Move over transgenic mice; your days as the go-to model for studying biology and disease outside humans may be numbered. While genetic modification has become one of the cornerstones in the effort to understand basic molecular biology, the problem for awhile has been that researchers have had to rely on relatively crude tools to engineer these genetic changes. Homologous recombination might have opened the world of transgenic mouse models, but these techniques weren't efficient enough to work well in other organisms. And alternative approaches to modifying the genome require making a host of relatively indiscriminate mutations and then trying to select for desired phenotypes. However, recent advances in genome engineering methods are changing the way researchers create genetic modifications as site-specific nuclease technologies are finally making it possible to edit any organism's genome.Putting the Zinc Finger on It
A decade ago, two articles were published that demonstrated how zinc finger nucleases (ZFNs) could be used to introduce mutations in specific locations in both Drosophila and in a human cell line (1, 2). The work signaled that with the right combination of DNA binding domains it might be possible to build a much wider range of tools for introducing knockouts and adding or editing out genetic mutations, providing wider genetic access to biological organisms.
ZFNs are composed of zinc finger domains that recognize DNA fused to the cleavage domain of the Fok I restriction endonuclease. Although zinc fingers are the most common DNA binding domain in the human genome, and include patterns of repeating cysteines and histidines that coordinate to zinc ions, each zinc finger recognizes a specific sequence of 3 nucleotides. When fused to Fok I, the endonuclease forms a dimer and each subunit clips the DNA at the location specified by the zinc finger binding.
Following initial demonstrations of the genetic editing abilities of ZFNs, research efforts took off to examine their full potential. Sangamo Biosciences, a biotechnology company in Richmond, CA, started working on ZFNs soon after the initial papers were published in 2002. “[ZFNs] allow you to place a cut in the DNA at a defined genomic address,” says Philip Gregory, chief scientific officer and vice president for research at Sangamo Biosciences. “Such a cut creates a very accelerated system for both the knockout or indeed the replacement of DNA precisely at the site of the break.” The repair process is error prone, which can lead to deletions or insertions that fully knock out the function of a gene of interest. But if a sequence of repair DNA or other new genetic information is included, that sequence can be incorporated at the site of these double stranded DNA breaks.
Initially, researchers thought that they would be able to string together fingers that bind combinations of three nucleotides to create arrays of zinc finger domains that recognize a DNA sequence of interest. But it turns out that constructing ZFNs that bind specific sequences requires a certain level of expertise. “While that's in part the case, if you start mixing and matching and changing the order of fingers in an array you change their specificity,” explains Dan Voytas of the University of Minnesota, who is also a consultant for Cellectis, a French company that develops genome engineering technologies. “They're not completely modular. You can't just stitch them together and get the specificity that you want.”
In an effort to work out some of the challenges of ZFN design, Voytas along with J. Keith Joung, MD, Ph.D. of Massachusetts General Hospital, co-founded an open source collaboration called the Zinc Finger Consortium that includes more than a dozen academic research labs who pool resources to better understand ZFNs. In 2008, Zinc Finger Consortium members published an article detailing a selection-based method called oligomerized pool engineering (OPEN) for designing active ZFNs (3). Although effective, the OPEN strategy is also labor intensive and still requires users to have a certain level of expertise with ZFNs. This challenge led the consortium to follow up OPEN work with another platform called context-dependent assembly (CoDA) that analyzes pairs of zinc fingers that work to string together in longer chains of functional fingers (4). While not as effective as OPEN, CoDA is software driven and avoids carrying out selections on libraries of proteins, according to Joung.