Low and variable efficiency is a major problem in targeted gene alteration, which is used as a primary tool in gene therapy and animal model studies. We tested several types of constructs, alone, or in combination with other factors, to introduce a point mutation into the alphaB-crystallin gene in one-celled mouse embryos. We found that co-injection of single stranded DNA (ssDNA) along with antibodies against Ku70/86, or supplementing the system with hRad51/hRad54, increases efficiency of targeted mutagenesis. These findings suggest that proteins in the homologous recombination DNA repair pathway contribute, and that proteins involved in the alternative non-homologous end joining pathway inhibit, ssDNA-mediated targeted mutagenesis. This is the first successful demonstration of targeted mutation in early mouse embryos. This novel methodology of supplying protein factors to stimulate gene modification in the nucleus has not been reported previously. For this initial study we have been using an extremely time consuming PCR-based restriction fragment length polymorphism (RFLP) assay to detect point mutations introduced into genomic DNA.? ? To improve the detection method, we have developed transgenic mouse lines in which correction of a mutation in a fluorescent protein gives an instant read-out, and many embryos can be screened simultaneously. We engineered a cassette encoding a bicistronic mRNA (mutated red fluorescent protein DsRed / an IRES element / green fluorescent protein). All mouse embryos expressing the transgene should fluoresce green (endogenous control for transgene expression), but only embryos in which the point mutation in DsRed has been corrected should also fluoresce red. We have used several enhancer/promoter elements, which have been used successfully in preimplantation stage mouse embryos or mouse ES cells, to drive expression of the cassette.? ? A successful reporter line would exhibit easily detectable levels of fluorescent protein expression at the earliest stages of embryo development. All promoter/fluorescent protein transgenes were constructed and tested for expression of EGFP and intact DsRed in both transient assays in mouse embryos and in F1 transgenic mouse embryos. Several promoters/enhancers were used in attempts attain early embryonic gene expression. A construct, whose expression is driven by the EF1alpha promoter, was able to generate green fluorescent protein in eight cell stage embryos (48 hr pf) and red fluorescent protein at 72 hr pf. A line of transgenic mice with the EF1alpha promoter / mutated red fluorescent protein / an IRES element / green fluorescent protein construct appears to be extremely promising at this time. Four transgenic mice lines with mutated DsRed protein were established. All four lines show expression of EGFP at 48 hr pf. Sequence analysis of a PCR-amplified region of the DsRed gene, from mouse genomic DNA, confirmed presence of the nonsense mutation in codon 15 of DsRed. RT-PCR analysis shows that three transgenic mouse lines have low copy number of the construct (2 copy) and one has a high copy number (26 copies of the construct). One line of mice (with 2 gene copies) is being bred for further experiments.? ? In addition to our work with ssDNA as a targeting vector, we are also working, in single celled mouse embryos, on another type of gene targeting which we have named Forced Homologous Recombination (FHR). In this case, we use a significantly larger targeting vector, similar to those currently used in ES cell targeting procedures, with regions of homology to a selected gene or chromosomal locus at both 5 and 3 ends. Co-microinjection of the targeting construct with key factors involved in homologous recombination (for example Rad51/Rad54 proteins, which has been shown be able to form a filament-complex with nucleotides in vitro) may increase gene targeting efficiency. We are planning to use a nucleoprotein filament complex instead of the naked DNA vector to further increase alteration efficiency. The presence of long homology arms on both ends should increase specificity of gene modification and also dramatically decrease any random integration of the targeting vector, compared to our ssDNA strategies. Several genes were selected for in vivo testing of this FHR system. Generation of several vectors for this project is presently underway.
