Targeted modification of the complete chicken lysozyme gene by poxvirus-mediated recombination.

We have developed a novel ex vivo system for the rapid one-step targeted modification of large eucaryotic DNA sequences. The highly recombinant environment resulting from infection of rabbit cornea cells with the Shope fibroma virus was exploited to mediate precise modifications of the complete chicken lysozyme gene domain (21.5 kb). Homologous recombination was designed to occur between target DNA (containing the complete lysozyme gene domain) maintained in a lambda bacteriophage vector and modified targeting DNA maintained in a plasmid. The targeting plasmids were designed to transfer exogenous sequences (for example, beta-galactosidase alpha-complement, green fluorescent protein, and hydrophobic tail coding sequences) to specific sites within the lysozyme gene domain. Cotransfection of the target phage and a targeting plasmid into Shope fibroma virus infected cells resulted in the poxvirus-mediated transfer of the modified sequences from plasmid to phage. Phage DNA (recombinant and nonrecombinant) was then harvested from the total cellular DNA by packaging into lambda phage particles and correct recombinants were identified. Four different gene-targeting pairings were carried out, and from 3% to 11% of the recovered phages were recombinant. Using this poxvirus-mediated targeting system, four different regions of the chicken lysozyme gene domain have been modified precisely by our research group overall with a variety of inserts (6-971 bp), deletions (584-3000 bp), and replacements. We have never failed to obtain the desired recombinant. Poxvirus-mediated recombination thus constitutes a routine, rapid, and remarkably efficient genetic engineering system for the precise modification of large eucaryotic gene domains when compared with traditional practices.

Analysis of an approach to oviduct-specific expression of modified chicken lysozyme genes.

The -2.7 kb enhancer (E) element of the chicken lysozyme gene domain appears to govern expression of the gene in macrophages but not in oviduct tubular gland cells, the only other site of lysozyme expression. The ultimate goal of our research was to determine whether lysozyme domain variants could be developed that would mainly be expressed in the oviduct so that transgenic birds could be produced that would deposit exogenous protein in the egg white. Accordingly, precise mutations were made by poxvirus-mediated gene targeting in FEF/PU.1 and CCAAT/enhancer-binding protein (C/EBP) transcription factor binding sites in the -2.7 kb E of cloned copies of a specific lysozyme gene variant that includes a hydrophobic pentapeptide tail encoding sequence inserted immediately prior to the stop codon. This variant contains the entire lysozyme domain and is cloned in a lambda bacteriophage vector (lambdaDIILys-HT); the novel tail sequence enables distinction in cell-based expression systems between transcripts of the variant and those of the endogenous gene. These various lysozyme domain mutants, in bacteriophage vector form, were tested for expression in cultured chicken blastodermal cells cotransfected with plasmids encoding the transcription factors C/EBP and v-Myb. In the absence of these plasmids, barely detectable levels of endogenous lysozyme gene transcription resulted in the blastodermal cells. In the presence of the plasmids, however, transcripts of the endogenous gene could be detected as well as varying levels (as evaluated by quantitative real-time PCR) of transcripts of all of the lysozyme domain mutants. These results are discussed in the context of the known role and occurrence of various transcription factors involved in gene expression in differentiating macrophage cells. The ultimate test of expression of the variants in macrophages vs. oviduct cells will be to use them to produce transgenic birds.

Measuring conservation of contiguous sets of autosomal markers on bovine and porcine genomes in relation to the map of the human genome.

Based on published information, we have identified 991 genes and gene-family clusters for cattle and 764 for pigs that have orthologues in the human genome. The relative linear locations of these genes on human sequence maps were used as "rulers" to annotate bovine and porcine genomes based on a CSAM (contiguous sets of autosomal markers) approach. A CSAM is an uninterrupted set of markers in one genome (primary genome; the human genome in this study) that is syntenic in the other genome (secondary genome; the bovine and porcine genomes in this study). The analysis revealed 81 conserved syntenies and 161 CSAMs between human and bovine autosomes and 50 conserved syntenies and 95 CSAMs between human and porcine autosomes. Using the human sequence map as a reference, these 991 and 764 markers could correlate 72 and 74% of the human genome with the bovine and porcine genomes, respectively. Based on the number of contiguous markers in each CSAM, we classified these CSAMs into five size groups as follows: singletons (one marker only), small (2-4 markers), medium (5-10 markers), large (11-20 markers), and very large (> 20 markers). Several bovine and porcine chromosomes appear to be represented as di-CSAM repeats in a tandem or dispersed way on human chromosomes. The number of potential CSAMs for which no markers are currently available were estimated to be 63 between human and bovine genomes and 18 between human and porcine genomes. These results provide basic guidelines for further gene and QTL mapping of the bovine and porcine genomes, as well as insight into the evolution of mammalian genomes.