Identification of NO-sensing protein domains that regulate bacterial pathogenesis and biofilm formation in Pseudomonas aeruginosa.

Abstract

Most species of bacteria preferentially grow in sessile communities, known as biofilms, rather than as free-living planktonic cells. Biofilms can be up to 1000 times more tolerant to antimicrobials compared to their planktonic counterparts, making them a growing problem within the medical and industrial settings. Biofilms formed by Pseudomonas aeruginosa are involved in chronic infections, such as those affecting the lungs of Cystic Fibrosis patients. High or low intracellular levels of a bacterial secondary messenger, bis-(3’-5’)-cyclic dimeric guanosine monophosphate (c-di-GMP), regulates biofilm formation and dispersal respectively. The synthesis of c-di-GMP is carried out by diguanylate cyclases (DGC) due to catalytic domains known as GGDEF domains. Whereas c-di-GMP degradation is carried out by phosphodiesterases (PDE) due to catalytic domains known as EAL domains. Low (non-toxic) concentrations of nitric oxide are known to induce a biofilm dispersal through a reduction in the c-di-GMP levels and an increase in the PDE activity. However the nitric oxide sensor and the protein responsible for the reduction in c-di-GMP levels is unknown.

Previously found to be involved in an NO-induced biofilm dispersal, the bi-functional enzyme RbdA (regulation of biofilm disposal) contains a GGDEF and an EAL domain in tandem. Using an enzymatic assay measuring PDE activity, we investigate the relationship between the tandem GGDEF and EAL domains. We find that the isolated EAL domain of RbdA has a higher PDE activity, suggesting that the tandem GGDEF domain negatively influences the activity of the EAL domain. We attempt to further investigate this at the molecular level using X-ray crystallography and structure determination. The structure of the EAL domain of RbdA wasdetermined and appears to be in a primed state for substrate binding, with a single Mg2+ ion bound within the active site. After comparisons to other EAL domain structures, we suggest a schematic for substrate binding to EAL domains.
We investigate an RbdA homologue, PA2072, as previous biological data indicates opposing physiological roles. By comparing the primary and secondary structures of RbdA and PA2072 we suggest that their physiological differences are caused by disparities between their periplasmic regions and / or their putative sensory PAS domains. Protein crystallisation of the PA2072 periplasmic region (a putative CHASE4 domain) and the PA2072 PAS domain were attempted but require further optimisation.
The first and second PAS domain of PA0285 were predicted to bind a haem-b and FAD (flavin adenine dinucleotide) cofactor respectively. We hypothesise that NO can be sensed by a haem-bound PAS domain. Using ultraviolet-visible spectroscopy we could only identify a very weak haem-b cofactor binding to the PAS1 domain of PA0285 and so requires further investigation. However, we identified a PA0285 PAS2 : FAD binding stoichiometry of approximately 2 : 1. Here we put forward models suggesting that, NO-induced changes to the redox potential are sensed by the FAD bound PAS2 domain, leading to changes in the enzymatic output of PA0285 and potentially biofilm dispersal. Restoring the sensitivity of bacterial cells to antimicrobials by inducing a biofilm dispersal, is thought to be a novel treatment strategy. This work lays some of the foundations required to understand the molecular mechanisms that lead to a biofilm dispersal in P. aeruginosa.

More information

Identifiers

ORCID for Jack William Craddock: orcid.org/0000-0001-9193-0366

ORCID for Jeremy Webb: orcid.org/0000-0003-2068-8589

Catalogue record

Export record