Gene targeting in normal and amplified cell lines.

Targeted recombination in mammalian cells is rare compared with non-homologous integration. In Saccharomyces cerevisiae the reverse is true. Differences in targeting efficiency could arise because a target of unique DNA is 200 times more dilute in mammalian genomes than it is in yeast. We tested this possibility by measuring gene targeting in normal CHO cells with two copies of the dihydrofolate reductase (DHFR) gene and in amplified CHOC 400 cells, which carry 800 copies. If the concentration of the target gene is critical, amplified cells should show an enhanced frequency of targeted recombination relative to non-homologous integration. Using a positive/negative selection protocol, we demonstrated that the efficiency of targeting into DHFR genes is indistinguishable in normal and amplified CHO cells. As targeting does not depend on the number of targets, the search for homology is not a rate-limiting step in the mammalian pathway of gene targeting. Thus, the difference in genome size is not the basis for the different outcomes of targeting experiments in S. cerevisiae and mammals.