Bacteria for dinner

For millennia, protein has come from animals or plants. But a third source may soon become available – proteins from bacteria. XiaoZhi Lim reports

Food production faces twin crises. To feed a population forecast to balloon to 9.7bn by 2050, food production must increase by 25-70% from current levels, according to a 2017 estimate (BioScience, doi: 10.1093/biosci/bix010).

But this expansion cannot be accompanied without a rise in greenhouse gas emissions. If food production were scaled up according to business-as-usual dietary trends, the associated emissions could bust the greenhouse gas budget for keeping the global temperature rise under 1.5°C, even if all non-food greenhouse gas emissions were halted immediately (Science, doi: 10.1126/science.aba7357).

Currently, food production contributes about 30% of global emissions, with the biggest contributor coming from protein. Traditional meat and seafood production methods are inefficient and emit powerful greenhouse gases like methane and nitrous oxide. Yet, protein is necessary for growth and more is needed in regions like East and Southern Africa, where protein inadequacy is strongly correlated to stunting in children (British Journal of Nutrition, doi: 10.1017/S0007114512002577). ‘The intensive farming of animals is viewed no longer as an option to plug that gap,’ says Alan Shaw, CEO of California, US, biotech Calysta.

Plant-based protein is slowly gaining ground. With a huge market need and a growing movement towards animal alternatives, a third, even more efficient source of protein could help: microbial cells.

Eating the cellular biomass of microbes means producing only the cells that are consumed – instead of growing an entire plant like soya to harvest just the beans, and worse, using the beans to raise animals just for their meat. ‘We are growing only those parts that have nutritional value and that brings a fundamentally new, higher level of efficiency to food production,’ says Pasi Vainikka, founder of Finnish food technology start-up company Solar Foods.

Technology has significantly advanced for fermenting microorganisms to build biomass. Companies are also banking on access to available feedstocks like natural gas, renewable energy and even waste gases. Selecting microbes that feed on greenhouse gases, for example, should lead to a carbon-negative protein production process that effectively transforms renewable power into calories. ‘We’re absorbing carbon from industrial sources, essentially removing CO2 that would end up in the atmosphere and converting that into [protein],’ says Brian Sefton, President of biotech NovoNutrients based in California, US.

The biggest advantage is that microbial protein does not need arable land, which is now often created by deforestation, or resources that conflict with existing food production. ‘We don’t take anything out of the human food chain,’ Shaw says. ‘It’s a complete new source of protein and therefore we see huge potential as the world tries to close this protein gap.’

Bacterial advantage

Microbial food isn’t new – indeed, microorganisms are already widespread in our food. We eat live bacteria in yoghurt and yeast in beer, bread, and cheese. Microalgae like spirulina and chlorella are already available on the market as nutritional supplements. Quorn, a mycoprotein meat substitute, is sold in 15 countries as an alternative to meat.

However, Shaw thinks the future of very large-scale production of microbial cells to serve as protein belongs to bacteria. ‘Bacteria can take a carbon source which is outside of the human food chain,’ he says. Specifically, C1 compounds that contain just one carbon atom like methane, carbon monoxide and carbon dioxide. In contrast, yeast, algae or fungi typically need a carbohydrate or simple sugars. Another advantage is some bacteria can grow extremely quickly – E. coli, for instance doubles its biomass every 20 to 30 minutes, several orders of magnitude faster than soya or pea crops.

The first commercial example of bacterial protein dates back to the 1970s, from UK company ICI (Microbial Biotech., doi: 10.1111/1751-7915.12369). ICI used the methanol-oxidising bacteria Methylophilus methylotrophus to create a flour called Pruteen. But despite success in feeding trials with livestock, ICI’s venture eventually wound down by the 1990s with increasing pressure from cheap fishmeal and soya feed, as well as rising oil and gas prices.

We’re absorbing carbon from industrial sources, essentially removing CO2 that would end up in the atmosphere and converting that into [protein].
Brian Sefton President, California-based biotech NovoNutrients

Calysta is trying to mass produce a methane-consuming microorganism, Methylococcus capsulatus, discovered in the Roman Baths in Bath, UK. The bacterium uses methane for both carbon to build its cells and for energy. Methane is currently advantageous because of existing natural gas infrastructure, low prices, and the fact natural gas already comprises 95% methane, says Shaw. ‘We just literally take the gas and we put it straight into the fermenter and we don’t need to use an additional source of energy, which someone using CO2 has to do.’ Danish biotech, Unibio, is also using M. capsulatus for its protein product.

A problem that others see with methane-consuming bacteria however, is ‘where are you going to get your methane from?’ says Nico Boon, a microbiologist at Ghent University in Belgium. While biogas could eventually be a sustainable source of methane, autotrophic microbes that live on carbon dioxide and hydrogen could be more sustainable in the long term, because all they need is a solar panel and an electrolysis unit to produce hydrogen from water. Once the infrastructure is in place, the microbes can grow with just water and air. ‘I think it’s a bit simpler,’ Boon says.

Solar Foods is growing one such autotrophic bacterium found in Finnish soil. NovoNutrients is also targeting similar bacteria that oxidise hydrogen for energy and use that energy to capture carbon dioxide for building their cells. But instead of selecting just one microbe, NovoNutrients is taking a consortium approach. The company has identified a suitable community of microbes it has been testing in feeding trials and is still seeking others. ‘Because we can change the member microbes, we are developing new and better consortia all the time,’ Sefton says.

Scaling up fast

Growing bacteria at scale is challenging. Traditionally, biofermentation takes place in a batch fashion, making it inefficient. ‘Even in the food industry today, there are very few, if any, examples of people who have successfully industrialised the biological fermentation of microbes to get high concentrations of biomass,’ says Shaw.

In particular, how well the microbe can consume and incorporate its substrate into its cells is key. Anything less than 90% would lead to a build-up of substrate, and manufacturers would have to stop and clean out the bioreactor, explains Shaw. ‘If it’s an inefficient process, it’s doomed to be batch.’

For Calysta, while the microbe is naturally occurring, its production process is novel and patented. Calysta acquired the associated IP from Norwegian oil company Statoil in 2016, which took Statoil engineers some 20 years and $400m to optimise. In one pass, the microbes can convert 99% of methane to its biomass, says Shaw. Then the cells are simply dried to yield the final protein product, FeedKind. Calysta has operated a pilot plant in Teesside, UK, for the last three years, producing 50t/year of Feedkind protein for feeding trials in powder or pellet form. The company will soon be breaking ground on a plant in China. That plant will operate in a continuous flow fashion and produce 20,000t/year of product, Shaw says. ‘That will set the standard for everybody else in the world.’

Shaw believes those growing autotrophs on carbon dioxide and hydrogen will face major obstacles in scaling up because carbon dioxide in the air is very dilute. ‘You can’t just take air and feed it into a fermenter,’ he says. Meanwhile, because carbon dioxide is an energy-deficient molecule, the microbes will need a second energy-rich chemical feedstock, typically hydrogen. Taken together, autotroph growers may require an additional carbon capture facility, an electrolysis operation to produce hydrogen from water, and sufficient renewable energy infrastructure. ‘If your feedstocks are hard to get, or if you have to invest in additional technology steps, your economics literally go out the window,’ Shaw says.

Vainikka acknowledges that investments are needed to build up renewable energy capacity, but counters that solar, wind or geothermal, depending on location, are rapidly becoming the cheapest sources of electricity, if not already. ‘We are taking advantage of that development,’ he says. The company has also conducted pre-engineering for a full-scale factory, including calculating the investments needed, Vainikka says. ‘It is in fact, a very attractive business case.’

Solar Foods’ ambitious plan is to feed carbon dioxide directly captured from air and hydrogen produced sustainably through electrolysis of water to its bioreactor. The harvested microbes are simply dried to become a flour product, Solein. The start-up’s pilot is currently producing 0.5kg/day of Solein and has raised $15m in funding to invest in a plant about 100 times larger somewhere in Finland. Construction will begin soon, and Vainikka expects the plant will be operating in about 30 months’ time.

NovoNutrients is counting on a key advantage: its consortium of microbes can handle untreated raw industrial flue gases. ‘The fact we don’t have to clean the flue gas is a big economic enabler,’ Sefton says. Microbes can be finicky about what they come into contact with – methanotrophs, for instance, are notoriously sensitive to carbon dioxide, while nitrogen-fixing bacteria need low oxygen levels. NovoNutrients’ consortium can tolerate flue gases from many sources, including cement plants, oil refineries, and power plants. The microbes can consume pollutants like hydrocarbons, oxides of sulphur, oxides of nitrogen, benzene or cyanide. ‘We’ve done this many times,’ Sefton says. ‘Right now in our laboratory, we have tanks of flue gas that were taken directly from an oil production facility and they’re uncleaned, unfiltered, and our microbes were able to capture over 98% of the CO2 in that gas in one pass in the bioreactor.’ The only pollutants to avoid are heavy metals like elemental mercury or arsenic, he adds.

NovoNutrients currently operates a laboratory facility that produces samples in kg quantities to support feeding trials. The company has designed a pilot plant that will produce about 12t/year of product, supporting very large feeding trials with multiple partners, Sefton says.

As with conventional farming, the microbes will need a source of nitrogen, an essential element in amino acids and proteins. Some specialised microbes could fix their own nitrogen from air (see C&I, 2021, 85(1), 9), but those being used so far by microbial protein companies require a ‘reactive’ form of nitrogen like ammonium or nitrate supplied to them. Solar Foods plans to use ammonia in the growth media, while NovoNutrients plans to include a proprietary nitrogen-fixing process to generate ammonia in situ.

Market potential

For now, the best opportunities for bacterial protein are animal feed or pet food markets. ‘At the beginning, we can probably use this protein as a replacement for soya for feed,’ says Boon.

Calysta and NovoNutrients are both targeting fish feed markets. Aquaculture, or fish farming, provides about half the global supply of fish and accounts for almost three-quarters of global fishmeal consumption. In turn, fishmeal is produced from small fish caught in the wild, and stocks of those little fishes are rapidly depleting. ‘Year to year, there’s a lot of volatility in the price,’ says Sefton, which depends on the size of the catch. Fishmeal prices have been growing steadily however, from a low of $500/t in the 1990s to $1500-$2000/t in recent years.

While soya feed helps to take some pressure off fishmeal and wild fish stocks, soya can only replace fishmeal to a limited extent because fish are obligate carnivores. For this reason, Shaw believes bacterial protein, with its low levels of complex carbohydrates, is a superior feed for fish.

Solar Foods hopes to make inroads into human food markets right away, by going for a neutral-tasting flour that the food industry can process into food items. The company has created meatballs, cheese, and chicken-like products from Solein flour, as well as bakery items, breakfast granola, and even sweets. ‘It’s quite a broad range,’ says Vainikka, spanning various items that can be consumed at every meal of the day. ‘Then, basically along the day, you get the nutrition of meat without ever going close to meat.’