Student disruptors invent new way to synthesise DNA
Frustrated at the slow pace of putting code together, two young researchers took things into their own hands. Elizabeth Finkel reports.
Researchers have borrowed from nature to develop a new way to synthesise DNA. It promises to be faster, cheaper and greener than the current 40 year-old chemical method.
The technique, published in the journal Nature Biotechnology, was invented by two frustrated graduate students at the US Department of Energy's Joint BioEnergy Institute, based at Lawrence Berkeley National Laboratory in California.
It promises to be a game-changer for synthetic biologists who design and build genes from scratch to tackle global challenges such as affordable drugs, biofuels and biodegradable plastics.
“What this means for science is that we can engineer biology much less expensively – and in new ways – than we would have been able to do in the past,” says Jay Keasling, a senior author of the paper.
Graduate students Sebastian Palluk and Daniel Arlow were drawn to synthetic biology by lofty ambitions. Arlow joined Keasling’s lab, which is famous for engineering yeast to make the anti-malaria drug artemisinin, to develop bacteria-based biosensors. Palluk, based at the Technische Universität Darmstadt in Germany, had been developing bacteria that would degrade plastic, as part of the international synthetic biology competition iGEM.
Synthetic biologists build circuits made of custom-designed genes to carry out pathways of chemical synthesis. Like all engineers, making progress relies on rapid iteration through cycles of design, build, and test.
But the students were frustrated by the amount of time it took to get their computer-designed gene circuits back from custom-order DNA synthesis companies.
“You might have to wait a month or more to test if your idea is going to work,” says Arlow.
The problem was a 40-year-old one. The companies were using a method that relied on linking up the individual letters of DNA (the nucleotides referred to as A,T,C, and G) one at a time via chemical synthesis.
It was error-prone, time-consuming, used hazardous chemicals and could reliably make only 200-letter-long strands at a time. These would need to be chemically joined together to make 1000 to 2000-letter-long genes.
Nature, they knew, did a much faster and more accurate job. The bacterium Escherichia coli replicates its DNA in a watery non-toxic medium at about 1000 nucleotides per second.
But there were two problems. First, the main enzyme used to synthesise DNA, known as DNA-polymerase, can only copy a pre-existing template. So it’s not much use to researchers who are trying to make a brand new sequence.
On the other hand, there was a weird type of enzyme known as terminal deoxynucleotidyl transferase (TdT), found in vertebrates and only used to make DNA in the immune system.
Its actual purpose is to deviate from the fidelity of the DNA code to introduce random changes in antibody genes so they can keep pace with fast evolving viruses and bacteria.
Able to spool out 200 bases per minute, it’s not as fast as the bacterial enzyme but still a great improvement on the chemical method.
Researchers had long eyed the potential of TdT. Indeed, Palluk had also joined the ranks of those trying to tame the enzyme to meet the needs of synthetic biology. The idea was that by feeding it nucleotides in the right order, it would grab them like a crochet hook and chain them up in the correct sequence.
But there was a problem. The enzyme tended to hook in more than one of the given letters at a time, so instead of ending up with a sequence of A-T-C-G, you might get A-A-T-T-T-T-C-C, and so on.
Palluk tried to solve the problem by putting a chain blocker on the end of each DNA letter, so only one at a time would be added. But the enzyme would not accept the blocked letters. In mid-2015, he moved to the Berkeley lab to join forces with Arlow.
The students came up with a very leftfield idea. Instead of tinkering with the components of the chain, they decided to tinker with the enzyme – the crochet hook.
They prefabricated four set of enzymes so that each was conjugated to just one of the DNA letters. This effectively restricted enzyme, so that it could only ever add one component.
DNA synthesis proceeded by adding these prefab units to the growing chain – in effect, adding not just the links, but the crochet hook as well. The hook would then be cleaved away before adding the next enzyme-letter unit.
Keasling applauds the students' disruptive idea: “Rather than reusing an enzyme as a catalyst, they said, ‘Hey, we can make enzymes really inexpensively. Let's just throw them away.' So the enzyme becomes a reagent rather than a catalyst.”
So far, the pair have used the approach to show they can make a DNA sequence of 10 letters but are confident that they will be able to eventually make one with 1000 bases at many times the speed of the chemical method.
"Our dream is to make a gene overnight," says Arlow.