Synthetic biology · Biocontainment and bioprocessing

Synvivia

A UC Berkeley spinout that built molecular on and off switches so engineered microbes die unless they are fed a specific chemical. A decade on, the same chemistry is aimed at a bigger, duller problem: the cost of making gene therapy vectors.

A lock and a key

In June 2015 a bioengineering graduate student at UC Berkeley published a way to make a bacterium dead by default.

Gabriel Lopez and his adviser, associate professor J. Christopher Anderson, took a laboratory strain of E. coli, picked genes the organism cannot live without, and rewrote them so the proteins they encode only work when a particular small molecule is present. Take the molecule away and the cell stops. The compound they used was benzothiazole.

“This approach is very robust and simple in that it only requires a few mutations in the genome,” Anderson told Berkeley News. “The molecule serves as the key, and we engineer the lock.”

Lopez reached for a car. “What we’re doing is putting an ignition switch onto a handful of the bacteria’s essential genes. Without the right key, the bug won’t live.”

The numbers were the point. According to the Berkeley account, switching a single gene improved viability a hundred millionfold, and stacking several locks into one organism pushed that to ten billionfold. The National Science Foundation and DARPA paid for the work. It ran in ACS Synthetic Biology, and UC Berkeley says it was picked as an Editor’s Choice in Science.

The containment problem

Engineered microbes are being designed to live in soil, in fermenters, and inside patients. The recurring safety question is not sabotage. It is what happens when an organism built for one place persists somewhere else.

Most earlier answers used a switch that poisons the cell on command. The Berkeley pair argued for the reverse logic: make death the resting state, so survival takes a deliberate act.

They were careful about what it does not do. The method does not stop engineered DNA moving between organisms, and it is not built to withstand someone deliberately undoing it. “We’re never going to be able to solve the problem of a human actively hacking away at our safety systems,” Lopez said in 2015. Anderson set the boundary this way: “We have to look at what it is that we’re worried about with which application in determining which biocontainment approach is relevant.”

One founder, one license

Lopez stayed on as a postdoc in Anderson’s lab and turned the method into a company. UC Berkeley’s technology transfer office dates the founding to 2015 and the Y Combinator cohort to 2016; a Cognition IP case study says 2016. Synvivia holds an exclusive license to the Berkeley intellectual property in all fields of use, with Lopez and Anderson as inventors. In 2021 it was incubated at the QB3 East Bay Innovation Center.

It never got big. Y Combinator’s directory listed a team of four in September 2026. The company’s own SEC filing in December 2023 described Lopez as the sole executive officer and sole director. The current NIH grant names a different principal investigator, Charlie Huang, who is also a co-inventor on the company’s viral vector patent application.

From containment to yield

In a 2018 trade interview Lopez described the goal in absolute terms. “Synvivia’s mission is to provide ultra-robust biocontainment for genetically engineered organisms so that synthetic biology can be used outside of the lab.” He compared the switches to the component that made modern electronics possible: “Our technology is to biology as transistors are to electronics.” Asked when it would be commercially available, he answered: “The short answer is now. The long answer is that it depends on your application.”

In 2020 it appeared at NVIDIA’s GTC conference, using GPU accelerated molecular dynamics to design a protein switch and screen drug candidates against a SARS-CoV-2 target.

Then the target moved. A patent application assigned to Synvivia covers engineered cyclic peptides that raise the titer and transduction efficiency of viral vectors. In December 2024 the National Institute of General Medical Sciences awarded a Direct to Phase II SBIR, R44GM158472, to build a platform for finding those peptides and using them to manufacture adeno-associated virus vectors. It obligated $1,640,169 in fiscal 2025 and $856,268 in fiscal 2026, and runs to November 30, 2026. Y Combinator’s listing now reads: “Synvivia applies chemical control over cell behavior. This allows us to optimize biomanufacturing of high-value biopharmaceutical products.”

The biocontainment work did not stop. It changed hands. On November 21, 2025 five authors, Lopez and Huang among them, posted a preprint describing a modular kit for making E. coli depend on theophylline, a common asthma drug, by putting a riboswitch in front of essential genes. Synvivia funded the work. The competing interest statement is unusually plain: “Synvivia, Inc. is not developing, and has no plans to develop, commercial products or licensing activities based on this toolkit.”

A company named for living together had published its safety kit and declined to own it.

What is proven, and what is still claimed

Read plainly: the science is real, published and still being done, and the company is still operating. What is missing is a market. Eight years after the founder said the containment system was available, there is no public trace of a customer, a licensee or a price. The money keeping the lights on is federal, and most of it is not for biocontainment. It is for making gene therapy cheaper to manufacture, a market with buyers.

What to watch

• What happens when R44GM158472 ends on November 30, 2026. A renewal or a commercialization award would show the AAV work is landing.
• Whether the cyclic peptide application issues as a granted patent, and where.
• Peer review of the November 2025 preprint, and whether anyone outside the company picks up the kit.
• Any Form D larger than $54,096, the first sign of outside equity at a size that changes the company.
• A named manufacturing customer. One disclosed partner would be worth a decade of grants.

In their words

Related companies

• 20n: Also in Industrial Bio. A Y Combinator biotech whose software invented a fermentation route to the active ingredient in Tylenol, then published its whole stack as open source in March 2017 and
• Ansa Biotechnologies: Also in Industrial Bio. An enzymatic DNA synthesis company in Emeryville that sells the genes other vendors decline
• 64x Bio: Also in Industrial Bio. A Harvard spinout that engineers the cells which manufacture viral vectors, rather than the therapies those vectors carry
• Lygos: Also in Industrial Bio. A Berkeley Lab spinout that taught yeast to make malonic acid without cyanide, then became a specialty chemicals supplier whose polymers end up in shampoo, diapers and

Sources

1. Public recordNIH RePORTER 1R43DK112475-01NIDDK · FY2016
2. Public recordNIH RePORTER 1R44GM158472-01 and 5R44GM158472-02NIGMS · FY2025 and FY2026
3. Public recordHHS TAGGS award detail, R44GM158472HHS · period December 2024 to November 2026
4. Public recordNSF SBIR Phase I award 2126430NSF · 2021 to 2022
5. Public recordForm D, Synvivia, Inc., CIK 0002003070SEC EDGAR · Dec 11, 2023
6. Public recordA Modular Toolkit For Theophylline-Dependent Synthetic AuxotrophsbioRxiv · Nov 21, 2025
7. Public recordPatents assigned to Synvivia, Inc.Justia Patents
8. IndependentScientists use molecular ‘lock and key’ for potential control of GMOsBerkeley News · Jun 16, 2015
9. InterviewProgrammed for safety: How Synvivia is commercializing biocontainment technologySynBioBeta · 2018
10. IndependentInception Spotlight: Synvivia Developing a Protein Switch for COVID-19NVIDIA Technical Blog · Oct 9, 2020
11. IndependentSynvivia Inc SBIR awardee profileinKnowVation · edited Jun 17, 2022
12. IndependentSynvivia company research noteWhiteford Research Biobase · 2024
13. CompanySynvivia, Inc. startup profile and timelineUC Berkeley IPIRA · accessed Sep 22, 2026
14. CompanySynvivia company page, status Active, team size 4Y Combinator · accessed Sep 22, 2026
15. CompanySynvivia websiteInternet Archive capture · Jun 14, 2026
16. CompanySynvivia case study with founder quotationsCognition IP · 2023