A more efficient RNAi inducible system for tight regulation of gene expression in mammalian cells and xenograft animals

Abstract

Two types of tetracycline-controlled inducible RNAi expression systems have been developed that generally utilize multiple tetracycline operators (TetOs) or repressor fusion proteins to overcome the siRNA leakiness. Here, we report a novel system that overexpresses the tetracycline repressor (TetR) via a bicistronic construct to control siRNA expression. The high level of TetR expression ensures that the inducible promoter is tightly bound, with minimal basal transcription, allowing for regulation solely dependent on TetR rather than a TetR fusion protein via a more complicated mechanism. At the same time, this system contains only a single TetO, thus minimizing the promoter impairment occurring in existing systems due to the incorporation of multiple TetOs, and maximizing the siRNA expression upon induction. In addition, this system combines all the components required for regulation of siRNA expression into a single lentiviral vector, so that stable cell lines can be generated by a single transduction and selection, with significant reduction in time and cost. Taken together, this all-in-one lentiviral vector with the feature of TetR overexpression provides a unique and more efficient tool for conditional gene knockdown that has wide applications. We have demonstrated the high degree of robustness and versatility of this system as applied to several mammalian cells and xenograft animals.

FIGURE 1.

The Tet-inducible siRNA expression system. (A) Schematics of the wild-type (a) and inducible mouse U6 promoters with one (b) or more copies of TetO sequences engineered around the TATA box, PSE, and DSE regions (c–f). (B) Schematic representation of the lentivector pSD31 for constitutive siRNA expression and pSD400 for conditional siRNA expression. pSD31 was modified from the pHIV-7 vector by introducing a BamHI site downstream of the 5′ -LTR for the cloning variant siRNA expression cassette. pSD400 was modified from pSD31 by deletion of the SV40-Puror sequence and introduction of a BglII-CMV-TetR-IRES-Puror-BglII cassette at the BamHI site. In this cassette a CMV promoter drives the bicistronic expression unit comprising TetR and the antibiotic puromycine resistance gene through an internal ribosomal entry site (IRES). The inducible siRNA expression cassette is inserted at the 3′- LTR via a newly introduced BamHI site that replaced the original BglII site by mutagenesis. (C) Western blot analysis of p53 knockdown in MCF-7 cells by siRNAs expressed from multiple TetO-engineering inducible mouse U6 promoters (c–f). Those siRNA expression cassettes were stably transduced into MCF-7 cells via the lentivector pSD31. There was no apparent p53 knockdown in either the presence (+) or absence (−) of Dox, indicating that multiple TetOs in the U6 promoter ruined its transcription activity.

FIGURE 2.

Taqman (A) and Western blot (B) analysis of TetR over-expression in stably transduced MCF-7 cells with the empty pSD400. The T-Rex cell line (Invitrogen) stably expresses TetR. Compared with T-Rex cells, the TetR expression in stably transduced MCF-7 cells was ∼15-fold higher on the mRNA level and ∼50-fold higher on the protein level.

FIGURE 3.

Western blot analysis of conditional siRNA expression against p53 in stably transduced MCF-7 cells cultured in the presence and absence of Dox (1 mg/mL). pSD31-p53siRNA (WT) was a constitutive siRNA expression vector with a wild-type mU6 promoter (Fig. 1A, a), serving as the positive control. pSD400-b was an inducible vector with a single TetO-modified mU6 promoter (Fig. 1A, b). pSD31-CTNLsiRNA (CNTL) was a negative control vector that constitutively expressed a scrambled siRNA. The pSD31-p53siRNA (WT) vector caused nearly complete silencing of p53 expression at the protein level in both the presence (+) and absence (−) of Dox. In the absence of Dox, the pSD400-b vector had little effect on p53, with its expression level at almost the same as that in negative control (CNTL) transduced cells. Addition of Dox caused a significant reduction in p53 protein expression by pSD400-b, reaching a level almost similar to that from the positive control cells. Here, actin was an internal protein for Western blot analysis.

FIGURE 4.

Taqman (A) and Western blot (B) analysis of the dose dependence of Dox on siRNA induction and target gene p53 knockdown. The p53 mRNA levels in transduced MCF-7 with various concentrations of Dox were normalized to that from nontransduced MCF-7 cells. GAPDH is an internal control for Western blot analysis. (C) Comparison of p53 knockdown by wild-type vector pSD31-a, inducible vector pSD400-b, and a constitutive vector containing the inducible cassette pSD31-b. The p53 protein knockdowns were quantitatively normalized to negative control siRNA (CNTL).

FIGURE 5.

The time-courses of siRNA expression (A) and repression (B) upon Tet induction and withdrawal. The stably transduced MCF-7 cells were first allowed to grow for 1 wk in Tet-free medium and then Dox (250 ng/mL) was added to induce siRNA expression for 0–8 d. After Dox was removed by changing into fresh medium, monitoring the p53 mRNA levels continued for 0–6 d. The p53 mRNA levels from various days were normalized to initial p53 without Dox addition.

FIGURE 6.

Single cell clones. (A) Taqman analysis of the p53 mRNA levels in six single cell clones from transduced MCF-7 pool cells that were cultured in the presence and absence of 1000 ng/mL Dox for 3 d. The data were reported relative to the p53 mRNA level in scrambled siRNA-transduced MCF-7 cells. (B) Western blot analysis of the p53 protein in single cell clones 9 and 10, which exhibited less siRNA leakiness under noninduced conditions and significant induction under induced conditions based on Taqman. (C) Western blot analysis of the Dox effect on TetR expression in single cell clones 9 and 10 to confirm that Dox has no effect on TetR expression.

FIGURE 7.

Conditional cancer target mTor silencing and tumor repression by the all-in-one inducible siRNA system in xenograft mice models. (A) Response of early-staged tumors to Dox induction. Eighteen and 22 nude mice were separately implanted with pSD400-CNTL- or pSD400-mTor-transduced PC3-MC6-Luc cells, 52×210⁶ cells per each mouse, and then 2.0 mg/mL Dox-containing water was given to half of the different xenograft mice, with the remaining given regular water to drink for 1 mo. No significant tumor repression phenotype was observed for mice with the control siRNA system in either the presence or absence of Dox and for mice with mTOR siRNA but drinking regular water, while 100% tumor repression was seen for mTOR siRNA-containing mice that drank Dox water, with 45% become tumor free. (B) Taqman analysis of the mTOR mRNA level in xenograft tumors. Twenty mice with later-staged tumors were divided into five groups with four in each group and were dosed with Dox at 0, 0.02, 0.1, 0.5, and 2 mg/mL in drinking water. Tumor samples were taken out 3 d later for Taqman analysis as previously described (Ke et al. 2006). The expression levels of mTOR mRNA in various xenograft tumors were normalized against that in noninduced control xenograft tumors.