Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Jan 16, 2007; 104(3): 1027–1032.
Published online Jan 5, 2007. doi:  10.1073/pnas.0610155104
PMCID: PMC1764220

Controlled expression of transgenes introduced by in vivo electroporation


In vivo electroporation is a powerful technique for the introduction of genes into organisms. Temporal and spatial regulation of expression of introduced genes, or of RNAi, would further enhance the utility of this method. Here we demonstrate conditional regulation of gene expression from electroporated plasmids in the postnatal rat retina and the embryonic mouse brain. For temporal regulation, Cre/loxP-mediated inducible expression vectors were used in combination with a vector expressing a conditionally active form of Cre recombinase, which is activated by 4-hydroxytamoxifen. Onset of gene expression was regulated by the timing of 4-hydroxytamoxifen administration. For spatial regulation, transgenes were expressed by using promoters specific for rod photoreceptors, bipolar cells, amacrine cells, Müller glia or progenitor cells. Combinations of these constructs will facilitate a variety of experiments, including cell-type-specific gene misexpression, conditional RNAi, and fate mapping of progenitor and precursor cells.

Keywords: Cre, retina, brain, development, RNAi

Gain-of-function and loss-of-function studies using transgenic animals have greatly advanced our understanding of the molecular and cellular mechanisms of development and disease. In particular, conditional gene activation and inactivation have been powerful methods for studies of gene function and for the labeling and manipulation of specific cell populations in vivo (1, 2). Site-specific recombination systems (Cre/loxP and Flp/FRT) are widely used to control gene expression in transgenic mice. By crossing two transgenic lines, one expressing Cre (or Flp) under tissue-specific and/or inducible control and the other carrying two loxP (or FRT) sites, DNA recombination can lead to inducible gene expression or loss of gene expression in restricted tissues at specific times (3). Although such strategies are very useful, they are time-consuming and cannot be applied to species that are difficult to manipulate genetically.

In vivo electroporation is a convenient technique for the introduction of genes into a variety of animals, including mouse, rat, and chick (4, 5). We previously reported that plasmid DNAs can be easily delivered to developing mouse/rat retinas by in vivo electroporation (6). The plasmid DNAs, electroporated from the scleral side of the postnatal day 0 (P0) retina, are preferentially introduced into retinal progenitor/precursor cells, which give rise to four different cell types; rod photoreceptors in the outer nuclear layer (ONL), bipolar and amacrine cells in the inner nuclear layer (INL), and Müller glial cells, which extend radial processes spanning the entire thickness of the retina [see supporting information (SI) Fig. 7]. The efficiency of electroporation into the developing postnatal retina is quite good, and transgene expression persists at least for a few months. Moreover, compared with other gene transfer methods, such as viral vectors, in vivo electroporation has several advantages. First, various types of DNA constructs, including RNAi vectors, are readily introduced to the retina without DNA size limitation. Second, more than two different DNA constructs can be introduced into the same cells at once. Despite such advantages, however, the means to conditionally control the expression of transgenes after in vivo electroporation have not been established, limiting the applications of this method.

We have adapted existing methods for the conditional expression of transgenes or small hairpin RNA (shRNA) from electroporated plasmids in the developing rat retina and mouse brain by applying the Cre/loxP system and/or cell-type-specific promoters. We demonstrate how these methods can be applied to studies of development, for which temporal and/or cell-type-specific expression is required.


Temporal Regulation Using Inducible Cre Recombinase.

To temporally regulate transgene expression in the retina, the conditionally active forms of Cre and Flp recombinases, CreERT2 (7), ERT2Cre, ERT2CreERT2 (8), and FlpeERT2 (9) [ERT2, mutated ligand-binding domain of estrogen receptor (ER)] were used. These recombinases were expressed under the control of the ubiquitous CAG promoter (Fig. 1A) (10) and are activated in response to 4-hydroxytamoxifen (4OHT). With Cre recombinase, a Cre-dependent expression vector (11) containing the CAG promoter, a floxed “stop cassette,” and a reporter gene (GFP or DsRed) was used (termed CALNL-GFP and CALNL-DsRed, respectively) (Fig. 1B). With Flp recombinase, a similar inducible expression vector that had FRT sites instead of loxP sites was used (termed CAFNF-GFP and CAFNF-DsRed, respectively) (Fig. 1B). The activities of CreERT2, ERT2Cre, ERT2CreERT2, and FlpeERT2 were tested in vivo in the rat retina (Fig. 1 C–K). When P0 rat retinas were coelectroporated with CAG-CreERT2, CALNL-DsRed (recombination indicator), and CAG-GFP (transfection control) and harvested at P21, very high background recombination (DsRed expression) was detected without 4OHT (Fig. 1 D and H). ERT2Cre also had very high background recombination activity (data not shown). FlpeERT2 had detectable background recombination activity without 4OHT, although it was lower than that of CreERT2 (Fig. 2E and I). In contrast, ERT2CreERT2 had no detectable recombination activity without 4OHT (Fig. 2 F and J). When 4OHT was i.p. injected into the transfected rats at P20, an induction of DsRed expression was clearly detected 24 h after 4OHT administration (Fig. 2 G and K). Similar results were observed when CreERT2, ERT2Cre, ERT2CreERT2, and FlpeERT2 were transfected into 293T cells (SI Fig. 8) or embryonic day 14.5 (E14.5) mouse brain (SI Fig. 9). These results indicate that, at least for in vivo electroporation studies, ERT2CreERT2, but not CreERT2 (ERT2Cre) and FlpeERT2, can lead to tight regulation of the onset of transgene expression.

Fig. 1.
Temporal regulation of gene expression in the retina by using inducible Cre and Flp recombinases. (A) 4OHT-responsive Cre and Flpe (the enhanced form of Flp) recombinases composed of Cre/Flpe and the mutated ligand-binding domain(s) of ER (ERT2) are expressed ...
Fig. 2.
Gene expression in the retina by using cell-type-specific promoters. P0 rat retinas were coelectroporated with two plasmids: CAG-GFP (transfection control) and retinal cell-type-specific promoter-DsRed. The two plasmids were mixed at a mass ratio of 1:2 ...

Cell-Type-Specific Regulation Using Specific Promoters.

To restrict transgene expression to specified cell types in the retina, several retinal cell-type-specific promoters were characterized. Regulatory sequences for rhodopsin (12), calcium binding protein 5 (Cabp5), and cellular retinaldehyde binding protein (Cralbp), were previously reported (6). In addition to these promoters, promoter regions of Nrl [expressed in rods (13)], Crx [expressed in photoreceptors and weakly in bipolars (14, 15)], N-myc downstream regulated gene 4 [Ndrg4; expressed in amacrines (C. Punzo and C.L.C., unpublished data)], clusterin [expressed in Müller glia (16)], Rax [expressed in progenitors (17, 18)], and Hes1 [expressed in progenitors (19)] were characterized for this study. Two types of constructs were made to characterize these promoters. One used the promoter to express DsRed. This type of construct was coelectroporated with CAG-GFP, as a transfection control, into P0 rat retinas (Fig. 2). The other type of construct used the cell-type-specific promoter to regulate expression of Cre. This type of construct was coelectroporated with CALNL-DsRed, a recombination indicator, and CAG-GFP as a transfection control (Fig. 3). The transfected retinas were harvested at P20, and the expression patterns of DsRed were analyzed. Fig. 2 shows the expression patterns of DsRed directly driven by the cell-type-specific promoters. Rhodopsin promoter-DsRed (Fig. 2A) and Nrl promoter-DsRed (Fig. 2B) expressed exclusively in rod photoreceptors. The former started to be detected at around P4 or P5, whereas the onset of the latter was slightly earlier (at around P2–P3), consistent with the expression profiles of rhodopsin and Nrl (2022). Crx promoter-DsRed was detected in bipolar cells at a high level and in photoreceptors at a very low level (Fig. 2C). This expression pattern of DsRed is inconsistent with that of the native Crx gene (14, 15) as well as with a previous report characterizing the Crx promoter in transgenic mice (23). The previous report used different regulatory constructs, which might explain this discrepancy, or perhaps the difference is attributable to nonintegrated plasmid regulation vs. integrated transgenes. Cabp5 promoter-DsRed (Fig. 2D) and Ndrg4 promoter-DsRed (Fig. 2E) were detected only in bipolar and amacrine cells, respectively. Cralbp promoter-DsRed (Fig. 2F) and clusterin promoter-DsRed (Fig. 2G) were detected only in Müller glia. Rax promoter-DsRed could not be detected at P20 (Fig. 2H), although weak expression of DsRed was detected in the retina at P2–P3 (data not shown), suggesting that the Rax promoter is active in progenitors and down-regulated in differentiated cells. Hes1 promoter-DsRed was detected in ≈50% of GFP-positive cells at P2 (Fig. 2I), and its expression was restricted to Müller glia at P20 (Fig. 2J).

Fig. 3.
Lineage-tracing experiments in the retina by using the Cre/loxP system and cell-type-specific promoters. P0 rat retinas were coelectroporated with three plasmids: CAG-GFP (transfection control), CALNL-DsRed (recombination indicator), and retinal cell-type-specific ...

Fig. 3 shows the expression patterns of CALNL-DsRed after recombination by the Cre induced by the retinal cell-type-specific promoter. In all cases, DsRed expression levels in the retina were higher than those observed when each cell-type-specific promoter was used to directly drive expression of DsRed. The rhodopsin promoter-Cre specifically induced the expression of DsRed in rods (Fig. 3A). Similarly, the Nrl promoter-Cre led to the expression of DsRed only in rods (Fig. 3B), indicating that both promoters are restricted to rods and not even transiently active in other cell types, including multipotent progenitors. When the Crx promoter-Cre was used, rods and bipolars were clearly labeled with DsRed (Fig. 3C). Interestingly, when the Cabp5 promoter-Cre was used, a subset of rods, as well as bipolars, were labeled with DsRed (Fig. 3D). The ratio of the number of DsRed-positive rods to that of DsRed-positive bipolars was ≈1:1. This ratio might suggest that the Cabp5 promoter is active in the progenitors that produce rod and bipolar cells. The Ndrg4 promoter-Cre induced the expression of DsRed in amacrines as well as in a subset of rods (Fig. 3E). The Cralbp promoter-Cre (Fig. 3F) and the clusterin promoter-Cre (Fig. 3G) induced the expression of DsRed in Müller glia and a small population of rod and bipolar cells. The Rax promoter-Cre (Fig. 3H) and the Hes1 promoter-Cre (Fig. 3I) induced the expression of DsRed in rod, bipolar, amacrine, and Müller glial cells (see also SI Tables 1 and 2).

Temporal and Cell-Type-Specific Regulation.

By combining the temporal regulation afforded by ERT2CreERT2 with the cell-type-specific regulation provided by the promoters, we next tried to inducibly express a reporter gene specifically in “differentiated” Müller glia. As shown in Fig. 4, the clusterin promoter was cloned upstream of ERT2CreERT2, and the resulting plasmid was coelectroporated with CALNL-DsRed and CAG-GFP into P0 rat retinas. The retinas were stimulated with 4OHT at P14, the time point when retinogenesis is complete (24, 25), and then analyzed at P16. Without 4OHT stimulation, DsRed expression was not detected (Fig. 4B). When 4OHT was injected into the rats, clear DsRed expression was detected in a subset of GFP-positive cells (Fig. 4C). The retinal sections showed that all of the DsRed-positive cells examined were morphologically Müller glia (Fig. 4D). These results show that ERT2CreERT2 can be used with retinal cell-type-specific promoters to achieve precise temporal and cell-type-specific regulation in the retina.

Fig. 4.
Inducible expression in Müller glial cells. P0 rat retinas were coelectroporated with three plasmids: clusterin promoter-ERT2CreERT2, CALNL-DsRed, and CAG-GFP. The three plasmids were mixed at a mass ratio of 1:3:2 (final concentration, 6.0 μg/μl). ...

Conditional RNAi.

To conditionally knockdown gene expression in the retina, Cre-dependent inducible RNAi vectors using the microRNA30 (mir30)-based shRNA expression system (26, 27) were made. The mir30-based shRNA expression system has several advantages over conventional RNAi vectors expressing shRNAs directly from pol III promoters, such as the U6 promoter. First, shRNA can be expressed using pol II promoters (e.g., CAG promoter). Second, it is technically simple to put the regulatory elements (e.g., transcriptional stop cassette) into the expression vectors. Fig. 5A shows the mir30-based vectors expressing shRNA under the control of the human U6 promoter (hU6-mir30) or CAG promoter (CAG-mir30 and CALSL-mir30). CALSL-mir30 is a Cre-dependent inducible RNAi vector carrying a floxed transcriptional stop cassette immediately after the CAG promoter. These vectors were used to express a shRNA against GFP (GFPshRNA) and tested in 293T cells transfected with CAG-GFP. As shown in Fig. 5 B–D, both hU6-mir30(GFPshRNA) and CAG-mir30(GFPshRNA) significantly suppressed the expression of GFP. The knockdown efficiencies of these two vectors were comparable. CALSL-mir30(GFPshRNA) efficiently knocked down the expression of GFP only in the presence of Cre (Fig. 5 E–F). To determine whether CALSL-mir30 also works in the retina, CALSL-mir30 was used with the rhodopsin promoter-Cre. P0 rat retinas were coelectroporated with four plasmids, CALSL-mir30(GFPshRNA), CAG-GFP, CAG-DsRed, and the rhodopsin promoter-Cre, and analyzed at P20 (Fig. 5 G–H). Without the Cre construct, transfected retinal cells were clearly labeled by both GFP and DsRed (Fig. 5G). With the rhodopsin promoter-Cre, GFP expression in the ONL was specifically silenced, whereas that in the INL cells was not affected (Fig. 5H), demonstrating that CALSL-mir30 can be used to conditionally knockdown gene expression in the retina. These data also demonstrate that there is a very high cotransfection efficiency. In order for cells to be both red and green, they must have been successfully electroporated with both CAG-GFP and CAG-DsRed. The coexpression rate of these two genes was nearly 100%, both in this case, where four plasmids were coelectroporated, and in many previous cases with two or three coelectroporated plasmids (6).

Fig. 5.
Inducible RNAi in the retina. (A) Diagram of mir30-based RNAi vectors. hU6-mir30: The mir30 expression cassette has the hairpin stem composed of siRNA sense and antisense strands (22 nt each), a loop derived from human miR30 (19 nt), and 125-nt miR30 ...

Controlled Misexpression of Transcription Factors in the Retina.

By using the Cre/loxP system, two transcription factors that prevent neural differentiation, Rax (28) and Hes1 (19, 29, 30), were inducibly expressed at two different developmental stages of rod photoreceptors (Fig. 6) and in the developing mouse brain (SI Fig. 10). When the Rax and Hes1 expression vectors driven by the CAG promoter were coelectroporated with CAG-GFP into P0 rat retinas, most GFP-positive cells localized to the INL and became Müller glia-like cells by P14 (Fig. 6 A and B). However, the morphology of the Rax-induced cells was slightly different from that of the Hes1-induced cells, which were more similar to native Müller glia. The rhodopsin promoter was used to target the late developmental stage of rod photoreceptors. Rax and Hes1 were misexpressed by using the Cre-dependent inducible expression vectors (CALNL-Rax and CALNL-Hes1, respectively). When P0 rat retinas were coelectroporated with CALNL-Rax, CALNL-GFP, and the rhodopsin promoter-Cre and analyzed at P20, all of the GFP-positive cells were abnormal rods lacking outer segments, and no GFP-positive cells were seen in the INL (Fig. 6D). When Hes1 was misexpressed by using the rhodopsin promoter-Cre, all GFP-positive cells were rods with shortened outer segments (Fig. 6E). Next, another rod-specific promoter, the Nrl promoter, was used to express Cre. Nrl is a direct upstream regulator of rhodopsin, and its expression precedes the expression of rhodopsin (Fig. 6C) (31). CALNL-Rax or CALNL-Hes1 was coelectroporated with CALNL-DsRed, CAG-GFP, and the Nrl promoter-Cre into P0 rat retinas, and the retinas were analyzed at P20. When Rax was inducibly expressed by using the Nrl promoter-Cre, less than half of the DsRed-positive cells were rods in the ONL, and the rest were Müller glia-like cells whose nuclei were in the INL (Fig. 6F). Hes1 misexpression using the Nrl promoter-Cre also led to the generation of both rods and Müller-like cells (Fig. 6G). These results indicate that the fate of immature rods (Nrl+/rhodopsin) can be partially respecified by Rax/Hes1, whereas, after rhodopsin expression turns on, the cell fate can no longer be changed by Rax/Hes1.

Fig. 6.
Targeted misexpression of Rax and Hes1 in rod photoreceptors. (A and B) Misexpression of Rax and Hes1 in the developing retina using a conventional expression system. CAG-Rax (A) or CAG-Hes1 (B) was coelectroporated with CAG-GFP into P0 rat retinas. The ...

Electroporation of CAG-Hes1 into the ventricular zone (VZ) of E14.5 mouse telencephalon led to generation of radial glia (like) cells whose cell bodies remained in the VZ (SI Fig. 10 A–D). When Hes1 was inducibly misexpressed in the E14.5 VZ-derived cells 2 d later (at E16.5) using the 4OHT-regulated Cre and CALNL-Hes1, radial glia (like) cells were not generated, but labeled cells showed abnormal migration patterns (SI Fig. 10 E and F).


The steroid ligand-regulated forms of Cre and Flp are widely used to regulate the timing of gene activation and inactivation in transgenic mice (3). However, we found that CreERT2 (ERT2Cre) and FlpeER T2 have substantial background recombination activities even in the absence of 4OHT when expressed in the rat retina or mouse brain by electroporation or in 293T cells by lipofection. Compared with a Cre allele with a single ERT2 domain, the ERT2CreERT2 double fusion had much lower background activity. In our experimental conditions, no recombination activity was detected in the rat retina at least for 3 weeks after electroporation. Heat shock protein 90 (Hsp90) interacts with the ER domain in the cytosol and thereby prevents the translocation of CreER fusion protein to the nucleus where DNA recombination occurs (32). ERT2CreERT2 double fusion may have a higher affinity for Hsp90 to form a tighter complex. Alternatively, the ERT2CreERT2 fusion may have less activity due to the double fusion and thus less background activity. It is also possible that degradation of CreERT2 (ERT2Cre) results in generation of “active Cre” lacking the regulatory domain, whereas ERT2CreERT2 is still inactive even after losing one regulatory domain.

The Cre/loxP recombination-dependent inducible vectors with the CAG promoter (CALNL) have several useful features. First, it is possible to inducibly express genes in specific cell types at specific time points. Second, after Cre/loxP-mediated recombination, strong “output signals” driven by the CAG promoter can be obtained. Expression of fluorescent reporter genes directly from cell-type-specific promoters frequently results in such low levels of expression as to render the constructs unusable for some applications, e.g., detection of live, labeled cells. We found that there was much greater sensitivity for the detection of expression when cell-type-specific promoters were used with Cre plus CALNL-DsRed. Finally, these vectors can be used to trace the fate of progenitor and precursor cells labeled upon Cre/loxP-mediated recombination. Using these vectors, we showed that several “cell-type-specific” promoters are weakly active in progenitors and/or other cell types. Two explanations for this observation, which are not mutually exclusive, can be considered. Lineage analyses using retroviral vectors have shown that clones with bipolar, Müller glia, and/or amacrine cells almost always also contain rods, even if the clone is of only two cells (33). If these promoters are weakly active in progenitors that give rise to multiple cell types, then one might see labeling of rods, along with labeling of the cell type that normally is the only cell type to express a particular promoter. This idea is supported by two previous observations. Most of the known Müller glia-specific genes of the adult retina are also expressed in progenitors (16), and two genes thought to be restricted to amacrine and horizontal cells could be observed in progenitors (34). An additional explanation is that these promoters, transiently introduced into the retina by electroporation, are slightly “leaky,” and such leakiness was detected by the more sensitive reporter (Cre plus CALNL-DsRed).

In addition to progenitor cells, Müller glia are also reported to have the potential to generate new neurons in response to acute retinal injury in the adult retina (35, 36). Therefore, it is of interest to label and trace the fate of Müller glia in living animals by using Müller glia-specific promoters. However, because of the aforementioned overlap of gene expression profiles between Müller glia and progenitors (16), it is difficult to isolate promoters restricted only to mature Müller glia. The two presumptive Müller glia-specific promoters, Cralbp and clusterin promoters, characterized in this study do not appear to be specific for “mature” Müller glia, at least when introduced by electroporation. However, by combining the clusterin promoter with inducible Cre, selective labeling of Müller glia upon 4OHT stimulation was achieved in the differentiated retina. This system will be useful for specific labeling of mature Müller glia in the adult retina, as well as for other cell types with gene expression profiles that overlap during development but not at later times.

Cre-dependent inducible RNAi vectors have been constructed from the U6 promoter-based conventional RNAi vectors (3740). Compared with these vectors, the CAG promoter/mir30-based inducible RNAi vector (CALSL-mir30) has several merits. First, it is possible to express a gene of interest for gain-of-function studies and shRNA for loss-of-function studies by using the same vector. Second, the ubiquitous CAG promoter can be replaced with other pol II promoters to increase the cell type specificity. We found that the rhodopsin promoter can be used to directly express shRNA with the mir30 cassette in rods (data not shown). Thus, it appears that the CAG promoter/mir30-based inducible RNAi vector is more suitable for in vivo electroporation.

We demonstrate that the use of the Cre/loxP system and cell-type-specific promoters broadens the application range of in vivo electroporation. These methods rely on the fact that multiple plasmids (e.g., Cre-expression vector plus Cre-dependent inducible expression vector) can be introduced into the same cells by electroporation. This is one of the advantages of in vivo electroporation over retroviral vectors, with which it is hard to introduce multiple expression units into the same cells. These methods are not only useful for neuroscience research, but also for studies of other tissues of various animal species in which in vivo electroporation can be applied.

Materials and Methods

DNA Construction.

Detailed DNA construction procedures are described in SI Materials and Methods.

In Vivo Electroporation and Tissue Processing.

In vivo electroporation of the rat retina was performed as described (6). In vivo electroporation of the mouse telencephalon was performed as described (41). Tissue processing and immunostaining were performed as described (6). Detailed procedures are described in SI Materials and Methods.

4OHT Stimulation.

For in vivo injection, 4OHT (Sigma, St. Louis, MO) was dissolved in ethanol at a concentration of 20 mg/ml and then diluted with 9 volumes of corn oil (Sigma). Diluted 4OHT (2 mg/ml) was i.p. injected (500 μl per animal) into the animals with a 26-gage needle. For 293T cells, 4OHT diluted in ethanol (0.39 mg/ml) was directly added to the cell culture media (final concentration, 1 μM = 0.39 μg/ml).

Supplementary Material

Supporting Information:


We thank Drs. I. Saito (University of Tokyo, Tokyo, Japan), J. Miyazaki (Osaka University, Osaka, Japan), R. Awatramani (Harvard Medical School), S. Dymecki (Harvard Medical School), D. Metzger (Institut de Génétique et de Biologie Moléculaire et Cullulaire, Illkrich, France), P. Chambon (Université Louis Pasteur, Strasbourg, France), D. Zack (The Johns Hopkins University, Baltimore, MD), R. Kageyama (Kyoto University, Kyoto, Japan), S. McConnell (Stanford University, Stanford, CA), and R. Molday (University of British Columbia, Vancouver, BC, Canada) for providing materials and Dr. K. Sanada for technical advice. This work was supported by National Institutes of Health Grant EYO-8064 and EYO-9676 and by the Howard Hughes Medical Institute.


calcium binding protein 5
cellular retinaldehyde binding protein
embryonic day n
estrogen receptor
inner nuclear layer
N-myc downstream regulated gene 4
outer nuclear layer
postnatal day n
small hairpin RNA.


The authors declare no conflict of interest.

This article contains supporting information online at www.pnas.org/cgi/content/full/0610155104/DC1.


1. Lewandoski M. Nat Rev Genet. 2001;2:743–755. [PubMed]
2. Nagy A, Gertsenstein M, Vintersten K, Behringer R. Manipulating the Mouse Embryo: A Laboratory Manual. Woodbury, NY: Cold Spring Harbor Lab Press; 2003.
3. Branda CS, Dymecki SM. Dev Cell. 2004;6:7–28. [PubMed]
4. Itasaki N, Bel-Vialar S, Krumlauf R. Nat Cell Biol. 1999;1:E203–E207. [PubMed]
5. Swartz M, Eberhart J, Mastick GS, Krull CE. Dev Biol. 2001;233:13–21. [PubMed]
6. Matsuda T, Cepko CL. Proc Natl Acad Sci USA. 2004;101:16–22. [PMC free article] [PubMed]
7. Feil R, Wagner J, Metzger D, Chambon P. Biochem Biophys Res Commun. 1997;237:752–757. [PubMed]
8. Casanova E, Fehsenfeld S, Lemberger T, Shimshek DR, Sprengel R, Mantamadiotis T. Genesis. 2002;34:208–214. [PubMed]
9. Hunter NL, Awatramani RB, Farley FW, Dymecki SM. Genesis. 2005;41:99–109. [PubMed]
10. Niwa H, Yamamura K, Miyazaki J. Gene. 1991;108:193–199. [PubMed]
11. Kanegae Y, Lee G, Sato Y, Tanaka M, Nakai M, Sakaki T, Sugano S, Saito I. Nucleic Acids Res. 1995;23:3816–3821. [PMC free article] [PubMed]
12. Zack DJ, Bennett J, Wang Y, Davenport C, Klaunberg B, Gearhart J, Nathans J. Neuron. 1991;6:187–199. [PubMed]
13. Swaroop A, Xu JZ, Pawar H, Jackson A, Skolnick C, Agarwal N. Proc Natl Acad Sci USA. 1992;89:266–270. [PMC free article] [PubMed]
14. Chen S, Wang QL, Nie Z, Sun H, Lennon G, Copeland NG, Gilbert DJ, Jenkins NA, Zack DJ. Neuron. 1997;19:1017–1030. [PubMed]
15. Furukawa T, Morrow EM, Cepko CL. Cell. 1997;91:531–541. [PubMed]
16. Blackshaw S, Harpavat S, Trimarchi J, Cai L, Huang H, Kuo WP, Weber G, Lee K, Fraioli RE, Cho SH, et al. PLoS Biol. 2004;2:E247. [PMC free article] [PubMed]
17. Furukawa T, Kozak CA, Cepko CL. Proc Natl Acad Sci USA. 1997;94:3088–3093. [PMC free article] [PubMed]
18. Mathers PH, Grinberg A, Mahon KA, Jamrich M. Nature. 1997;387:603–607. [PubMed]
19. Tomita K, Ishibashi M, Nakahara K, Ang SL, Nakanishi S, Guillemot F, Kageyama R. Neuron. 1996;16:723–734. [PubMed]
20. Treisman JE, Morabito MA, Barnstable CJ. Mol Cell Biol. 1988;8:1570–1579. [PMC free article] [PubMed]
21. Morrow EM, Belliveau MJ, Cepko CL. J Neurosci. 1998;18:3738–3748. [PubMed]
22. Blackshaw S, Fraioli RE, Furukawa T, Cepko CL. Cell. 2001;107:579–589. [PubMed]
23. Furukawa A, Koike C, Lippincott P, Cepko CL, Furukawa T. J Neurosci. 2002;22:1640–1647. [PubMed]
24. Young RW. Anat Rec. 1985;212:199–205. [PubMed]
25. Rapaport DH, Wong LL, Wood ED, Yasumura D, LaVail MM. J Comp Neurol. 2004;21:304–324. [PubMed]
26. Zeng Y, Wagner EJ, Cullen BR. Mol Cell. 2002;9:1327–1333. [PubMed]
27. Paddison PJ, Cleary M, Silva JM, Chang K, Sheth N, Sachidanandam R, Hannon GJ. Nat Methods. 2004;1:163–167. [PubMed]
28. Furukawa T, Mukherjee S, Bao ZZ, Morrow EM, Cepko CL. Neuron. 2000;26:383–394. [PubMed]
29. Ishibashi M, Moriyoshi K, Sasai Y, Shiota K, Nakanishi S, Kageyama R. EMBO J. 1994;13:1799–1805. [PMC free article] [PubMed]
30. Ohtsuka T, Sakamoto M, Guillemot F, Kageyama R. J Biol Chem. 2001;276:30467–30474. [PubMed]
31. Rehemtulla A, Warwar R, Kumar R, Ji X, Zack DJ, Swaroop A. Proc Natl Acad Sci USA. 1996;93:191–195. [PMC free article] [PubMed]
32. Picard D. Curr Opin Biotechnol. 1994;5:511–515. [PubMed]
33. Turner DL, Cepko CL. Nature. 1987;328:131–136. [PubMed]
34. Alexiades MR, Cepko CL. Development (Cambridge, UK) 1997;124:1119–1131. [PubMed]
35. Fischer AJ, Reh TA. Nat Neurosci. 2001;4:247–252. [PubMed]
36. Ohoto S, Akagi T, Kageyama R, Akita J, Mandai M, Honda Y, Takahashi M. Proc Natl Acad Sci USA. 2004;101:3654–3659.
37. Coumoul X, Li W, Wang RH, Deng C. Nucleic Acids Res. 2004;32:e85. [PMC free article] [PubMed]
38. Fritsch L, Martinez LA, Sekhri R, Naguibneva I, Gerard M, Vandromme M, Schaeffer L, Harel-Bellan A. EMBO Rep. 2004;5:178–182. [PMC free article] [PubMed]
39. Tiscornia G, Tergaonkar V, Galimi F, Verma IM. Proc Natl Acad Sci USA. 2004;101:7347–7351. [PMC free article] [PubMed]
40. Ventura A, Meissner A, Dillon CP, McManus M, Sharp PA, Van Parijs L, Jaenisch R, Jacks T. Proc Natl Acad Sci USA. 2004;101:10380–10385. [PMC free article] [PubMed]
41. Saito T, Nakatsuji N. Dev Biol. 2001;240:237–246. [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
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...


  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...