Development of anaerobic high cell density cultivation for sustainable single-cell protein production using denitrification

Abstract

Future food production faces the immense challenge of feeding a growing population with an increasing demand for resource-intensive food in a sustainable way. The high environmental footprint of conventional food production opens the floor to alternative sources of food and feed, with single-cell protein (SCP) emerging as a promising solution. SCP offers a high protein content (up to 80%), favorable amino acid profiles, and the presence of carbohydrates, vitamins, fats, and antioxidants increasing the nutritional value. The production is independent of seasonal and climate variations, has low requirements for arable land and resources, and has low greenhouse gas emissions.

SCP is generally produced by high cell density cultivations (HCDCs) to maximize the volumetric yield. However, conventional HCDC relies on aerobic respiration, which is limited by the low solubility and diffusion rate of oxygen in liquid. Oxygen limitations may result in anoxic zones accompanied by fermentation and accumulation of unwanted byproducts inhibiting growth. Denitrification, the stepwise reduction of nitrate (NO3-) to nitrogen (N2) via nitrite (NO2-), nitric oxide (NO), and nitrous oxide (N2O), is the most efficient form of anaerobic respiration and overcomes the rate-limitations inherent in aerobic respiration due to the high solubility of NO3- and NO2-. Previous attempts at HCDC with denitrification most likely failed due to denitrification-driven increases in pH necessitating tight pH control and the provision of NO3- as a salt resulting in salt accumulation.

This work presents a novel method for anaerobic HCDC that overcomes these challenges by using nitric acid (HNO3) as both the pH regulator and the provider of NO3- for denitrification. The bioreactor is operated as a pH-stat where denitrification-driven increases in pH trigger HNO3 injections, which in turn control the provision of the carbon source from a separate feed pump. The study was initiated by proof-of-concept experiments that demonstrated the viability of the method by growing the model denitrifier Paracoccus denitrificans Pd1222 to a density of more than 20 g dry weight L-1 (Paper I). We were able to reach the expected yield based on the provided substrate, but the growth rate was much lower than what we observed in low-density batch cultures. In addition to the low growth rate, we observed 1) the accumulation of polyhydroxyalkanoates (PHA), a storage and energy reserve typically encountered during carbon excess, and 2) a low expression of NirS, the enzyme responsible for the reduction of NO2- to NO.

These challenges were addressed in subsequent experiments where we aimed to develop the method further (Paper II). We switched to Paracoccus pantotrophus GB17, a closely related strain with a higher NirS pool, that was also shown to adapt better to the bioreactor environment. The compositions of the feed solutions were refined to prevent accumulation of byproducts, including PHA, and improve the growth conditions. We incorporated ammonium nitrate (NH4NO3) as the nitrogen source to remove the cost of NO3- assimilation and designed new acid solutions with minerals and trace elements. We hypothesized that the slow growth rate was due to the high concentration of carbon dioxide (CO2) in the liquid lowering the pH, thereby interfering with the provision of the substrates which is triggered by alkalization. This was investigated through in silico simulations of the NO3-, CO2, and pH kinetics in the bioreactor and rapid sampling of the bioreactor. The modeling and empirical data showed that the NO3-, which was injected at intervals of several minutes, was depleted within seconds. We aimed at improving the CO2 removal by dynamically changing the sparging flow rate and pH setpoint in response to changing partial pressure of CO2 in the off-gas and experimenting with lower pH limits. Although the process is still limited by a slow growth rate, we were able to reach densities of more than 60 g dry weight L-1 using P. pantotrophus. The biomass was analyzed and shown to have an average protein content of 75%.

While anaerobic high cell density cultivation is proven feasible, one major challenge for the upscaling is the high cost of HNO3. We have therefore validated HNO3 produced by N2 Applied using plasma technology, which is expected to become a cheaper and more sustainable alternative in the future (Paper III). Biomass produced using our process was also tested by the Center for Feed Technology (Fôrtek, Norwegian University of Life Sciences, Norway) in successful feed trials with Atlantic salmon. This demonstrates that SCP produced by our process can be used as a protein source in aquaculture feed.

The work presented in this thesis bridges fundamental research on microbial physiology with applied science, demonstrating the proof-of-concept, development, and commercial potential of a novel method for anaerobic high cell density cultivation. Although only tested for heterotrophic growth on glucose, the process has the potential to be adapted to a wide range of strains, substrates, and metabolic strategies. This work paves the way for anaerobic high cell density cultivation as a supplement to existing methods, advancing the field of sustainable food production.

Fremtidens matproduksjon står overfor den enorme utfordringen å skulle fø en voksende befolkning med en økende etterspørsel etter ressurskrevende mat på en bærekraftig måte. Det høye miljøavtrykket til matproduksjon åpner for alternative kilder til mat og fôr, hvor mikrobielt protein (SCP) fremstår som et godt alternativ. SCP har et høyt proteininnhold (opptil 80%), gunstig aminosyreprofil, og karbohydrater, vitaminer, fett og antioksidanter som øker den ernæringsmessige verdien. Produksjonen er uavhengig av sesong- og klimavariasjoner og har lave krav til dyrkbar mark og ressurser, samt lavt klimagassutslipp.