Full Paper can be found at https://www.nature.com/articles/s41564-018-0258-8
The human gut harbors a vast community of microbes, known as the human gut microbiota, with the ability to deconstruct and utilize a wide range of polysaccharides found in the human diet. The human gut microbiota collectively encode and express an incredible variety of carbohydrate active enzymes which are absent from the human genome. The human host and gut microbes form a symbiotic relationship in which the host benefits from microbial metabolites and the microbes are supplied with a plentiful nutrient source in the form of dietary polysaccharides. Foods that contain arabinogalctan (AG) include many soft drinks and confectionery as Gum Arabic AG, as well as vegetables which contain larch-type AGs.
Initial bioinformatic and transcriptomic data revealed two substantial polysaccharide utilization loci directed against AG (Figure 1) which are present in the genome of Bacteroides thetaiotaomicron (B. theta), a Gram-negative species found in the human gut microbiota. However, despite possessing every activity required for utilization of AG, B. theta is unable to utilize the polysaccharide without initial degradation with an endo-arabinogalactanase. Present in the large AG-PUL was an arabinogalactanase which could fulfill the surface oligosaccharide generating role however, it was found in the periplasm. In the degradation of AG by Bacteroides location is everything. It is not enough to produce every enzyme required for glycan utilization, they need to encounter the target polysaccharide in the correct order and in the correct location. This demonstrates the challenge of Gram-negative bacterial nutrient utilization and the requirement of partial deconstruction of the polysaccharide for transport to occur. What modifications must the substrate undergo to allow uptake? In this case the target polysaccharide must be hydrolyzed into oligosaccharides, which creates a new problem. How do you stop oligosaccharide loss at the cell surface? Is it even possible or worth doing so in the complex and competitive environment of the human gut? Bacteroides caccae and Bacteroides cellulosilyticus (B. cell) were found to possess surface endo-activity allowing both of these species to produce oligosaccharides from AG at the cell surface. Of all Bacteroides tested in our study just these two species could utilize undigested AG polysaccharide. Rather than utilizing all AG oligosaccharides produced at the cell surface these species both release oligosaccharides which are capable of supporting AG oligosaccharide species, like B. theta and Bacteroides ovatus. We showed these species do not just produce oligosaccharide for themselves, they also support other Bacteroides species. This is potentially a co-evolved relationship due the relative prevalence of oligosaccharide utilizing Bacteroides, with a few keystone species providing the oligosaccharide required for growth (Figure 2). Interestingly, when performing cross-feeding assays we found preferences showed by the recipients used in the assay. B. ovatus showed preference for oligosaccharides produced by B. caccae whereas B. theta preferred oligosaccharide produced by B. cell. Together, this suggests specific niches within the AG cross-feeding network of Bacteroides species in the gut.
We then created an integration mutant of B. theta that presents the arabinogalactanase from B. cell at the cell surface, called B. theta::Baccell_00844 to test the hypothesis that keystone status is dependent on presence of an endo-arabinogalactanase presented on the cell surface of the bacterium. Indeed, this engineered B. theta could utilize AG itself while also supporting growth of oligosaccharide utilizing B. ovatus, in co-culture with AG as the sole carbon source. This demonstrates that keystone status in Bacteroides polysaccharide utilization is conferred by the presence of an endo-acting enzyme at the cell surface.
Our study sheds light on the utilization mechanisms employed by Bacteroides to degrade and utilize AG, a complex polysaccharide present in the human diet. We also demonstrated interactions between different Bacteroides species during AG utilization highlighting potential nutrient cross-feeding networks between the individual species.
Figure 1 from Cartmell, Munoz-Munoz, Briggs and Ndeh et al. (2018) Nat. Microbiol.
Figure 2 by Jon Briggs
Full Paper can be found at: https://www.nature.com/articles/s41564-018-0258-8