

IN-001005-01-05

Mimic miR21: miRIDIAN Mimic hsa-miR-21

Dharmacon, Inc., Lafayette, CO, USA C-300492-03-0005

Control mimic : miRIDIAN microRNA Mimic Neg control

Dharmacon, Inc., Lafayette, CO, USA CN-001000-01-05

Real time-PCR primers:

Name Forward Reverse

human eEF1A1 AGCAAAAATGACCCACC AATG

GGCCTGGATGGTTCAGGATA

human Gus B CCACCAGGGACCATCCAA T

AGTCAAAATATGTGTTCTGGACAA AGTAA

HCV Jc1 Primers KK30 and KM3 (7).

HCV core TCCTAAACCTCAAAGAAA

AACCAAA TCCTGTGGGCGGCG

miR21-3p CAACACCAGTCGATGGGC

miR21-5p TAGCTTATCAGACTGATGTTG

miR16 ACAGCCTAGCAGCACGTAAAT

mm-u-sno234 GGCTTTTGGAACTGAATCTAAGTG

Hs_SNORD61 Hs_SNORD61_11 miScript Primer Assay (MS00033705, Qiagen AG, Hombrechtikon, Switzerland)

Other reagents:

Name Provider Catalogue number

Jetprime® Polyplus, Berkeley, CA, USA 114-01 Interferin® Polyplus, Berkeley, CA, USA 409-10 Lipofectamine 2000 Invitrogen AG, Basel Switzerland 11668-027

ECL reagent Amersham, Switzerland RPN2135

Dual-Luciferase Reporter Kit Promega, Dübendorf, Switzerland E1910 BODIPY™ 493/503 (4,4-

Difluoro-1,3,5,7,8-

Pentamethyl-4-Bora-3a,4a- Diaza-s-Indacene)

Invitrogen AG, Basel Switzerland D3922

Oil-Red O Sigma-Aldrich Chemie GmbH, Buchs, Switzerland

75087

Nucleospin RNA II Macherey-Nagel AG, Oensingen, Switzerland

740.955.50

QIAamp Viral RNA Mini Kit Qiagen AG, Hombrechtikon, Switzerland

52904

Transcriptor Universal cDNA Master kit

Roche Diagnostics AG, Rotkreuz, Swizerland

05 893 151 001

LightCycler® 480 SYBR Green I Master

Roche Diagnostics AG, Rotkreuz,

Swizerland 04 887 352 001

miRNeasy Micro Kit Qiagen AG, Hombrechtikon,

Switzerland 217084

miScript II RT Kit Qiagen AG, Hombrechtikon,

Switzerland 218161

miScript SYBR Green PCR kit

Qiagen AG, Hombrechtikon,

Switzerland 218073

triglyceride kit Roche Diagnostics AG, Rotkreuz,

Swizerland 12016648-122

Cholesterol/cholesteryl ester

quantitation kit Calbiochem, San Diego, CA, USA 428901-1

AAV8 production and infection

Adeno-associated virus serotype 8 (AAV8) carrying either the HCV-3a core or GFP genes under the control of albumin promoter (pAAV-ALB-core3a-IRES-GFP and pAAV-ALB-GFP) were produced by Vector Biolabs (Malvern, PA, USA) using the plasmid pIRES-EGFP-core 3a as template (1). Mice were injected retro-orbitally with 5x10¹¹ genome copies and sacrificed 10 days later for histopathological analyses of liver tissues.

Lentivirus production

The lentiviral particles were produced and collected as reported (8). Viral titers were estimated by real-time PCR (9). Cells transduced with a GFP-encoding lentivector were used as controls.

Cell transduction and transfection

Cells were seeded at 1.5x10⁴ cells/well in a 24-well plate and cultured 6 hours before lentiviral transduction. For transfection, cells were plated at 3.5x10⁴ cells/well in a 24-well plate and

Interferin® (Polyplus) 6 hours prior either lentiviral transduction or HCVcc infection. Plasmid transfection was performed using JetPrime® (Polyplus), according to the manufacturer protocol.

RNA isolation, reverse transcription and real-time PCR

Total intracellular RNA was extracted using the Nucleospin RNA II Kit (Macherey-Nagel AG, Oensingen, Switzerland). Extracellular RNA level of HCV particles was extracted with a QIAamp Viral RNA Mini Kit (Qiagen AG, Hombrechtikon, Switzerland). Relative quantification of RNA transcripts was then performed by real-time RT-PCR as described (10).

For microRNA measurement, RNA extraction, reverse transcription and real-time PCR were performed using miRNA-dedicated kits (miRNeasy Micro, miScript II RT, miScript SYBR Green PCR, Qiagen AG), according to the manufacturer instructions.

miR16/Sno234/SNORD61 were used to normalize miR-21 expression.

Immunoblot analyses

Cells were lysed in ice-cold RIPA containing 2 mM sodium orthovanadate,10 mM sodium fluoride, 2 mM EDTA and protease inhibitors. Equal amounts of proteins were separated by 10% SDS-PAGE, blotted to nitrocellulose membranes and proteins of interest blotted with specific antibodies (see above table) were revealed with ECL. Quantifications were performed using the ImageJ software.

In vitro transcription

Chimeric JFH1-J6 (Jc1) full length RNA was generated from pFK-J6/C3 (Jc1) (gift from R.

Bartenschlager, Heidelberg, Germany; (4)), by using the T7 Ribo Max Express Large scale RNA Production System (Promega, Dübendorf, Switzerland). In vitro transcripts were purified using the Nucleospin RNAII kit (Macherey Nagel), and their quality verified using the Agilent 2100 bioanalyzer.

HCV particles production, titration

HCV infectious particles were produced by electroporating Huh-7.5 cells (7.5 x 106) with 5 µg of HCV RNA transcribed from pFK-J6/C3 (Jc1) using Amaxa cell line nucleofector kit T (260 V, 950 µF; Lonza, Basel, Swizerland).. Particles were collected in culture supernatant 48 hours after transfection, filtered through 0.45 µm pore-sized PVDF membrane and titered by infecting Huh-7.5 cells with serial dilutions. Cells were fixed after 72 hours with -20°C methanol and immunostained using an anti-HCV-core (C7-50) antibody. TCID50/ml was calculated as reported (4). Huh-7 cells were infected with 2-5 MOI for 48-72 hours. Specific infectivity of

HCV particles produced by Huh-7 cells was calculated as ratio of infectivity titer (TCID50/ml) to viral load (extracellular RNA).

Cells/tissues Immunochemistry and lipid staining

Paraformaldehyde-fixed cells were incubated with anti-HCV core antibody for 30 minutes at RT, and subsequently with Alexa488-conjugated anti-mouse antibody and DAPI for nuclear staining, for 30 minutes at RT. Neutral lipids were stained with Oil-Red-O (ORO).

For mouse tissue, fresh livers were embedded and frozen in OCT using liquid nitrogen and 2- methyl-butane, and then stored at -80°C. Five µm-thick liver cryosections were fixed with 4%

of paraformaldehyde for 15 minutes at RT. Microwave-citrate antigen retrieval was performed for 10 minutes at 98°C, followed by blocking with AffiniPure Fab fragment goat anti-mouse IgG (H+L) and bovine serum. Liver cryosections were then incubated with anti-HCV core antibody for 30 minutes at RT, and subsequently with Alexa555-conjugated anti-mouse antibody and DAPI, for 30 minutes at RT. Neutral lipids were stained with Bodipy 493/503.

Images were acquired with a confocal LSM700 microscope (Carl Zeiss AG, Feldbach, Switzerland). The surface area of individual LD was calculated using the MetaMorph®

software (Molecular Devices, San Jose, CA, USA).

Meta-analysis

The NCBI Gene Expression Omnibus (GEO) and EMBL-EBI were searched for relevant with the search string “(microrna OR miRNA) AND (hepatitis C OR HCV)” on September 13th 2016. The primary outcome of interest was the difference in expression of miR-21-5p in hepatitis C positive human samples versus uninfected controls. A study was considered eligible if it performed a whole genome miRNA assessment in hepatitis C positive samples and uninfected controls, if miR-21-5p was measured and if the study was performed in human liver or serum/plasma samples. The full text of potential articles was retrieved and assessed for inclusion, any disagreements were resolved by the senior author FN.

We retrieved the individual studies in GEO or Array-Express. As multiple array platforms were used within the difference studies we used the data pre-processed as deposited in GEO or Array- Express, we collapsed all probes to the corresponding gene and scaled the observations of each gene to the z-score. To account for the difference in scales, we performed meta-analysis on the standardized mean difference using random effects meta-analysis due to expected heterogeneity of measures. Meta-analysis was performed using the metafor R package, and all statistical analyses were performed using the R statistical language.

II- Supplementary Figures

Supplementary Figure 1: Expression of miR-21 in the serum or plasma of HCV-infected patients. (A-B) Meta-analysis of publicly available datasets of miR-21 expression in HCV- infected human serum/plasma samples compared to controls. Table in A shows characteristics of the 2 studies included in the meta-analysis. (B) Forest plot shows absolute differences in normalized gene expression of individual studies and pooled estimate. Size of the square for each individual studies reflects weight of the study in the pooled estimate.

CI, confidence interval; SMD, standardised mean difference.

Supplementary Figure 2: Validation of the miR-21 activity test. (A) Schematic representation of the pCMV-Luc containing four miR-21 target sequences at the 3’-UTR of luciferase [pCMV-luc-miR21(P)]. (B-C) Luciferase activity measured in Huh-7 cells transfected with either miR21 mimic, control mimic (B); anti-miR21 and control anti-miR (C) nucleotides using the pCMV-luc-miR21(P) and the pCMV-luc-random as control (6).

Supplementary Figure 3: Specific infectivity of HCV particles produced by Huh-7 cells transfected with either anti-miR21, control anti-miR, miR-21 mimic or control mimic nucleotides was calculated as ratio of infectivity titer (TCID50/ml) to viral load (extracellular RNA).

III- Supplementary references

1. Abid K, Pazienza V, de Gottardi A, Rubbia-Brandt L, Conne B, Pugnale P, et al. An in vitro model of hepatitis C virus genotype 3a-associated triglycerides accumulation. J Hepatol 2005;42:744-751.

2. Clement S, Pascarella S, Conzelmann S, Gonelle-Gispert C, Guilloux K, Negro F. The hepatitis C virus core protein indirectly induces alpha-smooth muscle actin expression in hepatic stellate cells via interleukin-8. J Hepatol 2010;52:635-643.

3. Lohmann V, Hoffmann S, Herian U, Penin F, Bartenschlager R. Viral and cellular determinants of hepatitis C virus RNA replication in cell culture. J Virol 2003;77:3007-3019.

4. Pietschmann T, Kaul A, Koutsoudakis G, Shavinskaya A, Kallis S, Steinmann E, et al.

Construction and characterization of infectious intragenotypic and intergenotypic hepatitis C virus chimeras. Proc Natl Acad Sci U S A 2006;103:7408-7413.

5. Bartosch B, Dubuisson J, Cosset FL. Infectious hepatitis C virus pseudo-particles containing functional E1-E2 envelope protein complexes. J Exp Med 2003;197:633-642.

6. Zeng Y, Cullen BR. Sequence requirements for micro RNA processing and function in human cells. RNA 2003;9:112-123.

7. Takeuchi T, Katsume A, Tanaka T, Abe A, Inoue K, Tsukiyama-Kohara K, et al. Real- time detection system for quantification of hepatitis C virus genome. Gastroenterology 1999;116:636-642.

8. Dull T, Zufferey R, Kelly M, Mandel RJ, Nguyen M, Trono D, et al. A third-generation lentivirus vector with a conditional packaging system. J Virol 1998;72:8463-8471.

9. Sastry L, Johnson T, Hobson MJ, Smucker B, Cornetta K. Titering lentiviral vectors:

comparison of DNA, RNA and marker expression methods. Gene Ther 2002;9:1155-1162.

10. Clement S, Juge-Aubry C, Sgroi A, Conzelmann S, Pazienza V, Pittet-Cuenod B, et al.

Monocyte chemoattractant protein-1 secreted by adipose tissue induces direct lipid accumulation in hepatocytes. Hepatology 2008;48:799-807.