Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2012 Apr;40(8):3741-52.
doi: 10.1093/nar/gkr1214. Epub 2011 Dec 14.

Gene Targeting to the ROSA26 Locus Directed by Engineered Zinc Finger Nucleases

Affiliations
Free PMC article

Gene Targeting to the ROSA26 Locus Directed by Engineered Zinc Finger Nucleases

Pablo Perez-Pinera et al. Nucleic Acids Res. .
Free PMC article

Abstract

Targeted gene addition to mammalian genomes is central to biotechnology, basic research and gene therapy. For example, gene targeting to the ROSA26 locus by homologous recombination in embryonic stem cells is commonly used for mouse transgenesis to achieve ubiquitous and persistent transgene expression. However, conventional methods are not readily adaptable to gene targeting in other cell types. The emerging zinc finger nuclease (ZFN) technology facilitates gene targeting in diverse species and cell types, but an optimal strategy for engineering highly active ZFNs is still unclear. We used a modular assembly approach to build ZFNs that target the ROSA26 locus. ZFN activity was dependent on the number of modules in each zinc finger array. The ZFNs were active in a variety of cell types in a time- and dose-dependent manner. The ZFNs directed gene addition to the ROSA26 locus, which enhanced the level of sustained gene expression, the uniformity of gene expression within clonal cell populations and the reproducibility of gene expression between clones. These ZFNs are a promising resource for cell engineering, mouse transgenesis and pre-clinical gene therapy studies. Furthermore, this characterization of the modular assembly method provides general insights into the implementation of the ZFN technology.

Figures

Figure 1.
Figure 1.
Schematic representation of the genomic target sequence of the ROSA26 ZFNs. (A) Target sequence of the ROSA26 ZFNs. Located on mouse chromosome 6, the Gt(ROSA)26Sor (ROSA26) locus spans 9 kb at position 113,026,025 of NCBI Reference Sequence NC_000072.5. The ROSA26 ZFN pair was generated by a modular assembly approach and consists of six zinc finger domains in the left array and four zinc finger domains in the right array that target the first intron of the ROSA26 locus. (B) Amino acid sequences of the alpha helices within the zinc finger domains that recognize the indicated triplets of DNA.
Figure 2.
Figure 2.
Activity of the ROSA26 ZFNs in SSA assays. (A) Schematic representation of the vector used in Single Strand Annealing (SSA) assays. A central region in the EGFP cDNA is duplicated and both copies are separated by a sequence that introduces an early stop codon to render a truncated not-functional EGFP. This spacer sequence also contains the ROSA26 ZFN target site. Upon cleavage the endogenous SSA repair mechanisms are expected to remove the stop codon and restore the EGFP cDNA. (B) ROSA26 ZFNs efficiently induce SSA-mediated gene repair in an episomal reporter plasmid. Expression plasmids for ROSA26 ZFNs containing 3, 4, 5, or 6 modules in the left and right array were cotransfected with the SSA reporter vector into 293T cells and the levels of EGFP expression were measured at 48 h by flow cytometry (n = 3, mean ± stdev). ANOVA P = 2E-8. R3L4 versus R3L3 P = 0.001; R3L5 versus R3L4 P = 0.05; R4L3 versus R4L4 P = 0.002; R4L3 versus R4L5 P = 0.001. (C) ROSA26 ZFNs generate targeted genomic double-strand breaks. The SSA reporter construct was stably integrated into the genome of HEK293 cell lines. In the absence of ZFNs, spontaneous EGFP gene reconstitution and protein expression was detected in 0.3% of cells. Transfection of these cells with wild-type (WT) ZFNs or ZFNs containing the obligate heterodimer and Sharkey mutations (eZFNs) led to a significant increase in the number of EGFP-positive cells (n = 3, mean ± stdev). ANOVA P = 7E-19. R3L5 versus R3L4 P = 1E-4; R3L6 versus R3L5 P = 5E-5; R4L5 versus R4L4 P = 2E-5; R4L6 versus R4L4 P = 6E-5.
Figure 3.
Figure 3.
Activity of the ROSA26 ZFNs at the endogenous target locus. (A) Schematic representation of the Surveyor assay. Genomic DNA from cells transfected with ZFNs is isolated and used in a PCR reaction that amplifies wild-type DNA and DNA cleaved by the ZFNs and repaired by error-prone non-homologous end joining (NHEJ). The PCR product is melted and reannealed to facilitate the formation of heteroduplexes, which can be detected as cleaved fragments of the PCR product upon digestion with Cel-I nuclease. (B) Level of endogenous allele modification induced by ROSA26 ZFNs with a wild-type FokI domain. Neuro2A cells were transfected with ZFNs containing 3, 4, 5 or 6 modules in the left and right array. After 3 days the genomic DNA was isolated and the Surveyor assay was performed. No gene modification was detected in cells electroporated with control plasmid, whereas significant levels of modification were detected in all samples treated with ZFNs. Although there was no significant difference in the activity of ZFN pairs containing R3 or R4 arrays by this assay, ZFN activity increases when the L5 module is added to the L4 array. (C) Level of endogenous allele modification induced by ROSA26 eZFNs. The level of endogenous allele modification by the eZFNs was significantly higher than modification by wild-type ZFNs regardless of the number of zinc finger modules in either array. (D) Sequences of indels at the ROSA26 target site following ZFN-mediated cleavage and repair by NHEJ.
Figure 4.
Figure 4.
Levels of ROSA26 ZFN activity as a function of time, dose, and cell type. (A) Time-dependent activity of ROSA26 ZFNs. Neuro2A cells were transfected with ROSA26 R4L6 eZFNs and genomic DNA was isolated at 16, 32, 48, 72 and 120 h after transfection. The endonuclease activity of the ZFNs was readily detected at 16 h and reached near maximal levels of activity at 32 h. The percentage of gene modification remained effectively constant for 5 days. (B) Dose-dependent activity of ROSA26 ZFNs. DNA was isolated from Neuro2A cells transfected with 5, 10 or 20 μg of plasmid encoding ROSA26 ZFNs and the level of genomic modification was measured by the Surveyor assay. Targeted DNA cleavage increased with the dose of ROSA26 ZFN expression plasmid. (C) Cell type-dependent activity of ROSA26 ZFNs. The activity of the ROSA26 R4L6 eZFN was tested in a panel of murine cell lines, including NIH3T3 embryonic fibroblasts, C2C12 skeletal myoblasts, Neuro2A neuroblastoma cells, and primary skeletal myoblasts. The activity of the ZFNs was consistently highest in Neuro2A cells, however, significant levels of gene modification were detected in all cell lines tested.
Figure 5.
Figure 5.
Targeted homologous recombination directed by ROSA26 ZFNs. (A) Schematic representation of targeted integration via homologous recombination using ZFNs. The donor vector contains two 800 bp arms of homology with the ROSA26 locus adjacent to the ZFN target site. Genes of interest (GOI), such as the ZeocinR-EGFP cassette (ZeoEGFP), can be cloned into the multiple cloning site (MCS) between the two homology arms. When this vector is delivered together with the ZFNs, the endogenous repair mechanisms can use the donor vector as template to repair the double-strand break. This will result in integration of the gene of interest into the genome at the ZFN target site. (B) Integration of a 45 bp sequence into the ROSA26 locus using the ROSA26 R4L6 eZFN. Neuro2A cells were electroporated with control DNA, with ROSA26 R4L6 eZFN, with an empty donor vector alone or with ROSA26 R4L6 eZFN and the donor vector. PCR with primers that bind the MCS and the ROSA26 locus beyond the homology arms was performed. No specific integration events were detected following delivery of an empty vector, the ROSA26 eZFNs, or the donor vector alone. However when the ZFNs were transfected together with the donor vector, the MCS derived from the donor vector was integrated into the ROSA26 locus at the target site. (C) Integration of a 45 bp sequence into the ROSA26 locus using ZFNs of variably sized zinc finger arrays. Neuro2A cells were electroporated with expression plasmids for eZFN pairs containing 3, 4, 5 or 6 modules in the left and right arrays and the MCS donor vector. All the ZFN pairs were effective at mediating integration of the MCS into the genomic DNA at the correct target site, as determined by PCR with primers that bind the MCS and the ROSA26 locus beyond the homology arms.
Figure 6.
Figure 6.
Targeted integration of a ZeocinR-EGFP expression cassette into the ROSA26 locus. (A) Integration of ZeoEGFP in the ROSA26 locus using ROSA26 R4L6 eZFN in unselected populations. To determine whether the ROSA26 ZFNs facilitate targeted addition of larger expression cassettes at this locus, a ∼1.6 kb sequence consisting of a ZeocinR-EGFP gene (ZeoEGFP) driven by a CMV promoter was inserted between the homology arms of the donor vector. Control plasmid, ROSA26 R4L6 eZFN alone, donor vector alone, or donor vector together with the ROSA26 R4L6 eZFN were electroporated into Neuro2A cells. Targeted gene addition was tested using PCR of the genomic DNA with primers that bind in the Zeo-EGFP expression cassette and genomic DNA in the ROSA26 locus beyond the homology arms. The transgene was inserted correctly in the target site, as confirmed using sequencing of the PCR product, only when the donor vector was delivered with the ZFNs. (B) Efficiency of ZFN-mediated targeted gene addition. In cells transfected with only the donor vector, ZeoEGFP expression was detectable in less than 1% of transfected cells by 14 days post-transfection. However, ∼10% of the cell population expressed EGFP after 25 days in cells transfected with donor vector and the ROSA26 R4L6 eZFNs, representing sustained expression from integrated expression cassettes (ANOVA P = 2E-4; Student's t test P = 1E-4). (C) Clonal analysis of gene targeting by ROSA26 ZFNs. Neuro2A cells were transfected with the ZeoEGFP donor vector alone or cotransfected with the donor vector and the ROSA26 R4L6 eZFNs. The cells were selected with Zeocin and clonal populations were isolated and analyzed for presence of the ZeoEGFP transgene in the ROSA26 locus. All clones contained the desired integration event when the donor vector and the ZFNs were cotransfected but not when the donor vector was delivered alone. Further analysis using PCR with primers that detect the wild-type ROSA26 locus determined that the clones containing targeted ROSA26-ZeoEGFP integration also contained at least one wild-type allele. (D and E) Expression of ZeoEGFP from the ROSA26 locus. To determine the levels of EGFP expressed from the ZeoEGFP transgene stably integrated into the genome, the (D) 12 untargeted and (E) 12 targeted clonal cell lines were analyzed by flow cytometry. Clones derived from transfections with the donor vector alone expressed highly variable levels of EGFP. In contrast, the clones from cells treated with ZFNs and expression vector showed relatively homogenous levels of EGFP expression. (F) Quantification of variability of gene expression in untargeted and targeted clones. The reduced variability of gene expression achieved by ZFN-mediated gene targeting was quantified as a decrease in the mean of the coefficient of variation of cellular fluorescence intensity of each clonal population and a dramatic decrease in the standard deviation of the coefficient of variation for each clone (n = 12, mean ± SD).

Similar articles

See all similar articles

Cited by 35 articles

See all "Cited by" articles

References

    1. Koller BH, Smithies O. Altering genes in animals by gene targeting. Annu. Rev. Immunol. 1992;10:705–730. - PubMed
    1. Capecchi MR. Gene targeting in mice: functional analysis of the mammalian genome for the twenty-first century. Nat. Rev. Genet. 2005;6:507–512. - PubMed
    1. Zambrowicz BP, Imamoto A, Fiering S, Herzenberg LA, Kerr WG, Soriano P. Disruption of overlapping transcripts in the ROSA beta geo 26 gene trap strain leads to widespread expression of beta-galactosidase in mouse embryos and hematopoietic cells. Proc. Natl Acad. Sci. USA. 1997;94:3789–3794. - PMC - PubMed
    1. Soriano P. Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat. Genet. 1999;21:70–71. - PubMed
    1. Friedrich G, Soriano P. Promoter traps in embryonic stem cells: a genetic screen to identify and mutate developmental genes in mice. Genes Dev. 1991;5:1513–1523. - PubMed

Publication types

Feedback