Search

The Outer Membrane (OM) of Gram-negative bacteria is the primary barrier for the cell, with its asymmetric composition of lipopolysaccharides (LPS) at the outer leaflet and glycerophospholipids (GPL) at the inner leaflet. The stability of this membrane is important not only as a barrier but also for viability of the cell as it provides an anchor for all outer membrane components necessary for bacterial survival. Thus, defects in the outer membrane, exacerbated by the packing and flipping of GPLs from the inner leaflet to the outer leaflet, allow for the diffusion of compounds that are toxic to the cell, and critically membrane integrity. Despite the apparent importance of the outer membrane, the implications of OM defects on drug- or host-pathogen interactions is not well studied. Antibiotic resistance is a major health priority, which is exacerbated by the increased antibiotic use, rising rates of bacterial antibiotic resistance and a reducing pool of effective drugs. Whilst horizontal gene transfer remains a major route of AMR gene acquisition, chromosomally encoded intrinsic systems play a significant role in conferring protection against both antibiotics and the host immune response. In Klebsiella pneumoniae, the intrinsic transcriptional regulator, RamA, has been shown to regulate membrane permeability via the activation of increased efflux (acrAB) and decreased influx (ompF). Furthermore, RamA has been shown to alter host-pathogen interactions via changes to lipid A, which has been implicated in immune evasion and increased dissemination. As such, RamA remains a key regulator of membrane permeability and integrity in Klebsiella pneumoniae. Transcriptome analyses to define the RamA regulon and proteomic analyses to establish RamA-dependent membrane changes also demonstrate that RamA functions as an activator of the mla (maintaining lipid membrane asymmetry) operon. The recently described mla operon has been attributed with trafficking of GPLs to and from the OM in addition to promoting outer membrane vesicle biogenesis. The mla operon is also involved in host-/ drug-pathogen interactions, as it is responsive to iron limitation via the Fur repressor in vivo and contributes to antibiotic susceptibility. Taken together, the overriding hypothesis of this work is the RamA-mediated activation of the mla operon results in altered OM composition and stability thus contributing to changes in host-pathogen interactions. To address this hypothesis, we aimed to investigate the mla operon, and its contribution to the RamA phenotype in K. pneumoniae, by specifically characterising the changes in outer membrane stability and integrity. Therefore, the specific aims were, to determine the level of changes in membrane stability caused by mla overexpression in a ΔramR background, the effect of these outer membrane changes on drug- and host-pathogen interactions. The RamA overexpressing strain (Ecl8ΔramR) showed significant increase in membrane stability compared to the wildtype Ecl8, when plated on SDS-EDTA plates, a common OM integrity assay. This marked increase was in part due to the mla operon, as deletion of mlaCDEF in the Ecl8ΔramR showed a decrease in survival. The same decrease was also noted in the Ecl8 background in mlaCDEF deletions. Further to this, membrane hydrophobicity, an indicator of GPL accumulation at the OM, as measured via xylene and crystal violet assays, showed a marked decrease in the Ecl8ΔramR compared to Ecl8, whilst mla deletions in the Ecl8ΔramR and Ecl8 showing an increase in hydrophobicity. Changes to the GPL content at the OM and IM in the Ecl8, Ecl8ΔramR and mlaE deletions were investigated via Thin layer Chromatography and Gas Chromatography, to examine changes in head group and fatty acid tails respectively. A significant increase of phosphatidylethanolamine (PE), one of the most abundant GPLs, was seen at the OM in Ecl8ΔramR compared to Ecl8. A PagP assay, that cleaves palmitic acid tails from OM outer leaflet exposed GPLs, demonstrated that this increase of PE was confined to the inner leaflet of the OM, confirming that RamA overexpression leads to alterations of the OM to maintain stability. The mlaE deletion in an Ecl8 background demonstrated an increase in PE content at the OM, however the same increase was not noted in an mlaE deletion in the RamA overexpressing background Ecl8ΔramR. Via a PagP assay the mlaE deletion in both Ecl8 and Ecl8ΔramR backgrounds showed a significant increase in surface exposed GPLs, confirming the phenotypic data that mla expression is responsible for the reduction of GPL accumulation at the outer leaflet of the OM, maintaining membrane stability. Gas chromatography (GC) of the IM and OM fatty acid tails additionally demonstrated that Ecl8ΔramR showed a significant increase in palmitic acid levels at the OM compared to Ecl8, with a slight decrease noted in the Ecl8ΔramR mlaE deletion. The mlaE deletions in both backgrounds showed imbalance in the palmitic acid and vaccenic acid tails in the OM and IM. The lipidomic work indicates that overexpression of RamA led to changes in GPL and fatty acid content at the OM, some dependant of the mla operon and others via potential alterations of other pathways, specifically GPL and Fatty acid (FA) synthesis, leading to the reduced susceptibility phenotype. More importantly when a ramR deletion in Kp52145 was created, the same trends as Ecl8ΔramR were seen, demonstrating the changes were not strain dependant. Due to RamA dependant but mla independent changes being noted at the OM, it was important to investigate further the role that RamA could be having on GPL synthesis. This was carried out via metabolomics and proteomics, specifically focussing at FA and GPL synthesis proteins. Importantly, Ecl8ΔramR showed a significant increase in several FAS II proteins, with expression variation known to affect FA tail length, along with PssA the first committed step to PE synthesis. In addition, Glucose-6-phosphate dehydrogenase (zwf), the main producer of NADPH, also showed a significant increase in Ecl8ΔramR (8.1-fold), whilst in a gene reporter fusion assay it was significantly upregulated. With a 5.4-fold NADPH increase in Ecl8ΔramR compared to Ecl8. NADPH is a key cofactor for the FA elongation pathway and turnover. In conjunction with changes seen at the OM via TLC and GC, this demonstrated that RamA overexpression influenced GPL composition as well as having a tight control of the OM/IM via mla trafficking. As the OM is the primary point of contact it is given that any changes would immediately influence host-pathogen interactions. Using pooled human serum, it had already been established that Ecl8ΔramR was more resistant than Ecl8 to complement. With an mlaE deletion, an increase in sensitivity was observed relative to the wild type and RamA overexpressor. In vivo infections of G. mellonella, showed an increase in killing, melanisation and dissemination in Ecl8ΔramR compared to Ecl8, demonstrating an increase in virulence. Interestingly, the mlaE deletion in Ecl8 showed a significant increase in killing and melanisation, suggesting a role for the mla operon directly in virulence. However, the mlaE deletion in Ecl8ΔramR did not show a significant change post 24hrs for any of the assays. In conjunction with the complement assay, this would suggest that the RamA regulation of the mla operon plays an important role in immune evasion. Therefore, this study shows that the changes at the OM mediated by RamA have a key role in altering antibiotic susceptibility and host-pathogen interactions; where these perturbations are not limited to is the active efflux and influx systems but affect the actual outer membrane barrier which has significant implications to the host- and drug- pathogen interactions.

Sepsis is a life-threatening condition caused by the body's overwhelming response to an infection, such as pneumonia or urinary tract infection. It occurs when the immune system releases cytokines into the bloodstream, triggering widespread inflammation. If not treated, it can lead to organ failure and death. Unfortunately, sepsis has a high mortality rate, with studies reporting rates ranging from 20% to over 50%, depending on the severity and promptness of treatment. According to the World Health Organization (WHO), the annual death toll in the world is about 11 million. One of the main toxins responsible for inflammation induction are lipopolysaccharides (LPS, endotoxin) from Gram-negative bacteria, which rank among the most potent immunostimulants found in nature. Antibiotics are consistently prescribed as a part of anti-sepsis-therapy. However, antibiotic therapy (i) is increasingly ineffective due to resistance development and (ii) most antibiotics are unable to bind and neutralize LPS, a prerequisite to inhibit the interaction of endotoxin with its cellular receptor complex, namely Toll-like receptor 4 (TLR4)/MD-2, responsible for the intracellular cascade leading to pro-inflammatory cytokine secretion. The pandemic virus SARS-CoV-2 has infected hundreds of millions of humans worldwide since its emergence in 2019. The COVID-19 (Coronavirus disease-19) caused by this virus is associated with high lethality, particularly for elderly and immunocompromised people. As of August 2023, nearly 7 million deaths were reported worldwide due to this disease. According to some reported studies, upregulation of TLR4 and the subsequent inflammatory signaling detected in COVID-19 patients "mimics bacterial sepsis". Furthermore, the immune response to SARS-CoV-2 was described by others as "mirror image of sepsis". Similarly, the cytokine profile in sera from severe COVID-19 patients was very similar to those suffering from the acute respiratory distress syndrome (ARDS) and sepsis. Finally, the severe COVID-19 infection is frequently accompanied by bacterial co-infections, as well as by the presence of significant LPS concentrations. In the present review, we will analyze similarities and differences between COVID-19 and sepsis at the pathophysiological, epidemiological, and molecular levels.

Enterobactin (Ent)-mediated iron acquisition is critical for many Gram-negative bacteria to survive in the host. Given the bacteriostatic effect of lipocalin-2 resulting from its potent Ent-binding ability, immune interventions directly targeting Ent is a promising antimicrobial strategy against Gram-negative bacterial infections. Moreover, hyperimmune egg yolk antibody is an emerging passive immune agent for the control of bacterial infections. Oral administration of anti-Ent egg yolk antibody may confer passive immune protection against Ent-dependent enteric pathogens. In this dissertation study, multidisciplinary approaches in conjunction with different model systems (in vitro, in vivo, and ex vivo) were used to examine protective efficacies of the Ent-based immune interventions. Anti-Ent antibodies induced by an Ent conjugate vaccine were demonstrated to exert in vitro bacteriostatic effect on Ent-dependent Gram-negative bacterial pathogens, including diverse Escherichia coli and Salmonella enterica strains with different siderophore production profiles as well as Campylobacter spp. In the subsequent chicken trial aimed to assess active immune protection of the Ent conjugate vaccine against avian pathogenic E. coli (APEC) infection, the vaccine induced high titers of anti-Ent IgY in the serum as well as specific IgY in the intestine. As a result, the vaccination significantly alleviated the organ lesions caused by a highly virulent APEC and affected the chicken intestinal microbiota. Regarding the passive immunization, when a large quantity of anti-Ent hyperimmune egg yolk was orally administered as a feed additive to chickens, it failed to suppress C. jejuni colonization in the chicken intestine, consistent with the finding that the egg yolk antibody was not detected in gastrointestinal tracts and sera of the chickens. Further ex vivo evaluation of egg yolk antibody stability revealed that the antibody was efficiently degraded when entering chicken gizzard. Lastly, an epidemiological survey of E. albertii in broilers revealed chickens as an important reservoir for this emerging zoonotic pathogen. The in vitro growth assay also showed significant inhibitory effect of the anti-Ent antibodies on E. albertii. Collectively, this dissertation study demonstrated significant potential of the Ent-based immune interventions in controlling different Gram-negative bacterial infections.

Gram-negative bacterial infections pose a significant health threat due to their increasing resistance to all classes of currently used antibiotics. Gram-negative bacteria are especially challenging to kill because of their unique cell envelope. This envelope is composed of two membranes that are separated by an aqueous compartment known as the periplasm. The inner membrane is a phospholipid bilayer, and the outer membrane is an unusual asymmetric bilayer containing lipopolysaccharide on the outer leaflet and phospholipids on the inner leaflet. Lipopolysaccharide is a large, highly charged, glycolipid and lateral interactions between lipopolysaccharide molecules creates an electrolyte mesh that forms a permeability barrier that prevents the entry of drugs that are normally able to kill other types of bacteria. Lipopolysaccharide synthesis occurs in the cytoplasm and on the cytoplasmic face of the inner membrane. It is then flipped across the inner membrane by the ATP-powered flippase MsbA. Transport across both the periplasm and the outer membrane is mediated by the seven-protein lipopolysaccharide transport machine (LptB2FGCADE), which forms a bridge that connects the inner and the outer membranes. Transport by LptB2FGCADE is powered by ATP-hydrolysis by the inner-membrane components LptB2FGC. In this thesis, we present progress towards understanding how the lipopolysaccharide transporter works, how it coordinates lipopolysaccharide transport to phospholipid biogenesis and progress towards antibiotics that target lipopolysaccharide transport. In chapter two, we describe one of the first two reported MsbA inhibitors. We find that this compound decouples MsbA’s ATPase activity from its lipopolysaccharide transport activity. In chapter three, we report the development of new tools to observe intermediate lipopolysaccharide transport states in vivo and in vitro. These results are guided by a crystal structure of LptB2FGC. We use these newly developed tools to understand how lipopolysaccharide transport is coordinated to ATP hydrolysis, and how LptB2FGC maintains efficient LPS transport. In chapter four, we use the tools developed in chapter three to investigate the mechanism of action of three compounds that interfere with LptB2FGC in vivo. We show that polymyxin B, which is currently the clinical drug of last resort against multidrug resistant Gram-negative infections, interacts specifically with LptB2FGC in vivo and in vitro. This presents a new avenue for making improved polymyxin analogues. We find that the effect of novobiocin on Lpt depends on ATP concentration, which sheds light into how novobiocin interacts with LptB2FGC. We also show that inhibition of fatty acid biosynthesis causes an inhibition of lipopolysaccharide transport. We speculate that this may form the basis of how the cell coordinates phospholipid biogenesis with lipopolysaccharide biogenesis. In chapter five, we report that a novel macrocyclic antibiotic functions (MCP-1) by inhibiting LptB2FGC activity. We report a 3.1 Å resolution co-complex of lipopolysaccharide, LptB2FG and MCP-1. We show that MCP-1 has a novel non-competitive binding mode. Our structural and biochemical results provide insight both into how MCP-1 functions and how LptB2FGC functions. In chapter six, we report the cryo-EM structure of full membrane-to-membrane LptB2FGCADE protein bridge, with each component at 3.5 Å resolution or better. We find that the transmembrane helix of LptC serves to regulate LptB2FGC activity by sensing proper bridge assembly. Future work can leverage the chemical, biochemical and structural tools described in this thesis to gain a better understanding of how lipopolysaccharide is transported, how the cell coordinates lipopolysaccharide transport, and how to develop drugs that target lipopolysaccharide transport.

The outer membrane of Gram-negative bacteria (GN) makes them distinct among superbugs that are associated with the development of antibiotic resistance. The outer membrane, and inner membrane, separated by the periplasm, form a double-membrane barrier to the entry of antibiotics into the cell. Several studies have been conducted to examine the role of outer membrane modifications such as porins, lipopolysaccharides, and efflux pumps on antibiotic resistance. However, there is a paucity of knowledge on how antibiotics behave in the periplasm, to gain access into their target region. My thesis focuses on understanding the mechanism of antibiotic permeability through the cellular envelope of Gram-negative bacteria. I studied the distribution of fluoroquinolones in the two aqueous compartments (periplasm and cytoplasm) of Escherichia coli using fluorescence intensity measurement and minimum inhibitory concentration (MIC) test. We treated the bacteria cells with each antibiotic, allowed the antibiotic to accumulate in the cells, fractionated the cells and quantified the concentration of accumulated antibiotic through the measurement of its fluorescence intensity. The compound accumulation assay showed that the efflux-deficient strain (DtolC) accumulated more antibiotic than the wild type (WT) strain, for all nine fluoroquinolones tested. An analysis of the subfractions showed a greater accumulation of the antibiotic in the periplasm than in the cytoplasm. A positive correlation was observed between the MIC ratio (WT/DtolC) and the cytoplasmic accumulation ratio (DtolC/WT). This is an indication of the importance of measuring accumulation in the target rather than the whole cells. I also studied the impact of osmo-regulated periplasm glucans (OPGs) on antibiotic susceptibility in GN. We created E. coli and Salmonella Typhimurium strains deficient in OPG production. We also created an E. coli strain that produced neutral OPGs. The drug susceptibility test showed that the strains that are either deficient in OPG production or produce neutral OPGs were less susceptible to some of the positively charged aminoglycosides compared to the WT strain. A similar response was observed when the bacteria strains were treated with the fluoroquinolones and tetracyclines. We speculate that this behavior is due to the net positive charge carried by these antibiotics from complexes formed with Mg2+. In contrast, the strains grew slower in the presence of negatively charged cefuroxime and neutral linezolid. The observed MIC changes were not due to a leaky membrane. The OPG deficient strains produced significantly reduced amount of OPGs compared to the parent strain. Our study demonstrated that charge plays a significant role in OPG-mediated susceptibility to antibiotic. We also probed the role of OPGs on copper homeostasis in GN bacteria. We found that the disruption of the opgGH operon had an impact on the tolerance of GN bacteria to copper. Copper quenched the fluorescence of cytosolic GFP faster in the OPG- deficient strains compared to the WT strain. The GFP-quenching effect of copper ions was diminished with the increase of ionic strength, indicating that Cu2+ penetration into the cytoplasm slowed down under high salt condition.

Gram-negative bacterial infections are a significant threat to human health due to increasing resistance to all classes of clinically used antibiotics. Gram-negative bacteria are extremely challenging to kill because of their unique cell envelope, which consists of an inner phospholipid membrane, an outer membrane, and a layer of peptidoglycan in the aqueous space (periplasm) between the two membranes. The outer membrane is an unusual asymmetric bilayer containing lipopolysaccharide (LPS) in the outer leaflet and phospholipids in the inner leaflet. LPS is a large, highly charged glycolipid, and lateral interactions between LPS molecules create a dense electrolyte mesh that prevents entry of hydrophobic molecules, including most antibiotics. To maintain this protective barrier, cells must quickly and accurately transport millions of LPS molecules from their synthesis site at the cytoplasmic face of the inner membrane, across the aqueous periplasm, and through the outer membrane to the cell surface. The pathway that is responsible for this transport, the lipopolysaccharide transport (Lpt) machinery, consists of seven essential proteins that are proposed to form a transenvelope bridge to transport LPS. This thesis establishes biochemical systems that demonstrate the formation of a bridge across which LPS molecules move from the inner membrane to the outer membrane. We also address how the Lpt machinery recognizes the LPS substrate and achieves efficient transport. In chapter two, we describe a reconstitution of LPS transport from purified components that has enabled the first direct visualization of the Lpt protein bridge. Chapter three describes the development of a quantitative fluorescence-based assay to monitor membrane-to-membrane LPS transport and allow characterization of factors that influence the rate of LPS transport. In chapter four, we identify a cluster of residues required for LPS substrate recognition by the inner membrane Lpt complex and identify a role for LptC in coupling ATP hydrolysis and LPS transport. In chapter five, we characterize a suppressor mutation that allows survival in the absence of LptC and provides insight into its functional role in the Lpt pathway. Finally, in chapter six, we describe progress toward establishing a purification of the transenvelope Lpt protein bridge, with the goal of allowing future structural characterization. Together, these results show how the Lpt machinery accomplishes rapid and efficient LPS transport to the cell surface. Minor disruptions to LPS transport increase outer membrane permeability and cause antibiotic susceptibility, and major disruptions cause cell death, making the LPS transport machinery an attractive target for antibiotic development. Detailed structural and mechanistic knowledge of the transport machinery will greatly facilitate the development of novel antibiotics that disrupt LPS transport.

Urinary tract infections (UTIs) are among the most common infections caused by both Gram-negative and Gram-positive bacteria, acquired in the community and hospitals. There are three main groups of UTIs: (i) lower UTIs that affect the urethra or the bladder, (ii) upper UTIs that affect the kidneys, or (iii) asymptomatic bacteriuria (ABU). Group B Streptococcus (GBS) is a Gram-positive bacteria known to cause a variety of infections in neonates, pregnant women, the elderly or immunocompromised individuals. GBS has been estimated to cause 1-2% of all single organism UTIs. GBS has been shown to form biofilms, on abiotic and biotic surfaces, protecting it from killing by antibiotics or host immune cells and promotes host colonisation. Various factors have been shown to affect the biofilm forming ability of GBS. Here we determined that LB supplemented with glucose was the optimal media for biofilm formation by a strong biofilm forming strain. We then investigated the biofilm forming phenotype of 292 clinical GBS isolates that presented with asymptomatic, acute, or recurrent infection. We found that there was no significant difference in the biofilm forming ability across the clinical presentations. We also showed a significant reduction in the biofilm forming ability of a strong biofilm forming strain and its isogenic maeK and maeE mutants in LB supplemented with 1% glucose. A multiplex PCR screen for genes encoding bsaB, pil1, pil2a, and pil2b found that there was no significant difference in the number of strains that had the right sized fragments for all four genes across the three clinical presentations. We also found that there was a significant difference in the proportion of strains that had the right sized fragments for the pil genes across the three different levels of biofilm activity under shaking conditions. High biofilm forming strains had the lowest proportion of strains that possessed all four genes, compared to low and medium biofilm formers. Lastly, we assessed the haemolytic activity of the strains by growing them on tryptic soy agar plates supplemented with 5% horse blood and found that asymptomatic strains had a significantly higher number of strains with high haemolytic activity compared to acute and recurrent strains.

Gram-negative bacteria pose a major threat to human health as they demonstrate low sensitivity to most antibiotics. This lack of sensitivity to antimicrobials is largely due to the amphipathic membrane architecture and efficient efflux pumps. Additionally, antibiotics have been overused, which increases the prevalence of multidrug-resistant pathogens and thus, less efficient antibiotics. For these reasons, the need for broad- spectrum antibiotics has never been more important. In the early 1980s, a new class of antibiotics based on the natural product acyldepsipeptide (ADEP) was discovered. ADEPs target the well-conserved bacterial protease caseinolytic protease P (ClpP). Overactivation of this protease leads to uncontrolled protein degradation and, in many cases, cell death. However, the membrane-permeability and susceptibility to efflux preclude the use of ADEPs to target Gram-negative bacteria. Unfortunately, a lack of understanding exists for how to rationally “design in” Gram-negative activity and thus it remains very difficult to improve the spectrum of known chemotypes. Related to the aforementioned challenges, we have a research program focused on elucidating which physicochemical properties of small molecules influence Gram- negative membrane-permeation and, in turn, intracellular small molecule accumulation. As part of this initiative, we synthesized two libraries, 1) a peptidomimetic library of putative ClpP activators based on the ADEP pharmacophore and 2) a library based on the fluorescent probe N-phenyl-1-naphthylamine (NPN). In total, 46 new peptidomimetic analogues and 57 NPN derivatives were synthesized. During the construction of these libraries two new applications for the reagent DABAL-Me3 were developed. In addition, all compounds were assessed for minimum inhibitory concentrations against wild-type and compromised Gram-negative bacterial strains. The 103 compounds synthesized in this work represent an important component of a larger collaborative effort to determine molecular features and species-specific biology that dictate small molecule accumulation. Completion of these libraries sets the stage for incorporation into larger scale cheminformatics, accumulation assessments, and kinetic modelling of permeation and accumulation parameters.

Over the last century, the use of antibiotics has enabled many advances in modern medicine, making life as we know it possible. In recent years, however, emerging bacterial resistance to virtually all major antibiotic classes has resulted in a worldwide increase in morbidity, mortality, and financial burden associated with drug resistant infections. The antimicrobial resistance crisis presents an urgent need for new antimicrobials with distinct mechanisms of action from existing drugs. The current pharmaceutical pipeline of new antibiotics is limited due to three obstacles: a lack of understanding of resistance mechanisms, a dearth of novel mechanisms of action among new antibiotics, and drug discovery challenges unique to bacteria due to their cellular and physiological composition. My dissertation research addressed each of these challenges. The mechanisms of resistance to folate biosynthesis inhibitors in Staphylococcus aureus were explored from a genetic, biological, biochemical, and structural basis. Unexpected roles in resistance and fitness were attributed to various mutations in the sulfonamide target dihydropteroate synthase. This information currently guides the development of next-generation antifolates designed to avoid these characterized resistance strategies. In the subsequent chapter, a conditional metabolic screening approach was employed to discover inhibitors disrupting metabolic pathways related to folate biosynthesis. The testing conditions were optimized to measure the biological activity of antimetabolites, which is often masked by nutrients present in standard testing media. This screen yielded an inhibitor of cysteine synthase A in Escherichia coli, which was characterized in chapter four. Multiple experimental approaches yielded indications that the cysteine synthase A inhibitors have a false-product mechanism, resulting in a widespread impact on several key branches of metabolism beyond cysteine biosynthesis. The final research chapter describes the adaptation of the cellular thermal shift assay to explore target engagement in the Gram-negative cell system. Drug entry and accumulation are especially challenging to achieve in Gram-negative cells due to their unique dual membrane system and associated efflux machinery. This assay provided an early stage tool to quickly assess the ability of antimicrobial candidates to engage the intended target in the intact cell system and also measure efflux sensitivity. Taken together, this dissertation contributes to the fight against the antimicrobial resistance crisis from multiple angles, all within the context of bacterial metabolism which is a rich source of exciting new antibiotic targets.

The cell envelope of Gram-negative bacteria has a unique double-membrane architecture that limits the entry of antibiotics and allows them to colonize harsh environments. The cytoplasmic membrane and peptidoglycan cell wall, typical of all bacteria, are surrounded by an outer membrane with an outer leaflet composed primarily of lipopolysaccharide (LPS) rather than phospholipids. LPS is composed of hundreds of sugars and several saturated acyl chains, and the asymmetric distribution of LPS in the outer membrane makes Gram-negatives unusually impermeable to hydrophobic compounds. The cytoplasmic and inner-membrane enzymes involved in the biosynthesis of LPS have been studied in great detail, however, the proteins responsible for its assembly in the outer membrane have only been identified recently. In E. coli, seven essential LPS transport proteins, LptA-LptG, transport LPS from the outer leaflet of the inner membrane, across the aqueous periplasm, to the outer leaflet of the outer membrane. The cytoplasmic and inner membrane Lpt components form complex, LptB2FGC, which is a member of the ATP-binding-cassette (ABC) -transporter superfamily and uses the energy released in ATP binding and hydrolysis to extract LPS from the inner membrane and power its transport to the outer membrane. The mechanism by which cytoplasmic LptB couples ATP cycling to LPS extraction from the membrane remains mysterious, as do the means by which LptB2FGC ensures unidirectional LPS transport to the outer membrane. In the first part of this thesis, we use structural, biochemical, and genetic tools to interrogate the interactions between LptB and LptFG. We find that although LptB2FGC is an atypical ABC-transporter, it contains several motifs common to other members of the family and uses widely conserved interactions between the nucleotide binding domains and transmembrane domains to mediate LPS transport. During these studies, we make the serendipitous discovery that the antibiotic novobiocin binds to Lpt and is an agonist of LPS transport. As LPS is both essential in most Gram-negative bacteria and a primary factor in their antibiotic resistance, structural and mechanistic characterization of the Lpt proteins could enable the development of new antibiotics. In the second part of this thesis we report the first X-ray crystal structures of LptB2FGC and use information glean from them to guide functional studies of LPS transport. We identify the path LPS takes out of the membrane, into a cavity within the transporter, and on to the Lpt bridge to the outer membrane. We find that LPS entry into the cavity is ATP-independent whereas extraction from the membrane require ATP. Comparison of the structures reveals a protein gate that must open to allow LPS movement out of the cavity and onto the bridge. Together, our findings show how LptB2FGC acheives efficient unidirectional transport of LPS to the outer membrane.

Several Gram-negative bacteria modify their outer most surface structure, lipopolysaccharide (LPS), to evade immune surveillance and survive within the host. Many of these changes occur within the lipid A domain, a region that is recognized by the innate immune system via Toll-like receptor-4 (TLR4). One such pathogen, Porphyromonas gingivalis, orchestrates chronic inflammatory disease by disrupting immune homeostasis. P. gingivalis initially synthesizes a penta-acylated lipid A that functions as a weak TLR4 agonist but displays tetra-acylated forms that are either immunologically silent or TLR4 antagonists. The impact of lipid A modifications on downstream signaling and antigen-specific immunity are unclear. TLR4 signals from the plasma membrane through a MyD88-dependent pathway and intracellularly through a TRIF-dependent pathway. Here we show that expression of immunological silent or antagonistic lipid A enables P. gingivalis to evade TRIF-dependent signaling in dendritic cells (DCs). Evasion of TRIF signaling accelerated antigen degradation and impaired priming of pathogen-specific T cells. In contrast, a P. gingivalis strain expressing agonist lipid A potently activated TRIF signaling and delayed antigen degradation, thereby preserving peptides for optimal T cell activation. We propose that lipid A modifications control the endocytic activity of DCs and the efficiency at which microbe-specific T cells are primed. We next investigated the impact of purified P. gingivalis LPS on innate signaling and antigen presentation. All P. gingivalis LPS species induced a program of DC maturation that allowed for constitutive antigen uptake and cross-presentation. This was independent of TLR4 agonist activity and required CD14, a protein that transports TLR4 to endosomes where TRIF signaling can occur. Agonist LPS induced signaling through both MyD88 and TRIF and elicited T cell priming. Antagonistic LPS potently accelerated CD14 endocytosis and induced TRIF-biased signaling leading to comparable degree of cross-priming. Immunologically silent LPS promoted CD14 endocytosis but failed to activate signaling and induced T cell tolerance. Collectively, our results demonstrate that modification of lipid A structure enables Gram-negative bacteria to direct the host immune system towards tolerance or immunity. We propose that these findings can be harnessed for therapeutic modulation of the immune system to treat a variety of immune-mediated diseases.

Gram-negative bacteria are distinguished from Gram-positive bacteria by the secondary membrane that surrounds their peptidoglycan cell wall. The outer leaflet of this membrane is primarily composed of the glycolipid lipopolysaccharide (LPS), which has lipid A, core oligosaccharide, and O-antigen portions. LPS helps protect Gram-negative bacteria from hydrophobic toxins and, in pathogenic bacteria, from the host immune system. The membrane anchor portion of LPS (lipid A) is responsible for stimulation of the inflammatory response of the mammalian immune system by LPS via activation of the Toll-like receptor 4/myeloid differentiation factor 2 complex. In systemic infections, overstimulation of this receptor causes acute inflammation, which can cause septic shock. Modifications to the LPS, particularly to the lipid A portion, can help bacteria evade the host immune system by disguising the bacteria, modulating the inflammatory response, and inhibiting interactions with antimicrobial host factors. Lipid A is synthesized in the well characterized and largely conserved Raetz pathway in the cytosol and at the cytosolic face of the inner membrane. The non-repeating core oligosaccharide is synthesized on lipid A the cytoplasmic face of the inner membrane, and the repeating O-antigen polysaccharide is attached to the core-lipid A molecule at the periplasmic face of the inner membrane. The completed LPS molecules are then transported to the extracellular leaflet of the outer membrane. Structures of proteins involved in LPS synthesis have proved critical to our understanding of LPS synthesis and transport. Moreover, these structures provide targets for rational design of antibiotics targeting Gram-negative bacteria. As the Raetz pathway is the most conserved part of LPS synthesis, the enzymes of the Raetz pathway provide particularly promising targets for development of broad spectrum antibiotics, such as those needed to treat sepsis. Therefore, I studied the structures of two enzymes in the Raetz pathway (LpxH and LpxB). LpxH was crystallized with the α-helical substrate-binding cap domain in a displaced conformation, suggesting that this domain is highly mobile. The structural dynamics of this domain and their relevance to substrate binding were further explored by hydrogen-deuterium exchange mass spectrometry, molecular dynamics simulations, and activity assays. These data supported a model in which a loop in the core hydrolase domain acts as a wedge to promote opening of the capping helices and allow facile substrate binding between these helices. In addition, the first structure of LpxB was determined showing a Glycosyltransferase B superfamily (GT-B) fold modified by the formation of a novel C-terminally swapped dimer wherein the last 87 residues of one subunit complete the GT-B fold of the other subunit. Furthermore, the binding site of the sugar-donor substrate was identified by a structure of LpxB with the UDP product bound. Activity assays supported the formation of this C-terminally swapped dimer in solution and showed that a surface-exposed hydrophobic patch is critical for LpxB activity, which suggested this patch allows productive membrane association required for substrate binding. Thus, the present research has expanded our understanding of two enzymes important in Gram-negative bacterial physiology that are potential targets for antibiotic development.

Enterobacteriaceae strains producing the Klebsiella pneumoniae carbapenemase (KPC) have disseminated worldwide, causing an urgent threat to public health. KPC-producing strains often exhibit low-level carbapenem resistance, which may be missed by automated clinical detection systems. In these studies, eight Klebsiella pneumoniae strains, one Enterobacter cloacae strain and one Escherichia coli strain with heterogeneous resistance to imipenem were used to elucidate the factors leading from imipenem susceptibility (1-2ug/ml) to high-level resistance (>32ug/ml) as determined by clinical laboratory testing standards. These strains had highly similar heteroresistance phenotypes, yet were genetically diverse in their plasmid content other than carriage of the blaKPC-containing Tn4401. We showed that the heteroresistant strains had effective KPC-mediated inactivation of low levels of imipenem, combined with reduced or absent expression of ompK35, but as the imipenem concentration increased to 8- and 16-fold higher than the MIC, a majority (>99%) of the population was killed. Time-kill analysis with an inoculum as low as 3x106 cfu/ml showed that full recovery of the population occurred by 20 hours of incubation in the same drug concentrations. We showed that it was not the density of the cultures per se, but cells with a distinct physiology, present at a frequency of 2x10-7 to 3x10-6 in starting cultures, and selected by lethal concentrations of imipenem, that coordinate population recovery subsequent to the killing of the majority of the initial cells. Samples selected 2 hours after exposure to imipenem were as susceptible as the unexposed parental strain and produced the major outer membrane porin OmpK36. However, between 4 to 8 hours after exposure, OmpK36 became absent and the imipenem MIC increased at least 32-fold. Individual colonies isolated from cultures after 20 hours of exposure revealed both susceptible and resistant subpopulations. We hypothesize that these heterogeneous populations arose from the small population of cells that initially survived imipenem-mediated killing. There were two types of OmpK36 production among the K. pneumoniae heteroresistant strains. The majority (6 of 8) of the strains permanently abolished OmpK36 upon lethal imipenem exposure due to mobile insertion element interruptions in the coding region of ompK36. High-level imipenem resistance was maintained and OmpK36 remained absent even without continued carbapenem exposure. Two strains reverted to the heteroresistance phenotype and resumed production of OmpK36 once imipenem exposure was removed. Through transposon mutagenesis of a reverter-type heteroresistant K. pneumoniae strain, we showed that acquisition of specific nutrients was essential to abolish OmpK36 production and for population recovery. Addition of the KPC enzyme inhibitor phenylboronic acid (PBA) at any point during imipenem exposure prior to loss of OmpK36 inhibited subsequent OmpK36 loss and prevented population recovery, showing the essentiality of blaKPC in the expression of heteroresistance. These studies demonstrated the coordination between bacterial physiology, blaKPC and ompK36 expression that led to the rapid induction of high-level imipenem resistance from a population of bacteria that initially exhibited a carbapenem-susceptibility phenotype. The finding of specific nutritional requirements for full expression of heteroresistance highlights potential therapeutic avenues of exploration for these strains that pose an urgent and increasing threat to public health.

Brucella melitensis Rev 1 is the best vaccine available for the prophylaxis of small ruminant brucellosis and, indirectly, for reducing human brucellosis. However, Rev 1 shows anomalously high rates of spontaneous dissociation from smooth (S) to rough (R) bacteria, the latter being inefficacious as vaccines. This S-R instability results from the loss of the O-polysaccharide. To overcome this problem, we investigated whether some recently described mechanisms promoting mutations in O-polysaccharide genes were involved in Rev 1 S-R dissociation. We found that a proportion of Rev 1 R mutants result from genome rearrangements affecting the wbo O-polysaccharide loci of genomic island GI-2 and the wbkA O-polysaccharide glycosyltransferase gene of the wbk region. Accordingly, we mutated the GI-2 int gene and the wbk IS transposase involved in those arrangements, and found that these Rev 1 mutants maintained the S phenotype and showed lower dissociation levels. Combining these two mutations resulted in a strain (Rev 2) displaying a 95% decrease in dissociation with respect to parental Rev 1 under conditions promoting dissociation. Rev 2 did not differ from Rev 1 in the characteristics used in Rev 1 typing (growth rate, colonial size, reactivity with O-polysaccharide antibodies, phage, dye and antibiotic susceptibility). Moreover, Rev 2 and Rev 1 showed similar attenuation and afforded similar protection in the mouse model of brucellosis vaccines. We conclude that mutations targeting genes and DNA sequences involved in spontaneous O-polysaccharide loss enhance the stability of a critical vaccine phenotype and complement the empirical stabilization precautions taken during S Brucella vaccine production.

Abstract: The Cpx two‐component signal transduction pathway of Escherichia coli consists of an inner membrane‐localized sensor histidine kinase, CpxA, the response regulator, CpxR, and the novel periplasmic accessory protein, CpxP. These proteins mediate a bacterial stress response, sensing envelope perturbations including damage caused by misfolded periplasmic or inner membrane proteins, and regulating the expression of a host of factors that contribute to preservation of the envelope. While structural information exists for a number of bacterial two‐component systems, there is limited data to describe the constituents of the Cpx system. Since CpxP has no homologues of known function, we initially focused on its biophysical and structural characterization. Biochemical studies demonstrated that CpxP maintains a dimeric state, but may undergo a slight structural adjustment in response to the Cpx pathway inducing cue, alkaline pH. The crystal structure of CpxP, determined to 2.85 Å resolution, revealed an antiparallel dimer of intertwined α‐helices with a highly basic concave surface. We identified that the core fold adopted by CpxP is also found in a number of other periplasmic stress response proteins. Additionally, we proposed that the conserved LTXXQ motifs that define a family of proteins have a structural role in the formation of diverging turns. Finally, we identified several sites, including two solvent‐exposed residues and the charged surfaces of CpxP, which are likely involved in intermolecular interactions. In an effort to understand the molecular mechanism(s) by which CpxA detects specific inducing signals, we set out to describe the uncharacterized periplasmic sensor domain of CpxA biochemically and structurally. We demonstrated that the isolated sensor domain likely exists in a unique tetramer‐dimer equilibrium. Further, we suggested that the periplasmic domain of CpxA forms a PAS‐like PDC fold, and identified regions that could possibly be involved in a structural rearrangement upon stimulus perception. Lastly, we established preliminary crystallization conditions that will be optimized for high‐resolution structure determination. Ultimately, these studies will provide insight into the molecular characteristics of CpxA and CpxP that contribute to the regulation of the response to envelope stress in pathogenic bacteria such as E. coli.

There have always been reported cases of bacteria able to resist antibiotic treatment. Through recent history, infections with bacteria capable of resisting at least one type of drug treatment have become more commonplace [1,2,3]. The dearth of novel-acting antibiotics, combined with the resourcefulness of pathogenic bacteria under selective pressure means that humans are now faced with strains of bacteria that have developed resistance to multiple drugs. Of the few novel acting antibiotics developed in the last 30 years, most are targeted only to Gram positive bacteria, and reports of bacteria demonstrating resistance to these drugs have already surfaced [4,5,6]. The vast difference in the number of drugs available for treatment of Gram positive versus Gram negative bacterial infections is a direct reflection of the difficulty in targeting Gram negative bacteria with drugs. Drugs which treat Gram positive infections are often superfluous against Gram negative bacteria in large part due to the extra outer cell membrane these bacteria possess, which acts as an additional physical barrier and contains drug efflux pumps, ABC transporters and selective porins [7]. The structural integrity of the outer memberane structure has been shown to be essential to the viabililty of pathogenic Gram negative bacteria, which makes it an attractive target for the development of novel acting antibiotics [8].

TolC is a versatile, multifunctional protein and in Gram-negative bacteria such as E.coli, it has been shown to participate in several functions like drug efflux, toxin secretion and has also been demonstrated to be important for survival in iron limiting conditions. This multifunctional nature of TolC makes it an attractive target for drug design and therefore it becomes imperative to understand the mechanism of TolC's multifunctional nature.

Lipopolysaccharide (LPS), a glycolipid found in the outer membrane of Gram negative bacteria, induces the secretion of pro-inflammatory cytokines such as tumor necrosis factor alpha (TNF-a) and interleukin (IL )-1, by monocytes/macrophages. Excessive and uncontrolled secretion of these compounds leads to multiple pathological conditions, such as septic shock. LPS receptors have been shown to be CD14, TLR4 and MD-2. LPS interaction with these receptors mediates many monocyte/macrophage functions. Even though only CD14 was demonstrated to bind to LPS, and TLR4/MD-2 were capable of transducing signals, data only show that LPS and CD 14 were in close proximity to TLR4 and no direct binding was reported. Quite recently, moesin, a member of the ERM family of proteins, has been also found to function as a receptor for LPS. We have shown that anti-moesin antibody inhibited the release of TNFa by LPS stimulated monocytes. Moesin was also found to be necessary for the detection of LPS, where homozygous knockout mice exhibited 3-fold reduction in neutrophil infiltrates in LPS injected sites when compared to their wild type controls. When moesin gene expression was completely suppressed with antisense oligonucleotides, there was a significant reduction of LPS-induced TNF-a secretion. LPS stimulation of mononuclear phagocytes activates several intracellular signaling pathways including the phosphorylation of IKBa, mitogen-activated protein kinase (MAPK) pathways: extracellular signal-regulated kinases (ERK) 1 / 2 (P44/42), p38. These signaling pathways in tum activate a variety of transcription factors including NF-KB, which coordinates the induction of several genes encoding inflammatory mediators. [TRUNCATED]

The urease enzyme of the gastric pathogen Helicobacter pylori catalyzes the degradation of urea into ammonia and carbon dioxide. Urease is believed to protect the bacterium from gastric acid. Urease activity of H. pylori is essential for colonization of the stomach. An objective of this dissertation was to correlate the level of urease expression with loss of colonization of H. pylori isolated from infected piglets and subjected to rounds of in vitro passage. Using slot blot and reverse transcription-polymerase chain reaction (RT-PCR) in a thermocycler with real-time analysis, urease RNA was determined to be constitutively expressed regardless of the number of in vitro passages. Determination of urease activity of the isolates by the phenol-hypochlorite assay confirmed the RNA data. Lewis x (Lex) and Lewis y (Ley) antigens of Helicobacter LPS undergo phase variation during in vitro growth due to slip-strand mispairing of polycytidine (poly(C)) tracts in the alpha-3-fucosyltransferase genes, futA and futB, that specify the Lewis antigen phenotype. Hence, another objective of this dissertation is to evaluate whether the on/off status of futA and futB via slip-strand mispairing determines Lex and Ley phenotype. The lengths of the poly(C) regions were analyzed by PCR. The Lewis phenotypes of the samples were determined by immunoblots and enzyme-linked immunoabsorbent assay (ELISA) using anti-Lex and anti-Ley antibodies. This study failed to correlate futA/B expression with Lex expression. Because both futB and Ley were expressed in all strains, no relationship could be determined. However, futB is sufficient for Lewis x/y synthesis. The final objective of this dissertation was to evaluate the ability of H. pylori LPS and its O-antigen to modulate the CD4+-dependent host immune response to H. pylori. This was performed by culturing splenocytes from uninfected and SS1-infected C57Bl/6 mice and T-bet-deficient mice. The cultures were stimulated with sonicate preparations and LPS samples from the parental and O-antigen-deficient bacterial strains. The data from this study suggests that Lewis antigens cannot alone account for the strong stimulation of the immune system and may even down-regulate the inflammatory response to H. pylori based on the higher stimulation of interferon-ã of SS1:0826 LPS compared to SS1 LPS.

This study describes the biochemical activity and antimicrobial susceptibility patterns (antibiograms) of 104 gram-negative bacteria, represented by six species of bacilli, isolated from ten macaws of the genus Ara. Bacterial samples were acquired from fecal matter of six different species of macaws, ages one to three years, housed in a variety of locations. Bacteria from these samples were cultured onto selective media for classification. Identification and biochemical characterization were accomplished with the API 20E system. Escherichia (E.) coli accounted for 78% (81/1 04) of the total number of isolates. Antimicrobial susceptibility testing was done using the Kirby-Bauer method. Biotype and antimicrobial susceptibility data gathered in this study correspond with data of studies done on pigeons and llamas. These data show specific E. coli biotypes as members of the microflora of macaws.

A novel Gram negative, capsule-forming bacterium was previously isolated in Dr. G. Roland Vela's laboratory. The distribution of this bacterium in soils from various locations was investigated. Soil samples from 188 locations around the world were examined. Isolates of the bacterium were obtained from 50 of these soils, with 48 of the isolates found in soils from the southwestern United States and northern Mexico. This suggests that this region is the natural habitat of the bacterium. The other two isolates were obtained from Madrid, Spain and Taipei, Taiwan. None were found in soils from South America or Australia. A lack of variation in morphology and physiological properties in the isolates suggests that a homogeneous population exists, even from widespread geographical locations.

This study deals with the nature and characteristics of the Gram negative bacterial flora of the intestine of the domestic pigeon. Columba livia was selected for this study because of the abundance of this species around the Classroom building on the University of the Pacific campus.