Companies typically approach contract research companies with the aim to better understand single organisms, groups of organisms, metabolic pathways and the interaction with the host that make up the microbiome. In addition, companies search for compounds or modulators that could improve in the end the healthy state of the host. Such modulators could be prebiotics, probiotics, enzyme inhibitors, antimicrobials and more.
NIZO’s unique business model centres around performing industrially relevant research projects for the industry under strict confidentiality and with clear IP clauses.
The first approach, connect, focuses on identifying structure-function relationships of the microbiome. The microbiome experts at NIZO do this by finding linkages between genes, organisms, metabolites and host factors based on literature and expert data. From this overview we build a database or microbiome map that provides a structured overview of the relationship between the microbiome and health issue. Examples of maps we have developed in the past are ‘skin ageing’ and ‘atopic dermatitis’ (2).
The second approach, predict, makes use of algorithms to predict biological functions based on these microbiome maps, yielding a list of genes and organisms that may be targeted for modulation. Findings can then be corroborated by high-resolution genomic sequencing methods and metabolomics performed on clinical samples.
This analysis can improve the quality of modulators found, which means that less compounds are required in in vitro validation studies. Next to defining high quality modulators, the mode of action of a designed modulator can be described on the molecular level, providing a unique IP position. In the slide deck, an example is given of a client case in which active modulators were discovered to battle vaginal dryness (3).
Validate involves deploying and developing models that closely mimic real-life situations by using DNA material from donors or designing a mock community that resembles a microbiome based on expert knowledge. NIZO’s model MicroColon is a high throughput screening tool that uses faecal material as an inoculant and that can be spiked with pathogens including C. difficile (4). Outputs that are measured include:
Based on experience with infant faecal samples, it indicates that this model is suitable to measure the performance of prebiotics for stimulation of bifidobacterial, which, in normal screenings, do not give distinct differences in performance.
Ultimately, validation using bioinformatics tools, advanced statistics and biological interpretation in human trials is required to link shifts in the microbiome to relevant clinical outcomes as mentioned in our joint studies with ETH Zurich, which overcame the challenges of iron fortification in Kenyan infants (5).
The final approach, scale-up, is often a challenge. Companies select and validate their microbiome modulators but need to create a robust process at industrial scale. Whereas we focus on optimising processes on a lab scale by taking the requirements that need to be met at an industrial scale from the first experiment into account, thereby improving the chance of success of a full-scale trial. We have specifically developed expertise on strict anaerobes (non-GMP): upscaling, downstream processing and designing a suitable formulation for clinical trials. An example of such a project is the development of an oral formulation of Eubacterium halii, as performed for Caelus Health.
To conclude, using the approach described above, companies can create data-driven, expert-based modulation technologies for microbiome niches (6).