The benefits of modern gene editing techniques and tools come with some downsides. While the precision of new and improved CRISPR systems is continually on the rise, there are some fundamental aspects of the tools that scientists have long wished we could work around, in order to reduce even the most minimal risk of off-target changes to the genome.
How DNA Breaks
A primary example of this is the double-strand break or DSB. It has been a hallmark of gene editing, including with older tools like zinc finger nucleases (ZFNs) and TALENs, that in order to insert, remove, or otherwise alter the genetic code, both strands and complementary nucleotides of that code must be cut for the change to take place.
The problem with this is that, in most cases, we rely on the cell itself to repair the broken, cut strands after the alteration. With a direct addition and removal, this meant having two blunt ends of DNA that are often just ligated (welded) back together. Such a coarse repair method as this, termed non-homologous end joining (NHEJ), has a high risk of adding extra nucleotides to the point of ligation. And these sorts of mutations can be disastrous, since they are happening right next to the desired inserted or removed section of DNA code.
An attempt was made to avoid this with specific systems like CRISPR Cas12 (Cpf1) that created sticky, overhanging ends to each strand. This allowed the pieces of DNA to naturally rebind with each other, no repair needed. This method, called homology-directed repair (HDR), avoided many of the concerns that the prior systems engendered, but it wasn’t perfect.
CRISPR cleavage proteins, as a whole, are likely to cut more than once in a genome, especially if the targeted sequence in the guide RNA is short and can be found in more than one place. Even for Cas12, these additional cuts may not be HDR ones with sticky ends, but have the same sort of blunt end issues as with the other systems. The benefit here is that these cuts will usually be elsewhere in the genome and not affect the targeted and altered sections, but random insertion mutations from repair elsewhere could still prove extremely damaging.
Even without considering the issues that NHEJ bring to successfully incorporating gene editing changes, double-strand breaks also have the complication of being cytotoxic to the cell as a whole. Too many of them at once has a high likelihood of killing the cell, limiting the number of gene edits that can be made across the genome at the same time.
All of these and more explain why gene editing tools are often used only for broad, longer gene sequences changes and have difficulty in making smaller, more minute modifications.
A Singular Solution
Researchers at Yale University decided to try and make a new gene editing tool that avoided all of this. Past studies had shown that synthetically made oligonucleotides that were designed to be single-stranded and complementary to one of the target DNA strands (usually the lagging strand rather than the coding one) could be implemented as a way to make genetic adjustments without requiring double strand breaks.
Relatedly, the formation of a new mutation-causing tool several years back called multiplex automated genome engineering (MAGE) seemed promising. But it had mainly only been used in prokaryotes, especially E. coli. Though it allowed for multiple modifications to be made at the single nucleotide level with a greater than 40% accuracy, which is quite good when speaking on the level of changing one single base pair.
The process of single stranded oligonucleotides in eukaryotic cells, however, is less well understood and the DNA repair systems have greater complexity as well. Therefore, MAGE had yet to see fruitful usage in higher order organisms. Until now.
Deciphering Eukaryote Repair
Using Saccharomyces cerevisiae as a test for eukaryotic modeling, the researchers used single-stranded oligonucleotides with a single base pair change coded into them and the specific homology repair structure found in yeast. The genes involved in this mechanism were overexpressed and the genes used in mismatch repair were knocked out to improve the likelihood of incorporating the oligonucleotide. Though this only confirmed, from past study, that the efficiency increases are limited with this method.
They were able to get around this limitation by aiming for the change to be implemented during DNA replication at the replication fork (the part of the genome that is unwound so DNA polymerase can begin copying the strand). This process works independently of the homology-specific genes, specifically recombinase protein gene Rad51. Other recombinase genes, however, were shown to improve the likelihood of the oligonucleotide annealing to the DNA strand at the replication fork.
This study not only gave far deeper insight into the apparatus that eukaryotes use in their binding of oligonucleotides, it also was a proven success in using MAGE gene editing technology in eukaryotes. Having titled their new tool eMAGE in that regard, the researchers believe that it will prove highly consequential in understanding entire biochemical pathways and in developing synthetic chromosomes by allowing for dozens of single base pair modifications across the genome all at once without bringing harm to the cell.
These methods may even translate to efficient usage in plants and animals for precise mutation breeding without counting as transgenic changes. Only time (and the governmental regulatory system) will tell.
Photo CCs: Asco Candida albicans from Wikimedia Commons