The Cotter Lab has a long-standing interest in understanding molecular mechanisms used by bacteria to cause disease. We are especially interested in 1) how and why bacteria differentially regulate the expression of virulence factor-encoding genes in response to the various environments they encounter throughout the infectious process, and 2) the secretion and function of proteins secreted by the Two Partner Secretion (TPS) pathway. We study members of the Gram-negative genus Bordetella (which includes B. pertussis, the causal agent of whooping cough) as models because they are relatively easy to grow, they are genetically tractable, and B. bronchiseptica (a close relative of B. pertussis) has a broad host range that includes many animals commonly studied in the laboratory, such as rabbits, rats and mice. Using B. bronchiseptica with its natural hosts ensures that the information we obtain from our experiments is biologically relevant.
In Bordetella, expression of all known protein virulence factor-encoding genes is controlled by a sensory transduction system called BvgAS. Early work by us and others led to the conclusion that the so-called Bvg+ phase (the phenotypic state expressed when the BvgAS system is active) is necessary and sufficient for Bordetella to establish a persistent respiratory infection and that the Bvg– phase (when BvgAS is inactive) is required for survival outside the mammalian respiratory tract. Together with our collaborator Steve Julio at Westmont college, we discovered recently, however, that another signal transduction system, called PlrSR, is required for Bordetella to survive in the lower respiratory tract (LRT) and for the BvgAS system to remain active at this site. PlrSR is not required for bacterial survival in the nasal cavity, indicating that PlrSR is important for sensing and responding to conditions specific to the LRT. We showed that PlrSR is required for survival in the LRT even if BvgAS is rendered constitutively active, suggesting that BvgAS-independent, PlrSR-dependent genes exist that are critically important for bacterial survival at this site. Our goals now (Fig. 1) are to determine how the PlrS and PlrR proteins function mechanistically, including identifying the signals to which they respond, to identify the PlrSR-dependent genes that are required for bacterial survival in the LRT, and to determine the link between PlrSR and BvgAS activities in the LRT. In addition to shedding light on the intricate regulatory networks used generally by bacteria to navigate environments within and between hosts, our studies may reveal new targets for therapeutic and vaccine development.
One of the main virulence factors produced by Bordetella pertussis and Bordetella bronchiseptica is filamentous hemagglutinin (FHA), a large beta-helical protein that is both cell-associated and secreted into the extracellular environment during growth in vitro. FHA is secreted by the Two Partner Secretion (TPS) pathway and is, in fact, the prototypical member of this large protein family. FHA is first synthesized as a ~375 kDa preproprotein called FhaB. Its N-terminal signal sequence is removed during Sec-mediated translocation across the cytoplasmic membrane and its C-terminal ~130 kDa prodomain is removed during translocation across the outer membrane. Our lab has been instrumental in determining that although the N-terminus of FhaB is required for initiating interactions with the outer membrane channel protein FhaC, the C-terminus of FHA is located distally from the cell surface after maturation to FHA. Moreover, the prodomain remains in the periplasm and although it is degraded rapidly upon cleavage from FhaB, it is critical for FhaB/FHA function during infection. Our current goals are to determine how the prodomain subdomains function in FhaB/FHA processing and FhaB/FHA-mediated interactions with epithelial cells and phagocytic cells during respiratory infection (Fig. 2). These studies will provide mechanistic insight into the broadly-used TPS pathway and may also reveal how bacteria can distinguish and tailor their response to different host cell types.
Burkholderia pseudomallei and Burkholderia thailandensis are closely related Gram-negative soil saprophytes that are endemic to northern Australia and southeast Asia. Unlike B. thailandensis, B. pseudomallei can cause serious and sometimes fatal disease in humans. It has also been weaponized in the past and hence has been classified as a CDC Tier 1 select agent and it must be grown and manipulated only in a BSL-3 laboratory. In the mid-90s, we set out to test the hypothesis that the TPS proteins encoded in the B. pseudomallei genome function as virulence factors, analogous to Bordetella FHA. Instead, we discovered that these proteins function in contact-dependent interbacterial cooperation and competition (Fig. 3). Our studies led to the realization that these so-called CDI systems are widespread among Gram-negative bacteria and that the toxins present at the C-terminal ends of the large exoproteins are highly variable within and between species. Our work suggests that these systems function as multicolored greenbeards that shape community development and composition. We are currently investigating the mechanisms by which these proteins induce gene expression changes in neighboring ‘self’ cells. We have also recently discovered that the genes encoding these proteins in B. thailandensis are located on a currently mobile genomic island and we are investigating the mechanism by which this island produces copies of itself that are then transferred to other chromosomal locations within and between bacterial cells. Understanding sociomicrobiological community development at the molecular level (Fig. 4) may reveal strategies for controlling biofilm-mediated infections, for shifting microbiomes towards a health-promoting state, and for altering bacterial communities for industrial purposes.
Like B. pseudomallei and B. thailandensis, Burkholderia cepacia complex (Bcc) strains are common soil saprophytes. These bacteria, however, are opportunistic pathogens that cause especially severe disease in the lungs of people with cystic fibrosis (CF). We have discovered that some Bcc strains use CDI systems for intra-species competition. Moreover, some Bcc strains produce Type 6 Secretion Systems (T6SS), which are also contact-dependent interbacterial competition systems, but which can function in an interspecies manner. As CF lung infections are typically polymicrobial, understanding how Bcc strains cooperate and compete with other bacteria they encounter in this environment may lead to new therapeutic strategies (Fig. 5). Another challenge for understanding the pathogenesis of CF lung infection is the lack of suitable animal models. We are working closely with Wanda O’Neal and Richard Boucher here at UNC, who have created a beta-ENaC mouse line that recapitulates many features of the CF lung. We are currently working towards using these mice to develop a model for investigating the roles of CDI and T6SS in Bcc virulence.