Relevant Degree Programs
Graduate Student Supervision
Doctoral Student Supervision (Jan 2008 - May 2019)
Lipopolysaccharide (LPS) is a component of the outer membrane in most Gram-negative bacteria. The lipid A region of LPS anchors the molecule to the outer membrane and forms the first barrier between Gram-negative bacteria and the extracellular environment. Lipid A is also important for bacterial interaction and activation of host immune cells through binding to Toll-like receptor 4 (TLR4), activation of which results in a downstream inflammation response. The work presented in this thesis explores the genetic basis for different lipid A structures and the effects of this structural variability on activation of TLR4 and resistance to cationic antimicrobial peptides (CAMPs). Penta-acyl lipid A from B. pertussis strain BP338 is modified with glucosamine (GlcN), and mutational analyses revealed that LgmA, LgmB, and LgmC are required for this modification. Bioinformatic analysis suggests the following hypothetical model: LgmA transfers N-acetylglucosamine (GlcNAc) from UDP-GlcNAc to the carrier lipid C55P, LgmC removes the acetyl group, and LgmB transfers GlcN from C55P to the lipid A phosphate. Glycosyltransferase assays with LgmA-expressing E. coli membranes show that LgmA transfers GlcNAc onto a lipid, therefore supporting the first step in this model. Site-directed mutagenesis has also identified a putative active site in LgmA and LgmC. Killing assays show the GlcN modification in B. pertussis increases resistance to numerous CAMPs and to outer membrane perturbation by EDTA. Lipid A from B. pertussis strain 18-323 exhibits low levels of TLR4 activation. Lipid A of strain 18-323 has no GlcN modification (due to an incomplete lgm locus) and a shorter C3’ acyl chain compared to strain BP338 (due to a difference in LpxA). Complementation of 18-323 with BP338 lipid A-modifying genes lpxA and/or the lgm locus increase TLR4 activation, though the GlcN modification had a dominant effect. In hexa-acyl E. coli lipid A, shortening the C3 and C3’ acyl chains had been shown to decrease TLR4 activation and resistance to polymyxin B, but increase activation of the limulus amebocyte lysate assay. Thus, varying the structure of lipid A, in both B. pertussis and E. coli, can affect TLR4 activation and resistance of the bacteria to CAMPs.
Autotransporters are a superfamily of Gram-negative secreted proteins, composed of a Cterminaldomain which forms a β-barrel in the outer membrane and plays a significant role intranslocation of the N-terminal passenger domain to the cell surface. A subfamily ofautotransporters is N-terminally acylated during their biogenesis, a post-translationalmodification demonstrated to be essential for their function. Considering the autotransporterpassenger domain is secreted beyond the outer membrane, how translocation can occur in thepresence of N-terminal acyl groups is undetermined. The work in this thesis describes thebiogenesis of the Bordetella autotransporter F (BapF) from B. bronchiseptica RB50, which is Nterminallyacylated during the initial stages of its secretion to the cell surface. The acyl groupsattached to BapF at the Cys28 residue were shown to be subsequently removed by signalpeptidase 1 cleavage between residues Ala34 and Ala35, indicating that BapF forms an acylatedintermediate in its secretion pathway. Studying the secretion of BapF mutants in which Nterminalacylation was ablated revealed that the passenger domain was still capable of reachingthe cell surface; the surface-expressed passenger domain mediated phenotypes of cellularaggregation and an increased rate of culture sedimentation, similar to that observed in E. colicultures expressing the wild-type BapF protein. However, cells expressing these mutantsappear to have damaged outer membranes, potentially due to the observed increase in fulllengthprotein accumulating in the periplasm. Alternatively, E. coli cells expressing a BapFmutant in which signal peptidase 1 cleavage is blocked do not exhibit obvious aggregation andsedimentation phenotypes. Yet, an independent passenger domain is clearly produced. Basedon the results presented in this thesis, it can be hypothesized that sequential processing of theBapF signal peptide, producing an acylated intermediate in the secretion pathway, helps toregulate the passage of BapF through the periplasm ultimately permitting surface expression. Inaddition, bioinformatic and molecular analysis strongly suggest the BapF passenger domainfolds into a β-propeller structure, and if proven to do so, BapF will be the first autotransporterreported with this conformation
Master's Student Supervision (2010 - 2018)
Autotransporters (ATs) are an important family of proteins that are essential for the virulence of a variety of Gram-negative bacteria. The vast majority of ATs possess a classical right handed β-helical structure which facilitates the vectorial secretion of the protein. However, not all ATs possess this classical structure. Tracheal colonization factor (TcfA) of B. pertussis is one of only a few ATs that are predicted to be relatively unstructured or to possess a coil structure. It is not known what factors are important for the secretion of non- β-helical ATs. This study sought to characterize the secretion of TcfA which could also reveal more broadly applicable requirements for the secretion of ATs and other surface exposed proteins. This thesis characterized the secretion of TcfA both in E. coli as well as in B. pertussis. The study determined that TcfA has special secretion requirements that are not met when the protein is expressed in E. coli. The unique B. pertussis chaperone Par27 was identified as an important factor for secretion via both its disruption in B. pertussis as well as via its insertion in E. coli. However, it was also determined that there are additional factors that E. coli is lacking that are important for the secretion of TcfA. The study also sought to characterize potential virulence functions of TcfA. It is known that TcfA contributes to the pathogenesis of B. pertussis, but its specific role remains to be elucidated. This study used modeling to provide support for the theory that TcfA binds Factor H in B. pertussis. However, a factor H surface binding assay determined that TcfA is not the only factor that binds the complement regulatory protein Factor H in B. pertussis. Another uniquely structured AT, BapB, was hypothesized as the potential additional factor that binds Factor H. However, additional studies are required to determine the importance of BapB. Furthermore, the study determined that TcfA does not play a large role in the serum survival of B. pertussis. In summary, this thesis characterized the secretion and some potential virulence functions of TcfA, but it also raised many additional questions.
Protein secretion plays an essential role in the virulence of Gram-negative bacterialpathogens. Gram-negative bacteria have evolved multiple specialized secretion pathways inorder to navigate proteins across the Gram-negative cell envelope. The simplest and mostwidespread secretion pathway is the type V secretion system and autotransporters (ATs) (Va)represent the largest class of secreted proteins in Gram-negative bacteria. ATs are structurallycharacterized by the presence of three distinct domains; an N-terminal signal sequence thattargets the N-terminus of the polypeptide to the inner membrane, a passenger domain that oftenpossesses β-helix structure and carries out the virulence function(s) and a conserved C-terminaltranslocation unit (TU) consisting of a β-domain that forms a porin-like structure in the outermembrane (OM) through which the passenger domain is thought to be extruded and an α-helicallinker region that joins the β-domain to the C-terminus of the passenger domain. At the Cterminusof the majority of AT passenger domains is a conserved region termed theautochaperone (AC). The importance of conserved AC residues for passenger domain secretionand folding has been demonstrated for numerous ATs, yet the exact role of the AC region in OMtranslocation and passenger domain folding has yet to be clarified. In this study, the requirementof conserved C-terminal passenger domain motifs for the acquisition of passenger domainsecondary structure and the interchangeability of these conserved AC motifs was investigated. Acombination of far-UV CD spectroscopy and limited proteolysis with trypsin of full-length andAC-deleted Vag8, Ag43 and Smp passenger variants revealed that the requirement of conservedC-terminal AC motifs for the acquisition of passenger domain secondary structure varies amongATs. A cell wall fraction assay, in which the ability of BrkA, Ag43 and Smp AC regions torescue BrkA AC-deleted passenger folding was tested, indicated that an AC region with similarstructure to the cognate AC is necessary to rescue passenger domain folding. Altogether, theresults of this study highlight the involvement of multiple factors in passenger domain foldingand the likely variation that exists in the mechanism of passenger folding at the bacterial cellsurface.