This thesis presents two stories related to metabolite uptake by bacteria. The first focuses on sialic acid uptake, while the second focuses on organosulfonate uptake.

Sialic acids comprise a large family of structurally related α-keto acid sugars, commonly found occupying the terminal position of mammalian glycoconjugates that cover mucosal cell surfaces. The uptake and metabolism of sialic acids by pathogenic bacteria is linked to infection and disease in humans; hence, inhibition strategies targeting sialic acid metabolism present a novel avenue for antimicrobial development. Here I focus on the mechanisms of sialic acid uptake by bacteria.

Escherichia coli uses two closely related major facilitator superfamily transporters to import distinct forms of sialic acid, with the ability to transport more than one form thought to confer a competitive advantage. Here I investigate, for the first time in vitro, how EcNanX facilitates the import of 2,7-anhydro-N-acetylneuraminate (2,7-an-Neu5Ac), an anhydro form of sialic acid. An EcNanX green fluorescent protein fusion construct was used to monitor expression and assess detergent stability, however, this construct was unstable and unsuitable for characterisation. The final EcNanX construct, which included a C-terminal octa-histidine tag, readily formed dimers and was screened for structural determination using single-particle cryo-electron microscopy. A small dataset yielded promising 2D classes and a low-resolution reconstruction validating the dimerisation properties of EcNanX. A structural model of EcNanX was generated and used to dock the 2,7-an-Neu5Ac substrate, allowing residues specific to 2,7-an-Neu5Ac transport, as opposed to the most common sialic acid N-acetylneuraminate (Neu5Ac), to be identified. Through mutagenesis of the substrate-binding site, EcNanX was successfully engineered to import Neu5Ac via an in vivo bacterial growth assay. These findings lay the groundwork for further structural and biophysical characterisation of EcNanX and enhance our understanding of alternative pathways for sialic acid metabolism, which may be leveraged for the development of novel antimicrobials in future.

Sulfate-reducing bacteria, such as Desulfovibrio species, are implicated in human disease due to hydrogen sulfide production and have bioremediation potential for industrial applications. One key oxidised sulfur source is the organosulfonate isethionate, which is imported by the sulfate-reducing bacterium Oleidesulfovibrio alaskensis using a tripartite ATP-independent periplasmic (TRAP) transporter (OaIsePQM). This system relies on the substrate-binding protein (OaIseP) to scavenge isethionate and deliver it to the membrane transporter component (OaIseQM) for import into the cell.

OaIseP was studied to define the mechanism of isethionate binding, the first step in the TRAP transport cycle. The binding affinity of isethionate to OaIseP was determined by isothermal titration calorimetry and is comparable to other TRAP substrate-binding proteins. X-ray crystal structures of OaIseP in the ligand-free and isethionate-bound forms were obtained and showed that in the presence of isethionate, OaIseP adopts a closed conformation whereby two domains of the protein fold over the substrate. This conformation of OaIseP is stabilised by both interdomain contacts and contacts to isethionate. Serendipitously, two crystal forms with sulfonate-containing buffers (HEPES and MES) bound in the isethionate-binding site were discovered. However, these do not evoke domain closure, presumably because of the larger ligand size. Together, the data helps unravel the molecular details of how a TRAP substrate-binding protein binds a sulfonate-containing substrate, rather than a typical carboxylate-containing substrate.

The structure of isethionate-bound OaIseQM in complex with a megabody was determined to 2.98 Å (extending to 2.2 Å in some regions) via single-particle cryo-electron microscopy. As sialic acid TRAP transporters are the only members of the family that have been structurally characterised, the structure of OaIseQM contained novel features including a unique orientation of the fusion helix that links the Q- and M-subunits through the membrane, and only two Na+-binding sites (instead of three in other homologues). The OaIseQM structure allowed the isethionate-specific substrate-binding site to be defined, as well as three bound lipids to be resolved, consistent with those found in the sialic acid TRAP transporters. OaIseQM was shown to be functional using a proteoliposome transport assay, while models of the complete OaIsePQM complex were generated to shed light on the conformational changes undergone during the full TRAP transport cycle. The OaIsePQM models highlight a potential allosteric ‘scoop loop’ from the M-subunit to prompt isethionate release by OaIseP. This work expands the knowledge of TRAP transport, providing insight into the mechanisms underpinning transport which may inform future antibiotic development.