Complex chromosomal rearrangements (deletions, inversions, translocations) are a hallmark of human tumour cells. Yet, the generation of animal models for gross chromosomal abnormalities still presents a formidable challenge. Here, we describe a versatile procedure for chromosomal engineering that was used to generate an ES cell line with a megabase deletion encompassing the tumour suppressor gene neurofibromatosis-1 (Nf-1) on mouse chromosome 11, which is often deleted in tumours of neural crest origin. Homologous recombination into sites flanking Nf-1 was used to introduce artificial sequences (triple-helix, loxP, vector backbone) that can be employed for in vitro recovery of intervening sequences or the generation of in vivo deletions. This strategy may be developed into a scheme by which large chromosomal regions with precisely defined end points may be excised from mammalian cells and reintroduced after suitable in vitro modification.
Malignant cells often carry chromosomal deletions thought to encompass tumour suppressor genes (Johansson et al., 1993). Such deletions range from small intragenic deletions to loss of large chromosomal regions encompassing thousands of genes. However, using conventional methods of gene inactivation in the mouse, deletions in the megabase range are still difficult to reproduce.
Here, we describe an integrated methodological approach that allows predetermined chromosomal deletions to be achieved in the genome of mouse embryonic stell cells in vivo and in vitro and, at the same time, to recover intervening sequences in a clonable form. As a model, we have used a region on mouse chromosome 11, which contains the neurofibromatosis-1 tumour suppressor gene that is often deleted in tumours derived from neural crest cells (Viskochil et al., 1993).
We have previously described the construction and use of neomycin and hygromycin resistance cassettes containing an artificial triple-helix site and a synthetic lac operator for integration into mammalian genomes (Nehls et al., 1994). This strategy allowed restriction endonuclease cleavage at pre-defined sites in vitro and subsequent affinity purification. However, cloning of the isolated fragment required ligation to a vector backbone prior to transformation into a suitable host. To improve this basic principle, we have subsequently modified one targeting vector by the incorporation of a suitable vector backbone thus replacing a two-fragment ligation with an intramolecular circularization step (data not shown). In a further modification of this procedure described here, loxP sites have been added to the targeting vectors. This eliminates the need for any DNA ligation step since recombination at tandemly arranged loxP sequences catalyzed by cre recombinase produces circular molecules (Hochman et al., 1983). A further significant advantage of this strategy is that deletions can be generated in vivo. Previously, Cre-loxP recombination has been used to generate site-directed chromosomal translocations (Smith et al., 1995; van Deursen et al., 1995) and large deletions (Ramirez-Solis et al., 1995; Li et al., 1996). Our deletion strategy extends these findings by incorporating the facility to recover intervening sequences in clonable form.
We have previously determined the transcriptional orientation of the Nf-1 gene relative to physically linked sequence-tagged sites on chromosome 11 (Nehls et al., 1995); this information simplified the design of the targeted deletion experiment described here. Two consecutive homologous recombination events introduced the desired artificial sequences at markers D11Bhm100 and D11Bhm109, flanking the Nf-1 gene at either end (Figure 1a). The genetic distance between the two STS markers is estimated to be 0.18 cM (Montgomery et al., 1998); the physical distance is not exactly known but is greater than 500 kb (Nehls et al., 1995). After the first targeting event, the artificial sequence upstream (i.e. centromeric) of the Nf-1 gene (Figure 1b) contains a loxP sequence, followed by a tkneo fusion gene and a F′-plasmid based vector backbone, derived from the peloBAC vector (Shizuya et al., 1992). After the second targeting, the telomeric D11Bhm109 sequence (located downstream of the Nf-1 gene) is supplied with a loxP sequence and a hygromycin resistance gene. Because the second targeting event may affect either homolog of chromosome 11, cell lines with targeted integrations on one or two chromosomes, respectively, were subsequently selected by fluorescence in situ hybridization (Figure 2). Cell line D7-2 (Figure 2a) carries both insertions on one chromosome, whereas in F8-1 cells, the integrations occurred on different chromosomes. To achieve the desired intrachromosomal deletion, D7-2 cells were transfected with an expression plasmid encoding a GFP-cre fusion protein (Gagneten et al., 1997).
After 48 h, 2.5% of transfected cells were found to express GFP and could be purified by fluorescence activated cell sorting. Individual sub-clones were analysed after a further 7 days.
Cre recombinase activity results in intrachromosomal rearrangement and the deletion of genomic DNA encompassing the Nf-1 gene; the excised fragment is contained in an extrachromosomal episome carrying a F′-plasmid based origin of replication (Figure 3a). Because this episome lacks a centromere and possibly also a eukaryotic origin of replication, it will be lost during subsequent cell divisions. The resulting sub-clones will, therefore, be hygromycin resistant (as the hyg gene is located telomeric to the loxP site) but neomycin sensitive (as the tkneo gene is contained in the episome). Fourteen hygRneoS sub-clones of D7-2 cells derived after transient cre expression were tested for the deletion event by the PCR strategy shown in Figure 3a. Ten clones (for an example, see Figure 3b) contained the expected deletion as detected by primer combination 156 and 157; the parental D7-2 cell line is negative in this assay, because the relevant sequences are at least 500 kb apart. An amplimer diagnostic for the deleted fragment (primer combination 160 and 30) is not detectable, consistent with the notion that the episome is no longer present in these cells. This result is confirmed by Southern filter hybridization analysis (Figure 3c) using a tkneo probe and a probe derived from the telomeric end of the SacI fragment carrying the targeted insertion loxP-hyg cassette (c.f. Figure 1c). Although the in vivo generated BACs appear to be unstable in ES cells, it is conceivable that the provision of a suitable eukaryotic origin of replication may lead to a stable maintenance of the excised fragments in episomal form.
The intrachromosomal deletion event apparently occurs with such high frequency (about 70% in the case above) that enrichment for cre-expressing cells suffices to obtain sub-clones with the desired genomic modification, at least in case of the Nf-1 region. In less favourable situations, our experimental design allows for the enrichment of cell clones containing the targeted deletion by counter-selection with ganciclovir (as the tkneo fusion gene is contained within the unstable episome). We have also attempted to measure the frequency of cre-mediated chromosomal translocation in F8-1 cells. However, PCR analysis of a pool of 200 clones did not show unambiguous evidence for the desired inter-chromosomal rearrangement (data not shown). This is in agreement with the findings of other studies, which have reported translocation frequencies of less than 1 in 1000 (van Deursen et al., 1995; Smith et al., 1995). It is unclear at present, why the intra-chromosomal rearrangement occurs at a much higher frequency than the inter-chromosomal event.
In some instances, the in vitro recovery of the deleted DNA fragment for cloning purposes may be desirable. Our previously described strategy encompassed the facility for triple-helix-mediated chromosomal cleavage as both selection markers carry the previously described optimized THS-sequence in addition to an artificial lac operator sequence for subsequent affinity purification of DNA (Nehls et al., 1994). We, therefore, wished to ascertain whether the in vivo excision by cre recombinase of the Nf-1 region was also possible in vitro. Because of the size of the resulting episome, cre-mediated excision should take place in conditions minimizing shear forces. To this end, intact chromosomal DNA was prepared in agarose plugs and treated with a highly purified bacterial His-tag-cre fusion protein. DNA reaction products were purified and subjected to PCR analysis. The results shown in Figure 3b (right panel) indicate that the deletion product is readily detectable. However, in contrast to the in vivo situation, the assay designed to detect the resulting episome (primer combination 160/30) is also positive, indicating the presence of the Nf-1 BAC clone.
We have chosen a single-copy bacterial origin of replication for our present studies, as fragments cloned in such vectors have been shown to be stable in suitable E. coli hosts. This is often not the case with yeast artificial chromosomes where mammalian DNA tends to become rearranged during propagation in yeast cells. However, with the advent of rapid strategies for homologous recombination in bacteria (Kempkes et al., 1995; Zhang et al., 1998), the in vitro manipulation of such fragments before reintroduction into mammalian cells (previously a major advantage of the yeast system) has become a feasible undertaking. Future work, however, will have to address conditions by which in vitro generated BACs can be efficiently transferred into bacteria despite a large excess of genomic DNA; encouraging results have already been reported (Sheng et al., 1995; Zhu and Dean, 1999).
Because the targeted deletion leaves a single loxP site of known orientation in the chromosome, the experimental strategy exemplified here may ultimately be developed into a scheme to replace large chromosomal regions with precisely defined endpoints from one source by homologous regions from another source in a manner that preserves chromosomal context and orientation.
Gagneten S, Le Y, Miller J and Sauer B. . 1997 Nucl. Acids Res. 25: 3326–3331.
Hochman L, Segev N, Sternberg N and Cohen G. . 1983 Virology 131: 11–17.
Johansson B, Mertens F and Mitelman F. . 1993 Genes Chrom. Cancer 8: 205–218.
Kempkes B, Pich D, Zeidler R and Hammerschmidt W. . 1995 Proc. Natl. Acad. Sci. USA 92: 5875–5879.
Li ZW, Stark G, Gotz J, Rulicke T, Gschwind M, Huber G, Muller U and Weissmann C. . 1996 Proc. Natl. Acad. Sci. USA 93: 6158–6162.
Lichter P, Tang CC, Call K, Hermanson G, Evans GA, Housman D and Ward DC. . 1990 Science 247: 64–69.
Montgomery JC, Silverman KA and Buchberg MA. . 1998 Mammalian Genome 8: 5215–5240.
Nehls M, Krause S and Boehm T. . 1994 Mammalian Genome 5: 183–186.
Nehls M, Kyewski B, Messerle M, Waldschütz R, Schüddekopf K, Smith AJH and Boehm T. . 1996 Science 272: 886–889.
Nehls M, Lüno K, Schorpp M, Pfeifer D, Krause S, Matysiak-Scholze U, Dierbach H and Boehm T. . 1995 Mammalian Genome 6: 321–331.
Ramirez-Solis R, Liu P and Bradley A. . 1995 Nature 378: 720–724.
Sheng Y, Mancino V and Birren B. . 1995 Nucl. Acids Res. 23: 1990–1996.
Shizuya H, Birren B, Kim UJ, Mancino V, Slepak T, Tachiiri Y and Simon M. . 1992 Proc. Natl. Acad. Sci. USA 15: 8794–8797.
Smith AJ, De Sousa MA, Kwabi-Addo B, Heppell-Parton A, Impey H and Rabbitts P. . 1995 Nat. Genet. 9: 376–385.
van Deursen J, Fornerod M, Van Rees B and Grosveld G. . 1995 Proc. Natl. Acad. Sci. USA 92: 7376–7380.
Viskochil D, White R and Cawthon R. . 1993 Annu. Rev. Neurosci. 16: 183–205.
Zhang Y, Buchholz F, Muyrers JP and Stewart AF. . 1998 Nat. Genet. 20: 123–128.
Zhu H and Dean RA. . 1999 Nucl. Acids Res. 27: 910–911.
We thank Drs M Nehls and M Messerle for the initial isolation and characterization of genomic clones containing markers D11Bhm100 and D11Bhm109, Dr B Sauer for provision of the GFP-cre expression plasmid, and M Sator-Schmitt and H Kohler for excellent technical assistance. This work was initiated at the German Cancer Research Center, Heidelberg, and completed at the Max-Planck-Institute for Immunobiology, Freiburg. Financial support from the Deutsche Forschungsgemeinschaft is gratefully acknowledged.
About this article
Cite this article
Schlake, T., Schupp, I., Kutsche, K. et al. Predetermined chromosomal deletion encompassing the Nf-1 gene. Oncogene 18, 6078–6082 (1999) doi:10.1038/sj.onc.1203021
- cre recombinase
- chromosome engineering
Nature Reviews Genetics (2001)
Proceedings of the National Academy of Sciences (2000)
International Journal of Cancer (2000)