Introduction

Hepatitis C virus (HCV) is a positive stranded RNA virus replicating in intracellular phospholipid-enriched membrane domains. Several unbiased RNAi screens identified a panel of host factors required for HCV entry, replication and assembly1,2,3,4 but none of these previous approaches discriminates effects on mRNA translation. Host protein translation is initiated with the recruitment of the 40S ribosomal subunit to mRNA. This process mostly involves the recognition of a 5′ m7GpppN cap structure by eIF4E of the cap binding complex eIF4F5. Most eukaryotic mRNAs also contain a 3′ poly(A) tail, which is acting synergistically with the cap structure to enhance translation6,7,8. Initiation of cap-dependent translation is susceptible to regulation via eIF4F by eIF4E inhibitory proteins by phosphoinositide 3-kinase (PI3K)/Akt and mammalian target of rapamycin (mTOR) signaling pathways9. Furthermore, MAP kinase pathways modulate cap-dependent translation by phosphorylation of ribosomal protein S610 and by eIF4E phosphorylation via MAP kinase interacting serine/threonine kinase 1 (MKNK1)11,12. Furthermore, cap translation is inhibited by heatshock proteins13 and by protein kinase R (PKR) and PKR-like endoplasmic reticulum kinase (PERK), which are activated by double stranded viral RNA intermediates and ER-stress, respectively14. PKR and PERK are thus triggered by a cell in despair trying to prevent viral RNA replication and to activate repair mechanisms that rely on an alternative translation initiation mechanism mediated by internal ribosomal entry sequences (IRESs). IRES translation is thus of particular physiological importance when cap-dependent translation is compromised15,16,17,18, but which is also used by some positive strand RNA viruses including HCV5,19,20 promoting viral protein synthesis21. It has been demonstrated that miR-122 stimulates HCV IRES translation20,22 and that RACK1 controls the IRES-mediated translation of viruses including HCV23 but additional host factors which are critical for HCV IRES activity remain largely to be determined. Since cellular signaling events regulate key aspects cap-dependent translation9, miRNA expression24 and the HCV life cycle2,25 we studied the role of host kinases and protein phosphatases in IRES-dependent translation.

Results

To analyze the impact of gene silencing on IRES- and cap-dependent translation, respectively, we co-transfected reporter mRNAs (100 ng/0.3 cm2) in gene silenced hepatoma cells 48 h post siRNA transfection as described previously26,27 (Fig. 1): Renilla luciferase mRNA initiated by a m7G cap structure and firefly luciferase mRNA containing a non-physiological adenosine cap structure (‘A-cap’) and the HCV IRES element. The A-cap maintains stability of the mRNA, but is not recognized by the cap binding complex. Luciferase expression was assessed by a Mithras LB 940 (Berthold Technologies) using Dual-Luciferase Reporter Assay or Bright-Glo (Promega). Toxicity of gene silencing was assessed using MTT (Sigma) and Presto Blue (Sigma) for the tertiary screen. In the primary screen targeting 893 genes we identified 46 candidates that predominant impact HCV IRES-dependent over cap-dependent translation (Supplementary table S1). In a secondary validation screen using side-by-side transfection of firefly reporter mRNAs of cap and HCV IRES (Fig. 1) we validated 11 hits of the primary screen (Supplementary table S2) and thus confirmed that these genes predominantly affect HCV IRES- rather than cap-dependent translation. As HCV IRES translation is a key step in the viral life cycle we assessed whether the identified genes confirm as positive regulators of HCV infection. We validated the results from the two foregoing screens (performed with siRNA pools) in a tertiary screen by at least two of four individual siRNAs per target (Fig. 1) to minimize off-target effects and validated mRNA knockdown specificity of the final hits by qPCR (Supplementary figure S1). As a result we confirmed that silencing of 3 genes from the secondary screening have a reproducible and significant impact on HCV infection: phosphatidylinositol 4-kinase catalytical subunit beta (PIK4CB), MAP kinase interacting serine/threonine kinase 1 (MKNK1) and tumor protein D52-like 3 (NYD-SP25) (Supplementary table S3). We were not able to validate a specific silencing of NYD-SP25 mRNA expression and therefore cannot rule out that off-target effects being responsible for the impact in IRES-dependent translation. Using a specific inhibitor of MKNK128 we demonstrate a significant (p < 0.01, t-test) and preferential inhibition of IRES-dependent translation over cap-dependent translation of luciferase reporter genes (Fig. 2a) at absent cell toxicity (Fig. 2b). Long-term treatment with the MKNK1 inhibitor over three days significantly (p < 0.01, t-test) block HCV infection (Fig. 2c) demonstrating that the molecular mechanism of action of MKNK1 involves its kinase activity.

Figure 1
figure 1

High-throughput RNAi screen identifying human kinases and protein phosphatases withpredominant impact on IRES-dependent translation and HCV infection.

Full size image
Figure 2
figure 2

Kinase activity of MKNK1 promotes IRES-dependent translation and HCV infection.

(a) A specific inhibitor of MKNK1 kinase activity preferentially impairs IRES-dependent over CAP-dependent luciferase reporter translation. Huh7.5 cells incubated for one hour with increasing doses of MKNK1 inhibitor. Treated cells were co-transfected with reporter mRNAs (IRES-firefly and cap-renilla) for four hours prior measuring of the firefly and renilla luciferase activity as described in the methods section. Data are expressed as means of the ratio of firefly/renilla luciferase activity +/− SEM. *p < 0.01 (Student’s t-test, n = 15 of three independent experiments). (b) MKNK1 inhibitor has only a minor impact on cell viability of Huh7.5 cells. After 5 h incubation with increasing concentrations of MKNK1 inhibitor the cell number was assessed by counting and the cell viability was assessed using Presto Blue. Data are expressed as means +/− SEM (n = 3 of one representative experiment). (c) MKNK1 inhibitor significantly and dose-dependently inhibits HCV infection at absent cell toxicity. Huh7.5.1 cells were pre-treated for one hour with increasing concentrations of MKNK1 inhibitor prior infection with cell culture-derived HCV (strain Luc-Jc1). Infected cells were maintained in the presence of the respective MKNK1 inhibitor concentration prior cell lysis and the measurement of the luciferase activity at day three. Data are expressed as means +/− SEM. *p < 0.01 (Student’s t-test, n = 3 of one representative experiment). All inhibitor dilutions and controls in this figure were prepared in a constant background of 1% DMSO.

Full size image

Discussion

Understanding the IRES-mediated control of HCV protein synthesis is important for the understanding of HCV infection. Although much progress has been made in understanding HCV entry and replication translational control by the HCV IRES and the involved host factors remain largely unknown. For genes identified as HCV cofactors it was mostly not conclusive whether they affect HCV translation or -replication or both steps in the viral life cycle. An established unbiased multistep screening approach identified five genes modulating HCV IRES-dependent translation. Among the genes was MKNK1, a regulator of cap translation and recently described HCV entry factor29. MKNK1 regulates EIF4E affinity to cap mRNA in fibroblasts and is required for Herpes simplex virus (HSV-1) replication. For the first time we demonstrate that MKNK1 predominantly promotes IRES-dependent translation in hepatocytes (Supplementary table S1) that likely contributes to decreased core protein expression levels observed upon MKNK1 inhibition29. A specific MKNK1 inhibitor predominantly impairs IRES over cap-dependent translation demonstrating that the molecular mechanism of action of MKNK1 involves its kinase activity. Indeed, MKNK1 inhibitor specifically and dose-dependently impairs HCV infection highlighting its potential as antiviral target for IRES-dependent viruses. No phosphatase was identified suggesting that HCV IRES translation is mainly dependent on protein phosphorylation. Interestingly, the screen identified a phosphatidylinositol 4-kinase (PI4K), another key member of the phospholipid metabolism. PI4K generate membranes enriched in phosphatidylinositide 4-phosphate lipids, which serve as replication platforms for RNA viruses from the Picornaviridae and the Flaviviridae30. Strikingly, most members of these virus families replicate in such membranous platforms (poliovirus, coxsackievirus, Aichi virus, enterovirus, HCV)30,31 also rely on IRES-dependent translation. PI4K alpha and beta isoforms have been implicated in entry, replication and packaging of HCV1,2,3,4,32. Thus our data suggest an additional and previously unrecognized role of PI4K-beta for HCV IRES-mediated translation of the viral polyprotein and potentially also for other viruses that employ an IRES mechanism. Collectively, our data identify a novel pathway of HCV-host interactions with functional relevance for the HCV life cycle and IRES-mediated translation of viral RNA.

Methods

Transcription and transfection and RT-PCR

All transfections were performed at the High Throughput Screening platform of the Institut de Génétique et de Biologie Moléculaire et Cellulaire (IGBMC) in Illkirch, France. Gene silencing by RNAi for all screens was performed on 5000 cells/3.5 pmol siRNA/0.3 cm2 in 96 well plates by reverse transfection using Interferin (Polyplus Transfection) as described previously2,25. Non-targeting scrambled siRNA (Qiagen) and RISC-free siGlo RNA (Dharmacon) were transfected as mock controls. 100 ng/0.3 cm2 reporter mRNA was transfected 48 h post siRNA transfection using TransMessager reagent (Qiagen) as described previously26. In vitro transcription of mRNAs were described previously27. Reporter mRNAs were co-transfected in a ratio (IRES-firefly: cap-renilla) of 1:4 in the primary screen. In the secondary screen reporter mRNAs (IRES-firefly, cap-firefly) were transfected side-by-side. Luciferase expression was assessed by a Mithras LB 940 (Berthold Technologies) using Dual-Luciferase Reporter Assay System System or Bright-Glo (Promega). Toxicity was assessed using MTT assay (Sigma) and Presto Blue assay (Sigma) for the tertiary screen and the MKNK1 inhibitor experiments. RNA was extracted using RNeasy kit (Qiagen) and cDNA were generated using Maxima Reverse Transcriptase (Life Technologies). qPCR was performed using RT2 SYBR Green qPCR Mastermix (Qiagen) and a C1000 Touch Thermo Cycler (Bio-Rad).

Cells culture, virus, DNA, siRNA and small molecules

Cell growth conditions of hepatoma cell lines Huh7.5 and Huh7.5.1 were described33,34. Infection with HCVcc (strain LUC-Jc1) was described2. siRNAs for the primary and secondary screen comprised the Human Kinase RNAi Set V2.0 (pool of four siRNAs) targeting 691 kinases and associated proteins and siRNAs targeting 203 human phosphatases from the Human Druggable Genome siRNA Set Version 4.0 (pools of four siRNAs) from Qiagen. Individual siRNA (four individual siRNAs per target) for the tertiary screen were obtained from Qiagen. The plasmids encoding firefly luciferase pT3FireflyLuc(pA), the renilla luciferase control plasmid pT3RenillaLuc(pA) and pHCV-IRES-luc has been described previously27,35,36. qPCR primers were obtained from Qiagen (RT2 qPCR assays). MKNK1 inhibitor was obtained from Calbiochem (Merck Millipore) and solved in DMSO.

Screening hit selection

To evaluate the impact of gene silencing on IRES and cap-dependent translation, ratios of relative light units emitted by cells transfected with IRES firefly luciferase and capped renilla luciferase reporter mRNAs were formed and values normalized by the plate median of the relative light signals. Measurements were scored calculating z-scores using CellHTS2 software. Hits were assigned from the primary screen if z = <−1.29. Secondary screening results were normalized to % luciferase activity of control cells transfected with scrambled siRNAs. Hits were assigned if target silencing decreased IRES-dependent firefly luciferase activity and if the ratio of IRES- to cap-dependent firefly luciferase activities is <0.8. An impact of target-specific siRNA (four siRNA per target) on infection with HCVcc was identified in a tertiary screen using the method of strictly standardized mean differences (SSMD)37,38 using GUItars software39. A target gene with specific impact on HCVcc was considered a hit if at least two individual siRNAs impairs HCVcc infection with an SSMD < =−1.645 (“fairly strong” inhibition) and absent toxicity (cell viability higher than a threshold of 2x plate median of standard deviations). Tertiary screen hits were scored by products of SSMD values from individual siRNAs with SSMD < =−1.645.

Additional Information

How to cite this article: Lupberger, J. et al. PI4K-beta and MKNK1 are regulators of hepatitis C virus IRES-dependent translation. Sci. Rep. 5, 13344; doi: 10.1038/srep13344 (2015).