Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Cell Host Microbe. Author manuscript; available in PMC 2012 December 11.
Published in final edited form as:
PMCID: PMC3519368

Position-dependent Function for a Tandem MicroRNA miR-122 Binding Site Located in the Hepatitis C Virus RNA Genome


MicroRNAs usually interact with 3’ noncoding regions of target mRNAs leading to downregulation of mRNA expression. In contrast, liver-specific microRNA miR-122 is known to interact with the 5’ end of the hepatitis C virus RNA genome, resulting in increased viral RNA abundance. We found that the location of the viral miR-122 binding site dictates its effect on gene regulation, because insertion of the site into the 3’ noncoding region of a reporter mRNA lead to downregulation of mRNA expression. Furthermore, we discovered an adjacent, second miR-122 binding site, separated from the first by a highly conserved fourteen-nucleotide sequence. Results obtained with mutated viral genomes argue that both sites are occupied in the same molecule and cooperatively regulate target gene expression. These findings set a paradigm for dual functions of tandem microRNA binding sites in a position-dependent manner, and offer a potential antiviral intervention by targeting an oligomeric microRNA complex.


MicroRNAs are 21–22 nucleotide RNA molecules that are expressed in a wide range of eukaryotic organisms (Bartel, 2004; Grimson et al., 2007). Several hundred different human microRNA genes have been identified which show considerable specificity in both tissue type and developmental stage expression. Importantly, it has been estimated that each microRNA can interact with 100 target mRNAs (Brennecke et al., 2005), suggesting that a significant fraction of mRNAs are likely to be regulated by microRNAs (Lewis et al., 2005; Xie et al., 2005). Animal microRNAs are thought to function by binding to imperfectly complementary sites in the 3’ noncoding regions (3’NCRs) of target mRNAs and thereby repress mRNA expression. The mechanism by which microRNAs inhibit target mRNA expression is controversial and likely involves modulation of distinct steps in translation and mRNA turnover (Jackson and Standart, 2007; Nilsen, 2007).

Hepatitis C virus (HCV) is a hepatotrophic, positive-sense RNA virus, containing a 9.6kb genome that establishes persistent infections, leading to chronic liver disease (Hoofnagle, 2002). We recently demonstrated that a liver-specific microRNA, miR-122 (Chang et al., 2004), interacts with sequences in the 5’NCR of HCV RNA, and that this interaction is required to maintain high viral RNA abundance in cultured liver cells (Jopling et al., 2005). The function of miR-122 is absolutely dependent on an operational microRNA processing pathway, because siRNA-mediated knock down of Drosha, DGCR8, Dicer or each of the four Argonaute proteins greatly diminished the accumulation of viral RNA in HCV-infected cells (Randall et al., 2007). Interaction of miR-122 with the viral genome did not affect rates of translation and RNA turnover, arguing that rates of viral replication are augmented directly (Jopling et al., 2005). This unprecedented mechanism of upregulation of a target mRNA by a microRNA led us to examine more closely whether the HCV-miR-122 complex is specialized, or whether miR-122 could also mediate upregulation of a different mRNA. We found that interactions of miR-122 with the binding site from HCV placed in the 3’NCRs of reporter mRNAs led to downregulation of mRNA expression, arguing that the location of the miR-122 binding site, the function of the mRNA, or both can influence the mechanism by which target RNA molecules are regulated. Furthermore, we noted a second, juxtaposed binding site for miR-122 in the HCV genome that is important to ensure viral RNA accumulation. Genetic complementation experiments suggested that both sites are simultaneously occupied by miR-122 in a single viral mRNA, revealing a putative novel antiviral target.


The miR-122 binding site in HCV displays position-dependent function on gene expression when located in the 5’ or 3’NCR of reporter mRNAs

The binding site for miR-122 in the HCV 5’UTR shows characteristics of a typical miRNA binding site. Specifically, HCV RNA nucleotides 23 to 28 display Watson-Crick base pair complementarity to the microRNA seed sequence (Lewis et al., 2005), that encompasses nucleotides 2–7 (Jopling et al., 2005). We have shown recently that the interaction of miR-122 with the viral genome is essential for the accumulation of viral RNA in cultured Huh7 liver cells (Jopling et al., 2005).

To determine whether the upregulating function of the miR-122 binding site is dependent on the surrounding viral sequence context and its location at the 5’ end of the viral mRNA, we inserted nucleotides 1–347 of the HCV type 1a genome into the 5’NCR of a firefly luciferase reporter plasmid, designated pHCV-Luc (Fig. 1A). The plasmid was introduced into Huh7 cells together with a control Renilla luciferase expressing plasmid, pRL-SV40. Luciferase activity was measured in the presence of co-transfected 2’O-ribose methylated miR-122 antisense oligonucleotides (miR-122 antisense) with exact complementarity to miR-122, which were shown to sequester intracellular miR-122 molecules and to inhibit their activity in the RNA interfering pathway (Jopling et al., 2005). Randomized 2’O-methylated RNA oligonucleotides (Random-2’OMe), known to have no effect on miR-122 activity (Jopling et al., 2005), served as negative controls. The data in Figure 1B show that the intracellular abundance of miR-122 had no effect on the efficiency of reporter mRNA expression. Similarly, reporter gene expression was not affected after ectopic expression of synthetic wild-type miR-122 molecules (miR-122 wt) (Fig. 1B). Control Renilla luciferase activities were similar in each of these experiments (data not shown). These data argue that miR-122 binding sites do not affect the efficiency with which the HCV internal ribosome entry site directs protein synthesis when located at the 5’ end of reporter mRNAs.

Fig. 1
Expression of reporter mRNAs containing the HCV 5’ terminal sequences in their 5’ noncoding sequences. A: Diagram of capped Renilla luciferase-encoding reporter mRNA, expressed from plasmid pRL-SV40, and of firefly luciferase-encoding ...

To examine whether miR-122 affects reporter gene expression from miR-122 binding sites in 3’NCRs, HCV sequences 1–45 were inserted into the 3’NCR of the firefly luciferase reporter gene to yield plasmid pLUC-122x1 (Fig. 2A). Luciferase activity was measured in the presence of co-transfected Random-2’OMe oligonucleotides and set as 100 (Fig. 2C). Addition of miR-122 antisense oligonucleotides enhanced firefly luciferase activity by approximately 30% (Fig. 2C). In contrast, ectopic expression of synthetic miR-122 wt molecules (Fig. 2B) led to a 50% decrease in luciferase activity (Fig. 2C). Ectopic expression of mutated miR-122p3-4 RNAs (Fig. 2B), which should not bind to wildtype miR-122 seed match sequences (Jopling et al., 2005), had no effect on reporter mRNA expression (Fig. 2C). Thus, it appears that although the miR-122 binding site in HCV upregulates gene expression when residing in the 5’NCR of a replication-competent viral genome, it downregulates gene expression when located in the 3’NCR of a reporter mRNA, reminiscent of the outcome of a typical microRNA-mRNA interaction.

Fig. 2
Expression of reporter mRNAs containing miR-122 binding sites in their 3’ noncoding region. A: Diagram of capped Renilla luciferase-encoding reporter mRNA, expressed from plasmid pRL-SV40, and of firefly luciferase-encoding reporter mRNA, expressed ...

Identification of a second, adjacent microRNA binding site in the HCV genome

To determine whether additional miR-122 binding sites increased the effect on reporter gene expression, a second copy of HCV sequence 1–45 was inserted in the 3’NCR of pLUC-122x1, resulting in the plasmid pLUC-122x2 (Fig. 3A). As shown in the black bars in Figure 3B, sequestration of miR-122 by miR-122 antisense oligonucleotides resulted in an 80% increase in firefly luciferase activity compared to Random-2’OMe oligonucleotides. In contrast, ectopic expression of synthetic, wildtype miR-122 molecules caused a 70% decrease in luciferase production compared to mutant miR-122 p3-4 oligonucleotides (Fig. 3B). These findings indicate that miR-122-mediated repression of target mRNAs can occur by binding of miR-122 to tandem target sites, and that the repression is greater than that mediated by a single target site.

Fig. 3
Expression of reporter mRNAs that contain tandem HCV sequence elements. A: Diagram of a capped firefly luciferase-encoding reporter mRNA, expressed from plasmid pLUC-122x2, which harbors two HCV sequence elements in its 3’ noncoding region. Shown ...

To examine whether luciferase production in the chimeric mRNAs was modulated by direct binding of miR-122 to the tandem miR-122 binding sites, as opposed to binding to target sequences located in other mRNAs, the seed match sequences in both of the sites in pLUC-122x2 were mutated. Specifically, the nucleotides at positions 3 and 4 of each seed match sequence were mutated to their complement, i.e. U26C to A26G to yield S1:p3-4, such that the mutant sites are not predicted to bind wildtype miR-122 molecules. Instead, they should bind to mutant miR-122p3-4 molecules, restoring base pairing to the mRNA-miRNA pair. As shown by the striped bars in Figure 3B, miR-122 sequestration by methylated antisense miR-122 molecules still displayed an effect on reporter gene expression, although the effect was reduced relative to the effect observed with wildtype sequences. Furthermore, ectopic expression of either wildtype or mutant miR-122p3-4 microRNAs decreased reporter gene expression (Fig. 3B). These results suggest that additional binding sites for wildtype miR-122 may be present in the reporter mRNAs. Indeed, inspection of HCV nucleotide sequences 1–45 revealed a potential second seed match sequence that could form a Watson-Crick base pair interaction with miR-122. This putative seed match 2, spanning viral nucleotides 38–43, is highlighted in Figure 3A.

To test whether both seed match 1 and 2 functioned in the first 45 nucleotides of the HCV genome, positions 3 and 4 in both sequence elements were mutated to their complementary nucleotides, i.e. U26C to A26G and U41C to A41G, to yield plasmid S1+S2:P3-4 (Fig. 3A). Results from expression studies with this plasmid are shown by the white bars in Figure 3B. Neither sequestration of wildtype miR-122 nor overexpression of wildtype miR-122 had a significant effect on the expression of this reporter mRNA. However, ectopic expression of mutant miR-122p3-4 molecules caused an 80% decrease in luciferase activity. These findings provide genetic evidence for the presence of two functional miR-122 binding sites that can downregulate luciferase production.

Translational repression mediated by tandem miR-122 binding sites located in 3’ noncoding regions

There is some controversy about the mechanism by which microRNA-mediated repression occurs in mammalian cells. Studies have provided examples for microRNA-mediated repression at the translation initiation step, at the translation elongation step and at the post-elongation step. Furthermore, there are examples of microRNA-mediated mRNA degradation, possibly occurring as a result of translational repression (reviewed in (Jackson and Standart, 2007; Nilsen, 2007).

To examine the intracellular abundances of the reporter mRNAs, total RNA was extracted from Huh7 cells that had been transfected with plasmid pLUC-122x2 in the presence of the various small RNA modulators, and reporter mRNA abundances were quantified by real-time quantitative PCR. Whereas the ratio of firefly:Renilla luciferase activity changed with sequestration or overexpression of miR-122, as described above (Fig. 3B, black bars), no significant change in the ratio of firefly:Renilla mRNA was observed under these conditions (Fig. 3C). These findings argue that miR-122 mediates translational repression via two seed matches in the HCV genome when located in the 3’NCR of a luciferase mRNA, and that this repression likely occurs at a translational step and does not involve mRNA degradation.

Regulation of a tandem miR-122 binding site in the HCV genome by cellular miR-122

We have previously shown that mutation of seed match 1 diminished HCV RNA accumulation, which can be restored by mutant miR-122 molecules that can bind to the mutated seed match sequence (Jopling et al., 2005). To examine a functional role for the seed match 2 site in modulating HCV gene expression, defined mutations at the p3 and p4 positions of the seed match (Fig. 4A) were introduced into a type 1a HCV genome that contains adaptive mutations that allow it to replicate in Huh7 cells (Yi et al., 2006). In particular, double mutations (p3-4) in HCV were generated, in which seed matches 1 (S1) and 2 (S2) were both mutated at the p3 and p4 positions (Fig. 4A). RNA abundance of mutant RNAs was measured after electroporation into Huh7 cells. The Northern blot in Figure 4B (lane 1) shows that significant amounts of wildtype HCV RNA could be detected 5 days after electroporation, whereas neither mutant S2:p3-4 nor S1+S2:p3-4 RNA could be detected in transfected cells (lanes 2 and 4). However, both S2:p3-4 and S1+S2:p3-4 RNA could be detected after co-transfection of miR-122p3-4 RNAs, predicted to base pair with the mutant HCV genomes (Fig. 4B, lanes 3 and 5). Interestingly, ectopic expression of miR-122p3-4 repeatedly rescued the abundance of HCV RNAs that contained mutations at both S1 and S2 to near wildtype levels (Fig. 4B, lane 5), suggesting that ectopically expressed miR-122 RNAs display cooperative effects on HCV gene expression. Therefore, miR-122 interaction at seed match 1 alone is not sufficient to augment HCV RNA abundance. We conclude that miR-122 interactions at both seed matches are required for efficient HCV RNA accumulation.

Fig. 4
Effects of mutations in seed match sequences on abundance of replication-competent HCV RNA. A: Diagram of the introduced mutations in the two viral seed match sequences for miR-122. B,C,D: RNA abundances of wildtype or mutant HCV RNAs. Abundances were ...

To determine the effects of single-nucleotide mutations in seed match 1 and seed match 2, mutations at the p3 positions of both seed match sequences were tested. Similar to the findings with the double mutants, a single mutation in S2 (Fig. 4C, lane 2) or single mutations in both S1 and S2 (Fig. 4C, lane 4) abolished RNA accumulation, which could be restored by co-transfection of miR-122p3 molecules (Fig. 4C, lanes 3 and 5). These findings also argue that miR-122 binding to seed match 1 is not sufficient to allow viral RNA amplification.

To examine whether both seed match sequences are required to maintain HCV RNA abundance and are occupied by miR-122 in a single HCV RNA molecule, we introduced distinct mutations into S1 and S2 and monitored rescue of RNA abundance after transfection of distinct miR-122 molecules. Specifically, a p3 mutation in S1 and a p3-4 mutation in S2 were generated in the same viral RNA molecule (Fig. 4A). Transfection of this mutant RNA into Huh7 cells did not lead to detectable HCV RNA accumulation five days after electroporation (Fig. 4D, lane 2). Presumably, the endogenous wildtype miR-122 RNA had no effect on either mutant seed match sequence. Co-transfection of either miR-122p3 or miR-122p3-4 molecules was not sufficient to restore RNA abundance (Fig. 4D, lanes 3 and 4). In contrast, simultaneous transfection of both miR-122 mutant RNAs restored HCV RNA abundance (Fig. 4D, lane 5). This experiment suggests that both miR-122 binding sites need to interact with miR-122, either sequentially or concurrently, in a single HCV molecule to ensure viral RNA amplification.

Effect of a miR-122 seed match substitution on HCV gene expression

Next, we examined whether roles for miR-122 can be substituted by other microRNAs. Thus, we exchanged the miR-122 seed match 1 sequence element in the HCV genome with a seed match element that is predicted to base pair with microRNA miR-21 (Fig. 5A). Transfection of the mutant HCV-m21 genome into Huh7 cells failed to accumulate viral RNA (Fig. 5B). While miR-21 is expressed at a five-fold lower level than miR-122 in these cells (data not shown), HCV RNA abundance could not be rescued after ectopic expression of miR-21 duplexes (data not shown). These findings suggest that oligomeric miR-122 complexes have specialized functions in HCV replication or that the seed match substitution affected sequences whose integrity are essential for viral genome amplification.

Fig. 5
Effect of a seed match substitution on abundance of replication-competent HCV RNA. A: Diagram showing the locations of the introduced seed match swap. B: RNA abundances of wildtype or mutant HCV-m21 RNAs after electroporation into Huh7 cells. An autoradiograph ...

Modulation of viral gene expression by sequences encompassing the tandem miR-122 binding sites, including a conserved fourteen nucleotide spacer sequence element

To distinguish roles for sequences surrounding the miR-122 binding sites in microRNA binding or viral RNA replication, the replication-competence of HCV genomes carrying defined mutations was examined. It is known that nucleotide length between two tandemly located microRNA-binding sites can dictate the efficiency by which the microRNA regulates target mRNA expression (Doench and Sharp, 2004; Saetrom et al., 2007). In particular, optimal downregulation was found to be mediated by tandem microRNA binding sites in which the distance from the start of one seed to the start of the second seed sequence is 13–35 nucleotides (Saetrom et al., 2007). A fourteen nucleotide distance was observed between S1 and S2 in HCV RNA (Fig. 6A). Together with seed match sequences 1 and 2, this spacer sequence element is highly conserved between HCV genotypes, except that adenosine at position 36 is missing in genotypes 5 and 6 (Fig. 7A,B). To examine whether the conserved linker sequence is important for miR-122-mediated regulation of HCV RNA abundance, conserved nucleotides C30 and C31 were changed to adenosines (Fig. 6A). As shown in Figure 6B, transfected mutated viral genomes (pCC-AA) accumulated approximately two-fold less than wildtype RNA genomes (Fig. 6B, lanes 2 and 3). Ectopic expression of miR-122 wt duplexes did not rescue viral RNA abundance. More dramatically, mutant viral RNAs with nucleotides C30 and C31 deleted (pΔCC) failed to accumulate to detectable levels after transfection into Huh7 cells (Fig. 6B, lanes 4 and 5) and abundance could not be rescued by ectopic expression of miR-122 duplexes. Similarly, insertions of two or four uridines at position 32 (Fig. 6A) into the viral genome failed to generate significant levels of replication-competent RNAs (Fig. 6C, lanes 3 and 4). Insertion of the pCC-AA substitution, or the pΔCC deletion, into the 3’NCR of pLUC-122x1 (Fig. 2A) displayed translational repression in a microRNA-dependent manner that was similar to pLUC-122x1-derived mRNAs which contain wildtype microRNA binding sites and spacer sequences (Suppl. Fig. 1). These findings suggest that the spacer nucleotides contribute to the formation of a replication-competent RNA structure. Alternatively, engagement of the viral miR-122 binding sites with miR-122 has different effects on viral replication and reporter mRNA translation.

Fig. 6
Effects of mutations in spacer sequences on abundance of replication-competent HCV RNA. A: Diagram showing the locations of the introduced mutations in the spacer sequences. B,C: RNA abundances of wildtype or mutant HCV RNAs carrying mutations in the ...
Fig. 7
MiR-122 seed match sequences and predicted interactions with miR-122. (A) Locations of the miR-122 seed match sequences (highlighted in red) in HCV genotype 1a. The nucleotides linking the two seed match sequences are highlighted in blue. The borders ...


Most or all binding sites for microRNAs have been detected in 3’NCRs of target mRNAs (Lewis et al., 2005). Interactions of one or more microRNAs have been shown to lead to reduced expression of target mRNAs, and the mechanisms that repress mRNAs are potentially diverse (Jackson and Standart, 2007; Lewis et al., 2005). The positive effect of the interaction of miR-122 with HCV RNA on intracellular RNA abundance was surprising and raised the question whether a specialized miR-122-protein complex or the location of the miR-122 binding site at the 5’ end of the viral genome dictated its function. Thus, we examined the functional role of the HCV miR-122 binding site when located in the 5’ or 3’ noncoding regions of a reporter mRNA which was expressed by polymerase II. Results showed that reporter mRNAs, which lacked or contained miR-122 binding sites in their 5’NCRs, were translated with similar efficiencies in cultured human Huh7 cells. This result is in contrast to the finding of Lytle et al. (Lytle et al., 2007) who noted that let7 microRNA binding sites in 5’NCRs diminished translational efficiencies of reporter mRNAs in human HeLa cells. Because the affinity of microRNAs for target mRNA sequences is unknown, it is possible that the microRNA species or the composition of the RNA-induced silencing complex (RISC) can dictate the outcome by which the microRNA-mRNA complex is regulated. However, reporter mRNAs that contained miR-122 binding sites in their 3’NCRs were repressed at a post-transcriptional step in the presence of miR-122. This finding suggests that the location of the miR-122 binding site in an mRNA can dictate its function and that miR-122 does not associate with RISC which has a specialized upregulating function in liver cells.

MiR-122 binds to two adjacent sites in the HCV 5’NCR and binding to both sites is required to maintain RNA abundance (Fig. 7A,C). This finding was surprising because of the extreme proximity of the sites to each other. However, the fourteen nucleotide spacer sequence between the miR-122 binding sites is highly conserved among HCV genotypes (Fig. 7B). For both sites in the viral genome to be occupied concurrently would preclude base pair interactions outside the seed match sequences (Fig. 7C). Results from mutagenesis of the spacer region suggested that both spacer length and sequence context are important in allowing efficient accumulation of HCV RNA abundance in RNA-transfected cells. However, some of the spacer region mutants still repressed reporter mRNA translation in a microRNA-dependent manner. This observation argues that the spacer sequence contains RNA sequences that are essential for viral RNA replication. Crosslinking experiments, designed to test whether two miR-122 molecules can bind to spacer mutant RNAs, are currently being employed to distinguish between these possibilities.

While genetic evidence suggested that miR-122 needs to interact with both seed match site 1 and seed match site 2 in HCV to allow RNA accumulation, it is possible that miR-122 RISC complexes contact only one site, then dissociate and stably associate with the second site. While such a hit-and-run mechanism is possible, it is more likely that two miR-122 RISCs are anchored, predominantly by seed-seed match interactions, simultaneously on the same mRNA molecule. A recent study has shown that optimal repression of a target mRNA can be achieved by tandem microRNA binding sites that are separated by 13–35 nucleotides (Saetrom et al., 2007). Because effects of 3’ end sequences in the microRNA were minimized in this study, it was concluded that optimally spaced seed-seed match microRNA RISCs may cooperatively regulate gene expression (Saetrom et al., 2007). Similarly, it is possible that the tandem binding of miR-122 RISC to the HCV genome provides a RNA-protein scaffold for factors that enhance viral RNA abundance. Such a scaffold could provide a novel antiviral target. Studies in mice indicated that the normal function of miR-122 in liver is to regulate the expression of genes involved in cholesterol and fatty-acid biosynthesis (Esau et al., 2006; Krützfeldt et al., 2005). Whether mRNAs targeted by miR-122 are regulated by a similar cooperative mechanism is unknown, but it is curious that numerous microRNA-targeted mRNAs are predicted to bind several copies of microRNAs which can be either identical or distinct (Saetrom et al., 2007).

The observation that the miR-122 RISC-HCV complex can exert either positive or negative function on target mRNA expression is intriguing. When located in the 3’NCR of a reporter mRNA, the miR-122 RISC downregulates mRNA expression, as would be expected for any microRNA. However, location of the miR-122 RISC complex at the 5’ end of the viral genome upregulates viral RNA abundance. In this case, translation of the viral genome is not affected (Jopling et al., 2006; Jopling et al., 2005). Similarly, inserting the miR-122 binding site into the 5’NCR of a reporter mRNA had no effect on translational efficiency of the targeted mRNA. Thus, the upregulating function of miR-122 is, thus far, confined to a replicating viral mRNA. The mechanism by which viral RNA abundance is augmented by miR-122 will most likely involve modulating rates of RNA amplification and, possibly, intracellular targeting of miR-122 RISCHCV complexes.

Experimental Procedures

Oligonucleotides and DNA constructs

RNA oligonucleotides and 2’-O-methylated oligonucleotides were synthesized by Dharmacon Inc. (Lafayette, CO). The sequences were: 122-2’OMe, 5’-AGACACAAACACCAUUGUCACACUCCACAGC-3’; Rand-2’OMe, 5’-CACGUUAAAACCAUACGCACUACGAAACCCC-3’; miR-122wt, 5’-UGGAGUGUGACAAUGGUGUUUGU-3’; miR-122p3, 5’-UGCAGUGUGACAAUGGUGUUUGU-3’; miR-122p3-4, 5’-UGCUGUGUGACAAUGGUGUUUGU-3’; miR-122*, 5’-AAACGCCAUUAUCACACUAAAUA-3’. Duplexes were formed between the wt, p3 or p3-4 forms of miR-122 and its complement (miR-122*).

To generate the plasmid pHCV-LUC, nucleotides 1–374 of HCV type 1a were amplified by PCR from the vector pH77ΔE1/p7 (Yi et al., 2006) using the primers 5’-CGACTCAAGCTTGCCAGCCCCCTGATGGGG-3’ and 5’-GTTGGTGTTACCCATGGTTTTTCTTTGAGG-3’ and inserted between the HindIII and NcoI restriction sites in the plasmid pGL3con (Promega). To generate plasmid pGL3-MCS, a multiple cloning site cassette, 5’-GATGAAGCTTACTAGTGCGGCCGCTGCAGAATTCCCGGGCCCGCGGATCCATCAGCTAGA-3’, was ligated into the XbaI site in the SV40 promoter-based vector pGL3con (Promega), allowing insertion of HCV sequences into the 3’NCR. Nucleotides 1–45 of the HCV type 1a sequence were amplified by PCR from the vector pH77ΔE1/p7 and inserted between the SpeI and EcoRI sites in pGL3-MCS to create the plasmid pLUC-122x1. The plasmid pLUC-122x2 was constructed by PCR amplification and insertion of a second copy of nucleotides 1–45 of HCV, between the EcoRI and SacII sites in pLUC-122x1. The nucleotides that bind to p3-4 of miR-122 in seed match site 1 in each of the two sites in pLUC-122x2 were mutated to their complement to create S1:P3-4A: U26C to A26G. This mutation was also introduced into seed match site 2 in each of the two sites to generate S2:P3-4B: U41C to A41G. The p3-4 and p3 seed match mutations were created in the vector pH77ΔE1/p7 by overlap PCR of the region between the XmnI and KpnI restriction sites, using the outer primers 5’-GCTCATCATTGGAAAACGTTCTTCGGGGCG-3’ and 5’-GCCAAGGGTACCCGGGCTGAGCCCAGGTCC-3’. The mutagenic primers 5’-CCATGAATCACAGCCCTGTGAGGAACTAC-3’ and 5’-GTAGTTCCTCACAGGGCTGTGATTCATGG-3’ were used to generate the S2:p3-4 mutant, and primers 5’-CCATGAATCACTGCCCTGTGAGGAACTAC-3’ and 5’-GTAGTTCCTCACAGGGCAGTGATTCATGG-3’ to create S2:p3. The S1+S2:p3-4 and S1+S2:p3 mutants were generated by the same method using the previously published S1:p3-4 and S1:p3 mutants as templates (Jopling et al., 2005). Mutant HCV-m21 was generated by changing nucleotides 23–28 in the HCV genome from ACAUCC to UAAGCU.

The pCC-AA, pΔCC, 32-2xU and 32-4xU mutations were created by directed mutagenesis (Quickchange II XL, Stratagene). A PCR fragment spanning the XmnI and AgeI-restriction sites of the H77ΔE1/p7 plasmid (5’-TGCTCATCATTGGAAAACG-3’, 5’-GGCAATTCCGGTGTACTCAC-3’) was subcloned by TOPO-TA cloning into pCR2.1 (Invitrogen) and mutated according to the manufacturers instructions with the following primers: pCC-AA: 5’-CTGATGGGGGCGACACTCCAAAATGAATCACTCCCCTGTGAG-3’ and 5’-CTCACAGGGGAGTGATTCATTTTGGAGTGTCGCCCCCATCAG-3’, pΔCC: 5’-CTGATGGGGGCGACACTCCAATGAATCACTCCCCTGTGAG-3’ and 5’-CTCACAGGGGAGTGATTCATTGGAGTGTCGCCCCCATCAG-3’. 32-2xU: 5'-CTGATGGGGGCGACACTCCACCTTATGAATCACTCCCCTGTGAG-3' and 5'-CTCACAGGGGAGTGATTCATAAGGTGGAGTGTCGCCCCCATCAG-3'. 32-4xU: 5'-CTGATGGGGGCGACACTCCACCTTTTATGAATCACTCCCCTGTGAG-3' and 5'-CTCACAGGGGAGTGATTCATAAAAGGTGGAGTGTCGCCCCCATCAG-3'. The mutated sequences were inserted into the parental vector between the XmnI and AgeI sites.

Cell culture and transfection

Huh7 cells were cultured in DMEM supplemented with 10% FBS and 1% non-essential amino acids (Gibco). DNA constructs and oligonucleotides were introduced into cells in six-well plates using Lipofectamine 2000 (Invitrogen), according to the manufacturer’s instructions. The plasmid pRL-SV40 (Promega) was included as a transfection control. The 2’-O-methylated oligonucleotides and the miR-122 RNA duplexes were delivered to cells at a concentration of 20nM in the medium. Transfected cells were cultured for 24 hours before harvesting into Passive Lysis Buffer (Promega), according to the manufacturer’s instructions. The replication-competent HCV RNA, H77ΔE1/p7 (Yi et al., 2006), carrying a deletion in the envelope gene, was generated by in vitro transcription using the T7 Megascript kit (Ambion), according to the manufacturer’s instructions, and introduced into Huh7 cells by electroporation as described (Yi and Lemon, 2004). RNA was prepared and processed five days later. When included in experiments, miR-122 duplexes were introduced by Lipofectamine 2000-mediated transfection at one day prior to electroporation and one and three days subsequently.

RNA isolation and Northern analysis

Total RNA was extracted from cells using Tri Reagent (Sigma), according to the manufacturer’s instructions. To analyse mRNA levels, 5–20µg of total RNA was separated in 1% agarose gels containing 1x MOPS buffer and 2.2M formaldehyde (Ausubel et al., 1989), and transferred to Zeta-probe membrane (Bio-Rad). The membranes were hybridized overnight in Church-Gilbert solution (180mM Na2HPO4, 70mM NaH2PO4, 7% SDS) to random-primed 32P-labeled DNA probes complementary to nucleotides 84–374 of HCV and to nucleotides 685–1171 of γ-actin, as indicated.

Quantitative real-time PCR

Relative amounts of firefly and Renilla luciferase RNA were quantified by real-time PCR using a Rotor-gene SE6000 (Corbett Life Science). The primers used were 5’-TCGCCAGTCAAGTAACAAC and 5’-ACTTCGTCCACAAACACAA to amplify firefly luciferase RNA, and 5’-AACGCGGCCTCTTCTTATTT and 5’-GTCTGGTATAATACACCGCG for Renilla luciferase. PCR was carried out using Power SYBR green master mix (Applied Biosystems) according to the manufacturer’s instructions.

Supplementary Material


We are grateful to Richard Jackson (Cambridge University) in whose laboratory many of these experiments were performed. We thank Karla Kirkegaard for critical comments on the manuscript. We also thank Gus Zeiner for use of electroporation reagents. Work performed in the authors’ laboratories was supported by grants from the Wellcome Trust (C.L.J.), the Deutsche Forschungsgemeinschaft (Schu 22941/2-1 to S.S.) and the National Institutes of Health (GM069007, AI047365, AI069000 to P.S.).


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


  • Ausubel FM, Brent R, Kingston RE, Moore DD, Smith JA, Seidman JG, Struhl K. Current Protocols in Molecular Biology. New York: Current Protocols (Greene Publishing Associates, Inc. and John Wiley & Sons, Inc.; 1989.
  • Bartel DP. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell. 2004;116:281–297. [PubMed]
  • Brennecke J, Stark A, Russell RB, Cohen SM. Principles of microRNA-target recognition. PLoS Biol. 2005;3:e85. [PubMed]
  • Chang J, Nicolas E, Marks D, Sander C, Lerro A, Buendia MA, Xu C, Mason WS, Moloshok T, Bort R, et al. miR-122, a mammalian liver-specific microRNA, is processed from hcr mRNA and may downregulate the high affinity cationic amino acid transporter CAT-1. RNA Biology. 2004;1:106–113. [PubMed]
  • Doench JG, Sharp PA. Specificity of microRNA target selection in translational repression. Genes Dev. 2004;18:504–511. [PubMed]
  • Esau C, Davis S, Murray SF, Yu XX, Pandey SK, Pear M, Watts L, Booten SL, Graham M, McKay R, et al. miR-122 regulation of lipid metabolism revealed by in vivo antisense targeting. Cell Metab. 2006;3:87–98. [PubMed]
  • Grimson A, Farh KK, Johnston WK, Garrett-Engele P, Lim LP, Bartel DP. MicroRNA targeting specificity in mammals: determinants beyond seed pairing. Mol Cell. 2007;27:91–105. [PMC free article] [PubMed]
  • Hoofnagle JH. Course and outcome of hepatitis C. Hepatology. 2002;36:S21–29. [PubMed]
  • Jackson RJ, Standart N. How do microRNAs regulate gene expression? Sci STKE. 2007;2007:re1. [PubMed]
  • Jopling CL, Norman KL, Sarnow P. Positive and negative modulation of viral and cellular mRNAs by liver-specific microRNA miR-122. Cold Spring Harb Symp Quant Biol. 2006;71:369–376. [PubMed]
  • Jopling CL, Yi M, Lancaster AM, Lemon SM, Sarnow P. Modulation of hepatitis C virus RNA abundance by a liver-specific MicroRNA. Science. 2005;309:1577–1581. [PubMed]
  • Krützfeldt J, Rajewsky N, Braich R, Rajeev KG, Tuschl T, Manoharan M, Stoffel M. Silencing of microRNAs in vivo with 'antagomirs'. Nature. 2005;438:685–689. [PubMed]
  • Lewis BP, Burge CB, Bartel DP. Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets. Cell. 2005;120:15–20. [PubMed]
  • Lytle JR, Yario TA, Steitz JA. Target mRNAs are repressed as efficiently by microRNA-binding sites in the 5' UTR as in the 3' UTR. Proc Natl Acad Sci U S A. 2007;104:9667–9672. [PubMed]
  • Nilsen TW. Mechanisms of microRNA-mediated gene regulation in animal cells. Trends Genet. 2007;23:243–249. [PubMed]
  • Randall G, Panis M, Cooper JD, Tellinghuisen TL, Sukhodolets KE, Pfeffer S, Landthaler M, Landgraf P, Kan S, Lindenbach BD, et al. Cellular cofactors affecting hepatitis C virus infection and replication. Proc Natl Acad Sci U S A. 2007;104:12884–12889. [PubMed]
  • Saetrom P, Heale BS, Snove O, Jr, Aagaard L, Alluin J, Rossi JJ. Distance constraints between microRNA target sites dictate efficacy and cooperativity. Nucleic Acids Res. 2007;35:2333–2342. [PMC free article] [PubMed]
  • Xie X, Lu J, Kulbokas EJ, Golub TR, Mootha V, Lindblad-Toh K, Lander ES, Kellis M. Systematic discovery of regulatory motifs in human promoters and 3' UTRs by comparison of several mammals. Nature. 2005;434:338–345. [PMC free article] [PubMed]
  • Yi M, Lemon SM. Adaptive mutations producing efficient replication of genotype 1a hepatitis C virus RNA in normal Huh7 cells. J Virol. 2004;78:7904–7915. [PMC free article] [PubMed]
  • Yi M, Villanueva RA, Thomas DL, Wakita T, Lemon SM. Production of infectious genotype 1a hepatitis C virus (Hutchinson strain) in cultured human hepatoma cells. Proc Natl Acad Sci U S A. 2006;103:2310–2315. [PubMed]