Leveraging protein and gRNA engineering of diverse CRISPR systems to optimize adenosine base editor activity for correction of a causative monogenic disease mutation

I’m Sean Crosson, and I’ll be sharing how ElevateBio can help partners optimize adenine base editors for specific pathogenic SNPs when target access, editing potency, bystander editing, or off-target activity limit a therapeutic program. 

For base editing, this technology can be used to change a G nucleotide to an A nucleotide. And what the implications of that are is there are certain genetic diseases that have this mutation that causes a deleterious mutation that causes disease. So here at Elevate what we did is we were looking to develop a base editing platform to correct one of these deleterious mutations in gene A here. 

So if you look at this top panel, what our approach was first is here at Elevate, we have a bunch of different CRISPR proteins that can target different areas of the gene. And what we do first is we actually screen a whole collection. This is a representative data set showing three that we screened here. And select what our lead editor is going to be. 

After we did that, what we first discovered is while this positioned the base we were trying to correct in a good window, the actual potency of the editor was fairly low. It was down about 10%. So what we did is we tackled all three different parts of this gene editing system.  

So there’s the nickase that’s cutting the DNA. There’s the deaminase that’s actually doing the base change. And then there’s the guide RNA that’s targeting the editor to the specific spot in the genome.  

So here in this panel, what we did is do parallel engineering approaches to the nickase. And what that resulted in is by combining all of these we were able to increase the potency from 10%, up to 60% just by engineering that part of the editor. 

If you move down to the bottom panel. We also did some engineering on the guide RNA. And so this is the part that targets it to the actual position in the genome. And by modifying the backbone and the sequence and adding different modifications, we were able to further enhance the potency of this editing system. And if you move to the top right, we moved on to the third part of the editor, which is the actual deaminase that’s doing the G to A change. 

And we did an engineering campaign on that to both increase the potency and the specificity of this editor. And what we saw was that we could actually reduce some of these unintended bystander edits, while still increasing the potency for the on-target position. And as we move down, you can see in this last panel, when we combine all of these three elements together, the nickase, the deaminase, and the guide RNA, we can see that our final product has much enhanced potency over what we started with and also has less unintended edits. 

We also took a look at the off-target profile, which is here in the bottom left. And while this target actually had fairly low number of off target sites that were edited initially, as we increase the potency, we actually saw even less off-targets. So we were able to increase the efficacy of this editor at the target position without increasing any of the off-target sites. 

And in fact, we actually ended up decreasing them overall. So this really summarizes how this parallel engineering strategy for modifying the entire base editor enables this editing system for gene correction of this disease. 

This work shows how combining editor selection, protein engineering, and guide RNA optimization can help partners move from an initial base editing concept to a more potent, specific, and target matched candidate.