Biotech basics in the battle for beef's bloodiness - Part I

Impossible vs Motif via GM yeast

This post summarises some biotech concepts and industry movements I’ve learned as a newcomer to the field. It’s intended to be narrative, notes and references that past-me would have found interesting and useful.

Impossible Meat mimetics

Impossible Foods' claim to fame is producing a plant-based burger product which has some of the unique characteristics of meat - taste, texture, colour and cooking behaviour. They have a patent which covers a realm of art, and they are particularly well known for their production of heme.

Heme is a family of molecules that provide the rich red colouring1 in raw meat and denature at temperature, giving the satisfying experience of changing colour during cooking. Technically hemes are a subset of macrocycle tetrapyrolles, four pyrolles2 arranged in a circular shape, combined with an iron ion. Taking their name from the Greek for blood, hemes are a precursor to hemoglobin, myoglobin and other similar proteins.

Heme derivatives are found in all plants and animals. Impossible Foods' heme is in the form of soy leghemoglobin, but they don’t harvest it from soy plants, rather they produce it using a process called precision fermentation.

Side note on precision fermentation

I find precision fermentation interesting enough to warrant its own tangent3. Here’s how Impossible brought blood to food without the meat.

First they identified the heme producing genes within the a soy genome, and genetically modified a particular strain of yeast4 to include those same genes, as well as a few other useful traits to increase overall productivity. The idea is that this genetically modified yeast, given the right conditions, will grow efficiently and then, when signalled, will express a target molucule - in this case soy leghemoglobin. Different organisms are differently able to express desirable molecules5, so the first win is when your yeast is expressing your target in the lab.

Alongside this genetically enhanced yeast is a matching precision fermentation plant. In the plant, the yeast is cultivated in bulk in a bioreactor and the signal for target expression is provided. Once done, the whole mass is passed through a fairly involved set of steps known as downstream processing6 to extract just the desired target.

It appears the ideal of being “more efficient than an animal” is not yet proven at scale. A precision fermentation plant is essentially a yeast factory, with the target molecule often making up <5% of the output volume. The process can theoretically be optimised at almost any point along the way, though each has its challenges - from the genetic engineering7 to the selection of yeast feedstock8, to expression signalling9, to the size and shape of the bioreactor10, and of course each step in the downstream processing. Achieving productivity at a scale that makes the business case stack up is the result of a complex dynamic system that is notoriously fickle. Being 1% less productive in output can mean losing 20% or more of your target.

For the last few decades precision fermentation has been used for high value target proteins such as vaccines and other pharmacological ingredients. With heme, Impossible Foods was fairly early to adopt the technique to create a food magic ingredient at scale.

More companies are adopting precision fermentation to produce high value food ingredients. One of the challenges is scaling up the process from the lab to hundreds of thousands of litres. On that front BioBrew have the largest ambitions, to create very large scale facilities, focussing on adaptive control systems that can break the current mould where plant desin and operation needs to be tightly coupled to its microbe, feedstock, media and target. If it comes to fruition, it will be a significant shift in the value chain to the industry, commodifying what has been custom to date11.

A new Motif

Fast forward a few years and enter Motif Foodworks. Inspired by Impossible Foods and well backed, for example including by Bill Gates, they too have developed heme in the form of bovine myoglobin.

It’s a different heme molecule, derived from a calf muscle rather than from a soy root. They also use a precision fermentation process, modifying a yeast to express the bovine myoglobin. They launched last year with the level of fanfare expected from a well funded biotech startup. The launch was closely followed in March this year with a patent infringement case from Impossible Foods.

Naively I initially assumed that because the heme source was a different molecule from a different source then Motif should be in the clear, following Motif’s line that the two proteins are “fundamentally different, and deliver different flavour and aroma benefits”. But looking at Impossible’s patent, they cover “a beef substitute that uses a muscle replica including a heme-containing protein…" (among other ingredients) and lay claim to a wide variety of those heme-containing proteins:

hemoglobin, myoglobin, leghemoglobin, non-symbiotic hemoglobin, chlorocruorin, erythrocruorin, neuroglobin, cytoglobin, protoglobin, truncated 2/2 globin, HbN, cyanoglobin, HbO, Glb3, and cytochromes, Hell’s gate globin I, bacterial hemoglobins, ciliate myoglobins, flavohemoglobins

The legal case is likely to keep meat mimetic producers from buying heme derivative proteins for some time. Motif will need deep pockets to see it through and in the meantime Impossible could enjoy a near-monopoly.

Another alternative?

So much for heme for the next while. In Part II we’ll look at a patent from an Australian company which covers another ingredient to help non-meat look and taste like meat.

  1. I had thought it was the iron in hemoglobin that made meat red; actually it’s the heme structure that gives blood its red colour↩︎

  2. A pyrrole is C4H4NH, itself in a ring shape. ↩︎

  3. I’m a little biased because I participated in the CSIRO Biomakers program. ↩︎

  4. My understanding is that “fermentation” implies using a yeast, and this page assumes as much. However, others describe a precision fermentation process using other microbes such as bacteria or algae. ↩︎

  5. For example, bacteria have limited protein folding abilities. Yeast and insect cells have different patterns of post translational modifications (e.g. glycosylation), so may not be able to express proteins that are functional. Mammal cell cultures and genetically modified animals are most able to produce protein targets for human use, and plants have a similar protein synthesis and modification pathways to mammals. But the name of the game is to find the most cost effective way to get your target protein, and yeast is pretty cheap and relatively easy to grow. ↩︎

  6. The major differences between traditional fermentation (beer, yoghurt) and precision fermentation are: 1) genetic modification of the microorganism; and 2) the downstream processing. Downstream processing generally consists of: separation (centrifuge to separate out the yeast from its growth media); disruption (break apart the yeast cells; required if the target is not expressed directly into the culture medium); extraction (purify using filtration of molecules both larger and smaller than the target); and stabilisation (evaporation/drying to help prevent degradation). ↩︎

  7. There are only a few breeds of yeast with known viable characteristics, and they tend to be locked up behind IP. On the genetic modification side, the sheer scale of a genome and any modification’s affect on the organism is against you; only a small part of the genetic space has been explored. The process of choosing a modification to experiment with usually consists of reading papers to see what others have tried, and the techniques for making modifications are also often IP protected. ↩︎

  8. Feedstock can vary in format (e.g. crystal vs liquid) and purity, and availability of formats varies at different scales. ↩︎

  9. Signals come in multiple forms, such as temperature, pH, and commonly, changes to the chemical composition of the growth media. The signal mechanism of course needs to match your (modified) yeast strain. ↩︎

  10. For example the pressure and temperature differences between the bottom and the top of a large, tall bioreactor vs a smaller, rounder one. Aeration, oxygen transfer rates, management of cooling are all affected. ↩︎

  11. I have something of an interest in this sort of industry-level value chain shift↩︎