Recombination of an insertion vector into its chromosomal homologue is a conservative event in that both the chromosomal and the vector sequences are preserved. However, gene conversion may accompany homologous recombination of an insertion vector. To examine gene conversion in more detail we have determined the targeting frequencies and the structure of the recombinant alleles generated with a series of vectors which target the hprt gene in embryonic stem cells. We demonstrate that gene conversion of the introduced mutation does not significantly limit homologous recombination and that gene conversion occurs without a sequence specific bias for five different mutations. The frequency of the loss of a vector mutation and the gain of a chromosomal sequence is inversely proportional to the distance between the vector mutation and the double-strand break. The loss of a chromosomal sequence and the gain of a vector mutation occurs at a low frequency.

To develop an animal model for cystic fibrosis (CF), targeted gene disruption in embryonic stem (ES) cells was used to generate a duplication of exon 3 (cftrm1Bay allele) of the mouse CF gene. ES cells containing this mutation were used to generate chimeric animals that transmitted the mutant allele through the germline. Homozygous mutant animals display a severe phenotype, with approximately 40% dying within 1 week from intestinal obstruction. RNAase protection analysis of the cftrm1Bay allele did not detect any normal mRNA (< 1-2% of wild-type) in mutant animals. Pathologic changes in the intestines from mutant mice included mucus accumulation in the crypts and intestinal lumen, dilatation of the bases of the crypts, enlargement of goblet cells, and the presence of concretions in the crypts or between the villi. Changes were also present in the mucosal glands of the pharynx and the minor sublingual glands, where dilatation of acini and accumulation of eosinophilic material were evident. Atrophy of acinar cells that may be secondary to nutritional deficiency and mild inflammation in the main pancreatic duct were present in the pancreas of mutant animals. No changes were noted in the lung, trachea, liver, or male reproductive tract of mutant animals, and mutant males were fertile. Homozygous mutant mice showed defects in cAMP-mediated ion transport both in ileum and in cultured fetal tracheal explants. Thus, an additional mouse model for CF has been generated that should prove useful for the understanding of the pathogenesis and the development of treatments for CF.

We have analyzed the gene-targeting frequencies and recombination products generated by a series of vectors which target the hprt locus in embryonic stem cells and found the existence of alternative pathways that depend on the location of the double-strand break within the vector. A double-strand break in the targeting homology was found to increase the targeting frequency compared with a double-strand break at the edge of or outside the target homology; this finding agrees with the double-strand break repair model proposed for Saccharomyces cerevisiae. Although a double-strand break in the homology is important for efficient targeting, observations reported here suggest that the terminal ends are not always directly involved in the initial recombination event. Short terminal heterologous sequences which block the homologous ends of the vector may be incorporated into the target locus. A modification of the double-strand break repair model is described to account for this observation.

Homologous recombination has been used to introduce site-specific mutations into murine embryonic stem (ES) cells with both insertion and replacement vectors. In this study, we compared the frequency of gene targeting with various lengths of homology and found a dramatic increase in targeting with an increase in homology from 1.3 to 6.8 kb. We examined in detail the relationship between the length of homology and the gene-targeting frequency for replacement vectors and found that a critical length of homology is needed for targeting. Adding greater lengths of homology to this critical length has less of an effect on the targeting frequency. We also analyzed the lengths of homology necessary on both arms of the vector for gene replacement events and found that 472 bp of homology is used as efficiently as 1.2 kb in the formation and resolution of crossover junctions.

Gene targeting has been used to direct mutations into specific chromosomal loci in murine embryonic stem (ES) cells. The altered locus can be studied in vivo with chimeras and, if the mutated cells contribute to the germ line, in their offspring. Although homologous recombination is the basis for the widely used gene targeting techniques, to date, the mechanism of homologous recombination between a vector and the chromosomal target in mammalian cells is essentially unknown. Here we look at the nature of gene targeting in ES cells by comparing an insertion vector with replacement vectors that target hprt. We found that the insertion vector targeted up to ninefold more frequently than a replacement vector with the same length of homologous sequence. We also observed that the majority of clones targeted with replacement vectors did not recombine as predicted. Analysis of the recombinant structures showed that the external heterologous sequences were often incorporated into the target locus. This observation can be explained by either single reciprocal recombination (vector insertion) of a recircularized vector or double reciprocal recombination/gene conversion (gene replacement) of a vector concatemer. Thus, single reciprocal recombination of an insertion vector occurs 92-fold more frequently than double reciprocal recombination of a replacement vector with crossover junctions on both the long and short arms.