Classical genetics and gene targeting
Gene targeting has provided considerable insight into the functions of numerous genes since it was developed a decade ago (1-4). A listing of the diversity of targeted genes and breadth of mutant phenotypes characterized to date can be obtained through mouse mutation databases (http://www.bis.med. jhmi.edu/Dan/tbase.html and http://biomednet.com/cgi-bin/ mko/mkohome.pl) (5, 6). However, in order to take full advantage of this technology, classical genetic methods should be utilized to extend our knowledge of individual genes to genetic pathways. In this chapter, the significance of genetics in gene targeting and phenotype interpretation are discussed. We describe how Mendelian and quantitative genetics can be exploited to map modifier loci or generate animals carrying mutations in two or more genes. We also discuss the development and application of classical genetic approaches towards elucidating gene function such as generation of allelic series and creation of deletion complexes throughout the genome in ES cells and mice. Several genetic considerations should be taken into account during the initial stages of a gene targeting experiment. In order to maximize homologous recombination efficiency, both arms of a targeting vector should be isolated from the same strain of mice as that of the ES cells (see Chapter 1 for details). Although most ES cell lines have been isolated from 129 strains of mice, significant genetic variation exists among the different substrains, which is sometimes evident by pronounced differences in coat colour (see Chapter 4, Table 1) (7, 8). For instance, 129/Sv mice (Aw/Aw, +c-Tyr +p/+c-Tyr +p ) have a white-bellied agouti (Aw) phenotype, whereas 129/SvJ mice (Aw/Aw, Tyrc p/T yrc-ch p) have a cream colour owing to the effect of mutant tyrosinase (Tyr) and pink-eyed dilution (p) alleles which are epistatic to Aw (9). Molecular analysis of different 129 substrains using microsatellite markers has provided insight into their genetic differences and revealed that 129/SvJ is particularly divergent and actually contaminated with genomic regions of non-129 origin (7, 8). Before the significance of this heterogeneity was appreciated, many targeting vectors were constructed from 129/SvJ DNA for use in 129/Sv ES cell lines.