Both authors contributed equally to this work
Hairpin ribozymes in combination with siRNAs against highly conserved hepatitis C virus sequence inhibit RNA replication and protein translation from hepatitis C virus subgenomic replicons
Article first published online: 31 OCT 2005
Volume 272, Issue 22, pages 5910–5922, November 2005
How to Cite
Jarczak, D., Korf, M., Beger, C., Manns, M. P. and Krüger, M. (2005), Hairpin ribozymes in combination with siRNAs against highly conserved hepatitis C virus sequence inhibit RNA replication and protein translation from hepatitis C virus subgenomic replicons. FEBS Journal, 272: 5910–5922. doi: 10.1111/j.1742-4658.2005.04986.x
- Issue published online: 8 NOV 2005
- Article first published online: 31 OCT 2005
- (Received 17 May 2005, revised 5 September 2005, accepted 22 September 2005)
- constitutive transport element;
- hairpin ribozyme;
- hepatitis C virus;
- small interfering RNA;
- subgenomic replicon
Chronic hepatitis C virus (HCV) infection is a clinically important liver disease with limited therapeutic options in a significant proportion of patients. Therefore, novel efficient therapeutic agents are needed. Because the 5′- and 3′-untranslated regions (UTRs) of HCV are highly conserved and functionally important for HCV replication, they are attractive targets for RNA-cleaving ribozymes or small interfering RNAs (siRNAs). In this study hairpin ribozymes (Rz) targeting HCV 5′- and 3′-UTR sequences were expressed from a retroviral vector transcript under control of two different RNA polIII promoters (tRNAVal, U6). Ribozymes were evaluated in monocistronic, subgenomic I389/hyg-ubi/NS3-3′/5.1 HCV replicon cells as single agents or in combination with siRNAs against HCV 5′- or 3′-UTR recently demonstrated to inhibit HCV replicons. Additionally, ribozyme constructs were generated with the 3′-terminus of the ribozyme flanked by constitutive transport element (CTE) sequences, an RNA motif that has previously been shown to enhance cleavage activity of hammerhead ribozymes. In our study, tRNAVal as well as U6 promoter-driven Rzs markedly reduced HCV replicon RNA expression and HCV internal ribosome entry site (IRES)-mediated HCV NS5B protein translation from monocistronic subgenomic replicons. However, attachment of CTE sequences to the 3′-terminus did not significantly enhance activity of Rzs tested in this study. Interestingly, we detected additive HCV inhibitory effects for combinations of tRNAVal-driven Rzs and U6-derived siRNAs both directed against highly conserved 5′- and 3′-UTR sequence, suggesting that a dual strategy of ribozymes and siRNAs might become a powerful molecular tool to specifically silence HCV RNA replication.
constitutive transport element
hepatitis C virus
internal ribosome entry site
open reading frame
small interfering RNA
truncated human CD4
tRNAval promoter-driven ribozyme
U6 promoter-driven ribozyme
Chronic hepatitis C virus (HCV) infection is a major cause of human liver disease as, in most cases, persistent infection is observed with variable clinical outcomes including chronic hepatitis, cirrhosis and hepatocellular carcinoma . Despite significant therapeutic improvement with pegylated interferon-α and ribavirin combination therapy, effective antiviral agents are urgently needed to treat nonresponders or relapsers to standard treatment regimen .
HCV, a member of the Flaviviridae family of viruses, is an enveloped virus that includes a capsid containing the positive-stranded 9.6 kb RNA genome. HCV has a long open reading frame (ORF) which is flanked by the 5′- and 3′-untranslated regions (UTRs). The 5′-UTR functions as an internal ribosome entry site (IRES) forming a complex secondary structure which allows binding of cellular proteins to support translation of the HCV genome. The HCV ORF encodes a single precursor polyprotein of about 3000 amino acids that is processed by host signal peptidases and HCV viral protease functions into the structural proteins core, envelope proteins 1 and 2 (E1/E2) and nonstructural proteins NS2, NS3, NS4A, NS4B, NS5A and NS5B.
Because HCV RNA replicates in the cytoplasm of liver cells without integration into the genome, RNA-directed antiviral agents such as ribozymes or siRNAs are likely to successfully inhibit the HCV replication cycle. Ribozymes are small RNA molecules with endonuclease activity that can be engineered to specifically cleave complementary RNA sequences. Hairpin ribozymes (Rz) consists of four helical domains and five loops. The cleavage site (GUC) is flanked by the two substrate binding sequences formed between target RNA and ribozyme, allowing the design of trans-acting ribozymes for specific cleavage of a sequence of interest. Since the HCV RNA-dependent RNA polymerase NS5B lacks significant proofreading activity, viral replication is accompanied by the occurrence of mutations so that escape mutants to therapeutic strategies specific to these sequences might arise. In contrast to the observed variation in the coding sequence, HCV 5′- and 3′-UTRs are highly conserved among all HCV genotypes. Additionally, these regions are essential for translation and replication of the virus, making them an attractive molecular target. Therefore, in this study, we focused our interest on the HCV 5′- and 3′-UTRs and developed hairpin ribozymes under control of different promoters (tRNAVal, U6) to target theses sequences. The 3′-terminus of the ribozyme was flanked by constitutive transport element (CTE) sequences, an RNA motif that has previously been shown to interact with intracellular RNA helicases and thereby might enhance cleavage activity by RNA binding and unwinding activities of RNA helicase. CTE has been shown to facilitate nuclear export [3–7], thereby potentially increasing cytoplasmic ribozyme abundance and colocalization with HCV RNA [8,9]. The efficacy of ribozymes and the effect of 3′-CTE sequence was evaluated after endogenous expression in Huh-7 cells stably expressing monocistronic, subgenomic I389/hyg-ubi/NS3-3′/5.1 HCV replicons (Huh-mono cells) . We detected a significant reduction of HCV RNA and NS5B protein levels induced by active ribozymes compared with controls, an activity that, in contrast to studies previously published for hammerhead ribozymes [11–13], was not improved by attachment of CTE-sequence to the ribozyme 3′-terminus. Interestingly, we detected additive HCV inhibitory effects for combinations of tRNAVal-driven Rzs and U6-derived siRNAs both directed against highly conserved 5′- and 3′-UTR sequence, suggesting the potential for a successful dual antiviral strategy.
Construction of DNA-based retroviral vectors expressing CTE-linked hairpin ribozymes against the HCV 5′-UTR (IRES) and 3′-UTR
Several target sequences containing the GUC recognition triplet required for hairpin ribozyme cleavage were selected within the highly conserved regions of the 5′- (position 323) and 3′-untranslated regions of HCV (positions 9523 and 9543 of the positive strand; position 40 of the negative strand). The structure of the HCV monocistronic, subgenomic HCV replicon I389/hyg-ubi/NS3-3′/5.1 and positions of HCV-directed ribozymes are schematically depicted in Fig. 1A (for details on the target sequences see Table 1). Ribozyme sequences were cloned into the retroviral vector pLHPM (schematically depicted in Figs 1B; plasmid designated tRz), previously shown to allow for high copy cytoplasmic expression of hairpin ribozymes driven by a human tRNAVal promoter located on the 5′-terminus of the ribozyme [17–20]. Additionally, ribozymes were constructed with inserted constitutive transport element (CTE) sequences at the 3′-terminus of the ribozyme (tRz-CTE; Fig. 1B). A ribozyme directed against the genome of hepatitis B virus (tRzHBV) served as the control. Additionally, we reversed the orientation of the CTE sequence element in ribozymes tRz323 and tRz3′X (tRz323-CTEas, tRz3′X-CTEas) to control for specific effects of the sense orientation of CTE. Finally, ribozyme-CTE cassettes were inserted into a second retroviral vector (pSuppressorRetro vector) flanked by a U6 promoter at the ribozyme 5′-terminus (designated as uRz; Fig. 1C) to compare with the tRNAVal-containing constructs.
|Ribozyme||Target RNA||Position within target RNA (nt)||Target sequence (5′ to 3′)|
|Rz9523||HCV 3′-UTR (+)-strand||9523||CCCUA^GUCACGGCUAG|
|Rz9543||HCV 3′-UTR (+)-strand||9543||GAAAG^GUCCGUGAGCC|
|Rz3′X||HCV 3′-UTR (–)-strand||40||CUGCA^GUCAUGCGGCU|
Expression and cytoplasmic localization of DNA-based hairpin ribozymes in Huh-7 cells
Huh-7 cells were harvested 5 days after transfection of hairpin ribozyme vectors and the cytosol was separated from nuclei. Following RNA isolation and RT-PCR, PCR using specific oligonucleotides was performed on cytoplasmic extracts to detect expression of ribozymes with and without CTE. PCR on ribozyme-plasmid DNA served as positive controls. PCR with ribozyme-specific primers revealed cytoplasmic localization of ribozymes independent of CTE (Fig. 2A). Successful separation of nuclear and cytoplasmic fractions was demonstrated by PCR amplification of U6 small nuclear RNA which is known for its nuclear localization (Fig. 2B).
Hairpin ribozymes directed against HCV 5′-UTR (IRES) and 3′-UTR sequence inhibit luciferase reporter activity independent from CTE sequence
To evaluate cleavage activity of hairpin ribozymes on intracellular 5′- and 3′-UTR sequence as well as inhibitory effects on HCV IRES-mediated translation, we cotransfected Huh-7 cells with bicistronic reporter plasmid pRL-5′-FL-3′-UTR (Renilla luciferase-HCV 5′-UTR (HCV IRES)-firefly luciferase- HCV 3′-UTR), β-galactosidase and ribozyme expressing plasmids. Luciferase expression levels measured 72 h after transfection demonstrated the strongest inhibitory effect for tRz9543 with reductions in HCV IRES-dependent firefly luciferase levels of up to 50%, while tRz9523 and tRz3′X consistently demonstrated moderate inhibition of firefly luciferase in the range of 20%(Fig. 3A). tRz9523, tRz9543 and tRz3′X activities remained unchanged independent from CTE sequence flanking the ribozyme 3′-end (Fig. 3B) while reversing the orientation of the CTE sequence did not change the activity of ribozymes in this study (data not shown). No significant inhibition of IRES-dependent luciferase activity was found for tRz323 compared with control ribozyme transfected cells and CTE did not improve inhibitory activity in this assay. Results obtained from the U6 promoter-driven ribozymes (uRz) paralleled the observations from tRNAVal-driven Rzs indicating an IRES-inhibitory activity for constructs uRz9523, uRz9543 and uRz3′X, but no significant activity for uRz323 (data not shown).
DNA-vector-based ribozymes inhibit HCV replicon RNA in cultured Huh-7 cells without enhancing effects of CTE sequence
First, we examined tRNAVal-driven ribozymes without CTE sequences for intracellular effects in Huh-mono cells stably expressing monocistronic, subgenomic I389/hyg-ubi/NS3-3′/5.1 HCV replicons (replicon schematically depicted in Fig 1A). Huh-mono cells were transfected with control ribozyme or HCV-targeting ribozyme vectors together with pMACS4.1, a plasmid encoding a truncated cell surface marker (tCD4) suitable for selection of transfected cells by magnetic separation (MACSelect, details see Experimental procedures). Using this transfected cell selection system, transfection efficiencies of about 90% were obtained (data not shown). Cells were harvested five days after transfection and HCV replication levels were determined by northern blotting analysis. Ribozyme tRz9543 markedly reduced HCV RNA levels by 36%(Fig. 4A), while in this assay, tRz3′X (targeting HCV negative strand) consistently inhibited HCV replication by 31%. Both ribozymes did not reveal improvement in their activity by attachment of CTE sequence (Fig. 4B). In contrast, a 15–20% enhancement of activity was noted for tRz323 and tRz9523 with 3′-CTE sequence. Again, no differences in activity were detected for U6 promoter-driven ribozymes compared with the tRNA constructs. Again, reversing the orientation of the CTE sequence did not change the inhibitory activity on HCV replicon RNA (data not shown).
HCV NS5B protein translation is inhibited by ribozymes targeting HCV 5′- and 3′-untranslated regions independent from CTE sequence
HCV protein expression levels of Huh-mono cells after ribozyme transfection and magnetic cell selection were examined by western blotting analysis. In parallel to the inhibitory effects on RNA levels, a marked reduction of HCV NS5B protein was detected for tRz3′X with NS5B protein levels inhibited by 35%, respectively (Fig. 5A). For tRz323, tRz9532 and tRz9543, reductions in NS5B protein levels in the range of 25% were detected. These reductions were detected for all variants of these ribozymes [± CTE (Fig. 5B); U6 promoter]. From these data we conclude, that the HCV-directed ribozymes Rz323, Rz3′X, Rz9523 and Rz9543 enfold moderate, but specific inhibitory activity on HCV replicons. Unfortunately, in our studies, this effect was not enhanced significantly by attachment of CTE sequence at the 3′-terminus of the transcript.
Combinations of CTE-linked hairpin ribozymes under expression of U6 and tRNAVal promoter enhance the inhibitory effect on HCV replication cycle
We then examined whether inhibitory effects of single ribozymes Rz323 (5′-UTR-directed) and Rz9543 (3′-UTR-directed) on HCV replication cycle can be enhanced when combined. Combinations of U6 promoter-driven and/or tRNAVal promoter-driven ribozymes with attached CTE were transfected into Huh-mono cells and selected after 5 days by magnetic cell selection procedure. Analysis of HCV replicon RNA expression in northern blotting showed an additional reduction of 10% by ribozyme combinations (U6 or tRNAVal) compared with single ribozymes. The combination of Rz9543-CTE and Rz323-CTE expressed from different promoters revealed an enhanced inhibitory effect on HCV replicon RNA expression in the range of 15% in comparison with ribozyme combinations driven exclusively by U6 or tRNAVal promoter (Fig. 6). Analysis of HCV NS5B protein level in western blotting experiments showed an enhancement of the inhibitory effect on HCV up to 20% by the combination of uRz323-CTE and tRz9543-CTE expressed from different promoters (Fig. 6).
Combinations of CTE-linked hairpin ribozymes with siRNAs against HCV 5′- and 3′-UTR inhibit HCV NS5B protein translation in monocistronic replicon cells
Recently, we have successfully used siRNAs to target highly conserved HCV 5′- and 3′-UTR sequences and demonstrated inhibitory activity on HCV RNA and protein levels in monocistronic replicon expressing cells . Here, we investigated a dual strategy of ribozyme and siRNA simultaneously targeting distinct highly conserved HCV RNA sequences. Huh-mono cells were cotransfected with U6 promoter-driven siRNAs and tRNAVal-driven hairpin ribozymes with attached CTE (tRz-CTE). Following magnetic cell selection procedure, HCV NS5B protein levels were analyzed by western blotting. Again, tRz9543 induced a 30% decrease in NS5B protein levels. siRNAs driven from a U6 promoter reduced HCV NS5B expression by approximately 41% for siHCV-156, 53% for siHCV-207, 47% for siHCV-3′X, respectively. Surprisingly, when U6 promoter-driven siRNAs were combined with tRNAVal-driven hairpin ribozyme tRz9543-CTE, a consistent inhibitory effect of the ribozyme in the range of 25% was added to the inhibitory effect of each siRNA (Fig. 7). Concordant results were obtained from northern blot analyses of HCV replicon RNA (data not shown). From these data, we conclude that selected combinations of HCV-directed ribozymes with siRNAs targeting HCV UTR sequences provide additional antiviral effects on subgenomic HCV replicons.
Plasmid-based hairpin ribozymes and siRNAs do not trigger interferon production
To determine whether inhibitory effects on HCV replication cycle are based on the efficacy of hairpin ribozymes or siRNAs and are not caused by unspecific intracellular effects, we investigated interferon expression in cells following siRNA– and ribozyme–plasmid transfection. First, the Jak/STAT-pathway was tested for activation of the interferon-induced transcriptional factor STAT 1 in western blotting analysis. These experiments confirmed an enhanced STAT 1 level in cells transfected with plasmids expressing interferon regulatory factor 1 (IRF1) and 3 (IRF3) as positive controls. Additionally, interferon RNA levels were determined by northern blotting analysis following plasmid transfection of Huh-7 cells. Expectedly, cells transfected with interferon-inducing IRF1 and IRF3 vectors revealed enhanced interferon levels compared with background levels detected for ribozymes, siRNAs and the controls (Fig. 8).
Ribozymes have been demonstrated to enzymatically cleave and destroy RNA transcripts in a sequence-specific fashion in a variety of diseases and human disorders (recently reviewed in [21–25]). Hairpin ribozymes fold into a two-dimensional ‘hairpin’ structure, with the ribozyme binding arms hybridizing to sequences flanking the cleavage site (GUC) within the target RNA, thereby determining specificity of the recognized target sequence. In this study, we have demonstrated that hairpin ribozymes directed against highly conserved regions of HCV decrease subgenomic HCV RNA expression levels as well as protein translation in monocistronic replicon expressing cells. We evaluated four different ribozymes and found the greatest inhibitory activity for Rz9543 directed against target sequence flanking a GUC-site within the HCV 3′-UTR. As the highly conserved HCV UTRs fold to characteristic secondary structures [26,27] and cellular proteins bind to these functionally important regions, we anticipated this to prevent ribozymes from efficiently binding to the substrate RNA and enfolding optimal cleavage activity. However, with the tRNAVal promoter cassette allowing for high copy intracellular expression levels of the ribozyme and efficient cytoplasmic export of ribozymes, a considerable HCV inhibitory activity in the replicon expressing cells was detected on RNA and protein level. These results were confirmed for the U6 promoter-driven ribozymes suggesting that ribozyme activity is independent of the tRNAVal molecule flanking the ribozyme on the 5′-end . There might be additional factors explaining why we failed to induce higher inhibition of HCV target RNA expression: (a) the ratio of DNA chosen in our experiments (ribozyme–pMACS = 3 : 1) might not allow each tCD4-transfected cell to contain functionally active ribozyme; (b) magnetic cell selection procedure does not allow complete separation of transfected from nontransfected replicon expressing cells; (c) selection pressure added by presence of the ribozyme might lead to rapid development of escape mutants; (d) ribozyme position and length of binding arms might not be optimal; or (e) the kinetically unfavorable repetitive association and disassociation events of ribozymes and their target RNA could have limitations for highly structured HCV RNA.
Transfection of Huh-7 cells with plasmids may lead to the production of interferon by activation of Toll-like-receptor-pathway . In our experiments, we did not detect interferon activation following plasmid-mediated ribozyme or siRNA expression. In addition, as indicated by the inactive status of the transcriptional factor Stat1 (signal transducer and activator of transcription), the Jak/STAT-pathway is not activated. These data suggest that the effects observed in our studies are based on specific cleavage and not on nonspecific induction of interferon pathway.
Experiments with the HCV IRES-dependent bicistronic luciferase reporter system revealed no significant activity for ribozyme Rz3′X (directed against sequence position 40 within the 3′-UTR of the HCV negative strand RNA; Fig. 1A) compared with control ribozyme RzHBV-transfected cells. This was not surprising, since the cotransfected reporter transcript does not contain Rz3′X target sequence. In contrast, Rz3′X was found to be active in the replicon expression system (with HCV strands of both polarities), which might be explained by effects of Rz 3′X on expression of the proteasome α-subunit PSMA7, a 28 kDa protein of the 20S proteasome, found to modulate HCV IRES activity in cell culture . The incubation time of 72 h for luciferase assays – compared with 120 h for subgenomic replicon assays – might not be sufficient to detect a substantial effect of Rz3′X on IRES-mediated translation via down-regulation of PSMA7. Most likely, the half-life of cellular PSMA7 and the subsequent inhibition of HCV NS5B translation following Rz3′X-mediated depletion of PSMA7 requires prolonged incubation.
While several studies have shown that the activity of hammerhead ribozymes can be increased by attachment of the cis-acting CTE of Mason–Pfizer monkey virus [11,13], experience with hairpin ribozymes have not been published. Attachment of the CTE sequence to a tRNAVal-driven hammerhead ribozyme in cis allowed for efficient cleavage of sites that had previously been refractory because of local folding . Additionally, in these studies, the presence of the CTE also improved cleavage activity for hammerhead ribozymes that were already functionally active. The authors found this effect to be dependent on the presence of wild-type CTE in cis and attributed their observations to an interaction of CTE with RNA helicases which have RNA binding, sliding and unwinding activities [11–13] and not due to change in stability, expression or transport of the ribozyme. From their observations, the authors concluded that the use of CTE sequence could be exploited to overcome the problem of restricted target accessibility and should consequently enhance the efficacy of ribozymes. In our studies we delivered hairpin ribozymes by DNA-retroviral vector transfection and targeted highly conserved sequences within the HCV untranslated regions of subgenomic HCV replicons that sustain efficient replication of HCV in cell culture. However, we failed to identify a significant improvement of ribozyme activity by CTE sequence added in cis at the 3′-terminus of the ribozyme as measured by HCV IRES-mediated translation, replicon RNA or NS5B protein levels. For our initial experiments we used the strong human tRNAVal promoter, which has been demonstrated to allow efficient transport of ribozymes from the nucleus to the cytoplasm and a high number of ribozyme transcripts per cell [17–20], certainly a prerequisite for cytoplasmic colocalization of ribozymes and HCV RNA and successful cleavage. Because we failed to identify an enhancement of ribozyme activity by CTE in the tRNAVal-driven constructs (pLHPM), we also evaluated ribozymes with 3′-CTE sequence under the control of a U6 promoter, and again found no enhancing effects of CTE; at least, ribozyme activity was maintained. Several explanations might be considered for the similar levels of HCV knock-down observed for hairpin ribozymes with or without CTE: (a) the length and specific folding of HCV RNA, particularly of the UTR regions, make ribozyme–substrate–helicase interaction difficult; (b) secondary structures of HCV can not be sufficiently unwound by RNA helicases; (c) binding of translational HCV cofactors prevent both, RNA helicase and ribozymes from accessing the HCV UTRs; (d) because expression analyses were performed 4 days after cotransfection, initial differences in ribozyme activity might not be detected in this experimental setting; and (e) the monocistronic replicon cells used in this study might lack an additional factor required for CTE-helicase–target RNA interaction.
In an attempt to improve the efficacy of ribozyme-mediated cleavage, ribozymes were expressed from different promoters and combinations directed against different targets were investigated. Targeting HCV 5′-UTR and HCV 3′-UTR simultaneously slightly increased the inhibitory effect of single ribozymes against these targets. Interestingly, combinations of U6- together with tRNAVal-promoter-driven ribozymes revealed additional inhibitory benefit compared with U6- or tRNAVal-promoter-driven ribozyme-combinations. The use of different promoters to drive ribozyme expression might avoid the limitation of cellular transcriptional factors.
Furthermore we intended to enhance antiviral cleavage activity by combining ribozymes with RNA interference mediated by siRNAs. Our studies on combination of siRNAs with ribozymes revealed consistent improvement of siRNA inhibitory activity by ribozyme tRz9543 in the range of 25% independent of the level of inhibition already induced by the different siRNAs. These data underline that a dual strategy of tRNAVal-driven hairpin ribozymes combined with U6 promoter-derived siRNAs might be developed to specifically silence HCV RNA replication. Our current experiments are aimed at developing optimized siRNAs directed against HCV RNA and putative HCV cofactors combined with active hairpin ribozymes targeting highly conserved HCV sequence elements. In addition, experiments are undertaken to develop a single vector system for simultaneous high level expression of ribozymes and siRNAs.
In summary, we developed hairpin ribozymes active against highly conserved sequences within HCV 5′- and 3′-UTR and thereby inhibited HCV IRES-mediated translation and subgenomic RNA replication, independent from CTE sequence added to the 3′- terminus of the ribozyme. Cleavage activity of hairpin ribozymes combined with siRNA molecules directed against HCV RNA or putative HCV cofactors represents a promising dual strategy to silence HCV replication.
Construction of DNA-based retroviral vectors expressing hairpin ribozymes under the control of a tRNAVal-promoter
Several hairpin ribozyme target sites were selected within the highly conserved regions of the 5′- (RzHCV-323) and 3′-untranslated region of HCV (RzHCV-9523; RzHCV-9543) (for sequences, see Table 1; numbering according to hepatitis C virus strain H77; GenBank accession number AF011751). Additionally, RzHCV-3′X was designed to target the negative strand intermediate within the 3′-UTR (Fig. 1C). Hairpin ribozymes RzBR1 (directed against human hepatitis B virus)  and HCV-directed ribozymes were constructed by annealing of overlapping ribozyme-specific oligonucleotides (MWG-Biotech AG, Ebersberg, Germany) and cloned into BamHI–MluI sites of the Moloney retroviral genome based ribozyme expression vector pLHPM (containing a ribozyme expression cassette driven by the tRNAval promoter, Fig. 1B) as described previously [14,15]. The constitutive transport element (CTE) was taken from vector psp-MPMV8240 by PCR with forward primer 5′-MluI (5′-tgctacgcgtcacctcccctgtgagctag-3′) and reverse primer 3′-MluI (5′-gcctacgcgtccaagacatcatccgggc-3′) thereby generating flanking MluI-restriction sites. Digested fragments were cloned into the MluI site flanking the ribozyme 3′-end in pLHPM. Sequences were confirmed by DNA sequence analysis (MWG Biotech).
Construction of ribozymes driven by a U6 promoter
To express ribozymes under control of the human U6 promoter, ribozyme cassettes from pLHPM vectors were amplified by PCR using forward primers containing a SalI-site (underlined): 5′-gaaacagtcgacaggttgggagaagc-3′ (RzBR1), 5′-gaaacagtcgacagccgcatagaagc-3′ (RzHCV-3′X), 5′-gaaacagtcgacgtctacgaagaatc-3′ (RzHCV-323), 5′-tatagtgtcgacctagccgtagaaag-3′ (RzHCV-9523), 5′-gaaacagtcgacggctcacgagaat-3′ (RzHCV-9543). For amplification of cassettes without CTE, oligonucleotides containing an XbaI-site (underlined) were used as reverse primers: 5′-gaggtgtctagaaccaggtaatatac-3′ (3′-RzUni). For amplification of CTE oligonucleotides 3′-CTE were used (5′-gcaaaatctagaccaagacatcatc-3′). SalI- and XbaI-digested PCR products were ligated to the SalI- and XbaI-digested retroviral vector pSupressorRetro (Imgenex, San Diego, CA, USA) downstream of the U6 promoter (Fig. 1B). Correct sequences were confirmed by DNA sequence analysis (MWG Biotech).
Bicistronic luciferase reporter plasmid construction
Bicistronic reporter plasmid pRL-IRES-FL-3′-UTR (Renilla luciferase-HCV 5′-UTR-firefly luciferase-HCV 3′-UTR) was constructed as described previously .
The human hepatoma cell line Huh-7 were maintained in Dulbecco's modified Eagle's medium (DMEM), supplemented with 10% (v/v) heat-inactivated fetal bovine serum, 100 units·mL−1 penicillin/streptomycin, 1 mm sodium pyruvate, 1 mm nonessential amino acids (all Invitrogen, Karlsruhe, Germany) and cultured at 37 °C with 5% (v/v) CO2. Cells harboring monocistronic, subgenomic HCV replicons I389/hyg-ubi/NS3-3′/5.1 (Huh-mono) were generously provided by R. Bartenschlager (University of Heidelberg, Germany) . This monocistronic, subgenomic replicon is composed of HCV 5′-UTR, nucleotides 342–389 of core-encoding sequence, the hyg gene (encoding the hygromycin phosphotransferase), the ubiquitin-endcoding sequence (ubi), the HCV nonstructural proteins NS3 to NS5B and the HCV 3′-UTR. For Huh-mono cells culture media were additionally supplemented with 25 μg·mL−1 of hygromycin B (Sigma-Aldrich, Taufkirchen, Germany).
Transfection and luciferase assays
Huh-7 cells were transiently transfected with 0.03 µg·cm−2 of bicistronic reporter plasmid DNA, 0.03 µg·cm−2 of β-galactosidase plasmid (CMV β-Gal, Clontech, Heidelberg, Germany) and 0.24 µg·cm−2 of ribozyme-expressing plasmid DNA using Lipofectamine 2000 reagent (Invitrogen, Karlsruhe, Germany) in accordance with the manufacturer's instructions. Similarly, transfections were performed for EGFP expression analyses, except that pHygEGFP (Clontech) was used instead of bicistronic reporter. Luciferase activities were quantified using a luminometer (Lumat LB9501, Berthold, Wildbad, Germany) in cell lysates obtained 72 h after transfection. Levels of cotransfected β-galactosidase were quantified at 405 nm wavelength using a Lambda Scan 200e photometer (MWG-Biotech). Ratios of signal intensities for firefly luciferase and β-galactosidase present in control plasmid transfected cells were set to 100%. Assays were performed in triplicate, repeated at least three times, and the results were expressed as mean ± SD as percentages of the controls.
Transfected cell selection
Huh-mono replicon cells (plating density: 2.5 × 104 cells·cm−2) were transiently cotransfected with 0.1 µg·cm−2 of plasmid DNA expressing truncated CD4 (pMACS 4.1, Miltenyi Biotech, Bergisch Gladbach, Germany) and 0.3 µg·cm−2 of ribozyme-expressing plasmid DNA using Lipofectamine 2000 reagent (Invitrogen, Karlsruhe, Germany). For ribozyme or ribozyme-siRNA combination experiments, the total amount of plasmid DNA was increased to 0.6 µg·cm−2 with single ribozymes (or siRNAs) supplemented with control plasmid DNA at a 1 : 1 ratio. Transfected cell selection was performed according to the manufacturer's instructions (MACSelect, Miltenyi Biotech) 5 days after transfection. Selected cells were harvested and processed for western and northern blotting analyses.
Northern blotting analysis
Northern blotting analyses were performed as described previously . Probes for HCV, interferon-α and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were generated by PCR (primer sequences are available upon request) and labeled with [32P]dCTP[αP] by random priming method (high prime DNA labeling kit, Boehringer Mannheim, Mannheim, Germany). Quantitation was achieved by phosphorimager analysis and computer-assisted densitometry (Fujifilm bas-1000, PCBAS-Software). RNA transcript levels were normalized to GAPDH mRNA and values are expressed as percentages compared with the levels for controls (tRzHBV, tRzHBV-CTE, tRzHBV-CTEas, uRzHBV or uRzHBV-CTE).
Separation of nuclear and cytoplasmatic fractions of Huh-7 cells
Huh-7 cells were harvested 48 h after transfection in NaCl/Pi and centrifuged at 4 °C/1000 g for 5min. The pellet was resuspended in 200 µL NPBT-buffer [10 mm Tris/HCl (pH 7.4), 2 mm MgCl2, 140 mm NaCl, 1 mm Na3VO4, 2 mg·mL−1 aprotinin/0.5 mm dithiothreitol, 0.5mm phenylmethanesulfonyl fluoride, 0.5% (v/v) Triton X-100], loaded on 50% sucrose and centrifuged for 10 min at 4 °C/20 000 g. After removal of the supernatant, the nuclear pellet was resuspended in 100 µL buffer D [20 mm Hepes, pH 7.8, 25% (v/v) glycerol, 420 mm NaCl, 1.5 mm MgCl2, 0.2 mm EDTA, 1 mm dithiothreitol, 0.5 mm phenylmethanesulfonyl fluoride] and incubated for 30 min at 4 °C. Following centrifugation at 4 °C/20 000 g, the pellet was resuspended for western blotting and northern blotting analyses.
Western blotting analysis
Western blot analysis was performed on selected and harvested cells from MACSelect procedure as described previously . Briefly, membranes were incubated with monoclonal antibodies to HCV NS5B (kindly provided by D. Moradpour, University of Freiburg, Germany), STAT1 (Santa Cruz Biotechnology, Santa Cruz, CA, USA) and β-actin (clone AC-40, Sigma-Aldrich, Taufkirchen, Germany) for 3 h. Following incubation with a secondary antimouse IgG antibody conjugated to alkaline phosphatase, membranes were developed using a chemiluminescent substrate (WesternBreeze, kit, Invitrogen) in accordance with the manufacturer's instructions. Blots were exposed to film for 10 s, 30 s or 1 min. Band intensities were measured by using National Institutes of Health (NIH) image software and protein signals were normalized to β-actin levels. Experiments were repeated at least three times.
We thank Ralf Bartenschlager (University of Heidelberg, Germany) for generously providing Huh-mono cells harboring monocistronic, subgenomic HCV replicons I389/hyg-ubi/NS3-3′/5.1. Monoclonal HCV NS5B antibody was kindly provided by Darius Moradpour (University of Freiburg, Germany). This work was supported by grant KR 1706/2–1 from the Deutsche Forschungsgemeinschaft, Bonn, Germany and by Else Kröner-Fresenius-Stiftung, Germany.
- 22001) Current and future treatment of hepatitis C. Indian J Gastroenterol 20 (Suppl. 1), C47–C51., & (
- 211998) Ribozymes in cancer gene therapy. Gene Therapy of Cancer (LattimeEC & Gerson, SJ, eds), pp. 139–152. Academic Press, San Diego, CA., & (