Dynamic modelling of monogastric gut microbiota
Item statusRestricted Access
Embargo end date31/07/2022
The porcine gut microbiota is composed of a considerable number of microbial species with several levels of interaction with its host, including the production of energy from sources, otherwise not digested by the animal, or the production of metabolites, such as the short chain fatty acids (SCFAs). Moreover, a subset of these species, the porcine gut core microbiota (CM), which seems not to be affected by changes in host diet or breed, has been identified by several authors. Manipulations of the whole microbiota (WM), and as an extension of the CM could achieve the production of active molecules (e.g. anti-inflammatory), and a metabolite pattern advantageous for the host health and performance. Nevertheless, the methods to explore long term perturbations are either ethically/technically demanding or expensive. Thus, mathematical modelling could provide the tools for the in silico experimentation able to predict changes in the metabolites due to intervention strategies. During this project an innovative mathematical model depicting the WM dynamics was developed and validated in vitro, in parallel with the development of an innovative model describing the CM dynamics. A continuous fermentation experiment mimicking the porcine proximal colon was conducted to validate the WM model, which predicted well the dynamics of the major bacterial and archaeal communities and of the major SCFAs (i.e. acetate, propionate and butyrate) observed in vitro. De facto their modelled ratio was 47.98%, 28.06% and 23.96%, whereas analogous values observed during the experiment was 44.5%, 30.5% and 25.1%. The CM model showed a possible role of the core sub-community in forming the basis of the SCFA production, and a probable role of the core in providing acetate to the WM when compared to literature porcine microbiota studies. Both the models were then the object of in silico experimentation to simulate the effects of probiotic/prebiotic therapies on the metabolite pattern, anticipating SCFA changes upon these interventions. As a result, multiple scenarios were derived from different types of intervention, for instance, a combined therapy of non-starch polysaccharides (as prebiotic) and two CM strains (as probiotic) could increase colonic butyrate concentration, suggesting a potential positive effect on the host well-being. In conclusion, the models developed during this project produce accurate estimates of the major SCFAs and bacterial communities, thus providing a valuable tool to those interested in testing the microbiota response to different starting conditions, which could serve as a step preceding the necessary in vitro/in vivo investigations.