This article has been updated to correct details about the company's chip design.
BALTIMORE – British biotech startup Evonetix plans to commercially develop a platform for thermally controlled enzymatic DNA synthesis, following a version that relies on chemical synthesis.
Earlier this month, the Cambridge-based company said it has shown feasibility for the proprietary technology, which relies on selectively heating or cooling small regions on a semiconductor chip.
According to Evonetix, the technology is the "culmination of a three-year development program." It is “conceptually similar” to existing polymerase-based DNA synthesis methods, said Matthew Hayes, the company's interim CEO. However, instead of using acidic reagents to remove the 3'-blocking group on the nucleotide after each cycle, the approach deprotects the nucleotides via a thermal step, using the company’s semiconductor array-based thermal platform.
Evonetix hopes the DNA manufacturing method, still in its exploratory phase, will one day be applicable to library preparation, CRISPR or protein engineering, and eventually gene synthesis.
Initially built for phosphoramidite-based chemical DNA synthesis, Evonetix’s DNA writer centers around “a novel thermal control chip, which has a very large number of reaction sites on the surface,” said Hayes. The semiconductor chip — sandwiched together from ASIC (application-specific integrated circuit) and MEMS (micro-electrical-mechanical system) subcomponents — contains micro-reaction sites that are 100 microns in diameter and “capable of independently controlling the temperature of a little volume of liquid above” from 20 to 90 degrees Celsius, he said, while anchoring oligos directly on the surface via thermally cleavable linkers.
As a result, Hayes said, the chip can create “little islands" of micro-sites with independent temperatures called “virtual wells” within a continuously flowing liquid environment, enabling localized reactions “in the same way that other people localize them with physical wells.”
One of the main advantages of Evonetix’s technology, he said, is that it allows “highly parallel synthesis.” Compared with standard enzymatic methods where oligos are deprotected all at once by flooding them with acid, the company's process can “heat up sites where we want deprotection to occur, and we cool down sites where we don't want it to occur,” generating various end products simultaneously.
Using the technology, Hayes said Evonetix has previously achieved phosphoramidite DNA synthesis on the platform before testing its compatibility with enzymatic DNA synthesis. In both cases, to achieve the deprotections, Evonetix has engineered modified phosphoramidites or modified nucleotides that are “extremely sensitive to temperature,” he said.
As for the technology’s enzymology, the company declined to share more technical details at this point, citing pending patents. Other enzymatic DNA synthesis companies, including French startup DNA Script; University of California, Berkeley spinout Ansa Biotechnologies; San Diego-based Molecular Assemblies; and, recently, Twist Bioscience of South San Francisco, California, have been using some version of engineered terminal deoxynucleotidyl transferase, or TdT, to synthesize oligos de novo, meaning without a template.
However, Raquel Sanches-Kuiper, Evonetix’s VP of technology who spearheaded the firm’s DNA synthesis research efforts, said that although the company's enzyme is “template independent,” it is "not quite a TdT.”
But according to John Nelson, a senior principal scientist at General Electric who also works on enzymatic DNA synthesis, de novo enzymatic DNA synthesis, under normal circumstances, does require some version of terminal transferase. While Nelson said it is hard to gauge the specifics of Evonetix’s method based on the company’s public statements alone, the big question is “what is your error rate?”
For example, it remains to be seen how well the thermally controlled deprotection works, he said. If the process does not completely remove the blocking groups on the desired oligo strands during a cycle, the next base won't be successfully added to all strands, leading to deletion errors, which he called “a killer for enzymatic DNA synthesis.” Similarly, if the enzyme does not work efficiently to incorporate a base, it can also cause deletion errors. In addition, questions remain about the platform’s performance. For instance, if the system does not clean out well after each nucleotide cycle, the residual nucleotides could be added to the growing oligos by mistake, resulting in incorrect products.
Finally, Nelson noted that one disadvantage of thermally treating DNA is that it encourages deamination. “Certainly, deamination occurs much more quickly at high temperature,” he said. “In molecular biology, we hesitate to heat DNA up for too long because you end up with your Cs getting turned into Ts.” In the end, “all of these errors come up to an overall error rate, which will either be commercially acceptable or won't be commercially acceptable," he added.
Evonetix has not released any performance data for its thermally controlled synthesis technology yet, but Hayes said the phosphoramidite process has achieved “commercial equivalent error rates” and the enzymatic approach is “much earlier on in our development” and still needs to mature.
In terms of turnaround time, Hayes said the system currently takes “half an hour or so” per cycle. “Because we have this time element to our deprotection,” he acknowledged, “we don't have the fastest cycle time of all systems.” However, he said he believes the platform’s capability to synthesize a large amount of DNA “in a very parallel fashion” can compensate for its slower turnaround, and therefore, still achieve high throughput. Specifically, the current semiconductor chip prototype can accommodate 384 reaction sites, while company researchers “envisage much larger arrays, probably as many as 10,000 reaction sites.”
Moving forward, Hayes said Evonetix expects to release a version of the DNA synthesizer with phosphoramidite chemistry first, since “it’s a much more mature technology.” However, he did not provide a concrete timeline for the product launch, other than saying it will be “fairly soon.” While claiming an early-access program “will be imminent,” he also did not share a firm timeline.
The commercial platform will likely include three components, he said: the instrument, which he anticipates being “office printer size;” a cloud platform; and the consumables, which include the one-time-use semiconductor chips and reagents. He said the company has yet to settle on a price for the system, as it will “depend slightly on the customer base that we end up working with.”
As for the commercial prospects of the company’s enzymatic DNA synthesis method, Hayes said the technology “is much earlier in its journey, so it needs to mature before we integrate it into our product roadmap."