The assembly of the human gut microbiota (HGM) during infancy has a profound effect on the host health and perturbations of the early life HGM maturation are associated with life-long health effects. Mother’s milk, especially HMOs are key in the selection of the Bifidobacteria-dominated neonate HGM, which is potentiated by the efficient HMO utilization machinery of infant Bifidobacteria¹. With the introduction of solid food and simultaneous reduction of breastfeeding, fundamental changes are induced to HGM. Notably, HMO-specialist Bifidobacteria are replaced by Firmicutes as the top abundant phylum during a highly competitive maturation towards an adult HGM structure. This critical window for the assembly of a healthy and resilient adult HGM offers a unique opportunity for therapeutic interventions to overcome aberrant HGM states in early life that may have life-long negative health outcomes for the host. The question if the roles of HMOs extends to the critical transition of the HGM during weaning remains elusive.

How is this connected to our work?

Back in 2016, I started my PhD in the lab of Prof. Maher Abou Hachem (DTU, Denmark) who at this time focused on butyrate-producing Clostridiales, as a core abundant HGM group. This Firmicutes group has attracted attention due to their ability to ferment dietary fibers into butyrate – a short chain fatty acid that acts a prime energy source for colonocytes and which exhibits anti-carcinogenic and anti-inflammatory properties. In previous studies^2,3 Prof. Maher Abou Hachem and my colleague Maria Louise Leth explored how these bacteria degrade the two key dietary fibres xylan and β-mannan.

When I started my work, we were very intrigued by hints in the literature reporting the presence of Clostridium cluster XIVa members in the infant gut before full transition to solid food. Our previous work that demonstrated the competitiveness of Clostridium cluster XIVa in the utilization of dietary fibres due to their efficient glycan capture and oligosaccharide uptake systems, did not really explain the early presence in infants during the weaning period. We thought, could it be HMOs that promote their early colonization? Are these bacteria able to utilize HMOs with an unknown catabolic pathway? We selected members of Roseburia-Eubacterium group as model systems and started our quest for the missing HMO-utilization link! We identified possible gene-loci with a light HMO taste, based on the presence of a lacto-N-biose phosphorylase (GH112), and a good few hypothetical proteins. We started expressing these “hypothetical” proteins, but although we screened the obtained products over a great variety of potential substrates, we failed to detect any activity and started to worry that these proteins might remain truly “hypothetical” for us.  Suddenly however, we got very excited when Professor Shinya Fushinobu and Assistant Professor Chihaya Yamada from Tokyo University in collaboration with Professor Takane Katayma (Kyoto University, Japan) published a paper on a new exotic family of bifidobacterial HMO active enzymes (GH136 lacto-N-biosidases), which were distant homologues to our “hypothetical proteins”. The Japanese researcher could show that this enzyme family requires the expression of a “chaperone” protein to obtain an active enzyme - which was the reason why we could not measure any activity. We established a collaboration with the two Japanese research groups and together resumed our hunt for the missing HMO-utilization link.

The following three years of HMOs isolation, growth experiments, co-culture/cross-feeding studies, proteomic analyses, bioinformatics analyses, biochemical as well as structural characterization of the involved enzymes and transport system, we had the first complete picture of a previously unknown but highly prevalent gene loci that mediate the growth of Roseburia-Eubacterium members on distinct HMOs. Remarkably, we demonstrate that different Roseburia spp. specialize on different HMOs, which are not preferred by Bifidobacteria and which may allow them to coexist with the declining Bifidobacteria HMOs-specialists during weaning.

At least equally exciting, we show that the HMO-machinery and other loci are deployed to cross-fed on mucin O-glycans together with the model mucin degrader Akkermanisa muciniphila.

So where does this leave us? Our study reveals the mechanistic details by which a physiologically important core group from the early life HGM metabolizes HMOs and mucin O-glycans. These findings uncover a previously unknown role of HMOs in the promotion of Clostridiales and provides the basis for a better understanding of the crucial maturation process of the early life HGM.

The full paper can be found at https://www.nature.com/articles/s41467-020-17075-x

1. Sakanaka, M. et al. Evolutionary adaptation in fucosyllactose uptake systems supports bifidobacteria-infant symbiosis. Sci. Adv. 5, eaaw7696 (2019).
2. Leth, M. L. et al. Differential bacterial capture and transport preferences facilitate co-growth on dietary xylan in the human gut. Nat. Microbiol. 3, 570–580 (2018).
3. La Rosa, S. L. et al. The human gut Firmicute Roseburia intestinalis is a primary degrader of dietary β-mannans. Nat. Commun. 10, 905 (2019).