Position-dependent function for a tandem microRNA miR-122-binding site located in the hepatitis C virus RNA genome

Abstract

MicroRNAs usually interact with 3' noncoding regions (3'NCRs) of target mRNAs leading to downregulation of mRNA expression. In contrast, liver-specific microRNA miR-122 interacts with the 5' end of the hepatitis C virus RNA genome, resulting in increased viral RNA abundance. We find that inserting the viral miR-122 binding site into the 3' noncoding region of a reporter mRNA leads to downregulation of mRNA expression, indicating that the location of the miR-122 binding site dictates its effect on gene regulation. Furthermore, we discovered an adjacent, second miR-122 binding site, separated from the first by a highly conserved 14-nucleotide sequence. Mutational analysis demonstrates that both miR-122 binding sites in a single viral genome are occupied by the microRNA and function cooperatively to regulate target gene expression. These findings set a paradigm for dual, position-dependent functions of tandem microRNA-binding sites. Targeting an oligomeric microRNA complex offers potential as an antiviral-intervention strategy.

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 reporter mRNA, expressed from plasmid pHCV-LUC, harbouring HCV sequences 1-374 in its 5’ noncoding region. The nucleotides comprising the HCV 5’ noncoding region are shown in bold. B: Effects of methylated antisense miR-122 oligos (miR-122 antisense), methylated random (Random-2’OMe), or ectopically expressed wildtype (miR-122 wt) molecules on reporter mRNA expression. Firefly luciferase and control Renilla luciferase activities were measured in lysates, and results from triplicate experiments are displayed as a ratio of firefly:Renilla luciferase activity, expressed as a percentage of the values obtained from mock-treated cells. The error bars represent standard error of the mean. The expression of Renilla luciferase did not vary significantly between samples.

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 from plasmid pLUC-122x1, harbouring HCV sequences 1-45 in its 3’ noncoding region. The nucleotides comprising the miR-122 seed match sequence in the viral RNA are shown in bold. B: Nucleotide sequences of ectopically expressed wildtype miR-122 (miR-122 wt) and of a mutant miR-122 that contains mutations at positions 3 and 4 in the seed match (miR-122 p3-4). C: Effects of methylated random (Random-2’OMe), methylated antisense miR-122 oligos (miR-122 antisense), ectopically expressed wildtype (miR-122 wt) or mutant (miR-122 p3-4) miR-122 molecules on reporter mRNA expression. Firefly luciferase and control Renilla luciferase activities were measured in lysates, and results from triplicate experiments are displayed as a ratio of firefly:Renilla luciferase activity, expressed as a percentage of the values obtained in cells treated with Random-2’OMe. The error bars represent standard error of the mean. The expression of Renilla luciferase did not vary significantly between samples.

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 above are mutations at positions 3 and 4 of the miR-122 seed match in each copy of the HCV sequence element (seed match 1 and 2). B: Effects of various oligos and duplex mRNAs on reporter mRNA expression. See details in legend to Figure 1C. C: Effects of various oligos and duplex RNAs on RNA abundance of reporter RNAs. RNA abundance was measured by quantitative RT-PCR analysis. Results from triplicate experiments are displayed as a ratio of firefly:Renilla RNA abundance, and are expressed as a percentage of the values obtained in cells treated with Random-2’OMe. The error bars represent standard error of the mean.

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 measured by Northern blot analyses five days after electroporation into Huh7 cells. Duplex miR-122p3 or miR-122p3-4 molecules were introduced into cells one day prior to electroporation of HCV RNAs and again at one and three days after electroporation of viral RNA. Autoradiographs of Northern blots, displaying HCV and γ-actin RNA abundance from an example of triplicate experiments are shown.

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 of a Northern blot is shown. Experimental protocol was performed as described in the legend to figure 4.

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 spacer sequence element after electroporation into Huh7 cells. An autoradiograph of a Northern blot showing RNA abundance of wildtype or mutant RNAs is shown. Experimental protocol was carried out as described in the legend to figure 4. The ratio of HCV to γ-actin RNA from three independent experiments is shown above the autoradiograph. The error bars represent standard error of the mean.

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 of the conserved fourteen nucleotide spacer element, located between the two seed match sequence positions #1 in the viral genome are indicted by the arrows. (B) Seed match sequences (red) and surrounding sequences (blue) which are conserved in all HCV genotypes are shown. The missing adenosine nucleotide #36 in HCV type 5 and type 6 is indicated by a “-“ symbol. Consensus sequences were compiled with the aid of the HCV sequence database at http://hcv.lanl.gov/content/hcv-db/index. (C) Potential interactions of two miR-122 molecules, shown in green with the seed match sequences (red) in HCV are shown.