© 1998 Nature Publishing Group http://www.nature.com/naturegenetics • V. v.H. and N.D. H. are International Scholars of the Howard Hughes Medical Institute. Kiyoshi Miyagawa1.6, Jill Kent 1 , Adrian Moore 1 , Jean-Paul Charlieu 1 , Melissa H. Little\ Kathleen A. Williamson\ Anna Kelsey 2 , Keith W. Brown3, Shabbir Hassam 4 , Jakob Briner4, Yasuhide Hayashi 5 , Hisamaru Hirai 6 , Yoshio Yazaki6, Veronica van Heyningen 1 & Nicholas D. Hastie 1 'MRC Human Genetics Unit, Western General Hospital, Crewe Road, Edinburgh EH4 2XU, UK. 2Department of Pathology, Royal Manchester Children's Hospital, Pendlebury, Manchester M27 1 HA. UK. 3 CLIC Research Unit, Department of Pathology and Microbiology, School of Medical Sciences, University Walk, Bristol BSB 1 TO, UK. 4 Departement Pathologie der Universitat. lnstitut fur Klinische Pathologie, Universitatsspital ZOrich. Schmelzbergstrasse 12, CH-8091 Zurich, Switzerland. 5 Department of Paediatrics and 6Third Department of Internal Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113, Japan. Correspondence should be correspondence addressed to N.D. H. e-mail: oldnick@hgu.mrc.ac.uk 1. Hastie, N.D. Annu. Rev. Genet. 28, 523-558 (1994). 2. Call, K.M. eta/. Cell 60, 509-520 (1990). 3. Gessler, M. et al. Nature 343,774-778 (1990). 4. Drummond, I.A. et a/. Science 257, 674-678 (1992). 5. Larsson, S.H. eta/. Cell 81 , 392-401 (1995). 6. Weintraub, H. Cell 75, 1241-1244 (1993). 7. Olson, E.N. & Klein, W. H. Genes Dev. 8, 1-8 (1994) . 8. Little, M .H. et a/. Hum. Mol. Genet. Z, 259-264 (1993). 9. Brown, K.W. et a/. Oncogene 7, 763-768 (1992) . 10. Bickmore, W.A. eta/. Science 257, 235-237 (1992). No female embryonic lethality in mice nullizygous for Msh2 and p53 R ecently, it has been reported that male mice bearing targeted inactiva- tions of both p53 and Msh2 are viable, but rapidly succumb to lymphoma earlier than either p53 or Msh2 single mutants. In con- trast, it was reported that all female mice lacking p53 and Msh2 underwent develop- mental arrest at 9.5 days and died in utero 1 • The published study did not report a sig- nificant reduction in the number of female p53-l- animals, although the authors state that this is probably due to their small cohort size. We have also generated male mice lacking p53 and Msh2 which are viable and succumb to lymphoma at a sim- ilarly early age ( tX is 65 days). However, our results differ from the data of Cranston et a/. in that we have successfully generated 22 adult female mice nullizygous for both p53 and Msh2. The parental strains used to gen- erate our cohort differ from those used by Cranston et a/., but both have been well characterized previously2• 3 . Importantly, the single mutant Msh2 and p53 parental strains used in our cohort possess very sim- ilar reported phenotypes to those used by Cranston et at.Z- 5 • Data from our cohort show the male to female ratio in Msh2·1· pSJ-1- mice is similar to the ratio observed in Mshz+i+ p53-t- mice (Table 1). Thus, in our cohort Msh2 plays no detectable role in female development with the reduction in female mice in our Mshzi- psJ·I- cohort from expected Mendelian ratios being accounted for by p53-associated exen- cephaly and subsequent anencephaly6. We observed an identical phenotype for female Mshz i- p5J-I- mice as for their male counterparts, dying from lymphoma at a similar age. We have used male Msh2 -/- p5J·1· mice in successful breeding pairs, confirming that these mice are fertile. nature genetics volume 18 january 1998 Table 1 • Male to female ratio in live born mice pSJ+I+ psJ-1- 26112 (-37%) 52/21 (·42%) 61/22 (-47%) Msh2+ 1 + male/female Msh2+ 1 - MshL'- 262/254 (-2%) 118/94(-11%) 166/177 (+3%) Sex ratio of mice for each genotype group together with the percentage deviation from the expected 1:1 male to female ratio shown in brackets. The observed reduction in female mice doubly null for Msh2 and p53 is not significantly different from that related to p53 deficiency alone (X2 test). In this data set homozygosity for lack of p53, irrespective of Msh2 status, confers a significant reduction in the number of female progeny (p< O.Q1 , X2 test) . We suggest four possible explanations for these cohorts had been derived from ES the difference between our own data and cells engineered over an extended period in the work of Cranston et al. First, the dis- vitro to carry multiple mutations, but this crepancy may result from different levels of was not the case 1 - 5 . Finally, it is possible environmental insult. However, our colony that the targetted events differ in some way, is not maintained under barrier conditions either at the targetted locus itself or by the and we therefore consider this unlikely. influence of the targeted locus upon neigh- Second, this difference may arise through bouring genes. Although we cannot at pre- the use of different genetic backgrounds. sent confirm or refute any of the above Indeed, we have previously documented the explanations, it is clear from our data that, strain dependency of p53-related anen- at the very least, the reported female cephaly6. Against this possibility argues the embryonic lethality associated with Msh2 fact that both groups of mice were gener- and p53 deficiency is not fully penetrant. ated from outbred crosses derived from a mixture of mouse strains (our animals seg- regate for 129/0la, Balb-c and SWR genomes) and further, that our previous analysis showed different outcrossed strains to possess similar levels of p53-related embryonic death 6 • Notwithstanding these observations, it remains possible that the observed Msh2-related death is strain dependent, and we are currently carrying out appropriate backcrosses to address this point. Third, it is possible that the pheno- type observed by Cranston eta/. is arising as a consequence of a second mutation which is linked to one or other of the targeted alle- les. The likelihood of such a linked muta- tion would perhaps be increased if either of Acknowledgements We thank H. te Riele for supply of mice. A.R. C. is a Royal Society University Research Fellow. Neil J. Toft, Mark J. Arends, Andrew H. Wyllie & Alan R. Clarke Cancer Research Campaign Laboratories, Department of Pathology, University Medical School, Edinburgh, EHB 9AG, UK. Correspondence should be addressed to A.R.C. 1. Cranston, A. eta/. Nature Genet. 17, 114-118 (1997). 2. de Wind, N. eta/. Ce//82, 321-330 (1995). 3. Clarke, A.R. et al. Nature 362, 849-852 (1993). 4. Donehower, L.A. eta/. Nature 356, 215-221(1992). 5 Reitmair, A. H. eta/. Nature Genet. 11, 64-69 (1995). 6. Armstrong, J.F., Kaufman, M.H., Harrison, D.J. & Clarke A.R. Curr. Bioi. 5, 931-936 (1996). 17