Cyclic-di-GMP, a common bacterial second messenger, has been thought to help develop virulence and biofilms in bacteria, most specifically in Salmonella Typhimurium. By being able to dysregulate cyclic-di-GMP production, virulence may be better combatted. STM1987, an L-arginine-responsive diguanylate cyclase with a periplasmic sensory domain, dimerizes and generates the bacterial second messenger cyclic-di-GMP in response to the amino acid L-arginine in a pathway that also requires the periplasmic L-arginine-binding protein ArtI. Their biochemical responses to L-arginine and when they dimerize could help clarify this pathway, so I sought to develop a periplasmic dimerization sensor to better monitor these biochemical interactions. Similar to STM1987, the ToxR transcriptional regulator from Vibrio cholera is also activated by dimerization. By switching out the periplasmic domain of ToxR for the periplasmic regions of interest, I can better evaluate the cyclic-di-GMP response to L-arginine. This research aims to find the specific responses in this pathway to be able to use this in combatting bacterial virulence. I was able to successfully show that the STM1987 periplasmic domain dimerizes in response to L-arginine, providing an important insight into this signaling pathway.

Abstract Acinetobacter baumannii is a prevalent nosocomial where infections are typically secondary infections to patients that already have an infection or other source of being immunocompromised. Like many other infectious bacteria, A. baumannii is increasingly considered a multi-drug resistant pathogen. This eliminates the ability to treat A. baumannii infections with traditional antibiotics, hence the need for another method of treating A. baumannii. This research study was designed to find a way to affect the survival of A. baumannii such that it can be applied to a hospital setting to prevent further infections to immunocompromised patients. One mechanism potentially used by A. baumannii to persist on hospital surfaces is through the use of the bacterial second messenger cyclic-di-GMP (c-di-GMP). This nucleotide signal is regulated in response to environmental conditions, and then activates c-di-GMP-binding proteins that induce phenotypic changes. I hypothesized that by deleting these c-di-GMP-binding proteins that it will produce measurable differences in phenotype like biofilm formation, motility, and desiccation survival. Reducing phenotypes such as these may alter A. baumannii’s ability to persist on hospital surfaces, and potentially lead to future surface eradication. A. baumannii encodes two potential c-di-GMP-binding proteins of particular interest, one that contains a sole PilZ domain and another that pairs a PilZ domain with a hydrolase domain. PilZ domains bind c-di-GMP within a conserved binding site, regulating the conformational structure of the protein, and are named for the first studied PilZ domain within the pilus-associated PilZ protein. Pili are used in pilus-mediated motility and surface attachment, and they are A. baumannii’s primary method of motility due to not having flagellum. I hypothesized that by removing these c-di-GMP-binding proteins, I would interrupt the c-di-GMP signaling that might regulate motility. I am testing two A. baumannii strains: 5075, a recent military hospital isolate and 17978, an older lab strain. A notable difference between these two strains is that 5075 demonstrates twitching motility where it utilizes type IV pili, but 17978 demonstrates swarming motility that has unknown mechanisms. Both c-di-GMP-binding proteins were tested for their role in twitching or swarming motility of the respective strains. I found that swarming motility of 17978 is regulated by both c-di-GMP-binding proteins. While I am still generating the deletion strain for the c-di-GMP-binding hydrolase enzyme, the sole PilZ domain protein is also required for twitching motility in the 5075 strain. These results suggest c-di-GMP regulates both forms of motility in A. baumannii. Future plans include determining the role of the c-di-GMP-binding hydrolase enzyme in twitching motility and identifying the role that these proteins play through binding of c-di-GMP.

Cyclic di-GMP is a second messenger used by bacteria to regulate motility, extracellular polysaccharide production, and the cell cycle. Recent advances in the measurement of real time cyclic di-GMP levels in single cells have uncovered significant dynamic heterogeneity of second messenger concentrations within bacterial populations. This heterogeneity results in a wide range of phenotypic outcomes within a single population, providing the potential for population survival and adaptability in response to rapidly changing environments. In this chapter, we discuss some of the measurement technologies available for single-cell measurement of cyclic di-GMP concentrations, the resulting discovery of heterogeneous cyclic di-GMP populations, the mechanisms bacteria use to generate this heterogeneity, and the biochemical and functional consequences of heterogeneity on cyclic di-GMP effector binding and the bacterial population.

Lyme disease is the most common tick-borne disease in North America, with approximately 35,000 cases reported to the Centers for Disease Control in 2008. The genome of its causative agent, Borrelia burgdorferi, encodes for a set of genes involved in the metabolism and regulatory activities of the second messenger nucleotide, cyclic-di-GMP (c-di-GMP). Rrp1 is a response regulatory-diguanylate cyclase, and its regulatory capability is likely mediated via production of c-di-GMP, as it lacks a DNA-binding domain. One known class of c-di-GMP effector/binding proteins are those that harbor a PIlZ domain. The genome of B. burgdorferi strain 5A4 encodes for one chromosomally-carried PilZ domain, which we have designated PlzA. Additionally, certain B. burgdorferi strains encode for a second PilZ domain-containing protein (PlzB) which is plasmid-carried. Both PlzA and PlzB were found to bind specifically to c-di-GMP, and c-di-GMP binding by PlzA was found to be dependant upon arginine residues in the c-di-GMP binding region. Additionally, expression of PlzA was found to be upregulated by tick feeding and was constitutive in the mammalian host. We next constructed two deletion/allelic exchange mutants – one with the targeted deletion of PlzA, and on ethat replaced PlzA with PlzB in a strain lacking the plzB gene. Our studies demonstrated that ΔplzA was deficient in motility and was also non-infectious in the mouse model of B. burgdorferi infection. Additionally, this strain remained viable in larval Ixodes ticks. Also, B31-plzB KI was deficient in motility, as well as infectivity, demonstrating that PlzB is unable to complement for functions fo PlzA in vitro and in vivo and that it may play other roles in the biology of B. burgdorferi strains carrying the plzB gene. These studies represent the first identification of a c-di-GMP binding protein in any spirochete, but also represent the first demonstration of the importance of PilZ domain proteins in a spirochetal system. We additionally examined the effects of c-di-GMP synthesis and breakdown in the related bacterium, B. hermsii, a causative agent of tick-borne relapsing fever (TBRF). Deletion mutants in Rrp1 (B. hermsii’s sole diguanylate cyclase) and PdeA (B. hermsii’s only EAL domain-containing phosphodiesterase) were created. These strains were analyzed in order to determine: 1) the effect(s) of the losse of Rrp1/PdeA on intracellular spirochete c-di-GMP levels, and 2) the effects of Rrp1/PdeA on the establishment of murine infection and on gross motility/chemotaxis. It was demonstrated that c-di-GMP accumulates intracellularly in the cells lacking PdeA. Additionally, spirochetes were shown to chemotax towards N-acetyl-glucosamine (NAG) and they did not form soft agar swarms. In contrast, cells lacking Rrp1 did not accumulate detectable levels of c-di-GMP, demonstrated a reduced ability to chemotax towards NAG, and swarmed on soft agar in a fashion indistinguishable from wild type. Despite these differences in phenotype, both mutant strains display an attenuated murine infectivity. These results indicate that c-di-GMP is indeed important in the TBRF spirochete, B. hermsii and this vital second messenger plays key roles in virulence, motility, and chemotaxis. These studies also pave the way for future investigation of B. hermsii through use of targeted genetic manipulation.

Lyme disease is a tick-borne infection caused by Borrelia burgdorferi, B. garinii, and B. afzelii. These spirochetes experience environmental fluctuations as they are passed between mammalian and Ixodes tick hosts throughout their enzootic cycle. Recent studies have suggested cyclic diguanylate (c-di-GMP), a ubiquitous secondary messenger, is a key modulator of B. burgdorferi adaptive responses and may play a significant role in cycle progression. In this study, we examined the impact of the sole diguanylate cyclase (Rrp1), c-di-GMP binding proteins (PlzA and PlzB), and HD-GYP-containing phosphodiesterase (PdeB) in disease establishment of both murine and Ixodes tick systems. Strains harboring targeted gene deletions or plasmid-based constitutive gene expression constructs were generated. Rrp1 was required for tick colonization, yet overexpression abolished murine disease, thus implicating the requirement of finely regulated c-di-GMP levels for enzootic cycle progression. Deletion of rrp1 disrupted translational motion and swarming patterns by causing extended cell runs, eliminating stops/flexes, and reducing swarming capabilities. This was attributed to a defect in N-acetyl-D-glucosamine (NAG) metabolism and chemotaxis. NAG is a major source of nutrition for B. burgdorferi within the tick environment; therefore this defect would impede spirochete migration towards feeding ticks, as well as pathogen uptake and survival within the Ixodes vector. In contrast, the downstream c-di-GMP effector, PlzA, was critical for murine disease but nonessential for survival within ticks nor functionally complemented by PlzB. Deletion of plzA altered strain motility and swarming similarly to the rrp1 deletion mutant, yet had a distinct phenotype with significantly slower translational motion and no affect on NAG chemotaxis and metabolism. This indicates B. burgdorferi could possess alternate c-di-GMP effectors or Rrp1 could be directly influencing these cellular processes. Uniquely, PdeB did not abolish murine infection via needle inoculation, but wasrequired for natural transmission from ticks. This defect was linked to the decreased tick colonization efficiency upon pdeB deletion. Together, these analyses indicate that c-di-GMP signaling is an important virulence mechanism of Borrelia burgdorferi and demonstrate the complexity of this signaling pathway in an arthropod-borne pathogen. The data presented here additionally provide significant new insight into the gene regulatory mechanisms of the Lyme disease spirochetes.

Pseudomonas aeruginosa is a gram-negative pathogen that causes opportunistic infections in immunocompromised individuals. Whereas clinical isolates from acute infections are characterized by host cell cytotoxicity and motility, isolates from chronic infections are characterized by biofilm formation and persistence. The type III secretion system (T3SS) causes cytotoxicity by injecting effectors into host cells. T3SS gene expression is activated by ExsA, an AraC family transcriptional regulator. Transcription of exsA is controlled by two promoters, PexsC and PexsA, which are regulated by ExsA and the cAMP-Vfr system, respectively. Additional global regulatory systems also influence T3SS including the second messenger signaling molecule c-di-GMP and the RsmAYZ regulatory system. c-di-GMP signaling increases biofilm production and decreases acute virulence factor expression. A previous study found that c-di-GMP alters cAMP levels and affect cAMP-Vfr signaling. Other studies found that c-di-GMP signaling alters expression of the small non-coding regulatory RNAs, rsmY and rsmZ. The RsmAYZ post-transcriptional regulatory system regulates ExsA translation. We hypothesize that c-di-GMP regulates T3SS expression by altering exsA transcription through the cAMP-Vfr dependent PexsA promoter. Overexpression of YfiN, a c-di-GMP synthase, decreases T3SS reporter activity in PA103 and requires a functional GGDEF active site for full inhibition. Inhibition by YfiN does not require rsmYZ. YfiN expression decreases cAMP-Vfr signaling and coordinately inhibits PexsA-lacZ reporter activity. Consistent with the proposed model, YfiN expression in a vfr mutant does not further decrease T3SS reporter activity. These data indicate that the YfiN alters T3SS expression through transcriptional control of the cAMP-Vfr dependent PexsA promoter.

Klebsiella pneumoniae is a Gram negative, enteric bacterium that frequently causes disease in immunocompromised individuals. These types of infections are often associated with the presence of indwelling medical devices, which provide a site for the organism to attach and subsequently form a biofilm. A key component in K. pneumoniae biofilm formation in vitro is type 3 fimbriae. The two main components of this project have been to determine if type 3 fimbriae are an in vivo virulence factor using a mouse model of catheter associated urinary tract infection (CAUTI) and to examine the mechanism by which the production of type 3 fimbriae are regulated. Using a mouse model in which a silicone tube is implanted into the bladder of mice, mimicking the effects of catheterization, we have been able to show that type 3 fimbriae are required for colonization and persistence. Using different time points and conditions, we demonstrated that there are conditions when type 3 fimbriae alone are sufficient for colonization and other conditions where both type 1 and type 3 fimbriae have unique roles in colonization and persistence. Additionally, competition experiments showed that neither fimbrial mutant alone, or a double mutant in type 1 and type 3 fimbriae could compete with wildtype K. pneumoniae. In most animals, only wild-type bacteria were recovered by 24 hours post-inoculation. This work reinforced the role of type 1 fimbriae in pathogenesis and showed, for the first time, a role for type 3 fimbriae using an in vivo model. Our early work has indicated that type 3 fimbriae are regulated at least in part by the intracellular levels of the secondary messenger molecule cyclic di-GMP. Downstream from the type 3 fimbrial operon a gene encoding a phosphodiesterase is present; the product of this gene breaks down cyclic di-GMP. In the absence of this gene the levels of type 3 fimbrial expression are increased. Also adjacent to the mrk operon is a two-gene operon containing the determinants we have named mrkH and mrkI. mrkH encodes a PilZ domain containing protein, which we have shown binds cyclic di-GMP. Using a transcriptional fusion we have shown that the mrk gene promoter is activated modestly in the presence of MrkH, but when MrkH and MrkI are both present the activity is increased 100-fold. This has lead to the hypothesis that MrkH and MrkI interact, which we have been able to demonstrate using copurification procedures. This interaction appears to occur in a cyclic di-GMP dependent manner with the resulting protein complex binding to the mrk promoter region and activating the expression of type 3 fimbriae.

Acinetobacter baumannii is a prevalent nosocomial pathogen where, like many other infectious bacteria, A. baumannii is increasingly considered a multi-drug resistant pathogen. This research study was designed to find a way to affect the persistence of A. baumannii such that it can be applied to a hospital setting to prevent further nosocomial infections. One regulatory mechanism potentially used by A. baumannii to persist on hospital surfaces is through the use of the bacterial second messenger cyclic-di-GMP (c-di-GMP). This nucleotide signal is regulated in response to environmental conditions, and then activates c-di-GMP-binding proteins that induce phenotypic changes. One c-di-GMP-regulated phenotype is bacterial motility, and reducing motility may alter A. baumannii’s ability to colonize and persist on hospital surfaces. I hypothesized that A. baumannii uses c-di-GMP-binding proteins to regulate motility. A. baumannii encodes two potential c-di-GMP-binding proteins of interest, one that contains a sole c-di-GMP-binding PilZ domain and another that pairs a PilZ domain with a hydrolase enzymatic domain. I am also testing two A. baumannii strains: AB5075, a recent multi-drug resistant military hospital isolate and 17978, an established lab strain. A notable difference between these two strains is that AB5075 demonstrates twitching motility where it utilizes type IV pili, while 17978 demonstrates swarming motility that has unknown mechanisms. Both c-di-GMP-binding proteins were tested for their role in motility for the particular A. baumannii strain. While I am still generating the deletion strain for the c-di-GMP-binding hydrolase enzyme in AB5075, the sole PilZ domain protein is required for twitching motility, while both c-di-GMP-binding proteins are required for 17978 swarming motility. [PM1] Future plans include determining the role of the c-di-GMP-binding hydrolase enzyme in twitching motility and identifying the role that these proteins play through binding of c-di-GMP.

Salmonella is a leading bacterial cause of foodborne illness worldwide. During a previous study investigating the enzymes responsible for regulating cyclic-di-GMP concentrations, a mutant in the cyclic-di-GMP-specific phosphodiesterase STM3615 was identified that displayed a phenotype characterized by decreased survival on agar plates and a shorter bacterium length. I was able to determine that the periplasmic domain of STM3615 was responsible for this phenotype, not the enzymatic phosphodiesterase domain. Based upon a bioinformatic analysis of the protein, I then hypothesized that the periplasmic domain of STM3615 was interacting with a periplasmic protein to give rise to this phenotype. To identify this periplasmic protein partner, a transposon mutagenesis approach was taken to disrupt genes within the STM3615 mutant. Two mutants, rcsD and yrfG, within the STM3615 deletion mutant restored the WT phenotype and require further investigation. RcsD is an important partner of the transcription regulatory protein RcsB that controls expression of FtsZ, a key player in cell division.

Regulation of the bacterial second messenger cyclic-di-GMP in Salmonella Typhimurium allows it to delicately alter phenotypes to optimize invasion and survive intracellularly in epithelial cells and macrophages to become virulent and cause infection. The concentration of cyclic-di-GMP is regulated by the presence of external stimuli, sensor CMEs (diguanylate cyclases, DGCs, and phosphodiesterases, PDEs), and cyclic-di-GMP binding effectors. Previous studies established that maintenance of low cyclic-di-GMP concentrations is required for survival in macrophages and that the deletion of 3 active PDEs reduces this survival. This study aimed to further investigate the regulation of cyclic-di-GMP for survival in macrophages and epithelial cells. Salmonella Typhimurium mutants were generated and used for an infection assay with RAW 264.7 macrophage and HeLa epithelial cell lines to determine active CMEs via intracellular survival. Intracellular survival was quantified by plate counting of cell lysates at 1-, 4-, and 24-hours post-infection. Our result showed that the previously identified 3 PDEs also influenced the infection of epithelial cells. We re-established the decreased survival in the RAW 264.7 macrophage cell line and determined that the cyclic-di-GMP-binding cellulose synthase BcsA was responsible for decreased survival in macrophages. Finally, we identified an active DGC whose deletion within the 3xKO PDEs restores survival levels, suggesting that this enzyme is responsible for the synthesis of cyclic-di-GMP during macrophage infection. Further studies on how cyclic-di-GMP regulates Salmonella Typhimurium intracellular survival could lead to identifying a potential alternative drug target for treating its infections.