Examination of artificial MiRNA mimics with centered-site complementarity for gene targeting

Abstract

Background: MiRNA primarily acts to repress gene expression at the post-transcriptional level through imperfect complementarity of its 5' region to the "seed site" in the 3' untranslated region of target mRNAs, with its "3'-supplementary site" and "center site" also playing important roles under certain circumstances. The aim of this study was to test if artificial miRNA mimics (miR-Mimics) that are designed to target the "centered sites" without "seed sites" complementarity are able to repress gene expression as natural miRNAs.

Methods: We designed miR-Mimics carrying centered-site matches (CS-miR-Mimics) or seed-site matches (SS-miR-Mimics) and siRNA to two antiapoptotic genes BCL2 and AKT1. We tested the gene targeting of these constructs using real-time RT-PCR and Western blot to quantify mRNA and protein levels of BCL2 and AKT1, respectively, luciferase reporter gene assay to investigate the interaction between miR-Mimics and their target sites, and cell survival assay to study the functional outcomes of the miR-Mimics.

Results: We found that CS-miR-Mimic, SS-miR-Mimic and siRNA, all down regulated the mRNA and protein levels of their cognate target BCL2 or AKT1 in a concentration-dependent manner. Luciferase reporter gene assay further confirmed the functional interactions of CS-miR-Mimic, SS-miR-Mimic and siRNA with their target sites. We then observed that the miR-Mimics and siRNAs were all able to induce cell death, as indicated by the reduced survival rate of cells.

Conclusions: We have provided evidence for the feasibility of CS-miR-Mimics for post-transcriptional repression of genes, which can be designed to have reduced numbers of seed type off-target sites compared to the number of target sites from an average endogenous seed-site miRNA. CS-miR-Mimics may be a novel approach for miRNA research requiring miRNA gain-of-function.

Figure 1. Design of centered–site miRNA mimics (CS–miR-Mimics), seed–site miRNA mimics (SS–miR-Mimics), and siRNAs targeting BCL2 (a) and AKT1 (b) genes, respectively.

The centered sites (nucleotides 5–16 from 5′–end) are indicated by underlined blue letters in boldface, and the seed sites (nucleotides 2–8 at 5′–end) by boldface, red letters and the 3′–complementary sites (nucleotides 13–16 from 5′–end) by boldface, purple letters. GS: guide strand; PS: passenger strand.

Figure 2. Effects of miR-Mimics on the mRNA levels of their target genes BCL2 and AKT1 in H9c2 rat ventricular cells, determined by real-time quantitative RT-PCR.

(a) and (b) Concentration-response curves of BCL2 and AKT1, respectively. Measurements were made 24 hrs after transfection of cells with CS–miR-Mimics (centered site miR-Mimics), SS–miR-Mimics (seed site miR-Mimics), siRNA, or NC miR-Mimic (negative control miR-Mimic) using lipofectamine 2000. The concentrations of the constructs tested were 1, 10, and 100 nM, expressed in log10 scale. Control (Ctl) cells were mock-treated. Symbols are averaged experimental data and the curves are fits by Hill equation. For BCL2, EC₅₀ = 2.5 nM for CS–miR-Mimic, EC₅₀ = 4.6 nM for SS-miR-Mimic, and EC₅₀ = 1.6 nM for siRNA. For AKT1, EC₅₀ = 2.5 nM for CS–miR-Mimic, EC₅₀ = 10 nM for SS-miR-Mimic, and EC₅₀ = 2.1 nM for siRNA. Note that the constructs for AKT1 failed to affect BCL2 (c) and the constructs for BCL2 failed to affect AKT1 (d). *p<0.05 vs. Ctl; n = 5 for each group.

Figure 3. Effects of miR-Mimics on the protein levels of Bcl-2 (a) and Akt1 (b) in H9c2 rat ventricular cells, determined by Western blot analysis.

Upper panels: representative immunoblotting bands; lower panels: averaged band densities. Measurements were made 24 hrs after transfection of cells with varying constructs using lipofectamine 2000. CS/BCL2 = CS-miR-Mimic/BCL2 targeting BCL2; CS/AKT1 = CS–miR-Mimic/AKT1 targeting AKT1. The concentrations of the constructs tested were 10 nM. Control (Ctl) cells were mock-treated. *p<0.05 vs. Ctl; n = 4 for each group.

Figure 4. Interactions between miR-Mimics and their target sites in, as reported by luciferase activity assay with the pMIR-REPORT^TM luciferase miRNA expression reporter vector carrying the BCL2 or AKT1 3′UTR in H9c2 rat ventricular cells.

(a) and (b) Concentration-response curves. Measurements were made 24 hrs after transfection of cells with varying constructs using lipofectamine 2000. The concentrations of the constructs tested were 1, 10, and 100 nM, expressed in log10 scale. Control (Ctl) cells were transfected with the luciferase vector alone without miR-Mimics. Symbols are averaged experimental data and the curves are fits by Hill equation. For BCL2, EC₅₀ = 1.7 nM for CS-miR-Mimic, EC₅₀ = 1.3 nM for SS-miR-Mimic, and EC₅₀ = 1.7 nM for siRNA. For AKT1, EC₅₀ = 1.7 nM for CS-miR-Mimic, EC₅₀ = 2.9 nM for SS-miR-Mimic, and EC₅₀ = 2.7 nM for siRNA. Note that the constructs for AKT1 failed to affect BCL2 (c) and the constructs for BCL2 failed to affect AKT1 (d). *p<0.05 vs. Ctl; n = 4 for each group.

Figure 5. Effects of various miR-Mimics targeting BCL2 (a) or AKT1 (b) on H9c2 cell survival evaluated with MTT assay.

Cell death was induced by incubating with H₂O₂ (50 µM). “+” indicates H₂O₂ + construct. *p<0.05 vs. Ctl; ^Φ p<0.05 vs. H₂O₂ alone; n = 5 for each group.