The "Mystery Phases": Bordetella Adenylate Cyclase Toxin and biofilm in the Bvg-Intermediate Phase & Bordetella pertussis Motility and Flagella in the Bvg-Minus Phase

Bordetella pertussis is the Gram-negative bacterial pathogen responsible for the life-threatening disease, whooping cough (pertussis). The bacterium has been of increasing interest in the research community due to the reemergence of this vaccine-preventable disease. In fact, the numbers of pertussis cases reported to the CDC annually are similar to those reported prior to the introduction of the whole cell pertussis vaccine. The increased number of cases coincides with the transition to the acellular vaccine, comprised of just one to five B. pertussis antigens (pertussis toxin, filamentous hemagglutinin, pertactin, fimbriae 2 and fimbriae 3). This correlation, while not the only contributing factor to reemergence of the disease, highlights the fact that not enough was known about the biology of B. pertussis before the acellular vaccine was created. In fact, it has only recently been shown that B. pertussis forms biofilm during infection, and that biofilm may be a virulence trait of Bordetella. These biofilms form both in vitro on abiotic surfaces and in vivo in the mouse trachea and nasopharynx. Biofilms are defined as surface-associated bacterial growth, in which the bacteria are covered in extracellular polymeric substance (EPS or matrix), comprised of polysaccharides, eDNA, and proteins. These microbial communities are often associated with persistent infections and in some cases, asymptomatic infections.
The mechanisms by which biofilm formation occurs and are regulated are not completely understood in the Bordetella species, although several pathways and important components of the biofilm have been identified. One of the major observations that inspired this work was that a B. bronchiseptica strain from which the cyaA gene, encoding adenylate cyclase toxin (ACT), has been deleted, formed more biofilm than wild type bacteria, suggesting the possibility of an inhibitory effect of ACT. ACT is a host-directed, protein, bacterial toxin, and the toxin’s role as an inhibitor of biofilm is unprecedented. It was also discovered by Zaretzky et al that ACT binds Filamentous Hemagglutinin (FHA), a surface displayed adhesin required for biofilm formation. Until now, the consequences of this interaction were unknown. We have found that ACT binds to FHA via a direct interaction between the catalytic domain of ACT (AC domain) and the mature c-terminal domain of FHA (MCD). This protein-protein interaction results in the inhibition and disruption of biofilm formation. The AC domain is also able to inhibit biofilm of other bacterial species that express an FHA-like protein.
In fitting these findings into the overall body of knowledge regarding Bordetellae biofilm, we found major differences in the regulatory processes controlling B. bronchiseptica and B. pertussis biofilm. Flagella are essential for initial binding and mature biofilm formation by B. bronchiseptica. In contrast, B. pertussis forms biofilm despite being classically defined as a non-motile and non-flagellated bacterium. B. pertussis encodes all of the required genetic material required for flagella expression, but a stop codon in one of the flagella synthesis gene renders B. pertussis unable to express flagella, and therefore non-motile.
Under various conditions, we noted a differential expression of genes required for flagellar synthesis and function in B. pertussis. It was because of this apparent regulation of flagellar gene expression that we tested for motility. We found that B. pertussis is motile, as demonstrated as outward spreading in the soft agar motility assay, and that these motile bacteria express flagella. These data challenge the idea that B. pertussis is a non-motile organism and also verify that the regulatory mechanisms, involved in B. bronchiseptica flagella expression and the motile phenotype, are conserved in B. pertussis.
These data raise many questions about what we know and what we do not know about this pathogen, and require further research to discover the meaning of these two phenotypes. The findings described herein add to the general body of knowledge for B. pertussis and B. bronchiseptica, and alter the manner in which we think about bacterial pathogenesis during the age of next-generation vaccine development.