| Nucleic Acids Research | Article |
Simplified generation of targeting constructs using ET recombination
Introduction
Materials And Methods
DNA techniques
PCR product preparation
Bacterial transformation
ES cell culture and transformation
Results
Discussion
Acknowledgements
References
Simplified generation of targeting constructs using ET recombination
Received June 29, 1999; Revised and Accepted July 19, 1999
ABSTRACT ET recombination is a way to engineer DNA in Escherichia coli using homologous recombination. Here we develop the potential of ET recombination in two ways relevant to complex engineering exercises such as building gene targeting constructs. First, a targeting construct was made in a single step. Second, ET recombination was used to place two unique restriction sites at precise positions in a large genomic clone. Subsequently a complex targeting construct was created by ligation with a multifunctional cassette.
INTRODUCTION
Assembling DNA constructs for amplification in Escherichia coli is the starting point for many experiments in molecular biology. Existing methodologies employing restriction enzymes, PCR and ligation steps are well suited for many tasks but their limitations become evident when complex engineering exercises are desired. For example, these limitations often impede the construction of targeting constructs intended for modifications of vertebrate genomes, in particular, the mouse genome via embryonic stem (ES) cells.
The introduction of predetermined modifications into the mouse germ line via homologous recombination in ES cells is fundamental in the application of reverse genetics to mouse biology (1,2). Targeting constructs to manipulate the mouse genome often require complex cloning exercises. Minimally this involves positioning two sizeable fragments of mouse genomic DNA (3) either side of a selectable gene. In many cases, these cloning exercises are more complex since inclusion of additional elements, such as lacZ or GFP reporter genes, loxP sites or point mutations, are also intended. The use of long segments of genomic DNA requires extensive mapping to search for suitable restriction sites for construct design and DNA fragment assembling by DNA ligations in vitro.
We describe here two approaches to simplify complex construction exercises based on complementary applications of ET recombination, a recently described E.coli homologous recombination reaction (4), and conventional DNA methodologies. Since ET recombination permits the engineering of large intact DNA molecules regardless of the disposition of convenient restriction sites, it presents new possibilities for complex engineering. We developed this potential to create, in one step, a knock-out targeting construct for the Ssrp1 gene (5,6) and to insert an IRES-lacZ-selectable gene cassette (7) into a gene coding for a PHD fingers and SET domain-containing protein, called Nsd2 (P.-O.Angrand et al., submitted).
MATERIALS AND METHODS
DNA techniques
Large-scale plasmid DNA preparation was performed with the Qiagen Plasmid Kit (Qiagen). Restriction endonucleases and T4 DNA ligase were purchased from New England Biolabs. Plasmids were grown in E.coli strain XL1-blue [F[prime]::Tn10 proA+B+ lacIq [Delta](lacZ)M15/recA1 endA1 gyrA96 (Nalr) thi hsdR17 (rk- mk+) supE44 relA1 lac]. Complete nucleotide sequences and restriction maps of the plasmids used are available on the World Web Site http://www.embl-heidelberg.de/ExternalInfo/stewart/plasmids.html
PCR product preparation
All PCR reactions were performed in 50 µl reactions containing 5 U Amplitaq DNA polymerase (Perkin Elmer Cetus), 5 ng of plasmid and 1 pmol of each PCR primers for 30 cycles (94°C 30 s, 55°C 1 min, 72°C 1 min). The zeocin dual transcription unit was amplified from ScaI-linearized pSVZeoX1 using the primers 5[prime]-GTT CTG TCA AAG GCA GAT GTG ATC CAG GCC ACC GGA GAC GCC ATC TGC ATC TTC GTT TAA ACT CGT TAA TTA AAG GTG GCA CTT TTC GGG GAA ATG-3[prime] and 5[prime]-CTT TGC TCT TGA AGC TGT CAC TCA ACT GCC TGG ATG AAG ACT TGG ATG ACG ACG AAG CTT AGA CAT GAT AAG ATA CAT TG-3[prime] where homolgy arms to the exons 6 and 16 of the Ssrp1 gene are in bold. The chloramphenicol resistance gene from pMAK705 (8) was amplified using the primers 5[prime]-CTG TGT GAC AAG ACA GGC AGT CTC TAA CTG TGT GAG GGA CCC TGT TGT GGA TTC TAG TTT AAA CCC TGC CCT GAA CCG ACG ACC GGG T-3[prime] and 5[prime]-TAC ACA GGC TTC ATG GTA AAA CTT TCC ACA CTG ATT TAC CAC ACA ACG TTT CAC CTC GGC GCG CCT CGA ATA AAT ACC TGT GAC GGA AGA TG-3[prime], where the homology arms to Nsd2 are in bold and restriction sites for PmeI and AscI, respectively, underlined.
PCR products were purified using the Qiagen PCR Purification Kit and eluted with water, followed by digestion of any residual template DNA with DpnI. PCR products were then ethanol precipitated and resuspended in water (1 µg/µl).
Bacterial transformation
Escherichia coli strain JC8679 (9) [recB21, recC22, sbcA23, his-328, thr-1, ara-14, leuB6, [Delta](gpt-proA)62, lacY1, tsx-33, glnV44(AS), galK2, rpsL31, kdgK51, xylA5, mtl-1, argE3(Oc), thi-1, Lam-, Rac+, Qsr1+] was transformed by electroporation using a Bio-Rad Gene Pulser set at 2.5 kV, 200 [Omega] and 25 µF. The transformed cells were suspended in 600 µl L-broth and incubated for 1 h at 37°C before plating on L-agar containing zeocin (25 µg/ml) or chloramphenicol (100 µg/ml). Electroporation-competent bacterial cells were made as follows: saturated overnight JC8679 cultures were diluted 50-fold in L-broth, grown to an OD600 of 0.4 and chilled in ice for 30 min. Cells were centrifuged at 5000 r.p.m. for 15 min at -5°C. The pellet was resuspended in ice-cold 10% glycerol and centrifuged again (6000 r.p.m., -5°C, 15 min). This was repeated twice more and the cell pellet was then suspended in 300 µl of ice-cold 10% glycerol. Aliquots (50 µl) were frozen in liquid nitrogen and stored at -80°C. For electroporation, cells were thawed on ice and added to 1 µl mix containing vector DNA (0.5 µg) and PCR product (0.5 µg).
ES cell culture and transformation
E14 ES cells were grown in Dulbecco's modified Eagle's medium supplemented with 15% fetal calf serum, 100 µM non-essential amino acids, 1 µM [beta]-mercaptoethanol and leukemia inhibitory factor (LIF) (ESGROTM; Gibco BRL) at 37°C in a humidity-saturated 9% CO2 atmosphere. The cells were cultured on confluent feeder layers mitotically inactivated by treatment with mitomycin C except when they were grown in the presence of antibiotics. For transformation, E14 ES cells were trypsinized and resuspended in phosphate-buffered saline (PBS) at a concentration of 1 × 107/ml. For each individual transformation 0.8 ml cells (5.6 × 107) were mixed with 20 µg linearized DNA in an electroporation cuvette with a 0.4-cm electrode gap. Cells were electroporated with a Bio-Rad Gene Pulser set at 240 V, 500 µF. The cells were plated onto four 10-cm gelatinized plates. After 24 h cells were fed with medium additionally supplemented with the suitable antibiotics (Table 1). Southern analyses were performed by standard procedures.
Table 1. Antibiotic resistance conveyed by pSVKeoX1, pSVZeoX1 and pSVBsdX1
| Antibiotic | Escherichia coli | ES cells | |
| pSVKeoX1 | kanamycin | 30 µg/ml | - |
| G418 | - | 200 µg/ml | |
| pSVZeoX1 | zeocin | 25 µg/ml | 25 µg/ml |
| pSVBsdX1 | blasticidin | 50 µg/ml | 5 µg/ml |
RESULTS
As with previous ways to build targeting constructs, the examples described here began with identification of mouse genomic clones from isogenic [lambda] DNA libraries (10) encompassing the region to be targeted. Complete [lambda]DASH NotI-inserts were subcloned into pZErO2.1 (Invitrogen). In contrast to previous approaches, these complete [lambda] subclones provided the backbone for the targeting construct. This presents the first merit of our approach, namely the homology arms for recombination in ES cells were not subject to further manoeuvring, thus reducing the possibility that they acquire inadvertant mutations. The subclones were then modified by ET recombination using the methodology described by Zhang et al. (4).
In the first example, we explored the possibility that the selectable gene required for ET recombination could then be used a second time as the selectable gene required for homologous recombination in ES cells, thereby greatly simplifying assembly of vertebrate targeting constructs. Currently there are three selectable genes that convey antibiotic resistances in both E.coli and vertebrate cells. They are: (i) the neomycin resistance gene (neo) which conveys kanamycin resistance in E.coli and G418 resistance in vertebrate cells; (ii) the bleomycin resistance (ble) gene which conveys zeocin resistance in both E.coli and vertebrate cells; and (iii) the blasticidin S deaminase gene (bsd) which conveys blasticidin resistance in both E.coli and vertebrate cells. All three genes were cloned into a dual expression construct utilizing the E.coli bla promoter (blaP) for expression in E.coli and the SV40 early promoter (SVe) for expression in vertebrate cells (Fig. 1A). All three constructs conveyed the appropriate resistance in both E.coli, and in mouse E14 ES cells (Table 1).
Figure 1. (A) Schematic representation of the pSVKeoX1, pSVZeoX1 and pSVBsdX1 plasmids. In pSVKeoX1, the neo resistance gene is placed under the control of the [beta]-lactamase promoter (blaP) confering neo gene expression in E.coli, and under the control of the SV40 early enhancer/promoter region (SVe) confering neo gene expression in mammalian cells. The neo transcription unit which is derived from the pBK-CMV vector (Stratagene) is flanked by loxP sites (triangles). pA, thymidine kinase polyadenylation signal. pSVZeoX1 and pSVBsdX1 are derived from pSVKeoX1 by replacement of the neo coding sequence by the ble or bsd coding sequences from pcDNA3.1/Zeo (Invitrogen) or pcDNA6/V5-His (Invitrogen), respectively. (B) Schematic representation of the pIZKeoX1 vector. pIZKeoX1 is derived from pSVKeoX1 by the insertion of an IRES-lacZ-polyadenylation signal cassette from the pBV.IRES.LacZ.PA plasmid (14). Detailed map and sequence information of the plasmids used are available on the Web site: http://www.embl-heidelberg.de/ExternalInfo/stewart/plasmids.html
To demonstrate the principle of the dual selection strategy, a subclone carrying 16.95 kb of the Ssrp1 gene was modified by ET recombination (Fig. 2A). The genomic interval between Ssrp1 exons 6 and 16 was replaced with a PCR product made from pSVZeoX1 using oligonucleotides that included ~60 nt of homology to either exon 6 or 16, by selection for zeocin resistance in E.coli. As described elsewhere (4,11,12), this ET recombination product was readily obtained (Fig. 2B and C). The ET product was directly used for targeting in ES cells by excising the 12.4 kb NotI DNA fragment containing the resistance cassette flanked by 6 and 4.1 kb Ssrp1 homology arms. Out of eight zeocin-resistant ES clones analyzed, one exibited gene targeting at the Ssrp1 locus by homologous recombination (Fig. 2D). This experiment demonstrates that a knock-out construct can be made in a single step by ET recombination. The dual resistance genes of the plasmids shown in Figure 1A are also flanked by loxP sites. Since we only wanted to knock-out Ssrp1, the PCR primers used to amplify the dual zeocin resistance gene of pSVZeoX1 excluded the loxP sites. However, for more subtle applications (13) the loxP sites can be included by choice of different PCR primers.
Figure 2. Generation of a knock-out construct in a single step. (A) The general strategy. Linear DNA, synthesized by PCR to amplify zeocin resistance gene (ble) and including 50-nt homology arms (a and b) is co-transformed with a plasmid containing the target genomic DNA into a sbcA E.coli strain. Recombinants are identified by selection on zeocin. (B) Schematic representation of the Ssrp1 protein and gene. Position of the HMG box is shown. The isolated genomic clone contains the 17 exons of the Ssrp1 gene. The map of all the EcoRI restriction sites is indicated as well as the various EcoRI fragments (R1-R6) together with their size (in kb). The ble transcription unit synthesized by PCR from pSVZeoX1 is indicated at the bottom. ET recombination was performed in the sbcA JC8679 E.coli strain, and recombinants were isolated in the presence of 25 µg/ml zeocin (Invitrogen). (C) Agarose gel showing the EcoRI restriction pattern of the vector containing the genomic fragment of Ssrp1 (Genomic clone), and one of the recombinant products. The restriction fragments R1-R6 are shown as well as the EcoRI fragment corresponding to the pZErO2.1 vector. As expected, the restriction fragments R1 and R5 are missing, and a new 1.95 kb EcoRI fragment containing the ble transcription unit is present in the recombinant construct. (D) Southern blot analysis showing homologous recombination in E14 ES cells isolated in zeocin (25 µg/ml) after electroporation of the NotI-Ssrp1::zeo targeting DNA segment. Genomic DNA from wild-type (+/+) and targeted (+/-) E14 ES cells was cut with Asp718I and hybridized with a radiolabeled probe included in the R2 EcoRI restriction fragment. The wild-type allele is expected to give a 6.5 kb fragment and the targeted allele a 6.8 kb fragment (not shown).
In the first example, PCR was used to amplify a dual selectable gene and the homology arms for ET recombination were encoded by the oligonucleotides used. In principle, it is possible to apply the same approach to amplify more complicated cassettes that contain, in addition to a dual selectable gene, other elements such as a lacZ reporter gene. However, PCR is highly mutagenic. Whereas the functionality of the selectable gene is tested by the ET recombination step, the possibility that other elements amplified by PCR are mutated is an undesirable risk that increases with size of the region to be amplified. To circumvent this risk, we developed a second approach that combines the convenience of ET recombination with the most efficient conventional restriction/ligation methodology. In the second approach, ET recombination was used to position two restriction enzyme sites exactly where desired. The two sites were included in the oligonucleotides between the ET recombination homology arms and the PCR primer for amplification of the antibiotic resistance gene (Fig. 3A). Any gene that conveys antibiotic resistance in E.coli can be used in this application (not shown). Furthermore any restriction site(s) can be chosen by this approach. Sites that will be unique in the product and convenient for the next subcloning step are selected. The product of the ET recombination step is a plasmid that has two chosen restriction sites exactly where desired. Cleavage excludes the selectable gene leaving a fragment with ends suitable for a conventional ligation reaction.
Figure 3. Combinational use of ET recombination and conventional in vitro DNA manipulations in a two-step process. (A) The general strategy. A first step of ET recombination is performed using PCR primers containing unique restriction sites for PmeI and AscI, flanking the chloramphenicol resitance gene (Cm). The resulting ET recombination product is purified and the restriction sites PmeI and AscI used to clone a suitable cassette to complete the targeting construct. (B) Schematic representation of the NSD2 protein and its different domains. The genomic clone used contains the Nsd2 gene from exon 10 to 18, encoding regions of the protein from the PHD fingers to the SET domain. The EcoRI restriction map is indicated as well as the various EcoRI fragments (R1-R5) together with their size (in kb). ET recombination products were obtained by chloramphenicol selection (100 µg/ml) in the sbcA JC8679 E.coli strain (recombinant 1). An IRES-lacZ-containing cassette was isolated from the pIZKeoX1 vector and then ligated into the ET recombination product using the PmeI/AscI restriction sites in order to generate the final targeting construct (recombinant 2). (C) Agarose gel showing the EcoRI restriction pattern of the construct containing the genomic fragment of Nsd2 (genomic clone), as well as the two types of recombinants (recombinants 1 and 2). The restriction fragments R1-R5 are shown as well as the EcoRI fragment corresponding to the pZErO2.1 vector. As expected, the EcoRI restriction fragments R4 and R5 are modified in the two recombined constructs. In the final targeting construct (recombinant 2), the R5 fragment is cut into a 900 bp fragment, R4 contains the neo transcription unit giving a 4.1 kb fragment, and a new 3.6 kb EcoRI fragment containing the IRES-lacZ sequence is obtained as predicted.
The second approach was applied to the construction of a complex targeting vector for the mouse Nsd2 gene. A 15 kb DNA fragment containing part of the Nsd2 gene was isolated from a [lambda]DASH mouse genomic library and cloned into the pZErO2.1 vector. The genomic region between Nsd2 exons 11 and 13 was replaced with a cassette made by PCR from the pMAK705 plasmid (8) confering chloramphenicol resistance in E.coli. The PCR primers were designed to include unique restriction sites for PmeI and AscI (Fig. 3A). Consequently, the ET recombination product, obtained by chloramphenicol selection, contained chosen unique restriction sites positioned precisely where desired (Fig. 3B and C). In a second, in vitro, step a cassette containing an IRES-lacZ reporter and the selectable gene neo was isolated from the pIZKeoX1 plasmid (Fig. 1B), and cloned into the ET recombination product after cleavage of the introduced PmeI and AscI sites (Fig. 3).
DISCUSSION
ET recombination in E.coli was developed to engineer large DNA molecules, such as BACs, without restriction site requirements (4,11,12). With two examples, we show the application of ET recombination to simplify previously difficult engineering exercises. The examples concern the construction of targeting vectors for homologous recombination in ES cells, although the implications are also applicable to many engineering exercises. The first example illustrates the generation of a targeting construct from a subcloned fragment of genomic DNA in a single, ET recombination, step. This example relies on the use of dual prokaryotic/eukaryotic promoters to express antibiotic resistance from single coding regions. This is the simplest approach to construct a knock-out targeting vector yet described. The second example illustrates a two-step combination of new possibilities offered by ET recombination with the strengths of existing methodology. In the ET recombination step, unique restriction enzyme sites were introduced into the subclone at freely chosen positions. In the second step, these sites were used in a classical ligation subcloning to position a complex cassette between two genomic homology arms. In this approach, the final targeting construct did not contain PCR-derived elements, hence this source of potential mutation was avoided. Other variations of two-step, or more, combinations of ET recombination with classical restriction/ligation engineering, or PCR, are clearly possible. To date, we have successfully applied similar strategies to generate targeting vectors for five genes without limitation. As described this approach allows targeting construct engineering within 1 week, which is faster than using any other method, including those based on recA recombination in E.coli (14) or homologous recombination in yeast (15,16). Although the focus of this work has been the fluent generation of targeting constructs for the mouse, the principles described can be applied to any DNA engineering exercise, whatever the purpose.
ACKNOWLEDGEMENTS
We wish to thank X. W. Yang for the gift of the pBV.IRES.LacZ.PA construct and Y. Zhang, J. Muyrers and G. Testa for discussions. This work was supported in part by a grant from the Volkswagen Foundation, Program on Conditional Mutagenesis.
REFERENCES
*To whom correspondence should be addressed. Tel: +49 6221 387 562; Fax: +49 6221 387 518; Email: stewart{at}embl-heidelberg.de Present address: Pierre-Olivier Angrand, Institut de Génétique et de Biologie Moléculaire et Cellulaire, CNRS/INSERM/ULP, BP 163, F-67404 Illkirch Cedex, C. U. de Strasbourg, France
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: jnl.info{at}oup.co.uk
Last modification:
Copyright© Oxford University Press, 1999.
This article has been cited by other articles:
![]() |
R. Geng, S. F. Geller, T. Hayashi, C. A. Ray, T. A. Reh, O. Bermingham-McDonogh, S. M. Jones, C. G. Wright, S. Melki, Y. Imanishi, et al. Usher syndrome IIIA gene clarin-1 is essential for hair cell function and associated neural activation Hum. Mol. Genet., August 1, 2009; 18(15): 2748 - 2760. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Catela, D. Bilbao-Cortes, E. Slonimsky, P. Kratsios, N. Rosenthal, and P. te Welscher Multiple congenital malformations of Wolf-Hirschhorn syndrome are recapitulated in Fgfrl1 null mice Dis. Model. Mech., May 1, 2009; 2(5-6): 283 - 294. [Abstract] [Full Text] [PDF] |
||||
![]() |
P Sipila, J Jalkanen, I T Huhtaniemi, and M Poutanen Novel epididymal proteins as targets for the development of post-testicular male contraception Reproduction, March 1, 2009; 137(3): 379 - 389. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Chan, N. Costantino, R. Li, S. C. Lee, Q. Su, D. Melvin, D. L. Court, and P. Liu A recombineering based approach for high-throughput conditional knockout targeting vector construction Nucleic Acids Res., April 10, 2007; (2007) gkm163v1. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Glaser, J. Schaft, S. Lubitz, K. Vintersten, F. van der Hoeven, K. R. Tufteland, R. Aasland, K. Anastassiadis, S.-L. Ang, and A. F. Stewart Multiple epigenetic maintenance factors implicated by the loss of Mll2 in mouse development Development, April 15, 2006; 133(8): 1423 - 1432. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Aoyama, K. Agari, G.-H. Sun-Wada, M. Futai, and Y. Wada Simple and straightforward construction of a mouse gene targeting vector using in vitro transposition reactions Nucleic Acids Res., March 22, 2005; 33(5): e52 - e52. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. W. Bogue Genetic Models in Applied Physiology: Invited Review: Functional genomics in the mouse: powerful techniques for unraveling the basis of human development and disease J Appl Physiol, June 1, 2003; 94(6): 2502 - 2509. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. van der Weyden, D. J. Adams, and A. Bradley Tools for targeted manipulation of the mouse genome Physiol Genomics, December 3, 2002; 11(3): 133 - 164. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Orford, M. Nefedov, J. Vadolas, F. Zaibak, R. Williamson, and P. A. Ioannou Engineering EGFP reporter constructs into a 200 kb human {beta}-globin BAC clone using GET Recombination Nucleic Acids Res., September 15, 2000; 28(18): e84 - e84. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||









