Efficient Cre-loxP−induced mitotic recombination in mouse embryonic stem cells

FLP/FRT-induced mitotic recombination provides a powerful method for creating genetic mosaics in Drosophila and for discerning the function of recessive genes in a heterozygous individual. Here we show that mitotic recombination can be reproducibly induced in mouse embryonic stem (ES) cells, by Cre/loxP technology, at frequencies ranging from 4.2 10^-5 (Snrpn) to 7.0 10^-3 (D7Mit178) for single allelic loxP sites, and to 5.0 10^-2 (D7Mit178) for multiple allelic lox sites, after transient Cre expression. Notably, much of the recombination occurs in G2 and is followed by X segregation, where the recombinant chromatids segregate away from each other during mitosis. It is X segregation that is useful for genetic mosaic analysis because it produces clones of homozygous mutant daughter cells from heterozygous mothers. Our studies confirm the predictions made from studies in Drosophila¹ that suggest that X segregation will not be limited to organisms with strong mitotic pairing, because the forces (sister-chromatid cohesion) responsible for X segregation are an elemental feature of mitosis in all eukaryotes. Our studies also show that genetic mosaic analysis in mice is feasible, at least for certain chromosomal regions.

Figure 1. Recombination cassettes, chromatid segregation patterns and targeted loci. a, Hprt minigene recombination cassettes. The 3' cassette contains a PGK promoter (black box), Hprt exon 1 and 2, the loxP-Hprt intron (a loxP in the XbaI site of human HPRT1 intron 2), a PolII promoter (green box) that drives neor expression, and a bovine growth hormone polyA site (yellow box). The 5' cassette contains a puror selection marker, the same loxP-intron, Hprt exons 3−9 and an SV40 polyA site (gray box). Cre/loxP-induced mitotic recombination produces a wildtype Hprt minigene and linked puror and neor selection markers. Arrows denote loxP sites. b, Segregation patterns of recombinant chromatids after G2 recombination. Note that only X segregation makes homozygous all loci distal to the recombination event. Homologous chromosomes are represented as red and blue lines; an asterisk denotes a mutation. c, Location of genetic loci targeted by the Hprt minigene recombination cassettes. Genetic maps are drawn to scale (for exact map locations see the Mouse Genome Informatics web site). Full Figure and legend (35K)

	Top

To induce and detect mitotic recombination events at multiple sites in mouse ES cells, we constructed two recombination cassettes (3', 5') that contained complementary, but nonfunctional, halves of a human HPRT1 minigene (Fig. 1a). Recombination between loxP sites embedded in the intron of each cassette produced a wildtype HPRT1 minigene that could be scored by the ability of Hprt-deficient AB2.2 ES cells to grow in HAT (hypoxanthine/aminopterin/thymine) medium (HAT resistance, HAT^r; Fig. 1a)²¹. Recombinant chromatids follow either G2 X or G2 Z segregation after G2 recombination (Fig. 1b). In initial experiments, we targeted the two recombination cassettes to allelic positions at the proximal end of chromosome 7 (D7Mit178; Fig. 1c). We chose chromosome 7 because it contains several coat-color genes such as tyrosinase (Tyr; also known as c) and pink-eyed dilution (p) that can be used to score for mitotic recombination events in the whole animal. Cre recombinase was then transiently expressed after electroporation into doubly targeted cells (DT1E9 cells; Table 1), and the recombinant cells were selected in HAT medium 48 h later. The mitotic recombination frequency, calculated by dividing the number of HAT^r colonies by the total number of colonies surviving electroporation in the absence of HAT selection, was 7.0 10^-3 at the D7Mit178 locus. This frequency is tenfold greater than the highest frequency of interchromosomal recombination reported previously for mouse ES cells²⁰.

Table 1. Recombination efficiency under transient Cre expression Full Table

To determine whether biased segregation of recombinant chromatids occurs in mouse cells, we needed a marker that could distinguish both chromosome 7 homologs. As AB2.2 ES cells are derived from an inbred mouse strain²², we used a differentially methylated imprinted region that is present in the Snrpn promoter²³ and maps 28 cM distal to the D7Mit178 locus (Fig. 1c). A SacII restriction site in this region is unmethylated on the paternal allele but is methylated on the maternal allele in ES cells (Fig. 2a). Restriction fragments of SacII-digested genomic DNA can therefore be used to follow segregation of the recombinant chromatids. Analysis of 240 HAT^r colonies from doubly targeted DT1E9 ES cells (Table 1) showed, unexpectedly, that all possessed only the paternal Snrpn allele (Fig. 2a and data not shown). We subsequently examined 192 HAT^r colonies from other double-targeted cell lines, and again all of the colonies retained only one of the two parental Snrpn alleles (Fig. 2a and data not shown). These results suggest that all induced mitotic recombination at the D7Mit178 locus occurs in G2 and is followed by X segregation, at least at this level of detection. Depending upon which parental allele is targeted by the 5' cassette, the resulting HAT^r cells are either di-paternal or di-maternal for chromosome 7.

Figure 2. D7Mit178 recombination occurs in G2 and is followed by X segregation. a, X segregation of D7Mit178-targeted recombinant chromatids. DNA from DT1E9 HATr colonies was digested with SacII/TaqI and hybridized with a probe from a differentially methylated imprinted region located in the Snrpn promoter. Note that wildtype AB2.2 ES cells have both paternal (p) and maternal (m) Snrpn alleles, whereas all DT1E9 HATr colonies are di-paternal (dp) for chromosome 7 (lanes 1−8). The faint maternal band represents contaminating feeder cells. In 50% of the ES cells, the 5' Hprt minigene cassette is targeted to the maternal chromosome 7. HATr colonies derived from these ES cells are di-maternal (dm) for chromosome 7 (lanes 9−16). b, Southern blot of HATr colonies from triply targeted cell lines (3', 5' and bsdr). Recombinant chromatids can be traced by the loss or retention of the targeted allele (11.0-kb EcoRV fragment) of the Igf1r locus. HATr colonies from cell line 1E9-396-G11 all lose the bsdr cassette and are thus sensitive to blasticidin (lanes 1−8), whereas cell line 1E9-396-B4 gives rise to HATr colonies that all retain the bsdr cassette (lanes 9−16). c, DNA samples from AB2.2, dp (1E9-1, 1F2-1) or dm (3G5-2, 4E11-1) chromosome 7 ES cells were examined for Snrpn gene dosage using Snrpn and Bcl11a control probes. No change in Snrpn dosage was observed. Full Figure and legend (35K)

To determine whether the frequency of induced mitotic recombination is the same for other proximal chromosome regions, we targeted the Hprt 3' and 5' cassettes to allelic positions in the proximal region of chromosome 11 (D11Mit71; Fig. 1c). The frequency of induced mitotic recombination at D11Mit71 after transient Cre expression in doubly targeted DT26D11 cells was 3.5 10^-4 (Table 1). This frequency is similar to that reported for nonallelic loxP sites on distal chromosome 11 (4.7 10^-4)²⁰, but is 20-fold lower than the frequency of mitotic recombination at the D7Mit178 locus. The frequency thus varies for different proximal chromosomal regions.

Cis recombination between two loxP sites separated by a few kilobase pairs of DNA varies dramatically for different loci²⁵ and is thought to reflect position effects. To determine whether the low frequency of mitotic recombination at D11Mit71 reflects a position effect, we targeted the Hprt 3' and 5' cassettes to another chromosome 11 locus, Wnt3, that has a high cis recombination frequency (Fig. 1c)²⁰. The frequency of induced mitotic recombination at Wnt3 in doubly targeted DT17E8 cells was 5.1 10^-4 (Table 1), which was no greater than the frequency of mitotic recombination at the D11Mit71 locus. The low frequency of mitotic recombination at Wnt3, a locus with a high cis-recombination frequency, thus suggests that simple position effects do not account for the differences in induced mitotic recombination between chromosomes 7 and 11.

Figure 3. Segregation of D11Mit71 recombinant chromatids. G2 X segregation events are marked by the disappearance of the wildtype Wnt3 allele (20.0-kb EcoRI fragment). Full Figure and legend (30K)

Mouse chromosome 7 contains a number of well defined imprinting domains homologous to the Prader-Willi syndrome (PWS) and Angelman syndrome (AS) regions on human chromosome 15q11−q13 and the H19/IGF2 imprinting domain on 11p15. Preferential S-phase pairing of the PWS and AS imprinted regions has been observed in mouse fibroblasts²⁶. Preferential S-phase pairing has also been observed in human T lymphocytes by three-dimensional fluorescence in situ hybridization (FISH)²⁷. Loss of imprinting in PWS patients disrupts this association. It is conceivable that the efficient mitotic recombination observed during the G2 phase of the cell cycle at the D7Mit178 locus is mediated by this S-phase pairing. To address this question, we targeted the Hprt 3' and 5' cassettes to allelic positions in intron 2 of the Igf1r locus. We also targeted these cassettes to allelic positions in intron 4 of Snrpn, which is located within the PWS/AS imprinted domains (Fig. 1c). If S-phase pairing occurs between imprinting domains in ES cells, then the frequency of induced mitotic recombination at these loci might be greater than at D7Mit178, as these loci map within or closer to the imprinting domains than D7Mit178. This was not the case, however, as the frequency of mitotic recombination in doubly targeted Igf1r (DT40B2) and Snrpn (DT35C4) cells was greatly reduced compared with doubly targeted D7Mit178 (DT1E9) cells (Table 1). Further analysis of HAT^r colonies from recombination at the Snrpn locus indicated that 23% of them were G2 X events and 77% were G1 or G2 Z events (data not shown).

We also measured the frequency of induced mitotic recombination in two HAT^r cell lines that have di-paternal or di-maternal chromosome 7s generated in previous mitotic recombination experiments at the D7Mit178 locus. Each cell line has a reconstituted HPRT1 minigene in one D7Mit178 allele and an Hprt 5' cassette in the other allele (Fig. 1b). To measure the frequency of mitotic recombination in these uniparental disomy cell lines, we replaced the HPRT1 minigene in each cell line with an Hprt 3' cassette by another round of gene targeting at the D7Mit178 locus. These cells (DT38B12, di-paternal for chromosome 7; DT39A11, di-maternal for chromosome 7; Table 1) were then subcloned once to avoid contamination from the parental HAT^r cells and transfected with the Cre expression plasmid. Notably, the induced frequency of mitotic recombination in the di-paternal or di-maternal cells was similar to that observed in DT1E9 cells, which carry both a paternal and a maternal copy of chromosome 7 (Table 1). Genomic imprinting is therefore not responsible for the high frequency of induced mitotic recombination at the D7Mit178 locus.

To determine whether a similar arrangement of loxP sites might also increase the frequency of Cre-mediated mitotic recombination in mouse ES cells, we replaced the single loxP site in the Hprt 5' intron with a 500-bp fragment of DNA that was flanked by loxP sites at each end. We then targeted this modified cassette to one D7Mit178 allele and an unmodified Hprt 3' cassette to the other allele. The frequency of induced mitotic recombination in these doubly targeted cells was 1.2%, which represents nearly a twofold increase over that observed for single loxP sites in DT1E9 cells. Analysis of 192 HAT^r colonies showed that all were di-maternal or di-paternal for chromosome 7, indicating that this arrangement of loxP sites does not affect the subsequent segregation of recombinant chromatids (data not shown) because the cis recombination is a much more efficient process than the trans recombination between homologous chromosomes.

To determine whether adding more lox sites might increase the frequency even further, we constructed two new recombination cassettes that contained neo^r and puro^r flanked by three lox site variants: lox5171, lox2272 and either lox66 (Hprt M3') or lox71 (Hprt M5'; Fig. 4a). These variant lox sites can efficiently recombine with themselves but not with each other (with the exception of lox66 and lox71, which can recombine to generate a mutant lox site that will not recombine with itself or a wildtype loxP site)²⁸. These lox site variants were used to prevent cis recombination between the tandemly linked lox sites. Cis recombination between the variant lox sites located on each side of neo^r or puro^r or trans recombination between variant lox sites on homologous chromosomes can still occur, however, thus increasing the Cre recombinase target size from one loxP site in the original cassettes to three lox sites in the modified cassettes. We then targeted the M3' and M5' cassettes to allelic positions at the D7Mit178 locus and measured the frequency of induced mitotic recombination after transient Cre expression (DT31G8). Notably, 5% of the doubly targeted DT31G8 cells containing these modified cassettes underwent recombination upon transient Cre expression (Table 1).

Figure 4. Multiple lox sites increase the frequency of Cre-induced mitotic recombination. a, Structure of the modified Hprt M3' and M5' targeting cassettes (see Fig. 1, legend). Efficient Cre recombinase−mediated cis recombination between the variant lox sites results in the excision of both neor and puror (data not shown). Less efficient trans recombination generates a wildtype Hprt minigene with three variant lox sites in the Hprt intron. b, HATr colonies from DT31G8 doubly targeted cells examined by Southern-blot analysis using an Snrpn promoter-region probe. Approximately 35% of the colonies retained both maternal (m) and paternal (p) alleles. The rest were di-maternal for chromosome 7. Full Figure and legend (29K)

To determine whether the use of multiple lox site variants alters the frequency of G2 X segregation, we analyzed the DT31G8 HAT^r colonies by Southern-blot analysis using the Snrpn promoter probe. Notably, approximately 35% of the HAT^r colonies carried both maternal and paternal copies of chromosome 7 (G1 recombination or G2 recombination followed by Z segregation; Fig. 4b). This contrasts our observations on DT1E9 cells with one loxP site, where all the HAT^r colonies were di-maternal or di-paternal for chromosome 7. Although this may reflect a true increase in the frequency of G1 recombination or G2 Z segregation, a much more likely explanation is that recombination still occurs in G2, but these extra lox sites allow for a second exchange between the sister chromatids. This second exchange would be very efficient, as it involves sister chromatids rather than non-sister chromatids. If the second exchange were followed by X segregation, the end result would appear as G1 recombination or G2 Z segregation (Fig. 6b). In support of this hypothesis, we also examined the HAT^r colonies generated from doubly targeted cells that contained several tandemly linked wildtype loxP sites in the intron of the Hprt 3' and 5' recombination cassettes (no DNA stuffer fragment). Again, the frequency of G2 X segregation decreased with an increased number of loxP sites (data not shown).

Figure 6. Mitotic chiasma formation directs recombinant chromatid segregation (adapted from Beumer et al.1). Homologous chromosomes are depicted as either red or blue lines. a, Cre/loxP catalyzes recombination between two homologous chromosomes in G2. Short thin lines between sister chromatids depict cohesion complexes. Exchange of non-sister chromatids leads to mitotic chiasma formation. When a cross-over point is near the centromere, the majority of original sister chromatids are still held together by cohesion. Recombinant chromatids are forced to segregate to different daughter cells because of the physical constraints imposed by the mitotic chiasma (X segregation). b, If an exchange occurs between two sister chromatids after recombination between non-sister chromatids, then the position of the sister chromatids in the mitotic chiasma will be switched. This results in segregation products that are identical to those of G2 Z or G1 recombination. Full Figure and legend (29K)

Figure 5. Induced mitotic recombination under constitutive Cre expression. Several doubly targeted ES cell lines analyzed in transient Cre experiments (Table 1) were electroporated with linearized PL318 containing the PGK-bsdr-bpA and PGK-cre-bpA cassettes, and the number of Bsdr HATr colonies was measured at various times after electroporation. Note that nearly 100% of DT31G8 Bsdr colonies contained HATr cells after 8 d of Cre expression. Full Figure and legend (22K)

	Top

Genetic mosaics created by FLP/FRT-induced mitotic recombination have provided a powerful tool for genetic studies in Drosophila. Here we show that mitotic recombination can be reproducibly induced in mouse ES cells with Cre/loxP technology. The frequency varies across the genome, however, and for four of five loci tested in these experiments the frequency was low, ranging from 4.2 10^-5 (Snrpn) to 5.1 10^-4 (Wnt3) for single allelic loxP sites after transient Cre expression. This low frequency of mitotic recombination is consistent with previous experiments showing that homologous mouse chromosomes are not paired in the nucleus, unlike in Drosophila, where homologous chromosomes are paired in the interphase. The frequency at the fifth locus, D7Mit178, was 13 times more frequent (7.0 10^-3) than the next best locus, Wnt3 (5.1 10^-4). D7Mit178 was also exceptional in that all recombination occurred in G2 and was followed by X segregation. In contrast, only 60% of the recombination at D11Mit71, and 23% of the recombination at Snrpn, was G2-X. Perhaps only D7Mit178 alleles are able to transiently associate during the late S-G2 phase of the cell cycle, and this transient association promotes the high frequency of G2 recombination observed at the D7Mit178 locus.

Increased mitotic recombination has also been reported in mice with Bloom syndrome, which carry a mutation in the mouse homolog of the Escherichia coli RecQ DNA helicase gene²⁹. The frequency of LOH on chromosome 11 (per cell per generation) in mice with Bloom syndrome (4.2 10^-4) is similar to the frequency of Cre/loxP-induced mitotic recombination on chromosome 11 in transient Cre experiments (3.5−5.1 10^-4). However, this frequency is 100−fold lower than the highest frequency of mitotic recombination we obtained on chromosome 7 with multiple lox sites (5.0 10^-2). Even with this mitotic recombination frequency, greatly accelerated tumorigenesis has been observed in Blm-deficient mice that also carry a heterozygous mutation in Apc²⁹. Another advantage of induced mitotic recombination is that the site and the timing of the recombination can be controlled. When recombination is induced near the centromere, all loci distal to the recombination will become homozygous. In mice with Bloom syndrome, recombination can occur anywhere on the chromosome, and proximal loci will become homozygous less frequently than distal loci.

	Top

Genomic DNA fragments used to generate targeting constructs were isolated from a phage FIXII mouse 129/SvEv genomic library (Stratagene) and subcloned into pBluescript. We isolated D7Mit178 clones using a single-copy probe that maps near the CA repeat, which defines the D7Mit178 locus; Snrpn clones using a probe from intron 4; and Igf1r clones using a PCR product from Igf1r intron 3. We obtained the D11Mit71 genomic clone from B. Zheng (Baylor College of Medicine). The recombination cassettes were targeted to a BglII site in D7Mit178 (with a 1.0-kb BglII deletion), an XbaI site in Snrpn intron 4, a HindIII site in Igf1r intron 2, an NcoI site in D11Mit71 (with a 0.8-kb NcoI deletion) and a ClaI site in Wnt3 exon 4. The targeting frequency ranged from 5% at D7Mit178 to over 80% at Wnt3 in AB2.2 ES cells with positive and negative selection. We tested at least two independent lines from each doubly or triply targeted cell type in each experiment.

We obtained AB2.2 ES cells, STO feeder cells, an HPRT1 minigene and PGK-puro^r-bpA and PoII-neo^r-bpA selection cassettes (Sanger Centre, UK). We cultured ES cells and identified targeted clones as described²¹. We analyzed all targeted clones to ensure that only one integrated cassette was present in a cell and in the targeted locus. We did not observe any background recombination in all the targeted cell lines (no HAT^r colonies in the absence of Cre recombinase). In transient Cre expression experiments, we electroporated 1.0 10⁷ ES cells with 40 g PGK-Cre-bpA plasmid. HAT selection generally started 48 h after electroporation. We used HT (hypoxanthine and thymidine) medium to release HAT selection.

We constructed PGK-bsd^r-bpA from a blasticidin-resistance plasmid (pEM7-Bsd, Invitrogen). ES cells expressing Bsd were selected with 12.5 g ml^-1 of blasticidin (Invitrogen). We made plasmid PL318 by ligating the PGK-bsd^r-bpA and PGK-Cre-bpA selection cassettes into pBluescript.

We carried out constitutive Cre experiments as follows. First, we electroporated doubly targeted cells with 20 g KpnI (or KpnI/SacII)−linearized PL318 plasmid and divided the electroporation mixture equally into six plates with bsd^r STO feeder cells. We applied bsd selection 24 h after electroporation, and then added HAT medium to the plates, 2, 4 and 8 d after electroporation.

	Top

1. Beumer, K.J., Pimpinelli, S. & Golic, K.G. Induced chromosomal exchange directs the segregation of recombinant chromatids in mitosis of Drosophila. Genetics 150, 173−188 (1998). | PubMed  | ISI | ChemPort | Add to Connotea (beta)  |
2. Rossant, J. & Spence, A. Chimeras and mosaics in mouse mutant analysis. Trends Genet. 14, 358−363 (1998). | Article | PubMed  | ISI | ChemPort | Add to Connotea (beta) |
3. Favor, J. & Neuhauser-Klaus, A. Genetic mosaicism in the house mouse. Annu. Rev. Genet. 28, 27−47 (1994). | Article | PubMed  | ISI | ChemPort | Add to Connotea (beta) |
4. Xu, T. & Rubin, G.M. Analysis of genetic mosaics in developing and adult Drosophila tissues. Development 117, 1223−1237 (1993). | PubMed  | ISI | ChemPort | Add to Connotea (beta)  |
5. Pimpinelli, S. & Ripoll, P. Nonrandom segregation of centromeres following mitotic recombination in Drosophila melanogaster. Proc. Natl Acad. Sci. USA 83, 3900−3903 (1986). | PubMed  | ChemPort | Add to Connotea (beta)  |
6. Stevens, N.M. A study of the germ cells of certain Diptera, with reference to the heterochromosomes and the phenomena of synapsis. J. Exp. Zool. 5, 359−383 (1908).
7. Metz, C.W. Chromosome studies on the Diptera II: the paired association of chromosomes in the Diptera, and its significance. J. Exp. Zool. 21, 213−279 (1916).
8. Golic, K.G. Site-specific recombination between homologous chromosomes in Drosophila. Science 252, 958−961 (1991). | PubMed  | ISI | ChemPort | Add to Connotea (beta)  |
9. Xu, T., Wang, W., Zhang, S., Stewart, R.A. & Yu, W. Identifying tumor suppressors in genetic mosaics: the Drosophila lats gene encodes a putative protein kinase. Development 121, 1053−1063 (1995). | PubMed  | ISI | ChemPort | Add to Connotea (beta)  |
10. Cavenee, W.K. et al. Expression of recessive alleles by chromosomal mechanisms in retinoblastoma. Nature 305, 779−784 (1983). | Article | PubMed  | ISI | ChemPort | Add to Connotea (beta) |
11. Mortensen, R.M., Conner, D.A., Chao, S., Geisterfer-Lowrance, A. & Seidman, J.G. Production of homozygous mutant ES cells with a single targeting construct. Mol. Cell. Biol. 12, 2391−2395 (1992). | PubMed  | ISI | ChemPort | Add to Connotea (beta)  |
12. Shao, C. et al. Mitotic recombination produces the majority of recessive fibroblast variants in heterozygous mice. Proc. Natl Acad. Sci. USA 96, 9230−9235 (1999). | Article | PubMed  | ChemPort | Add to Connotea (beta) |
13. Bateman, A.J. A probable case of mitotic crossing-over in the mouse. Genet. Res. 9, 375 (1967). | PubMed  | ISI | ChemPort | Add to Connotea (beta)  |
14. Panthier, J.J., Guenet, J.L., Condamine, H. & Jacob, F. Evidence for mitotic recombination in Wei/+ heterozygous mice. Genetics 125, 175−182 (1990). | PubMed  | ISI | ChemPort | Add to Connotea (beta)  |
15. Fisher, G., Stephenson, D.A. & West, J.D. Investigation of the potential for mitotic recombination in the mouse. Mutat. Res. 164, 381−388 (1986). | Article | PubMed  | ISI | ChemPort | Add to Connotea (beta) |
16. German, J. Cytological evidence for crossing-over in vitro in human lymphoid cells. Science 144, 298−301 (1964). | PubMed  | ISI | ChemPort | Add to Connotea (beta)  |
17. Ellis, N.A. et al. The Bloom's syndrome gene product is homologous to RecQ helicases. Cell 83, 655−666 (1995). | Article | PubMed  | ISI | ChemPort | Add to Connotea (beta) |
18. Johnson, R.D. & Jasin, M. Double-strand-break-induced homologous recombination in mammalian cells. Biochem. Soc. Trans. 29, 196−201 (2001). | Article | PubMed  | ISI | ChemPort | Add to Connotea (beta) |
19. Herault, Y., Rassoulzadegan, M., Cuzin, F. & Duboule, D. Engineering chromosomes in mice through targeted meiotic recombination (TAMERE). Nature Genet. 20, 381−384 (1998). | Article | PubMed  | ISI | ChemPort | Add to Connotea (beta) |
20. Zheng, B., Sage, M., Sheppeard, E.A., Jurecic, V. & Bradley, A. Engineering mouse chromosomes with Cre-loxP: range, efficiency, and somatic applications. Mol. Cell. Biol. 20, 648−655 (2000). | Article | PubMed  | ISI | ChemPort | Add to Connotea (beta) |
21. Ramirez-Solis, R., Liu, P. & Bradley, A. Chromosome engineering in mice. Nature 378, 720−724 (1995). | Article | PubMed  | ChemPort | Add to Connotea (beta) |
22. Matzuk, M.M., Finegold, M.J., Su, J.G., Hsueh, A.J. & Bradley, A. -inhibin is a tumour-suppressor gene with gonadal specificity in mice. Nature 360, 313−319 (1992). | Article | PubMed  | ISI | ChemPort | Add to Connotea (beta) |
23. Yang, T. et al. A mouse model for Prader-Willi syndrome imprinting-centre mutations. Nature Genet. 19, 25−31 (1998). | Article | PubMed  | ISI | ChemPort | Add to Connotea (beta) |
24. Gabriel, J.M. et al. A transgene insertion creating a heritable chromosome deletion mouse model of Prader-Willi and angelman syndromes. Proc. Natl Acad. Sci. USA 96, 9258−9263 (1999). | Article | PubMed  | ChemPort | Add to Connotea (beta) |
25. Baubonis, W. & Sauer, B. Genomic targeting with purified Cre recombinase. Nucleic Acids Res. 21, 2025−2029 (1993). | PubMed  | ISI | ChemPort | Add to Connotea (beta)  |
26. Riesselmann, L. & Haaf, T. Preferential S-phase pairing of the imprinted region on distal mouse chromosome 7. Cytogenet. Cell Genet. 86, 39−42 (1999). | Article | PubMed  | ISI | ChemPort | Add to Connotea (beta) |
27. LaSalle, J. M. & Lalande, M. Homologous association of oppositely imprinted chromosomal domains. Science 272, 725−728 (1996). | PubMed  | ISI | ChemPort |