H2020 HoloFood Project: A holistic solution to improve animal feed production by deconstructing biomolecular interactions between feed, gut microorganisms, and animals in relation to performance parameters

Ensuring that food production remains sustainable while meeting the needs of a growing population is a key challenge that must be overcome. Gut microorganisms play a critical role in animal health and well-being, and a balanced gut microbiota is essential for optimal feed production. Feed additives have been shown to be effective in modulating microbiomes in many systems, although their efficacy often varies. One likely reason underlying such inconsistency is the limited knowledge we have about their specific means of action. Ideally, feed additive companies could employ in vitro and in vivo experiments to test the additive's properties and benefits.

However, the informative capacity of these approaches is limited, as the in vivo response of microorganisms can be radically different from in vitro conditions, since the microorganisms of interest are found in different physicochemical conditions and could interact with hundreds of other microbial taxa, as well as with the host. Therefore, feed additive improvement often relies on trial-and-error approaches, which limits the development of a complete understanding of the reasons behind additive success or failure. Consequently, procedures for improving feed additive products are inefficient, and it is unlikely that truly optimal products can be found without drastically modifying the adopted approach. HoloFood overcomes these limitations by leveraging state-of-the-art computational and laboratory developments to provide direct insights into the effects of biomolecular interactions between feed additives, gut microorganisms, and hosts on food production. This was achieved by implementing a novel holomics framework, which exploits the decrease in the cost of generating omics datasets, along with cutting-edge biomolecular and statistical methods.

HoloFood is based on two observations:

1. Animals and their associated microorganisms behave differently depending on the environment and the animal's stage of development.
2. Many biological processes of animals and their associated microorganisms are affected, or even conditioned, by the other partner in the symbiosis.

This implies that the effect of any feed additive could differ depending on the animal's genomic background and developmental stage, microbiome composition, and production environment. Our framework analyzes the genomes, transcriptomes, and metabolomes of animals and their associated gut microorganisms, and then links these to animal health and physiology, as well as food production performance parameters. To showcase the potential of different food production systems and measure the commercial and societal impact of our approach, HoloFood conducted commercial-scale trials for poultry and salmon that analyzed biomolecular interactions between microorganisms and animals fed different feed additives in relation to food production parameters related to food quantity, quality, safety, as well as food production sustainability and welfare.

By understanding the physiological effects of biomolecular interactions that occurred when administering different feed additives, HoloFood optimized its use by adapting its delivery to the genomic context and gut microbiome of animals, as well as the agricultural environment. As such, the goal was to increase the efficiency of feed additives and, therefore, improve final food products. In the future, we believe that other food production systems could be optimized through these holomic approaches. In this way, HoloFood will act as a paradigm for approaches and, in doing so, will foster the creation of jobs related to scientific research and the food sector, both for producing companies and associated industries. Furthermore, the open resources and data availability resulting from the HoloFood project will be of value to research and industry stakeholders beyond the project's timeframe.