Compared to previous results involving either DNA microinjection or retroviral infection, the latest results of Gaiano and Hopkins represent a 20- to 30-fold increase in the efficiency of generating transgenic zebrafish. If the retroviral constructs insert at random into the genome, then they may be potent insertional mutagens for screening the zebrafish genome. If the same high frequency of integration can be achieved by a lacZ- or GFP-expressing retroviral vector, then living F1 embryos can screened directly, and positive ones raised to sexual maturity and subsequently bred to homozygosity. This makes it feasible to breed thousands of retroviral insertions to homozygosity in zebrafish, allowing a large-scale insertional mutagenesis screen. The great advantage of such a screen is that any interesting mutation that is found will be readily cloned by transgene tagging or plasmid rescue.
Hopkins and her colleagues suggest that 4-6 people can generate 10,000 to 20,0000 founder fish that will carry 100,000-200,000 transgenes in about 3 mo (33). Given the size of the zebrafish genome (1.6 x 109 bp), then 200,000 random insertions would on average represent an insertion every 8 kb. This potenially could be an insertion in every gene. This assumes that the insertions are indeed random, which is as yet unknown. Analysis of some of the insertions already generated does suggest that the integrations of the retroviral sequences are in, or near, transcribed regions (34,35). Determining the mutant phenotypes of different insertions either involves generating homozygotes, which will be labor-intensive, or screening haploids generated from the F1 transgenic fish. These fish may carry multiple insertions and so could have complex phenotypes. An alternative is to modify the retroviral vectors so that they can be used as gene traps. These would contain a reporter gene which can only be expressed after integration into a transcribed endogenous gene.
Was this article helpful?