Biochemical Journal

Research article

Ribavirin-induced intracellular GTP depletion activates transcription elongation in coagulation Factor VII gene expression

Atsuo Suzuki , Yuhri Miyawaki , Eriko Okuyama , Moe Murata , Yumi Ando , Io Kato , Yuki Takagi , Akira Takagi , Takashi Murate , Hidehiko Saito , Tetsuhito Kojima

Abstract

Coagulation FVII (Factor VII) is a vitamin K-dependent glycoprotein synthesized in hepatocytes. It was reported previously that FVII gene (F7) expression was up-regulated by ribavirin treatment in hepatitis C virus-infected haemophilia patients; however, its precise mechanism is still unknown. In the present study, we investigated the molecular mechanism of ribavirin-induced up-regulation of F7 expression in HepG2 (human hepatoma cell line). We found that intracellular GTP depletion by ribavirin as well as other IMPDH (inosine-5′-monophosphate dehydrogenase) inhibitors, such as mycophenolic acid and 6-mercaptopurine, up-regulated F7 expression. FVII mRNA transcription was mainly enhanced by accelerated transcription elongation, which was mediated by the P-TEFb (positive-transcription elongation factor b) complex, rather than by promoter activation. Ribavirin unregulated ELL (eleven-nineteen lysine-rich leukaemia) 3 mRNA expression before F7 up-regulation. We observed that ribavirin enhanced ELL3 recruitment to F7, whereas knockdown of ELL3 diminished ribavirin-induced FVII mRNA up-regulation. Ribavirin also enhanced recruitment of CDK9 (cyclin-dependent kinase 9) and AFF4 to F7. These data suggest that ribavirin-induced intracellular GTP depletion recruits a super elongation complex containing P-TEFb, AFF4 and ELL3, to F7, and modulates FVII mRNA transcription elongation. Collectively, we have elucidated a basal mechanism for ribavirin-induced FVII mRNA up-regulation by acceleration of transcription elongation, which may be crucial in understanding its pleiotropic functions in vivo.

  • coagulation Factor VII
  • eleven-nineteen lysine-rich leukaemia 3 (ELL3)
  • gene up-regulation
  • positive-transcription elongation factor b (P-TEFb)
  • ribavirin
  • transcription elongation

INTRODUCTION

FVII (Factor VII), a vitamin K-dependent plasma glycoprotein, is synthesized in the liver and secreted into the blood as a single-chain zymogen at a concentration of 500 ng/ml in plasma [1,2]. Upon vascular injury and in the presence of calcium, FVII forms a one-to-one stoichiometric complex with its cell-surface receptor and cofactor TF (tissue factor). Once in a complex with TF, FVII is rapidly cleaved to its active form FVIIa, and converts zymogen Factor IX and Factor X into active enzymes [3,4]. The formation of an active complex between TF and FVIIa is widely thought to be the primary stimulus for blood coagulation.

Haemophilia is an X-linked recessive bleeding disorder caused by a deficiency in the activity of coagulation Factor VIII or IX. One of the treatments for haemophilia is replacement of lacked coagulation Factor VIII (haemophilia A) or IX (haemophilia B); however, the development of an alloimmune antibody inhibiting Factor VIII or IX currently represents the most serious complication, particularly if the antibody is classified as highly responding, where substitution with Factor VIII or IX fails to provide adequate haemostasis [5]. In haemophilia patients with high responding inhibitors, haemostatic treatments include clotting Factor VIII- or IX-bypassing agents, such as a prothrombin complex concentrate, plasma-derived activated prothrombin complex or rFVIIa (recombinant FVIIa) [6]. In particular, rFVIIa eliminates the risk of human blood-transmitted diseases and acts by enhancing the natural coagulation pathway through activation of prothrombinase complex formation and has a local action only in areas where TF and/or phospholipids are exposed [3,7].

In 2006, Yamamoto et al. [8] reported that the anti-HCV (hepatitis C virus) agent ribavirin elevated the activity of FVII, leading to reduced doses of clotting factors that are used in haemostatic therapy for patients with chronic hepatitis C. They demonstrated increased mRNA expression of the F7 (FVII) gene by ribavirin treatment in HepG2 (human hepatoma cell line) and in cultured human hepatocytes; however, the detailed molecular mechanisms for this phenomenon remain unknown.

Ribavirin is a purine nucleoside analogue used in anti-HCV therapy. The proposed mechanisms for the antiviral actions of ribavirin include inducing error catastrophe, inhibiting the activity of RNA-dependent RNA polymerase of HCV, inhibiting the activity of IMPDH (inosine-5′-monophosphate dehydrogenase) to decrease the GTP pool and modulating the immune system [9]. Although the actions of ribavirin are partially understood, the mechanisms are not completely clear.

The aim of the present study was to investigate further the molecular mechanisms of increased FVII mRNA expression by ribavirin treatment in HepG2 cells. To date, there have been few reports on the effects of ribavirin on endogenous gene expression in mammalian cells. It is important to elucidate the molecular mechanisms of ribavirin before understanding its in vivo functions.

EXPERIMENTAL

Cell culture and reagents

The HepG2 cell line was purchased from the A.T.C.C. (Manassas, VA, U.S.A.) and cultured as described previously [10]. HepG2 cells were incubated with ribavirin, MPA (mycophenolic acid) or 6-MP (6-mercaptopurine) (Sigma) in DMEM (Dulbecco's modified Eagle's medium) (Wako) supplemented with 10% (v/v) FBS (fetal bovine serum) (JRH Biosciences) and 5 μg/ml vitamin K1 (Isei). DRB (5,6-dichlorobenzimidazole 1-β-D-ribofuranoside) was purchased from Sigma.

Total RNA isolation and qRT-PCR (quantitative real-time PCR)

RNA preparation and qRT-PCR were performed as described previously with minor modifications [10]. Total RNA was extracted from the cells using the RNeasy mini kit (Qiagen) and the first strand cDNA was prepared from 1 μg of total RNA using the PrimeScript™ RT reagent kit (TaKaRa Bio). qRT-PCR was performed with SYBR® Premix ExTaq II to detect GusB, HPRT1, ELL (eleven-nineteen lysine-rich leukaemia) 3 and TCEB3 mRNA, and with SYBR® Premix Dimer Eraser to detect FVII and ELL2 mRNA, using the primers described in Supplementary Table S1 at http://www.BiochemJ.org/bj/449/bj4490231add.htm. The level of mRNA expression in all experiments was calculated as relative values of the respective mRNAs normalized to GusB mRNA.

Protein extraction and Western blotting

Nuclear protein was extracted from HepG2 cells treated with or without 100 μg/ml ribavirin using the CelLytic™ NuCLEAR™ Extraction Kit (Sigma) according to the manufacturer's protocol.

Western blotting was performed as reported previously [10]. Membranes were blocked with excess protein (1%, w/v, non-fat dried skimmed milk powder) in PBS supplemented with 0.05% Tween 20. Primary antibodies against CDK (cyclin-dependent kinase) 9, histone H1 (Santa Cruz Biotechnology), Pol II (RNA polymerase II) 8WG16 (Covance), Pol II CTD (C-terminal domain) pSer2 (pSer is phosphorylated serine), Pol II CTD pSer5 (Abcam), ELL3 (Abnova) or α-tubulin (Sigma) were used. When using anti-pSer antibodies, membranes were blocked with 0.5% BSA in Tris-buffered saline supplemented with 0.05% Tween 20. Signals were visualized using Immobilon-Western chemiluminescent substrate (Millipore).

Luciferase reporter assay

The human F7 promoter region was prepared by PCR amplification from human genomic DNA. The PCR products from −1593 to −1 or −722 to −1 (with A of the ATG translation initiation codon as +1) were inserted into the pGL4.10 vector (Promega) (F7–1593/pGL4 or F7–722/pGL4 respectively). The PROS1 (protein S) gene promoter region was cloned previously [10], and the PROS1 promoter fragments were recombined into pGL4.10 vectors.

HepG2 cells were seeded into 35-mm-diameter dishes at a concentration of 105 cells in DMEM supplemented with 10% FBS. After 16 h, the appropriate F7 luciferase reporter plasmids (200 ng) or empty vector and 50 ng of pGL4.74 vector (Promega) were transiently co-transfected using the Lipofectamine 2000™ reagent (Invitrogen), according to the manufacturer's protocol. Following the 4-h transfection period, the culture medium was exchanged for fresh DMEM supplemented with 10% FBS and vitamin K1, with or without 100 μg/ml of ribavirin. The cells were treated with ribavirin and harvested at 24 h. Luciferase activities were subsequently determined with a luciferase assay system (Promega) according to the manufacturer's protocol, with luciferase activity normalized to the activity of co-transfected hRluc (human-optimized Renilla luciferase) as an internal control for transfection efficiency.

Nuclear run-on assay

The nuclear run-on assay was performed as reported previously with minor modifications [11]. HepG2 cells were grown in 100-mm-diameter dishes and cultured with or without ribavirin (100 μg/ml) or MPA (100 μM) for 24 h. The cells were washed twice with chilled PBS and intact nuclei were collected. Total RNA was isolated using the RNeasy mini kit. The biotinlabelled RNA-binding pull-down assay was performed using Dynabeads® M-280 Streptavidin (Invitrogen). The RNA-bound beads were suspended in 10 μl of RNase-free water, and the bound RNA was used as a template for reverse transcription and subjected to qRT-PCR as described above.

Nascent RNA capturing

Newly synthesized RNA transcripts were determined by the Click-iT® Nascent RNA Capture Kit (Invitrogen) according to the manufacturer's protocol. HepG2 cells were cultured with or without ribavirin (100 μg/ml) for 16 h and subsequently pulsed with 0.2 mM EU (5-ethynyl uridine) at 37°C for 8 h. The cells were washed and collected, and total RNA was extracted. A Click reaction was performed using 5 μg of EU-labelled RNA and 0.5 mM biotin azide; the mixture was incubated at room temperature (25°C) for 30 min. Following RNA precipitation, the RNA was dissolved in 50 μl of RNase-free water. Biotin-labelled EU–RNA-binding pull-down assay was performed using 50 μl of Dynabeads® MyOne™ Streptavidin and the bound RNA was washed and used as a template for reverse transcription. The captured nascent RNA was analysed using qRT-PCR as described above.

We also used the Click-iT® Nascent RNA Capture kit to determine mRNA decay. HepG2 cells were treated with ribavirin (100 μg/ml) and 5-EU (0.2 mM) at 37°C for 24 h, and the medium was replaced with growth medium without 5-EU. Total RNAs were isolated at several time points (0, 12 and 24 h after exchanging the medium) and subjected to the Click reaction. The reaction was performed using 1 μg of EU-labelled RNA and 0.25 mM biotin azide. A biotin-labelled EU–RNA pull-down assay was performed using 15 μl of Dynabeads® MyOne™ Streptavidin.

Microarray

Gene expression profiles of HepG2 cells treated or untreated with ribavirin were obtained by microarray analysis using the Human 3D-Gene Oligo chip 25K from outsourcing to Toray.

RNAi (RNA interference)

HepG2 cells were transfected with ELL3-specific siRNA (small interfering RNA) (siGENOME SMARTpool siRNA, human ELL3; Dharmacon) or NS (non-specific) siRNA (siGENOME Non-Targeting siRNA#3; Dharmacon) using Lipofectamine RNAiMAX (Invitrogen) by the reverse transfection protocol according to the manufacturer's protocol. The siRNAs (20 μM) were combined with Opti-MEM (Invitrogen) and mixed with Lipofectamine RNAiMAX reagent. After incubation at room temperature for 20 min, 2.5×105 cells were seeded into 60-mm-diameter dishes and cultured for 24 h. The cells were then treated with ribavirin (100 μg/ml) for 24 h and nuclear protein or total RNA was extracted as described above.

ChIP (chromatin immunoprecipitation)

ChIP was performed as described previously with minor modifications [10]. HepG2 cells were cultured with or without ribavirin (100 μg/ml) for 18 h, and fixed with 1% formaldehyde at 37°C for 10 min. The cross-linking reaction was stopped by the addition of glycine, at a final concentration of 0.125 M. The chromatin supernatants were precleared with Dynabeads® Protein G and immunoprecipitated with respective antibodies specific to Pol II 4H8, Pol II CTD pSer2, Pol II CTD pSer5 (Abcam), CDK9, AFF4, ELL3 and NS normal rabbit IgG (Santa Cruz Biotechnology) at 4°C for 3 h. The protein–antibody complexes were incubated with Dynabeads® Protein G at 4°C for 1 h. The beads were extensively washed and the immune complexes were extracted from the beads using elution buffer, and the protein–DNA cross-linking was reversed by incubation at 65°C for overnight. Bound DNA was purified with a High Pure PCR Cleanup Micro Kit (Roche Applied Science) and used as a template for subsequent quantitative PCR performed with a SYBR® Premix ExTaq II reagent. The primers used for quantitative PCR are listed in Supplementary Table S2 at http://www.BiochemJ.org/bj/449/bj4490231add.htm.

Statistical analysis

Data are presented as means±S.D. and are representative of at least three independent experiments. Significant differences (P<0.05) between experimental groups in the qRT-PCR and luciferase assays were observed using Student's t test.

RESULTS

Effects of ribavirin treatment on FVII mRNA expression in HepG2 cells

In the present study, we investigated further the molecular mechanisms of up-regulated F7 expression by ribavirin treatment. We first examined FVII mRNA levels in HepG2 cells treated with various concentrations of ribavirin (for 24 h) and observed increased FVII mRNA expression in a dose-dependent manner (Figure 1A). We also analysed the effects of ribavirin treatment (100 μg/ml) at various time points in HepG2 cells and observed increased FVII mRNA levels in a time-dependent manner (Figure 1B). The referenced GusB gene expression was slightly increased by ribavirin treatment, but not in a dose-dependent and time-dependent manner (Supplementary Figure S1 at http://www.BiochemJ.org/bj/449/bj4490231add.htm). Therefore we used ribavirin at a concentration of 100 μg/ml for 24 h in the following experiments.

Figure 1 Quantification of FVII mRNA in HepG2 cells treated with or without ribavirin

qRT-PCR was performed to determine FVII mRNA levels in HepG2 cells. (A) Dose-dependent increase in FVII mRNA expression in HepG2 cells treated with ribavirin for 24 h. The cells were treated with several different concentrations of ribavirin (10, 25, 50, 100 or 500 μg/ml). (B) Time-course analysis of FVII mRNA levels under ribavirin treatment in HepG2 cells. The cells were treated with 100 μg/ml ribavirin for 6, 12, 24, 48 or 72 h and total RNA from the cells were isolated as described in the Experimental section. The expression levels of FVII mRNA normalized to GusB mRNA were shown as relative values to those of untreated cells. *P<0.05 compared with untreated cells. **P<0.01 compared with untreated cells.

Effects of other IMPDH inhibitors on F7 expression

Ribavirin is metabolized in the cytoplasm to RMPs (ribavirin monophosphates), which are known to inhibit IMPDH in de novo synthesis of purine metabolisms. As the immunosuppressants MPA and 6-MP are also known to be IMPDH inhibitors, we examined the effects of MPA and 6-MP on F7 gene expression in HepG2 cells (Figure 2A). MPA and 6-MP were used at concentrations of up to 100 μM to treat HepG2 cells for 24 h. We found that these IMPDH inhibitors induced an approximate 3–4-fold increase in FVII mRNA levels at concentrations ranging from 1 to 100 μM.

Figure 2 Induction of FVII mRNA expression in HepG2 cells by treatment with MPA and retraction by treatment with guanosine

(A) HepG2 cells were treated with MPA or 6-MP (0–100 μM) for 24 h and total RNA was isolated from the cells. FVII mRNA levels were determined by qRT-PCR and expressed as relative values to that of untreated HepG2 cells. (B) FVII mRNA induced by ribavirin or MPA treatment was retracted by the addition of guanosine in HepG2 cells. The expression levels of FVII mRNA normalized to GusB mRNA were shown as relative values to those of untreated cells. *P<0.05 compared with untreated cells. **P<0.01 compared with untreated cells.

As treatment with the IMPDH inhibitors resulted in reduced intracellular guanine nucleoside pools, we tested whether the addition of guanosine would reverse FVII mRNA up-regulation (Figure 2B). The increased FVII mRNA expression in the HepG2 cells was retracted following supplementation of guanosine (100 μM) in the presence of ribavirin or MPA. These results confirmed that IMPDH inhibitors induced intracellular GTP depletion and increased FVII mRNA expression, which was retracted by the addition of guanosine by retrieving the guanine nucleotides.

Ribavirin-induced FVII mRNA up-regulation through transcription elongation

To investigate how ribavirin contributes to FVII mRNA up-regulation in HepG2 cells, we analysed the effects of ribavirin treatment on F7 transcription activity. We assessed newly synthesized FVII mRNA transcripts with the nuclear run-on assay under ribavirin treatment (Figure 3A). The nuclear run-on assay demonstrated that in vitro F7 transcriptants in ribavirin-treated HepG2 cells were five times more abundant than those in the untreated cells. We measured HPRT1 mRNA as a ribavirin-insensitive control, which showed no significant increase by ribavirin treatment. We also observed that MPA augmented newly synthesized FVII mRNA in HepG2 cells. This observation was confirmed by measuring the nascent FVII mRNAs using the Click-iT® Nascent mRNA Capturing Kit. Nascent FVII mRNA synthesis was increased ~8-fold in HepG2 cells treated with ribavirin compared with untreated cells (Figure 3B), whereas HPRT1 mRNA synthesis did not show any significant changes. We performed luciferase-reporter analyses, and observed a small increase (~1.5-fold) in F7 promoter activity by ribavirin treatment (Supplementary Figure S2A at http://www.BiochemJ.org/bj/449/bj4490231add.htm); however, we also observed a similar increase (~1.3-fold) in ribavirin-insensitive PROS1 promoter activity (Supplementary Figures S2B and S2C). These results suggested that the up-regulation of F7 promoter activity by ribavirin might be caused mainly by a background effect, probably as a result of the pGL4.10 vector itself, in the present study. Meanwhile we could not find significant changes in FVII mRNA stability (Figure 3C).

Figure 3 De novo synthesis of FVII mRNA in ribavirin-treated HepG2 cells

The mechanisms of ribavirin-induced FVII mRNA up-regulation in HepG2 cells were investigated. (A) Nuclear run-on assay was performed using HepG2 cells treated with or without ribavirin and MPA for 24 h, and intact nuclei were isolated. Newly synthesized transcripts were labelled with biotin-16-UTP and collected using Streptavidin-conjugated magnet beads. Relative levels of HPRT1 and FVII mRNAs normalized to GusB mRNA were determined in vitro by qRT-PCR. (B) Measurement of nascent mRNA in HepG2 cells treated with ribavirin was conducted as described in the Experimental section. Relative expression levels of HRPT1 and FVII mRNAs were normalized to GusB mRNA. (C) FVII and HRPT1 mRNA stability in HepG2 cells treated with or without ribavirin. The mRNA stability was determined using Click-iT® Nascent RNA Capture kit as described in the Experimental section. The results show mRNA decay of FVII and HPRT1 in HepG2 cells. (D) Inhibition of transcription elongation by DRB. HepG2 cells were treated with 50 μM DRB for 8 h before exposure with ribavirin (100 μg/ml) for 16 h. DMSO was added as the vehicle control for DRB. Relative expression levels of HPRT1 and FVII mRNAs were normalized to GusB mRNA. *P<0.05 compared with untreated cells.

To assess the impact of ribavirin at the late stages of transcription, we examined the effects of the transcription elongation inhibitor DRB on FVII mRNA up-regulation in ribavirin-treated HepG2 cells (Figure 3D). DRB is known to inhibit the activity of CDK9, which is a major component of P-TEFb [12]. In untreated HepG2 cells, DRB (50 μM) caused only a slight decrease in FVII mRNA expression. By contrast, DRB abolished the increased FVII mRNA expression in HepG2 cells treated with ribavirin.

Microarray analyses for gene expression profiles induced by ribavirin treatment in HepG2 cells

To investigate how ribavirin modulates F7 transcript elongation, we used cDNA microarray analysis to search for candidate molecules associated with increased FVII mRNA expression in HepG2 cells. Microarray analysis screened 25 394 human genes and found that 842 genes were up-regulated following ribavirin treatment in HepG2 cells. As shown in Table 1, the transcription elongation factors ELL2 gene (ELL2), ELL3 gene (ELL3) and elongin A gene (TCEB3) were found as putative up-regulated genes related to ‘transcription and translation’. We examined the expression levels of these three genes by qRT-PCR and found no significant increase in ELL2 and elongin A mRNA expression, whereas ELL3 mRNA expression was significantly increased in HepG2 cells treated with both ribavirin and MPA.

View this table:
Table 1 Microarray screening for whole-gene expression in HepG2 and HuH7 cells treated with ribavirin and confirmed by qRT-PCR

Analysis of mRNA expression by microarray and qRT-PCR. Microarray data show selected genes whose expression was changed more than 2-fold in cell-to-cell comparisons (cells treated with ribavirin compared with without ribavirin). For reverse transcription–PCR (primers are listed in Supplementary Table S1 at http://www.BiochemJ.org/bj/449/bj4490231add.htm), values were normalized to GusB mRNA and are represented relative to that of untreated HepG2 cells. *P<0.05 compared with untreated cells.

Ribavirin induced ELL3 mRNA expression before up-regulation of FVII mRNA expression

Next, we determined ELL3 expression levels at several time points in ribavirin-treated HepG2 cells. We found an approximate 3-fold increase in ELL3 mRNA expression at 3–12 h following ribavirin treatment, which increased 8-fold at 24 h post-treatment (Figure 4A). This is in contrast with the increased FVII mRNA expression at 24 h after ribavirin treatment (Figure 1B). There were no significant changes in ELL3 and FVII mRNA expression in untreated HepG2 cells. This ELL3 up-regulation was also induced by MPA treatment, and restored by guanosine supplementation at both mRNA and protein levels (Figures 4B and 4C).

Figure 4 Time-course analysis for ELL3 expression induced by ribavirin treatment and knockdown of ELL3 by RNAi

(A) The expression levels of ELL3 and FVII mRNA in HepG2 cells treated with ribavirin at various time points were determined by qRT-PCR. (B) IMPDH inhibitors up-regulated ELL3 expression and supplementation of guanosine retraced the ELL3 up-regulation. (C) ELL3 protein induction by IMPDH inhibitors was cancelled by the addition of guanosine. U, untreated; R, ribavirin-treated; M, MPA-treated. (D) The effects of ELL3 RNAi in HepG2 cells were analysed by Western blotting. The ELL3-specific siRNA (iELL3) or NS siRNA (iNS) was transfected into HepG2 cells at a final concentration of 10 nM and treated with ribavirin (100 μg/ml) for 24 h. The nuclear extracts were isolated from the cells and subjected to Western blotting. Histone H1 was used as a loading control. U, untreated; R, ribavirin-treated. (E) Quantification of ELL3 mRNA in HepG2 cells transfected with iNS or iELL3 (10 nM). After transfection, cells were treated with or without ribavirin (100 μg/ml) and total RNA was isolated and subjected to qRT-PCR. The mRNA levels were expressed as relative values to that of untreated iNS-transfected cells. (F) FVII mRNA levels were examined in siRNA-transfected HepG2 cells treated with or without ribavirin. The expression levels were normalized to GusB mRNA, and shown as relative values to that of untreated cells. *P<0.05 compared with untreated cells or iNS-transfected cells; **P<0.01 compared with untreated cells.

To verify the function of ELL3 in the up-regulation of F7 expression by ribavirin, we performed RNAi experiments on ELL3. We transfected HepG2 cells with ELL3-specific and NS siRNA and analysed ELL3 expression levels by Western blotting and qRT-PCR. We observed that ribavirin-induced ELL3 protein up-regulation was abolished in the ELL3 siRNA-transfected cells (Figure 4D). We also observed diminished expression of ELL3 mRNA levels in the ELL3 siRNA-transfected cells (Figure 4E). Next, we investigated the effects of ELL3 knockdown on F7 expression by qRT-PCR, and observed an attenuated ribavirin-induced up-regulation of FVII mRNA expression in ELL3 siRNA-transfected HepG2 cells, whereas there was no significant difference in FVII mRNA levels between the NS siRNA and ELL3 siRNA-transfected HepG2 cells (Figure 4F).

Ribavirin treatment resulted in modulation of Pol II CTD phosphorylation and enhanced the recruitment of Pol II and elongation factors in HepG2 cells

As shown in Figures 3 and 4, increased FVII mRNA expression was likely to be dependent on transcription elongation associated with P-TEFb and to some extent with ELL3. We performed ChIP analyses to evaluate the recruitment of Pol II to the F7 gene, the status of serine phosphorylation of the Pol II CTD, and the P-TEFb and/or ELL3 contribution to transcription elongation efficacy by Pol II on F7 gene expression (Figures 5A and 5B). Indeed, ribavirin treatment increased Pol II occupancy and the level of phosphorylation of CTD Ser2. Ribavirin treatment did not increase the levels of CTD pSer5, but decreased slightly at some points of F7 gene context. Ribavirin treatment also increased the recruitment of both CDK9 (P-TEFb) and ELL3, which displayed a similar distribution pattern. In addition, we observed ribavirin treatment-enhanced AFF4 recruitment to the FVII, which was occupied by CDK9 and ELL3.

Figure 5 Recruitment of Pol II and transcription elongation factors induced by ribavirin treatment

(A) The genomic locus of human F7 (Gene ID: 2155; sequence derived from GenBank®) with the primer positions we used in the present study. Numbers (transcription start site as +1) indicated the 5′ end as the forward primer. The primers are described in Supplementary Table S2 at http://www.BiochemJ.org/bj/449/bj4490231add.htm. The arrow indicates transcription start site of the F7 gene. The relative quantities of pSer2 and pSer5 were normalized to Pol II levels. Ex, exon. (B) ChIP–qRT-PCR analyses to assess the occupancy of Pol II, pSer2 of CTD, pSer5 of CTD, CDK9, ELL3 and AFF4 on the F7 using specific antibodies. The density was determined by quantitative PCR and is represented as a percentage of input DNA, with values subtracted from that of normal IgG immunoprecipitation. The x-axis indicates the position of quantitative PCR amplification in F7. Values are the mean±S.D. for at least five independent experiments. (C) Western blot analyses for nuclear extracts from HepG2 cells treated with IMPDH inhibitors (ribavirin, MPA) using specific antibodies against Pol II CTD, CTD pSer2, CTD pSer5, CDK9, ELL3 and histone H1. Histone H1 was used as the loading control. Molecular masses are indicated in kDa to the right of the blots.

We assessed the status of pSer residues of Pol II CTD and expression levels of elongation factors in HepG2 cells treated with IMPDH inhibitors by Western blotting (Figure 5C). No change was observed in the total amount of Pol II CTD in both the IMPDH inhibitor-treated and -untreated cells, whereas ribavirin or MPA treatment increased the level of CTD pSer2. The levels of CTD pSer5 were slightly increased. We also checked the expression levels of CDK9 and ELL3 and found that ribavirin or MPA treatment clearly increased the ELL3 protein levels. By contrast, there was no significant change in the expression level of the 42-kDa CDK9 protein, which is a predominant form of CDK9.

DISCUSSION

The major finding of the present study is that intracellular GTP depletion induced by IMPDH inhibitors modulated Factor VII gene transcription, mainly in the elongation phase rather than transcription initiation.

Ribavirin, a synthetic nucleoside analogue, is used in combination with pegylated IFN (interferon)-α as the standard treatment for patients with chronic hepatitis C. A previous study demonstrated that ribavirin treatment increased FVII activity in HCV-infected haemophilia patients as well as increased FVII mRNA expression in both normal human hepatocytes and a human hepatocarcinoma cell line (HepG2) in vitro [8]. In the present study, we investigated further the molecular mechanisms of ribavirin on FVII mRNA up-regulation in HepG2 cells.

We observed that ribavirin treatment increased FVII mRNA expression in HepG2 cells in a dose- and time-dependent manner. Because the ribavirin metabolite RMP inhibits IMPDH, we tested other IMPDH inhibitors (MPA and 6-MP) and found that they also up-regulated FVII mRNA in HepG2 cells. Interestingly, it was reported that ribavirin treatment induced the expression of particular ISGs (IFN-stimulated genes) by activating transcription [13,14], whereas MPA induced the expression of ISGs by promoter activation [13,15]. Thus there may be a close relationship between intracellular GTP depletion caused by sequential inhibition of IMPDH and modulation of various gene expressions. Indeed, guanine nucleoside repletion restored ribavirin-induced FVII mRNA up-regulation.

In general, control of gene expression in eukaryotic cells involves regulatory events at multiple transcriptional and post-transcriptional stages. Transcriptional regulation by Pol II is a multifaceted process, requiring the combined action of a large number of factors for the steadfast synthesis of full-length mRNA [1620]. Transcription by Pol II is divided into four stages: initiation, promoter clearance, elongation and termination. For many years, transcription initiation was considered as the rate-limiting step to the whole transcription process; however, a large number of studies have demonstrated that the elongation stage of transcription, regulated by a number of factors, is essential for productive transcription [17,18,21]. Recently, the elongation phase of transcription has gained increasing importance as numerous positive and negative elongation factors have been identified [18]. We analysed which step of transcription was critical for ribavirin-induced FVII mRNA up-regulation. We found that ribavirin treatment did not affect the FVII mRNA stability. In the luciferase analysis, we observed an approximate 1.5-fold increase in F7 promoter activity by ribavirin treatment, but also found a similar increase of ‘ribavirin-insensitive’ PROS1 promoter activity, suggesting that we might overestimate the effects of ribavirin in our luciferase assay system. By contrast, ribavirin treatment sufficiently increased nascent FVII mRNA synthesis in HepG2 cells, as analysed by the nuclear run-on and nascent mRNA capturing assay, which was abolished by inhibiting CDK9 using DRB. These findings indicate that ribavirin-induced F7 up-regulation is mainly induced by acceleration of transcription elongation, which is associated with P-TEFb, rather than by promoter activation. The marginally enhanced transcription initiation could contribute to some extent in the ribavirin-induced F7 up-regulation, but the enhanced transcription by elongation factors would perform a critical contribution in the up-regulation.

Previous evidence has indicated that elongation is tightly linked to post-transcriptional events such as mRNA capping, splicing, polyadenylation and nuclear export [2224]. The CTD of the largest subunit of Pol II plays a critical role in the integration of these events. The CTD comprises multiple heptad repeats (YSPTSPS motifs, 52 in mammals) each containing two main phosphor-acceptor sites, Ser2 and Ser5. Several CTD kinases have been identified, with CDK7 and CDK9 being the most prominent among them. CDK7 is part of general transcription factor TFIIH and catalyses Ser5 phosphorylation [25]. CDK9 is the catalytic subunit of the P-TEFb, with site specificity to preferentially phosphorylate Ser2 or Ser5 depending on the experimental paradigm used [2630]. CDK9 kinase activity of P-TEFb phosphorylates Ser2 of the CTD and signals the release of the stalled Pol II into productive transcription elongation [31,32].

In the present study, we found that IMPDH inhibitors increased nascent FVII mRNA expression in HepG2 cells, possibly by acceleration of transcription elongation. The increase in FVII mRNA expression was inhibited by DRB treatment, indicating that CDK9 is critical for up-regulation of F7 expression by acceleration of transcription elongation. Consistently, ChIP assays revealed that recruitment of Pol II and CDK9, and phosphorylation of Ser2 of the Pol II CTD were reinforced by ribavirin treatment. To investigate further the elongation steps necessary for increased FVII mRNA expression, we searched ribavirin-responsive genes by cDNA microarray analysis and found that transcription elongation factor ELL3 expression was significantly up-regulated in HepG2 cells treated with ribavirin, and this ELL3 up-regulation was induced in response to intracellular GTP depletion. In addition, we found that ribavirin-induced an increase in ELL3 mRNA expression before an increase in FVII mRNA expression in HepG2 cells, suggesting that induced ELL3 may contribute to FVII mRNA up-regulation.

ELL3, a member of the Pol II elongation factors, is an approximately 400-amino-acid protein with high homology for the C-terminal sequences of ELL and ELL2 [33]. To test the effects of ELL3 on F7 transcription, we performed an RNAi analysis of ELL3 and observed that diminished ELL3 expression by ELL3-specific siRNA and that ribavirin-induced FVII mRNA up-regulation was attenuated by ELL3 knockdown in HepG2 cells. A ChIP assay indicated that ELL3 was bound to F7 in a similar distribution pattern to that of CDK9, suggesting that CDK9 and ELL3 may work together in the transcription elongation phase of ribavirin-induced F7 gene up-regulation.

A previous study reported that ELL proteins are a component of the SEC (super elongation complex) that contains P-TEFb [34]. AFF (AF4/FMR2) proteins are also activators of transcription elongation and interacts CDK9 and ELL [35]. It was concluded that AFF4 was a component of the transcription elongation complex containing P-TEFb and ELL and that AFF4 was required for ELL–P-TEFb complex stability. We found that CDK9, ELL3 and AFF4 were recruited to the F7 gene, as shown by the ChIP assay. A study by Shilatifard and co-workers suggested that these elongation factors form the SEC in the elongation stage [36]. Consistent with previous findings, our findings suggested that ELL3 is involved in SEC with P-TEFb and AFF4 for F7 transcription elongation. We proposed that ELL3 has an important role in ribavirin-induced FVII mRNA up-regulation, even though it may not be essential. It is possible that other molecules could contribute to increased FVII mRNA expression, or the function of ELL3 may be compensable in ribavirin-induced F7 transcription elongation. Our proposed model for the up-regulation of F7 gene transcription is shown in Figure 6.

Figure 6 A proposed model for the acceleration of F7 transcription elongation by ribavirin treatment

IMPDH inhibitors induce the transcription elongation on F7. Intracellular GTP depletion accelerates newly synthesized FVII mRNA by recruiting the SEC, consisting of P-TEFb, AFF4 and ELL3. For a detailed explanation, see the Discussion.

We wonder whether F7 gene expression was specifically modulated by intracellular GTP depletion. In fact, we observed that some other genes of blood coagulation factors were also fairly up-regulated by ribavirin exposure, and now we are investigating the molecular basis of their gene expressions.

In summary, we found that ribavirin increased FVII mRNA expression by acceleration of its transcription elongation in HepG2 cells and that CDK9 kinase activity of P-TEFb has a crucial role in F7 transcription elongation. Although further study is required to determine why intracellular GTP depletion causes accelerated F7 transcription elongation, our findings have contributed to elucidate the precise mechanisms of ribavirin-induced F7 up-regulation. Finally, the present study highlights the possible development of novel pharmaceutical therapies for haemophilia.

AUTHOR CONTRIBUTION

Atsuo Suzuki designed and performed the experiments, and drafted the paper. Yuhri Miyawaki, Eriko Okuyama, Moe Murata, Yumi Ando, Io Kato and Yuki Takagi analysed and interpreted the data and contributed to the analytical methodology. Akira Takagi and Takashi Murate designed the experiments and analysed the data. Hidehiko Saito supervised the project and edited the paper before submission. Tetsuhito Kojima designed the project, analysed data and wrote the paper. All authors were involved in a critical reading of the paper before submission.

FUNDING

A.S. is a research fellow of the Japan Society for the Promotion of Science (JSPS). This study was supported in part by grants-in-aid for JSPS Fellows [grant number 23-5159] (to A.S.)] and by Scientific Research (C) [grant number 22590524 (to T.K.)] from the Japanese Ministry of Education, Culture, Sports, Science and Technology, and for Research on Measures for Intractable Diseases to T.K. from the Japanese Ministry of Health, Labour and Welfare.

Acknowledgments

We thank Dr Nobuaki Suzuki, Dr Shinji Kunishima, Dr Akira Katsumi and Dr Tadashi Matsushita for their helpful discussions.

Abbreviations: 6-MP, 6-mercaptopurine; CDK, cyclin-dependent kinase; ChIP, chromatin immunoprecipitation; CTD, C-terminal domain; DMEM, Dulbecco’s modified Eagle’s medium; DRB, 5,6-dichlorobenzimidazole 1-β-D-ribofuranoside; ELL, eleven-nineteen lysine-rich leukaemia; EU, 5-ethynyl uridine; FBS, fetal bovine serum; FVII, Factor VII; HCV, hepatitis C virus; IFN, interferon; IMPDH, inosine-5′-monophosphate dehydrogenase; ISG, IFN-stimulated gene; MPA, mycophenolic acid; NS, non-specific; Pol, II, RNA polymerase II; P-TEFb, positive-transcription elongation factor b; pSer, phosphorylated serine; qRT-PCR, quantitative real-time PCR; rFVIIa, recombinant FVIIa; RMP, ribavirin monophosphate; RNAi, RNA interference; SEC, super elongation complex; siRNA, small interfering RNA; TF, tissue factor

References

View Abstract