While significant progress has been made in recent years to optimize microbial strains and drive efficiency in biomanufacturing, a disconnect between upstream and downstream work continues to hold back progress, claims Boston Bioprocess cofounder Dr. Michael Tai.
Founded in Boston in 2022 by Tai (formerly head of bioprocess at Motif FoodWorks) and Ted Netland, Boston Bioprocess moved to Champaign, Illinois, in 2023.
The firm, which offers fermentation and downstream processing services at benchtop and pilot-scale, supports a wide range of companies from startups to established players looking to optimize their processes or develop new products.
AgFunderNews (AFN) caught up with Tai (MT) at the SynBioBeta conference in San Jose to get his take on what is holding back the bioeconomy.
AFN: What problems were you trying to solve when you founded Boston Bioprocess?
MT: When we started Boston Bioprocess there was a significant relative weakness of offerings in the bench development and processing scale up area, compared with relatively strong capabilities in strain engineering and development in the front end.
You could be very familiar with the performance [of your strain/bioprocess] in a 500-liter bioreactor, but there is a significant gap to be able to accurately predict performance in a 500,000-liter bioreactor.
Likewise, it’s very hard to tell how [upstream] fermentation performance can be integrated with downstream performance. So if I have a 10g-per-liter protein and you ask me what is your final harvesting yield after multiple stages of, let’s say, TFF [tangential flow filtration], it’s still very hard to predict.
So one thing I believe is critical is the integration between upstream and downstream work. There is still a significant resource poured into optimizing fermentation but relatively speaking an insignificant amount of resource going into the downstream process to solve questions as small as anti-foam [a common additive in bioprocessing to prevent or reduce foam formation] interference.
So when an upstream engineer says, if I put five more milliliters of anti-foam in my process, I can increase my fermentation yield or titer by 10% before the downstream guys know about it, then you might have to sacrifice quite a bit of downstream yield.
So I do think that it’s the integration and the talent pipeline between the fermentation and downstream area that is, in my mind, probably the top priority if we want to build a better biotech industry.
AFN: Do we need a broader range of microbial workhorses for industrial bioprocessing?
MT: We have two choices. One is that you keep further engineering [existing industrial microbial hosts] based on a very strong chassis, for example, Pichia [yeast], which is probably one of the most established ones. Then you keep expanding and developing the toolbox to the point where you know every single thing about it and you can engineer every single thing about it.
The second approach is putting more resources to further explore [wild microbes] and to domesticate [those] that might not be well-studied but have strong, unique capabilities.
There are pros and cons for both approaches. Obviously, if you have a well-studied chassis, you might be able to brute force it [to make it express a target protein, say]. So that answers the yes or no question, but it does not answer the question of how much or how good.
The more copies of plasmids [DNA molecules used to carry and express foreign genes in microbes], you put it in, cells can start losing their productivity, and it’s very hard to maintain the healthy levels that you want in a longer fermentation.
Conversely, if you are trying to domesticate wild type microbes, Trichoderma is a great example… If you are trying to put natural host enzymes in there, you can easily get to 180 grams per liter of proteins, which I will say the majority of startups would be very jealous of! But the problem is that if you put exotic protein genes in Trichoderma, you may not be able to express or secrete them without having significant problems.
So I think we just need to study more to see what type of molecule or what type of expression pathways would fit an existing chassis with further optimization, or would fit better into a wild type exploration that would enable you to express these molecules better.
AFN: Could continuous fermentation be a gamechanger if it can be used for more products? Â
MT: There are strong application cases in some areas. The easiest example is biomass fermentation, where you are looking at quantity, maybe certain nutritional values, or protein percentages.
But where you need a certain specific functionality over certain expressed proteins or molecules, or if you need to apply that into a certain induction mechanism coupled with growth areas, then you might face significant challenges.
When you run, let’s say, eight weeks of fermentation, then try to have a seed tank completely in a growth model [where the cells proliferate], and then the moment you feed them [the microbes] into the main fermentation tank [where you want them to express the target protein or substance], you want to maintain them at peak productivity for eight weeks, that is a significant challenge.
Further reading:
🎥 From jumping genes to designer genomes: Anthology’s blueprint for smarter bioproduction strains
The post What is holding back the bioeconomy? In conversation with Boston Bioprocess CEO Dr. Michael Tai appeared first on AgFunderNews.
