MOLECULAR MEDICINE TODAY, SEPTEMBER 1999 (VOL. 5)
Knockout mice for dummies Gene targeting has become an extremely popular method for generating mouse models of human disease. The targeting event is carried out in vitro in pluripotent embryonic stem (ES) cells, which are then used to generate chimeric founder animals by either fusion with early-stage embryos or direct injection into donor blastocysts. Strategies used range from the simple insertion of vector sequences into the chosen gene or replacement of crucial exons with vector sequence, to complicated spatial or temporal knockouts and chromosomal rearrangements. All these methods depend on homologous recombination between the endogenous ES cell genomic DNA and the targeting construct, which must share a minimum stretch of DNA sequence. Consequently, they all require extensive preliminary physical mapping and restriction mapping before designing a construct for use in the targeting event. This is followed by subcloning of the desired restriction fragments into a vector backbone. Zheng and colleagues1 have tried to circumvent some of the barriers to knockout mouse production by producing libraries of preconstructed targeting vectors. Each library has been constructed from genomic mouse DNA in a modified l phage vector. The two libraries contain different resistance genes (neomycin and puromycin) plus either the 59 or 39 halves of the hprt gene. These sequences are flanked by loxP sites, which recombine on expression of the enzyme Cre recombinase in bacteria, to automatically excise the construct as a plasmid and eliminate laborious subcloning. So, you choose your gene of interest, screen the phage libraries for a clone containing parts of that gene, excise the plasmid and then use that construct for your gene targeting experiments. In addition, by using the two libraries to target physically distant regions, you can express Cre in the doubly targeted ES cells and promote
Gene therapy for HIV infection: safe if not efficacious
recombination between the two loci. This results in the reconstitution of the hprt gene and so gives a method of selecting for recombinants. The authors also utilize a method called Ôgap repair targetingÕ: this makes use of the doublestrand gap repair that occurs during insertional events. A gap is engineered in the genomic insert of the chosen construct and during homologous recombination the gap is repaired from the chromosomal DNA template. This results in a duplication of the genomic insert as well as insertion of the vector sequences, and provides an easy detection system for the recombination event through the use of a probe from within the gap. It is also likely to result in a null allele through transcript instability and/or truncation. The icing on the cake for this system is the inclusion of coat-color tags in the targeting vectors. The Agouti minigene works particularly well because it is dominant over the wild-type allele. This means that heterozygous pups bred from the chimeric founder can be visibly genotyped by coat color. As more of the mouse genome is sequenced, large-scale knockout experiments and the recreation of some of the large chromosome rearrangements seen in human diseases will become more commonplace. The methods and resources described here are ideal for highthroughput knockout of single genes, as well as the rapid systematic generation of targeted chromosomal rearrangements such as nested chromosome deletions. Although the coat color tags need some work as the change in color is dependent upon the position of the insertion, these vector libraries are a useful and timely addition to the knockout tool box. 1 Zheng, B., Mills, A. and Bradley, A. (1999) A system for rapid generation of coat color-tagged knockouts and defined chromosomal rearrangements in mice, Nucleic Acids Res. 27, 2354Ð2360 Lucy R. Osborne PhD [email protected]
1357-4310/99/$ - see front matter © 1999 Elsevier Science Ltd. All rights reserved.
The expression of anti-HIV genes in hematopoietic stem cells (HSCs) has the potential to generate multiple cell types (including mature T cells, macrophages and dendritic cells) resistant to HIV-1 infection. To test this approach to HIV gene therapy, Kohn and colleagues1 isolated HSCs (CD341 cells) from the bone marrow of four HIVpositive pediatric subjects. The CD341 cells were cultured ex vivo and exposed to retroviral vectors expressing either a marker gene or a marker gene plus an anti-HIV gene, a rev-responsive element (RRE) Ôdecoy geneÕ that works by mopping up Rev protein, thereby preventing it from binding to the bona fide RRE in the integrated viral genome. Cells transduced with the marker vector and the marker plus RRE decoy vector were re-administered to each subject with the aim of seeing whether cells carrying the anti-HIV gene had a selective advantage. Between 7 and 30% of the reinfused cells carried either one or other of the retroviral vectors. However, one day after transplantation only two subjects showed the presence of vector-positive peripheral blood leukocytes (1Ð3 cells per 10 000 cells). At all subsequent time points the marker gene vector only could be detected, and then at a very low frequency (1 cell per 100 000 cells). Although these gene-transfer and engraftment frequencies are low, there are some encouraging aspects to this trial. There were no adverse effects, including no detectable increase in HIV-1 levels even though the re-infused cells were subjected to cytokine activation. Given recent improvements in retroviral-mediated gene transfer of HSCs, perhaps future trials will produce a more clinically relevant outcome. 1 Kohn, D.B. et al. (1999) A clinical trial of retroviral-mediated transfer of a revresponsive element decoy gene into CD34+ cells from the bone marrow of human immunodeficiency virus-1 infected children, Blood 94, 368Ð371 Natasha J. Caplen PhD [email protected]