Gene trap expression and mutational analysis for genes involved in the development of the mammalian nervous system

We have used a large-scale gene trap approach for the isolation and mutation of genes that might play roles in the developing nervous system. After in vitro integration of two different gene trap vectors (pGT1.8geo: Skarnes et al. [1995] Proc. Natl. Acad. Sci. USA 92:6592-6596; IRES beta geo: Chowdhury et al. [1997] Nucleic Acids Res. 25:1531-1536) in mouse embryonic stem (ES) cell lines, we created 64 transgenic mouse lines. The expression analysis of the reporter gene during embryogenesis of heterozygous embryos revealed 47 lines with a variety of patterns. Around one-third (36%) of these gene trap lines showed spatiotemporal expression that was either restricted predominantly in the developing nervous system (11 lines; 17%) or widespread but with very high levels of expression in the nervous tissue (12 lines; 19%). In most cases, a correlation was found between the in vitro and the in vivo patterns of the reporter gene expression. Thus far, preliminary mutant analysis of 16 gene trap lines with potentially interesting expression patterns in the developing nervous system showed that mice homozygous for eight (50%) insertions were lethal, whereas the homozygous mice from five gene trap lines (31%) showed a lower than expected Mendelian ratio of live homozygous animals. Analysis of beta-galactosidase reporter gene expression during embryogenesis has shown that four transgenic lines are useful lacZ in situ markers for specific regions of the developing nervous system. Here, we discuss some in vivo and in vitro selection criteria that may increase the number of the trapped genes potentially involved in the control of neural development and some future strategies to improve further the efficiency of the gene trap approach.