There has recently been a shift in water treatment practices in Canada to a source-to-tap approach on water quality. With this approach comes the understanding that our water distribution systems are covered in biofilms which could contain potentially hazardous opportunistic pathogens. Meanwhile, regulations on control of disinfection byproducts in drinking water have also induced the phase-out of chlorine and its replacement with either chlorine dioxide or monochloramine in conjunction with UV treatment. These changes are based solely on the reduction of known disinfection byproducts and have not considered the impact of changing the disinfection regime on the ecology of biofilms within the distribution system. The objective of this study was to determine what, if any, influence changes in water disinfection would have on the mix of bacterial species in the distribution system biofilms with particular reference to cast iron and polycarbonate as pipe substrata. Classical culture-based methods can reveal only a fraction of the bacterial content of biofilms because of our rudimentary understanding of the nutritional requirements of the organisms present and their inter-dependency. In contrast, newer techniques in molecular biology have become the norm for studying microbial ecology as they are not subject to the limitations of the culture methods, and thus can provide a much better profile of bacterial populations in biofilms. Although these methods have their own biases, PCR-DGGE was selected to monitor any changes in the profiles of the biofilms obtained under different disinfectant regimes and identify similarities and differences. Identification of the bacterial species would then be obtained by the sequencing of cloned bands, and matching them to the online databases, BLAST and RDP II. Clear differences were observed in the biofilms from the two pipe materials tested. Aquabacterium parvum, Escherichia coli, Dechloromonas sp., Methylobacillus flagellatus, Phyllobacterium sp., Rhodocyclus sp., and Sphingomonas sp., were only identified in biofilms from cast iron coupons, while Chitinophaga sp., was found in biofilms from only polycarbonate coupons. This confirms that the pipe material can influence the types of organisms growing on its surface. In general, the bacterial profiles were similar in the presence or absence of upstream UV treatment, except for the uncultivable Flavobacterium spp., which was detected only in the absence of UV treatment. This indicated that UV treatment has a relatively minor impact on altering the biofilm composition. While a direct comparison between the impacts of the chemical disinfectants was not possible due to the design of the experimental set-up, the data obtained showed that several aspects of the bacterial profiles remained similar irrespective of the dosage levels of chlorine, chlorine dioxide, and monochloramine used.

Background. The sepsis syndrome is characterized by such clinical changes as hypotension, hypothermia or fever, metabolic acidosis, pulmonary hemorrhage, and death within several hours to days. Antibiotics are a mainstay of treatment and have been repeatedly demonstrated to improve survival in both human studies and animal models of sepsis. Appropriate fluid resuscitation is an important component of the initial therapy for severe sepsis. The antibody against TNF-alpha has improved survival in certain models of sepsis (Reinhart et al. 2001). Hypothesis. Timing of antibiotics is pivotal in survival outcome of sepsis and early fluid resuscitation is crucial in the maintenance of survival rates. Materials & methods. Male Balb/c mice (23--25g) received the cecal ligation and puncture. During the surgery, 1/3 cecum was ligated and 2 18G needle punctures were made from which feces were expressed. During antibiotic treatment strategies, cefotaxime was firstly given at different time points (0, 3, 6 & 12h post-surgery) and thereafter every 6 hours for up to 72 hours. All the mice were given 1ml 0.9% saline fluid at 0 hour post-surgery and continuously every 6 hours for up to 72 hours. For fluid resuscitation 1ml 0.9% saline fluid was started at 0, 3, 6, 9 or 12 hours post-surgery respectively by subcutaneous injection and continuously every 6 hours for up to 72 hours. All the mice were given cefotaxime at 0 hour post-surgery and continuously every 6 hours. Finally, we repeated cefotaxime 12 hours (Group 1) and fluid 12 hours post-surgery (Group 2) groups. Anti-TNF-alpha monoclonal antibody was given at 6 hours post-surgery for both groups. Results. After surgery, untreated mice showed signs of sepsis, such as anorexia, hypothermia and dehydration, and the mortality was 100% within the first 30 hours. The mortalities in cefotaxime 6 & 12 hours post-surgery groups were nearly 100% while cefotaxime 0 & 3 hours post-surgery groups were 0% and 25% respectively. The mortality in fluid resuscitation 12 hours post-surgery group was nearly 100% whereas the other groups (0, 3, 6 & 9 hours) were 12.5% respectively. The mortality of group 1 decreased significantly (Fisher's exact test) compared to the same treatment group without anti-TNF-alpha while the mortality of group 2 did not. Discussion. This model provides consistent results of sepsis, which could enable us to control for treatment factors that are not controllable in human sepsis. In this model we can clearly show the impact of timing of antibiotics on survival. Despite early antibiotics administration (time 0), mortality of this model is still high when fluid resuscitation is delayed. Anti-TNF-alpha can decrease the mortality under condition that fluid resuscitation are given at an early time point.

Pseudomonas aeruginosa is an opportunistic pathogen prevalent in nosocomial infections and patients with cystic fibrosis. P. aeruginosa shows a high degree of antibiotic tolerance which in part can be attributed to the formation of biofilms. The increased antibiotic resistance seen in biofilms can be attributed to several factors including differential gene expression within biofilms that can lead to biofilm specific mechanisims of antibiotic resistance. The ndvB gene is important for biofilm specific antibiotic resistance in P. aeruginosa. It is important in signaling and regulation of gene expression in other pathogens. Microarray analysis comparing gene expression between wildtype and a ndvB deletion mutant was performed to identify genes that might be regulated by the gene product, believed to be cyclic glucans, that contribute to biofilm specific antibiotic resistance. The array analysis identified 24 genes that were differentially regulated by ndvB, including a response regulator, agmR, as well as most of the genes which it regulates. Quantitative real-time polymerase chain reactions using primers specific to agmR and its associated genes confirmed that they were expressed in a ndvB related manner. Minimal bactericidal concentration assays were performed and confirmed that these genes are important in resistance to tobramycin and ciprofloxacin.

Listeria monocytogenes is a Gram-positive, facultative intracellular pathogen that causes gastroenteritis, which in the young, the elderly, and the immunocompromised can progress to severe invasive disease with high mortality. While previous studies have largely elucidated the bacterial and host mechanisms necessary for the bacterium to access its replicative niche in the host cell cytosol, the L. monocytogenes factors required for adaptation to life within this restrictive environment are poorly understood. In this dissertation, I describe a fluorescence-activated cell sorting (FACS)-based differential fluorescence genetic screening technique for the identification of L. monocytogenes genes necessary for optimal intracellular replication. Bacteria harboring deletions in identified genes were defective for intracellular replication, plaque formation, and in vivo virulence, validating the ability of the screening method to identify novel intracellular replication-defective mutants. Minor alteration of the FACS-based screening strategy allowed the detection and differentiation of bacterial mutants displaying varying severities of actin-based motility defects. A preliminary FACS-based genetic screen to identify actin-based motility mutants isolated multiple independent insertions within internal control genes, demonstrating the potential utility of FACS-based differential fluorescence genetic screening methods for the identification of L. monocytogenes genes important for multiple virulence phenotypes. Lastly, my characterization of the X-prolyl aminopeptidase PepP, a novel virulence factor identified by the FACS-based genetic screen to discover genes necessary for optimal intracellular replication, revealed this enzyme plays an unexpected role in L. monocytogenes virulence gene regulation.

Bacterial biofilm formation, the construction of dense, protective, multicellular communities, is a widely conserved behavior. In some bacteria, such as the Gram-positive model organism Bacillus subtilis, the genetics controlling biofilm formation are well understood. In other bacteria, however, including the Gram-negative opportunistic pathogen Pseudomonas aeruginosa, the identities or roles of many biofilm genes remain unknown. Importantly, many proposed applications of biofilm research, particularly in the medical field, require knowledge not only of biofilm assembly but also of biofilm disassembly, the latter being a recent and underdeveloped area of study. It was previously reported that B. subtilis biofilms disassemble late in their life cycle due to the incorporation of four D-amino acids (D-leucine, D-methionine, D-tryptophan, and D-tyrosine, or D-LMWY) into peptidoglycan. It was further argued that D-LMWY specifically inhibits and disassembles the biofilms of diverse bacterial species, including B. subtilis and P. aeruginosa. Here I present a contrasting report. I describe how what had been perceived as D-LMWY-mediated biofilm inhibition is actually D-tyrosine-mediated toxicity. B. subtilis is sensitive to growth inhibition by D-tyrosine due to the absence of D-tyrosyl tRNATyr deacylase (Dtd), an enzyme that prevents the misincorporation of D-tyrosine and other D-amino acids into nascent proteins. By repairing the gene for Dtd, I was able to render B. subtilis resistant to both growth inhibition and biofilm inhibition by D-tyrosine and D-LMWY. In parallel, I recovered spontaneous mutants of B. subtilis that survive in the presence of D-LMWY. These isolates harbored mutations in pathways that regulate tRNATyr charging. Three of these mutations enhanced the expression of the gene (tyrS) for tyrosyl-tRNATyr synthetase (TyrRS), while a separate mutation improved the stereoselectivity of TyrRS. I concluded that these spontaneous D-LMWY resistance mutations were compensating for the absence of Dtd. In addition to my research on B. subtilis biofilm regulation, I demonstrated a new, non-destructive screening approach for identifying P. aeruginosa biofilm genes. Using this screen, I was able to recover a wide range of known biofilm genes as well as the new biofilm gene candidates ptsP, PA14_16550, and PA14_69700. These three genes are the focus of an ongoing study dedicated to characterizing P. aeruginosa biofilm formation, particularly as it relates to the secondary messenger cyclic di-GMP. In summary, this dissertation covers aspects of biofilm formation and dispersal in two bacterial species. My work offers mechanistic insight into D-amino acid resistance, resolves the relationship between D-amino acids and biofilms, and establishes a new tool for understanding the complexities of biofilm genetics and regulation. / Biology, Molecular and Cellular

With eight genomic segments encoding at least thirteen proteins, influenza virus can subvert a cell into a virus-producing factory. To study influenza infection, I utilize versatile fluorescent-based technologies to non-invasively probe a range of biological processes and targets in both static and living systems. This thesis covers three broad areas aimed towards unraveling the complexities of influenza infection: understanding how host proteins regulate viral infection, developing assays to study infection heterogeneity, and applying live cell imaging to study antiviral mechanism of action. First, I discuss how two cellular proteins—COPI complex and CD81—facilitate influenza infection. Genome-wide knockdown screens identified COPI complex and CD81 as influenza host dependency factors, but their specific function remained unclear. Applying imaging and flow cytometry methods, I found COPI siRNA knockdown inhibited virus entry during internalization and transport to late endosomes, and late stage infection during viral membrane protein trafficking. However, acute pharmacological treatment only recapitulated membrane protein trafficking defects, suggesting COPI directly facilitates late stage infection, not entry. In contrast, CD81 facilitates both viral fusion during entry and scission during egress. Single particle tracking studies revealed ~50% of virus particle fusion events occurred within CD81+ endosomes, and CD81 is recruited to progeny virus budding zones to facilitate viral scission. Second, while influenza infection encompasses a complex mixture of incomplete infection events, no influenza infection assay has thus far comprehensively evaluated infection heterogeneity. To study the prevalence and composition of incomplete infection events, I developed a multiplexed immunofluorescence assay to probe viral protein expression from all eight genomic segments. I found that influenza infection heterogeneity is highly prevalent, and the composition of different infection states exhibits correlated expression patterns amongst a subset of viral proteins. Lastly, I include a study applying live cell imaging to dissect the mechanism of a newly identified influenza antiviral drug, clotrimazole. I found clotrimazole inhibited WSN influenza infection between viral fusion and replication. Altogether, a multi-pronged approach is required to study the complex influenza infection cycle. Deciphering the complexities will guide development of much needed influenza therapeutics and vaccines. / Medical Sciences

Vibrionaceae family members are interesting models for studying DNA replication initiation as they contain two circular chromosomes. Chromosome II (chrII) replication is governed by two evolutionarily unique, yet highly conserved elements: the origin DNA sequence, oriCII, and the initiator protein, RctB. The minimum functional region of oriCII, oriCII-min, contains multiple elements that are bound by RctB in vitro, but little is known about the specific requirements for individual elements during oriCII initiation. We utilized undirected and site-specific mutagenesis to investigate the functionality of oriCII-min mutants, and assessed binding to various mutants by RctB by electrophoretic mobility shift assays. Our analyses showed that deletions, point mutations, and changes in RctB target site spacing or methylation all impaired oriCII-min-based replication. RctB displayed reduced affinity for most low efficacy origins tested, although its characteristic cooperative binding was generally maintained. Mutations that removed or altered the relative positions of origin components other than RctB binding sites (e.g., AT-rich region, DnaA target site) also abolished replicative capacity. A comprehensive mutagenesis and deep-sequencing-based screen (OriSeq) allowed identification of a previously uncharacterized methylated domain in oriCII that is required for origin function. We also examined RctB-oriCII interactions with DNaseI footprinting, pelleting, and electron microscopy. Along with the binding studies, these data suggest that RctB is able to generate large, oligomeric structures on DNA, which are sequence independent at high proteins concentrations. Together, our results reveal the remarkable evolutionary honing of oriCII and provide new insight into the complex interplay between RctB and oriCII. / Medical Sciences

The Staphylococcus aureus cell envelope is a large, complex structure essential for cell shape and protection from the environment. It consists of membrane lipids, glycan polymers, and proteins, many of which have not been characterized despite decades of research. One method for elucidating biological function is to identify synthetic lethal interactions. Specific chemical inhibitors of known proteins are powerful tools in these studies. Here I describe a systematic approach to identify small molecule inhibitors useful for synthetic lethal interaction mapping. I first probe a transposon mutant library with an inhibitor of wall teichoic acid (WTA) biosynthesis and perform transposon insertion sequencing (Tn-seq) to refine a growing network of genetic interactions centered on WTAs. I next carried out a whole-cell pathway-directed high-throughput chemical screen for inhibitors of proteins within the WTA interaction network based on differential growth inhibition of WTA-deficient cells versus wild-type. To identify hits, I developed a flexible method of analysis that ranks hit compounds by likelihood of being a true positive. Through this screen I found a direct relationship between teichoic acid D-alanylation and permeability to positively charged antibiotics such as aminoglycosides. I also identified amsacrine as an inhibitor of WTA-deficient strains and established its target as the D-alanyl aceyltransferase DltB. I showed that amsacrine phenocopies a D-alanylation-deficient strain in sensitizing S. aureus to aminoglycosides and preventing biofilm formation. My research represents the first iteration of a “discovery cycle” for using a specific small molecule probe to identify genetic interactions, then exploiting those genetic interactions to perform a chemical screen that identifies additional probes. This work describes a rapid and adaptable method for exploring the complex interactions within cell wall biosynthesis. / Medical Sciences

Francisella tularensis is a Gram-negative, intracellular pathogen and the causative agent of tularemia. Due to its low infectious dose, ability to cause potentially fatal disease, and ability to be easily aerosolized, several countries have developed F. tularensis as a potential bioweapon. Three proteins, MglA, SspA, and PigR, and the small molecule guanosine tetraphosphate (ppGpp), are transcription factors critical for the virulence of this organism. These regulators function coordinately to positively regulate the expression of genes present on the Francisella pathogenicity island, as well as many other genes that are required for the virulence of this organism. MglA and SspA form a complex that associates with RNA polymerase (RNAP); the interaction between the MglA-SspA complex and RNAP is thought to be critical for MglA and SspA to regulate gene expression. PigR, a putative DNA-binding protein, associates with the RNAP-associated MglA-SspA complex and may stabilize the binding of RNAP at regulated promoters. The interaction between the MglA-SspA complex and PigR in F. tularensis has been shown to be promoted by ppGpp. A direct interaction between the MglA-SspA complex and PigR had previously been found using a modified version of an E. coli two-hybrid assay, referred to as the bridge-hybrid assay, that permits the detection of interactions between a protein of interest and a protein complex. However, the role of this direct interaction in controlling gene expression in F. tularensis had not been investigated. Conflicting reports in the literature over the ability of PigR to interact with the MglA-SspA complex led to differing models of how PigR regulates virulence gene expression in F. tularensis. To address the importance of the interaction between the MglA-SspA complex and PigR in regulating gene expression, we used a combination of genetic approaches to identify mutants of either MglA or SspA that are specifically defective for interaction with PigR. The identified mutants of MglA and SspA were unable to functionally substitute for MglA or SspA, respectively, and were unable to promote expression of MglA- and SspA- regulated genes in F. tularensis. These results indicate that the interaction between the MglA-SspA complex and PigR is critical for expression of virulence genes in F. tularensis. Our work also identified a surface on the MglA-SspA complex that is important for the interaction with PigR and which may constitute a binding site for PigR. The small molecule ppGpp has previously been shown to promote the interaction between the MglA-SspA complex and PigR in F. tularensis. It is unknown if ppGpp directly or indirectly promotes this interaction. We determined that ppGpp is required to detect an interaction between the MglA-SspA complex and PigR in the E. coli bridge-hybrid assay, indicating that ppGpp is either directly involved in promoting this interaction or works through an indirect mechanism that is conserved between F. tularensis and E. coli. One potential conserved mechanism through which ppGpp may be influencing the interaction between the MglA-SspA complex and PigR is through regulation of the levels of the molecule polyphosphate. However, we determined that polyphosphate is not required in order for the MglA-SspA complex and PigR to detectably interact with one another in the E. coli bridge-hybrid assay. Furthermore, analysis of the role of polyphosphate in gene expression in F. tularensis revealed that polyphosphate is a negative regulator of virulence gene expression. / Medical Sciences

Antibiotics are often credited with being one of the major forces behind the expansion of human life expectancy in the past 60 years. Yet at the root of this advancement lies its potential undoing: using antibiotics promotes the emergence and spread of resistant strains, reducing the efficacy of the drugs. Now, rising antibiotic resistance threatens to undo much of the progress of modern medicine. To halt the rise of resistance and preserve the activity of antibiotics, we must find ways to neutralize, modulate, or even invert the evolutionary advantage of resistant strains. Chapter 1 reviews three strategies to overcome antibiotic resistance through the sequential or concurrent use of multiple drugs: resistance mechanism inhibitors, synergistic, antagonistic, and suppressive drug interactions, and collateral sensitivity. Collateral sensitivity occurs when a bacterium acquires a mutation or gene that provides resistance to one drug, but makes them more susceptible to others. This new vulnerability can therefore be exploited to select against resistance mechanisms. Chapter 2 describes a screening strategy, based on direct competition between antibiotic resistant and susceptible strains, for identifying compounds that select against antibiotic resistance genes and cassettes. Using this approach we identified two compounds—β-thujaplicin and disulfiram—that select against the TetA tetracycline resistance pump in E. coli. Furthermore, we demonstrate a two-phase treatment paradigm in which β-thujaplicin drives a tetracycline resistant population back to susceptibility, allowing successful second-phase treatment with tetracycline. Chapter 3 examines the consequences of linking two antibiotics—ciprofloxacin and neomycin—into one hybrid compound. We compared the cross-resistance and genotypic profiles of strains evolved in the hybrid to those evolved in mixtures of its two components. We find that the hybrid inhibits bacterial growth through its ciprofloxacin moiety, but prevents resistance through its neomycin moiety by avoiding a common multiple antibiotic resistance pathway. As a result, strains evolved in the hybrid gain less resistance than those evolved in an unlinked mixture. This indicates that linking two drugs can surpass traditional unlinked combination therapy in its ability to prevent resistance. Finally, Chapter 4 discusses the implications of this work and possible directions for future research in treating antibiotic resistance. / Chemical Biology