Transgenic animals: techniques.

Abstract

The methodology for generating transgenic mice has opened up new vistas in genetic manipulation of the mammalian genome. Transgenic mice can be generated by injection of DNA into the zygote pronucleus (Palmiter and Brinster 1986) retroviral infection of early embryos (Soriano et al. 1986) or introduction of DNA into pluripotent embryonic stem (ES) cells in culture followed by colonization of the germ line in chimeras (Robertson et al. 1986). Most of the work utilizing transgenic mice has involved defining cis-acting regulatory elements that control tissue-specific and developmental expression of genes. This is still an important area of research and two of the workshop presentations dealt with such studies: one involving use of an in situ detectable reporter gene construct for analyzing tissue-specific gene expression and the other, the role of long-range sequences in controlling gene expression. However, transgenic animals are increasingly being considered as tools for manipulating endogenous gene expression to investigate gene function. Two possible approaches, use of antisense RNA or targeted mutagenesis, were described. Transgenic mice also offer the possibility of obtaining new mutations by insertional mutagenesis and may also be used to identify new genes active in embryogenesis by integration of reporter constructs. Dr. Lesley Toth described studies on the regulation of the expression of one of the mouse homeo-box genes, Hox-3. I. When 5' regions of this gene were linked to the Escherichia coli P-galactosidase gene and used to generate transgenic mice, tissue localization of gene expression could be detected by staining with X-gal (Zakany et al. 1988). Blue staining cells were detected in a region of the developing spinal cord similar to that defined for the endogenous gene by in situ hybridization. Expression outside this region was detected, suggesting that not all control sequences were included in the construct. Further definition of the spinal cord controlling region is underway. Two common observations in transgenic mice are that level of transgene expression is not directly related to copy number and varies with site of integration. Dr. Lee Wall described how, for the P-globin locus, such effects can be overcome by inclusion of sequences located at DNAse I superhypersensitive sites removed from the close vicinity of the gene (Grosveld et al. 1987). These sequences may allow formation of chromosomal domains that protect the transgene from effects due to integration sites. Whether such sequences flank other genes is not yet known. The possibility of manipulating gene expression by injection of antisense RNA into eggs was discussed by Dr. Bob Erickson. His group has shown that antisense P-glucuronidase RNA injected in large excess into mouse zygotes can significantly suppress P-glucuronidase enzyme activity throughout preimplantation development (Bevilacqua et a1 . 1988). Similar studies utilizing antisense nRNA to one of the gap junction protein genes are underway, and effects on uncoupling of cell communication have been observed. This approach has potential for investigating the role of several genes active in the early embryo. A more definitive means of analyzing the function of any known gene in development is to mutate the gene and study the phenotype of the mutation. Dr. Tom Doetschman described the development of targeted mutagenesis in mammals by homologous recombination in pluripotent embryonic stem cells. Once a mutation has been produced in the gene of interest, the stem cells can be injected into blastocysts to generate chimeric mice which will hopefully contain germ cells derived from the embryonic stem cells and thus carry the mutant gene. By this means, a mutation induced in culture can be transferred into the mouse stock for genetic analysis. Dr. Doetschman has demonstrated the feasibility of such approaches by both curing HPRT(Doetschman et al. 1987) mutations and generating HPRTmutations by homologous recombination in ES cells. Strategies for applying this approach to other genes were discussed and are underway in a number of laboratories. Dr. Janet Rossant described a novel approach to identifying new genes active in embryogenesis which is very similar to an approach being developed in Drosophila. Vectors containing the E. coli lacZ gene driven by a weak promoter were introduced to ES cells, and a number of clonal lines were used to generate chimeras. Analysis of P-galactosidase expression in midgestation revealed that the lacZ gene could be activated in specific spatial patterns, indicating that cis-acting regulatory elements close to the site of integration were activating the gene. By choosing the most interesting patterns of expression and cloning out the host gene that activates lacZ, it should be possible to identify a new set of genes important in embryonic development. Discussion at the workshop demonstrated that this whole area of genetic manipulation of the mammalian gene line has a lot of potential for elucidating problems in all aspects of mammalian genetics.