siRNAs can function as miRNAs

Abstract

With the discovery of RNA interference (RNAi) and related phenomena, new regulatory roles attributed to RNA continue to emerge. Here we show, in mammalian tissue culture, that a short interfering RNA (siRNA) can repress expression of a target mRNA with partially complementary binding sites in its 3' UTR, much like the demonstrated function of endogenously encoded microRNAs (miRNAs). The mechanism for this repression is cooperative, distinct from the catalytic mechanism of mRNA cleavage by siRNAs. The use of siRNAs to study translational repression holds promise for dissecting the sequence and structural determinants and general mechanism of gene repression by miRNAs.

Figure 1

siRNAs translationally represses a target mRNA. (A) Schematic of the proposed interaction between a binding site engineered into the 3′ UTR of the target mRNA and the antisense strand of the CXCR4 siRNA. The thymidines at the 3′ end of the siRNA are deoxynucleotides. (B) Dual luciferase assay of transfected HeLa cells. Three Renilla reniformis luciferase (Rr-luc) constructs were used in this assay. One was unmodified (“no sites”), one contained a binding site perfectly complementary to the siRNA strand shown in A (“1 perfect”), and one contained four of the binding sites shown in A in tandem repeat (“4 bulged”). A Photinus pyralis luciferase (Pp-luc) served as an internal transfection control. The cells were transfected with no siRNA (black bars), a nonspecific (targeting GFP) siRNA (white bars), or the CXCR4 siRNA (gray bars). The ratios of Rr-luc to Pp-luc expression were normalized to the no siRNA transfections, ±S.E. from three independent experiments. (C) RT–PCR of harvested RNA. Total RNA was harvested from cells transfected with the constructs described in B, transfected with or without the CXCR4 siRNA. Control experiments demonstrate that DNA was successfully removed from the RNA preparation and that the PCR was in the linear range of amplification (data not shown). (D) Schematic of the proposed interaction between the sense strand of the CXCR4 siRNA and a designed binding site. (E) RNA analysis of Pp-luc with four bulged CXCR4 binding sites (shown in A), targeted for translational repression, transfected either with the CXCR4 siRNA (+) or no siRNA (−). RNA was detected by Northern analysis, probing for either Pp-luc or β-actin.

Figure 2

Analysis of sequence and structure rules for siRNA: mRNA interaction. HeLa cells were transfected with constructs containing four binding sites in tandem repeat with imperfect complementarity to either the antisense (A) or sense (B) strand of a GFP siRNA. The effect on luciferase expression is shown by the white bars, ±S.E. from two independent experiments, normalized to cells transfected with no siRNA (black bars). A different siRNA was then used to produce different bulges, shown in gray with arrows. These new interactions were assayed and are depicted with gray bars.

Figure 3

Comparison of RNAi and translational repression. (A) Titration of Renilla reniformis luciferase (Rr-luc) constructs containing zero (○), two (▪), four (×), or six (●) of the bulged binding sites, for pairing with the antisense strand of the CXCR4 siRNA, as depicted in Figure 1A. The level of repression achieved is plotted, normalized to cells transfected with no siRNA. (B) Titration of Pp-luc constructs containing zero (○), one (▪), two (×), or three (●) binding sites perfectly complementary to the antisense strand of the CXCR4 siRNA (Fig. 1A). (C) Analysis of the relative repression each site contributes for the data presented in A, normalized to the construct with two binding sites, ±S.E. (D) Analysis of the relative repression each site contributes for the data presented in B, normalized to the construct with one binding site, ±S.E.