Pat Adams

Up to 50 people will receive $70k to give away, as microgrants, to create new scientific fields. Good opportunity!

This is a simple method to make "bacterial chimeras" (or trimeras) by moving ~2 million bases of DNA between bacteria in a single step. The method is called ACE, or Additive Conjugative-CAST Engineering. It's described in a new paper from Kaihang Wang's group at Caltech. They used this method to make, in one case, an organism with 4.6M bases of DNA from E. coli DH10B, 2M bases of DNA from Syn61 (a recoded strain of E. coli) and about 900kb from Shigella (meaning that this works across bacterial genera.) When I joined Caltech in 2018 to start my PhD, I was actually the first graduate student in Kaihang’s lab! I remember unboxing the equipment, getting the new pipettes out of the box, and installing the DNA sequencing machine. My bench was like two rows down from David Baltimore's group. This is the first project talked about when I joined, meaning that they've been working on this for at least the last 8 years. The way it works is that you have a donor strain and a recipient strain. You decide which DNA from the donor you want to transfer, and you flank that sequence with two bits of DNA, called CassetteSTART and CassetteEND. The donor is also engineered to express a helper plasmid, which encodes the machinery needed to move the DNA to the other strain via conjugation. Once the DNA moves into the recipient strain, a *second* helper plasmid splices it into a chosen location in the genome (via a CRISPR-associated transposase.) They use this simple approach to move up to 2 million bases of DNA in a single step. When they tested this on a 1 million-base strand of DNA, they got 1,000 successful colonies per experiment. (Meaning this works REALLY well.) Across three target sites, 98.1–99.7% of mapped insertions were at the intended site, too. And when they did RNA sequencing on these chimeras, they found that all the genes from all the donors were expressed. So why is this useful? Well, we can already move large segments of DNA, but DNA is fragile and can shear. You can order oligos or genes and assemble big sequences, but that takes a lot of time and money. With this tool, you could instead just identify a biosynthetic pathway in a cell and transfer it into a strain that’s easier to grow or use for biomanufacturing. This is a cheaper, more scalable way to "port" capabilities between cells. (But only bacteria of relatively closely-related genera for now.) This also works with synthetic DNA. Theoretically, then, you could "install" DNA into a cell and then use that strain as a sort of perpetual donor to transfer to other organisms, instead of rebuilding and re-transforming cells each time. I'm sure this will have other, as-yet-unseen uses as well.