The new gene-editing tool homes in on single problematic letters in someone's genetic code. (The Broad Institute)
CRISPR/Cas9 has been justifiably ballyhooed as the best gene-editing tool scientists have ever had to correct broken genes.
But it is by no means perfect. For instance, it only works correctly a small percentage of the time -- as little as a half-percent. And what is worse, it can add mistakes to the DNA of some cells on top of the ones it has failed to correct.
Recently, researchers affiliated with Harvard, MIT and the Broad Institute reported in the journal Nature that they had made some major improvements to CRISPR/Cas9. By homing in on single letters within the human genetic code, these tools corrected DNA significantly more often than the older version of the gene-editing tool, leaving behind a lot fewer mistakes as well. Below, you can see Broad Institute researcher David Liu describing this programmable "machine" he and others have created. Liu says two-thirds, or roughly 33,000, of the known genetic variants associated with a disease "are single-letter changes that are thought to play a role in human diseases."
Like the original CRISPR/Cas9, what's being called an Adenine Base Editor, or ABE, fixes the DNA of a gene. With an ABE, scientists corrected their target DNA a whopping 34-68 percent of the time, depending on the mutation that’s targeted, with not even 0.1 percent of cells showing a newly created mistake.
Sponsored
That’s compared to a correction rate of only about .5 to 4 percent with CRISPR/Cas9 sent to corresponding DNA sites. The new-mistake rate for CRISPR is roughly 3 to 11 percent.
CRISPR/Cas9 targets specific sites in DNA, but not necessarily specific enough.
How are ABE and CBE different from the Original CRISPR/Cas9?
The instructions in DNA are written using four molecules (or bases) called adenine, guanine, thymine and cytosine. These are usually abbreviated as A, G, T and C.
Many genetic mistakes consist of a single alteration in the combination of these letters. For example, where a G may be normally found, a patient has an A instead. In that case, the goal of editing the gene would be to change the problematic A to a G.
CRISPR/Cas9 does this in a roundabout way. When it is sent to that A by scientists, it makes a clean cut somewhere near in the DNA. This incision sets off alarms bells in the cell, prompting it to immediately begin closing the cut.
In order for this natural process to result in a corrected gene, scientists need to package CRISPR/Cas9 with a piece of DNA that contains a G in the correct position.
The cell uses the added DNA as a model to correct the sequence.
CRISPR/Cas9 then essentially swaps out a large part of the DNA around that erroneously positioned A with the added DNA that has a correctly situated G. You’ll notice the emphasis in that last sentence, the point being: This if fairly inefficient, as a lot more than just an A is replaced.
But ABE neither makes an incision to the existing DNA nor adds the "template" DNA that CRISPR/Cas9 does. Instead, it homes in on one base, the A, then turns it into a G. A companion tool to ABE likewise transforms a C into a T.
In these base editing tools, the Cas9 molecule can no longer make a complete cut in the DNA. Instead, the researchers attach an enzyme, a protein from the cell, to Cas9. In the case of ABE, that enzyme converts an adenine (A) into an inosine (I). (The cell can't tell the difference between an I and a G and so treats the I as a G.)
Why Is This Approach Safer?
A big problem with CRISPR/Cas9 is that cut DNA is so harmful to the cell that it moves to repair it as fast as it can. So fast, in fact, that it often introduces a lot of mistakes. When this occurs, instead of correcting errors, CRISPR creates new ones.
The end result can be a hodgepodge of cells with different configurations of DNA -- some of the cells will still have the broken DNA, some will have newly damaged DNA, and a small fraction will include the repaired DNA -- that G that used to be an A, for example.
But since ABE does not cut the DNA, cells are not sent into a frenzy of repair. Without that incision in the DNA, the chance of introducing new errors drops by as much as a factor of 100.
Will These New Tools Replace CRISPR/Cas9?
Not completely. These new gene-editing tools appear to be very good at turning As into Gs and Cs into Ts, but they won’t work for other kinds of mutations.
Sometimes a genetic disease occurs when a bit of DNA is missing, called a deletion, or when there is an extra bit of DNA, called an insertion. The most common mutation in the genetic disease cystic fibrosis, for example, results from a deletion. The original CRISPR/Cas9 system can fix these kinds of mistakes, but the new editing tools can’t.
Another New Tool Targets RNA
A different tool, created by a second group of Broad Institute, Harvard and MIT researchers, changes the messenger RNA, or mRNA, in a gene, instead of changing the gene itself. This tool is called REPAIR, for RNA Editing for Programmable A to I Replacement. The best version of REPAIR corrected 13-27 percent of mRNAs.
In order for a gene to make changes in a cell, it must first be read by the cellular machinery. The first step, called transcription, is when the gene is copied into a short-lived molecule called mRNA. This mRNA usually goes on to be translated into a protein. It is that protein that does the work in the cell.
Up until now, gene editing has been done at the DNA level, where scientists are permanently changing the instructions found in the gene’s DNA. Every mRNA that emanates from the altered gene will include the change.
But DNA and RNA are actually very similar in that they both use bases (although RNA uses a U instead of a T). The new tool changes an A to a G in the mRNA coming from a particular gene.
So the DNA itself, the original instructions, are not changed. But the disease is treated because the repair has been made in the mRNA.
This system might be used for temporary conditions like inflammation. The technique is also ostensibly safer than CRISPR/Cas9 because any changes made are not permanent. If the base editor makes a mistake, those changes will be short-lived. In contrast, any changes made in DNA are forever.
