Biological basis for restriction of microRNA targets to the 3' untranslated region in mammalian mRNAs

Abstract

MicroRNAs (miRNAs) interact with target sites located in the 3' untranslated regions (3' UTRs) of mRNAs to downregulate their expression when the appropriate miRNA is bound to target mRNA. To establish the functional importance of target-site localization in the 3' UTR, we modified the stop codon to extend the coding region of the transgene reporter through the miRNA target sequence. As a result, the miRNAs lost their ability to inhibit translation but retained their ability to function as small interfering RNAs in mammalian cells in culture and in vivo. The addition of rare but not optimal codons upstream of the extended opening reading frame (ORF) made the miRNA target site more accessible and restored miRNA-induced translational knockdown. Taken together, these results suggest that active translation impedes miRNA-programmed RISC association with target mRNAs and support a mechanistic explanation for the localization of most miRNA target sites in noncoding regions of mRNAs in mammals.

Figure 1. MiRNA-mediated repression is abolished in extended ORFs

The structure of the reporter constructs used in this study. pGL3-control containing no miRNA target sites; pGL3-3'UTR with two tandem mir-30 targets sites located in 3' untranslated region; and pGL3-ORF with upstream stop-codon abolished and mir-30 target sites covered by extended ORF. Grey box represents the ORF of FF-luciferase gene. Dark box represents the tandem mir-30 target sites with six base-pairs in between. Positions of upstream (original) stop-codon and downstream stop-codon are indicated by solid and dotted arrows, respectively. (A) Schematic illustration of the interactions between mir-30 target sequence and guiding strand sequence of sh-mir-30 and sh-mir-30P, respectively. (C) NIH3T3 cells were co-transfected with plasmids, as described above. Sh-RNA expressed from U6-driven cassette was detected by Northern blot using either a probe against mir-30 (Up) or a probe against mir-30P (Down). Due to sequence similarity, cross-hybridization was observed. Endogenous U6 snRNA was also detected as an internal control. (D)HEK293 cells and (E) NIH3T3 cells were co-transfected with different combinations of plasmids, and dual-luciferase assays were performed 36hr post-transfection. FF-luciferase activities were normalized with RL-luciferase, and the percentage of relative enzyme activity compared to the negative control (treated with sh-scramble) was plotted. Error bars represent the standard deviation from three independent experiments, each performed in triplicate. (F) Protein analysis by Western blot was performed in transfected 3T3 cells. A protein band of β-actin was used as an internal control. Positions of the bands representing wild-type or mutant FF-luciferase were indicated by arrows. A non-specific band was indicated by an asterisk. (G) RNA levels of either FF-luciferase or RL-luciferase from transected 3T3 cells were detected by ribonuclease protection assay (RPA). Full-length probes and protected bands were indicated in the figure. A band labeled with an asterisk is possibly due to a truncated RL probe and, therefore, corresponding to RL-luciferase mRNA level.

Figure 2. MiRNA-mediated repression studies were concordant in mouse liver in vivo

(A) The plasmids described in Figure 1 were transfected into mice by hydrodynamic tail injection (N = 5 per group, except group 4, N = 4, one animal died after injection). Real-time transgene expression was determined four days after injection. Levels of luciferase reporter activities were quantified as shown in each image. (B) A control plasmid, RSV-hAAT was co-transfected within each sample as an internal control for transfection efficiency. The FF-luciferase activities were normalized to serum hAAT levels measured by ELISA. Percentage of relative luciferase activity compared to negative controls (treated with sh-scramble) was plotted. Error bars represent the standard deviation.

Figure 3. Insertion of rare-codons upstream of the extended miRNA ORF rescues miRNA-mediated knockdown

(A) The maps of the reporter constructs used in this study. Plasmids containing tandem mir-30 target sequence in either 3'UTR (a1) or ORF (a2) are the same as those described in Figure 1. A cluster of rare codons (represented as a dark box) were inserted either upstream (a3) or downstream (a4) of mir-30 target sequences. In another construct, the upstream rare codons (a3) were replaced with optimal-codon sequences that code for the same peptide sequence. The arrows and grey box represent the position of the miRNA target sequences. (B) HEK293 cells and (C), (D) NIH3T3 cells were transfected with the reporter constructs illustrated in (A). Dual-luciferase assays were performed 36hrs post-transfection. FF-luciferase activities were normalized with RL-luciferase, and the percentage of relative enzyme activity compared to the negative control (treated with sh-scramble) was plotted. Error bars represent standard deviation from three independent experiments, each performed in triplicate. (E) Protein levels of reporter genes were analyzed by Western blot in transfected 3T3 cells. (F) NIH3T3 cells were transfected with constructs as indicated in the figure. Insertion of rare-codon cluster (dark box) upstream of mir-30 targets sites in the 3'UTR did not substantially change the miRNA-induced repression. (G) RNA levels of reporter genes were analyzed by Ribonuclease Protection Assay. The loading sequence of line 1 to 11 is same as noted in (E).

Figure 4. Insertion of rare codons increases the accessibility of downstream sequences to RNase H-mediated cleavage

(A) Experimental strategy. Cells were transfected with the luciferase reporter constructs containing the cluster of rare or optimal codons (Figure 3A). After fixing the ribosomes on the mRNA by the addition of cycloheximide, one of six oligos corresponding to the region between rare/optimal codons and target 3'UTR was added into the cell extracts. The hybridization of DNA oligos at the target site within the mRNA results in cleavage mediated by the endogenous RNase H activity in the cell extracts. The extent of the cleavage represents the relative RNA accessibility, which was quantified by real-time RT-PCR using two primers flanking the cleavage sites. (B) Quantification of RNase H-mediated cleavage. The values are presented as the relative PCR signal compared to control samples treated with a scrambled oligonucleotide and normalized for a GFP mRNA obtained from a co-transfected control plasmid.