Abstract

Antibiotic resistant bacteria are becoming increasingly difficult-to-treat resulting in increased morbidity and mortality, especially for common infections such as urinary tract infections (UTIs). Globally, UTIs afflict over 400 million people a year. Numerous different gram-negative and gram-positive bacterial species cause UTIs, including uropathogenic Escherichia coli (UPEC), Klebsiella pneumoniae, and Acinetobacter baumannii. Many of these bacterial species are listed as pathogens of urgent concern by the Center for Disease Control due to the dangerous rise and prevalence of antibiotic resistance necessitating development of new therapeutic measures. Bacteria utilize adhesin factors, such as chaperone-usher pathway (CUP) pili to gain a foothold in the bladder, where if left untreated, progress to pyelonephritis and sepsis. A model CUP pilus is the UPEC type 1 pilus which is a long extracellular fiber tipped with the two-domain mannose binding adhesin FimH. FimH facilitates UPEC attachment to the bladder epithelium and is essential to UPEC bladder infection. Due to structural interactions between the two domains, the UPEC FimH adhesin exists in a conformational equilibrium between a high-affinity relaxed state and a low-affinity tense state. Naturally occurring FimH sequence variants can allosterically shift the conformational equilibrium towards higher or lower affinity, but shifting the equilibrium to either direction leads to attenuation of UPEC in mouse models of UTI. Paradoxically, K. pneumoniae encodes a nearly identical FimH protein that is essential in bladder pathogenesis yet binds with very low-affinity. Using a combination of crystallography, in vitro binding assays, and mouse infection experiments, I delineated the structure-function relationship between the K. pneumoniae FimH conformational equilibrium and UTI pathogenesis. Despite high sequence conservation, I identified naturally occurring K. pneumoniae FimH sequence variants from human catheter infections that allosterically shifted the two-domain FimH towards higher affinity. These higher affinity FimH variants increased bladder titers in uncomplicated UTI, but not CAUTI, partially explaining the epidemiology of K. pneumoniae UTI. To develop antibiotic-sparing treatments of UTI, we developed monoclonal antibodies (mAbs) to E. coli and K. pneumoniae FimH. We identified high-affinity mAbs to purified FimH lectin domain truncate protein, but discovered that the natural conformational equilibrium of two-domain FimH decreased mAb binding affinity. Many mAbs displayed binding preference for the relaxed FimH structural state due to the structural interactions between FimH and the mAbs. We found a subset of mAbs with the same binding epitope that prevented UPEC infection in mice in an Fc-independent manner. Together, we identified a new potential treatment of E. coli and K. pneumoniae UTI and outlined a road map for future development of mAbs to bacterial adhesins. While the development of anti-adhesin therapeutics, such as ligand mimetics and mAbs, have been successful, we lack the biochemical ligand binding information needed to apply ligand mimetics to other adhesins and development of mAbs is time and labor intensive. Recent advancements in deep learning and protein structural prediction have led to the design of de novo proteins, proteins that do not exist in nature, with novel functions. We utilized protein design to create miniprotein inhibitors to the A. baumannii adhesins Abp1D and Abp2D involved in catheter-associated UTI (CAUTI). I crystallized and structurally characterized a cross-reactive miniprotein to the binding pocket of Abp1D and Abp2D. The leading miniprotein therapeutic prevented A. baumannii infection in our murine CAUTI model. Protein design offers an efficient method to create therapeutics for bacterial virulence factors and allows precise targeting of less characterized bacterial adhesins. In addition, there are numerous CUP pili among gram-negative bacteria that have diverse structures resulting in differential ligand specificity. Further showing the structural diversity of adhesins, I solved the crystal structure of the Yeh-like adhesin (YhlD) receptor binding domain highlighting a unique CUP pili adhesin binding domain structure with a mobile flap that may contribute to E. coli attachment in the gastrointestinal tract (GIT). Together, this work highlights the structural biology of CUP pili adhesins and their involvement in UTI and GIT colonization. My dissertation outlines common themes in CUP pili adhesin function, conformational dynamics, and inhibition that establish a foundational paradigm for understanding bacterial attachment in disease and therapeutic intervention.

Committee Chair

Scott Hultgren

Committee Members

Andrew Kau; Christina Stallings; David Hunstad; Michael Caparon

Degree

Doctor of Philosophy (PhD)

Author's Department

Biology & Biomedical Sciences (Plant & Microbial Biosciences)

Author's School

Graduate School of Arts and Sciences

Document Type

Dissertation

Date of Award

7-4-2025

Language

English (en)

Download

Included in

Biology Commons

Share

COinS