Adherence to host tissue is an important first step in the process of bacterial colonization and the pathogenesis of bacterial disease, and a detailed understanding of the determinants of adherence may lead to medical interventions aimed at preventing disease. The St. Geme lab investigates the molecular determinants of adherence by nontypeable (nonencapsulated) Haemophilus influenzae and Kingella kingae, two common pediatric pathogens.

1. H. influenzae

Nontypeable H. influenzae is a common cause of otitis media and sinusitis in children and is also an important etiology of exacerbations of underlying lung disease, such as cystic fibrosis, bronchiectasis, and chronic obstructive pulmonary disease. Isolates of nontypeable H. influenzae express a variety of adhesive proteins that interact with the host respiratory epithelium and facilitate the process of colonization. Examples under study in the St. Geme lab include HMW1, HMW2, Hia, and Hap, all of which are transported to the bacterial surface via the type V secretion system as either members of a two-partner secretion system or autotransporters.

HMW1 and HMW2 are highly homologous glycoproteins that are prototypic members of a two-partner secretion system and are translocated across the outer membrane by cognate outer membrane proteins called HMW1B and HMW2B, respectively. The HMW1 and HMW2 adhesins form short helical fibers on the bacterial surface and facilitate high levels of adherence to cultured human respiratory epithelial cells. HMW1 and HMW2 have another partner involved in their maturation, a glycosyltransferase encoded by the hmw1C or hmw2C gene. HMW1 and HMW2 are modified in the cytoplasm with monosaccharides and disaccharides at multiple sites and are then transported into the periplasm by the Sec secretion system. In the periplasm the N-terminal TPS domain targets HMW1 and HMW2 to HMW1B and HMW2B, facilitating translocation across the outer membrane and localization on the bacterial surface. In ongoing research we are investigating the recognition between HMW1 and HMW1B and between HMW2 and HMW2B, the mechanism of HMW1 and HMW2 tethering to the bacterial surface, the purpose of glycosylation of HMW1 and HMW2, and the conservation of HMW1C-like glycosyltransferases in other gram-negative bacteria.

This figure shows the hmww1 and hmw2 loci, which are in physically separate locations on the chromosome. Each locus encodes an adhesin (hmw1A encodes HMW1, and hmw2A encodes HMW2), an outer membrane translocator (HMW1B or HMW2B), and a glycosyltransferase (HMW1C or HMW2C).

This figure shows the surface of E. coli strain BL21omp8 expressing the hmw1 locus using quick-freeze deep-etch electron microscopy. This high-resolution microscopy technique highlights the HMW1 fibers, which are short, helical structures and are typically present on the bacterial surface in pairs.

HMW1 and HMW2 are highly homologous but have distinct binding specificities, reflecting the fact that they have distinct binding domains corresponding to the regions of maximal amino acid sequence dissimilarity. These light micrographs show adherence to Chang conjunctival cells by E. coli HB101 expressing HMW1 or HMW2.

The Hia adhesin is a trimeric autotransporter and is present in approximately 25% of nontypable H. influenzae strains, including almost all H. influenzae strains that lack HMW1 and HMW2 adhesins. Like other trimeric autotransporters, Hia has an N-terminal signal peptide, an internal passenger domain, and a C-terminal translocator domain (also designated a β-domain) and folds into a trimeric structure with three identical faces. The Hia passenger domain harbors two homologous binding domains referred to as HiaBD1 and HiaBD2 and mediates high affinity adherence to human respiratory epithelial cells. The presence of three identical faces provides potential for three identical binding pockets and a multivalent interaction with the host cell surface, resulting in increased avidity and more stable interaction compared with a monovalent interaction. This increase in avidity may be especially important in allowing organisms to overcome mechanical forces in the host, including the mucociliary escalator, coughing, and sneezing. Beyond facilitating multivalent binding, a trimeric architecture may also confer stability to proteases and detergents. In ongoing research, we are investigating the Hia host cell receptor and are examining the relationship between Hia-mediated adherence and epithelial cell cytokine production.

The purified Hia passenger domain demonstrates high affinity binding to human epithelial cells, as highlighted by fluorescence microscopy.

The HiaBD1 binding domain has a mushroom-shaped structure with a broad stem at the N-terminal end and an elongated cap at the C-terminal end.

The HiaBD1 binding domain has a binding pocket that contains a patch of acidic residues, as highlighted in purple.

Hap is a conventional autotransporter protein and has an N-terminal signal peptide, an internal passenger domain, and a C-terminal translocator domain in a monomeric architecture. This protein was first identified based on its capacity to promote bacterial adherence and entry in assays with cultured human epithelial cells. In addition, Hap mediates bacterial adherence to extracellular matrix proteins and bacterial aggregation and microcolony formation. The C-terminal 511 amino acids of the Hap passenger domain are responsible for interactions with fibronectin, laminin, and collagen IV, and the C-terminal 311 amino acids of the Hap passenger domain are responsible for interactions with epithelial cells and for mediating Hap-Hap interaction, triggering bacterial aggregation and microcolony formation. Hap also contains an N-terminal serine protease domain, with a canonical chymotrypsin family catalytic triad including His98, Asp140, and Ser243. The serine protease domain mediates autoproteolysis and release of the Hap passenger domain from the bacterial surface, modulating bacterial adherence and aggregation. Autoproteolytic activity is inhibited by secretory leukocyte inhibitor (SLPI), a serine protease inhibitor that is present in varying amounts in the upper and lower respiratory tract and that results in enhanced Hap adhesive activity. In ongoing work, we are examining the relationship between Hap-mediated microcolony formation and biofilm formation, the mechanism by which lipopolysaccharide biosynthesis affects Hap stability in the outer membrane, and the influence of Hap on HMW1- and HMW2-mediated adherence.

The panel on the left shows the ribbon crystal structure of the Hap passenger domain as it extends from the C-terminal β-barrel membrane anchor and the bacterial surface, with the N terminus of the passenger domain located distally. The scanning electron micrographs on the right show expanding bacterial microcolonies after incubation of H. influenzae strain DB117 expressing Hap for 1, 2, or 3 hours in a culture tube.

2. Kingella kingae

Kingella kingae is an emerging pediatric pathogen that is being recognized increasingly as a cause of septic arthritis, osteomyelitis, and bacteremia in young children. The St. Geme lab is currently studying the interplay between three surface factors that influence K. kingae adherence to host tissue, namely, type IV pili, a trimeric autotransporter called Knh, and a polysaccharide capsule. According to our current model, type IV pili mediate the initial interaction with the host cell surface and then undergo retraction, drawing the organism closer to the host cell and displacing the polysaccharide capsule. Displacement of the capsule exposes Knh, allowing Knh to interact with an as yet unknown host cell receptor. Work to validate this model is ongoing using genetic, biochemical, and cell biological approaches. In additional work, we are characterizing the structural and antigenic heterogeneity and the functional properties of the K. kingae polysaccharide capsule. We are also using a rat model of invasive disease to examine the contribution of a variety of putative virulence factors to K. kingae virulence.

Kingella kingae produces type IV pili, demonstrated as long, flexible, peritricous fibers after staining with uranyl acetate and examination by transmission electron microscopy.

Kingella kingae produces a polysaccharide capsule, demonstrated as a halo of electron dense material after staining with cationic ferritin and examination by transmission electron microscopy.