In today’s #apaperaday, Prof. Aartsma-Rus reads and comments on the paper titled: CRISPR-Based Therapeutic Gene Editing for Duchenne Muscular Dystrophy: Advances, Challenges and Perspectives
Today’s pick is a review paper on CRISPR based genome editing approaches for Duchenne by Chen et al in MDPI Cells. The review gives an introduction to Duchenne and dystrophin and genome editing, but I’ll focus on the editing types and challenges. Doi 10.3390/cells11192964
The review contains an overview of many CRISPR approaches tested in vitro and in vivo for Duchenne in case people are interested in an overview table. I’ll outline the different editing types authors discuss for Duchenne starting with the exon excision double cut.
This uses 2 CRISPR (guides) to cut at both sides of an exon to cause this exon to be deleted in order to restore the reading frame (readability of the dystrophin transcript). This allows production of a Becker-type dystrophin. This can also be used to delete multiple exons.
This double cut approach has been shown feasible for removing one and multiple exons. However, removing one exon is more efficient. The challenge of this approach is that using two CRISPRS can lead to more unexpected events than using just one.
Single cut genome editing for Duchenne can be done in 2 ways to restore readability of the dystrophin gene transcript. 1. Making the exon invisible for the splicing machinery by destroying the exon recognition site (cutting at the splice site, which is then mutated during repair)
The second approach is to cut inside the exon and then change the shape of the exon such that the readability is restored. This would in theory happen in 1/3 of cases, but for some targets the desired shape occurs more often (and in others it occurs less often).
Both single cut approaches rely on the fact that after cutting DNA, the DNA repair mechanism of the cell will repair the DNA, but this will be with errors. Normally that is bad, but here the errors are used to make the exon invisible or to reshape it so readability is restored
Knockin approaches aim to add DNA to replace missing parts for deletion mutations. This means that you have to deliver yet another component: 1. The cas9 (scissor), 2. The guide RNAs (CRISPR) 3. The template you want to be inserted into the DNA.
This approach is feasible in cultured cells and animal models but very high doses of AAV were needed to achieve dystrophin restoration in vivo. Finally authors discuss approaches that rely on a single stranded break and that are less likely to cause permanent unwanted DNA changes
First base editing allows changing 1 base pair in the DNA. This can be used to correct point mutations OR to make exons invisible. Prime editing allows reshaping of exons, but also correcting small deletion or insertion mutations. Both approaches are currently very inefficient.
Authors discuss 4 challenges:
- Safety related to AAV. The doses currently needed to deliver genome editing components are not tolerable by humans. So either the efficiency needs to improve (or I personally add, we need better AAV that go to muscle more specifically)
- Immunogenicity to AAV and to Cas proteins. About 20-50% of people have preexisting immunity to AAV and about 60-80% of people have preexisting immunity to Cas9. It is clear that for AAV this is a problem, what the effect of Cas9 immunity will be in humans is not sure. However, Dongshengh Duan already showed that it leads to lower efficiency in a mouse model.
- Off target activity – it is known there will be cuts at unwanted regions. Authors outline that attempts are made to improve specificity. Personally I think that for systemic treatment with a viral vector, long term Cas9 expression in multiple tissues is a risk. There is no undo button for unwanted edits. Authors suggest using tissue specific promotors to ensure Cas9 is only expressed in muscle
- Durability. Authors correctly point out that with genome editing partially functional dystrophins are produced, so likely there will be turnover of muscle and a transient effect. They propose targeting the satellite cells to generate a permanent pool of reframed cells
If that were possible it would be a very good solution. However, until now it has been very difficult to target the satellite cells. They are inactive and thus do not participate in taking up stuff.
Authors are positive these challenges can be overcome and there will be a clinical translation in the near future. Near future is a flexible term of course…Regardless, I am not so sure that systemic genome editing will be in clinical trials for Duchenne soon. Hope I’m wrong…
Finally, a comment about the introduction: the authors use the ‘c-word’ (cure). Genome editing can in theory cure Duchenne. No it cannot! First, approaches mainly aim at restoring a partially functional dystrophin. That is not curative. But more importantly, even when normal dystrophin is restored (e.g. with duplications), the accumulated damage will not be undone. All muscle lost will remain lost. Function lost will remain lost. That is not the definition of a cure.
Given the risks involved with genome editing (not certain it will be effective, there are risks of unwanted edits), it is important we (researchers) paint a realistic picture of the potential benefit so IF a trial comes, patients/families can make an informed decision.
Overpromising can lead to patients and families taking on risks they would never find acceptable had they known this is not a cure.