Gene transfer: Knockouts with a touch of a zinc finger

signaling update home

SOURCE:

Capitalizing on the imperfection of error-prone DNA repair, researchers generated knockout mammalian cell lines by targeting a nuclease to a gene of interest via a zinc-finger DNA-binding domain.

Faced with a new gene product, researchers often want to know the phenotype of a cell that is forced to do without it. Knocking out these target genes by homologous recombination is now routine in mouse embryonic stem cells. But as Philip Gregory, vice president of research at Sangamo BioSciences, points out, "most other cell types are quite recalcitrant to homologous recombination–mediated gene knockout, so the efficiency of that process is very low."

But there is another way. In mammalian cells, in addition to homologous recombination, double-strand breaks can be repaired by nonhomologous end joining—an imperfect repair process that helps cells survive in times of stress but results in alteration of genetic information at the site of the break. To take advantage of this imperfection to create a knockout cell line, the researchers needed to introduce a double-strand break within a gene of interest.

Sangamo is a company focused on developing zinc-finger technologies, and for this project they used an engineered zinc-finger nuclease that comprises a zinc-finger DNA-binding domain and an endonuclease domain that cleaves DNA upon dimerization. Thus, two zinc-finger domains are designed to recognize a specific sequence within the target gene; when they bind their target sites, two fused nuclease domains are brought together and create a double-strand break. While repairing this lesion by the endogenous nonhomologous end joining process, the cell effectively 'makes' the knockout.

As proof of concept, Gregory and colleagues targeted the DHFR gene in a Chinese hamster ovary (CHO) cell line, which is often used for production of therapeutic recombinant proteins. After transiently transfecting these cells with plasmids encoding different optimized zinc-finger nucleases, they found that 7% of the isolated clones had mutant DHFR alleles in one experiment and 3% in another. Sequencing of the PCR-amplified target regions revealed that about one-third of these clones had a disruption of both alleles of the target gene.

One advantage of this approach is that no selection marker is used. Also, in traditional homologous recombination–based knockout approaches, often only one allele is mutated, which requires marker excision and retreatment to isolate a homozygous knockout—a time-consuming process. The surprise in this work was that the frequency of both alleles being mutated was relatively high. It turns out, as Gregory explains, that "cells that have nucleases that are expressed and are sufficiently active to cleave at one allele tend to also cleave at the other, so both alleles are simultaneously disrupted at a high frequency." As a result, he estimates that the time required for this approach is less than half of that required for current homologous recombination–based methods, such as those that use adeno-associated virus (AAV)-mediated delivery.

The hardest part of both of these approaches is making the targeting tools, and neither is an off-the-shelf system. "At the moment most labs that find it difficult to make an AAV vector would also find it difficult to make a zinc-finger protein," Gregory notes. "But both are increasingly available through commercial entities and will be available to researchers." Sangamo's zinc-finger proteins and zinc-finger protein nucleases are now being distributed by Sigma-Aldrich.

In the meantime, Sangamo scientists are partnering with Pfizer and others in the industry to use this method to create improved cell lines for manufacturing proteins for human therapeutics. But another direction for this work, according to Gregory, is that it might be a method to create transgenic animals. "Because of the simplicity of the method—the fact that you only have to make a double-stranded break—and the frequency with which you get biallelic targeting, it's an exciting possibility that this will be a method for the creation of transgenic animals."

Irene Kaganman

References

Santiago, Y. et al. Targeted gene knockout in mammalian cells using engineered zinc-finger nucleases. Proc. Natl. Acad. Sci. USA 105, 5809–5814 (2008) | Article | PubMed |