In this talk, Jack Newman, co‑founder of Amyris Biotechnologies, explains how synthetic biology can replace shark‑derived squalene with yeast‑produced hydrocarbons for high‑performance lubricants. He begins with a World War II narrative, showing how Japan’s reliance on high‑performance fuels for its Zero fighters drove a need for squalene, a compound difficult to scale from sharks. Newman then outlines the engineering challenge: reprogramming industrial yeast, which naturally ferments sugar to ethanol, to instead synthesize long‑chain hydrocarbons. He quantifies the genetic changes required—approximately 120,000 base pairs added, 41,000 deleted, and 450 single‑nucleotide edits—to swap the “ethanol app” for a “hydrocarbon app.” The core methodology is a rapid design‑build‑test‑learn cycle powered by Living Foundries, using automated DNA assembly (ligase cycling reaction) and high‑throughput fermentation screening. He showcases the Brotas, Brazil production site co‑located with a sugar‑cane mill, where iterative cycles (over 11,000 design iterations) drove the cost of bio‑derived squalane from $12 / L to under $3 / L, making it competitive for lubricants. Partnering with Novi, Amyris converts these biophene hydrocarbons into base oils that outperform petroleum‑based groups (III, II, IV) in thermal‑oxidative stability, demonstrated by viscosity‑time curves from extreme testing environments (desert, Arctic, marine). The resulting lubricants meet military specifications, are biodegradable, and comply with new environmental regulations. Newman concludes that the “eureka moment” is a myth; progress comes from cumulative engineering cycles that recode living systems, enabling a materials revolution without reliance on sharks.
Chapters
Jack Newman was one of the founders of Amyris Biotechnologies. They're very well known for using synthetic biology to create artemisinin acid, a cure for malaria. They've made, I believe, over 120 million cures to date. I have the really large pleasure of introducing Jack because he is coming to DARPA as a program manager because he believes in this vision, and he's going to tell you about it. Thank you.
Thanks, Alicia. Thanks for that nice welcome. On December 7th, 1941, Japan, a nation engaged in a massive and expansive war effort in Asia, attacked the United States at Pearl Harbor. Virtually no one in Japan wanted a war with the United States, and practically everyone who was paying attention to world affairs agreed that Japan had little chance of winning such a war. To understand why a nation engaged in a massive war would take on a second front, you have to know a little bit about what was going on at the time. In 1940, about 80% of the oil and gas that was imported into Japan came from the United States. The war had doubled its demand for oil, and Japan knew it only had about a year and a half left of oil supplies for a war in China that would take at least 3 years. In 1941, the United States embargoed important imports into Japan, including oil, gas, and rubber. Japan had little choice but to attack to maintain access to key materials that it needed for the war effort and to maintain its expanding economy. The fighters used to carry out that attack were these, the AM6 Type Zero, commonly known as the Zeros. The Zeros ironically used the very same high-performance fuels and lubricants that the United States embargo denied. There was, however, a biological source in this squalus shark, which makes this compound called squalene. Squalene makes an excellent lubricant for fighters because of its very cold freezing point and because of its very high thermal stability, which enabled about 1 minute of overdrive that wouldn't be possible before the engines of a Zero would freeze. The problem with this biotechnology, it's difficult to scale. Even today, where squalane is used only in the highest end of emollients for skin softeners, it's really difficult to source. What we want is something easier than sharks. We want yeast, which can perform in industrial realms like this ethanol facility. In fact, last year delivered over 95 billion liters of ethanol. And simply, why does yeast make ethanol? Yeast makes ethanol because it is hard-coded into the DNA program that's what that organism is supposed to make. What we want is the high production efficiency of a yeast organism or a microorganism like this married with the genetic code that allows us to get to high-performance materials like hydrocarbons that make squalane, which today sells on the emollient market for about $30 a year, but even go beyond that to get to low-cost lubricants that can be even higher performance than the squalane that was used in those Zeros. How do we do that? Basically, the difference between yeast making ethanol and yeast making high-performance hydrocarbons is simply this. About 120,000 base pairs added, about 41,000 base pairs of DNA deleted, and about 450 single nucleotide base pair changes. That's it. It's an information technology. Sounds simple. So we can go into the operating system of yeast. Here I show the vast chemistry of all the chemical reactions that go on inside of yeast. The operating system, and here the app that turns sugar into ethanol. So if we want to make hydrocarbons, pretty simple. Delete the ethanol app, introduce the hydrocarbon app, make farnesine. That's about 12,000 base pairs added, 6,000 deleted. Not much. And for that, you get the eureka moment. The eureka moment usually comes in the form of a chromatogram like this. Where you look at what's been made in your yeast fermentation. Usually requires some amount of explanation. No, it's not that. It's actually the little bump over there. But it's the first time you've made this product in your living factory. Maybe it's the first time this chemical's ever been made at all. But in my opinion, and I hope of what I'll convince you of today, is the real eureka moment— oops— is this one. Where you've actually started to make enough of this material at the kiloton scale where you can start to change the way we look at how we source materials. That requires a bit more engineering. This is more what it looks like. 121,000 base pairs added, 41,000 deleted, 450 base pairs changed. And to get to that kind of efficiency, what you need is an iterative DNA coding and learning system, the one that Alicia mentioned. You start with a design, you build it, you test it, you learn something. And what drives progress for this design-build-test-learning cycle is how much code you can add or delete per cycle to learn something, but also how many times you can go around this cycle. And how much you can learn, the learning driven by the iterations through those cycles. This is what DNA coding now looks like at Amherst thanks to Living Foundries. Sitting around having a cup of coffee coding into a laptop designs that will end up in a liquid handling system that are going to bring together molecules of DNA and then using a new way of putting together these molecules called the ligase cycling reaction, also developed under Living Foundries. Now those ideas that are in your head as a coder of DNA turn into the thousands of designs that end up in your yeast. End to end, what we've done here is nothing short of turning the zeros and ones that occur in a DNA— in a computer program and turning those into the Gs, As, Ts, and Cs that now program life. What we've done this— what we've done here is done it basically 10 to 100 times faster, but we can trace the history of this sort of of technology back to some of the examples that have been pointed out today. It is the recoding of DNA in the form of recombinants that gave us insulin from yeast rather than cadavers or farm animals. Biologics, vitamins, vaccines, all those things, and now low-cost malaria drugs. This molecule squalane is made this way, 1,3-propanediol. And every day we are reminded that this revolution in DNA coding will continue to deliver new products from morphine to jet fuel if we can continue to work down this cost curve and bring kiloton scale of these new materials. Now there's one more ingredient that you need in order to bring these molecules to market. Of course you need the very high efficiency organism, but you also need the production facility that is going to make these molecules possible. So here I show the Amaris production facility in the background here, co-located with a sugar cane crushing plant, which basically takes sugar cane out of the fields, crushes it into juice, which then goes down this pipeline, and bagasse, the fiber— here you see a truck pulling in— which is burnt to make electricity for the area, power the plant, dewater the juice, and then feed that sugar into this highly contained and advanced low-cost manufacturing site which is able to handle a highly engineered organism. If you remember nothing else from this talk, remember this homely slide. This is the impact that Living Foundries had on the production of hydrocarbons at Brotas, Brazil in the 2 to 3 years that it was running. We started in Brotas with a generation of yeast that it had been well over 11,000 design cycles and iterations. That got us to a product that was about $12 a liter, which is fine if you want to make squalane for cosmetics. A few thousand cycles later, we get down to $4 a liter. Great for your margins, still in cosmetics. A few thousand iterations lower and you get under $3 a liter. That gets you into the high-performance lubricants, solvents, and performance materials that we're after in this materials revolution. Enter Novi. This is a high-performance lubricants and fluids company that takes the Amris biophene hydrocarbons and converts those to base oils and lubricants for all of the products that you see here: engine oils, gear lubricants, two-cycle oils, compressor oils, hydraulic fluid, transmission oils. All of the high-performance fluids that we source from petroleum, we can make better from biology now that we've gained control of the programming. The key attribute for a high-performance lubricant is a concept I've already introduced, and that is this thermal oxidative stability. And what I'm showing you here is a lubricant formulation where we've looked at the various petroleum cuts, Group III, Group II, Group IV, and compared them against the same formulation using the hydrocarbon made from yeast, that is Novuspec. And what you see here on the Y-axis is the 200% viscosity. That's basically the point where your engine oil has turned into sludge. And your engine is toast. And time on the x-axis, and what these new materials allow you is a much longer lifetime for your engine oil and a much higher thermal and oxidative stability. Zeros flying with this could do a lot better than a minute in overdrive. Now, taking that base oil and formulating it into an advanced engine oil and comparing it against all the other advanced engine oils out there, you get something even better than the best product on the market today. That's what new materials can do, and do it at a cost-competitive price. Because the market and nobody else is going to use it if it's not cost-competitive. These engine oils have been tested in rigorous environments, in the deserts in Las Vegas, on the north slope of Alaska, and they outperform all mineral oil products. These same high-performance products are used in hydraulic fluids and greases which conform to military specifications that are very, very high. And finally, in marine environments, not only do these products conform to spec, meeting and exceeding the performance requirements, but they also comply with new environmental regulations that require that when we use these oils that might contact water, that they're biodegradable. And so they don't stick around in the environment when we don't want them to. Well, if there's anything that I've convinced you of today, I hope it's that the eureka moment is a bit of a fallacy. It's great for movies, but actually, as scientists and engineers, we are always standing on the shoulders of last year's giants. And by building on everything that we've learned in information technology, and taking those paradigms and using it to recode living systems, we can source new materials without the sharks. Thanks very much.
