Exploring the Polyclonal Antibody Responses Induced by Klebsiella pneumoniae Derived Outer-Membrane Vesicles

Abstract

Antimicrobial resistance (AMR) is a major global health concern; six key pathogens have been highlighted for causing a disproportionate number of AMR-related deaths globally. These “ESKAPE” pathogens are Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter spp. Of these, Klebsiella pneumoniae stands out due to its vast genomic diversity with only a small percentage of the genome being dedicated to the core genes, while the remainder is drawn from a diverse pan-genome. K. pneumoniae can cause a wide range of pathologies including pneumonia, sepsis, urinary tract infections, and others. A wide variety of virulence factors are used by K. pneumoniae including a thick outer capsule, lipopolysaccharide (LPS), siderophores, and key metabolism and adhesion genes. These virulence factors assist in resisting a wide range of immunological and environmental stressors (complement, phagocytosis, antimicrobials, etc.). Currently, two major phenotypes of concern are emerging: multi-drug resistance and hypervirulence, which make K. pneumoniae a key pathogen of concern for global health. No vaccine against K. pneumoniae has been licensed at time of writing, with various approaches being tested without success. Two aspects of K. pneumoniae remain understudied. Firstly, there exists a gap in the literature between research highlighting the genomic diversity of K. pneumoniae and research examining pathologies and novel therapeutics, which often rely on a small handful of isolates. Secondly, the role of antibodies in K. pneumoniae immunity is currently poorly understood. The aims of this thesis were to examine antibody binding patterns by producing five sets of polyclonal serum using outer-membrane vesicles (OMVs) before characterising the resulting antibody response. Each set of polyclonal serum was then screened against a curated collection of K. pneumoniae isolates (the Warne collection) before examining various antibody effector functions. A method of consistently harvesting OMVs from a range of K. pneumoniae isolates was established, with five sets being taken for further use. OMVs had been selected for use as a means of generating a polyclonal antibody response based on the wide variety of surface antigens present within these structures. The vesicles within these five sets contained a wide variety of different proteins including a sizeable proportion not localised to the bacterial cell surface, due in part to contaminants which could potentially be removed in future work with additional purification steps. These five sets of OMVs were used to immunise mice resulting in five sets of polyclonal serum. The antibodies contained within each serum group bound both homotypically and heterotypically to vesicles from different isolates and were primarily dominated by the subclass IgG1, reducing the likelihood that functional responses would be observed due to inhibitory role IgG1 plays in mice. The antibodies bound a wide variety of different antigens including capsule, LPS, and several different protein targets. The identification of these protein targets was revealed using immunoprecipitation, however the proportion of proteins identified is likely small compared to the number truly targeted by each serum group due to the techniques used. When screening each serum group against whole cells, each group was capable of binding both homotypically and heterotypically against different isolates. However, when each group was screened against the Warne collection, only a small proportion of isolates were bound strongly by each serum group. This binding was despite the likely vast diversity of antibodies present within each serum group. Sharing specific characteristics with the isolates used to produce each serum group was found to increase the likelihood of strong binding but was not a precise correlation. When examining effector functions of the antibodies within each serum group, no group enhanced complement deposition while only a small number of serum group and isolate interactions enhanced phagocytosis. Pre-coating K. pneumoniae cells with antibody had no impact on infection in a pneumonia mouse model; however, immunising with OMVs did prevent systemic dissemination of bacterial cells. This thesis demonstrates the importance of retaining the genomic diversity of K. pneumoniae within experimental design as well as providing novel insights into polyclonal antibody responses against K. pneumoniae. Future work needs to focus on developing high throughput techniques to retain K. pneumoniae diversity within experiments, as well as establishing correlates of protection so that future work examining novel immunotherapies, or the role of antibody responses can be more effective.