Reversible gene knockdown in mice using a tight, inducible shRNA expression system

Abstract

RNA interference through expression of short hairpin (sh)RNAs provides an efficient approach for gene function analysis in mouse genetics. Techniques allowing to control time and degree of gene silencing in vivo, however, are still lacking. Here we provide a generally applicable system for the temporal control of ubiquitous shRNA expression in mice. Depending on the dose of the inductor doxycycline, the knockdown efficiency reaches up to 90%. To demonstrate the feasibility of our tool, a mouse model of reversible insulin resistance was generated by expression of an insulin receptor (Insr)-specific shRNA. Upon induction, mice develop severe hyperglycemia within seven days. The onset and progression of the disease correlates with the concentration of doxycycline, and the phenotype returns to baseline shortly after withdrawal of the inductor. On a broad basis, this approach will enable new insights into gene function and molecular disease mechanisms.

Figure 1.

Schematic representation of the inducible system and gene targeting strategy. (A) Principle of tetR-mediated control of shRNA expression. Transcription of the RNA polymerase III-dependent promoter is blocked in cells expressing the tet repressor (tetR). Upon induction by doxycycline, tetR is removed from the tet-operator sequences (tetO) inserted into the promoter, allowing transcription of shFluc. ShRNA expression leads to RNAi-mediated knockdown of the target gene firefly luciferase. Renilla luciferase is used for reference to quantify firefly luciferase activity. (B) Scheme of the targeting strategy. ShRNA and reporter constructs were independently inserted into the rosa26 locus by homologous recombination in ES cells. Genes encoding the Renilla (Rluc) and firefly luciferases (Fluc) along with an adenovirus splice acceptor sequence and a polyadenylation signal (pA) were placed downstream of the endogenous rosa26 promoter. The Fluc-specific shRNA is expressed under the control of the U6-tet or H1-tet promoter, and terminated by five thymidines (shRNA). The loxP-sites flanking the shRNA expression cassettes were used to generate a negative control through cre-mediated recombination in ES cells. (C) Southern blot analysis of genomic DNA from transfected ES cell clones containing the shRNA- (lane #1 and #2) or the reporter-constructs (lanes #3 and #4). Homologous recombination at the rosa26 locus is detectable by using EcoRV-digested genomic DNA and probe 1, resulting in a 11.7 kb band for the wt and a 2.5 kb band for targeted allele. E: EcoRV; X: XbaI; neo: FRT-flanked neomycin resistance gene; hyg: FRT-flanked hygromycin resistance gene. (D) Western blot analysis from protein extracts of ES cells expressing either the wt tetR or the itetR using tetR- or β-Actin-specific antiserum. Control: protein extracts from wt ES cells.

Figure 2.

Efficiency of H1/U6-shRNA-mediated firefly luciferase (Fluc) knockdown in mice expressing the codon-optimized tetR. Each configuration (control, H1-tet shRNA, U6-tet shRNA) was analyzed using two to four mice at the age of 8–10 weeks, respectively. Percentages of shRNA-mediated repression of firefly luciferase activity with standard error of the mean are shown for untreated controls (gray bars) and in animals upon 10 days feeding with 2 mg/ml doxycycline in the drinking water (white bars). In negative control animals (black bars), the shRNA expression cassettes are removed through cre-mediated recombination. Relative values of relative firefly luciferase expression level in different organs were calculated by using Renilla luciferases activities for reference. (A) H1-tet promoter-driven shRNA expression. (B) shRNA expression through the U6-tet promoter.

Figure 3.

Single vector configuration for INSR gene knockdown in vivo. (A) RMCE through Flp^e-mediated recombination using the exchange vector generates the rosa26(RMCE exchanged) allele. The exchange vector carries the shRNA coding region under the control of the H1-tet promoter, the codon-optimized tetRi gene under the control of the CAGGS promoter and a truncated neo^R gene for positive selection of clones upon successful RMCE. The insr-specific shRNA sequence IR5 is shown in the figure. (B) Detection of processed IR5-shRNA. shRNA-transgenic ES cell clone #1 (IR5) and control cells transfected with an RMCE exchange vector lacking the shRNA-expression cassette (control) were cultured in the absence and/or presence of 1 µg/ml doxycycline as indicated. RNA extracts were analyzed by northern blot using IR5-shRNA and 5S-rRNA-specific oligonucleotide probes. (C) Doxycycline dose response of INSR knockdown in mice. IR5 transgenic animals were treated in the absence or presence of 2 μg/ml, 20 μg/ml or 2 mg/ml doxycycline for 10 days, as indicated. Protein extracts were prepared from various tissues and subjected to western blot analysis using INSR- or AKT-specific antiserum. Control: protein extracts from animals carrying a firefly luciferase-specific shRNA. (D) Serum glucose levels in shRNA-transgenic mice treated with 2 μg/ml (triangles), 20 μg/ml (bars) or 2 mg/ml doxycycline in the drinking water (circles) for the indicated number of days. Control animals (rectangles) carry a firefly luciferase-specific shRNA construct. Serum glucose levels ± SEM are shown. All assays were performed using groups of six mice at the age of two months.

Figure 4.

Reversible induction of hyperglycemia in mice. A group of six 2-month old, shIR5-transgenic mice were fed for 10 days with 20 µg/ml doxycycline in the drinking water and subsequently kept in the absence of doxycycline for the following 21 days. (A) Blood glucose levels were determined in venous blood samples at the indicated day of treatment. (B) Serum insulin concentrations were measured as indicated. Each bar represents the mean value ± SEM. (C) Glucose tolerance test were performed using IR5-transgenic mice before (triangles) and 21 days after dox treatment (rectangles). Results are expressed as mean blood glucose concentration ± SEM from at least six animals of each group. (D) Reversible knockdown of the insulin receptor using 20 µg/ml doxycycline for 10 days and 21 days after removal of dox in two independent mice. Protein extracts prepared from liver were subjected to western blot analysis using an INSR-specific antiserum and an AKT-specific antiserum.