• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Feb 14, 2006; 103(7): 2252–2256.
Published online Feb 6, 2006. doi:  10.1073/pnas.0511034103
PMCID: PMC1413752
Genetics

A transgenic approach for RNA interference-based genetic screening in mice

Abstract

Genetic screening is the most powerful method through which to uncover gene function. It has been applied very successfully in lower organisms but seldom attempted in mammalian species because of their long generation time. In this study, we exploit RNA interference (RNAi) for its potential use in genetic screening in mice. We show that RNAi-induced gene knockdown can be generated through introducing small hairpin RNA-expressing constructs into the mouse as transgenes via conventional pronuclear injection. The knockdown effect can be transmitted for many generations in these transgenic animals. In a small-scale screening for developmental defects in the kidney, we uncovered a potential role of Id4 in the formation of the renal medulla. Our results demonstrate the feasibility of using RNAi for genetic screening in mice.

Keywords: Id4, p57kip2, cyclin D1, kidney

RNA interference (RNAi) induced by short dsRNA has gained widespread application in biological research (13). The discovery of RNAi was rooted in the desire to alter gene function by an antisense approach in Caenorhabditis elegans. It was observed that introducing dsRNA homologous to a specific gene resulted in the posttranscriptional silencing of that gene (4). This dsRNA-induced gene silencing was termed RNAi. It occurs through two main steps. The dsRNA is initially recognized by an enzyme of the RNase III family of nucleases, named Dicer, and processed into small double-stranded molecules, termed small interfering RNA (siRNA) (5). siRNA is then incorporated into an inactive large multisubunit nuclease complex, RISC (RNA-induced silencing complex). RISC unwinds the siRNA in an ATP-dependent step and finds homologous target mRNAs by using the siRNA sequence as a guide and cleaves these mRNAs (6, 7).

The RNAi effect can be induced in mammalian cell cultures via transient transfection of double-stranded siRNA (1) or stable expression of small hairpin RNA (shRNA), which is processed to siRNA inside cells (8). This technical advancement has made it possible to perform loss-of-function genetic screens in cultured cells by using shRNA libraries (9, 10). However, there are limitations to the types of biological processes that can be studied in cultured cells. Many genes function in processes that cannot be recapitulated in cultured cells. Therefore, it is desirable to apply RNAi technology in the whole mouse, preferably in semihigh through-put fashion as has been done so successfully in C. elegans (11, 12).

Transgenic RNAi mice have been produced in two ways. One way has been to first generate mouse ES cell clones that show RNAi effect and then to use the ES cell clones to produce mice (13). The other is to infect or inject one-cell mouse embryos with lentiviruses carrying a shRNA-expressing cassette (14). However, both methods are not easily adaptable for routine laboratory uses, not to mention large-scale genetic screens. Here, we report the production of transgenic RNAi mice through conventional pronuclear injection of shRNA-expressing constructs. We performed a small-scale RNAi screen in mice aimed at identifying genes whose function are important for the development of the kidney.

Results

To make the generation of transgenic RNAi mice easier and faster for its potential use in large-scale genetic screens, we designed and constructed a transgenic RNAi vector (pTshRNA, Fig. 1A). The expression of shRNA from the vector is driven by the human H1 promoter (8). The genomic sequence (0.7 kb) of the mouse H1 gene 3′ to its termination signal was placed downstream of the shRNA sequence. shRNA-encoding sequences were cloned into the vector as synthetic linkers. An EGFP expression cassette was also included in the vector to mark embryos that have received the transgenic DNA. The shRNA expression cassette and the EGFP cassette were purified away from the plasmid backbone and injected into one-cell mouse embryos. Fig. 1B shows a transgenic embryo that expresses GFP.

Fig. 1.
Generation of the transgenic shRNA vector. (A) Diagram of pTshRNA. (B) A GFP-positive embryo produced by using pTshRNA.

To test the efficacy of pTshRNA in knocking down gene expression, we generated an shRNA construct to target p57KIP2 (15). Embryos injected with the construct were harvested and analyzed at embryonic day 18.5 without establishing transgenic lines first. Two GFP-positive embryos were obtained that did not show any gross abnormalities. p57 is expressed in a number of tissues during mouse development (15, 16). We analyzed its expression in the nontransgenic littermates and the GFP-positive embryos by immunofluorescence staining. The expression of p57 was reduced in the eye lens (Fig. 2A), the kidney (Fig. 2B), and other tissues (data not shown) of the transgenic embryos. In the kidney, the reduction in expression was so strong that the expression of p57 was undetectable with the immunostaining method. A similarly strong reduction of p57 expression was seen in glomeruli as well (data not shown). However, no reduction of expression in the differentiating chondrocytes was observed (Fig. 2C). It is unclear whether this lack of RNAi effect was caused by the inability of chondrocytes to perform RNAi because of missing key components of the process or a result of failed expression of the shRNA. These cells did express GFP (data not shown), however, indicating the transgenes were not shut down in these cells. Similar results were obtained in both GFP-positive embryos. Next, we targeted another cell cycle-related gene, cyclin D1, for transgenic knockdown. As shown in Fig. 3, there was a clear reduction in cyclin D1 expression in the transgenic retina. These results demonstrate that shRNA expression constructs delivered by the conventional transgenic technique can produce RNAi effect in mice.

Fig. 2.
Silencing the expression of p57KIP2 with transgenic shRNA. Sections from embryonic day 18.5 transgenic and nontransgenic embryos were immunostained for the expression of p57 in the ocular lens (A), the medulla of the kidney (B), and the costal bone ( ...
Fig. 3.
Reducing cyclin D1 expression with transgenic shRNA. Sections from embryonic day 18.5 transgenic and nontransgenic (Non-TG) embryos were immunostained for the expression of cyclin D1 in the retina. (Magnifications: ×400.)

To determine whether the RNAi effect observed in the injected embryos can be transmitted through germ line, we established a transgenic line that carried the p57 shRNA construct and examined the expression of p57 in the embryos derived from this line. We found a similar knockdown of p57 expression in this transgenic line as in the transgenic embryos obtained transiently. To assess the RNAi effect more quantitatively, we dissected selected tissues out of embryonic day 18.5 embryos derived from a cross between a transgenic male and a nontransgenic female, and the expression of GFP and p57 in the tissues was analyzed by Western blotting. Significant reductions in the expression of p57 were seen in the skeletal muscle, lung, and small intestine (Fig. 4). The reduction was less pronounced in the adrenal gland (Fig. 4). We have bred the p57 shRNA transgenic line up to four generations and have not observed a reduction in the RNAi effect. Despite the observed RNAi effect, we did not observe any developmental defects in either p57 or cyclin D1 shRNA transgenic mice as seen in the respective knockout mice (1618), indicating that the residual amount of these two genes’ product is sufficient to support the normal developmental processes in which they are involved.

Fig. 4.
Western blot analysis of p57KIP2 expression in embryonic tissues. Day 18.5 embryos were obtained from a mating between a p57 shRNA transgenic male and a wild-type female mouse. Non-TG, nontransgenic; TG, transgenic.

Having demonstrated the feasibility of generating transgenic RNAi mice with conventional technology, we next wanted to perform genetic screens to identify genes that are required for developmental processes involved in the formation of the kidney. To avoid the cost of establishing and maintaining transgenic lines, we used F0 embryos obtained directly from pronuclear injections. As a positive control for this approach, we chose to target Pax6, because it is haploinsufficient for development. A reduction of 50% in its expression through transgenic RNAi should produce the phenotypes seen in Pax6+/− mice. The phenotypes include small eyes and a reduced body size (19). As expected, the transgenic Pax6 RNAi embryo harvested at 18.5 days after the injection of the shRNA construct displayed a smaller body size (Fig. 5A) and the eyes were smaller as well (Fig. 5B). Histological analysis of the eye also indicated defects in the lens (data not shown).

Fig. 5.
Developmental defects caused by shRNA-mediated gene silencing. (A) Reduced body size of the Pax6 shRNA transgenic embryo. (B) Small lens in the same transgenic embryo shown in A. (C) Failed elongation of the medulla of the kidney in the Id4 shRNA transgenic ...

Target genes for screening were chosen from our microarray experiment that identified genes that are highly expressed in the developing kidney (P.Z., unpublished results). Two transgenic shRNA constructs targeting different regions of a gene were generated for each gene (see Table 1 for the name of the genes and the target sequences), and each construct was injected once (into ≈50 one-cell embryos). Around 30% of the embryos survived the injection, and 10% of the survivors were transgenic. The transgenic embryos were identified under an epi-fluorescent dissecting microscope. GFP-negative (as control) and GFP-positive embryos from the same litter were fixed and processed for histological examination. If no GFP-positive embryos for both constructs of a gene were obtained, there could be two reasons. One could simply be that the injection was unsuccessful. The other could be that knocking down the expression of the gene is early embryonic lethal. Therefore, the lines in which we did not see any GFP-positive embryos after the injection were reexamined in secondary screens with additional constructs. Those that showed a phenotype were also reexamined to confirm the phenotype with the original as well as new shRNA constructs and to demonstrate the reduction of expression.

Table 1.
Summary of transgenic shRNA screening

So far we have injected 30 constructs for 15 genes and analyzed 30 transgenic embryos. We found that the Id4 shRNA transgenic embryo showed a striking defect. The medulla of the kidney failed to elongate (Fig. 5C), a phenotype reminiscent of that of p57 knockout mice (16). No other abnormalities were noticed in the rest of the kidney or the other organ systems. Id4 is a member of the ID (inhibitor of DNA binding) protein family (20), which regulates cell proliferation and differentiation in many different developmental pathways by antagonizing basic helix–loop–helix protein function (21, 22). It regulates the proliferation and differentiation of neural progenitors and is required for the development of the central nervous system (23, 24). Only 20% of Id4−/− mice were able to survive to adulthood. The rest either died in uteri or postnatally (24). However, whether the kidney is defective in these animals was not examined. Thus, our transgenic RNAi screen identifies a potential role of Id4 in kidney development.

Discussion

The standard approach for loss-of-function studies in mice is to generate a knockout mouse in which the gene of interest is inactivated through homologous recombination in mouse ES cells. The recombinant ES clones are then mixed with host embryos at the blastocyst stage and develop along with the host embryo, generating chimeric mice. Subsequent breeding of the chimeras will result in mice derived from the ES clones and hence germ-line transmission of the knockout allele. Although this approach is very powerful, it suffers two drawbacks. One is that it is very time consuming. A knockout construct has to be built from genomic clones isolated from a suitable genomic library. The construction involves many steps of molecular manipulations. Even with our advanced cloning technology (25), the generation of a knockout construct could still be a rate-limiting step. Further, after the construct is made, there are still months of time required for ES cell culturing, clone identification, chimera production, and breeding for germ-line transmission. The other drawback of the knockout approach is that the so-produced mutant allele is null. Although the total loss of function of a gene is very powerful in revealing its function, it is not always the most informative, especially when it causes early embryonic lethality. In many circumstances, a hypomorphic allele is more informative. For example, a complete loss of BubR1 is early embryonic lethal, but a hypomorphic allele revealed that BubR1 is required for preventing premature aging (26), which could not have been achieved even with a conditional knockout. Transgenic RNAi may serve as the “hypomorphic allele.” In fact, one could generate a number of transgenic lines with different degrees of loss of expression/function, thus creating many hypomorphic alleles. Thus, transgenic RNAi is a quicker and more versatile approach complementary to the knockout in the study of gene function in mice. Furthermore, it is impractical to use gene knockout in genetic screens in mice, whereas RNAi could be adapted for that purpose, as it has been successfully done with cultured cells (27).

Since the discovery of RNAi (4), there had been some earlier attempts to induce RNAi with long dsRNA in mice (2830), but the success was limited to preimplanting embryos because of the protein kinase R (PKR) response in older embryos as seen in cultured cells. PKR is activated by dsRNA (31, 32), and the activated PKR causes activation of NF-κB and global gene silencing through blocking translation, leading to cell death by apoptosis (33). This response is now largely circumvented through the use of short dsRNA or siRNA (1), which made it possible to produce transgenic RNAi mice. Two methods were developed to generate such mice. In the first method, ES cell clones that show RNAi effect are generated and used to produce the mice (13). However, this method is almost as time-consuming as the conventional knockout. The second method is to infect or inject one-cell mouse embryos with lentiviruses carrying a shRNA-expressing cassette (14), which requires the production of high titer lentiviruses, a laborious process.

Using the conventional transgenic technique, we demonstrated that transgenic RNAi mice could be created just like any other transgenic mice. We have succeeded in using the method in a small-scale genetic screen and identified an additional role for Id4 in kidney development. Large-scale genetic screens aimed at identifying gene function in embryogenesis are feasible with this method. The effect of knocking down gene expression on development can be scored directly in the embryos injected with a construct. In essence, the mouse embryos are used in a way analogous to the cultured cells for phenotypic observations. No transgenic lines need to be established, which eliminates a significant amount of animal costs. Furthermore, conditional RNAi using the Cre-LoxP system (34) can be accomplished easily with our method without resorting to the lentiviral system.

Materials and Methods

Generation of pTshRNA.

pTshRNA is based on pSUPER (8). We added the 3′ region of the mouse H1 gene, which was amplified with PCR from 129 mouse genomic DNA. The PCR product was sequence-verified. An EGFP expression cassette in which EGFP (Clontech) is driven by PGK promoter and followed by simian virus 40 intron, and a poly(A) signal was placed 5′ to the H1 promoter but in the transcriptionally opposite orientation, an arrangement similar to the endogenous H1 gene where the H1 promoter overlaps with that of PARP-2 (35).

The shRNA was designed according to Brummelkamp et al. (8). Briefly, a cDNA sequence was scanned for the presence of AA(N)19TT and AA(N)19AA [or AA(N)19 NN if the former two could not be found] to choose a 19-nt sequence for RNAi targeting. Oligos were synthesized and annealed to generate linkers encoding shRNAs with the 19-nt sequence forming head-to-head repeats separated by a 9-nt (TTCAAGAGA) spacer. The oligo linkers were placed into the pTshRNA vector in the HindIII and BglII sites to generate transgenic shRNA constructs. Sequence-verified constructs were digested with NotI and PvuII, gel-purified, and used in pronuclear injections. For p57KIP2, we used 5′-CGACTTCTTCGCCAAGCGC-3′ as the RNAi targeting sequence, and for cyclin D1, we used 5′-GATGAAGGAGACCATTCCC-3′.

Generation of Transgenic Mice.

C57BL/6 female mice (3–4 weeks old) were superovulated and mated with stud C57BL/6 males. The females were killed the next morning, and the fertilized eggs were selected for the injection of DNA. The injected one-cell embryos were placed back in the ovaries of pseudopregnant CD1 female mice. The embryos were allowed to development to full term to produce transgenic founders or were harvested before term for direct phenotypic analyses.

Analysis of the Transgenic shRNA Embryos.

Embryos were harvested at desired developmental stages and processed for paraffin embedding and sectioning as described (36). Immunofluorescent staining and Western blotting analysis of p57 was performed according to Zhang et al. (16) by using anti-p57 antibody M-20 purchased from Santa Cruz Biotechnology. Anti-cyclin D1 antibody was from Zymed.

Acknowledgments

We thank Dr. R. Agami from The Netherlands Cancer Institute (Amsterdam) for providing the pSUPER plasmid, Dr. J. Jiang at Harvard Medical School for sharing reagents and ideas, and Dr. R. Zhang for her excellent work in microinjection. This work is supported by National Institutes of Health Grants EY014745 and DK063964 (to P.Z.).

Glossary

Abbreviations:

RNAi
RNA interference
siRNA
small interfering RNA
shRNA
small hairpin RNA.

Footnotes

Conflict of interest statement: No conflicts declared.

References

1. Elbashir S. M., Harborth J., Lendeckel W., Yalcin A., Weber K., Tuschl T. Nature. 2001;411:494–498. [PubMed]
2. Hannon G. J., Rossi J. J. Nature. 2004;431:371–378. [PubMed]
3. Hannon G. J. Nature. 2002;418:244–251. [PubMed]
4. Fire A., Xu S., Montgomery M. K., Kostas S. A., Driver S. E., Mello C. C. Nature. 1998;391:806–811. [PubMed]
5. Nykanen A., Haley B., Zamore P. D. Cell. 2001;107:309–321. [PubMed]
6. Tijsterman M., Plasterk R. H. Cell. 2004;117:1–3. [PubMed]
7. Martinez J., Patkaniowska A., Urlaub H., Luhrmann R., Tuschl T. Cell. 2002;110:563–574. [PubMed]
8. Brummelkamp T. R., Bernards R., Agami R. Science. 2002;296:550–553. [PubMed]
9. Paddison P. J., Silva J. M., Conklin D. S., Schlabach M., Li M., Aruleba S., Balija V., O’Shaughnessy A., Gnoj L., Scobie K., et al. Nature. 2004;428:427–431. [PubMed]
10. Berns K., Hijmans E. M., Mullenders J., Brummelkamp T. R., Velds A., Heimerikx M., Kerkhoven R. M., Madiredjo M., Nijkamp W., Weigelt B., et al. Nature. 2004;428:431–437. [PubMed]
11. Fraser A. G., Kamath R. S., Zipperlen P., Martinez-Campos M., Sohrmann M., Ahringer J. Nature. 2000;408:325–330. [PubMed]
12. Sonnichsen B., Koski L. B., Walsh A., Marschall P., Neumann B., Brehm M., Alleaume A. M., Artelt J., Bettencourt P., Cassin E., et al. Nature. 2005;434:462–469. [PubMed]
13. Carmell M. A., Zhang L., Conklin D. S., Hannon G. J., Rosenquist T. A. Nat. Struct. Biol. 2003;10:91–92. [PubMed]
14. Rubinson D. A., Dillon C. P., Kwiatkowski A. V., Sievers C., Yang L., Kopinja J., Rooney D. L., Ihrig M. M., McManus M. T., Gertler F. B., et al. Nat. Genet. 2003;33:401–406. [PubMed]
15. Matsuoka S., Edwards M. C., Bai C., Parker S., Zhang P., Baldini A., Harper J. W., Elledge S. J. Genes Dev. 1995;9:650–662. [PubMed]
16. Zhang P., Liegeois N. J., Wong C., Finegold M., Hou H., Thompson J. C., Silverman A., Harper J. W., DePinho R. A., Elledge S. J. Nature. 1997;387:151–158. [PubMed]
17. Yan Y., Frisen J., Lee M. H., Massague J., Barbacid M. Genes Dev. 1997;11:973–983. [PubMed]
18. Sicinski P., Donaher J. L., Parker S. B., Li T., Fazeli A., Gardner H., Haslam S. Z., Bronson R. T., Elledge S. J., Weinberg R. A. Cell. 1995;82:621–630. [PubMed]
19. Hill R. E., Favor J., Hogan B. L., Ton C. C., Saunders G. F., Hanson I. M., Prosser J., Jordan T., Hastie N. D., van Heyningen V. Nature. 1991;354:522–525. [PubMed]
20. Benezra R., Davis R. L., Lockshon D., Turner D. L., Weintraub H. Cell. 1990;61:49–59. [PubMed]
21. Yokota Y. Oncogene. 2001;20:8290–8298. [PubMed]
22. Ruzinova M. B., Benezra R. Trends Cell Biol. 2003;13:410–418. [PubMed]
23. Yun K., Mantani A., Garel S., Rubenstein J., Israel M. A. Development (Cambridge) 2004;131:5441–5448. [PubMed]
24. Bedford L., Walker R., Kondo T., van Cruchten I., King E. R., Sablitzky F. Dev. Biol. 2005;280:386–395. [PubMed]
25. Zhang P., Li M. Z., Elledge S. J. Nat. Genet. 2002;30:31–39. [PubMed]
26. Baker D. J., Jeganathan K. B., Cameron J. D., Thompson M., Juneja S., Kopecka A., Kumar R., Jenkins R. B., de Groen P. C., Roche P., van Deursen J. M. Nat. Genet. 2004;36:744–749. [PubMed]
27. Westbrook T. F., Martin E. S., Schlabach M. R., Leng Y., Liang A. C., Feng B., Zhao J. J., Roberts T. M., Mandel G., Hannon G. J., et al. Cell. 2005;121:837–848. [PubMed]
28. Wianny F., Zernicka-Goetz M. Nat. Cell Biol. 2000;2:70–75. [PubMed]
29. Svoboda P., Stein P., Hayashi H., Schultz R. M. Development (Cambridge) 2000;127:4147–4156. [PubMed]
30. Svoboda P., Stein P., Schultz R. M. Biochem. Biophys. Res. Commun. 2001;287:1099–1104. [PubMed]
31. Meurs E., Chong K., Galabru J., Thomas N. S., Kerr I. M., Williams B. R., Hovanessian A. G. Cell. 1990;62:379–390. [PubMed]
32. Minks M. A., West D. K., Benvin S., Baglioni C. J. Biol. Chem. 1979;254:10180–10183. [PubMed]
33. Gil J., Esteban M. Apoptosis. 2000;5:107–114. [PubMed]
34. Ventura A., Meissner A., Dillon C. P., McManus M., Sharp P. A., Van Parijs L., Jaenisch R., Jacks T. Proc. Natl. Acad. Sci. USA. 2004;101:10380–10385. [PMC free article] [PubMed]
35. Ame J. C., Schreiber V., Fraulob V., Dolle P., de Murcia G., Niedergang C. P. J. Biol. Chem. 2001;276:11092–11099. [PubMed]
36. Zhang P., Wong C., Liu D., Finegold M., Harper J. W., Elledge S. J. Genes Dev. 1999;13:213–224. [PMC free article] [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...