Raphael Thys
Engineering pathways is still an art; commercializing the products is a black art. The protein production pathway is a remarkable exception, scaling reliably and offering new kinds of functionality for products. With a greater agility and a spectrum of innovative commercial applications as examples, protein is SynBio’s killer app.
SynBio is fighting a headwind of complexity, resource intensive product development and barriers to market. A generation of fast, profitable companies could change that Image credit: U.S. DOE. 2006. Breaking the Biological Barriers to Cellulosic Ethanol: A Joint Research Agenda, DOE/SC/EE-0095
With vivid images images of petroleum bubbling to the surface of an algae covered pond, Synthetic biology promises to be the philosopher stone of 21st Century Industry — turning sugar into just about anything imaginable — plastics, fuel, flavorings, therapeutics, bringing the trillion dollar petrochem economy to sustainability. Even so SynBio has had a hard time delivering.
SynBio Products have multiple barriers to market
Complexity is always the enemy of progress. In the case of SynBio, even modest increases in the number of enzymes in a pathway increases the work and time required exponentially. For each enzyme, additional variables like cofactors, competing pathways and cell viability enter the equation. When microbial strains that make any one of millions of compounds are created, they start producing tiny amounts — often around a few milligrams per liter.
After adding the metabolic pathway, scaling the production in fermentation, and then navigating the sales chain of multi-ton market for commodity chemicals have been significant problems for getting SynBio products to market as well.
Each class of compounds presents almost as a unique problem and for even pathways of modest length, substantial time and resources must be brought to bear to discover new solutions.
There are successful companies making bio-chemicals. Zymergen, among other companies, is an amazing example of an exception to the current rule of the struggling SynBio startup.
SynBio’s manifest destiny is real; the petro to bio conversion of commodities will happen. Just as spreadsheets are not all of software, SynBio could use a killer app… a generation of companies whose products are so needful , work so consistently, they establish biotech as a ubiquitous technology platform opening the way to broad, large and accessible markets.
Protein production scales by a work function
When it comes to the issues of strain engineering, scale, and market, one type of pathway is a remarkable exception. Producing protein has proven to be a leading basis for a new generation of biotech companies and has enormous untapped potential.
Rapid protein turnover is related to cell survival, and so protein production can run in high gear in any organism. Protein-making strains can be nearly perfect factories — in some cases converting 90+% of the dry mass carbon of feedstock into product.
This is good news. A killer app with log scaling can drive large markets. As production is scaled up, costs and efficiencies must change by orders of magnitude for the app to make business sense. The classic example of this reality is found in computing power and the software industry: in going from hertz to gigahertz, Moore’s Law scaled across eight orders of magnitude. DNA sequencing is biotech’s first famous example, but biology has other such scaling laws, making it a powerful force for the 21st century.
Scaling for protein production is a work function — the more work you put into a protein producing strain the better the yield. In the crude estimate in the figure, I attempt to include both fermentation scaling and strain engineering costs over time. Typically over the course of 10+ years, the costs associated with producing a gram of protein drops by up to a factor of 10,000 and can fall to as little as dollars per pound. This has amazing implications for proteins as commodity, but the status of protein as a killer app is that proteins have new classes of functionality in addition to scale.
The Protein Tech Future
In terms of user adoption, proteins are only just starting to see their potential realized. Proteins and small molecules are the yin and yang of cellular function. Small molecules are synthesized and transformed by proteins, while proteins are activated and modulated by small molecules as well as by each other. In the end though, the vast majority of biological function is performed by proteins.
Proteins are extremely versatile in their structure and function. Proteins make devices and materials available that are completely new that is simply not possible before at commercial scale. New proteins change rules, creating new products. We can only begin to imagine what kinds of technologies will be brought about in the SynBio protein space. Working with a single family of proteins opens up spectacularly large commercial spaces.
New proteins change rules, creating new products.
Already many companies are breaking ground with enormous promise. Several are fashion and food companies which have found markets much more easily than most SynBio companies.
Bolt Threads, Spiber and AmSilk are making spider silk commonly available — stronger than steel and light as cotton. Millions of spiders were needed to produce a spider silk cape in 2009. The spiders were literally yoked and milked for their silk by hand. As these companies edit and access the silk genes from the tens of thousands of spiders around the world, materials of differing adhesion, strength and elasticity will be possible. Imagine bulletproof outerwear, performance sportswear, construction cranes that fold into a moving van.
GelTor, a company I helped to fund, is making vegan gelatin through fermentation strains, but more profoundly it is a biomaterials platform of incredible versatility. Collagen is already a material that is used in nearly everywhere from food to cosmetics to reconstructive surgery. Usually derived by boiling a carcass in acid, GelTor can provide any of the millions of collagen genes from millions of animals and microbes in bulk, there are limitless new materials whose cost will drop in time. Performance pharmaceutical delivery systems, scaffolds for organ reconstruction, biodegradable leather materials for building, fashion or consumer packaging.
Proteins are largely responsible for the texture and mouthfeel, flavor, nutritional profiles of food, making new kinds of foods possible. Clara Foods is discovering the direct connection between proteins and the culinary uses for eggs. Taking the various separate protein components of egg whites they have made meringues that do not weep, lower cost but firmer pastas. In french cuisine the chef’s toque has 100 folds for the 100 ways to cook an egg. Maybe it should be 300. MiraculeX is shipping protein based sweeteners being produced by multiple systems but also demoing in many products.
Using proteins as a tech platform has been taken up by larger companies tpp. Hampton Creek has become a plant protein materials company; Modern Meadow now declares the basis of its technology is collagen.
The action and interaction of proteins will continue to astonish us as companies leap ahead with discovering new products rather than replacements for the ones we know, creating new economic vistas.
Consider gluten. Although a humble wheat protein, it has generated cuisines across continents. It activates when wetted, forms networks that create bubbles, fibers, and the scaffold for many of the most appealing foods we know. Engineered gluten has the potential to mitigate allergic reactions, create new textures, and new categories of food. But can it also inspire building and fabrication materials? Bioplastics? Self-assembling devices? All of these possibilities are up-and-coming, and proteins can be scaled much more quickly from prototype to product than can a tech device.
Given all this, there is an incredible need for better tools. Already several platforms are being created to jumpstart protein production to a much higher initial productivity than a starter strain. Right now the best industrial strains have beaten 100 grams/liter in production. How can we reach those sort of results more quickly and consistently and for any chosen sequence? Can we find even better systems and methods for protein production from the standpoint of sustainability, desirable post-translational modifications?
These are challenges, but possible a next wave of companies that will really bring SynBio to global impact.
This post was first featured on the SynBioBeta Blog Sept 25, 2017.