How close are we to implementing gene targeting in animals other than the mouse?

T paper by Kubota et al. in this issue of PNAS (1) describes several significant contributions that bring us much closer to extending ‘‘gene targeting’’ to mammalian species other than the mouse. Gene targeting now provides the means for creating new strains of mice with mutations in virtually any gene. First, the desired mutation is introduced into a cloned copy of the chosen gene by standard recombinant DNA technology. The mutation then is transferred to the genome of a pluripotent mouse embryoderived stem (ES) cell by means of homologous recombination between the exogenous mutated DNA sequence and the cognate DNA sequence in the ES cell chromosome. By microinjection of ES cells containing the targeted mutation into blastocysts and allowing the embryo to come to term in foster mothers, chimeric mice can be generated that are capable of transmitting the targeted mutation to their offspring. The power of gene targeting is that the investigator chooses both which genes to modify and how to modify them. The investigator has virtually complete control over the way in which the gene’s DNA sequence or surrounding regulatory elements are modified. Thus, gene targeting should not be thought of as a means only for the inactivation of genes, but as a method for altering gene activity in any purposeful manner. This technology permits evaluation of deliberately selected gene functions in the intact mouse and a systematic dissection of the most complex of biological processes, such as development and learning. Because nearly all biological phenomena are mediated or influenced by genes, this technology has had a profound impact on all areas of biological research in mammals, including the study of cancer, immunology, neurobiology, and human genetic disease. The extension of gene targeting to other mammalian species would provide an additional boost to the study of mammalian biology. To date, however, efforts to isolate embryo-derived stem cells capable of contributing to the formation of the germ line in mammalian species, other than the mouse, have failed. This failure precludes the use of such cells to generate chimeras of other mammalian species, and thus use of the gene targeting strategy that has been so successful in the mouse. However, Wilmut’s successful cloning of the sheep, Dolly, by nuclear transfer of a somatic cell nucleus into an enucleated sheep oocyte (2) provides an alternative route to generating animals with targeted mutations in their germ line. Targeted mutations can be introduced readily into any somatic cell in culture. (In fact, our experience suggests that if the experiments are done appropriately, using isogenic DNA, optimizing the tissue culture selection procedures for the chosen cell line, etc., most mammalian cells in culture show closely equivalent capacities to mediate homologous recombination between exogenous and endogenous DNA sequences.) The transfer of nuclei from cultured somatic cells containing targeted mutations to enucleated oocytes could provide the desired alternative route. The potential for success of such an approach is supported by the experiments reported by Kubota et al. (1). Those authors have generated cloned animals of a 17-year-old bull by using nuclei of fibroblasts cultured from a biopsy of his skin. Culture of these fibroblasts in vitro for 10–15 passages ('20–30 cell doublings) did not compromise their capacity to generate healthy animal clones. These new results have made two significant contributions to the prospects for use of this procedure as the means of generating animal clones containing targeted gene modifications. First, the authors used the most easily obtainable cells as a source of transplanted nuclei. Under the commonly used culture conditions for tissue explants, the cell type that generally outgrows all others is the fibroblast. The selection of fibroblasts as a source of nuclei for producing animal clones makes this procedure particularly practical. Second, as demonstrated by Kubota et al. (1), these fibroblasts can be cultured in vitro for 10–15 passages without losing their capacity to generate healthy bull clones is very significant because this number of passages exceeds the number needed to generate purified populations of cells containing targeted mutations. Although all mammalian somatic cells have the capacity to mediate homologous recombination (a requirement to generate targeted mutations), the efficiency of this reaction is low relative to the competing nonhomologous recombination reaction (3). Overcoming this deficit requires the use of tissue culture selection procedures to enrich for cells containing the desired homologous recombination event (4). However, because even under the most extreme conditions the absolute targeting frequency is approximately one in 105-106 starting cells (i.e., fewer than 20 cell doublings), the ability to culture the donor fibroblasts for at least 30 cell doublings, while still retaining their animal cloning potential, provides an ample cushion to carry out sophisticated gene targeting procedures with these cells. This procedure could include even the possibilities of generating cells homozygous for the targeted modification, or ones containing conditional mutations (5). A problem associated with all animal cloning procedures reported to date is high rates of embryonic and neonatal lethality. The source of this lethality has not been established. However, a possible culprit is improper execution of the complex changes in the patterns of genomic demethylation and methylation that normally accompany the process of early embryogenesis. Proper execution of this program is required to maintain balanced growth between extraembryonic and fetal tissues (6). This process normally is accomplished by differential epigenetic imprinting of selected genes from the maternal and paternal genomes responsible for control of balanced growth. A cloning procedure, based on the introduction of somatic nuclei into the cytoplasmic environment of the oocyte, requires a repro-