C OMMENT TIG NOVEMBER 1998 VOL. 14 NO. 11 438 0168-9525/98/$ – see front matter © 1998 Elsevier Science All rights reserved. PII: S0168-9525(98)01579-0 Genetics is a science of the aberrant. By studying mutant alleles, we hope to understand better the wild-type function of a gene. The more alleles analyzed for a specific gene, the greater the functional spectrum that is acquired. The only limitation is generating the alleles to form a series. Historically, this has been easy in phage, bacteria, yeast and fruit flies, simply because it is practical and economical to produce vast numbers of these organisms, yet mouse geneti- cists have been hesitant to attempt such experiments owing to financial and temporal constraints. Nowadays, however, there are many ways to generate allelic series relatively quickly and cheaply in the mouse. The advent of gene targeting 1 cleared the path for rapid, directed mutagenesis and has yielded a treasure-trove of new mouse alleles, most of which are knock-outs. But the null mutation coupled with the wild-type hardly constitutes an allelic series. To increase the diversity, it is possible to recapitulate subtle alleles by homologous recombination in the mouse 2,3 . Gene targeting can also be used to incorporate the recombinase- recognition sequences loxP and frt interspersed among exons, so that a single transgenic mouse can spin off a variety of allelic progeny when mated with appropriate recombi- nase-expressing mouse lines. Such ‘allelogenic’ mice have recently been used to create novel alleles of Fgf8, a fibroblast growth factor gene 4 , and N-myc, a proto-oncogene 5 . This strat- egy can produce hypomorphic alleles that allow for phenotypic analysis at later developmental stages that would have been precluded by the early-death phenotype of the null mutation. However, if a more classical approach is preferred – or one would rather study point mutations than recombinase-induced deletions – there is always chemical mutagenesis. The point mutagen ethylnitrosourea 6 can induce missense mutations in the mouse, often allowing a gene prod- uct to retain some residual function. A benefit of this strategy is that the investigator does not need any mol- ecular knowledge of the gene to build a series (i.e. there is no need to worry about what type of an engineered point mutation will generate a hypo- morph versus a hypermorph versus a neomorph versus an amorph). In fact, the gene itself doesn’t even have to be cloned. Instead, chemical mutagenesis allows for a random mutational spectra to unfold, unen- cumbered by any preconceived and limited notions of the observer. In a simple screen, the germ cells of a male are mutagenized and then the mouse is mated with a series of females that are already homozygous mutant for the gene of interest (usually a null mutation). Newly induced alleles from the male will fail to complement the female’s genotype and the progeny will be recognizable as mutants. This scheme, called the specific locus test, can rapidly generate an extensive allelic series 7 . In fact, since the 1950s, specially designed test stocks of mice have been used in large-scale specific locus tests at the Oak Ridge National Laboratory to determine the muta- genicity of different types of radiation and chemicals 8 . The result of these massive screens have been literally hundreds of alleles at seven specific loci, two of which are dilute and short-ear. In a second type of screen, de- letion stocks, currently those that occur at coat-color loci, have been used for rapidly generating point mutations by chemical mutagenesis to form an allelic series in the func- tional units, restricted to the deleted segment 9,10 . Nowadays, chromosome engineering in the mouse can create deletions at any defined site in the genome 11,12 . Additionally, these de- letions can be purposefully marked with coat-color genes that allow the manipulated chromosomes to be fol- lowed throughout the crosses 13 . By applying chemical mutagenesis to Mouse alleles: if you’ve seen one, you haven’t seen them all ALLAN PETER DAVIS AND MONICA J. JUSTICE davisap@bio.ornl.gov • mjustice@bcm.tmc.edu LIFE SCIENCES DIVISION, OAK RIDGE NATIONAL LABORATORY, OAK RIDGE, TN 37831-8080, USA. 7 Holländer, G. et al. (1998) Science 279, 2118–2121 8 Carrel, L. and Willard, H. (1998) Nat. Genet. 19, 211–212 9 Cattanach, B.M. and Beechey, C.V. (1990) Development (Suppl.) 63–72 10 Zhang, Y. et al. (1993) Am. J. Hum. Genet. 53, 113–124 11 Dao, D. et al. (1998) Hum. Mol. Genet. 7, 597–608 12 Mitsuya, K. et al. (1997) Hum. Mol. Genet. 6, 2243–2246 13 He, L. et al. (1998) Oncogene 16, 113–119 14 Dean, W. et al. (1998) Development 125, 2273–2282 15 Panning, B. and Jaenisch, R. (1998) Cell 93, 305–308 16 Tada, T., Takagi, N. and Adler, I.D. (1993) Genet. Res. Camb. 62, 139–148 17 Howe, M., Dimitri, P., Berloco, M. and Wakimoto, B.T. (1995) Genetics 140, 1033–1045 18 Tweedie, S., Charlton, J., Clark, V. and Bird, A. (1997) Mol. Cell. Biol. 17, 1469–1475 19 Shemer, R. et al. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 6371–6376 20 Nan, X. et al. (1998) Nature 393, 386–389 21 Feinberg, A.J. (1993) Nat. Genet. 4, 110–113 22 Tycko, B. in Genomic Imprinting: an Interdisciplinary Approach (Ohlsson, R., ed.), Springer (in press) 23 Moulton, T. et al. (1994) Nat. Genet. 7, 440–447 24 Leighton, P. et al. (1995) Genes Dev. 9, 2079–2089 25 Cui, H. et al. (1997) Cancer Res. 57, 4469–4473 26 McGowan, R., Campbell, R., Peterson, A. and Sapienza, C. (1989) Genes Dev. 2, 1669–1676 27 Adam, G. et al. (1996) Development 122, 839–847 References added in proof 28 Bix, M. and Locksley, R. (1998) Science 281, 1352–1354 29 Ohlsson, R. et al. Dev. Genes Evol. (in press)