WHY THIS MATTERS IN BRIEF
As more companies develop their own synthetic DNA products this new innovation will lower costs and increase the speed of manufacturing new genetic products.
In a time when our ability to create six and even eight base pair DNA synthetic organisms, that have no natural equals, improves, and when we can see a time when artificial humans are created, and when people question whether the pandemics we’re experiencing, like COVID-19, are natural or manmade, we now have news that the amount of time it takes to write new synthetic DNA code has just gotten a whole lot faster.
It’s long been known that scientists can read DNA sequences faster than ever before, something that was recently used to our advantage to create the world’s first ever real time genetic map of the COVID-19 pandemic, but their ability to write DNA hasn’t kept pace. As a result companies who want to order tailored DNA strands, companies such as Microsoft who recently spent millions of dollars buying strands that formed the foundation of their new ultra-dense DNA cloud storage platform with, have to make do with short strands, synthesised in a slow and expensive chemical process.
And now that’s all changing after researchers from a French biotechnology startup announced at a synthetic biology meeting in San Francisco, California, that by using close relatives of the DNA-writing enzymes in living things, they can build DNA strands as long as 150 “letters,” or nucleotide bases. That’s up from a record of 50 nucleotides just a few months ago, and nearly on par with the standard chemical approach.
“This is a significant milestone,” says George Church, a geneticist at Harvard Medical School in Boston, who was not involved with the work and is working to develop similar techniques. The result puts enzymatic DNA synthesis “on a curve that looks to grow exponentially.” If it does, Church and others say, researchers wanting to create new genomes or use DNA to archive vast troves of information will soon have longer DNA fragments made faster and more cheaply.
Although traditional chemical DNA synthesis has been miniaturized and automated, the underlying technique, known as phosphoramidite chemistry, is virtually unchanged since it was developed in the early 1980s. It involves adding nucleotide bases one at a time, each capped with a protecting group that stops it from reacting to extend the strand until scientists remove the cap and add the next base.
The approach isn’t perfect. With each letter added, there’s a 0.5% chance of an error. The longer the chain, the greater the chance that it will contain an error, effectively limiting DNA strands to about 300 bases long. As a result, researchers who hope to write genes containing thousands of letters must laboriously stitch fragments together. Enzymatic synthesis promises to do better by co-opting polymerases, the enzymes living things use to fasten nucleotides together into long, virtually error-free sequences.
Although enzymatic DNA synthesis efforts started only this decade, about a half-dozen companies are working on the approach, according to Church. Others are pushing data storage and biomedical applications to be ready when DNA printing rates improve. According to Michael Kamdar, CEO of Molecular Assemblies, an enzymatic DNA synthesis company in San Diego, California, the market for traditional chemical DNA synthesis today is about $1 billion annually. The market for data storage is more than $14 billion, although DNA likely suits only a small fraction of it.
“In the next 2 to 3 years you will see applications on the market with enzymatic DNA synthesis, if not by us then someone else,” says Sylvain Gariel, co-founder and chief operating officer of DNA Script in Paris.
Shifting from chemical synthesis to polymerases has brought challenges though. In living cells, most polymerases start with a template strand and create one that’s complementary, pairing As with Ts and Gs with Cs. A specialty polymerase in immune cells, called terminal deoxynucleotidyl transferase (TdT), works without a template, making it scientists’ most common choice of polymerase for enzymatic synthesis. Yet because TdT adds new DNA letters randomly, researchers have had to find ways to force it to add just the desired letter, one at a time. Gariel’s team does so by equipping each DNA base with a proprietary protecting group that, as in chemical synthesis, prevents TdT from adding more than one letter at a time to the growing strand. After the correct letter is added and its protecting group is removed, the cycle repeats. According to Gariel, adding each letter takes only 5 minutes and occurs with an accuracy of 99.5%.
“That’s great,” says William Efcavitch, co-founder and chief scientific officer of Molecular Assemblies. That speed and accuracy, along with the 150-nucleotide-long strands, bring enzymatic synthesis nearly on par with the 5- to 10-minute cycle of conventional phosphoramidite DNA synthesis, and many experts believe the enzymatic approach has plenty of room for improvement.
“The potential of enzymatic synthesis far surpasses chemical synthesis,” Kamdar says. Ultimately, DNA Script CEO Thomas Ybert says he expects his company will write 1000-base strands in a day very soon and that they hope to begin selling automated enzymatic DNA synthesizers later this year.
Such improvements could cut the cost of writing DNA by one to two orders of magnitude, Efcavitch predicts. That would make it easier and cheaper for synthetic biologists to design and test new genes for developing everything from new catalysts to medicines. It could also revolutionize DNA data storage, which, in theory, could capture all the world’s information in a volume smaller than a shoebox.
For the last several years, such applications seemed fanciful, because fast and cheap enzymatic DNA synthesis seemed over the horizon, and Kamdar adds: “The reality now is that the horizon is here.”