Overexpression of exportin 5 enhances RNA interference mediated by short hairpin RNAs and microRNAs

Abstract

Plasmids or viral vectors that express short hairpin RNAs (shRNAs) have emerged as important tools for the stable inhibition of specific genes by RNA interference. shRNAs are structural and functional homologs of pre-microRNAs, intermediates in the production of endogenously encoded microRNAs (miRNAs). Therefore, overexpressed shRNAs could inhibit miRNA function by competing for a limiting level of one or more factors involved in miRNA biogenesis or function. Here, we demonstrate that overexpressed shRNAs can saturate the activity of endogenous Exportin 5, a factor required for nuclear export of both shRNAs and pre-miRNAs. While shRNA overexpression can therefore inhibit miRNA function, simultaneous overexpression of Exportin 5 reverses this effect. Moreover, Exportin 5 overexpression can significantly enhance RNA interference mediated by shRNAs. These data have implications for the future clinical utilization of shRNAs and also provide a simple method to enhance RNA interference by shRNAs in culture.

FIGURE 1.

Exp5 overexpression enhances RNAi mediated by overexpressed miR-30a miRNA. 293T cells (24-well plates) were cotransfected with 10 ng of the pCMV-Luc-miR-30(AP) indicator plasmid, 1 ng of the pRL-CMV R-luc internal control, 100 ng of the pCMV-miR-30 effector plasmid, and 50 ng of pKmyc-Exp5 or 100 ng of pCK-Drosha or pK-T7-Dicer. The parental pBC12/CMV or pKmyc plasmids served as negative controls. At 40 h after transfection, the cells were lysed and aliquots used for Western analysis (A) or for assay of F-luc and R-luc activity (B). The endogenous Tap protein served as a loading control in panel A. The data in panel B were compiled from three experiments and were corrected for minor variations in the R-luc internal control. Data are presented relative to the culture transfected with pCMV-Luc-miR-30(AP) in the absence of any miRNA or protein expression plasmid (Pos).

FIGURE 2.

Exp5 overexpression enhances endogenous miRNA but not siRNA function. (A) 293T cells were transfected with pCMV-Luc (10 ng), pRL-CMV (1 ng), 100 nM of a synthetic siRNA specific for F-luc, and 50 ng of pKmyc-Exp5 or the parental pKmyc plasmid. An siRNA specific to the HIV-1 tat gene served as a negative control. At 40 h, cells were harvested and F-luc and R-luc activities determined. Data are presented relative to the culture transfected with pCMV-Luc alone (Pos). The average of three independent transfections with standard deviation is indicated. (B) 293T cells were transfected with pCMV-Luc or with pCMV-Luc-miR-30(AP) (10 ng), with the pRL-CMV internal control (1 ng) and with pKmyc-Exp5 or the parental pKmyc plasmid (50 ng). At 40 h, cells were lysed and F-luc and R-luc activities determined. Data are presented relative to the culture transfected with the pCMV-Luc or pCMV-Luc-miR-30(AP) indicator plasmid in the absence of Exp5 overexpression and were corrected for the R-luc internal control. The average of three experiments with standard deviation is indicated.

FIGURE 3.

Exp5 overexpression rescues the inhibition of miRNA function caused by an shRNA expression plasmid. (A) Cells were transfected with pCMV-Luc-miR-30(AP) (10 ng), pRL-CMV (1 ng), pCMV-miR-30 or the parental pBC12/CMV plasmid (100 ng), 100 ng of pSuper-CCR-5 or pSuper, and 50 ng of pKmyc-Exp5 or pKmyc. At 40 h, cells were harvested and F-luc and R-luc activities determined. Data are presented as in Figure 2A, where Pos represents the culture transfected with pCMV-Luc-miR-30(AP) together with the pBC12/CMV, pKmyc, and pSuper negative control plasmids. (B) Same as panel A, except that 293T cells were transfected with the pCMV-Luc-miR-21(P) indicator plasmid and the pCMV-miR-21 human miR-21 expression plasmid.

FIGURE 4.

Overexpression of Exp5 enhances expression of pre-miR-30 and mature miR-30. 293T cells were transfected as described in Figure 1. At 40 h, wells were pooled into two equal fractions and total cell or cytoplasmic RNA isolated and subjected to Northern analysis. The relative mobility of the 63-nt pre-miR-30 RNA and 23-nt mature miR-30a miRNA are indicated. 5S rRNA served as a loading control.

FIGURE 5.

Overexpression of Exp5 improves RNAi mediated by an shRNA expression plasmid. 293T cells were mock-transfected or transfected with pSuper-Tat-SF1 (300 ng) and 100 ng of pKmycExp5 or the parental pKmyc plasmid. At 40 h and at 72 h after transfection, individual cell cultures were lysed and subjected to Western blot analysis using antisera specific for the endogenous Tat-SF1 or Tap proteins.

FIGURE 6.

Analysis of Exp5 protein or mRNA expression levels. (A) Lysates were prepared from a variety of cell lines, normalized by total protein concentration and 50 μg of each preparation analyzed by Western blot using a polyclonal antibody against full-length human Exp5 or a monoclonal antibody specific for tubulin. Abbreviations and tissue origins are as follows: REF, rat embryo fibroblasts; MD, MDCK canine kidney; 293T, human embryonic kidney; BHK, baby hamster kidney; HeLa, human cervical carcinoma; 3T3, NIH 3T3 murine fibroblasts; COS, African green monkey kidney; RW, RWPE normal human prostate; PC-3, human prostate cancer; BT, BT474 human breast cancer; T47, T-47D human breast cancer; C42, human prostate cancer; LnC, LnCAP human prostate carcinoma; M10, MCF10A human breast epithelial cells; M7, MCF7 human breast cancer. (B) Semiquantitative RT-PCR analysis of Exp5 mRNA expression in human tissues. This experiment utilized a commercially prepared, prenormalized human multiple tissue cDNA panel. A similar level of a 293T cell cDNA sample was used as a positive control and to establish amplification conditions that gave rise to levels of the amplified cDNA fragment that were linearly related to the level of input cDNA. A twofold dilution series, using cDNA prepared from poly(A)⁺ 293T cell RNA, is shown.