DEFINING THE FUNCTION OF AN UNCHARACTERIZED CBASS GENE IN THE REGULATION OF 3’3’-CGAMP SIGNALING IN VIBRIO CHOLERAE By Soo hun Yoon A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Microbiology and Molecular Genetics – Doctor of Philosophy 2022 ABSTRACT Bacteria compete with phages in a myriad of environments for survival. This constant arms race has led to the acquisition of cyclic oligonucleotide based antiphage signaling system (CBASS) throughout all bacterial phyla. Phage infection activates CBASS which results in altruistic suicide of the host to save the clonal community, a process termed abortive infection. In Vibrio cholerae El Tor, CBASS is comprised of DncV, CapV, Vc0180, and Vc0181. DncV is activated following phage infection to synthesize 3’3’- cGAMP which activates CapV, a phospholipase that degrades the cellular membrane. Though evidence suggests VC0180 and VC0181 allow for response against broader range of phages, their function in relation to DncV remains undefined. To determine their role, we investigate the effect of VC0180 and VC0181 on the stability of DncV in vivo. During this pursuit, we discovered a novel protein which we named Bumo, for Bacterial Ubiquitin Modifier, encoded upstream of capV that is a component of the CBASS operon. We show a novel regulatory network of DncV, in which Bumo protects DncV from degradation and that VC0181 is a protease that degrades DncV. We also show evidence that VC0180 interacts with DncV. To expand our knowledge of CBASS, we also explored the function of CBASS systems encoding HNH-SAVED effectors in other Gram-negative bacteria. We discovered E. coli EDEC13E and P. fluorescens SRM1 have active CBASS systems that affect the growth capacity of heterologous hosts. We also show their nucleotide substrate specificity. Interestingly, the E. coli EDEC13E HNH-SAVED effector is inactivated following the addition of nucleotide signals, which is distinct from all previously described CBASS systems. In conclusion, Bumo is a newly discovered component of the CBASS operon that protects DncV from degradation by Vc0181. VC0180 and VC0181 both regulate DncV in separate ways. Moreover, the function of CBASS is conserved in other Gram-negative species, even though they may have adapted to respond differentially to cyclic oligonucleotide signals. The results of my thesis work lead to a better appreciation of the complexity in the regulation of CBASS signaling and the diversity in evolution of CBASS across species. This thesis is dedicated to my parents and family both biological and non-biological. Thank you all for your love, support, and encouragements throughout the years. Merci pour votre courages et support durant mon entrainement. 엄마, 아빠, 누나, 동생과 가족들 사랑과 기도 감사합니다. iv ACKNOWLEDGEMENTS I want to acknowledge first and foremost my family and close friends who have been there through the rough times and the good times, for their unconditional love and support. The time that you spend encouraging, supporting and loving me have helped me persevere. I want to thank Dr. Chris Waters, my PI, who have been so kind in guiding and sharing his wisdom with me throughout my training and creating a safe space for me to reach my potential. I especially want to acknowledge the patience and understanding he showed when I was going through multiple health crises throughout my training and how supportive he is of my aspirations and vision. I want to acknowledge members of the Waters lab, both former and current who have been there to create a safe, loving, and supportive environment that helped me thrive and achieve all that I did. I especially want to thank Brian Hsueh who have been a great friend and colleague I could bounce ideas with and share in the down and up sides of research and life. I also want to specially acknowledge my undergraduate associates Ram Sanath-Kumar and Elise Trost who have been amazing student, enthusiastic about science, ready to learn and being great implementers of knowledge. I want to thank Geoffrey Severin, who have been an awesome lab mate, encourager, and scientist to bounce ideas off of with. I would like to thank my close friend Alex Moauro, Meng-lun Hsieh and Michael Maiden who have made this journey worthwhile and memorable. I would like to acknowledge my close friends from around the world who gave me their time, encouragement, memories of travel and food. I would like to acknowledge the Chauvancy family for their unconditional love, encouragement, and hospitality. I would also like to give a shout out to my undergraduate mentor Dr. Greg Powell and Dr. Cynthia Powell for their wisdom, guidance and encouragement that helped me persevere. v I would like to thank my committee members Dr. Kristen Parent, Dr. Neal Hammer, Dr. Sean Crossen and Dr. Shannon Manning for their guidance, wisdom, and support. I would like to acknowledge Dr. Anthony Schilmiller and Dr. Lijun Chen of the Michigan State University Metabolomics and Mass Spec Core for the patience and help in my project throughout the years even though I continue to struggle with Mass Spec. vi TABLE OF CONTENTS LIST OF ABBREVIATIONS…………….………………………………………………………....……viii CHAPTER 1: INTRODUCTION...………………….……………………………………………….…...1 1.1 Vibrio cholerae: The Diarrheal Pathogen………………………………………………2 1.2 Diverse world of Cyclic di nucleotide signaling……………………………………….5 1.3 Cyclic GMP-AMP Signaling and CD-NTases in Bacteria……………………..…10 1.4 Adaptive Phage Defense and CBASS in Pathogenic V. cholerae……………..14 1.5 Ubiquitin-like Systems in Bacteria………………………………………………………18 1.6 Chapter 1 Summary…………………………………………………………………………..20 CHAPTER 2: DEFINING THE FUNCTION OF AN UNCHARACTERIZED GENE IN THE REGULATION OF 3’3’-CGAMP SIGNALING IN VIBRIO CHOLERAE EL TOR......................................................................................................21 2.1 Introduction…………………………………………………………………………………….22 2.2 Results…………………………………………………………………………………………….28 2.3 Discussion……………………………………………………………………………………….57 2.4 Materials and Methods………………………………………………………….………….61 CHAPTER 3: DETERMINING THE FUNCTION OF CBASS IN OTHER GRAM- NEGATIVE BACTERIA……………………………………..……………………………………………..70 3.1 Introduction…………………………………………………………………………………….71 3.2 Results…………………………………………………………………………………………….75 3.3 Discussion……………………………………………………………………………………….89 3.4 Materials and Methods……………………………………………………………………..96 CHAPTER 4: BROADER IMPACTS AND CONCLUSIONS…………………..……….102 REFERENCES…………………………………………………………………………………………..….….112 APPENDIX…..………..………………………………………………………………………………………..127 vii LIST OF ABBREVIATIONS Bumo Bacterial ubiquitin-like modifier cAMP Cyclic AMP CBASS Cyclic/oligonucleotide based antiphage signaling system cdN Cyclic di nucleotide c-di-AMP Cyclic di adenosine monophosphate c-di-GMP Cyclic di guanosine monophosphate cDNA Complementary DNA CD-NTase cGAS/DncV-like nucleotidytransferase cGAMP Cyclic guanosine monophosphate - adenosine monophosphate CIP Calf intestinal phosphatase coN Cyclic oligonucleotide CT Cholera toxin ctN Cyclic trinucleotide DAC Diadenylate cyclase dCMP Deoxycytidine monophosphate dCTP Deoxycytidine triphosphate DHF Dihydrofolate DGC Diguanylate cyclase DUB Deubiquitinase dUMP Deoxyuridine monophosphate EDEC E. coli EDEC13E EOP Efficiency of plaquing NTP Nucleotide triphosphates viii OMV Outer membrane vesicles PCR Polymerase chain reaction PDE Phosphodiesterase Pf P. fluorescens SRM1 PLE phage-inducible chromosomal island-like elements PUP Prokaryotic ubiquitin-like modifier SUMO Small ubiquitin-like modifier TCP Toxin-coregulated pilus THF Tetrahydrofolate TLC Thin layer chromatography TSS Transcriptional start site Ubl Ubiquitin-like protein Vc Vibrio cholerae VPI-1 Vibrio pathogenicity island 1 VSP-I/II Vibrio seventh pandemic island I or II ix CHAPTER 1: INTRODUCTION Part of this chapter was published in: Yoon S hun, Waters CM. 2019. Vibrio cholerae. Trends Microbiol 27:806–807. Yoon SH, Waters CM. 2021. The ever-expanding world of bacterial cyclic oligonucleotide second messengers. Curr Opin Microbiol 60:96–103. 1 1.1 Vibrio cholerae: The Diarrheal Pathogen Vibrio cholerae is the primary bacterial pathogen responsible for the Cholera pandemic worldwide (1). Plaguing many developing nations and areas of poor sanitation, it causes about 2.9 million cases and ~95,000 deaths annually that are worsened in times of natural disasters and humanitarian crises (2, 3). Characterized by its curved-shaped rod, this Gram-negative bacterium causes “rice-watery” stools upon consumption of contaminated food or water (Fig. 1.1). A previously healthy host can succumb to the illness due to two major virulence factors released: toxin-coregulated pilus (TCP) and cholera toxin (CT) (1, 4). TCP is important for colonization of the host gut mucosal layer where CT is released and passes through the gastro-endothelial wall to cause watery diarrhea (5, 6). CT binds to and 2 activates the G-alpha protein via ADP-ribosylation which induces adenylate cyclase to synthesize cAMP (7–9). CT induction leads to cAMP levels hundreds of fold over normal physiological levels leading to downstream activation of chloride ion channels, primarily the cystic fibrosis transmembrane conductance regulator (CFTR) (8, 10, 11). This results in massive efflux of chloride ions and expulsion of water, sodium, potassium and bicarbonates and diminished absorption that leads to dehydration and diarrhea characteristic of the disease (12, 13) (Fig. 1.1). If not adequately attended to, mass dehydration can lead to hypovolemic shock and metabolic acidosis that can prove fatal. Oral rehydration therapy is the main course of treatment with antibiotics as adjunct in case of prolonged or severe diarrhea (1, 3, 12). However, water sanitation and proper hygiene are the best proven preventative measures for cholera. Proper hygiene methods and treatment can decrease mortality from 50% to less than 1% (3, 13, 14). Though easy to treat and less lethal than other well-known bacterial pathogens such as Mycobacterium tuberculosis, cholera continues to be a threat in resource poor and endemic regions especially during times of natural disasters, wars as in Yemen and seasonal heavy rainfall as seen in many endemic Sub-Saharan African countries (2, 3). Vibrio cholerae is found in a myriad of environments. It often resides in saline waters along coasts and estuaries in association with shellfish and zooplanktons. Biofilm formation provides protection and a means of transportation for the bacteria to colonize and survive in these diverse environments (15, 16). They aggregate as biofilms, entering a dormant state to safeguard against varying pH, temperature, low nutrients or predations and propagate when conditions are optimal (13, 16–18). Chitin is the main carbon source of Vibrio cholerae in the aquatic environment and the lack thereof can facilitate biofilm formation (19, 20). V. cholerae is used as a model organism to study virulence in Gram- 3 negative bacteria due to its high infectious dose, ease of genetic manipulation, and rapid replication time (21–23). V. cholerae is also a model for studying chemical signaling such as cell-to-cell communication via quorum sensing and cyclic di-nucleotide signaling, both of which facilitate the lifestyle switch between motile and biofilms states of the bacterium in response to the environment (15, 24). Cholera is caused by two serogroups of Vibrio cholerae, the O139 and O1 serogroups, with the latter being the main causative agent. The O1 serogroup can be further divided into the classical and El Tor biotypes (1, 25). The El tor biotype is the primary agent accountable for the current 7th Cholera pandemic while the classical biotype was responsible for the previous six pandemics from the early 1800s to the start of the current pandemic in 1961 (26–29). Unlike the classical biotype, it had acquired two pathogenic islands, VSP-1 and VSP-2 that allow for its prominence (26, 30). The functions of the majority of the genes encoded in the islands are unknown. Recent work has shown the presence of two potential bacteriophage defense islands, AvcID and CBASS (31, 32). My thesis research used V. cholerae as a model system to better understand cyclic dinucleotide signaling and its connection to phage defense. I will review this topic in this chapter. 4 1.2 Diverse world of Cyclic di nucleotide signaling Elucidating the molecular underpinnings of how bacteria sense and respond to their environment is central to understand their evolution and adaptation. Given their ecological diversity, it is therefore not surprising that bacteria encode a myriad of strategies to respond to ever changing challenges including transcription factors(33, 34), two-component and phosphorelay pathways (35), mechanosensing (36), and chemotaxis (37). Of primary importance to environmental sensing and adaptation are second messenger signaling pathways. These pathways recognize and respond to a “first” signal that modulates the activity of a synthesis or degradation enzyme that controls the level of an intracellular signal molecule. Second messengers in bacteria are primarily purine derived modified nucleotides, which exert global regulatory effects by altering transcription, translation, or even protein activity (38). Second messenger systems have long been recognized as global regulatory networks in bacteria (and all living systems), but we have just begun to fully appreciate their diversity and central importance to many aspects of bacterial physiology. The original second messenger systems discovered include cyclic AMP, which is primarily associated with the regulation of carbon utilization (39, 40) as well as other central traits such as biofilm formation, virulence, and central metabolism (41–43), and guanine penta/tetraphosphate that drives the bacterial stringent response (44, 45). However, this chapter will focus on the more recent additions to this chemical lexicon, cyclic di- and trinucleotide (cdN and ctN) second messengers (Fig. 1.2). 5 6 Cyclic di-GMP (c-di-GMP) controls transitions from motility to biofilm formation. The first described cdN in any living organism is 3’-5’, 3’-5’ c-di-GMP. This second messenger was first identified and characterized by Moshe Benziman’s laboratory studying cellulose biosynthesis in Komagataeibacter xylinus (46–48), although the prevalence of this cdN was not widely appreciated until 2000 – 2005 with the advent of bacterial genome sequencing and a renewed interest in understanding the molecular mechanisms that regulate biofilm formation ( 4 8 ) . Analysis of bacterial genomes revealed a widely conserved domain of unknown function that was homologous to Benziman’s c-di-GMP synthases (49–51). This domain, which possess diguanylate cyclase (DGC) activity, was named the “GGDEF” domain for the key amino acids in its active site (52). The corresponding c-di-GMP phosphodiesterase (PDE) that degrades c-di-GMP was soon shown to be another widely conserved enzymatic domain that was named the “EAL” domain (53, 54). Moreover, a second PDE domain termed the HD-GYP domain was demonstrated to also degrade c-di-GMP ( 5 5 ) . GGDEF, EAL, and HD-GYP enzymes are present in the vast majority of bacteria, and in addition to being widely conserved, bacterial genomes can encode up to dozens of enzymes involved in c-di-GMP synthesis or degradation (56). Seminal studies of biofilm formation and motility in Escherichia coli, Vibrio cholerae, Salmonella Typhimurium sp., and development in Caulobacter crescentus demonstrated that c-di-GMP regulated by DGCs and PDEs was associated with the transitions between motile and sessile lifestyles (50, 57, 58). C-di-GMP signaling enzymes typically are multidomain proteins consisting of an N-terminal sensory domain and a C-terminal enzymatic domain (48). The N-terminal domain is proposed to bind specific environmental cues that regulate the synthesis and degradation of c-di-GMP, such that in specific environments a sessile lifestyle is 7 initiated or inhibited. Although a handful of environmental cues are known (for example oxygen ( 5 9 ) , spermine ( 6 0 ) , bile and bicarbonate (61), and reducing conditions (62)), the majority of environmental signals regulating DGCs and PDEs remain to be discovered (Fig. 1.3). Changes in the intracellular concentration of c-di-GMP is sensed by a myriad of transcription factors, riboswitches, and protein complexes whose activity is controlled by directly binding to c-di-GMP (48, 63). Although classically known as an inducer of biofilm formation and repressor of motility, c-di-GMP functions as a global regulator controlling numerous phenotypes including but not limited to bacterial predation (64), virulence (65), development (66), DNA repair (67), and cell shape (68). C- di-GMP signaling systems are primarily found in bacteria, although stalk cell formation in the slime mold Dictyostelium discoideum is dependent on c-di-GMP (69). 8 Cyclic di-AMP (c-di-AMP) The next cdN to be discovered in 2008 was 3’-5’, 3’-5’ c-di-AMP, first observed to be synthesized by the DNA integrity scanning protein DisA (70). As excellent reviews of c- di-AMP have been recently published, we will only briefly summarize it here (71, 72). C-di-AMP is found in many bacterial and archaeal species, and it is notable for having been identified in many Gram-positive bacteria not known to use c-di-GMP (73, 74). C- di-AMP is synthesized by diadenylate cyclase enzymes (DAC) and degraded by PDEs that have a DHH-DHHA1 domain (71) (Fig. 1.2). C-di-AMP has notable differences from c- di-GMP including that most bacteria encode only one or a few DACs (73), and c-di- AMP is essential in many but not all bacterial species including the Firmicutes (71, 75). This requirement for growth stems from c-di-AMP as a regulator of cellular osmolarity that controls the import and export of potassium and other osmoprotective molecules (76). Although the mechanism of this sensing is not fully understood, high extracellular concentrations of potassium increase DAC activity and intracellular c-di-AMP, which then functions to limit potassium uptake while promoting potassium export (76–78). C-di- AMP plays an analogous role in controlling the intracellular concentrations of other water-soluble osmoregulatory molecules such as certain amino acids and sugars (77, 79). Thus, in the absence of c-di-AMP, cells are not able to appropriately balance their osmotic state. Accordingly, c-di-AMP is not essential in rich media with lower salt concentrations or defined minimal media (76, 80). 9 1.3 Cyclic GMP-AMP Signaling and CD-NTases in Bacteria Bacterial 3’-5’, 3’-5’ cGAMP (3’3’cGAMP), the third known cdN was first discovered in 2012 from the El Tor Biotype of Vibrio cholerae, which is the causative agent of the 7th and current cholera pandemic (30). 3’,3’ cGAMP is synthesized by DncV, which interestingly is encoded on the Vibrio Seventh Pandemic-1 (VSP-1) genomic island that is unique to El Tor (81). DncV was reported to influence V. cholerae motility and colonization, but the receptor for 3’,3’cGAMP in V. cholerae was unknown (30). However, we recently demonstrated that 3’,3’ cGAMP directly binds to and activates CapV, a patatin-like phospholipase encoded directly adjacent to dncV on the VSP-1 island. 3’,3’ cGAMP binding to CapV activates this enzyme, leading to degradation of the cell membrane (82). 3’,3’ cGAMP synthesized by a DncV homolog also modulates biofilm formation in Escherichia coli ECOR1, an animal commensal, suggesting this signal may regulate broader functions (83). DncV is structurally analogous to the eukaryotic enzyme cyclic GMP-AMP synthase (cGAS), and both enzymes evolved from a common ancestor (84). Together, they belong to a novel protein superfamily termed cGAS/DncV like nucleotidyltransferase (CD-NTases), which is genetically conserved throughout all bacterial phyla (85, 86). In stark contrast to DGC and DAC protein families, CD-NTases produce an extensive array of nucleotide signals utilizing all four ribonucleotides to form linear oligonucleotides to cdN and ctN molecules, allowing for specificity and diversity in downstream pathways (86). In bacteria, recent in vitro studies revealed the first examples of pyrimidine containing cdNs including cyclic UMP-AMP and cyclic di-UMP, and ctNs such as cyclic tri-AMP (cAAA) and cyclic AMP-AMP-GMP (cAAG) (86–88). A bioinformatic analysis by Burroughs et al predicted such nucleotide synthases might participate in biological conflicts (85). Indeed, Cohen et. al. showed that dncV and capV 10 initiate altruistic suicide by restricting cell growth upon phage infection, thereby aborting phage replication and limiting phage spread within the bacterial population (89). Furthermore, 3’,3’-cGAMP synthesis by DncV is induced upon phage infection via an unknown mechanism (89). CD-NTases like DncV are found genetically associated with putative effector proteins that suggest their primary function is phage defense (Fig. 1.2). Such signaling modules were renamed as cyclic oligonucleotide (coN)-based anti-phage signaling systems (CBASS) (89). Millman et al proposed a classification system that organizes CBASS systems according to their operonic architecture, effector function and dominant signaling nucleotide (31). However, a major outstanding question is how phage infection modulates nucleotide synthesis by CD-NTases. The enzymes that degrade 3’,3’ cGAMP remain relatively unstudied although a few examples of 3’,3’ cGAMP PDEs have emerged. 3’,3’ cGAMP is degraded via hydrolysis by the V-cGAP1/2/3s in V. cholerae and PmxA in Myxococcus xanthus, both of which are HD-GYPs (90, 91) . VcEAL is the first 3’,3’ cGAMP specific EAL type PDE encoded by V. cholerae (20). Unlike the V-cGAP1/2/3 PDEs, VcEAL and PmxA hydrolyze both c-di- GMP and 3’,3’ cGAMP. 3’,3’ cGAMP and Hypr GGDEFs (20, 91). Bacterial 3’3’-cGAMP has also been studied in delta-proteobacteria. Although, in these bacteria 3’3’-cGAMP is not synthesized by a CD-NTase, but by hybrid promiscuous GGDEF (Hypr GGDEF), a GGDEF-like enzyme with active site residue variations that coordinate synthesis of 3’,3’ cGAMP rather than c-di-GMP (Figure 3) (92). 3’,3’ cGAMP binds riboswitches to modulate iron(III) oxide metal reduction in Geobacter sulfurreducens and mediates osmotic stress response in M. xanthus (92, 93). Interestingly, in Geobacter c-di-GMP activates biofilm formation and energy production on electrode surfaces, suggesting 3’,3’ cGAMP and c-di-GMP are antagonistic signaling pathways that induce alternative 11 lifestyles in this bacterium (94). Structural analyses revealed Hypr-GGDEFs have a symmetric active site whose nucleotide product is dependent on substrate availability, in contrast to DncV, which has an asymmetric active site that preferentially binds ATP and GTP as substrates (94, 95). However, 3’,3’ cGAMP is the major nucleotide signal for both enzymes. 2’,3’ cGAMP as an immune modulator. In human cells, a structural isomer of bacterial 3’3’- cGAMP, 2’-5’, 3’-5’-cGAMP, is synthesized by the CD-NTase cGAS upon binding cytosolic DNA introduced via viral invasion or intracellular damage (96). 2’3’-cGAMP then activates the STING receptor which upregulates Type I interferon production, leading to an anti-viral or anti-cancer response (97, 98). STING is also capable of sensing extracellular c-di-AMP released by the intracellular pathogen Listeria monocytogenes, leading to increased interferon response (74, 97). A second eukaryotic cdN receptor named RECON, which specifically recognizes adenine containing cdNs, was also recently discovered, suggesting an intricate recognition and response of eukaryotic cells to cdNs (99). CD-NTase – the newest enzyme family that synthesizes coNs. In contrast to c-di-AMP and c-di-GMP, 3’,3’ cGAMP regulation and effector functions are not as well described. Given the diversity of CD-NTases, characterization of these signaling modules is an exciting new frontier in understanding the evolution of coN signaling systems across biological kingdoms and its adaptation to modulate specific response pathways in different environments from seawater to human hosts. Many CD- NTases that have been identified do not exhibit nucleotide synthesis activity and their function is unknown(86, 100). Evidence of CD-NTases mediating signaling pathways vital for cellular adaptation and virulence present novel therapeutic targets amid growing 12 concerns with antibiotic resistance. In theory, abortive infection driven by CBASS can be manipulated to limit infection by human pathogens. DncV and cGAS, cGAMP synthases have been shown to be regulated by two mechanisms thus far. cGAS, the eukaryotic synthase, is activated upon binding of DNA or RNA nucleic acids to its protein backbone to initiate cGAMP production(98). This results in regulated response to invading pathogenic nucleic acids in the cell cytoplasm such as viruses and endocytosed bacteria (98, 101). In V. cholerae folate-like molecules inhibit cGAMP production by binding to corresponding backbone pocket (95). Thus, CD-NTase regulation is categorized into to DNA/RNA activated Type I or Folate inhibited Type II systems. 13 1.4 Adaptive Phage Defense and CBASS in Pathogenic V. cholerae Bacteria live in the environment in conjunction with predatory viruses known as phage in the environment (102). Bacterial phages are in a constant arms race with their hosts for survival leading to a continuous cycle of adaptation and co-evolution. Horizontal gene transfer has allowed for acquisition of novel DNA for the survival of V. cholerae (23). Pathogenic V. cholerae is distinguished by the production of TCP and CT that are encoded on mobile pathogenicity island, VPI-1 and lysogenic bacteriophage within its chromosome (4, 102). Pathogenic strains of V. cholerae employ various phage defense mechanisms such as decoy outer membrane vesicles (OMV) (103, 104), phage-inducible chromosomal island-like elements (PLEs) (18), and the cyclic/oligonucleotide based antiphage signaling system (CBASS) (82, 89). The 7th pandemic pathogenic El Tor strains notably have two genetic islands VSP I and II (for Vibrio Seventh Pandemic) that can also excise from the chromosome to make circular intermediates (81). Although the functions of the majority of these genes are unknown, recent evidence has shown the presence of two potential phage defense systems, AvcID and CBASS. The constant battle for survival of environmental Vibrio phage with these elements may have led to the prominence of three lytic vibrio phages, ICP1, 2 and 3, in clinical samples allowing for detailed exploration of phage defense mechanism in this system (18, 23, 105). V. cholerae release OMV coated in specific receptors that neutralize all three ICP phages (103, 104). OMV isolated from V. cholerae encoding mutant O1-antigen (104), specific for ICP1 or mutant ompU (23), specific for ICP2, showed lower levels of inhibition 14 while increased expression led to higher inhibition. Though the receptor for ICP3 is unknown, WT vesicles was able to neutralize ICP3 albeit at a lower level than ICP1. Phage-inducible chromosomal island-like elements (PLE) are V. cholerae specific mobile genetic islands that protects the bacterium from ICP1 infections (18, 105). PLE are excised from the chromosome following ICP1 infection and expressed to reduce phage replication and induce earlier cell lysis resulting in lower phage generation (18). Excised PLEs are packaged into virions and are transferred to neighboring cells and integrated to host genomes in an O-antigen dependent manner (18). This leads to sharing of phage defense islands among neighboring populations. PLEs are activated in response to ICP1 encoded CRISPR-Cas system (105). CRISPR-Cas is a form of bacterial adaptive immune response in which a small effector RNA “guides” Cas endonucleases to specific genetic targets for degradation (106). Though typically used by bacteria to identify and degrade invading phage DNA, studies of phage-host evolution have revealed that ICP1 phages utilizing a CRISPR-Cas system to counteract and degrade PLEs in V. cholerae (23, 105). On the VSP-1 island, our lab has recently uncovered the function of AvcID, a novel toxin antitoxin phage defense system in V. cholerae (32). AvcD is a dCTP/dCMP specific deaminase that decreased the dCTP and dCMP pools while increasing cellular dUMP following phage infection (32). However, in the absence of stimuli, AvcD is inhibited by a small RNA AvcI found upstream of AvcD. The AvcI degradative enzyme and mechanism of phage defense is yet undefined; however, it is hypothesized it biases cellular dNTP pool to decrease phage replication (32). A rather novel form of adaptive bacterial phage response is CBASS. First described in 2019, these are operons characterized by a cyclic/oligonucleotide synthase (CD-NTase) and its effector at its core (89). The CD-NTase is a synthase that utilize nucleotide 15 triphosphates (NTPs) to produce cdNs and ctNs that act as signals to bind and activate the associated effector in response to phage infection (31, 89). CD-NTases are associated with a variety of putative effectors such as phospholipases, endonucleases, and transmembrane proteins. Patatin-like phospholipases are the most common effector found associated with CD-NTases (31). Many effector enzymes are fused to putative cdN and ctN sensing domains such as STING and SAVED, and binding of the nucleotide signal to these domains modulates effector function (85, 88). The SAVED domain, which contains two CRISPR-associated Rossman folds (CARF), binds the specific nucleotide signal produced by their cognate CD-NTase to activate the fused effector function (88). This was recently demonstrated with the widespread Cap4 endonuclease that is activated after specifically binding to cyclic tri-AMP synthesized by its CD-NTase, CdnD, following phage infection (87). The majority of CBASS exist as a two gene operon, comprised of a CD-NTase and effector, predicted to confer phage defense (31). Since its introduction, CBASS is classified into 4 types. Type 1 comprises the core CD-NTase and effector operon (31). Type 2 contains ubiquitin like E1/E2 and DUB-like ancillary genes (31). Type 3 s TRIP13 and HORMA-like ancillary genes (31). The most rare type 4 is only found in Archaea and Firmicutes with potential nucleotide modifying enzymes encoded ancillary to the core (31). Many bacterial CBASS operons exist with ancillary domains such as eukaryotic-like ubiquitination or HORMA/TRIP13 ancillary systems predicted to modulate cdN or ctN synthesis in phage infection (31, 107). These accessory genes enhance bacterial resistance to more phage in addition to the CD-NTase and its effector. Bacterial HORMA1 initiates cAAG synthesis by binding its associated CD-NTase and TRIP13 dissociates the two. The role of eukaryotic ubiquitination-like systems, however, remains undefined in bacteria 16 (107). In eukaryotes, cGAS sensitivity to cytosolic DNA is modulated by small ubiquitin- like modifiers (SUMO) through E3 ubiquitin ligase and deubiquitinase (108). DncV does not sense cytosolic DNA, however, crystal structure analyses showed folate-like molecules inhibiting the enzyme in the corresponding structural backbone (95). Thus, a hypothetical function of eukaryotic-like ancillary proteins might be to modulate CD-NTase sensitivity to activating nucleic acids as in cGAS, or inhibitory molecules such as folates in DncV. 17 1.5 Ubiquitin-like Systems in Bacteria Ubiquitin-like (Ubl) post-translational protein modification pathways modify protein activities or target them for degradation by proteases (109, 110). Extensive studies in eukaryotes have shown Ubl signaling systems compose of E1, E2 and E3 enzymes that work together to activate, conjugate and transfer the Ubl modifier to their target proteins (110–114). Many systems include a deubiquitinase (DUB) which cleaves the substrate from the target (115). Ubiquitination involves activation of the Ubl modifier by an E1-like enzyme (110). Ubl have a characteristic C-terminal glycine (-GG) motif or in some, serine glycine (-SG) motifs (111, 116–118). The E1 enzyme adenylates the C-terminal glycine using ATP, forming a thiosester bond with an E1 cysteine (111, 117, 119). The activated Ubl is then to transferred to an E2 conjugating enzyme via a transesterification reaction (120–122). In the presence of an E3 enzyme, E2 conjugates the Ubl to the E3 which identifies and catalyzes the ligation of Ubl to a lysine residue on its target (110, 114, 120). Proteins can be monoubiquitinated as in many signaling pathways or polyubiquitinated when destined for degradation. In bacteria, a prokaryotic ubiquitin-like protein (Pup) that is structurally distinct from other Ubls, tags proteins for degradation (123, 124). Pup is activated at its C- terminus similar to Ubl proteins, however a -GGE/Q motif is found unlike the Ubl characteristic -GG motif (123, 125, 126). The terminal asparagine residue is ligated to its substrate at a lysine residue via the PafA ligase, which then directs the substrate to the Mpa-protease complex for degradation (118, 126, 127). Pup is cleaved by depupylase (125). Pup is ligated and cleaved in a biochemical mechanism similar to eukaryotic Ubl however using just one enzyme in the ligation process (126). This hints at the possibility that 18 ancestral ubiquitin-like systems were more simplified, less involved versions of the now more involved eukaryotic process found. In the degradative Ub-ligase pathway, the E3 ligase transfers the modifier to a target protein to signal it for degradation by proteasome complexes (110). Polyubiquitination allows for specific targeting and protein recycling. Sae1 and Sae2, E1 and E2-like yeast enzymes, form a complex with SUMO to modulate target I-kappa B- alpha protein activity blocking NF kB dependent transcriptional activation (128). Unlike other Ubl systems, Sae1 and Sae2 can conjugate SUMO to its target without the presence of an E3 ligase in vitro (128). MoeB and ThiF, molybdenum and thiamine synthase respectively, are bacterial precedents of eukaryotic E1 with beta-grasp folds (119, 129– 131). They are activated by sulfur transfer upon binding of their ubiquitin-like modifiers MoaD and ThiS which allows for conjugation with E2 equivalents (119, 129, 131–133). These E2 equivalents then transfer the modifier, a thiocarboxylate group, to a target protein to alter their activity without an E3 ligase (119, 129, 131–133). V. cholerae’s CBASS have vc0180 and vc0181 ancillary genes downstream of dncV. VC0180 is predicted to have an E2-like N-terminal domain and a C-terminal E1-like domain. VC0181 encodes a JAMM motif DUB-like domain. Together they are suggesting a ubiquitin-like post-translational modification of CBASS. However, V. cholerae’s CBASS system lack an E3 ligase suggesting a more modulatory role of these enzymes than targeting for degradation. 19 1.6 Chapter 1 Summary In summary, Vibrio cholerae is an pathogen of concern in developing nations, especially in areas of poor sanitation and healthcare access (134, 135). It causes the diarrheal disease cholera through release of various virulence factors such as TCP and CT (4, 136). It survives in various environment such as aquatic reservoirs and humans hosts through lifestyle modification mediated by cyclic di-nucleotide signaling (15, 16). In a constant battle for survival with phage predators, they have acquired various genetic island that encode phage defense mechanisms such as AvcID and CBASS (31, 32, 89, 102). In particular V. cholerae CBASS has produces cGAMP in response to phage infection in order to activate CapV, which degrades the host cell membrane (82). This altruistic suicide of the infected host leads to survival of the clonal community leading to aborted infection. Even though the function of DncV and CapV is well studied, the function of cognate genes VC0180 and VC0181, E1/E2-like enzyme and DUB-like enzyme are not well defined. We do know that they confer defense against a broader range of phages, however, their mechanism and actions are not yet known (89). Thus, to better understand the regulation of 3’3’-cGAMP signaling in response to phage predation, we explore the function of VC0180, VC0181 and an uncharacterized gene I discovered in modulating CBASS response. 20 CHAPTER 2: DEFINING THE FUNCTION OF AN UNCHARACTERIZED GENE IN THE REGULATION OF 3’3’-CGAMP SIGNALING IN VIBRIO CHOLERAE EL TOR For this research portion of my thesis, many individuals were involved. Undergraduate associate of Waters lab Elise Trost cloned and performed cell viability assays of various Vibrio cholerae El Tor mutants. Another undergraduate associate Ram Sanath-Kumar performed cloning of Vibrio cholerae El tor CBASS for phage studies, infection assays and transposon screening under the combined guidance of Kaylee Wilburn. 21 2.1 Introduction Vibrio cholerae is a diarrheal pathogen found in a large array of environments from the human host, to shellfish and chitin surfaces of aquatic reservoirs (15, 16, 19). It is able to adapt to and survive in different conditions due to cyclic di-nucleotide signaling pathways that respond to environmental stimuli (24, 68). Cyclic di-GMP, modulates the switch between bacterial motile and biofilm lifestyles through various transcriptional and post-translational mechanisms (137, 138). 3’3’-cyclic GMP-AMP (cGAMP) is another cyclic di-nucleotide in V. cholerae synthesized by the enzyme DncV to mediate phage defense in a system known as CBASS (cyclic/oligonucleotide based antiphage signaling system) (30, 95). CBASS activity commences with activation of DncV upon phage infection via an unknown signaling mechanism (89). cGAMP then binds and activates CapV, a patatin-like phospholipase that specifically degrades phosphotidylethanolamine and phosphotidylglycerol, the major phospholipids found in V. cholerae, cleaving off 16:1 and 18:1 free fatty acids (82) (Fig. 2.1). High activation of CapV leads to degradation of the membrane, making the cell no longer viable (82). This response has been coined abortive infection as phage infection is halted via host suicide thereby preserving neighboring cells from phage attack. DncV and CapV together form the core part of the CBASS system (89). 22 CBASS is encoded in V. cholerae on a genetic island called VSP-1, unique to the current and 7th pandemic strains of V. cholerae El Tor (30, 81). The function of many of the genes encoded on VSP-1 has not been determined. However, it has recently been discovered that VSP-1 harbors a second antiphage defense system mediated by the avcID in which the AvcD cytidine deaminase is activated upon phage infection to deplete dCTP and dCMP nucleotides in the cell, thereby inhibiting phage replication (32). VspR, a transcriptional regulator of CBASS, is also encoded on VSP-1 (30) (Fig. 2.2). One limitation to studies of CBASS antiphage defense is that study of these systems has almost exclusively been carried out in heterologous hosts such as Escherichia coli, and not in the native context in which they are discovered (88, 89, 139, 140). This 23 limitation is because there is often a scarcity of phages available to infect the natural hosts, whether they be V. cholerae or other species, while coli phages are generally more well characterized and available. Moreover, in V. cholerae, there are only three predominant phages found in the environment to infect pandemic strains of V. cholerae, ICP1, ICP2, and ICP3, all three of which our laboratory has shown are not effected by CBASS or the aforementioned AvcID (23, 32, 105). We are therefore interested in studying CBASS defense island in the context of its native V. cholerae host. In that regard, our laboratory has established that sulfamethoxazole treatment can be used as a proxy to study DncV activity in its native V. cholerae context (Severin, in preparation). V. cholerae El tor is 1,000 – 5,000 times more sensitive to sulfamethoxazole than classical V. cholerae (141). The main difference between the two strains being inclusion of VSP-1 and VSP-2 islands in V. cholerae El Tor, we investigated the possibility of VSP-1 being responsible for this difference (26, 81). Sulfamethoxazole inhibits dihydropteroate synthetase which synthesizes dihydropteroic acid, an important precursor of dihydrofolate (DHF) (141, 142). DHF is then synthesized into tetrahydrofolate (THF) (142, 143). DncV is inhibited by THF-like molecules, specifically 5’-methylTHF and 5’-methylTHF-diglutamate, both downstream products of THF synthesis pathways (95). Our lab established that deletion of VSP-1 rendered V. cholerae 24 El Tor more resistant to sulfamethoxazole treatment equivalent to the resistance of classical strains (Severin, in preparation). Thus, we hypothesize sulfamethoxazole treatment reduces the concentrations of intracellular folates that repress DncV, and therefore sulfamethoxazole sensitivity serves as a proxy for DncV synthesis of cGAMP and CapV activation. In this report, we support this hypothesis showing that DncV activation of CapV is the main modulator of sulfamethoxazole sensitivity. I also demonstrate that the accessory CBASS genes vc0180 and vc0181 impact CBASS activation. Given how detrimental it would be if DncV and CapV were unregulated, it is likely they are activated by a specific stimulus that is yet unidentified. The ancillary genes, vc0180 and vc0181 encoded downstream of DncV likely play a role in this regulation by conferring broader phage defense as CapV/DncV homologs only provided protection against 1/10 lytic phage in E. coli whereas addition of VC0180 and VC0181 homologs increased that protection to 6/10 phage (89). The mechanism by which they function is unknown. As VC0180 encodes E1 and E2-like ubiquitin modifying domains and VC0181 encode deubiquitinase-like domain, it is plausible they play a regulatory role in modifying CBASS response via protein modification (31, 89). E1, E2c and DUBs are part of the ubiquitin post-translational modification system (110, 114, 120). However, a vital component of this system is a ubiquitin like protein modifier. A ubiquitin-like protein is activated at its C-terminal -GG or -SG motif via adenylation reaction by E1 enzymes, which then conjugates and passes it on to E2 conjugating enzyme (116, 118–120, 144). In most well studied pathways, the Ubl is then passed onto an E3 ligase that works to identify then ligate the Ubl to its target protein (114, 125). Ubl have been shown to polyubiquitinate a protein to signal it for protease complexes that will degrade the protein (110). However, small ubiquitin like modifiers 25 (SUMO) can modify protein activity without driving degradation following binding to their target substrate, a process called SUMOylation (108, 128, 145, 146). In either system, a DUB-like enzyme exists to counteract this signaling, allowing for protein recycling or reversal of SUMOylation (115, 147). VC0180 and VC0181 likely function similar to their eukaryotic counterparts in activating, conjugating, and reversing those reactions. It is important to note no E3-like ligase enzyme exists in CBASS or V. cholerae; however, Ube2I, a SUMO E2-like enzyme previously called Ubc9, has been shown to directly bind its corresponding ubiquitin to ligate it to its protein targets (148). With the lack of an E3- like enzyme, and the existence of both E1 and E2-like domains, VC0180 might represent an ancestral prokaryotic post-translational system that did not yet diverge into multiple component systems similar to that of eukaryotes. VC0181 encodes a DUB-like domain that might function to reverse the effects of VC0180. In addition to no identified E3 ligase, no Ubl encoding gene has been identified for the V. cholerae CBASS (30, 89). Thus, the exact mechanism of VC0180 and VC0181 regulation, their targets, and the specific modifier remain elusive. It is important that we explore the targets and potential modifiers of these enzymes to better understand how they modulate 3’3’-cGAMP signaling to confer broader phage response. In search of the potential Ubl, we have identified two possible candidates. The most probable is DncV. Ubl proteins are generally small (<100 aa) and have characteristic C-terminal -GG motifs or - SG motifs in some bacteria and archaea. Though DncV is not small, and is a cGAMP synthase, it does have a C-terminus -SG motif similar to Ubl-proteins (118). Ubl domains have also been found encoded within multi-domain proteins to confer Ubl-like functions (147). Thus, it is not unfathomable that DncV might have a Ubl-like C-terminus that might be a target for activation and modification by VC0180 and reversal by VC0181. Another 26 candidate bumo (bacteria ubiquitin-like modifier), a small open reading frame upstream of CapV that I discovered during my thesis research that will be described below. It consists of 284 bases (84 amino acids), consistent with the size range of ubiquitin-like proteins, and its proximity to the CBASS operon makes it a likely candidate as a modifier. The two other flanking genes are less likely, as vspR is a transcriptional regulator of CBASS expression and vc0182 encodes an integrase-like domain (30). Thus, DncV and this uncharacterized gene represent two potential candidates to start studying VC0180 and VC0181 regulation of cGAMP signaling. We have designed both a broad top down, unbiasedand target-specific bottom-up approach to determine whether these two proteins are targets of VC0180 and VC0181 modification. Our laboratory has a transposon library of VSP1 which can be used to screen for mutants that reverse phage susceptibility to non-CBASS harboring levels (82). These selected mutants can be sequenced for identification of mutation sites and further exploration of VSP1 island genes responsible for phage defense. The more targeted approach is to manipulate these genes specifically, creating single mutants and testing phage susceptibility patterns. Our efforts show evidence that bumo is part of the CBASS operon and protects DncV from degradation by VC0181. We also show evidence that VC0181 is not on the same transcript as the rest of CBASS and seems to be working independently of VC0180 in modifying DncV. 27 2.2 Results 2.2.1 El tor CBASS leads to defense against T2, T4 and T5 phages Previous studies on CBASS systems have not utilized the CBASS operon from the laboratory strain of V. cholerae we use, El Tor C6706. To first explore whether this CBASS confers phage defense, a cosmid encoding the VSP-1 island, containing CBASS, which was previously isolated from a genetic screen, was inserted into Escherichia coli DH10B cells, which lack the island (82). This cosmid contains both the upstream and downstream intergenic regions that preserve its natural promoters, terminators, and other regulatory sequences. The VSP-1 harboring E. coli cells were then infected with a panel of 10 E. coli lytic phages to determine whether it conferred any difference in phage susceptibility. We tested host range against double stranded DNA phages: T2, T4, and T6 phage of the Myoviridae family, T3 of the Autographiviridae family, T5, lambda-vir, SECphi18 and SECphi27 of the Siphoviridae family, T7 of the Podoviridae family, and the single stranded DNA phage SECphi17 of the Microviridae family . The VSP-1 island conferred phage defense against multiple families of phages. There was a 1000-fold reduction in T2 phage titer and 100-fold reduction in T5 phage titer (Fig. 2.3). Smaller reductions of ~10 fold were seen for phages T4 and T7. No other phage showed significant difference in comparison to the control empty vector. 28 To determine which part of the VSP-1 island was responsible for the phage defense, the VSP-1 encoding cosmid was subjected to mutagenesis by a Tn5 transposome in vitro, electroporated into E. coli cells, and screened for efficiency of plaquing (EOPs). Cosmids that showed reversion of phage titers to that of empty vector were then sequenced using a primer to the transposon to determine the location of the transposon insertion. There were 54 clones that were sequenced for T2 and T5. There were 15 positive hits in VSP-1, 12 of which had differential EOPS for both T2 and T5 infections positive hits, 1 transposon insertion that altered defense for only T2, and 2 transposon insertions that altered defense for only T5. Most of the transposon insertions in VSP-1 were in the CBASS operon as shown in (Fig. 2.4). Transposon mutations that reduced infection of both T2 and T5 were found in each gene of CBASS including the upstream intergenic region with single hits on other genes of the VSP-1 island. This supports CBASS as the major contributor of 29 phage defense in the VSP-1 cosmid in these conditions. Interestingly, two of the transposon insertions that significantly impaired phage defense were in the upstream intergenic region of capV and vspR upstream of the uncharacterized gene (Fig. 2.4). 2.2.2 An uncharacterized gene, bumo, upstream of capV is part of the CBASS operon The intergenic region between vspR and capV is 913 base pairs. Prediction for putative open reading frames (ORFs) identified an uncharacterized ORF of 284 bp that we named bumo for Bacterial Ubiquitin-like Modifier (Fig. 2.5). To establish whether the upstream intergenic region including bumo is part of the CBASS operon, we extracted 30 RNA and used random hexamers to generate complementary DNA (cDNA). The cDNA was then PCR-amplified using primers spanning two genes to determine whether they were originally encoded on the same mRNA transcript and thus present in an operon. Primers specific for vspR, bumo, CBASS genes, and vc0182 were used to confirm which genes transcribed with CBASS. Genomic DNA was used a positive control for the PCR reaction, and a no reverse transcriptase reaction functioned as a negative control for DNA contamination of the RNA sample. As expected, no bands were seen in reactions using primers spanning vspR and capV (Fig. 2.5). Similarly, no bands were seen for reactions with primers spanning vspR 31 and bumo suggesting that vspR is not part of an operon with CBASS. However, a band of the expected size for primers spanning bumo and capV was amplified from the cDNA, confirming bumo is indeed part of the CBASS operon (Fig. 2.5 yellow box). Bands of the expected sizes amplified from cDNA were seen for genes spanning capV and dncV, and dncV and vc0180 confirming they are on the same transcript (Fig. 2.5). Unexpectedly, no bands were seen when primers spanning vc0180 and vc0181 were used on the cDNA, suggesting they are not on an operon. Interestingly, a band is present when primers spanning vc0181 and vc0182 were used to amplify the cDNA. Since vc0182 is encoded on the opposite strand, these genes are not in an operon, but this result suggests that transcription of vc0181 extends into vc0182. To confirm that the absence of a band from vc0180 to vc0181 is not due to primer error or low abundancy of cDNA, PCR amplifications were repeated for longer cycles and 32 using internal primers to confirm RNA was originally present when the cDNA was generated. Additional cycles resulted in brighter bands on gel electrophoresis for dncV to vc0180; however, no bands were apparent in wells containing the reaction for vc0180 to vc0181 (Fig. 2.6). Moreover, internal primers for both vc0180 and vc0181 yielded positive bands similar to the genomic DNA control suggesting that both genes were transcribed. These results show for the first time that vc0181 is not part of the CBASS operon, and more generally suggest that the DUB component of Type II CBASS systems may not be cotranscribed with the rest of the genes. With evidence that bumo is on the same transcript as capV, dncV, and vc0180 while vc0181 is not, I performed 5’RACE (rapid amplification of cDNA ends) to determine the transcriptional start site (TSS) of the CBASS operon and vc0181. We were unsuccessful in determining the TSS of vc0181 using both vc0181 internal and external primers. However, only one TSS of the CBASS operon was detected using reverse primers that annealed to capV. This TSS was found encoded 210 bp upstream of bumo further 33 supporting the fact that it is part of the CBASS operon (Fig. 2.7). The TSS is situated about midway between vspR and bumo. 2.2.3 Nonsense mutation of bumo leads to sulfamethoxazole resistance To ascertain whether the uncharacterized gene impacts CBASS activation, we first made a strain with bumo deleted and challenged it with sulfamethoxazole. We had established that El Tor strains encoding VSP-1 were more sensitive to sulfamethoxazole treatment (100 µg/mL). This sensitivity is dependent on dncV as deletion of the cGAMP synthase is highly resistant to sulfamethoxazole (Serverin et Al. , in preparation) (Fig. 2.8A). Therefore, sensitivity to sulfamethoxazole can be used as a proxy for CBASS activation to study CBASS in its native V. cholerae context. The Dbumo mutant exhibited earlier and lower cell growth following sulfamethoxazole treatment in comparison to WT strains and ∆dncV, suggesting this mutation increased DncV activity (Fig. 2.8A). 34 We hypothesized that such increased sensitivity upon deletion of bumo could be due to an impact on expression of the CBASS operon. To test this, we extracted mRNA from WT and Dbumo to measure gene expression. qRT-PCR revealed increased 35 abundance in CBASS (two to four-fold) gene transcripts, suggesting the earlier and lower cell death might be a result of increased CBASS transcription in the cells (Fig. 2.8B). There was no difference in CBASS and bumo expression profiles between sulfamethoxazole treated and non-treated cells confirming sulfamethoxazole does not affect CBASS gene expression (Fig. 2.9). To mitigate the effect of the deletion, we made a nonsense mutation to the first codon of bumo (bumoM1*). The nonsense mutant showed bacterial proliferation similar to ∆cbass and ∆dncV strains in contrast to WT strains that showed cell death following sulfamethoxazole treatment (Fig. 2.8C, Fig. 2.10). This result suggests that DncV is not active in the bumoM1* mutant. However, qRT-PCR of extracted mRNA from this mutant showed four-to-sixteen-fold reduction CBASS mRNA at both 2 and 4 hours post treatment and a 32-fold decrease in the transcript of bumo as expected (Fig. 2.8D). This 36 suggests that the proliferation of the cell in the bumo mutant upon sulfamethoxazole treatment might be due to decreased expression of CBASS genes and not a direct result of the deletion mutation. We tried to compliment the nonsense phenotype by introducing a vector overexpressing either the WT or nonsense mutant of bumo. Although the transcript of bumo increased four to eight-fold when expressed from the complementation strain, there was no significant difference in the expression of the other CBASS genes, suggesting that the bumoM1* mutation functions in cis to alter CBASS expression (Fig. 2.11). Even though nonsense mutation of bumo leads to sulfamethoxazole resistance similar to a CBASS deletion, due to its differential effects on CBASS expression, it is unclear how bumo impacts sulfamethoxazole resistance. This would have to be further explored. To assess the role of vc0180 and vc0181 in regulation of DncV activity, I generated a double deletion mutant of these genes. The ∆vco180-vc0181 also showed increased 37 resistance to sulfamethoxazole, similar to ∆dncV, suggesting they might be important together with bumo for enhancing DncV activity (Fig. 2.10). 2.2.4 bumo encodes for a protein To explore the nature of bumo, I fused its coding sequence to an intein-tag containing chitin binding domain expression system and induced its expression in E. coli BL21. I affinity purified the product using chitin resin in a column. The product was cleaved from the intein tag using DTT to purify bumo alone. The resulting elution showed a robust band on SDS-PAGE gel corresponding to its predicted molecular weight of ~10 kDa (Fig. 2.12). 38 Moreover, immunoblotting of the protein fusion tagged with intein-CBD domains with an antibody directed against the CBD detects a similar sized band from cell lysates (Fig. 2.13). Although this is not direct evidence that bumo encodes a protein, we can state that the peptide produced from bumo can be stably expressed in a cell. This gave us some confidence that the uncharacterized gene produces a stable protein rather than a small RNA or substrates. 39 2.2.5 Bumo is stable in vivo To explore whether Bumo is stable in vivo and whether it is impacted by the absence of other CBASS components, a tagged Bumo was cloned into both an intein tag and 6XHis-tag overexpression vectors introduced and introduced into V. cholerae A1552 strains. A1552 encodes an IPTG-inducible T7 RNA polymerase on its genome, allowing induction of proteins using the T7 expression system. C-terminally intein/Chitin Binding Protein (CBP)-tagged Bumo was stable in both WT and ∆dncV strains, with two bands seen in both strains. A lower band corresponding to the intein/CBP tag (~27 kDa) and the Bumo fusion as a monomer (~37 kDa total) (Fig. 2.13A). Analysis of C-terminal 6XHis- tagged Bumo revealed only one major band ~10 kDa in WT, Bumo M1*, ∆vc0181 and bumoM1*∆vc0181 corresponding to the size of a Bumo monomer with the C-terminus 6X-His tag (Fig. 2.13B). Together, this result suggests the stability of Bumo is not dependent on the presence of other components of CBASS. However, Bumo stability remains to be tested in the ∆capV and ∆vc0180 strains. 40 2.2.6 VC0180 and VC0181 stabilize DncV DncV is predicted to be modified by VC0180 and VC0181 in an undefined manner. Before we explore whether bumo works in relation with VC0180 and VC0181, we needed to define their function in cGAMP signaling. We first studied the effect of VC0180 and VC0181 on DncV stability. A 6x histidine C-terminal tagged fusion of DncV that is expressed using the T7 expression system from pET28 was electroporated into various CBASS mutant strains generated in V. cholerae El Tor A1552. DncV was then induced using IPTG in WT, ∆vc180, and ∆vc0181 mutants and detected from cellular lysates via Western immunoblotting using an anti-6X-His primary antibody. 41 The EV control did not generate a signal. Interestingly, there were no bands seen in ∆vc0180 lysates, suggesting VC0180 is vital for DncV stability in the cell. The ∆vc0181 strain showed both DncV bands observed from the WT at higher intensities (Fig. 2.14). The lower ~50 kDa band was darker than the higher band. These results suggests that VC0180 positively influences abundance of DncV while VC0181 negatively influences its abundance. E1 and E2 domains (i.e., VC0180) are the mediators of protein interaction in ubiquitin like networks with DUBs (i.e., VC0181) acting to undo the interaction. Interestingly, the first few amino acids of vc0181 overlaps with the C-terminus of vc0180 with no other potential start codons further downstream for vc0181. This overlap suggests there might be interdependency between VC0180 and VC0181. We tested a double KO 42 strain of VC0180 and VC0181 to determine how complete removal of this system affects DncV abundance. Given the potential role of VC0180 as an activator and conjugator and VC0181 as a negative driver of DncV stability, we hypothesized that DncV would be detected only in its monomeric form. Unexpectedly, no DncV bands were detectable in the ∆vc0180-vc0181 mutant, suggesting that VC0180 is required for the stability of DncV in vivo. Given differential dependency of DncV on these ancillary proteins, we next explored the impact of bumo on its stability. 2.2.7 Bumo is required for DncV stability Following our findings that both VC0180 and VC181 are required for DncV stability and existence in vivo, we tested to see whether bumo also affects DncV stability. Similar to the above experiments, a T7-promoter overexpression vector of 6X-His tagged dncV was electroporated into V. cholerae encoding the Dbumo (Fig. 2.15B)or bumoM1*(Fig. 2.18B) mutations. Induced cell lysates of these mutants were then blotted for DncV using an anti-6X-His primary antibody. Both bumo mutations showed a complete loss of the DncV signal similar to that of ∆VC0180, suggesting a relationship in the role of VC0180 and bumo in the abundance of DncV in the cell (Fig. 2.15 and Fig. 2.18). 43 2.2.8 VC0180 is stabilized by Bumo, DncV and VC0181 Following the dependency of DncV on Bumo and VC0180 and the role of VC0181 in destabilizing DncV, we tested whether VC0180 stability is dependent on the presence of the other CBASS components. A C-terminus 6X-His tagged VC0180 was induced from a T7-promoter vector and cell lysates collected for detection via western immunoblotting. Three major bands were visible in the WT strain: ~70 kDa corresponding to a VC0180 monomer, another between ~130 and 180 kDa and another >~180 kDa suggesting the presence of VC0180 multimers. A DcapV mutant also exhibited similar bands to the WT (Fig. 2.16). 44 However, the ∆vc0180, ∆dncV, ∆vc0180, and ∆VC0181 mutants all showed a very faint band at ~70 kDA corresponding to VC0180 (Fig. 2.16). Given VC0180 is being expressed from a plasmid, it is important to note that the lack of detection is not due to polar effects. However, it is interesting the ∆vc0180 mutant does not look like the WT, as expression of VC0180-His should complement this mutation, suggesting that either deletion of vc0180 is polar on expression of vc0181 or the fusion protein is not functional. To address a possible polar effect of ∆vc0180 on vc0181 expression, we made a first codon nonsense mutant of vc0180. However, the same phenotype was seen in this VC0180 M1STOP mutant (Fig. 2.16). This result suggests translation of vc0180 might be important for expression of vc0181. Overall, immunoblot imaging for VC0180 suggests CapV is not necessary for its stability; however, Bumo, DncV and VC0181 are important in its abundance and stability to form multimers. 45 2.2.9 VC0181 stability is dependent on VC0180 We also tested whether VC0181 stability is dependent on the presence of the other CBASS components. A C-terminus 6X-histidine tagged VC0181 was induced from a pTAC IPTG inducible promoter containing pEVS143 vector (Fig. 2.17A) and T7-promoter vector pET28b (Fig. 2.17B), and cell lysates collected for detection via western immunoblotting. Expression from pEVS143 showed a single faint band slightly larger than ~15 kDa in both the ∆dncV and ∆vc0181 strains with no bands visible for WT, ∆capV or ∆vc0180 (Fig. 2.17A). Due to the poor signal in this expression system, I fused the his-tagged VC0181 into a vector to drive expression using T7 polymerase and electroporated this vector into the V. cholerae A1552 strains. I did subsequent blots of cell lysates from bumoM1* and 46 ∆vc0181. Blot analyses of the WT strain showed a major band >~15 kDa, corresponding to a VC0181 monomer, predicted to be ~18 kDa (Fig. 2.17B). Two fainter bands were seen, one between ~35 kDa and ~45 kDa, and the other between ~55 kDa and ~70 kDa corresponding to possible dimer and tetramer forms of VC0181. There are a couple of higher but very faint bands suggesting a higher order complex of VC0181 (Fig. 2.17B). The Bumo M1* mutant strain and the ∆vc0181 strain had similar bands. Both blots taken together suggests VC0181 stability is dependent on VC0180 and possibly CapV, while its stability is independent of Bumo, DncV, and WT chromosomal VC0181. 2.2.10 DncV is depleted by VC0181 The stability of DncV is impacted by Bumo and VC0180, which are required for DncV to be observed, and VC0181 which reduces its stability. C-terminal 6X-His-tagged DncV is detectable as both a monomer and multimer in the WT and ∆vc0181 strains, with the latter showing greater abundance (Fig. 2.14). However, no bands were visible in the bumoM1* (Fig. 2.14), the ∆bumo (Fig. 2.15B) and ∆vc0180 strains (Fig. 2.14). We thus explored how these ancillary proteins controlled DncV. Given VC0181 encodes deubiquitinase-like domains, we hypothesized that VC0180 and Bumo function together to modify and stabilize DncV while VC0181 undoes that modification to destabilize the enzyme. To test whether VC0180 and Bumo can form a complex with DncV, we tested the ability of DncV to form a multimer in the absence of Bumo, the bumoM1* and ∆bumo mutants, the Dvc0181 mutant, and a double bumoM1*∆vc0181 mutant. In all strains, VC0180 was present, and thus we could test the necessity of Bumo and its effect on DncV stability in the presence and absence of VC0181. C-terminal His-tag fusion analyses showed the presence of both the monomer and two multimer bands in the WT strains 47 (Fig. 2.15B, Fig 2.21C). These DncV bands were missing in the two Bumo mutants, suggesting Bumo is important for DncV existence in vivo (Fig. 2.15B, Fig 2.18B) . The absence of VC0181 lead to a highly intense monomer band ~55 kDa and ~70 kDa with a faint band between ~130 kDa and ~180 kDa (Fig. 2.15, Fig 2.18B). Surprisingly, the double mutant of Bumo and VC0181 revealed major monomer and multiple higher bands, one between ~ 130 kDa and ~180 kDa as seen in WT and ∆vc0181 but also another higher band >~180 kDa (Fig. 2.15, Fig 2.18B). This suggests the formation of higher order structures that do not contain Bumo or VC0181. 48 To confirm these findings, we tested the stability of DncV under its natural expression in these mutants. To do this, I generated a chromosomal C-terminal FLAG tag fusion of DncV under control of its natural promoter and collected cell lysates at similar time points. FLAG-tagged DncV was detected in all strains. The WT strain had a very faint band between ~55 kDa and ~70 kDa corresponding to monomeric DncV. This band was missing in the bumoM1* strain but revived in both the ∆vc0181 and bumoM1*∆vc0181 strains in line with the previous two studies (Fig. 2.18A). The DncV signals were more intense in both ∆vc0181 and BumoM1STOP∆vc0181 compared to WT with ∆vc0181 exhibiting the most intense bands (Fig. 2.18A). This confirms that expression of dncV from its native promoter is unstable in the absence of Bumo and is even more stable in the additional absence of VC0181, confirming our findings expressing DncV from a plasmid. The sum of this evidence support that DncV abundance is depleted by VC0181 and Bumo negatively impacts this interaction. This suggests Bumo is not functioning through VC0180 as the presence of a C-terminus fusion tag on Bumo did not impact its ability to prevent VC0181 degradation of DncV. Given that DncV signal is depleted in vivo for both N-terminal and C-terminal fusions, VC0181 seems to be acting as a protease in degrading DncV regardless of placement of the fusion tag and that the cleavage target of VC0181 might be internal to the protein or nonspecific. We were thus interested in exploring the role of VC0181 as a potential protease. 49 2.2.11 VC0181 is a DncV protease Protein structure predictions by I-TASSER revealed similarities of VC0181 to other protease-like enzymes in both in silico predictions and structural database-based predictions. One well studied candidate found was 26S proteasome regulatory subunit N11-like protein in yeast with well characterized active sites. Amino acid sequence alignment with VC0181 revealed E39, a glutamine residue as an important protease active site residue (not shown). To determine whether VC0181 is a DncV specific protease, I made chromosomal mutation of the active site glutamine residue into alanine (E39A). C-terminus 6X- Histidine-tagged DncV vector was cloned into this strain and compared to the WT strain. The VC0181 E39A strain revealed a band between ~55 and ~70 kDa similar to known monomeric his-tagged DncV. However, no bands were detected in the normal WT strain (Fig. 19). This confirms that VC0181 is a protease that targets DncV. Given previous evidence that DncV stability is enhanced in a double deletion of bumo and vc0181, we then tested whether a double bumo and vc0181 E39A mutant would affect DncV activity. In line with previous results, there were stronger bands in this double mutant (Fig. 19). DncV also exhibited higher order bands in the absence of Bumo or VC0181, supporting previous findings that VC0181 might be acting as a peptidase to deform multimeric complexes and that Bumo is not required for these multimers. 50 Another line of evidence to support that VC0181 degrades DncV is the lack of monomeric DncV isolated when purified from V. cholerae. We have attempted purification of DncV from both BL21 E. coli and A1552 V. cholerae strains both containing IPTG inducible T7 polymerase promoter. Following column elution, a major band corresponding to ~50 kDa can be isolated from Bl21 cells with some contamination of noncleaved fusion protein and the intein-tag itself. However, following protein production, and affinity column purification of cell lysates from V. cholerae A1552 cells, 51 I saw many multiple bands throughout the entire well (Fig. 2.20). After multiple attempts at optimizations and purifications, all attempts at isolating DncV from A1552 was abandoned. However, VC0181 is lacking in E. coli Bl21 but present in V. cholerae A1552, resulting in DncV produced in V. cholerae being degraded by VC0181. This might explain the difficulty associated with purification and the multiple bands seen upon SDS- PAGE staining of purified column elution of V. cholerae isolated samples. 52 2.2.12 Bumo inhibits formation of DncV and VC0180 heteromultimers. VC0181 has a predicted DUB-like domain. NCBI BLAST search of VC0181 showed other DUB-like and protease-like domain containing enzymes. With evidence that Bumo, DncV ,and VC0180 might interact, we explored whether Bumo negatively impacts the interaction of DncV and VC0180. Due to previous findings that DncV in the ∆vc0181 and bumoM1*∆vc0181 mutants was produced at high levels to overshadow the DncV signal from the WT strain, we exposed the Western blot for a longer period of time (13 min vs 30s) to assess DncV signal in strains in which it was not as highly expressed. In the raw immunoblot alone, we can see that three bands exist in WT strains, the lowest corresponding to a possible monomer, a higher band between ~100 kDa and ~130 kDa presenting a possible dimer or multimer of DncV and VC0180 and a third high band >~180 kDa (Fig. 2.21C). No DncV signal was visible even at greater exposure levels for bumoM1* strain (Fig. 2.21C). However, in the ∆vc0181 strain, the two lower bands of DncV were visible (Fig. 2.21A&C). All three bands however can be seen at high intensities in the double bumoM1*∆vc0181 strain (Fig. 2.21C). When we look at VC0180 levels however, we can see that though there are reduced levels in all mutant strains, especially ∆vc0181, the band indicative of a monomer and possible dimers and heteromer are found in all strains (Fig. 2.21D). In ubiquitination pathways, E2 holds onto to their Ubl or substrates until they find an E3 ligase or target protein to pass it onto. With this logic in mind, it is possible, that in the WT strain where both VC0180 and VC0181 exists, you find all three forms of DncV: DncV alone, conjugated to E2, and in a complex with VC0180 and either Bumo or VC0181. 53 54 I have previously established that Bumo and VC0181 remain stable in these strains (Fig. 2.13). Thus, depletion of DncV in the absence of Bumo but with functional VC0180 and VC0181 present show that Bumo protects DncV from degradation. In the double bumoM1*∆vc0181 mutant, detection of DncV and VC0180 yield higher bands of similar MW (Fig. 2.21B&C lane 9, Fig. 2.21D lane 9). This suggests the higher bands might be a heterodimer of DncV and VC0180. That band however is absent in the ∆vc0181 strain where bumo is present but present in the bumoM1*∆vc0181 mutant indicating that Bumo might inhibit the formation of these heteromultimers. I have tested if only the monomeric DncV band is detected in ∆vc0180, but DncV is not produced in this mutant rendering this experiment unfeasible. Thus, further experiments are necessary as DncV is not detected in a ∆vc0180 background (Fig. 2.14 lane 8). 2.2.13 VC0180 specifically utilizes ATP E1-like enzymes activate their target via an adenylation reaction utilizing ATP. We tested whether VC0180 can adenylate target substrates using ATP in vitro by detecting nucleotide triphosphates (NTP) turnover. I purified VC0180 and incubated it with ATP, GTP, CTP and UTP alone and in different combination. In vitro reaction of VC0180 with different triphosphate nucleotides led to turnover of only ATP containing reactions when visualized on silica gel TLC (Fig. 2.22). This confirmed to us that VC0180 is functioning like other E1 like enzymes in vitro in preferentially turning over ATP for possible adenylation reaction. 55 56 2.3 Discussion Bacteria adapt to various environments through cdN signals that regulate numerous adaptive phenotypes (15, 24, 65). V. cholerae must respond to multiple challenges in aquatic conditions including viral predators for survival (102, 105). cGAMP leads to communal protection from phages via abortive suicide of the infected host (89). Upon phage infection, cells are killed by DncV synthesis of cGAMP which leads to membrane degradation by CapV (82). Neither the activating signal nor the full regulation of this signaling network has been identified. Previous work on cGAMP’s antiphage function showed that VC0180 and VC0181, two ancillary genes found, contribute to broader phage defense, hinting at a modulatory function of these enzymes, but their exact mechanism has yet to be determined (30, 85, 89). Moreover, no ubiquitin-like protein or modifier have been identified to be associated with CBASS, though such modifiers are an important component of ubiquitination biochemistry. We identified DncV and an uncharacterized upstream gene we name bumo as potential candidates for VC0180 and VC0181 modifications. DncV has the characteristic ubiquitin -SG motif in its C-terminus while bumo is a small ORF in the vicinity of CBASS. We show that bumo is part of the CBASS operon being present on the same RNA transcript as capV, dncV, and vc0180 (Fig 2.5). Surprisingly, vc0181 is not found to be on the same transcript as the other members of the CBASS operon (Fig. 2.5 & 2.6). It is possible that Q-RT PCR cannot amplify the junction between vc0180 and vc0181 even if they are on the same transcript due to the formation of hairloop pins or other structures that inhibit synthesis of continuous cDNA. However, is it interesting that the transcript of vc0181 extends into the ORF of vc0182 (Fig. 2.5), although on the opposite strand, suggesting transcription termination of vc0181 extends well past the translation stop site 57 for this gene. Given the proximity of bumo to capV and their existence on the same transcript, it is highly possible that VspR also negative regulation bumo expression. In line with our observation, previous studies into VspR, a transcriptional inhibitor of CBASS only showed negative regulation of vc0178 (capV), vc0179 (dncV) and vc0180 (30). This is supported by our observation that vc0181 is not on the same transcript as the rest of CBASS thus is not coordinately regulated. Analysis of Pseudomonas fluorescens SRM1 and Escherichia coli ECOR31 containing orthologous VC0180 and VC0181 ancillary domains show there are also overlaps in the ORFs indicating that differential transcriptional regulation and post-translational interdependency might be evolutionarily conserved. Bumo plays an important role in promoting DncV stability within the cell. A nonsense mutation in bumo decreases CBASS transcription while a deletion of bumo increased CBASS transcription (Fig. 2.8 B & D). These effects on CBASS expression can be seen phenotypically by the survival of nonsense mutants and earlier cell death of deletion mutants (Fig. 2.8 A & C). We were unable to complement these mutations suggesting they impact transcription in cis. Due to the differential effect of the mutation on sulfamethoxazole viability, it is not a suitable model to study the phenotypic effects of Bumo in relation to CBASS. I show that Bumo is a post translational modifier of DncV that protects it from degradation specifically by VC0181. In the WT strain, DncV and its multimeric bands are present, which are completely missing in either the Dbumo or bumo M1STOP mutants(Fig. 2.15 & 2.18). Evidence suggests that VC0181 targets and cleaves DncV as an endopeptidase as both C-terminal and N-terminal tagged fusions of DncV were completely cleared in the absence of Bumo. Furthermore, no peptide fragments 58 containing the tag were visible on the Western blot indicating that the protein was completely degraded. This phenotype is also independent of CBASS expression, as both bumoM1* and Dbumo lack DncV (Fig. 2.15B and 2.18B). It is interesting to note that the same higher bands of DncV are present when detecting VC0180, suggesting these higher bands might be heterodimers and multimers of DncV and VC0180 (Fig. 2.21). It is quite remarkable that DncV is the potential target of VC0180, as most E1 and E2 enzymes target ubiquitin-like small proteins for activation and conjugation as seen by the presence of the same higher bands when detecting for DncV or VC0180. In contrast DncV, is a larger protein and does not encode a predicted Ubl domain like some larger protein. Regulation of DncV by VC0181 also represents a rather novel mechanism of control. Instead of acting like other DUBs that only remove ubiquitin tags from target lysine residues or dissociate E1/E2 enzymes from SUMO, it acts to degrade DncV entirely in the absence of Bumo (115). This is in contrast to other prokaryotic deubiquitinase-like enzyme DOP which remove PUP from its target protein or DRJAMM, another JAMM motif containing DUB-like enzymes, known to cleave MoaD and MoaE (149). It is still possible that VC0181 cleaves VC0180 from DncV, which would be best investigated in vitro using a double terminal tagged DncV, in which cleavage of the C-terminal end would not affect its ability to be detected using an antibody against the N-terminal tag. Bumo seem to be inhibiting formation of even higher order multimers of DncV and VC0180 as seen by the lack of high bands in the ∆vc0181 mutant but presence in the double BumoM1STOP, ∆vc0181 mutant (Fig. 2.21). Though the DncV and VC0180 heterodimer is present, the higher multimeric bands are not (Fig. 2.21). E1 enzymes MoaD and ThiS generally function as a homodimer, so it is likely that the two higher bands represent a single DncV and VC0180 heterodimer and then a dimer of DncV and 59 VC0180 heterodimers (129, 131, 132). This is supported by the fact that these higher multimeric bands are present in both DncV and VC0180 immunoblots. The absence of the band indicative of a dimer of dimers in the presence of Bumo but absence of VC0181 (∆vc0181 strain) imply that Bumo inhibits the formation of the DncV-VC0180 dimer. The purpose of this inhibition might be to avoid targeting by VC0181. This can easily be tested with complementation of VC0181 in vivo. In vitro, these individual components can be combined to determine whether these higher order structures form based on MW band shifts of different combinations. We can also test VC0181 endopeptidase activity by adding purified enzyme and seeing whether DncV his-tag signal disappear. VC0180 preferentially utilizes ATP, supporting that it might be activating its substrate with adenylation (116). This is also further supported by the fact that the fusion of the C-terminus of DncV to a rather bulky 3X FLAG TAG does not lead to a formation of the higher bands comprising the multimer. However, the use of the smaller His-tag, the higher bands are formed. CBASS is an important antiphage defense system for bacteria to counteract phage in different environments. Though activating signals of CBASS have not been identified, we now know that activity of the DncV is regulated at multiple levels. VspR regulates capV, dncV, and vc0180 expression at the transcription level while at the post-translational level, VC0180 and Bumo stabilize DncV with the latter also protecting DncV from degradation by VC0181, a DncV-specific protease. We can further appreciate the diversity in ubiquitin-like post translational modification systems in bacteria to regulate vital pathways include that of 3’3’-cGAMP and the importance of cyclic di nucleotide signaling in bacterial survival. 60 2.4 Materials and Methods Strains and Growth Conditions Escherichia coli DH10b (Invitrogen) was used for cloning, protein expression, transposon library screening and phage plaque assays. Vibrio cholerae El Tor strain C6706 and its genetic derivatives were used for sulfamethoxazole treatment assay, mRNA extractions, cGAMP quantification and protein expression. E. coli BL21(DE3) was used for protein expression and production. V. cholerae A1552 and its genetic derivatives were used for protein production and expression studies. All strains were grown in 2 mL of Luria-Bertani (LB) broth (0.5 % yeast extract, 1% tryptone, 1% NaCl, pH 7.5) overnight at 37˚C unless indicated otherwise. LB agar medium was used with 1.5% agarose. Antibiotic selections were at the following dosages: Ampicillin [100 µg/mL], Kanamycin [100 µg/mL], Gentamycin [10 µg/mL], Spectinomycin [100 µg/mL], Sulfamethoxazole [100 µg/mL], and Trimethoprim [30 µg/mL]. Mutations For DNA base substitutions, we followed the previously established method called SPRINP as described (150). Plasmids harboring our gene(s) of interest were PCR amplified in sets of twos with each containing either a forward or reverse base substituted primer. They were then combined into a single reaction tube and incubated as described. They were then diluted with water on dialysis membranes and electrotransformed into E. coli DH10B cells as described previously (82). They were then recovered in 500 µL of SOC for 1 – 2 hours and spread on LB selective agar. They were incubated overnight, and positive colonies were PCR amplified and sequenced for the mutagenesis. 61 Genetic manipulation in V. cholerae El Tor was done following chitin-competent transformation and EXO-MUGENT protocol (151–153). V. cholerae strains were grown overnight then sub-cultured (1:100) into fresh LB media. They were grown to an OD600 of 0.5 – 0.9 and collected at 15,000 g for 5 min. The top media was aspirated, and the pellet resuspended in 1X Instant Ocean (IO). In a 1.5 mL Eppendorf tube, 150 µL of chitin slurry, 750 µL of 1X IO, and 100 µL of resuspended cultures were mixed together and incubated at 30˚C for 18 hrs. For plasmid transformations, 0.5 µL to 1 µL were added and mixed to transform into cells. For EXO-MUGENT ~3000 bases upstream and downstream of the gene of interest were PCR amplified and ligated together using Gibson cloning. They were then PCR amplified using internal primers for the combined product. For genomic editing, 10 – 15 µL of ligated PCR are added along with 1 µL of either Spec or Trim PCR amplified antibiotic cassettes which replaces Vc01807 a frameshifted transposase on the chromosome. For gene KI or replacement, primers spanning ~2-3 kb US to ~2-3 kb DS of the gene of interest are used to amplify the region then added along with 1 µL of PCR amplified antibiotic cassettes. The mixture is incubated at 30˚C for 18 hours. 500 µL of LB was added to each mixture and transferred into sterile glass tubes. The culture was the grown at 37˚C for 2-3 hours shaking at 210 rpm. Each culture (300 – 500 µL) was then spread on LB selective agar plates and incubated at 30˚C overnight. Isolated colonies were checked for insertions or deletions via colony GoTaq PCR and checked on agarose gel. Colonies with appropriate deletion or insertion size were confirmed via sequencing. 62 Phage plaque assay E. coli DH10B clones were challenged with T2, T3, T4, T5, T6, T7, lambda vir, SECphi17, SECphi18 and SECphi27 coli phages. Subsequent infection studies utilized T2 and T5 phages. Overnight culture of E. coli DH10b was subcultured 1:1000 into 10-15 mL of LB and grown to OD600 of 0.02 – 0.08. 250 µL of grown culture was transferred into a new test tube and mixed with liquified 15 – 18 mL of MMB agar. The mixed solution was immediately poured into large agar plate and swirled to cover the entire surface. While the plate solidifies, we prepared 10-fold dilutions of each phage to be tested. After solidification of the agar, we added 5 µL of each phage dilution onto the agar surface. The plate was incubated for 18 hours and then observed for plaque formation. mRNA extraction and cDNA synthesis Overnight cultures were back diluted to OD600 = 0.01 in 3 mL of LB and grown in experimental conditions in large glass tubes. Sample cultures (1.5 mL) were collected at designated timepoints and pelleted at 15,000 g for 5 min at RT. The solution was aspirated, and pellet was redissolved in 1 mL of Trizole (InvitrogenTM Cat.# 15596018). mRNA was extracted following manufacturer guidelines to “Isolate RNA.” Extracted mRNA were quantified by NanoDrop and stored at -20˚C. DNA contaminations in the mRNA extract was removed using TURBO DNA-freeTM Kit (InvitrogenTM Cat.# AM1907) following manufacturer protocol. cDNA was synthesized using the SuperScript™ III Reverse Transcriptase (InvitrogenTM Cat.# 18080044) following manufacturer protocol. Prepared 2x volume for each RNA sample 63 to be used in reverse transcription reaction and half-volume no reverse transcriptase control reactions. qRT-PCR qRT-PCR of cDNA samples were performed using SYBR™ Green PCR Master Mix (Applied Biosystems™ Cat.# 43-091-55) following the manufacturer’s protocol. No template and no RT samples were included as negative controls and gyrA as a positive control for standardization of results. Sample data was collected using QuantStudio3 following previous methods (67, 68). 5’ RACE 5’ rapid amplification of cDNA ends (5’ RACE) was carried out using Template Switching RT Enzyme Mix (NEBTM Cat.# M0466L) following the manufacturer’s protocol. For CBASS 5’ RACE, Random Primer (PromegaTM Cat.#C1181) was used for the initial second strand synthesis and CapV qPCR Rvs primer was used for cDNA PCR amplification and sequencing. For Vc0181 5’ Race, Vc0181 qPCR Rvs Primer and Vc0182 US primers were used for cDNA PCR amplification. Sulfamethoxazole Treatment Assay Overnight cultures were back diluted to OD600 of 0.01 in 3 mL of fresh LB liquid media in large glass tubes. Cultures requiring induction were inoculated with IPTG [100 µg/mL]. They were grown for 1 hr at 37˚C shaking at 210 rpm then treated with sulfamethoxazole [100 µg/mL]. Cultures were continuously grown and collected every 64 two hours until the 6th hour post treatment. The OD600 was recorded for each sample at their designated time points. Protein Purification Proteins of interests were purified using the IMPACTTM Kit (NEBTM Cat.# E6901S) following the manufacturer’s protocol. Each gene was cloned to be expressed by a T7 polymerase promoter containing AmpR vectors. For C-terminal fusions of intein-CBD tag, the gene was cloned into the pTXB1 vector at the NdeI and SapI restriction sites. For N- terminal fusion of intein-CBD tag, genes were cloned into the SapI site of pTYB21 vector. Overnight cultures of E. coli BL21 or V. cholerae A1552 were sub-cultured 1:1000 into 1 L of LB liquid media supplemented with Amp (100 µg/mL). They were then grown shaking at 210 rpm for at 37˚C until OD600 reached ~0.500. The culture was then induced with IPTG (100 µg/mL) and grown shaking at 210 rpm for 4-6 hours at 30˚C or 37˚C depending on the optimal conditions determined for each protein. Bumo was induced for 4 hrs at 30˚C. DncV and Vc0181 were induced for 6 hrs at 30˚C. And Vc0180 was induced for 6 hrs at 37˚C. The cultures were then collected in 300 mL centrifuge bottles and pelleted at 5000 g for 1 5 min at 4˚C. Cells were harvested in 100 mL of column buffer (20 mM HEPES/Tris-HCl, 500 mM NaCl pH 8.5) and homogenized using Microfluidics M-110P. They were then spun down at 25,000 g for 20 min at 4˚C. A chitin resin (NEBTM Cat.# S6651S/L (20 mL/100 mL) column was prepared, and samples were prepared and loaded following manufacturer recommendation. Cleavage buffer consisted of 100 mM DTT in the column buffer used. Loaded columns were incubated at either 4˚C or 23˚C for 40 – 60 hours before elution. A third of the column volume was eluted using column buffer. Eluted proteins were quantified by Quick Start™ Bradford 1x Dye 65 Reagent (BIO-RADTM Cat.# 5000205) following manufacturer protocol using BSA standards. Protein samples were concentrated using Amicon® Ultracentrifugal Filters following manufacturer protocol. Utilized Amicon® Ultra-4 Centrifugal Filter Units 3 kDa (MilliporeTM Cat.# UFC800324), 30 kDa (MilliporeTM Cat.# UFC801024) and 50 kDa (MilliporeTM Cat.# UFC805024) and Amicon® Ultra-15 Centrifugal Filter Units 3 kDa (MilliporeTM Cat.# UFC900308), 30 kDa (MilliporeTM Cat.# UFC903008) and 50 kDa (MilliporeTM Cat.# UFC905008). SDS-PAGE Each sample was mixed with 4X SDS-PAGE dye (200 mM Tris-HCl pH6.8, 8% SDS, 4.3 M glycerol, 6 mM bromophenol blue) with or without 400 mM DTT to 1X and boiled at 60˚C for 10 min or 95˚C for 5 min. Boiled samples were loaded onto 4–20% Mini-PROTEAN® TGX™ Precast Protein Gels, 10-well, 50 µl (BIO-RAD Cat.# 4561094) and ran for 90 min at 90 V constant or 30 min at constant 200 V. Samples were run in 1X TG-SDS PAGE running buffer (10X buffer: 0.2501 M Tris base, 1.924 glycine, 0.03467 M SDS). PageRuler™ Prestained Protein Ladder (Thermo ScientificTM Cat.# 26616) was used for all gels. Protein Gel Coomassie Staining Protein gels were stained using Coomassie stain (0.1% Coomassie Brilliant blue, 50% MeOH, 10% glacial acetic acid (vol/vol), 40% water) overnight shaking at 100 rpm at RT. They were then decanted and de-stained using warm destaining solution (20% MeOH, 10% glacial acetic acid in water) overnight. Destaining solution was refreshed at 66 least once when solution seemed concentrated. Images were captured using a personal iPhone 12 Pro Max. Western Blot Overnight cultures were sub-cultured (1:100) into LB liquid broth (3 mL) and grown until the OD600 reached ~1.0. Cultures needing induction were induced with IPTG (100 µg/mL) before subculture. Samples (1.5 mL) were centrifuged at 15,000 g for 3 min and aspirated. Pellets were normalized in 1X Dulbecco’s Phosphate Buffered Saline (SIGMATM Cat.# RNBH5973) to an OD600 of 1.0. Protein gels were transferred onto nictrocellulose membranes in transfer buffer (20% MeOH in 1X Tris-glycine buffer) for 2 hrs at 250 Amp using a transfer sandwich or semi-dry blot transferred at 250 Amp for 1 hr. Membranes were blocked in 5% milk 1X TBST (1% Tween20, 200 mM Tris, 1.5 M NaCl pH 7.6) for 2 hours shaking at 100 rpm at RT. Each membrane was then blocked-in primary antibodies diluted into 5% milk 1X TBST buffer at 4˚C for 18 hrs. They were then rinsed in 1X TBST buffer for 5 min shaking at 100 rpm at RT three times. Membranes needing secondary antibody treatment were blocked in secondary antibody mixed into 5% milk 1X TBST solution for 2 hours at RT shaking at 100 rpm. They were then rinsed in 1X TBST buffer for 5 min shaking at 100 rpm at RT three times. Membranes were imaged using the Amersham Imager 600 using semi-auto exposure to determine optimal exposure periods. Images were prepared using Microsoft Powerpoint and BioRender. His-tag proteins were detected using Mouse IgG αHA (GenScript® Cat.# A00186) at 1:5000 and αFLAG HRP-conjugated antibody (GenScript® Cat.# A01869) was used at 1:5000 in 5% milk 1X TBST solution. αRabbit/Mouse IgG H&L (HRP) (abcamTM Cat.# 67 ab6721/ab97023) secondary antibody was diluted 1:5000 in 5% milk 1X TBST solution. Bands were detected using Pierce™ ECL Western Blotting Substrate kit (Thermo ScientificTM Cat.# 32106) following manufacturer protocol. cGAMP Extraction Overnight cultures were sub-cultured (1:100) into LB liquid broth (3 mL) in large glass tubes and grown shaking at 210 rpm at 37˚C until the OD600 reached ~1.0. Cultures needing induction were induced with IPTG (100 µg/mL) before subculture. Samples (1.5 mL) were centrifuged at 15,000 g for 3 min and aspirated. They were resuspended in 100 µL of cdN extraction buffer (MeOH:MeCN:water:0.1N Formic acid, 4:4:4:0.2, vol:vol:vol:vol) and incubated at -20˚C for 30 min. The samples were centrifuged at 15,000 g for 5 min then the supernatants were transferred into new 1.7 mL Eppendorf tubes. The solvent was then evaporated via rotary evaporation overnight. Dried pellets were stored at -20˚C until needed for mass spec. Sulfamethoxazole samples were prepared following the appropriate protocol and harvested following the protocol above. We made 10-fold dilutions of each culture and pipetted 5 µL of each onto LB plates to determine CFU/mL of our cells. Mass Spectrometric Analysis Following cGAMP extraction as outlined above, thawed pellets were resuspended in 100 µL of HPLC-grade water. They were then analyzed on a Waters Xevo TQ-S using UPLC/MS/MS at the Michigan State University Mass Spectrometry and Metabolomics Core following the method outlined in Massie et al (154). cGAMP was detected with 68 electrospray ionization in negative ion mode at m/z 674.1 -> 337.05. Standards were prepared ranging from 1.95 to 125 nM. Nucleotide Utilization Assay Purified Vc0180 [5 µM] was mixed with MgCl2 [2.5 mM], ATP [1.25 mM] in column buffer (20 mM HEPES, 500 mM NaCl pH 8.5) in a PCR reaction tube for a total of 50 µL. No enzyme control and no substrate controls were also prepared of equal total volume. They were incubated at 37˚C for 16 hrs before TLC analysis. Thin Layer Chromatography Samples were blotted (10 µL) onto silica gel thin layer chromatography glass plates (250um, THICK, 60A. FLUORES. IND.) (ANALTECH Cat. C09622) with nucleotide standards (1 µL of 10 mM or 100 mM) . The TLC plate was incubated with running solution of n-propanol:ammonium hydroxide:water in 11:7:2 (vol:vol:vol) until the solvent front was about 1 cm from the top. The plate was then dried and visualized under UV light (254 nM). Images were captured using an iPhone 12 Pro Max and prepared using BioRender and Microsoft PowerPoint. 69 CHAPTER 3: DETERMINING THE FUNCTION OF CBASS IN OTHER GRAM-NEGATIVE BACTERIA For this research portion of my thesis, many individuals were involved. John Dover and Kendal Tinney of the Parents laboratory conducted growth curves and phage infection assays of Escherichia coli EDEC13E, Vibrio cholerae 2631-78 and Pseudomonas fluorescens SRM1 CBASS systems cloned into Shigella flexineri Y and Salmonella typhimirium. Q-TOF Mass Spectrometry was performed by Dr. Anthony Schilmiller of the Michigan State University Metabolomics and Mass Spec Core. 70 3.1 Introduction Bacterial pathogens must adapt to a wide variety of environmental conditions and predators. Phages compete with their bacterial hosts in these micro-environments to ensure their survival (23). This constant battle between the bacterial host and phage for survival has led to the development of various phage defense mechanisms. Many bacteria have acquired the cyclic-oligonucleotide based antiphage system (CBASS) that utilizes a nucleotide signal to mediate a phage infection response (31, 89). These systems have two core components comprising of the cGAS/DncV-like nucleotidyltransferase (CD-NTase), and its effector that senses and is activated by that signal (31, 89). Activation of the effector typically results in the death of the infected host, a process termed abortive infection, in which the infected host commits altruistic suicide to prevent infection of its clonal neighbors (31, 89). Majority of CBASS have these two primary components while many have ancillary domains hypothesized to play a crucial role in enhancing or regulating the CBASS response. CBASS has been categorized into Types 1, 2, 3 and 4 (31). Type 1 comprises just the synthase and effector, type 2 comprises E1/E2-like and JAB-like ancillary domains, type 3 comprises HORMA/TRIP13 ancillary domains, and Type 4 contains potential nucleotide modifying enzyme domains (31). There are variations of Type 2 and Type 3 called Type 2 short in which only a short E2-like domain is found and no E1 or JAB-like domains and Type 3 double HORMA domains (31). Putative CD-NTases have been identified in all bacterial phyla (31, 85, 88, 89). cGAS and DncV, cGAMP synthase in humans and bacteria, respectively, belong to 2 of 7 clades of the CD-NTase protein family divided based on 10% nucleotide sequence similarities (31, 85, 89, 100). CD-NTases are also classified into two different classes 71 based on whether they sense and respond to DNA or RNA or are inhibited by folate-like molecules (95, 98, 100). cGAS and OAS, a cGAS-like enzyme that produces 2’5’- oligoadenylate (pApA), are activated upon sensing double stranded DNA and RNA, respectively (31, 97, 155, 156). DncV, a cGAS-like bacterial structural homolog was shown to have folates attached in the corresponding backbone region and was active in the absence of such stimuli. In vitro screening of numerous bacterial CD-NTases showed synthesis of nucleotide signaling molecules utilizing all four nitrogenous bases, contrary to the previous thought that only purines were utilized as was the case for cyclic di-GMP, cyclic di-AMP, and cyclic GMP-AMP (cGAMP) (86, 100). CD-NTases can make a diverse set of cyclic oligo-nucleotides as in vitro c-di-UMP is synthesized by CdnE02 from Legionella, cUMP-AMP (cUAMP) and cAAA are synthesized by CdnD and CdnC from various E. coli strains (86, 88, 139). This demonstrates that we are just beginning to understand the variety of cdN produced in bacteria and their role in bacterial physiology. CD-NTases are found associated with a variety of potential effector domains. There are phospholipases, transmembrane domains, endonuclease domains, toll-like receptor domains, phosphorylase domains, and peptidase domains (31). Abortive infection has been observed experimentally in some of these systems but only hypothesized in others (87–89). The majority of Type 1 and 4 systems harbor transmembrane-like domains, type 2 systems encode patatin-like phospholipase domains, while type 3 system harbor endonucleases (31). There are three subtypes of endonucleases found throughout the CBASS system, the HNH-SAVED endonuclease family, NucC nuclease family, and the PF14130 family (31). It is interesting to note that patatin-like phospholipases, such as CapV, and HNH-SAVED endonuclease have only been identified in Type 1 and Type 2 systems (31, 88). Patatin-like phospholipase is the most abundant effector followed by 72 HNH-endonuclease (31). My primary goal of this research is to explore the function of the HNH-SAVED effector and its relation to the Type 2 CBASS systems. Though cells may harbor CBASS systems, is it not definitive that they are active. Given the vastness of CBASS, we have developed a funnel approach in first screening for active systems before further investigating their cyclic di-nucleotide (cdN) and effector function. Active CBASS systems were screened by induced expression in heterologous hosts. Previous studies in our lab have shown overexpression of El Tor CBASS led to reduced viability of the host (Severin, in preparation). Thus, we first acquired three different strains of Gram-negative bacteria harboring yet uncharacterized CD-NTases associated with HNH-SAVED effectors of Type 1, Type 2, and Type 2 short. These three CBASS systems came from Pseudomonas fluorescens strain SRM1 (157)(Type 2B), Escherichia coli strain EDEC13E (158) (Type 1), and Vibrio cholerae strain 2631-78 (159) (Type 2 short B). These CBASS operons were inserted into inducible vectors and cloned into E. coli DH10b and Shigella flexinerii cells to see whether they impacted cell viability. Moreover, we also studied phage infection by infecting the host with coli or ICP phages expressing the CBASS under their native promoter in their respective hosts. ICP phage 1, 2, and 3 are the dominant Vibrio cholerae specific phages isolated from pandemic cholera samples (23, 105). The reason for their dominance in both clinical samples and environment are not yet known, but they are seen to infect pathogenic V. cholerae. For those CBASS systems with activity, we tried to determine their signaling nucleotide used in vitro via mass spectrometry and thin layer chromatography. In vitro detection of novel nucleotides was achieved via purification of the CD-NTase and synthesis of the novel nucleotide in vitro with different combinations of nucleotide triphosphates (NTPs): ATP, UTP, CTP and GTP. The resulting reaction would be analyzed 73 via thin layer chromatography for a novel product band and quantitative time-of-flight mass spectrometry to determine novel mass profiles. For this chapter of my thesis, I identify active CBASS systems in E. coli EDEC13E and P. fluorescens SRM1 that affect heterologous host viability. We show EDEC CD- NTase utilizes ATP and GTP in vitro to produce two novel bands on TLC screening while P. fluorescens utilizes ATP in vitro to produce novel bands that are susceptible to phosphatase treatment. We show that EDEC HNH-SAVED endonuclease activity is inhibited by the addition of cdN synthesis mixture, indicating a possible inhibitory role of the unidentified cdN. 74 3.2 Results 3.2.1 In vivo overexpression P. fluorescens SRM1 and E. coli EDEC13E CBASS systems leads to growth defects in heterologous systems Overexpression of DncV and CapV within an E. coli heterologous host leads to growth defects caused by membrane degradation. I assessed whether the putative CBASS systems from other species impacted growth within E. coli DH10B cells. I constructed overexpression vectors of P. fluorescens SRM1, E. coli EDEC13E (EDEC) and V. cholerae 2631-78 (VC) CD-NTase systems and overexpressed these in a laboratory strain of E. coli DH10b or Shigella flexneri serotype Y. Overexpression of the putative CD-NTase signaling operon of P. fluorescens resulted in growth inhibition in E. coli DH10b (Fig. 3.1F), but no observable growth changes were observed with EDEC and V. cholerae operons (Fig. 3.1C & 3G). However, the Parent Laboratory observed growth defect upon overexpression of the EDEC CD- NTase operon in S. flexneri (Sf) (Fig. 3.1B). These results suggest that the P. fluorescens and EDEC CD-NTase are active in different species of bacteria and warrants future analyses within multiple bacterial species. Preliminary analyses identified these two out of three CD-NTase operons exhibiting in vivo activity in a heterologous system, supporting screening of in vivo activity as a useful tool in prioritizing CD-NTase systems for further study. 75 76 3.2.2 HNH-SAVED is the main perpetrator of P. fluorescens SRM1 CBASS growth inhibition Following growth defects seen with overexpression of P. fluorescens CBASS in E. coli DH10B cells, we next investigated which component was the main perpetrator of the phenotype. Given the core component are the CD-NTase and HNH-SAVED effector, I generated active site mutations in the HNH motif of HNH-SAVED and the DND aspartic active sites of the CD-NTase then tested these individual component impacts on cell viability. The CD-NTase D79A, D81A active site mutant showed growth defect in its heterologous host similar to that of CBASS expressing cells (Fig. 3.2). The HNH H56A H91A mutant however, showed cell viability similar to that of EV control. This suggests HNH-SAVED is active in the cell and is the main cause of the cell viability defect seen in WT CBASS and HNH-SAVED expressing cells and this effector, when overexpressed, does not need a functional synthase to produce a cdN. A double mutant displayed no discernible cell viability defect. 77 78 3.2.3 Pseudomonas fluorescens SRM1 CD-NTase utilize ATP as substrates in vitro When purified P. fluorescens CD-NTase was incubated with nucleotide triphosphates (NTPs), only ATP containing reactions showed novel bands, indicating formation of a novel cyclic or linear oligonucleotide (Fig. 3.3A). These bands were similar to c-di-UMP and slightly higher than c-di-GMP. Interestingly it was much lower than c-di-AMP. The top band was lined up with the AMP control on subsequent TLCs, as well as 5’-pApA-3’. Reactions with novel bands were repeated and treated with Calf intestinal phosphatase (CIP) to determine whether the cdN products contain any free 5’ or 3’ phosphates or P1 nuclease, a ssDNA and RNA nuclease with 3’- phosphomonoesterase activity to determine whether they have 3’-5’ cyclic peptide bonds. These ATP utilizing products were CIP-susceptible indicating they have 5’ or 3’ exposed phosphates with complete turnover into a single band corresponding to adenosine (Fig. 3.3B). C-di-AMP was not susceptible to CIP treatment as expected. We are continuing to enhance Quadrupole Time-of-Flight mass spectrometry (Q-TOF MS) detection methods and analyses to identify the masses of our products. 79 80 3.2.4 E. coli EDEC13E CD-NTase utilize ATP and GTP as substrates in vitro I observed novel bands on silica gel TLC upon screening the products of EDEC CD- NTase incubated with the NTPs ATP, GTP, CTP and UTP. The EDEC CD-NTase showed two novel bands on TLC (Fig. 3.4A). The first band product was observed when GTP was added as a substrate, with ATP and GTP yielding increased band intensity. This product was resistant to CIP and P1 cleavage indicating that it might be cyclized without a 3’5’ peptide bond and does not have an exposed 5’ and 3’ phosphate (Fig. 3.4 & 3.5). This band did not align with any known dinucleotide standards suggesting it might be a novel cdN. The second band product utilized ATP as the main substrate and was CIP- susceptible (Fig. 3.4B). This band however was later shown to be aligning with ADP, indicating the possibility of an ADP by-product being detected instead. 81 Many CD-NTase enzymes function in the presence of dimetal compounds, thus EDEC activity was tested in the presence of different divalent cations. EDEC activity is dependent on MgCl2 and MnCl2 based on the nucleotide turnover analysis on TLC however, ZnSO4, NicL2 and CuCl2 also showed some levels of nucleotide turnover (Fig. 3.6). 82 83 3.2.5. Mass spec analyses show E. coli EDEC13E CD-NTase metabolize ATP and GTP into AMP, ADP, GMP and GDP. To confirm our in vitro findings that the EDEC CD-NTase utilizes ATP and GTP to form a novel product, we analyzed the nucleotide synthesis reaction using Quadrupole Time-of-flight mass spectrometry in which all mass of compounds in a sample are identified. We prepared two controls, an enzyme control with ATP and GTP added but no enzyme and a substrate control with enzyme, but no substrates added to discern ATP, GTP and novel signals from background reaction mixture. Our reaction mixture that contained both ATP, GTP, and the purified CD-NTase showed reduction of ATP and GTP substrates with signals for AMP, ADP, GMP and GDP detected (Fig. 3.7). There was a greater reduction of the ATP signal compared to the GTP signal. In depth analyses for specific masses predicted for potential cyclic di nucleotides showed no signals corresponding masses of c-di-AMP, c-di-GMP, cGAMP or pApA. Moreover, no obvious signal indicating a novel compound was detected, keeping the identity of the nucleotide formed unknown. The signals at times ~1.6 and ~1.88 with mass 237.09 are HEPES molecules. We performed a similar analysis of P. fluorescens CD-NTase reaction which showed depletion of ATP signals and a rise in AMP signals. However, no novel signals were detected (Data not shown) 84 3.2.6 EDEC HNH-SAVED is inhibited by addition of the CD-NTase reaction molecule Following the unsuccessful studies to identify the nucleotide signaling molecule, we then investigated whether the HNH-SAVED domain encoded an actual endonuclease. To investigate this, we first incubated purified EDEC CD-NTase with NTPs to synthesize the cdN product. I then mixed this reaction product with purified EDEC HNH-SAVED enzyme in the presence of double stranded DNA substrates to determine whether any 85 degradation occurs. The resulting reaction were analyzed by gel electrophoresis to determine if there were any changes to the nucleic acid substrate (Fig. 3.8). The substrate only control showed three major bands indicating the presence of the pTXB1 vector in different colloidal forms (lane 1). Upon the addition of purified HNH-SAVED, this led to the appearance of multiple bands and smearing on the gel (lane 2). Unexpectedly, the incubation of the cdN products synthesized in vitro with HNH-SAVED led to the prominence of just one band with faint bands (lane 4) similar to the endonuclease absent control mixture (lane 3). This was observed a total of three times. 86 87 3.2.7 T2 and T5 Coli Phages infect E. coli EDEC13E but ICP phages do not infect V. cholerae 2631-78. To better study these CBASS systems in their native environment, we explored whether any of our phage libraries infect CBASS-containing hosts. EDEC13 E. coli cells were vulnerable to T2 and T5 phage infection suggesting this CBASS system is not sufficient to block these phages from infecting. V. cholerae 2631-78 CBASS showed no indication of an active system based on viability phenotype. Thus, we tested whether any of ICP phages could infect the cell. ICP 1, 2 and 3 were not able to infect V. cholerae 2631- 78, indicating invulnerability of the cell to ICP phages. Whether or not the inability of ICP 1 to infect V. cholerae 2631-78 is due to the CBASS system or other factors such as receptor expression remains to be determined. 88 3.3 Discussion CBASS is found in all bacterial phylogenies, including many human pathogens (89, 160). Having to adapt to various environments, they have acquired various genetic islands including various phage defense systems (23, 102). CBASS has a nucleotide synthase which produces a cyclic dinucleotide/oligonucleotide signaling molecules that binds to and activates an effector which kills off the host (86, 100). This form of abortive infection, though detrimental to the host altruistically saves the rest of the surrounding bacterial community. The CD-NTase family is a unique family of cyclic di nucleotide synthases as they produce a wide array of nucleotide molecules, from cyclic di purines or pyrimidines to tricyclic nucleotides, utilizing all four nucleotide peptides (86, 100). These CD-NTase are found associated with various effectors, the most common of which are patatin-like phospholipases in bacteria, with the original founding member of this family, CapV, being discovered in V. cholerae (31). CapV binds its cognate nucleotide signal and becomes activated to degrade the cellular membrane killing the host (82). However, many CD- NTases are also found associated with effectors containing a SAVED domain that is a specific nucleotide receptor (85, 88). These are fused to effector domains such as endonucleases and proteases. The majority of SAVED domains are found fused to HNH- endonuclease domains which cleave host and phage nucleic materials upon binding of a nucleotide signal to the SAVED receptor domain (88). This effector differs from the CapV mode of abortive infection but helps us better appreciate the diversity in CBASS effector response. These systems are found in many instances with ancillary domains hypothesized to regulate or aid the CD-NTase. HORMA/TRIP13 systems attach to their cognate CD- NTase to upregulate cdN synthesis and TRIP13 removes HORMA, suppressing the 89 activation (107). The majority of ancillary systems discovered are E1/E2-like ubiquitin like systems which have not yet been characterized (31, 89). Prior to the publication of the mechanism of HNH-SAVED effectors, we started looking into these effector domains to determine their functions. We decided to explore this HNH-SAVED effector containing system with and without full and short length E1/E2 ancillary domains to better understand their function. The majority of CBASS systems are predicted to be inactive in vitro (107). Thus, we started our investigation by identifying potentially active systems in vivo. Previous studies in our lab discovered that overexpression of V. cholerae El Tor CBASS in E. coli led to decreased cell viability or earlier death. Thus, we overexpressed EDEC, V. cholerae 2631- 78 and P. fluorescens SRM1 CBASSs in an E. coli or S. flexneri heterologous host and found P. fluorescens CBASS inhibited growth in E. coli cells while the EDEC CBASS inhibited growth in S. flexinerii cells. We then took P. fluorescens SRM1 and made inactivating mutations to either the CD-NTase or HNH domain to determine which component was responsible for this growth inhibition. Mutation of P. fluorescens SRM1 CD-NTase active sites D79A and D81A did not rescue CBASS expressing cells from growth inhibition. However, active site mutation to the HNH domain H56A H91A alone rescued the sick phenotype to that comparable to the empty vector control. As expected, a double mutant of HNH-SAVED and the CD-NTase led to a similar growth phenotype to that of HNH-SAVED single mutant and EV control. Inhibition of cdN synthesis should affect its potential effector activity, however, none was seen in the CD-NTase mutant. Thus, it would be important to verify that the mutations made for the CD-NTase did inactivate the resulting cdN synthesis activity. Another explanation is that the HNH-SAVED domain binds to multiple 90 signaling nucleotides. Even if it binds to similar signaling nucleotides at a lower affinity, in a highly induced state, there might be enough HNH-SAVED activated to induce abortive infection. Since both EDEC13E and SRM1 CD-Ntases showed active phenotypes, we purified and investigated their nucleotide products in vitro. Each enzyme was incubated with various combinations of NTPs and analyzed on TLC and MS for novel products. On TLC analysis, EDEC CD-NTase exhibited two novel bands different from the NTP and boiled enzymes controls. Thus, reactions were treated with CIP phosphatase and P1 nucleases to further characterize their molecular structure. The lower band product of EDEC appeared in any ATP containing reactions. This band was susceptible to CIP treatment but not P1 nuclease treatment. This result indicated that it was a linear nucleotide with exposed phosphates with a resulting product that lined up with adenosine on TLC. The higher band however was only present in GTP containing reactions with the combination of ATP and GTP yielding much higher yields. This product was resistant to both CIP and P1 nuclease treatment, indicating that it does not have any exposed phosphate groups and that it may be a cyclic compound without 3’-5’ peptide bonds. Mass spectrometric analysis confirmed both observations above with turnover of both ATP and GTP substrates with a greater utilization of GTP. This result indicates that the novel product may be a cyclic nucleotide containing both AMP and GMP with non 3’5’ phosphodiester bond resistant to P1 cleavage. It does not seem to be 3’3’-cGAMP nor 2’3’-cGAMP as both are susceptible to P1 cleavage at both or one bonds, and they appear lower on TLC than our band of interest. Thus, this novel compound might be a multimer or represent a non-phosphate containing cdN. Mass spec analyses thus far have detected turnover of ATP and GTP, including phosphorylation of HEPES in the buffer solution, however, no novel signal has 91 been identified. This novel signal could be hidden by the presence of even greater signals in the retention time zone. Thus, we could narrow our search by changing buffer conditions to not include HEPES and analyzing the band in isolation rather than the entire reaction to determine to accentuate the signal of interest. A different approach would be to treat the reaction with a 2’5’ phosphomonoesterase to determine whether there is an atypical cyclic peptide bond. P. fluorescens SRM1 yielded products only in reactions containing ATP as substrates. Treatment of these reaction with CIP led to complete turnover to bands similar to adenosine, with no effect from P1 nuclease treatment. This result indicates that the product of P. fluorescens SRM1 is a linear oligonucleotide containing 5’ or 3’ exposed phosphates or is just a hydrolyzed ADP product of ATP. It is difficult however to discern one from the other on TLC as ADP has similar Rf values to pApA or pAp. When we subjected the reaction products to mass spectrometry analysis, we saw complete turnover of ATP and ADP, however, no novel signals were apparent. Similar to the above analyses, this novel signal could be hidden by the presence of even greater signals in the retention time zone or hard to discern due to the similarity in mass to ADP. For future mass spectrometry studies, we hope to obtain samples of pAp and cAMP and compare those standards and retention time to both ADP and pApA to better discern signals. We also hope to utilize HPLC columns before mass spec analyses to better separate molecular compounds and detect specific signals. We were interested in studying these systems in their native environments. Thus, we tested to see whether our phage libraries for E. coli and V. cholerae could infect EDEC and V. cholerae 2631-78 expressing strains. EDEC was infectable with T2 and T5 phages, however, ICP phage did not form any plaques upon infection of V. cholerae 2631-78. This 92 result could be because CBASS may be active and prevent phage infection. Thus, we are planning to make a CBASS deletion strain to see if they are vulnerable to ICP phage infection. We will next determine whether T2 and T5 phage vulnerability of EDEC is mediated by EDEC CBASS by making CBASS deletion mutants and investigating whether that mutation changes its vulnerability to T2 and T5 phage infections. A broader approach would be to utilize transposon mutagenesis to identify gene disruptions that would change the plaque phenotype. We are also working on expanding our coli phage library to identify CBASS vulnerable phages. As no ICP phages infected our V. cholerae 2631-78 strain, we are working on acquiring other Vibrio phages from other laboratories and also isolating environmental Vibrio phages from water samples where V. cholerae is endemic. We currently do not have any Pseudomonas phages, thus are working on acquiring these phages from potential collaborators. In the meantime, we have inserted the SRM1 CBASS operon with its US and DS intergenic regions to preserve native promoter and regulatory sequences into a non-inducible plasmid and cloned these into E. coli host to study phage infection with coli phages. However, there has been difficulties in plaque assays as the top agar forms a hazy, uneven layer. This can be mitigated by doing liquid broth phage infection assays, in which we infect liquid cultures of bacteria and track their growth OD600 overtime. Due to the higher potential of EDEC CBASS forming a cyclic nucleotide signaling molecule, we investigated whether this product is the activator of its cognate HNH- SAVED. Incubation of purified EDEC HNH-SAVED with plasmid substrates in vitro led to smearing and appearance of multiple bands on agarose gel indicating the enzyme to be active and an endonuclease. However, the disappearance of such smearing when the nucleotide synthesis reaction is added to the mixture, and the similarities of the band to 93 the vector only and no nuclease controls indicate the possibility that the signaling nucleotide is actually an inhibitor of HNH-SAVED. If this held true for other HNH- SAVED containing systems, this would make sense of the phenotype seen with our P. fluorescens SRM-1 CD-NTase mutant, which had no effect on heterologous host viability. If the inhibiting signal is no longer there, the HNH-SAVED could continue to cause death of the host under high expression. This would not explain the lower viability in the WT SRM1 CBASS expressing cells. There is the possibility that due to overexpression, there might more HNH-SAVED than can be inhibited by signaling nucleotides thus causing death of the cells. It will be important to purify P. fluorescens SRM1 HNH-SAVED and perform a similar assay to determine whether nuclease activity is detected without the addition of the nucleotide signaling product. We would like to note that HNH-SAVED mechanism and regulation was investigated and published in 2019 (88). Similar to our observations, HNH-SAVED target double stranded nucleic acids. Au contraire, they saw that the signaling nucleotides produced by cognate CD-Ntases activated the HNH-SAVED, leading to disappearance of the nucleic acid on electrophoretic gel (88). They identified the nucleotide recognition sites of the SAVED domain, thus we hope to introduce SAVED mutations into EDEC HNH-SAVED and determine whether its nuclease activity is indifferent of the nucleotide signal (88). Given that EDEC CBASS contains two components compared to the four component CBASS systems of SRM1 in our study and Enterobacter cloacae and Acinetobacter baumannii used in the HNH-SAVED study, HNH-SAVED could have evolved to respond differentially to signaling nucleotides. Our study confirms that not all CBASS systems are active at least in the conditions we tested, and investigations are limited by the availability of CBASS responsive phages. 94 However, we can better appreciate the evolution of CBASS in different species with some like EDEC maintaining just two core components, but P. fluorescens SRM1 acquiring ancillary domains. Each CD-NTase utilizes and produces signaling molecules different from each other. However, both exhibit abortive infection-like phenotypes further supporting the preservation of CBASS function across phyla but its adaptation to specific microenvironments and nucleotide availabilities. Further investigation is needed to confirm that all HNH-SAVED are activated by their cognate signaling nucleotides and further characterize their regulation in simpler two component CBASS systems. 95 3.4 Materials and Methods Strains and Growth Conditions Escherichia coli DH10b (Invitrogen) was used for cloning, protein expression, transposon library screening and phage plaque assays. E. coli BL21(DE3) was used for protein expression and production. EDEC13E and V. cholerae 2631-78 were used for phage infection assays. All strains were grown in 2 mL of Luria-Bertani (LB) broth (0.5 % yeast extract, 1% tryptone, 1% NaCl, pH 7.5) overnight shaking at 210 rpm. EDEC13E and V. cholerae 2631- 78 were grown at 37˚C and P. fluorescens SRM1 was grown at 30˚C. LB agar medium was used with 1.5% agarose. Antibiotic selections were at the following dosages: Ampicillin [100 µg/mL], Kanamycin [100 µg/mL] and Sulfamethoxazole [100 µg/mL]. Generation of site-directed mutations For DNA base substitutions, we followed the previously established method called SPRINP as described (150). Plasmids harboring our gene(s) of interest were PCR amplified in sets of twos with each containing either a forward or reverse base substituted primer. They were then combined into a single reaction tube and incubated as described. They were then diluted with water on dialysis membranes and electrotransformed into E. coli DH10B cells as described previously. They were then recovered in 500 µL of SOC for 96 1 – 2 hours and spread on LB selective agar. They were incubated overnight, and positive colonies were PCR amplified and sequenced for the mutagenesis. Viability Assay Overnight cultures of DH10b cells harboring different CBASS operons on vectors, were back diluted to OD600 of 0.01 in 3 mL of fresh LB liquid media in large glass tubes. Cultures requiring induction were inoculated with IPTG [100 µg/mL]. Cultures were continuously grown at 37˚C shaking at 210 rpm and measured every two hours until the 8th hour. OD600 was recorded for each sample at their designated time points. Protein purification I purified full recombinant E. coli EDEC13E and P. fluorescens SRM1 CD-NTase via intein-tag purification (IMPACTTM Kit New England Biolabs (NEB) #E6901S). Each gene was cloned into pTXB1, a T7 polymerase promoter containing AmpR vectors at NdeI and SapI restriction sites for C-terminal fusion of intein-CBD tag in E. coli DH10b cells. Overnight cultures were sub-cultured 1:1000 into 1 L of LB liquid media supplemented with Amp (100 µg/mL) and grown to OD600 ~0.500 at 37˚C shaking at 210 rpm. They were then induced with IPTG (100 µg/mL) and grown further shaking at 210 rpm for 4-6 hours at 30˚C or 37˚C depending on the optimal conditions determined for each protein. EDEC and P. fluorescens CD-NTase were induced for 4 hrs at 30˚C and EDEC HNH-SAVED was induced for 6 hrs at 37˚C. The cultures were then collected in 300 mL centrifuge bottles and pelleted at 5000 g for 15 min at 4˚C. Cells were harvested in 100 mL of column buffer (20 mM HEPES/Tris-HCl, 500 mM NaCl pH 8.5) and homogenized using Microfluidics M-110P. They were then spun down at 25,000 g for 20 97 min at 4˚C and the supernatant was loaded on to a chitin column. A chitin resin (NEBTM Cat. # S6651S/L (20 mL/100 mL) column was prepared, and samples were prepared and loaded following manufacturer recommendation. Cleavage buffer consisted of 100 mM DTT in the column buffer used. Loaded columns were incubated at either 4˚C or 23˚C for 40 – 60 hours before elution. A third of the column volume was eluted using column buffer. Eluted proteins were quantified by Quick Start™ Bradford 1x Dye Reagent (BIO- RADTM Cat. # 5000205) following manufacturer protocol using BSA standards. Protein samples were concentrated using Amicon® Ultracentrifugal Filters following manufacturer protocol. Utilized Amicon® Ultra-4 Centrifugal Filter Units 30 kDa (MilliporeTM Cat. # UFC801024) and 50 kDa (MilliporeTM Cat.# UFC805024) and Amicon® Ultra-15 Centrifugal Filter Units 30 kDa (MilliporeTM Cat.# UFC903008) and 50 kDa (MilliporeTM Cat.# UFC905008). SDS-PAGE Each samples was mixed with 4X SDS-PAGE dye (200 mM Tris-HCl pH6.8, 8% SDS, 4.3 M glycerol, 6 mM bromophenol blue) with or without 400 mM DTT to 1X and boiled at 60˚C for 10 min or 95˚C for 5 min. Boiled samples were loaded onto 4–20% Mini-PROTEAN® TGX™ Precast Protein Gels, 10-well, 50 µl (BIO-RAD Cat.# 4561094) and ran for 90 min at 90 V constant or 30 min at constant 200 V. Samples were run in 1X 98 TG-SDS PAGE running buffer (10X buffer: 0.2501 M Tris base, 1.924 glycine, 0.03467 M SDS). Protein Gel Coomassie Staining Protein gels were stained using Coomassie stain (.1% Coomassie Brilliant blue, 50% MeOH, 10% glacial acetic acid (vol/vol), 40% water) overnight shaking at 100 rpm at RT. They were then decanted and de-stained using warm destaining solution (20% MeOH, 10% glacial acetic acid in water) overnight. Destaining solution was refreshed at least once when solution seemed concentrated. Images were captured using mobile device. In vitro Nucleotide Synthesis Reaction Purified protein [5 µM] was mixed with MgCl2 [2.5 mM], ATP/GTP/CTP/UTP each at [1.25 mM] in different combinations in column buffer (20 mM HEPES, 500 mM NaCl pH 8.5) in a PCR reaction tube (50 µL total). No enzyme control and no substrate controls were also prepared. They were incubated at 37˚C for 16 hrs before TLC analysis. In vitro Nuclease Reaction Nucleotide synthesis reaction mixture was incubated with purified EDEC HNH- SAVED [5-10 µM], MgCl2 [5 mM], nucleic acid substrate (1000 ng total) in 50 µL total. A no nucleotide signal control and no HNH-SAVED controls were supplemented with column buffer for a total of 50 µL. The reaction mixture was incubated at 37˚C for 1 hr then mixed with 6X EZ-Vision® Dye-as-Loading Buffer (VWR® Life Science Cat. #: 99 97063-166) to 1X. Each sample was run on 1% agarose gel for 45 min at 120 V and visualized using BIO-RAD ChemiDOC XRS+ Imager. Thin Layer Chromatography Samples were blotted (10 µL) onto silica gel thin layer chromatography glass plates (250um, THICK, 60A. FLUORES. IND.) (ANALTECH Cat. C09622) with nucleotide standards (1 µL of 10 or 100 mM). The TLC plate was incubated with running solution of n-propanol: ammonium hydroxide:water in 11:7:2 (vol:vol:vol) until the solvent front was about 1 cm from the top. The plate was then dried and visualized under UV light (254 nM). Images were captured using an iPhone 12 Pro Max and prepared using BioRender and Microsoft PowerPoint. Mass spectrometry Samples were analyzed using a Waters Acquity UPLC interfaced with a Waters Xevo G2-XS QTof mass spectrometer. 10 uL of sample was injected onto a Waters Acquity BEH-C18 UPLC column (2.1 x 100mm) held at 40°C. Compounds were separated using the following gradient run a 0.3 ml/min: initial conditions were 99% mobile phase A (8 mM dimethylhexylamine and 2.8 mM acetic acid in water) and 1% mobile phase B (methanol), linear ramp from 1% B at start to 40% B at 10 min, return to 1% B at 10.5 min and hold at 1% B until 15 min. Compounds were ionized by electrospray ionization operated in negative ion mode with a capillary voltage of 2.0 kV, cone voltage at 35V, source temperature of 100°C, desolvation temperature of 350°C, cone gas flow of 40 L/hr and desolvation gas flow of 600 L/hr. Data were acquired over the m/z range of 50-1500 using an MSe method with switching of collision energy (20-80V ramp) on and off 100 between successive 0.2 second scans. Lockmass calibration was performed using leucine enkephalin as the lockmass compound. Phage plaque assay EDEC13E clones were infected with T2, T3, T4, T5, T6, T7, lambda vir, SECphi17, SECphi18 and SECphi27 coli phages. Subsequent infection studies utilized T2 and T5 phages. V. cholerae 2631-78 was challenged with ICP1, ICP2, and ICP3 phages. Overnight cultures of bacteria were subcultured 1:1000 into 10-15 mL of LB and grown to OD600 of 0.02 – 0.08. Grown culture (250 µL) was then transferred into a new test tube and mixed with liquified 15 – 18 mL of MMB agar. The mixed solution was immediately poured into a large agar plate and swirled to cover the entire surface. While the plate solidified, we prepared 10-fold dilutions of each phage to be tested. After solidification of the agar, we added 5 µL of each phage dilution onto the agar surface. The plate was incubated for 18 hours and then observed for plaque formation. 101 CHAPTER 4: BROADER IMPACTS AND CONCLUSIONS 102 Bacteria and phage continuously compete for survival in a wide variety of environments (60, 61, 68, 102). This can be seen by the rise of antimicrobial resistance in the community in response to increased antibiotic usage through communal spreading of advantageous genes (134, 161–163). It is thus crucial that we better understand some of these advantageous genetic elements to better design antimicrobial therapeutics amidst the rise of antibiotic resistance across the world. Antibiotic resistance is far outpacing the production of novel antibiotic therapeutics. Thus, taking advantage of bacterial predators such as phages to treat infections, known as phage therapy, might prove an effective method of bacterial control (164, 165). However, bacteria have several antiphage defense mechanism such as CRISPR-CAS, exonucleases, and CBASS to name a few (23, 31, 88, 89, 139, 160, 166, 167). To aid in the design of effective phage therapies, it is crucial we understand the defense mechanisms that might render them ineffective. I sought to explore the regulation of cGAMP in the antiphage response of CBASS systems to help develop better antiphage therapeutics that bypasses these defense pathways. Specifically, I explored the role of the ancillary VC0180 and VC0181 proteins on the stability of DncV in vivo and investigated how they interact with a previously uncharacterized gene I named bumo in modulating DncV stability. I then broadened our horizons by investigating whether CBASS abortive functions are conserved in three other Gram-negative bacteria and exploring the function of another common effector, the HNH-SAVED nuclease, in abortive infection. I also attempted to identify the nucleotide signaling molecules synthesized by these uncharacterized CD-NTases and determine how they affect HNH-SAVED activity. 103 In chapter 2 of my theses, we identified a novel component of CBASS operon, a small uncharacterized gene named Bumo for bacterial ubiquitin-like modifier, upstream of CapV. Similar proteins or genes have not been identified before in other CBASS systems. I confirmed that bumo is on the same transcript as capV, dncV and vc0180 in the CBASS operon. This was further supported by 5’ RACE which identified the transcriptional start 104 site upstream of bumo, midway between vspR and bumo. Contrary to our previous understanding of the CBASS operon in Vibrio cholerae El Tor, I discovered that vc0181 may not be on the same transcript as the rest of the CBASS operon but still play a crucial role in modulating that response. Unlike other deubiquitinase domain encoding enzymes, my data suggest vc0181 is a DncV-specific protease, as mutation of the active site glutamine (E39A) led to restoration of DncV in vivo. However, the protease does not seem to specifically target the N- or C-terminus of DncV, as fusions of histidine, intein nor a FLAG tag affected the decrease of DncV we observed in the presence of WT VC0181. Moreover, my data suggests that Bumo is somehow inhibiting VC0181 from degrading DncV, as the absence of Bumo leads to a disappearance of DncV. How Bumo inhibits VC0181 remains to be determined. My in vivo western blot analyses suggest that DncV and VC0180 form a heterodimer and higher order complexes of these dimers as evidenced by the presence of similar non-monomeric bands when detecting for either protein. However, contrary to my hypothesis that the interaction is happening at the C-terminus, the higher order structures form regardless of whether a tag is fused in the C or N terminus indicating other interactive sites might exist. In Chapter 3, I explored the role of HNH-SAVED effector containing CBASS systems in other Gram-negative bacteria. We identified both E. coli EDEC13E and P. fluorescens SRM1 containing active CBASS systems that affected viability of heterologous hosts. However, we are still optimizing conditions to identify the exact molecules they form. EDEC CD-NTase synthesizes two products on TLC analyses, one, an ATP produced compound susceptible to phosphatase CIP and a second product produced from ATP and GTP indifferent to CIP and 3’-phosphomonoesterase, indicating a possible cyclic molecule with non 3’5’-phosphopeptide bond. P. fluorescens SRM1 HNH-SAVED is the 105 main cause of the lower cell viability in heterologous host as H56A H91A mutations restored EV-like growth, but a CD-NTase mutation had no effect. The EDEC HNH- SAVED nuclease was active in vitro in the absence of the signaling nucleotide, which instead inhibited nuclease activity as seen on agarose gel electrophoresis. My data suggests, contrary to four component CBASS systems, the simpler two component EDEC CBASS system might regulate its effector differently. HNH-SAVED nucleases have been shown to be activated by cyclic tri-AMP in the Enterobacter cloacae CBASS system. Similar to our observations, HNH-SAVED targeted cleavage of double stranded nucleic acids (88). However, in our in vitro analysis of EDEC HNH-SAVED, we saw the loss of activity with the addition of nucleotide signal. They identified the nucleotide recognition sites of the SAVED domain, and thus we hope to introduce SAVED mutations that would disrupt signal binding into EDEC HNH-SAVED and determine whether it is active and unaffected by the addition of the signaling nucleotide. Given that EDEC CBASS encodes two components compared to the four component CBASS systems of SRM1 in our study and Enterobacter cloacae and Acinetobacter baumannii used in previous study, HNH-SAVED could response differentially to signaling nucleotides (87–89). Thus, we would made mutants P. fluorescens SRM1 HNH-SAVED and hope to determine whether its activity is affected by the addition of a nucleotide signal. My PhD thesis thus far adds further complexity to the regulation of DncV in the CBASS response. What we once thought was only a VC0180 and VC0181 regulated system, I discovered a third component, Bumo, that inhibits VC0181 from degrading DncV. Even though bumo-like genes have not been identified in extensive analyses of CBASS operons across species, we are encouraged by the fact that all studies that show functional CBASS 106 systems included large US and DS intergenic regions of CBASS in their expression constructs, and such constructs containing the V. cholerae El Tor CBASS included bumo (89). However, the presence of Bumo is not enough to protect DncV from degradation from VC0181 as genetic deletion of VC0180 leads to the absence of DncV. This suggests VC0180 is important in either stabilizing DncV or mediating Bumo’s inhibition of VC0181. My evidence suggests that VC0180 and VC0181 are interdependent as they are not detected in vivo in the absence of each other. Interestingly, although dncV is detectable with ∆vc0181, it is no longer detectable when both vc0180 and vc0181 are deleted. Thus, the loss of DncV in the ∆vc0180 mutant is not suppressed by the deletion of vc0181, suggesting they are functioning as two separate regulatory mechanisms. An important question is how does VC0181 and Bumo work in conjunction with VC0180 to stabilize and regulate DncV activity. To better decipher these regulatory networks, we would be better investigating these as two separate regulatory networks. Even though we have evidence suggesting DncV and VC0180 form heteromers, additional studies are needed to confirm they do indeed form heterodimers and higher order complexes. This could be achieved by performing co-immunoprecipitation of either DncV or VC0180 and analyzing their bound substrates via mass spectrometry. These multimers can be separated by size and charge using HPLC then quantifying the mass of each individual signal using mass spectrometry to identify the protein substrate. To determine whether VC0180 affects DncV activity, we would quantify the synthesis of cGAMP in vivo in both WT and VC0180 KO mutants. Another approach would be to study 107 DncV and VC0180 in vivo in a heterologous host without CBASS. This would insulate them from the regulatory effects of Bumo and VC0181. To better study the regulation of Bumo and VC0181, we could introduce them along with DncV into a heterologous system not containing CBASS to avoid the effects of VC0180 and repeat Western blot analyses to determine how the presence of Bumo affects DncV stability. My data suggests Bumo inhibits VC0181 via an unknown mechanism. One possibility is that Bumo binds to VC0181. However, my western blot analyses do not show evidence that Bumo binds to VC0181 directly. Their interaction might be limited due to the inclusion of c-terminal tags. This could be mitigated by utilizing the N-terminal tagged VC0181 for studies or utilizing a Bumo or VC0181 specific antibody. We did try to generate a Bumo specific antibody; however, the resulting sera bound nonspecifically with the presence of bands in both WT and ∆bumo strains. We sent column purified elution of VC0181 for proteomic analysis, which did not detect any Bumo. Proteomic analysis of DncV elution was inconclusive as DncV was not the major product detected, and many other products were also detected. This is not surprising because V. cholerae purified DncV has many bands with smears on SDS-PAGE gel further suggesting degraded DncV. We plan to confirm VC0181 specificity for DncV by incubating purified VC0181 with different protein substrates in vitro and observing degradation on SDS-PAGE gels. Another possibility is that Bumo protects DncV from VC0181 proteolysis via direct binding, as SUMO does for mammalian cGAS. This could be tested by observing whether DncV forms a heavier MW band in vivo when Bumo is present in a heterologous host lacking CBASS and VC0181. To address the possibility that VC0180 might be mediating their interaction, the interaction of DncV and Bumo will be investigated also in the presence of VC0180. Since JAMM-like deubiquitinases depend on Zn2+ for activity, we 108 are interested in determining whether this holds true by assessing VC0181 protease activity in vitro with zinc supplementation. CBASS’ function as an abortive infection system is thought to be conserved throughout all CBASS containing systems (31, 100). This has been illustrated many times for CBASS systems from E. coli, V. cholerae, E. cloacae and P. aeruginosa among many others. Moreover, the function of ancillary domain HORMA/TRIP13 is conserved between E. coli and P. aeruginosa CBASS in which HORMA activates Cdn synthesis by binding to the CD-NTase and TRIP13 dissociates the two (107). Thus, it is plausible that the function of VC0180, E1/E2-like enzyme, and VC0181, deubiquitinase-like enzyme is conserved in other systems. As mentioned previously they have not identified a small ubiquitin like modifier encoded nearby. Our transposon mutagenesis and 5’RACE led us to explore Bumo as a potential small ubiquitin-like modifier. It is important to establish whether other bumo-like small open reading frames (ORF) exist within intergenic regions of other four component CBASS systems with ubiquitin-like ancillary domains. Working with collaborators who are experts in bioinformatic analysis, we hope to determine whether bumo-like ORFs exist in other CBASS and will be crucial to establish if it is a conserved feature of four component CBASS systems with ubiquitin like ancillary domains. Preliminary analysis of P. fluorescens SRM1 CBASS intergenic region show multiple contenders, but none have sequence similarity to El Tor bumo. Moreover, similar to VC0180 and VC0181, the C-terminus of P. fluorescens E1/E2-like gene overlaps with the first few amino acids of the deubiquitinase-like domain encoding gene out of frame, indicating a high chance those two ancillary domains function differentially similar to that of El Tor. It would be important to verify the functions of VC0180-like and VC0181-like enzymes of P. fluorescens using similar experiments to that used for El Tor to 109 demonstrate their functions and regulation of CD-NTase is conserved across bacterial species. Our work not only raises new questions to explore in studying CBASS, but also establishes a standard screening method for active CD-NTases using silica gel thin layer chromatography. The majority of CBASS studies utilized radiolabeled nucleotide triphosphates to determine protein activity and identify cyclic/oligonucleotide products on TLC. Though that technique remains much more sensitive than silica gel TLC, it is more time consuming and labor intensive. Silica gel TLC is comparably safer with faster results demonstrating nucleotide turnover. Thus, using purified CD-NTase we can quickly screen for NTP substrate specificity and identify potential products using standards. This technique is limited to a screening tool as some molecules such as c-di-UMP, pApA and ADP appear similarly on silica gel TLC and thus require further analysis to identify the exact molecular make-up of the compound. We were the first to study the function of CBASS under its native host condition. Though we were not able to demonstrate CBASS responsive phage infection in Vibrio cholerae El Tor, we investigated the nature of DncV stability and its regulation under its native host environment. Other studies clone CBASS systems into heterologous hosts with the advantage of there being established phage libraries and isolation of these systems from other modifiers in the host. Our studies in the V. cholerae El Tor’s natural genomic context add confidence to our findings that VC0181 is a DncV specific protease and allowed for the discovery of bumo, a previously uncharacterized part of CBASS. Our data 110 also add novel understanding that VC0181 might not be on the same operon as the rest of CBASS but still crucial in regulating the response. Overall, my work greatly advances our understanding of DncV regulation. I have identified bumo as a missing piece of the ubiquitin-like CBASS system. I demonstrate that VC0181 is a DncV protease and not just an endopeptidase as its deubiquitinating domain suggests and illustrate Bumo inhibits VC0181 proteolysis of DncV via an unknown mechanism. I also demonstrate that VC0180 and VC0181 differentially regulate DncV, with VC0181 working with Bumo to regulate proteolysis of DncV and VC0180 forming multimers with DncV with yet to be defined effects, contrary to traditional ubiquitin-like systems in which E1/E2 like and DUB-like enzymes would be expected to oppose each other’s functions on the same target. DncV is inhibited by folate-like molecules binding to a protein cleft on the opposite side of the active site; however, we establish DncV regulation is more complex with two more degrees of control: 1) Bumo and VC0181 control DncV levels via proteolysis, and 2) VC0180 binds to DncV to potentially affect its activity. Our findings would help better understand how to exploit these defense networks for antibacterial activity and design more effective phage-carrier genetic therapies that bypasses these defense networks to establish clinically significant effects on human 111 REFERENCES 1. Clemens JD, Nair GB, Ahmed T, Qadri F, Holmgren J. 2017. Cholera. Lancet 390:1539–1549. 2. Ali M, Nelson AR, Lopez AL, Sack DA. 2015. Updated global burden of cholera in endemic countries. PLoS Negl Trop Dis 9:1–13. 3. Ganesan D, Gupta S Sen, Legros D. 2020. Cholera surveillance and estimation of burden of cholera. Vaccine 38:A13–A17. 4. Pérez-Reytor D, Jaña V, Pavez L, Navarrete P, García K. 2018. Accessory toxins of vibriopathogens and their role in epithelial disruption during infection. Front Microbiol 9:1–11. 5. Herrington DA, Hall RH, Losonsky G, Mekalanos JJ, Taylor RK, Levine MM. 1988. Toxin, toxin-coregulated pili, and the toxR regulon are essential for Vibrio Cholerae pathogenesis in humans. J Exp Med 168:1487–1492. 6. Crawford JA, Krukonis ES, DiRita VJ. 2003. Membrane localization of the ToxR winged-helix domain is required for TcpP-mediated virulence gene activation in Vibrio cholerae. Mol Microbiol 47:1459–1473. 7. Miller VL, Taylor RK, Mekalanos JJ. 1987. Cholera toxin transcriptional activator ToxR is a transmembrane DNA binding protein. Cell 48:271–279. 8. Gill DM, Meren R. 1978. ADP-ribosylation of membrane proteins catalyzed by cholera toxin: basis of the activation of adenylate cyclase. Proc Natl Acad Sci U S A 75:3050–3054. 9. Mekalanos JJ, Swartz DJ, Pearson GDN, Harford N, Groyne F, De Wilde M. 1983. Cholera toxin genes: Nucleotide sequence, deletion analysis and vaccine development. Nature 306:551–557. 10. Miller VL, Mekalanos JJ. 1984. Synthesis of cholera toxin is positively regulated at the transcriptional level by toxR. Proc Natl Acad Sci U S A 81:3471–3475. 11. E. GS, N. BK, H. KB, C. BR, Jackson SM. 1994. Cystic Fibrosis Heterozygote Resistance to Cholera Toxin in the Cystic Fibrosis Mouse Model. Science (80- ) 266:107–109. 12. Field M, Fromm D, al-Awqati Q, Greenough WB. 1972. Effect of cholera enterotoxin on ion transport across isolated ileal mucosa. J Clin Invest 51:796– 112 804. 13. Keusch GT, Jacewicz M. 1977. Pathogenesis of shigella diarrhea: VII. Evidence for a cell membrane toxin receptor involving β1 → 4-linked N-Acetyl-D-glucosamine oligomers*. J Exp Med 146:535–546. 14. Fry SK. 1992. Cholera Prevention and Control : Water Supply , Sanitation 2030:2–4. 15. Krukonis ES, DiRita VJ. 2003. From motility to virulence: Sensing and responding to environmental signals in Vibrio cholerae. Curr Opin Microbiol 6:186–190. 16. Faruque SM, Biswas K, Nashir Udden SM, Ahmad QS, Sack DA, Balakrish Nair G. 2006. Transmissibility of cholera: In vivo-formed biofilms and their relationship to infectivity and persistence in the environment. Proc Natl Acad Sci U S A 103:6350–6355. 17. Boin MA, Austin MJ, Häse CC. 2004. Chemotaxis in Vibrio cholerae. FEMS Microbiol Lett 239:1–8. 18. O’Hara BJ, Barth ZK, McKitterick AC, Seed KD. 2017. A highly specific phage defense system is a conserved feature of the Vibrio cholerae mobilome. PLoS Genet 13:1–17. 19. Meibom KL, Blokesch M, Dolganov NA, Wu CY, Schoolnik GK. 2005. Microbiology: Chitin induces natural competence in vibrio cholerae. Science (80- ) 310:1824–1827. 20. Yadav M, Pal K, Sen U. 2019. Structures of c-di-GMP/cGAMP degrading phosphodiesterase VcEAL: Identification of a novel conformational switch and its implication. Biochem J 476:3333–3353. 21. Escudero JA, Mazel D. 2017. Genomic Plasticity of Vibrio cholerae. Int Microbiol 20:138–148. 22. Azman AS, Rudolph KE, Cummings DAT, Lessler J, Tor E. 2014. The incubation period of cholera: A systematic review supplement. J Infect 66:432–438. 23. Yen M, Camilli A. 2017. Mechanisms of the evolutionary arms race between Vibrio cholerae and Vibriophage clinical isolates. Int Microbiol 20:116–120. 24. Waters CM, Lu W, Rabinowitz JD, Bassler BL. 2008. Quorum sensing controls 113 biofilm formation in Vibrio cholerae through modulation of cyclic Di-GMP levels and repression of vpsT. J Bacteriol 190:2527–2536. 25. Alam M, Kasan NA, Sadique A, Bhuiyan NA, Ahmed KU, Nusrin S, Nair GB, Siddique AK, Sack RB, Sack DA, Huq A, Colwell RR. 2006. Seasonal cholera caused by Vibrio cholerae serogroups O1 and O139 in the coastal aquatic environment of Bangladesh. Appl Environ Microbiol 72:4096–4104. 26. Grim CJ, Choi J, Chun J, Jeon YS, Taviani E, Hasan NA, Haley B, Huq A, Colwell RR. 2010. Occurrence of the vibrio cholerae seventh pandemic VSP-I island and a new variant. Omi A J Integr Biol 14:1–7. 27. Ring J. 1804. Mr. Ring, on Diarrhœa and Cholera Morbus. Med Phys J 12:102– 108. 28. Anderson WS. 1819. An Account of Cholera Morbus Epidemic in India, in 1817 and 1818. Edinburgh Med Surg J 15:354–372. 29. J S. 1856. Cholera and the Water Supply in the South Districts of London in 1854. J public Heal Sanit Rev 2:239–257. 30. Davies BW, Bogard RW, Young TS, Mekalanos JJ. 2012. Coordinated regulation of accessory genetic elements produces cyclic di-nucleotides for V. cholerae virulence. Cell 149:358–370. 31. Millman A, Melamed S, Amitai G, Sorek R. 2020. Diversity and classification of cyclic-oligonucleotide-based anti-phage signalling systems. Nat Microbiol 1–8. 32. Severin GB, Hsueh BY, Elg CA, Dover JA, Rhoades CR, Wessel AJ, Ridenhour BJ, Top EM, Ravi J, Parent KN, Waters CM. 2021. A Broadly Conserved Deoxycytidine Deaminase Protects Bacteria from Phage Infection. bioRxiv 2021.03.31.437871. 33. Cohn M. 1964. Non-inducible mutants of the regulator gene in the “lactose” system of Escherichia coli. J Mol Biol 8:582–592. 34. Wilson CJ, Zhan H, Swint-Kruse L, Matthews KS. 2007. The lactose repressor system: Paradigms for regulation, allosteric behavior and protein folding. Cell Mol Life Sci 64:3–16. 35. Laub MT, Goulian M. 2007. Specificity in two-component signal transduction pathways. Annu Rev Genet 41:121–145. 114 36. Persat A, Inclan YF, Engel JN, Stone HA, Gitai Z. 2015. Type IV pili mechanochemically regulate virulence factors in Pseudomonas aeruginosa. Proc Natl Acad Sci U S A 112:7563–7568. 37. Berg HC. 2017. The flagellar motor adapts, optimizing bacterial behavior. Protein Sci 26:1249–1251. 38. Hengge R, Gründling A, Jenal U, Ryan R, Yildiz F. 2016. Bacterial signal transduction by cyclic di-GMP and other nucleotide second messengers. J Bacteriol 198:15–26. 39. Rickenberg H V. 1974. Cyclic AMP in prokaryotes. Annu Rev Microbiol 28:353– 369. 40. MAKMAN RS, SUTHERLAND EW. 1965. Adenosine 3’,5’-Phosphate in Escherichia Coli. J Biol Chem 240:1309–1314. 41. Petersen S, Young GM. 2002. Essential role for cyclic AMP and its receptor protein in Yersinia enterocolitica virulence. Infect Immun 70:3665–3672. 42. Fong JCN, Yildiz FH. 2008. Interplay between cyclic AMP-cyclic AMP receptor protein and cyclic di-GMP signaling in vibrio cholerae biofilm formation. J Bacteriol 190:6646–6659. 43. D’Ari R, Jaffé A, Bouloc P, Robin A. 1988. Cyclic AMP and cell division in Escherichia coli. J Bacteriol 170:65–70. 44. Haseltine WA, Block R. 1973. Synthesis of guanosine tetra and pentaphosphate requires the presence of a codon specific, uncharged transfer ribonucleic acid in the acceptor site of ribosomes. Proc Natl Acad Sci U S A 70:1564–1568. 45. Cashel M, Gallant J. 1969. Two compounds implicated in the function of RC gene. Nature 221:838–841. 46. Römling U, Galperin MY. 2017. Discovery of the second messenger cyclic di-GMP. Methods Mol Biol 1657:1–8. 47. Ross P, Weinhouse H, Aloni Y, Michaeli D, Weinberger-Ohana P, Mayer R, Braun S, De Vroom E, Van Der Marel GA, Van Boom JH, Benziman M. 1987. Regulation of cellulose synthesis in Acetobacter xylinum by cyclic diguanylic acid. Nature 325:279–281. 48. Romling U, Galperin MY, Gomelsky M. 2013. Cyclic di-GMP: the First 25 Years of 115 a Universal Bacterial Second Messenger. Microbiol Mol Biol Rev 77:1–52. 49. Ryjenkov DA, Tarutina M, Moskvin O V., Gomelsky M. 2005. Cyclic diguanylate is a ubiquitous signaling molecule in bacteria: Insights into biochemistry of the GGDEF protein domain. J Bacteriol 187:1792–1798. 50. Chan C, Paul R, Samoray D, Amiot NC, Giese B, Jenal U, Schirmer T. 2004. Structural basis of activity and allosteric control of diguanylate cyclase. Proc Natl Acad Sci U S A 101:17084–17089. 51. Ausmees N, Mayer R, Weinhouse H, Volman G, Amikam D, Benziman M, Lindberg M. 2001. Genetic data indicate that proteins containing the GGDEF domain possess diguanylate cyclase activity. FEMS Microbiol Lett 204:163–167. 52. Hecht GB, Newton A. 1995. Identification of a novel response regulator required for the swarmer-to- stalked-cell transition in Caulobacter crescentus. J Bacteriol 177:6223–6229. 53. Schmidt AJ, Ryjenkov DA, Gomelsky M. 2005. The ubiquitous protein domain EAL is a cyclic diguanylate-specific phosphodiesterase: Enzymatically active and inactive EAL domains. J Bacteriol 187:4774–4781. 54. Christen M, Christen B, Folcher M, Schauerte A, Jenal U. 2005. Identification and characterization of a cyclic di-GMP-specific phosphodiesterase and its allosteric control by GTP. J Biol Chem 280:30829–30837. 55. Ryan RP, Fouhy Y, Lucey JF, Crossman LC, Spiro S, He Y-W, Zhang L-H, Heeb S, Cámara M, Williams P, Dow JM. 2006. Cell-cell signaling in Xanthomonas campestris involves an HD-GYP domain protein that functions in cyclic di-GMP turnover. Proc Natl Acad Sci U S A2006/04/12. 103:6712–6717. 56. Galperin MY. 2004. Bacterial signal transduction network in a genomic perspective. Environ Microbiol 6:552–67. 57. Simm R, Morr M, Kader A, Nimtz M, Römling U. 2004. GGDEF and EAL domains inversely regulate cyclic di-GMP levels and transition from sessibility to motility. Mol Microbiol 53:1123–1134. 58. Tischler AD, Camilli A. 2004. Cyclic diguanylate (c-di-GMP) regulates Vibrio cholerae biofilm formation. Mol Microbiol 53:857–869. 59. Tuckerman JR, Gonzalez G, Sousa EHS, Wan X, Saito JA, Alam M, Gilles- Gonzalez MA. 2009. An oxygen-sensing diguanylate cyclase and phosphodiesterase couple for c-di-GMP control. Biochemistry 48:9764–9774. 116 60. Sobe RC, Bond WG, Wotanis CK, Zayner JP, Burriss MA, Fernandez N, Bruger EL, Waters CM, Neufeld HS, Karatan E. 2017. Spermine inhibits Vibrio cholerae biofilm formation through the NspS–MbaA polyamine signaling system. J Biol Chem 292:17025–17036. 61. Koestler BJ, Waters CM. 2014. Bile acids and bicarbonate inversely regulate intracellular cyclic di-GMP in vibrio cholerae. Infect Immun 82:3002–3014. 62. Hufnagel DA, DePas WH, Chapmana MR. 2014. The disulfide bonding system suppresses CsgD-Independent cellulose production in Escherichia coli. J Bacteriol 196:3690–3699. 63. Hsieh M, Hinton DM, Waters CM. 2020. Cyclic di-GMP Regulation of Gene Expression 379–394. 64. Hobley L, Fung RKY, Lambert C, Harris MATS, Dabhi JM, King SS, Basford SM, Uchida K, Till R, Ahmad R, Aizawa SI, Gomelsky M, Sockett RE. 2012. Discrete cyclic di-GMP-dependent control of bacterial predation versus axenic growth in Bdellovibrio bacteriovorus. PLoS Pathog 8. 65. Tischler AD, Camilli A. 2005. Cyclic diguanylate regulates Vibrio cholerae virulence gene expression. Infect Immun 73:5873–5882. 66. Tschowri N, Schumacher MA, Schlimpert S, Chinnam NB, Findlay KC, Brennan RG, Buttner MJ. 2014. Tetrameric c-di-GMP mediates effective transcription factor dimerization to control streptomyces development. Cell 158:1136–1147. 67. Fernandez NL, Srivastava D, Ngouajio AL, Waters CM. 2018. Cyclic di-GMP Positively Regulates DNA Repair in Vibrio cholerae. J Bacteriol 200:1–13. 68. Fernandez NL, Hsueh BY, Nhu NTQ, Franklin JL, Dufour YS, Waters CM. 2020. Vibrio cholerae adapts to sessile and motile lifestyles by cyclic di-GMP regulation of cell shape. Proc Natl Acad Sci U S A 117:29046–29054. 69. Chen Z, Schaap P. 2012. Dictyostelium uses the prokaryote messenger c-di-GMP to trigger stalk cell differentiation. Nature 488:680–683. 70. Witte G, Hartung S, Büttner K, Hopfner KP. 2008. Structural Biochemistry of a Bacterial Checkpoint Protein Reveals Diadenylate Cyclase Activity Regulated by DNA Recombination Intermediates. Mol Cell 30:167–178. 71. Stülke J, Krüger L. 2020. Cyclic di-AMP Signaling in Bacteria. Annu Rev Microbiol 74:159–179. 117 72. He J, Yin W, Galperin MY, Chou SH. 2020. Cyclic di-AMP, a second messenger of primary importance: Tertiary structures and binding mechanisms. Nucleic Acids Res 48:2807–2829. 73. Corrigan RM, Gründling A. 2013. Cyclic di-AMP: Another second messenger enters the fray. Nat Rev Microbiol 11:513–524. 74. Woodward JJ, Iavarone AT, Portnoy DA. 2010. c-di-AMP secreted by intracellular Listeria monocytogenes activates a host type I interferon response. Science 328:1703–5. 75. Luo Y, Helmann JD. 2012. Analysis of the role of Bacillus subtilis σ(M) in β- lactam resistance reveals an essential role for c-di-AMP in peptidoglycan homeostasis. Mol Microbiol 83:623–39. 76. Gundlach J, Herzberg C, Kaever V, Gunka K, Hoffmann T, Weiß M, Gibhardt J, Thürmer A, Hertel D, Daniel R, Bremer E, Commichau FM, Stülke J. 2017. Control of potassium homeostasis is an essential function of the second messenger cyclic di-AMP in Bacillus subtilis. Sci Signal 10:1–10. 77. Pham HT, Nhiep NTH, Vu TNM, Huynh TAN, Zhu Y, Huynh ALD, Chakrabortti A, Marcellin E, Lo R, Howard CB, Bansal N, Woodward JJ, Liang ZX, Turner MS. 2018. Enhanced uptake of potassium or glycine betaine or export of cyclic-di-AMP restores osmoresistance in a high cyclic-di-AMP Lactococcus lactis mutant. PLoS Genet 14:1–23. 78. Zarrella TM, Metzger DW, Bai G. 2018. Stress suppressor screening leads to detection of regulation of cyclic di-AMP homeostasis by a Trk family effector protein in Streptococcus pneumoniae. J Bacteriol 200:1–15. 79. Schuster CF, Bellows LE, Tosi T, Campeotto I, Corrigan RM, Freemont P, Gründling A. 2016. The second messenger c-di-AMP inhibits the osmolyte uptake system OpuC in Staphylococcus aureus. Sci Signal 9:1–31. 80. Zeden MS, Schuster CF, Bowman L, Zhong Q, Williams HD, Gründling A. 2018. Cyclic di-adenosine monophosphate (c-di-AMP) is required for osmotic regulation in Staphylococcus aureus but dispensable for viability in anaerobic conditions. J Biol Chem 293:3180–3200. 81. Hu D, Liu B, Feng L, Ding P, Guo X, Wang M, Cao B, Reeves PR, Wang L. 2016. Origins of the current seventh cholera pandemic. Proc Natl Acad Sci U S A 113:E7730–E7739. 82. Severin GB, Ramliden MS, Hawver LA, Wang K, Pell ME, Kieninger AK, 118 Khataokar A, O’Hara BJ, Behrmann L V., Neiditch MB, Benning C, Waters CM, Ng WL. 2018. Direct activation of a phospholipase by cyclic GMP-AMP in El Tor Vibrio cholerae. Proc Natl Acad Sci U S A 115:E6048–E6055. 83. Li F, Cimdins A, Rohde M, Jänsch L, Kaever V, Nimtz M, Römling U. 2019. DncV synthesizes cyclic GMP-AMP and regulates biofilm formation and motility in Escherichia coli ECOR31. MBio 10:1–21. 84. Kranzusch PJ, Wilson SC, Lee ASY, Berger JM, Doudna JA, Vance RE. 2015. Ancient Origin of cGAS-STING Reveals Mechanism of Universal 2’,3’ cGAMP Signaling. Mol Cell 59:891–903. 85. Burroughs AM, Zhang D, Schäffer DE, Iyer LM, Aravind L. 2015. Comparative genomic analyses reveal a vast, novel network of nucleotide-centric systems in biological conflicts, immunity and signaling. Nucleic Acids Res 43:10633–10654. 86. Whiteley AT, Eaglesham JB, de Oliveira Mann CC, Morehouse BR, Lowey B, Nieminen EA, Danilchanka O, King DS, Lee ASY, Mekalanos JJ, Kranzusch PJ. 2019. Bacterial cGAS-like enzymes synthesize diverse nucleotide signals. Nature 567:194–199. 87. Lau R, Ye Q, Patel L, Berg K, Mathews I, Watrous J, Whiteley A, Lowey B, Mekalanos J, Kranzusch P, Jain M, Corbett K. 2019. Structure and mechanism of a cyclic trinucleotide-activated bacterial endonuclease mediating bacteriophage immunity. Mol Cell 694703. 88. Lowey B, Whiteley AT, Keszei AFA, Morehouse BR, Mathews IT, Antine SP, Cabrera VJ, Kashin D, Niemann P, Jain M, Schwede F, Mekalanos JJ, Shao S, Lee ASY, Kranzusch PJ. 2020. CBASS Immunity Uses CARF-Related Effectors to Sense 3′–5′- and 2′–5′-Linked Cyclic Oligonucleotide Signals and Protect Bacteria from Phage Infection. Cell 182:38-49.e17. 89. Cohen D, Melamed S, Millman A, Shulman G, Oppenheimer-Shaanan Y, Kacen A, Doron S, Amitai G, Sorek R. 2019. Cyclic GMP–AMP signalling protects bacteria against viral infection. Nature 574:691–695. 90. Gao J, Tao J, Liang W, Zhao M, Du X, Cui S, Duan H, Kan B, Su X, Jiang Z. 2015. Identification and characterization of phosphodiesterases that specifically degrade 3’3’-cyclic GMP-AMP. Cell Res 25:539–550. 91. Wright TA, Jiang L, Park JJ, Anderson WA, Chen G, Hallberg ZF, Nan B, Hammond MC. 2020. Second messengers and divergent HD-GYP phosphodiesterases regulate 3′,3′-cGAMP signaling. Mol Microbiol 113:222–236. 119 92. Hallberg ZF, Wang XC, Wright TA, Nan B, Ad O, Yeo J, Hammond MC. 2016. Hybrid promiscuous (Hypr) GGDEF enzymes produce cyclic AMP-GMP (3’, 3’- cGAMP). Proc Natl Acad Sci U S A 113:1790–1795. 93. Kellenberger CA, Wilson SC, Hickey SF, Gonzalez TL, Su Y, Hallberg ZF, Brewer TF, Iavarone AT, Carlson HK, Hsieh YF, Hammond MC. 2015. GEMM-I riboswitches from Geobacter sense the bacterial second messenger cyclic AMP- GMP. Proc Natl Acad Sci U S A 112:5383–5388. 94. Hallberg ZF, Chan CH, Wright TA, Kranzusch PJ, Doxzen KW, Park JJ, Bond DR, Hammond MC. 2019. Structure and mechanism of a hypr GGDEF enzyme that activates cGAMP signaling to control extracellular metal respiration. Elife 8:1–36. 95. Zhu D, Wang L, Shang G, Liu X, Zhu J, Lu D, Wang L, Kan B, Zhang J ren, Xiang Y. 2014. Structural Biochemistry of a Vibrio cholerae Dinucleotide Cyclase Reveals Cyclase Activity Regulation by Folates. Mol Cell 55:931–937. 96. Sun L, Wu J, Du F, Chen X, Chen ZJ. 2013. Cyclic GMP-AMP synthase is a cytosolic DNA sensor that activates the type I interferon pathway. Science 339:786–91. 97. Li X, Shu C, Yi G, Chaton CT, Shelton CL, Diao J, Zuo X, Kao CC, Herr AB, Li P. 2013. Cyclic GMP-AMP Synthase Is Activated by Double-Stranded DNA-Induced Oligomerization. Immunity 39:1019–1031. 98. Liu H, Moura-Alves P, Pei G, Mollenkopf H, Hurwitz R, Wu X, Wang F, Liu S, Ma M, Fei Y, Zhu C, Koehler A, Oberbeck-Mueller D, Hahnke K, Klemm M, Guhlich- Bornhof U, Ge B, Tuukkanen A, Kolbe M, Dorhoi A, Kaufmann SH. 2019. cGAS facilitates sensing of extracellular cyclic dinucleotides to activate innate immunity . EMBO Rep 20. 99. McFarland AP, Luo S, Ahmed-Qadri F, Zuck M, Thayer EF, Goo YA, Hybiske K, Tong L, Woodward JJ. 2017. Sensing of Bacterial Cyclic Dinucleotides by the Oxidoreductase RECON Promotes NF-κB Activation and Shapes a Proinflammatory Antibacterial State. Immunity 46:433–445. 100. Kranzusch PJ. 2019. cGAS and CD-NTase enzymes: structure, mechanism, and evolution. Curr Opin Struct Biol. Elsevier Ltd. 101. Kato K, Omura H, Ishitani R, Nureki O. 2017. Cyclic GMP-AMP as an endogenous second messenger in innate immune signaling by cytosolic DNA. Annu Rev Biochem 86:541–566. 102. Faruque SM, Mekalanos JJ. 2012. Phage-bacterial interactions in the evolution of 120 toxigenic Vibrio cholerae. Virulence 3:556–565. 103. Wang Z, Lazinski DW, Camilli A. 2017. Immunity provided by an outer membrane vesicle cholera vaccine is due to O-antigenspecific antibodies inhibiting bacterial motility. Infect Immun 85:1–9. 104. Reyes-Robles T, Dillard RS, Cairns LS, Silva-Valenzuela CA, Housman M, Ali A, Wright ER, Camilli A. 2018. Vibrio cholerae outer membrane vesicles inhibit bacteriophage infection. J Bacteriol 200. 105. Hays SG, Seed KD. 2019. Dominant Vibrio cholerae phage exhibits lysis inhibition sensitive to disruption by a defensive phage satellite. bioRxiv 1–24. 106. Koonin E V., Makarova KS, Zhang F. 2017. Diversity, classification and evolution of CRISPR-Cas systems. Curr Opin Microbiol 37:67–78. 107. Ye Q, Lau RK, Mathews IT, Birkholz EA, Watrous JD, Azimi CS, Pogliano J, Jain M, Corbett KD. 2020. HORMA Domain Proteins and a Trip13-like ATPase Regulate Bacterial cGAS-like Enzymes to Mediate Bacteriophage Immunity. Mol Cell 77:709-722.e7. 108. Hu MM, Yang Q, Xie XQ, Liao CY, Lin H, Liu TT, Yin L, Shu HB. 2016. Sumoylation Promotes the Stability of the DNA Sensor cGAS and the Adaptor STING to Regulate the Kinetics of Response to DNA Virus. Immunity 45:555–569. 109. Vierstra RD. 2012. The expanding universe of ubiquitin and ubiquitin-like modifiers. Plant Physiol. American Society of Plant Biologists. 110. Song L, Luo ZQ. 2019. Post-translational regulation of ubiquitin signaling. J Cell Biol 218:1776–1786. 111. Lee I, Schindelin H. 2008. Structural Insights into E1-Catalyzed Ubiquitin Activation and Transfer to Conjugating Enzymes. Cell 134:268–278. 112. Noda NN, Satoo K, Fujioka Y, Kumeta H, Ogura K, Nakatogawa H, Ohsumi Y, Inagaki F. 2011. Structural basis of Atg8 activation by a homodimeric E1, Atg7. Mol Cell 44:462–475. 113. Stewart MD, Ritterhoff T, Klevit RE, Brzovic PS. 2016. E2 enzymes: More than just middle men. Cell Res 26:423–440. 114. Hennell James R, Caceres EF, Escasinas A, Alhasan H, Howard JA, Deery MJ, Ettema TJG, Robinson NP. 2017. Functional reconstruction of a eukaryotic-like 121 E1/E2/(RING) E3 ubiquitylation cascade from an uncultured archaeon. Nat Commun 8. 115. Mevissen TET, Komander D. 2017. Mechanisms of deubiquitinase specificity and regulation. Annu Rev Biochem 86:159–192. 116. Lake MW, Wuebbens MM, Rajagopalan K V., Schindelin H. 2001. Mechanism of ubiquitin activation revealed by the structure of a bacterial MoeB-MoaD complex. Nature 414:325–329. 117. Miranda H V., Nembhard N, Su D, Hepowit N, Krause DJ, Pritz JR, Phillips C, Söll D, Maupin-Furlow JA. 2011. E1- and ubiquitin-like proteins provide a direct link between protein conjugation and sulfur transfer in archaea. Proc Natl Acad Sci U S A 108:4417–4422. 118. Lehmann G, Udasin RG, Livneh I, Ciechanover A. 2017. Identification of UBact, a ubiquitin-like protein, along with other homologous components of a conjugation system and the proteasome in different Gram-negative bacteria. Biochem Biophys Res Commun 483:946–950. 119. Schmitz J, Wuebbens MM, Rajagopalan K V., Leimkühler S. 2007. Role of the C- terminal Gly-Gly motif of Escherichia coli MoaD, a molybdenum cofactor biosynthesis protein with a ubiquitin fold. Biochemistry 46:909–916. 120. Hershko A, Heller H, Elias S, Ciechanover A. 1983. Components of ubiquitin- protein ligase system. Resolution, affinity purification, and role in protein breakdown. J Biol Chem 258:8206–8214. 121. Burroughs AM, Jaffee M, Iyer LM, Aravind L. 2008. Anatomy of the E2 ligase fold: Implications for enzymology and evolution of ubiquitin/Ub-like protein conjugation. J Struct Biol 162:205–218. 122. Huang DT, Hunt HW, Zhuang M, Ohi MD, Holton JM, Schulman BA. 2007. Basis for a ubiquitin-like protein thioester switch toggling E1-E2 affinity. Nature 445:394–398. 123. Müller AU, Weber-Ban E. 2019. The bacterial proteasome at the core of diverse degradation pathways. Front Mol Biosci. Frontiers Media S.A. 124. Striebel F, Imkamp F, Özcelik D, Weber-Ban E. 2014. Pupylation as a signal for proteasomal degradation in bacteria. Biochim Biophys Acta - Mol Cell Res 1843:103–113. 125. Delley CL, Müller AU, Ziemski M, Weber-Ban E. 2017. Prokaryotic Ubiquitin-Like 122 Protein and Its Ligase/Deligase Enyzmes. J Mol Biol. Academic Press. 126. Iyer LM, Burroughs AM, Aravind L. 2008. Unraveling the biochemistry and provenance of pupylation: A prokaryotic analog of ubiquitination. Biol Direct 3:1– 7. 127. Striebel F, Imkamp F, Sutter M, Steiner M, Mamedov A, Weber-Ban E. 2009. Bacterial ubiquitin-like modifier Pup is deamidated and conjugated to substrates by distinct but homologous enzymes. Nat Struct Mol Biol 16:647–651. 128. Desterro JMP, Rodriguez MS, Kemp GD, Ronald T H. 1999. Identification of the enzyme required for activation of the small ubiquitin-like protein SUMO-1. J Biol Chem 274:10618–10624. 129. Dahl JU, Urban A, Bolte A, Sriyabhaya P, Donahue JL, Nimtz M, Larson TJ, Leimkühler S. 2011. The identification of a novel protein involved in molybdenum cofactor biosynthesis in Escherichia coli. J Biol Chem 286:35801–35812. 130. Xi J, Ge Y, Kinsland C, McLafferty FW, Begley TP. 2001. Biosynthesis of the thiazole moiety of thiamin in Escherichia coli: Identification of an acyldisulfide- linked protein - Protein conjugate that is functionally analogous to the ubiquitin/E1 complex. Proc Natl Acad Sci U S A 98:8513–8518. 131. Xu X, Wang T, Niu Y, Liang K, Yang Y. 2019. The ubiquitin-like modification by ThiS and ThiF in Escherichia coli. Int J Biol Macromol 141:351–357. 132. Leimkühler S, Wuebbens MM, Rajagopalan K V. 2001. Characterization of Escherichia coli MoeB and Its Involvement in the Activation of Molybdopterin Synthase for the Biosynthesis of the Molybdenum Cofactor. J Biol Chem 276:34695–34701. 133. Lehmann C, Begley TP, Ealick SE. 2006. Structure of the Escherichia coli ThiS- ThiF complex, a key component of the sulfur transfer system in thiamin biosynthesis. Biochemistry 45:11–19. 134. Jones CJ, Utada A, Davis KR, Thongsomboon W, Zamorano Sanchez D, Banakar V, Cegelski L, Wong GCL, Yildiz FH. 2015. C-di-GMP Regulates Motile to Sessile Transition by Modulating MshA Pili Biogenesis and Near-Surface Motility Behavior in Vibrio cholerae. PLoS Pathog 11:1–27. 135. Valentini M, Filloux A. 2016. Biofilms and Cyclic di-GMP (c-di-GMP) signaling: Lessons from Pseudomonas aeruginosa and other bacteria. J Biol Chem 291:12547–12555. 123 136. Lau RK, Ye Q, Birkholz EA, Berg KR, Patel L, Mathews IT, Watrous JD, Ego K, Whiteley AT, Lowey B, Mekalanos JJ, Kranzusch PJ, Jain M, Pogliano J, Corbett KD. 2020. Structure and Mechanism of a Cyclic Trinucleotide-Activated Bacterial Endonuclease Mediating Bacteriophage Immunity. Mol Cell 77:723-733.e6. 137. Yoon Sh, Waters CM. 2021. The ever-expanding world of bacterial cyclic oligonucleotide second messengers. Curr Opin Microbiol 60:96–103. 138. Northrup RS, Doyle MA, Feeley JC. 1972. In vitro susceptibility of El Tor and classical Vibrio cholerae strains to trimethoprim and sulfamethoxazole. Antimicrob Agents Chemother 1:310–314. 139. Bushby SRM. 1973. Trimethoprim-sulfamethoxazole: In vitro microbiological aspects. J Infect Dis 128:S442–S462. 140. Bach MC, Finland M, Gold O, Wilcox C. 1973. Susceptibility of recently isolated pathogenic bacteria to trimethoprim and sulfamethoxazole separately and combined. J Infect Dis 128:S508–S533. 141. Chen Z, Pickart CM. 1990. A 25-kilodalton ubiquitin carrier protein (E2) catalyzes multi-ubiquitin chain synthesis via lysine 48 of ubiquitin. J Biol Chem 265:21835–21842. 142. Yang Y, He Y, Wang X, Liang Z, He G, Zhang P, Zhu H, Xu N, Liang S. 2017. Protein SUMOylation modification and its associations with disease. Open Biol. Royal Society Publishing. 143. Cuijpers SAG, Willemstein E, Vertegaal ACO. 2017. Converging Small Ubiquitin- like Modifier (SUMO) and ubiquitin signaling: Improved methodology identifies Co-Modified target proteins. Mol Cell Proteomics 16:2281–2295. 144. Cao S, Engilberge S, Girard E, Gabel F, Franzetti B, Maupin-Furlow JA. 2017. Structural Insight into Ubiquitin-Like Protein Recognition and Oligomeric States of JAMM/MPN+ Proteases. Structure 25:823-833.e6. 145. Tahmasebi S, Ghorbani M, Savage P, Gocevski G, Yang XJ. 2014. The SUMO conjugating enzyme Ubc9 is required for inducing and maintaining stem cell pluripotency. Stem Cells 32:1012–1020. 146. Cai J, Pan C, Zhao Y, Xu H, Tian B, Wang L, Hua Y. 2022. DRJAMM Is Involved in the Oxidative Resistance in Deinococcus radiodurans. Front Microbiol 12:1–10. 147. Edelheit O, Hanukoglu A, Hanukoglu I. 2009. Simple and efficient site-directed mutagenesis using two single-primer reactions in parallel to generate mutants for 124 protein structure-function studies. BMC Biotechnol 9:1–8. 148. Dalia AB, McDonough EK, Camilli A. 2014. Multiplex genome editing by natural transformation. Proc Natl Acad Sci U S A 111:8937–8942. 149. Dalia TN, Yoon SH, Galli E, Barre FX, Waters CM, Dalia AB. 2017. Enhancing multiplex genome editing by natural transformation (MuGENT) via inactivation of ssDNA exonucleases. Nucleic Acids Res 45:7527–7537. 150. Dalia TN, Hayes CA, Stolyar S, Marx CJ, McKinlay JB, Dalia AB. 2017. Multiplex Genome Editing by Natural Transformation (MuGENT) for Synthetic Biology in Vibrio natriegens. ACS Synth Biol 6:1650–1655. 151. Massie JP, Reynolds EL, Koestler BJ, Cong JP, Agostoni M, Waters CM. 2012. Quantification of high-specificity cyclic diguanylate signaling. Proc Natl Acad Sci U S A 109:12746–12751. 152. Lopp A, Reintamm T, Kuusksalu A, Tammiste I, Pihlak A, Kelve M. 2010. Natural occurrence of 2′,5′-linked heteronucleotides in marine sponges. Mar Drugs 8:235–254. 153. Eskildsen S, Justesen J, Schierup MH, Hartmann R. 2003. Characterization of the 2′-5′-oligoadenylate synthetase ubiquitin-like family. Nucleic Acids Res 31:3166– 3173. 154. Lo R, Stanton-Cook MJ, Beatson SA, Turner MS, Bansal N. 2015. Draft Genome Sequence of Pseudomonas fluorescens SRM1, an Isolate from Spoiled Raw Milk. Genome Announc 3:10–11. 155. Hazen TH, Sahl JW, Redman JC, Morris CR, Daugherty SC, Chibucos MC, Sengamalay NA, Fraser-Liggett CM, Steinsland H, Whittam TS, Whittam B, Manning SD, Rasko DA. 2012. Draft genome sequences of the diarrheagenic Escherichia coli collection. J Bacteriol 194:3026–3027. 156. Watve SS, Chande AT, Rishishwar L, Mariño-Ramírez L, Jordan IK, Hammer BK. 2016. Whole-genome sequences of 26 Vibrio cholerae isolates. Genome Announc 4:4–6. 157. Høyland-Kroghsbo NM. 2019. Cyclic Nucleotide Signaling: A Second Messenger of Death. Cell Host Microbe 26:567–568. 158. Baker-Austin C, Oliver JD, Alam M, Ali A, Waldor MK, Qadri F, Martinez-Urtaza J. 2018. Vibrio spp. infections. Nat Rev Dis Prim 4. 125 159. Thorpe KE, Joski P, Johnston KJ. 2018. Antibiotic-resistant infection treatment costs have doubled since 2002, now exceeding $2 billion annually. Health Aff 37:662–669. 160. Lin J, Nishino K, Roberts MC, Tolmasky M, Aminov RI, Zhang L. 2015. Mechanisms of antibiotic resistance. Front Microbiol 6:1–24. 161. Pallasch TJ. 2003. Antibiotic resistance. Dent Clin North Am 47:623–639. 162. Torres-Barceló C. 2018. The disparate effects of bacteriophages on antibiotic- resistant bacteria. Emerg Microbes Infect 7. 163. Yen M, Cairns LS, Camilli A. 2017. A cocktail of three virulent bacteriophages prevents Vibrio cholerae infection in animal models. Nat Commun 8:1–7. 164. McDonald ND, Regmi A, Morreale DP, Borowski JD, Fidelma Boyd E. 2019. CRISPR-Cas systems are present predominantly on mobile genetic elements in Vibrio species. BMC Genomics 20:1–23. 165. Doron S, Melamed S, Ofir G, Leavitt A, Lopatina A, Keren M, Amitai G, Sorek R. 2018. Systematic discovery of antiphage defense systems in the microbial pangenome. Science (80- ) 359:eaar4120. 166. Helene Thelin K, Taylor RK. 1996. Toxin-coregulated pilus, but not mannose-sensitive hemagglutinin, is required for colonization by Vibrio cholerae O1 El Tor biotype and O139 strains. Infect Immun 64:2853–2856. 167. Yildiz FH, Schoolnik GK. 1998. Role of rpoS in stress survival and virulence of Vibrio cholerae. J Bacteriol 180:773–784. 168. Shao Y, Bassler BL. 2012. Quorum-sensing non-coding small RNAs use unique pairing regions to differentially control mRNA targets. Mol Microbiol 83:599–611. 126 APPENDIX Table 1. Chapter 2 Strain list Name Origin? Genotype V. cholerae El Tor C6706 Helen et WT Peru clinical strain, StrepR al.(166) V. cholerae El Tor C6706 Severin et al. ∆VSP-I∆VSP-II CR03 (82) V. cholerae El Tor C6706 SHY ∆cbass, TrimR ∆cbass V. cholerae El Tor C6706 Severin et al. ∆dncV ∆dncV (82) V. cholerae El Tor C6706 This study. ∆vc0180∆vc0181, TrimR ∆vc0180∆vc0181 SHY V. cholerae El Tor C6706 This study. ∆bumo, TrimR ∆bumo SHY V. cholerae El Tor C6706 This study. bumoM1*, TrimR bumoM1* SHY V. cholerae El Tor A1552 Yildiz et al WT GentR (167) V. cholerae El Tor A1552 This study. WT with pTXB1 EV, AmpR pTXB1 SHY V. cholerae El Tor A1552 This study. pTXB1 Tac promoter, EV, AmpR, TrimR ∆dncV pTXB1 SHY V. cholerae El Tor A1552 This study. pTXB1 Tac promoter, C-terminal pTXB1 Bumo SHY InteinCBD fused Bumo, AmpR V. cholerae El Tor A1552 This study. ∆dncV pTXB1 Tac promoter, C- ∆dncV pTXB1 Bumo SHY terminal InteinCBD-Bumo, AmpR,TrimR V. cholerae El Tor A1552 This study. pET28B T7 promoter EV, KanR pET28b SHY V. cholerae El Tor A1552 This study. pET28B T7 promoter EV, KanR, TrimR ∆bumo pET28b SHY V. cholerae El Tor A1552 This study. pET28B T7 promoter EV, KanR, TrimR bumoM1* pET28b SHY V. cholerae El Tor A1552 This study. pET28B T7 promoter EV, KanR, TrimR ∆vc0181 pET28B SHY V. cholerae El Tor A1552 This study. pET28B T7 promoter EV, KanR, TrimR ∆bumo∆vc0181 pET28b SHY V. cholerae El Tor A1552 This study. pET28b T7 promoter C-terminal fused pET28b Bumo SHY to Bumo-6XHIS, KanR V. cholerae El Tor A1552 This study. pET28b T7 promoter C-terminal fused bumoM1* pET28b Bumo SHY to Bumo-6XHIS, KanR, TrimR V. cholerae El Tor A1552 This study. pET28b T7 promoter C-terminal fused ∆vc0181 pET28b Bumo SHY to Bumo-6XHIS, KanR, TrimR 127 Table 1. (cont’d) V. cholerae El Tor A1552 This study. pET28b T7 promoter C-terminal fused bumoM1*∆vc0181 pET28b SHY to Bumo-6XHIS, KanR, SpecR Bumo V. cholerae El Tor A1552 This study. pET28B T7 promoter 6XHIS fused to pET28b DncV SHY C-terminal of dncV, KanR V. cholerae El Tor A1552 This study. pET28B T7 promoter 6XHIS fused to ∆bumo pET28b DncV SHY C-terminal of dncV, KanR V. cholerae El Tor A1552 This study. pET28B T7 promoter 6XHIS fused to ∆bumo∆vc0181 pET28b SHY C-terminal of dncV, KanR DncV V. cholerae El Tor A1552 This study. pET28B T7 promoter 6XHIS fused to pET28b DncV SHY C-terminal of dncV, KanR V. cholerae El Tor A1552 This study. pET28B T7 promoter 6XHIS fused to bumoM1* pET28b DncV SHY C-terminal of dncV, KanR V. cholerae El Tor A1552 This study. pET28B T7 promoter 6XHIS fused to ∆vc0180 pET28b DncV SHY C-terminal of dncV, KanR, TrimR V. cholerae El Tor A1552 This study. pET28B T7 promoter 6XHIS fused to ∆vc0181 pET28b DncV SHY C-terminal of dncV, KanR, TrimR V. cholerae El Tor A1552 This study. pET28B T7 promoter 6XHIS fused to ∆vc0180∆vc0181 pET28b SHY C-terminal of dncV, KanR, SpecR DncV V. cholerae El Tor A1552 This study. pET28B T7 promoter 6XHIS fused to bumoM1*∆vc0181 pET28b SHY C-terminal of dncV, KanR, SpecR DncV V. cholerae El Tor A1552 This study. pET28b T7 promoter C-terminal pET28B VC0180-6XHIS Geoffrey 6XHIS fused VC0180, KanR Severin V. cholerae El Tor A1552 This study. pET28b T7 promoter C-terminal ∆bumo pET28B VC0180- SHY 6XHIS fused VC0180, KanR, TrimR 6XHIS V. cholerae El Tor A1552 This study. Nonsense mutation in first codon of bumoM1* pET28B VC0180- SHY bumo with pET28b T7 promoter C- 6XHIS terminal 6XHIS fused VC0180, KanR, TrimR V. cholerae El Tor A1552 This study. pET28b T7 promoter C-terminal ∆capV pET28B VC0180- SHY 6XHIS fused VC0180, KanR, TrimR 6XHIS V. cholerae El Tor A1552 This study. pET28b T7 promoter C-terminal ∆dncV pET28B VC0180- SHY 6XHIS fused VC0180, KanR, TrimR 6XHIS V. cholerae El Tor A1552 This study. pET28b T7 promoter C-terminal ∆vc0180 pET28B VC0180- SHY 6XHIS fused VC0180, KanR, TrimR 6XHIS 128 Table 1. (cont’d) V. cholerae El Tor A1552 This study. Nonsense mutation in first codon of vc0180M1STOP pET28B SHY vc0180 with pET28b T7 promoter C- VC0180-6XHIS terminal 6XHIS fused VC0180, KanR, TrimR V. cholerae El Tor A1552 This study. pET28b T7 promoter C-terminal ∆vc0181 pET28B VC0180- SHY 6XHIS fused VC0180, KanR, TrimR 6XHIS V. cholerae El Tor A1552 This study. pET28b T7 promoter C-terminal bumoM1*∆vc0181 pET28B SHY 6XHIS fused VC0180, KanR, SpecR VC0180-6XHIS V. cholerae El Tor A1552 This study. pET28b T7 promoter C-terminal pET28B VC0181-6XHIS SHY 6XHIS fused VC0181, KanR V. cholerae El Tor A1552 This study. Nonsense mutation in first codon of bumoM1* pET28B VC0181- SHY bumo with pET28b T7 promoter C- 6XHIS terminal 6XHIS fused VC0181, KanR, TrimR V. cholerae El Tor A1552 This study. pET28b T7 promoter C-terminal ∆capV pET28B VC0181- SHY 6XHIS fused VC0181, KanR, TrimR 6XHIS V. cholerae El Tor A1552 This study. pET28b T7 promoter C-terminal ∆dncV pET28B VC0181- SHY 6XHIS fused VC0181, KanR, TrimR 6XHIS V. cholerae El Tor A1552 This study. pET28b T7 promoter C-terminal ∆vc0180 pET28B VC0181- SHY 6XHIS fused VC0181, KanR, TrimR 6XHIS V. cholerae El Tor A1552 This study. pET28b T7 promoter C-terminal ∆vc0181 pET28B VC0181- SHY 6XHIS fused VC0181, KanR, TrimR 6XHIS V. cholerae El Tor A1552 This study. pET28b T7 promoter C-terminal bumoM1*∆vc0181 pET28B SHY 6XHIS fused VC0181, KanR, SpecR VC0181-6XHIS V. cholerae El Tor A1552 This study. Chromosomal C-terminal fusion of 3X- dncV-3XFLAG tag SHY FLAG tag onto dncV, SpecR V. cholerae El Tor A1552 This study. Chromosomal C-terminal fusion of 3X- bumoM1* dncV-3XFLAG SHY FLAG tag onto dncV, SpecR tag V. cholerae El Tor A1552 This study. Chromosomal C-terminal fusion of 3X- ∆vc0181 dncV-3XFLAG tag SHY FLAG tag onto dncV, SpecR V. cholerae El Tor A1552 This study. Chromosomal C-terminal fusion of 3X- bumoM1*∆vc0181 dncV- SHY FLAG tag onto dncV, SpecR 3XFLAG tag V. cholerae El Tor A1552 This study. Substitution of E39A of vc0181 on the vc0181 E39A pET28b SHY chromosome with pET28b EV, , SpecR 129 Table 1. (cont’d) V. cholerae El Tor A1552 This study. Double mutants of bumoM1* and bumoM1* vc0181 E39A SHY substitution of E39A of vc0181 on the pET28b chromosome with pET28b T7 promoter EV, KanR SpecR V. cholerae El Tor A1552 This study. Substitution of E39A of vc0181 on the vc0181 E39A pET28b DncV SHY chromosome with pET28b T7 promoter 6XHIS C-terminal tagged dncV, KanR SpecR V. cholerae El Tor A1552 This study. Double mutants of bumoM1* and bumoM1*vc0181 E39A SHY substitution of E39A of vc0181 on the pET28b chromosome with pET28b T7 promoter 6XHIS C-terminal tagged dncV, KanR SpecR V. cholerae El Tor A1552 This study. pTYB21 T7 promoter N-terminal pTYB21 DncV SHY Intein-CBD fused DncV, AmpR V. cholerae El Tor A1552 This study. pTYB21 T7 promoter EV, AmpR pTYB21 SHY V. cholerae El Tor A1552 This study. pTYB21 T7 promoter EV, AmpR, TrimR ∆bumo pTYB21 SHY V. cholerae El Tor A1552 This study. pTYB21 T7 promoter EV, AmpR ∆bumo∆vc0181 pTYB21 SHY V. cholerae El Tor A1552 This study. pTYb21 T7 promoter N-terminal ∆bumo pTYB21 DncV SHY Intein-CBD fused DncV, TrimR V. cholerae El Tor A1552 This study. pTYB21 T7 promoter N-terminal ∆bumo∆vc0181 pTYB21 SHY Intein-CBD fused DncV AmpR, SpecR DncV E. coli DH10b pLAFR Shao et al pLAFR, , TetR (168) E. coli DH10b pCCD7 Severin et al V. cholerae El Tor VSP-I cloned into (82) pLAFR, TetR E. coli DH10b pCCD7 Severin et al Tranposome cosmid library , TetR cosmid library (82) E. coli BL21 pTXB1 This study. pTXB1 T7 promoter EV, AmpR SHY E. coli BL21 pTXB1 Bumo This study. pTXB1 T7 promoter Bumo, AmpR SHY E. coli BL21 pTYB21 VC0180 This study. pTYB21 T7 promoter with N-terminal SHY inteinCBD fused VC0180, AmpR 130 Table 2. Chapter 3 Strain list Name Origin? Genotype S. flexineri pEVS141 John Dover pEVS143 EV, KanR (Parent lab) S. flexineri pBMW1 John Dover E. coli EDEC13E CBASS cloned into (Parent lab) pEVS143, KanR E. coli DH10b pEVS141 Severin et al pEVS143 EV, KanR (82) E. coli DH10b pBMW4 This study. E. coli EDEC13E CBASS cloned into Bella Waters. pEVS143, KanR E. coli DH10b pEVS143 V. This study. V. cholerae 2631-78 CBASS cloned into cholerae 2631-78 CBASS SHY pEVS143, KanR E. coli DH10b pSHY34 This study. P. fluorescens CBASS cloned into SHY pEVS143, KanR E. coli DH10b pEVS143 P. This study. P. fluorescens CBASS cloned into fluorescens CD-NTase SHY pEVS143 with active site mutations D79A, D81A D79A and D81A in CD-NTase, KanR E. coli DH10b pEVS143 P. This study. P. fluorescens CBASS cloned into fluorescens hnh-saved SHY pEVS143 with active site mutations in H56A, H91A HNH motif (H56A and H91A) in hnh- saved, KanR E. coli DH10b pEVS143 P. This study. P. fluorescens CBASS cloned into fluorescens CBASS Double SHY pEVS143 with active site mutations mutant D79A and D81A in CD-NTase HNH motif (H56A and H91A) in hnh-saved, KanR E. coli BL21 pTXB1 Pf CD- This study. pTXB1 with C-terminal inteinCBD NTase SHY fused to CD-NTase of P. fluorescens SRM1, AmpR E. coli BL21 pTXB1 Ec CD- This study. pTXB1 with C-terminal inteinCBD NTase SHY fused to CD-NTase of E. coli EDEC13E, AmpR E. coli BL21 pTXB1 Ec hnh- This study. pTXB1 with C-terminal inteinCBD saved SHY fused to hnh-saved of E. coli EDEC13E, AmpR 131