NgAgo: Genome-editing by a DNA-guided enzyme

      While much of the scientific community was very busy optimising CRISPR-Cas9 technology for their affectionate model systems (mammalian cells, plants, organisms- mouse, zebrafish), a recent article in Nature Biotechnology has left everyone baffled.

The article was published online ahead of print in Nature Biotechnology journal last month, authored by Chunyu Han lab members from China. The authors demonstrate that a protein of Argonaute class can be utilised to edit genome using a DNA guide sequence. The enzyme belongs to the bacterium Natronobacterium gregoryi and has been christened  as NgAgo (Natronobacterium gregoryi Argonaute). By the way, I like rhyming NgAgo with San Diego, so if you are writing a rap song about genome-editing, just give it a thought!

But, seriously, why so much buzz about an enzyme that is guided by a  24-nt DNA sequence rather than 20-nt RNA sequence as in CRISPR-Cas9 system?

Well, it appears that NgAgo system has many advantages over Cas9 system.

• DNA synthesis of 24-nt guide DNA (gDNA) is cheaper than RNA synthesis. Also, RNA is a very delicate and less-stable molecule requiring more attention when setting up an experiment.

• NgAgo system is simpler and requires just the 24-nt guide DNA that is modified at 5′ end (phosphorylated, companies can do that for just a few extra bucks). On the other hand, Cas9 system requires 20-nt guide RNA (gRNA) plus a longer tracrRNA molecule (around 70-nt long). Although, the two RNA molecules of Cas9 system can be expressed together as fusion single guide RNA (sgRNA). But that requires cloning of gRNA into a plasmid (phew cloning) or in vitro synthesis (even worse for postdoc/PhD students!).

•  Target site recognition does not require PAM (3-letters NGG) or any other sequence of that kind. This makes life easier for researcher to virtually target any part of the genome without bothering about the availability of PAM sequence. All you have to do is copy 24-nt DNA sequence from your desired locus and send it to an oligo company. No need to use a gRNA prediction tool. Also, some parts of the genome (such as at the start and end of the gene) are AT-rich and you might be unlucky to find NGG, making it difficult to design gRNA for Cas9 system for such regions.

• NgAgo system is powerful system to edit GC-rich regions. Authors have shown in the article that NgAgo system can penetrate GC-rich, obscure regions of genome and outperforms Cas9 system which can hardly make any impression there. This might have more to do with guide loading onto enzyme: gRNA with high GC content are prone to self-annealing and form secondary structure before they are loaded onto Cas9 enzyme.

• NgAgo system is more stringent and cannot tolerate mismatches in gDNA/target DNA binding. Making a cut at undesired sites (off targets) is a big issue when it comes to gene therapy (and molecular biology in general). This article highlights the fact that NgAgo system does not like mismatches in guide DNA sequence (a single mismatch out of 24-letters can compromise cutting activity whereas 3-letters mismatch completely abolishes enzyme’s ability to target a DNA sequence). Cas9 can tolerate up to 4-5 mismatches and will cut off-targets as well. This off-target issue might not seem very important for regular genome-editing of completely messed up cancer-cell lines but could be a decisive factor when thinking about editing genome in a patient (we are far away from that).

Before Cas9 supporters hit me , let me tell you that everything is not good about NgAgo and there are some Cas9 moments as well.

• gRNA loading onto Cas9 can occur at room temperature (in vitro) as well as physiological temperature (37 ⁰C) without affecting Cas9 activity. NgAgo is more fussy when it comes to gDNA loading:
  1. in vitro loading: If you want to use recombinant, pre-assembled NgAgo/gDNA for transfections, you will have to incubate NgAgo plus gDNA at 55 ⁰C for 1 hour. This weakens enzyme which can no longer make a double-stranded break but can only make a nick (cuts one strand). A nick is not very useful for genome-editing because it gets repaired by cells ‘repair machinery’ without leaving any changes at cut-site.
  2. in vivo loading: I am not sure if one can express gDNA (single-stranded DNA) in mammalian cells, and that too 5′ phosophorylated ! Therefore, you will have to transfect cells with NgAgo-expression plasmid plus gDNA. You must be wondering do I have to boil cells at 55 ⁰C for 1 hour for gDNA loading? Good news is that your cells do not have to go through such ordeal. It appears that gDNA can be loaded onto NgAgo enzyme  when the protein is being freshly translated/expressed.
  3. CRISPR-Cas9 is powerful tool for gene activation/suppression, tracking etc. However, we must remember that it took 3-4 years for Cas9 to celebrate its success in wide applications. No doubt that the NgAgo system can be engineered to do the same.

• Genome-wide studies have shown that after all Cas9 is not that bad when it comes to cutting at off-targets. Only head-to-head genome-wide studies in future will tell if NgAgo is actually better than Cas9 in terms of off-targets.

• Authors have shown that NgAgo-mediated indel frequency for many tested genes is 20-40%. Cas9 system was not tested in parallel for those genes. I would assume that Cas9 is more potent tool for gene disruption because many studies have reported up to 60-70% indel frequencies.

Will NgAgo overshadow CRISPR-Cas9?

Hold on guys, give it some time. NgAgo is just one article and a month old. It surely appears cost-effective and simpler but needs to be scrupulously tested head-to-head with Cas9 system. The biggest concern with NgAgo is the impossible task of expressing 5′ phosphorylated single-stranded guide DNA in cells. Cas9 guide RNA enjoys various methods of delivery: in vitro pre-assembly, plasmid-mediated co-expression, virus-mediated expression etc. But I am sure scientists will come up with better variants of NgAgo or who knows, altogether new class of guided-endoncleases!

We have come a long way- since the discovery of DNA as the genetic material to mapping of the human genome (3,200,000,000-letters) and now to editing any letter in the human genome. The last 10 years have seen so much progress in the field of genome-editing. The RNA guide-based system, CRISPR-Cas9, has revolutionised genome-editing area and made us more ambitious to target many more genes and organisms in less time and money. Future studies will unleash the potential of NgAgo system and show whether it can overtake powerful CRISPR-Cas9.