Poly(A)-binding protein modulates mRNA susceptibility to cap-dependent miRNA-mediated repression

Abstract

MicroRNAs (miRNAs) regulate gene expression post-transcriptionally through binding specific sites within the 3' untranslated regions (UTRs) of their target mRNAs. Numerous investigations have documented repressive effects of miRNAs and identified factors required for their activity. However, the precise mechanisms by which miRNAs modulate gene expression are still obscure. Here, we have examined the effects of multiple miRNAs on diverse target transcripts containing artificial or naturally occurring 3' UTRs in human cell culture. In agreement with previous studies, we report that both the 5' m(7)G cap and 3' poly(A) tail are essential for maximum miRNA repression. These cis-acting elements also conferred miRNA susceptibility to target mRNAs translating under the control of viral- and eukaryotic mRNA-derived 5' UTR structures that enable cap-independent translation. Additionally, we evaluated a role for the poly(A)-binding protein (PABP) in miRNA function utilizing multiple approaches to modulate levels of active PABP in cells. PABP expression and activity inversely correlated with the strength of miRNA silencing, in part due to antagonism of target mRNA deadenylation. Together, these findings further define the cis- and trans-acting factors that modulate miRNA efficacy.

FIGURE 1.

An m⁷G-cap and poly(A) tail are required for maximum miRNA-mediated repression. (A) (Left panel) Schematic representation of reporter mRNAs utilized. Each reporter contains eight tandem miR-30 target sites (black ovals) in the 3′ UTR and either ∼30 nt of vector-derived sequence, the HCV IRES, or a mutated HCV IRES lacking subdomain IIIf (asterisk; IRES^mut) in the 5′ UTR. The predicted base-pairing between miR-30 and a single target site is shown. (Right panel) Raw RLU values for capped and polyadenylated HCV IRES reporter mRNAs after transfection into 293T cells. (B) miR-30 repression levels of transfected mRNA constructs. The indicated mRNAs were co-transfected with either miR-21 (nontargeting) or miR-30 (targeting) duplex RNA, and the repression levels were calculated. The nontarget control construct lacks miR-30 target sites. Error bars represent calculated values for standard deviation. (C) Analysis of c-myc IRES-containing constructs. (Left panel) Schematic depiction of c-myc IRES RLuc mRNAs. A stem–loop (SL) was inserted adjacent to the 5′ end of the cap-SL-myc-IRES-p(A) construct, and ∼200 nt of vector sequence separates the SL and IRES. (Right panel) Repression levels for individual reporters. Transfections were performed in triplicate in three separate experiments.

FIGURE 2.

miR-30 mediated repression of Pol II-driven reporter constructs. (A) Raw RLU values for the wild-type and mutant HCV IRES Pol II constructs. (B) Expression of miR-21 (upper panel) and miR-30 (lower panel) was evaluated by primer extension analysis. miR-21 or miR-30 expression plasmids were transfected into 293T cells, and primer extension was performed on RNAs harvested 24 h later. No miRNA was expressed in mock-transfected samples. (C) 293T cells were co-transfected with miR-21 or miR-30 plasmids and the indicated reporter construct. Cells were harvested for analysis at the indicated time points. (Darker shaded bars) Time points used for characterization of target mRNA integrity. A plasmid encoding FLuc was used as a transfection control. (D) Quantitative RT-PCR (qRT-PCR; upper panel) for RLuc mRNA was normalized to endogenous GAPDH mRNA levels. Target mRNAs co-transfected with miR-21 are set to 100%. Northern blots (lower panel) for RLuc mRNAs were performed as described in the Materials and Methods. rRNAs from corresponding samples are shown for loading control. qRT-PCR experiments were repeated on at least three separate occasions.

FIGURE 3.

miR-155 represses cap- and IRES-driven targets containing the authentic BACH1 3′ UTR. (A) Schematic of reporter constructs containing the BACH1 3′ UTR with four predicted miR-155 sites (black ovals). The sequence of mature miR-155 is shown with the seed sequence in capital letters. (B) Expression of miR-155 was evaluated in 293T cells by primer extension. (C) miR-155 repression levels of the indicated constructs. (D) qRT-PCR (upper panel) and Northern blot (lower panel) analyses were performed. (*) Background band consistently observed with the nontarget control.

FIGURE 4.

Changes in PABP expression and activity modulate miRNA repression. (A) miR-30-mediated repression was evaluated in 293T cells transfected with increasing amounts of myc-PABP expression plasmid. The repression level in cells transfected with control vector and miR-30 plasmids was set to 100% and then compared with repression data from cells similarly transfected with miR-30 and PABP (wild-type and M161A) expression plasmids. The total amount of transfected cDNA was kept constant between samples. Western blot analyses were carried out to determine PABP and tubulin levels. Myc-tagged PABP is visible as a distinct band that migrates slightly above endogenous PABP. (B) Northern blot analysis of target mRNA after myc-PABP overexpression. (C) miR-30 activity was evaluated in the context of Paip2 overexpression. Cells were transfected with increasing amounts of Flag-Paip2 plasmid, and repression of miR-30 target mRNA was examined as in A. Flag-Paip2 migrates slightly above endogenous Paip2. (D) miR-30 repression levels during knockdown of Paip2 along with Western blots for Paip2 and tubulin are shown. (E) miR-155 repression of the BACH1 3′ UTR reporter upon PABP, PABP (M161A), and Paip2 overexpression. (F) qPCR of the indicated miRNAs after control or myc-PABP expression plasmid transfection. This experiment was repeated three times, and a representative experiment is shown.

FIGURE 5.

miR-21 targeting of endogenous PDCD4 is ablated when active cellular PABP is increased. 293T cells were transfected with control or Paip2 siRNAs and then transfected with miR-21 or miR-30 ∼20 h later (left). After another 20 h incubation, cells were lysed and protein levels evaluated by Western blot. Effects of PABP overexpression were similarly evaluated (right). Quantification of PDCD4 normalized to tubulin levels in Western blots is shown (lower panels).

FIGURE 6.

The effect of miR-30 on target mRNA adenylation status after manipulation of active PABP. The LM-PAT assay was used to test poly(A) tail length (see Materials and Methods for details). 293T cells were transfected under conditions described in Figure 4, A or D. (A) The effects of PABP and PABP (M161A) ectopic overexpression on the poly(A) tail length of the miR-30 8× target mRNA 20 h post-transfection. (B) miR-30 8× target mRNA assayed as in A at 12 h post-transfection. (C) The adenylation status of a control mRNA (GAPDH) was assayed with the highest amount (800 ng) of PABP and PABP (M161A) cDNA transfected at 20 h and 12 h post-transfection. (D) LM-PAT assays of miR-30 8× target mRNA (upper panel) and control mRNA (GAPDH; lower panel) after knockdown of Paip2.