DNA writing is an aspect of our industry that I’ve been closely watching for several years because it is a critical component of so many groundbreaking capabilities, from cell and gene therapies to DNA data storage. At the SynBioBeta Conference in 2018, the co-founder of a new startup that was barely more than an idea gave a lightning talk on enzymatic DNA synthesis — and I was so struck by the technology the company was aiming to develop that I listed them as one of four synthetic biology startups to watch in 2019. I watched them, and I wasn’t disappointed.
Ansa Biotechnologies, Inc. — the Emeryville, California-based DNA synthesis startup using enzymes instead of chemicals to write DNA — announced in March the successful de novo synthesis of a 1005-mer, the world’s longest synthetic oligonucleotide, encoding a key part of the AAV vector used for developing gene therapies. And that’s just the beginning. Co-founder Dan Lin-Arlow will be giving another lightning talk at this year’s SynBioBeta Conference in just a few weeks. I caught up with him in the lead up and was truly impressed by what Ansa Biotechnologies has accomplished in just 5 years.
Why DNA writing needs enzymes
Synthetic DNA is a key enabling technology for engineering biology. For nearly 40 years, synthetic DNA has been produced using phosphoramidite chemistry, which facilitates the sequential addition of new bases to a DNA chain in a simple cyclic reaction. While this process is incredibly efficient and has supported countless innovative breakthroughs (a visit to Twist Bioscience’s website will quickly educate you on exciting advances in drug discovery, infectious disease research, cancer therapeutics, and even agriculture enabled by synthetic DNA) it suffers from two main drawbacks: its reliance on harsh chemicals and its inability to produce long (read: complex) DNA fragments.
This latter problem is particularly tricky because that requires multiple DNA fragments to be individually synthesized and stitched together to form the final, full-length product any time anyone wants to work with fragments longer than ~200 bases. That stitching process is usually done by the research lab, not the synthetic DNA vendor, and sometimes it simply doesn’t work.
Enzymatic DNA synthesis leverages a polymerase enzyme called terminal deoxynucleotidyl transferase (TdT), unique among polymerases in that it doesn’t require a template in order to generate a new DNA chain. Its natural role is to generate a highly diverse repertoire of antibodies that our bodies can use to fight off a variety of threats. But as all good synthetic biologists will, several innovators in our field have begun leveraging TdT to provide a better alternative to phosphoramidite-based DNA synthesis.
Enzymatic DNA synthesis produces less waste, boasts higher fidelity, and can produce longer DNA molecules than phosphoramidite chemistry in a very rapid timeframe. When Lin-Arlow and Ansa Bio’s co-founder Sebastian Palluk first described their novel approach to enzymatic DNA synthesis in Nature Biotechnology in 2018, their TdT enzyme-deoxyribonucleoside triphosphate (dNTP) conjugates could add a new base every 10-20s (extrapolated to the 1005-mer they recently synthesized, that’s a little over five and a half hours).
Enabling a revolution
The seed for Ansa was planted in Lin-Arlow’s mind when the computer scientist-mathematician joined Jay Keasling’s lab to gain some experience engineering biology — something he had spent quite some time supporting by developing enabling software. It didn’t take long for him to run into a source of serious frustration: getting the DNA he needed to engineer his cells turned out to be pretty painful.
Contrasting his experience with what he knew about computer science, Lin-Arlow remembers thinking that “If it took a month to recompile code, we wouldn’t have all this tech that we now enjoy and that’s changed the world — but that’s what it feels like in many cases when you’re engineering cells.” He found that all too often, vendors couldn’t make some part of the DNA construct that he wanted, and he wouldn’t find out for weeks that they’d failed, or they could make it but then when he got all the parts he couldn’t stitch them together. Back to square one — and often the same issues would happen again. So, he made a decision. “I decided I was going to step back from engineering cells and instead work on tools for DNA synthesis, because I knew that for this revolution to happen, we really needed to tighten up the feedback loop,” he says.
Shortly after making that decision, Lin-Arlow met Palluk, who was working on using enzymes to synthesize DNA. It didn’t take long for them to develop a proof of concept and report their first synthetic oligo produced enzymatically in Nature Biotechnology. It was 10 base pairs long — a good start — but their dream was to synthesize 1000-mers so that scientists could expand their discovery potential and stop struggling with gene parts that couldn’t be assembled into the full-length product.
“There’s so much biology out there that’s unexplored because you can’t get access to complex genes,” he says. “Nowadays people are pursuing more and more biologically complex applications that require large and complex DNA constructs, like cell therapies with finely tuned regulatory circuits so the cells don’t activate in the wrong context. What we’re trying to do is build a new DNA manufacturing platform from the bottom up that can handle all of those types of sequences, and our vision is to give our customers their sequences exactly as designed, on the first try.”
Now, just five years after reporting that first 10-mer, Ansa has nearly met their goal of routinely providing 1000-mers to the market.
The next 30 years of (synthetic) biology
In an important next step toward reaching that goal, Ansa launched their early access program last month, which provides synthetic DNA constructs to a select group of partners. The company asked these partners to send in their most challenging sequences that they hadn’t been able to get anywhere else, and at the time of our chat, Lin-Arlow said they’d been able to make nearly everything they’d received up to that point.
Right now the product is up to 500 base pairs long and delivered in Ansa’s plasmid — only half of the company’s goal but still hundreds of base pairs longer than the industry standard. And that length is going to increase over the next year, says Lin-Arlow. He plans to talk about some of these examples during another lightning talk at this year’s SynBioBeta conference in just a couple of weeks. In just five years the company has come full-circle, with many of the things that were just dreams for Lin-Arlow and Palluk during the first lightning talk now reality. But there’s still work to do.
“There’s been tremendous progress in engineering biology, even over the time that Ansa has existed, but it still is falling short of the promise of being able to program cells in the way that we program computers,” Lin-Arlow reminded me. “There’s so much promise and possibility for the field that could be unlocked by advances in the foundational technologies. I think this is really just the beginning.”
Eventually, he says, Ansa will offer pooled oligos, asking me to imagine an array synthesizer that can make hundreds of thousands or even millions of very large gene-length oligos. It might seem like a pipe dream, but people are asking for it, says Lin-Arlow — and he plans to deliver.
Importantly, he says, Ansa Biotechnologies doesn’t plan to ever make and sell DNA printers. By providing fully QC’d, gene-length synthetic DNA directly to customers, not only can the company ensure that these powerful oligos aren’t misused, but they can also provide the researcher what they really want, the desire he himself had years ago in Keasling’s lab: removing the need to even think about how they get the DNA to do their experiments. Keasling, who was there from the very beginning, says that this is “going to be great for all the synthetic biologists who want cheap, high quality DNA.” His own research will benefit greatly from access to gene-length oligos, he says.
“Reading DNA has defined the last 30 years of biology, and writing DNA will define the next 30 years,” Lin-Arlow told me as we parted ways. “The best DNA fab will enable the most advanced, cutting-edge applications. And we’re trying to build that fab.”
If the next 30 years is led by companies like Ansa, then our industry has a truly bright future. I can’t wait to see what we accomplish.
Thank you to Embriette Hyde for additional research and reporting on this article. I’m the founder of SynBioBeta and some of the companies I write about, including Ansa Biotechnologies and Twist Bioscience, are sponsors of the SynBioBeta conference. For more content, you can subscribe to my weekly newsletter and follow me on Twitter and LinkedIn.