Prime editors: interesting science & aggressive PR

Rumors had it – scientists at recent conferences were discussing prime editing with curiosity. Ten days ago I noted in my task manager:

And here we are now, a day after prime editors flew through the sky and caused a sonic boom.

Is prime editing really that big?

In case you missed it: the media virtually freaked out. Some outlets wrote pieces about CRISPR fixing “almost all genetic diseases”, whereas other wrote that the same tool is better than CRISPR.

To be fair – it’s easy to be confused about the relation between prime editing and CRISPR. Authors of the discovery avoid references to CRISPR and try hard to contrast prime editing and Cas9 (CRISPR-associated protein 9). They have a point here, as they put a lot of work to design/evolve their molecular complex. But at the same time, they are still relying on Cas9 (variant H840A), so we can clearly describe prime editing as a member of CRISPR toolbox.

Publication about the tool is available here, with lengthy supplemental section here.

Prime editor is a clever combination of Cas9 nickase, M-MLMV reverse transcriptase, and exclusively designed pegRNA. In short, part of pegRNA provides a sequence to search in the genome, Cas9 finds it and cuts a single strand of the genomic DNA, then reverse transcriptase synthesizes DNA from the second part of pegRNA (containing corrections, insertions, or deletions), this creates a stretch of hanging DNA, which – with help of nicking by Cas9 – can replace the targeted sequence.

Capabilities of prime editing are pretty outstanding: it performs all 12 transitions and transversions, as well as insertions up to 44 bp and deletions up to 80 bp. All of that without explicit double strand breaks. From that point of view, prime editors are a big thing. One would use the Swiss Army knife metaphor, if it wouldn’t be overused in regards to CRISPR.

That’s where the headlines about “curing all diseases” come from. As the authors calculated, prime editor could target 89% pathogenic genetic variants in the ClinVar database. Obviosly, that it is not an equivalent of 9/10 genetic diseases, but this is a nuance. The real question is: how close is it to the clinical setting?

Prime editing was not tested in vivo

The study worked on cell cultures only. There were no experiments with animals. We simply don’t know if prime editors will work efficiently and precisely in living organisms.

It may be not so easy to use prime editing in humans. The complex has reportedly large size. Delivery of such system into tissues is a major issue. The authors acknowledge presence of the problem and assure that they are “working hard with hope to deliver prime editing to animals”. In the light of that fact, headlines about fixing almost all genetic diseases are at least exaggerated.

Not all gene therapies require in vivo administration. There are approaches such as CAR-T which utilize ex vivo genetic engineering – cells can be collected from a patient, modified in a laboratory, and then transplanted back into the body. The study proposes use of prime editing for sickle cell anemia. The most popular mutation behind this disorder – A to T in HBB gene – was corrected in human cells with 44% efficiency and 4.8% frequency of undesired insertions/deletions. However, the caveat is that the mutation was corrected in human embryonic kidney cells (HEK293T, pictured underneath) – which are quite far from hematopoietic stem cells targeted in current therapies. As tests in other cell lines yielded much lower efficiencies (down to 10%), we can be mildly concerned about potential efficiency in comparison to current therapies which reach 25% efficiency.

On the other hand, it’s worth noting that prime editing seems precise in comparison to previous CRISPR methods. Lack of double-stranded breaks significantly limits unwanted insertions/deletions. There were hardly any off-target effects at sites known to be affected by standard CRISPR-Cas9, but we are waiting for evaluations on the whole genome level.

How much PR does science need?

Prime editing was heavily marketed to the public by the Broad Institute. There is no doubt that the work was embargoed and sold to journalists as a revolution. In one hour, we’ve seen dozens of articles with similar quotes and descriptions – and pretty wild headlines, as mentioned before.

There is a reason for that massive PR campaign. The scientists secured a patent for prime editing and set up two companies. Although anyone in academia is allowed to use prime editing without any hurdle, it’s totally different for any commercial entities. Exclusivity granted, at least for now in human gene therapies, to companies named Beam Therapeutics and Prime Medicine run by the team behind the paper is quite a game-changer in the whole story. Selling it as a revolutionary tool better-than-CRISPR, which will cure almost all diseases is an ideal catch for investors.

That is not to downplay the science – it’s sound and beautiful, but still limited and far from anything therapeutic.

We’ve seen similar PR campaign around the previous big discovery in the same lab. David Liu’s scientists made base editors in 2016. Since then, the original paper was cited over 1,000 times – an equivalent of large recognition in this area. But at the same time, since then, major problems with base editors came to the surface, for instance – substantial off-target effects and severe limitations of activity. After that story, we should be at least more careful in believing and propagating Broad Institute’s PR machine.

Publication: Andrew V. Anzalone, Peyton B. Randolph, Jessie R. Davis, Alexander A. Sousa, Luke W. Koblan, Jonathan M. Levy, Peter J. Chen, Christopher Wilson, Gregory A. Newby, Aditya Raguram & David R. Liu (2019). Search-and-replace genome editing without double-strand breaks or donor DNA. Doi:10.1038/s41586-019-1711-4.

Leave a Reply