- - - - - - - - - - - - - - - - - - - - - - - - - - ................................ ................................ ................................ ......................... ................................ ................................ ................................ ......................... ................................ ................................ ................................ ........ ................................ ................................ ................................ .................... ................................ ................................ ................................ ................................ .. ..................... - ................................ ................................ ................ ................................ ................................ ............................ ................................ ................................ . ................................ ............ ................................ ................................ .... - ................................ ...... ................................ ................................ ......................... - ................................ ............. ................................ ................................ .. - ................................ ................ - ................................ ................................ ................................ ................................ ................................ ................................ ................................ ...................... - ................................ ................................ ................................ ................ ................................ ................................ ................................ ................................ ................................ ................................ ................................ ............................... ................................ ................................ ................................ ......................... ................................ ................................ ................................ ....... ................................ .................. ................................ ................................ ................................ . ................................ ................................ ...... - ................................ ................................ ... ................................ ................................ ................................ . ................................ ................................ ................. ................................ ................................ ................................ ................................ . - ................................ ................................ ................................ .............................. ....................... - ................................ ...... ................................ . - ................................ ................................ ................................ ...................... ................................ ................................ ............................ .............................. ................................ ................................ ................................ ........................... ................................ ................................ ................................ ..... ................................ ................................ .............................. ................................ ................................ ............................. ................................ ................................ ................................ ............ ................................ ................................ .... ................................ ................................ .. ................................ ................................ ................................ ................ ................................ ............... - ................................ ..................... ................................ ................................ ................................ .......................... ................................ ................................ ............................... ................................ ................................ ............................ ................................ ........... ................................ . - ................................ ................................ ....... - ................................ ..................... ................................ ................................ ................................ .................... - ................................ ............. - ................................ ............. - ................................ ................................ ................................ .............................. - ................................ ................................ ..... ................................ ................................ ................................ ................................ ..... .......................... ................................ ................................ ................................ ....... - - ................................ ................................ ................................ .................. - - ................................ ................................ ..... ................................ ................................ ................................ ................................ ....... ....................... .................. ......... ....................... ................................ ....... 10 - MTHF 10 - Formyltetrahydrofolate 5,10 - MTHF 5,10 - Methylenetetrahydrofolate 5 - MTHF 5 - Methyltetrahydrofolate ABS absorbance ATP adenosine triphosphate BER Base Excision Repair BSA bovine serum albumin cAAG cyclic - AMP - AMP - GMP CBASS cyclic - oligonucleotide - based antiphage signaling system CDA cytosine deaminase c - di - AMP cyclic di - AMP c - di - GMP cyclic di - GMP cdN cyclic di - nucleotide cGAMP cyclic - GMP - AMP DAC diadenylate cyclase DCD deoxycytidylate deaminase dCMP deoxycytidin e monophosphate dCTP deoxycytidine triphosphate DGC diguanylate cyclase dTMP thymidine monophosphate dTTP thymidine triphosphate dUMP deoxyuridine monophosphate dUTP deoxyuridine triphosphate GMP guanosine monophosphate GTP guanosine triphosphate HCD high - cell density iNOS inducible nitric oxide system IPTG - D - 1 - thiogalactopyranoside LCD low - cell density MSHA mannose - sensitive hemagglutinin NK nucleoside/nucleotide kinase NMP nucleotide monophosphate NTP nucleotide triphosphate OD optical density ORF open reading frame PBS phosphate buffere d saline PDE phosphodiesterase QS quorum sensing RBS ribosome binding site TLD thymine - less death UDG uracil - DNA glycosylase UNG uracil - N - glycosylase VSP Vibrio Seventh Pandemic Island WT Wild - type Chapter 1: 1.1: - - 1.2: - - - - - - - - - - Schematic depiction of VSP - 1 and VSP - 2 based on the annotated El Tor V. cholerae N16961 genome. VSP - 1 is ~14 kb in length covering vc0175 vc0185 . The previously unannotated VSP - 1 ORF difV is l abeled in gray.VSP - 2 is ~25 kb in length covering vc0490 vc0516 . Not to scale. - - - - - - - - - - - 1.3: - - - - - - - - - - - - - - - - cdN synthases are in blue, the cdN products they make are in red, and examples of important biological processes these products regulate are in green. - 1.4: - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 1.5: - - - - - - - - - - - - 1.6: - - - - - - - - - - - - Left: The more conserved pair of enzymes, dncV ecoli and capV ecoli , predominantly synthesize and respond to cyclic - GMP - AMP, respectively. Right: The more degenerate pair, cdnE and capE , predominantly synthesize and respond to cyclic - UMP - AMP, respectively. 1.7: - - - - - - - - - - - - 1.8: - - - - - - - - - - - - The four previously known cdN molecules were all composed of purine nucleotides (c - di - GMP, c - di - - GMP - - GMP - AMP). With the identification of the bacterial CD - NTase family of enzymes, Whiteley et al. has expanded this signaling lexicon to include cyclic - UMP - AMP, cyclic - UMP - CMP, cyclic - di - UMP, cyclic - GMP - UMP, and the cyclic trinucleotide cyclic - AMP - AMP - GMP. Three possible cycli c dinucleotides remain to be discovered; cyclic - GMP - CMP, cyclic - CMP - AMP, and cyclic - di - CMP. - - - - - - - - - - - - - - - - - - - - - - - - 1.9: - - - - - - - - - - - - - - - - - - - - - - - - 1.10: - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 1.11: - - - - - - - - - - - - 1.12: - - - - - Chapter 2: - 2.1: - - - - 2.2: Chapter 3: - 3.1: - 3.2: The El Tor biotype of the Gram - negative bacterial pathogen Vibrio ch olerae is responsible for initiating and perpetuating the longest cholera pandemic in recorded history (1961 - current). Two genetic features that distinguish the El Tor biotype from strains of the classical biotype, responsible for the previous six pandemic s, are the presence of two genomic islands VSP 1 & 2. It was recently demonstrated that a four gene operon in VSP - 1 ( capV - dncV - vc0180 - vc0181 ) constitutes an anti - phage defense system, called CBASS, which is coordinated by cyclic - GMP - AMP (cGAMP) synthesized by DncV. Despite the significance of this finding many of the ~3 6 ORFs encoded in the VSP islands remain uncharacterized. We developed a bioinformatic pipeline to uncover other gene networks within the VSP islands by looking for the co - occurrence of islan d gene products within bacterial genomes. In addition to the known CBASS system, our analysis predicted dncV was involved in a gene network with the putative deoxycytidylate deaminase vc0175 , renamed here - in as d eoxy c ytidylate d eaminase V ibrio ( dcdV ). Whil e ectopic expression of dcdV in WT El Tor V. cholerae revealed no readily distinguishable phenotypes , a strain lacking VSP - 1, where dcdV is natively encoded, demonstrated reduced growth yield and a filamentous cell morphology. This filamentous phenotype is also inducible in laboratory strains of E. coli ectopically expressing dcdV that can be complemented back to wild - type cell morphology by supplying a single copy cosmid containing VSP - 1. This complementation was later attributed to a previously unannotate dcdV , named herein as dcdV insensitivity factor Vibrio (DifV). DcdV is a two domain protein containing a conserved deoxycytidylate deaminase (DCD) C - terminal domain and a putative nucleoside/nucleotide kinase (NK) N - terminal domain. The catalytic activity of the DCD domain performs the unique deamination of both dCMP and dCTP substrates, producing dUMP and dUTP. While the activity of the NK domain remains to be characterized, both the NK and DCD domain promote the in corporation of genomic dU and their combined activity is required to induce cell filamentation. While the biological utility of DcdV and its connection to DncV remain to be elucidated we discuss their relationship in the context of a novel phage - defense sy stem. 3.3: Vibrio cholerae , the etiological agent behind the diarrheal disease cholera, is a monotrichous, crescent shaped, Gram - negative bacterium found ubiquitously in marine environments. There have been seven recorded pandemics of cholera, begi nning in 1817, the first six of which are believed to have been caused by strains of the classical biotype. The seventh pandemic, which began in 1961 and continues to plague vulnerable populations today, was initiated and perpetuated by circulating strains of the El Tor biotype. Numerous phenotypic and genetic characteristics are used to distinguish the classical and El Tor biotypes [54] , but it is Vibrio Seventh Pandemic Islands 1 and 2 (VSP - 1 and 2) [4] , prior to the start of the seventh classic b iotype in modern cholera disease [3] . Combined, VSP - 1 and VSP - 2 represent ~3 6 putative ORFs encoded in ~39kb of genetic material that are typically separated by ~330 kb on the larger of the two El Tor V. cholerae chromosomes [4, 5] ( A ) ( A ) . While the majority of the genes in these two islands remain to be fully characterized, it has been hypothesized that the biological functions they encode may contribute to environmental persistence [55] and/or the pathogenicity [9] of the El Tor biotype. The ~26 ORFs encoded within VSP - 2 ( A ) , which include putative condensins, a chitinase, a pseudo - pilin, and numerous transcriptional r egulators, have yet to be interrogated for their specific contributions to El Tor fitness. Moreover, the genet ic composition of VSP - 2 appears to be more fluid across different strains of the El Tor biotype and non - pathogenic V. cholerae than VSP - 1 [8, 56] . In 2012, the first two genes characterized of the ~11 putative ORFs encoded within VSP - 1 were dncV ( vc0179 ) and a t ranscriptional repressor named v spR ( vc0176 ) [9] ( A ) . At the time DncV represented the first cyclic dinucleotide (cdN) synthase found in any living - AMP (cGAMP). cdN second messengers such as cGAMP, cyclic di - GMP, and cyclic di - AMP are intracellularly The intracellular concentration of a cdN dictates changes in diverse adaptive behaviors including biofilm formation, motility, stress response, and osmotic homeostasis at the levels of transcription initiation, post - transcription regulation, and allosteric interactions with effector proteins (reviewed in [10, 11, 13] ) . As described in Chapter 2, we found that an elevated intracellular concentration of cGAMP in El Tor V. cholerae resulted in cell toxicity, leading to the discovery of the cGAMP - activated phospholipase CapV ( vc0178 ) [18] . Both dncV and capV are encoded as part of a putative four gene operon ( capV - dncV - vc0180 - vc0181 ) in VSP - 1 and stimulation of CapV phospholipase activity by cGAMP leads to rapid degradation of the inner bac terial membrane and ultimately cell death. The function of the dncV - capV cGAMP signaling pathway in El Tor V. cholerae , along with vc0180 and vc0181 , has recently been suggested to coordinate an anti - phage defense system called the c yclic - oligonucleotide - b ased a ntiphage s ignaling s ystem (CBASS) [43] . CBASS executes an antiphage process termed abortive replication where an infected bacterium performs altruistic autolysis to prevent the virus from completing its replication cycle and contain the infection for the good of the bacte ria community [43] . Su ch functions for dncV and capV were demonstrated by expression of these genes in the heterologous host Escherichia coli followed by infection with E. coli specific phage , but such phage protection mediated by dncV and capV in their native genome locus in V . cholerae has not yet been documented. It has also been shown that dncV - like proteins, referred to as CD - NTase [28] or SMODS [27] , are capable of synthesizing a diverse array of nucleic acid compounds which are likely to specifically regulate the activity of neighboring effector proteins. CBASS is not limited to El Tor V. cholerae V SP - 1 as networks of homologous genes are shared in diverse mobile genetic elements found in other bacteria [43, 46] . The obse rvation that antiphage defense systems, such as CBASS, are frequently packaged together in mobile genetic elements, called defense islands [44, 57] , suggest that the seven remaining putative VSP - 1 ORFs, which include three transcriptional regulators ( vc0176 , vspR [9] , and vc0182 ), three xer like recombinases ( vc0183 , vc0184 , vc0185 ), and a deoxycytidyl ate deaminase - like protein ( vc0175 ), may also participate in anti - phage activities. Deoxycytidylate deaminases (DCD), part of the cytidine deaminase (CDA) family of enzymes, play a vital role in pyrimidine biosynthesis in many organisms by catalyzing the d eamination of deoxycytidine monophosphate (dCMP) to form deoxyuridine monophosphate (dUMP) [50] ( A ). With the exception of the deoxycytidine triphosphate (dCTP) deaminase enzymes [58] , the CDA deamination reaction is catalyzed in a conserved Zn - binding (A.) The activity of the c - amino group of cytidine to form uracil and free ammonium. (B.) Simplified generic de novo thymidine triphosphate biosynthesis pathway from dCTP and dCMP precursors. Enzymes involved in this proce ss are italicized and substrates in bold. The activities of DcdV are depicted in hatched lines; red for the C - terminal DCD domain and blue for the N - terminal NK domain. 5,10 - methenyltetrahydrofolate (5,10 - MTHF), tetrahydrofolate (THF). - amino group to form uridine and free ammonium ( A,B ) . The activity of these enzymes is critical for providing t he necessary dUTP/dUMP building blocks for deoxythymidine triphosphate (dTTP) synthesis. However, accumulation of these dU intermediates poses a risk to cells as DNA polymerases poorly discriminate between dUTP and dTTP allowing for the miss incorporation of dU in place of dT during DNA replication [59] . The activity of deoxyuridine phosphatases (dUTPase) reduce the likeliho od of erroneous incorporation of genomic dU by rapidly hydrolyzing available dUTP to dUMP, which is then converted into dTMP by thymidylate synthase (TS) using 5,10 - methenyltetrahydrofolate (5,10 - MTHF) ( B ) [50] . The delicate balance of enzymatic activity across the pyrimidine biosynthesis pathway can be corrupted by viruses that deploy their own DCD, dUTPase, and TS enzymes to hijack host nuc leotide biosynthesis to ensure the appropriate ratio of deoxyribonucleotide precursors for replicating their own A + T rich viral genomes [51, 60 63] . To predict novel biological pathways contained within El Tor V. cholerae VSP - 1 and VSP - 2 we performed a correlogy based bioinformatic analysis using a software package we call Correlogy , inspired by previou sly published method s [64, 65] . By mining the NCBI RefSeq non - redundant protein database for the co - occurrence of VSP island gene product homologs in bacterial genomes we identified numerous instances of co - occurring genes, called gene networks , across the bacterial phyla. Genetic components that make up a gene network are likely to function in a shared biological pathway or network. Our previous finding of the biological connection between dncV and capV [18] was identified as a gene network , validating this approach. Another identified VSP island gene network was a predicted association between the VSP - 1 genes dncV and the putative deoxycytidylate deaminase - like protein vc0175 [re named herein as d eoxy c ytidylate d eaminase in V ibrio ( dcdV )]. DcdV encodes both a predicted C - terminal DCD domain and an N - terminal nucleotide kinase (NK) domain. We found that ectopic expression of dcdV in El Tor strains lacking VSP - 1 and a heterologous Es cherichia coli host resulted in severely filamentous cells. This DcdV induced filamentous morphology is dependent on the catalytic activities originating from both the conserved C - terminal DCD, which deaminates both dCMP and dCTP substrates, and the less c onserved structural features present in the predicted N - terminal NK domain, whose activity has yet to be elucidated. We further show that DcdV induced cell filamentation is abrogated in WT El Tor V. cholera e and E. coli complemented with a cosmid containi ng VSP - 1, and this negative DcdV dcdV locus [renamed herein as D cdV i nsensitivity f actor in Vibrio ( difV )]. While the biochemical make up of DifV has yet to be determined, DcdV rep resents the first CDA family enzyme, to our knowledge, to be allosterically regulated by a sRNA or small peptide. We discuss the possibilities for dcdV dncV in bacterial genomes as well as the biological functions behind its activities a nd regulation as they relate to phage defense. 3.4: 3.4.1: The V. cholerae strains constructed in this study were derived from the El Tor biotype strain C6706str2 using the pKAS32 suicide vecto r, as described previously [66] . All vectors were constructed by Gibson Assembly (NEB). The vectors used for chromosomal deletions were generated by Gibson Assembly using three fragments: 500 bp of sequence upstream of the gene of interest, 500 bp of sequence downstream of the gene of interest, and cloned into the KpnI and SacI restriction sites of pKAS32. The ectopic expression vectors were construct ed by Gibson Assembly using PCR - amplified inserts, and pEVS143 and pBMM67EH linearized by EcoRI and BamHI, and pET28b digested with NcoI and XhoI. Site - directed mutagenesis used for the construction of DcdV variants were performed using the SPRINP method p reviously described [67] . Plasmids were introduced into V. cholerae through biparental conjugation using E. coli BW29427. For expression in E. coli, plasmids were transformed into DH10b by electroporation. E. coli BL21DE3 was used for the e xpression of pET vectors utilized in the in - vitro cell lysate deamination assay. Unless otherwise stated, all E. coli and V. cholerae were grown in Luria - Bertani (LB) at 35 o C. Medium was supplemented with the following as needed: ampicillin (100 µg/mL), k anamycin (100 µg/mL), tetracycline (10 µg/mL), and isopropyl - - D - thiogalactoside (IPTG) (100 µg/mL). E. coli BW29427, a diaminopimelic acid (DAP) auxotroph, was . - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 3.4.2: Overnight cultures were diluted 1:1000 into LB supplemented with antibiotics and IPTG in a 96 - well microplate (Costar®). The cultures were grown for ~15 hours in a BioTek Synergy HTX Plate Reader, with OD 600 measurements every 15 m in. 3.4.3: Cells were imaged as previously described [74] . Briefly, overnight cultures were diluted 1:1000 into LB supplemented with antibiotics and IPTG. Cultures were grown and induced for 7 - 8 hour, at which point cells were diluted to an OD 600 of 0.5 in 1X PBS, then membrane stain N - (3 - Triethylammoniumpropyl) - 4 - (6 - (4 - (Diethylamino) Phenyl) Hexatrienyl) Pyridinium Dibromide (FM4 - 64) (Sigma) was added to a final concentration of 20 µg/mL. 1% agarose pads in deionized water were cut into squares of approximately 20 x 20 mm and placed on microscope slides. 2 µl of diluted cultures were spotted onto a glass coverslip and then gently placed onto the agarose pad. FM4 - 64 signal was visualized using a Leica DM5000b epifluorescence microscope with a 100X - brightfield objective under RFP fluorescence channel. Images were captured using a Spot Pursuit CCD camera and an X - cite 120 Illumination system. Each slide was imaged with at least 20 fields of view for each biological replicate. Cell lengths were processed using the Fiji plugin MicrobeJ, and data were visualized and analyzed using R by quantifying the length of the curvilinear (medial) axis of detected cells. 3.4.4: - Cell Lysate Prep: Overnight cultures were sub - cultured 1:333 and grown at 35° C and 210 rpm shaking to an OD 600 of ~0.5 - 1.0. Cultures were induced with 1mM IPTG, supplemented with 100 µM ZnSO 4 , and grown for an additional 3 hr. Cell pelle ts from 100 mL of induced cultures were harvested in two successive 15 min centrifugation steps at 4k x g and 4° C. Supernatants were decanted and pellets were snap frozen in an ethanol and dry ice bath and stored at - 80° C. Pellets were thawed on ice and suspended in 2 mL of lysis buffer (50 mM NaPO 4 - mercaptoethanol, 20% glycerol). 1 mL of cell suspension was transferred to a microcentrifuge tube and sonicated on ice (20% amplitude, 20 sec total, 2.5 sec on, 2.5 sec off). Crude lysates were centrifuged at 15k x g for 8 min at 4° C and clarified lysates were transferred to fresh microcentrifuge tubes on ice. Clarified lysates were normalized for total protein to 1.9 mg/mL using Bradford reagents and a BSA standard. 26.5 µL reacti ons composed of lysis buffer, nucleic acid substrates, and 3.5 µL of normalized clarified lysates were assembled in PCR strip tubes, mixed by gentle pipetting, and incubated at 21° C for 60 minutes. NH 3 Cl solutions at the indicated concentration were disso lved in lysis buffer and substituted for nucleic acid substrates as positive ammonium controls. Ammonium Detection: The evolution of NH 4 + by deamination of the nucleic acid substrates was observed using a phenol - hypochlorite reaction to produce indophenol in a clear 96 - well microtiter plate and modified from Dong et al. 2015 [75] and the work of Ngo et al. [76] was considered when designing the lysis buffer . 50 µL of Reagent A (composition below) was added to each well followed by 20 µL of the completed in vitro deamination reaction described above. The phenol - hypo chlorite reaction was initiated by the addition and gentle mixing of 50 µL Reagent B (composition below) to the wells. The reaction was incubated at 35° C for 30 min and the ABS 630 was measured using a plate reader. Reagent A = 1:1:0.04 (v/v/v), water: 0. 5% (w/v) sodium nitroprusside (Sigma) in water: phenol solution (Sigma, P4557) Reagent B = 1:1 (v/v), 6% (w/v) sodium hydroxide (Sigma) in water: 1.5% (v/v) sodium hypochlorite solution (Sigma, reagent grade) in water . 3.4.5: Overnight cul tures were sub cultured 1:500 into fresh media supplemented with antibiotics and grown at 35 °C with 220 RPM shaking. After 1 h cultures were induced with 100 µM IPTG and allowed to continuing growing for another 1 h 45 m. 10 mL aliquots of each culture wa s removed prior to the addition of 50 µg/mL chloramphenicol (stop translation) and 250 µg/mL rifampicin (stop transcription). 10 mL culture aliquots were also removed 8, 16, and 32 min after addition of chloramphenicol and rifampicin. Each culture aliquot was immediately centrifuged at 4700 x g for 4 m at 22 °C, the supernatant was quickly removed by aspiration, the pellet was plunged into an EtOH + dry ice bath, and stored at - 80 °C. Cell pellets were thawed on ice and suspended in 1 mL lysis buffer (50 mM - DNAse I.. Cell suspensions were transferred to a microcentrifuge tube and sonicated on ice (20% amplitude, 20 sec total, 2.5 sec on, 2.5 sec off). Crude lysates were normalized for total protein using a Bradford assay. 34 µg of total protein per sample was loaded onto a single 10% acrylamide gel and proteins were transferred to Optitran reinforced nitrocellulose blot. The blot was bloc ked using skim milk and incubated with 1:5000 THETM His Tag Antibody, mAb, Mouse (GenScript) followed by 1:4000 Goat Anti - Mouse IgG Antibody (H&L) [HRP], pAb Amersham 600 Imager. Protein signal intensity was quantified using ImageJ. 3.4.6: The genomic uracil incorporation in vitro assay was carried out as described previously with slight modifications [77] . Briefly, genomic DNA was purified from cells grown in LB supplemented with antibiotics and IPTG for 7 - 8 hours using Wizard ® Genomic D NA Purification Kit (Promega). The purified genomic DNA (3 µg) was digested overnight at 37° C with 25 U uracil DNA glycosylase from E. coli (UDG) and 50 U human AP endonuclease 1 (APE 1) (New England Biolabs) in 1X NEB - buffer 4. Approximately 2.5 µg of ea ch reaction was loaded on a 0.8% agarose gel stained with EZ - Vision® Dye (VWR), and images were taken using GelDoc system (Bio - rad). 3.5: 3.5.1: - The two unique genomic regions, VSP - 1 and VSP - 2, are present in the seventh pandemic Vibrio cholera El Tor strains but not previous pandemic isolates [4] . The acquisition of these islands appears to be the final defining genomic event which allowed the El Tor lineage to emerge as the current pandemic strain [3] . However, the source of these genomic islands is unknown. BLAST searches for the entire VSP - 1 region using nucleotide or amino acid sequence does not return any hits in the current NCBI database, suggesting this order of genes is not present in any other sequenced genomes. A novel approach informed by previous work [64, 65] was developed to help define the biological function and groups of genes that function in a common pathway for these regions that lacked informative annotations. Many biological processes are the results of a set of gene products interacting to achieve a biological task such as metabolic pathways, cell wall synthesis, or molecular machines. The sets of gene products that accomplish a biological task can be n e s may physically interact or function in a common pathway. Many of these gene networks have deep evolutionary history and are widespread in many diverse taxa. We hypothesized that genes in a gene network will co - occur together in the genomes of diverse t axa at a higher frequency than chance alone would predict, and we used this reasoning to identify putative gene networks in VSP - 1. Kim and Price [64] previously explored genetic co - occurrence across the sequenced microbial datasets and - mathematical approach was never widely adopted nor was it implemented in publicly available software. genes that preferentially co - occur across the sequenced bacterial domain and assigns them to putative maximum relatedness subnetworks (MRS). These MRS can be considered as putative gene networks, with each MRS containing genes that preferentially co - occur in di verse taxa and may contribute to the same biological task. Importantly, Correlogy can only determine co - occurrence of genes across entire bacterial genomes; it does not take into account nor provide information on the spatial organization of these genes wi thin a given genome. To establish MRS for the VSP islands genomic region s we performed a BLASTP amino acid sequence search for each gene against the NCBI non - redundant protein database - ( A. ) Cartoon schematic of VSP - 1 from El Tor V. cholerae N16961. Not to scale. ( B. ) Gene network predictions for VSP - 1 where arrows show highest partial correlation (in italics) for each VSP - 1 gene (repre sented by ovals). Arrows pointing in opposing directions between the same two genes have the same . MRS = maximum relatedness subnetworks with an E - value cutoff of 10 - 4 . The BLAST results were limited to bacterial genomes, and all taxa belongi ng to the genus Vibrio were removed to avoid bias from closely related vertical inheritance. The BLAST results were used to generate a presence or absence matrix of VSP - 1 homologues with all species along one axis and VSP genes on the other axis. Next, a p airwise Pearson correlation value was calculated between all VSP - 1 genes i and j using binary data from the above - mentioned presence/absence matrix : - ( A. ) Cartoon schematic of VSP - 2 from El Tor V. cholerae N169 61. Not to scale. ( B. ) Gene network predictions for VSP - 2 where arrows show highest partial correlation (in italics) for each VSP - 2 gene (represented by ovals). Arrows pointing in opposing directions between the same two genes have the same . MRS = maximum relatedness subnetworks where N is the total number of unique species returned from the BLAST search and the number of species with co - occurrence of genes i and j. While a Pearson correlation is warranted for a normally di stributed binary data set it does not account for indirect correlation. For example, if genes i and j individually associate with a third gene a Pearson correlation will incorrectly calculate a correlation between i and j. To help correct for indirect corr elation we calculate a partial correlation w ij from the Pearson : where the ( i, j ) element of the inverse matrix of Pearson is . The partial correlation correction has the advantage of normalized output to a range of - 1 to 1. For exampl e, a of - 1 reveals genes i and j never occur in the same species, while a value of 1 demonstrates genes i and j always co - occur in the same species. A of 0 is the amount of co - occurrence expected between unrelated genes i and j drawn fr om a normal distribution. Using the above - mentioned approach, we calculated a partial correlation value for all genes i to j in VSP - 1 ( Supplemental File ) and VSP - 2 (Supplemental File 4 ) . Next, we chose to use the single highest value fo r each VSP - 1 gene to represent an edge ( i.e. line) in a visual MRS that suggests putative gene networks. This correlation - based visualization for VSP - 1 is shown in B and VSP - 2 in B . VSP - 2 island MRS networks are not explored further in this manusc ript. The MRS for VSP - 1 calculated capV as co - occurring most often with dncV , which reflects CapV activation by the DncV derived secondary messenger cGAMP [18] . Three other VSP - 1 genes ( vc0175 [ dcdV ] , vc0180, and vc0181 ) likewise have strongest correlogy with dncV ( Supplemental File 3 , ), sugg esting they too may be candidates for cGAMP regulation or involved in cGAMP signaling. Furthermore, the high of 0.501 shared between dncV and vc 0180 suggests the possibility of a shared activity . Previous work using well - classified E. coli gene n etworks showed that partial correlation values > 0. 0 45 were highly correlated with shared biological functions [64] . The predicted gene network composed of capV , dncV , vc0180 , vc0181 is consistent with their CBASS phage defense activity [43] . Curiously, the putative deoxycytid ylate deaminase dcdV , encoded distal to the CBASS , was found to co - occur with dncV ( of 0.147) ( , Supplemental File 3 ) . Recognizing this association might be indication of a novel biological function for dncV outside of CBASS we sought to understand the biological activity of dcdV and its relationship to dncV . 3.5.2: Based on our previous result that ectopic expression of dncV inhibited growth by activation of CapV [18] , we hypothesized that ectopic expression of DcdV could have a similar effect. To test this, we performed growth curves in both wild - type El Tor V. cholerae C6706str2 (WT), encoding dncV and capV, and a double VSP island deletion strain - 1/2) lacking dncV and capV, each over - expressing dcdV from a multi - copy plasmid under control of the P tac promoter (pDcdV) or an empty vector control (Vector). Contrary to our prediction, DcdV overexpression did not impact WT growth, but it did red - 1/2 background compared to the Vector control ( A ). Hypothesizing that changes in cell morphology could manifest in reduced growth yield we measured the cellular dimensions of e ach of the four strains in the growth curve after 8 h using fluorescence microscopy after staining the cell with membrane dye FM4 - 64. Strikingly, background yielded a filamentous cell morphology not found in the other t hree strains ( B ). Quantification of the cell - ( A. ) Growth curves of WT El Tor V. cholerae and VSP - 1/2 strains maintaining an empty vector plasmid (Vector) or P tac - inducible dcdV plasmid (pDcdV) grown in the presence of 100 µM IPTG. Error bars represent standard error mean from three biological replicates. ( B. ) Representative images of WT El Tor V. cholerae and VSP - 1/2 cultures maintaining an empty vector plasmid (Vector) or P tac - inducible dcdV plasmid (pDcdV) grown in the presence of 100 µM IPTG for 8 h. Cells were stained with FM4 - 64 prior to imaging. Scale represents 2 µm. ( C. ) Violin plots of c ell length distributions from ~2500 - 5000 cells pooled from three biological replicates with summary statistics: mean (diamonds), median (horizontal black line), interquartile range (box), and data below and above the interquartile range (vertical lines) Di ssimilar letters represent statistically significant differences between strains (p <0.05) determined by Two - - hoc multiple comparisons test. - ( A. ) Representative images of E. coli cultures maintaining an empty vector plasmid (Vector) or P tac - inducible dcdV plasmid (pDcdV) grow n in the presence of 100 µM IPTG for 8 h. Cells were stained with FM4 - 64 prior to imaging. Scale represents 2 µm. ( B. ) Distribution of cell lengths measured from three biological replicates of E. coli cultures carrying an empty vector (Vector) or P tac - indu cible dcdV plasmid (pDcdV) in addition to either an empty vector single copy cosmid (pLAFR) or pLAFR containing VSP - 1 (pCCD7) grown in the presence of 100 µM IPTG for 8 h. Distributions represent ~1000 - 2000 cells measured per strain Dissimilar letters repr esent statistically significant differences between strains (p <0.05) determined by One - - hoc multiple comparisons test. lengths measured in these four populations reve increased Vector, WT pDcdV, and WT Vector ( C ). SP - 1 and VSP - 2, we performed the same growth curve and image analysis in single island knock - origin of filamentation resistance. - 2 maintained a WT cell morphology when ectopically - expressing dcdV ( C ). Similarly, overexpression of pDcdV in a laboratory strain of E . coli also induced cell filamentation which could complemented to a non - filamentous morphology by providing VSP - 1 in a single copy cosmid (pCCD7) but not by an empty vector cosmid (pLAFR) ( A,B. ). Taken together, these results indicated that DcdV overexpression severely impact s cell physiology but only in the absence of a VSP - 1 encoded resistance factor we named d cdV i nsensitivity f actor in V ibrios ( difV ). 3.5.3: - Knowing difV was encoded in VSP - 1, we sought to identify its prec ise location by screening partial VSP - 1 island deletions for filamentation following pDcdV expression. Three sections of VSP - 1 were deleted based on gene orientations and organization including dcvD - vc0176 , vc0177 - vc0181, and vc0182 - vc0185 ( A ). Of the three partial VSP - 1 deletion strains, expression of pDcdV only induced filamentation in the dcdV - vc0176 mutant and this phenotype could be comp lemented back to a non - filamentous cell morphology by co - expression of dcdV - vc0176 from a separate P tac overexpression plasmid (p dcdV - vc0176 ) ( B ). To further narrow down the location of difV, we construct ed further deletions of dcdV , vc0176 , and the 452 nucleotide intergenic region (IG) between the two loci ( A) . When challenged with pDcdV expression both strains lacking either dcdV or vc0176 maintained WT cell morphology while the IG mutant became filamentous ( C ). Interestingly, IG is not filamentous in the absence of pDcdV expression indicating the laboratory conditions used throughout this study are not sufficient for the ( A. ) Schematic representation of VSP - 1 from El Tor V. cho lerae N16961 . Hatched lines depict deletions within VSP - 1 used to find difV . The intergenic region between vc0175 and vc0176 ( IG ) is highlighted in light gray while orf1 is highlighted in dark gray. Not to scale. ( B. ) Distribution of cell lengths measured from three biological replicates of gene deletions within VSP - 1 ( dcdV - vc0176, vspR - vc0181, and vc0182 - vc0185 ) maintaining an empty vector (Vector) or P tac - inducible dcdV plasmid (pDcdV) and complemented with a P tac - inducible dcdV - vc0176 plasmid (p dcdV - vc0176 ) grown in the presence of 100 µM IPTG for 8 h. ( C. ) Distribution of cell lengths measured from three biological replicates of gene deletions within VSP - 1 ( dcdV, vc0176, and the intergenic region between dcdV and vc0176 [ IG ]) maintaining a P tac - ind ucible dcdV plasmid (pDcdV) and complemented with a P tac - inducible orf1 plasmid (p orf1 ) grown in the presence of 100 µM IPTG for 8 h. Complementation of IG was done using pGBS80 and pGBS87. All cell length distributions represent ~700 - 3000 cells measured per strain. Dissimilar letters represent statistically significant differences between strains (p <0.05) determined by Two - - hoc multiple comparisons test. native dcdV locus to recapitulate a filamentous morphology althou gh it is possible that dcdV is not expressed in the IG mutant. S everal transcriptomic studies had previously identified IG as a hotspot for unannotated transcriptional activity including a putative transcriptional unit [78] , a putative transcriptional start site [79, 80] , a non - protein coding RNA [81] , and a number of putative sRNAs [82] . Taking these studies into account, we identified a potential small 222 NT ORF ( orf1 ) enc oded in the dcdV as a possible difV candidate ( A ). Remarkably, ectopic expression of orf1 was indeed sufficient to prevent pDcdV induced IG backgrou nd ( B). To test if translation of orf1 was necessary for inhibition of DcdV, we constructed an orf1 mutant where a stop codon was substituted for the native start codon, orf1 stop , but surprising ly this con struct also prevented pDcdV induced filamentation (data not shown) . These results indicate that the 222 NT in orf1 contains the necessary ge netic components for regulating DcdV activity and difV Distribution of cell lengths measured from three biological replicates of hapR and luxO maintaining an empty vector (Vector) or P tac - inducible dcdV plasmid (pDcdV) grown in the presence o f 100 µM IPTG for 8 h. Cell length distributions represent ~1300 - 3000 cells measured per strain. Letters represent statistical significance across strains (p <0.05) determined by Two - - hoc multiple comparisons test. is ei ther a sRNA or a small pept ide encoded in this region (< 74 AA) translated from an interior start codon. To determine if DifV is a sRNA we expressed pDcdV in an El Tor V. cholerae mutant with a deletion of hfq , a bacterial RNA binding protein that facilita tes many sRNA - mediated posttranscriptional gene regulations ( reviewed in [83] ) . Hfq pl ays a critical role in the stability of the four Qrr sRNAs that operate at the fulcrum of quorum sensing, helping to determine whether pathogenic Vibrios adopt a low - cell or high - cell density lifestyle [84] . We hypothesized that if DifV were an Hfq dependent sRNA the loss of hfq limit p DcdV induced filamentation. Interestingly, induction of DifV inhibition of DcdV induced filamentation is unlikely to occur at the level of translation or protein stability. ( A. ) Western blot a nalysis of DcdV 6xHis abundance in cell lysate from WT V. cholerae and VSP - 1 strains maintaining a P tac - inducible dcdV plasmid collected 0 32 min following chemical inhibition of transcription and translation that was preceded by 2.75 h of growth in the presence of 100 µM IPTG. ( B. ) Relative abundance of DcdV 6xHis measured from A. and calculated by dividing the protein signal detected at each time point by the initial protein signal (0 min) for each strain. Data represent a single biological replicate. hfq to filament indicating DifV is not an Hfq - dep endent sRNA (data not shown). S trains lacking either of the quorum sensing master transcriptional regulators luxO or hapR , which respectively coordinate low and high cell density gene expression, each presented mild pDcdV induced filamentation ( ). While these results do not rule out the possibility that DifV is a sRNA, it is clear that Hfq is not required for its activity and this activity is unlikely to be regulated by either quorum sensing s tate . 3.5.4: The regulatory mechanism DifV imparts on DcdV activity appears to be negative in nature but the molecular action utilized for this control is not clear. Three mechanisms are equally likely to occur; allost eric regulation of DcdV catalytic activity, inhibition of DcdV translation including transcript destabilization, or enhancing DcdV degradation. In regard to the second mechanism, most sRNAs that regulate translation interact with the ribosome binding site (RBS) of the target mRNA to prevent translation initiation. Translation of DcdV from the pDcdV overexpression plasmid is driven by a non - native RBS, yet DcdV activity from this plasmid can be inhibited by DifV. This finding suggests that the RBS of dcdV mR NA is not the target of DifV, although it remains possible that if DifV is a sRNA it could interact with other parts of the dcdV mRNA. To understand if protein production or stability of DcdV was impacted by DifV, we over - expressed a 6 x His C - terminal tag ged construct of DcdV (pDcdV 6xHis ) in both WT El Tor and - 1 during exponential growth for an hour, chemically halted transcription and translation, and collected protein samples at regular intervals for 32 minutes. Addition of the affinity tag in DcdV 6 xHis - 1 predictably filaments while WT does not when challenged with ectopic expression of this construct (data not shown). Western blot analysis of soluble protein extracts reveal DcdV 6xHis is initially abundant in bot h strains and detectable throughout the time course of the experiment ( A). Because DcdV levels are not decreased at time 0, the function of DifV is not to reduce DcdV production in the cell via a post - tran scriptional regulatory mechanism. Surprisingly, the rate of DcdV 6xHis degradation is - 1 background , while its abundance is relatively unchanged in the WT strain over the course of the experiment ( B). While this experiment was only performed a single time , these results suggest that DifV does not regulate DcdV activity by destabilizing the protein. Additionally, the limited degradation of DcdV 6xHis in the WT strain suggests the activity of DifV holds DcdV inactive while simultaneously protecting it from pr oteolysis. These experiments rule out two potential mechanism for DifV regulation of DcdV activity, that of decreasing DcdV production or increasing its degradation, and suggest that DifV allosterically controls the enzymatic activity of DcdV. 3.5.5: - We next sought to determine how DcdV activity leads to filament formation. DcdV is a 532 AA poly peptide composed of two putative domains; an unannotated N - terminal domain and a putative DCD - like C - terminus. We utilized the Phyre2 protein modeling suite [85] and psiBLAST to aid in the identification of conserved structural and primary sequence features present in DcdV that would give clues to its activity. Phyre2 confidently matched the DcdV termini with two independent protein families; the N - terminus contained features of Nucleoside/Nucleotide Kinases (NK) super family enzymes, while the C - terminus closely ali gned to structural homologs of DCD enzymes which are part of the zinc - dependent CDA family ( A). - phosphate from a nucleotide triphosphate donor to a diverse group of substrates, depending on the enzyme class, including deoxynucleotide mono - phosphates. Three structural features commonly found in these enzymes include a P - loop/Walker A motif {GxxxxGK[ST]} and a two helical LID module that together stabilize the donor nucleotide triphosphates, and a Walker B motif {hhhh[D/E], where h represents a hydrophobic residue} is partly involved in coordin ating Mg 2+ [86, 87] . Interrogation of the Phyre2 DcdV mod el and psiBLAST primary sequence alignments revealed these three features are likely present in the N - terminal domain ( B). This observation - - ( A. ) Cartoon model of DcdV from Phyre2. NK - like domain features: Walker A = yellow, Walker B = violet, and the LID module = blue. DCD - like domain features: Zn 2 + active site = red, and the non - conserved dNTP regulator motif: green. ( B. ) Close - up view of the residues identified for mutation within conserved features of the NK - like domain. ( C. ) Close - up view of the residues identified for mutation within conserved features of the DCD - like domain. suggests the N - terminus of DcdV is an NK super family domain involved in binding nucleotide substrates and performing a phosphotransfer reaction. The zinc - dependent CDA active site motif, [CH] - A - E - X (21 - 37) - P - C - X (2 - 8) - C [88] , is highly conserved in the C - terminal domain of DcdV ( C ). The constellation of residues that make up the Zn 2+ binding pocket is composed of three critical amino acids; H382, C411, and - - Distribution of cell lengths measured from three biological replicates of E. coli main taining a P tac - inducible dcdV plasmid (DcdV WT ) or a variety of DcdV variants grown in the presence of 100 µM IPTG for 8 h. E. coli cell length distribution for WT DcdV is depicted in dark gray while variants in the Walker A motif are in white, Walker B in light gray, and the DCD Zn 2+ active site are in medium gray. Cell length distributions represent ~1700 - 3000 cells measured per strain. Letters represent statistical significance across strains (p <0.05) determined by Two - - hoc multiple comparisons test. C414. This Zn 2+ is required for the catalytic deprotonation of water by E384 for the hydrolytic deamination of a cytosine base to uridine. Members of the CDA enzyme family specifically catalyze the deamination of a diverse assortment of cytosine containing substrates including free cytidine, deoxy cytidine mono and trisphosphate, RNA and ssDNA polynucleotides [51, 52, 89, 90] . DCD enzymes are unique among the CDAs for their allosteric regulation by dCTP and dTTP which activate and repress the deamination of dCMP, respectively, through a G[Y/W]NG allosteric site motif [91, 92] . Interestingly, DCD allosteric regulation by dNTPs may not be preserved in DcdV as this motif is composed of a divergent GCND. Lysates collected from E. coli expressing DcdV or DcdV E384A incubated with 12 nucleic acid substrates (1.9 mM NH 3 Cl as a positive control, 37.7 mM cytidine, and 7.5 mM for all other substrates). The evolution of NH 4 + resulting from substrate deamination wa s detected by measuring the solution ABS 630 microtiter plates. Data represent the mean and standard deviation of three biological replicate cal significance (p<0.05) determined by Two - - hoc multiple comparisons test. Error bars represent standard deviation. Hypothesizing that one of the two domains present in DcdV was responsible for cell filamentation in the absence of difV we made site - specific mutations in the conserved residues in both the NK and CDA domains. Six variant constructs were generated from the NK domain; four in the Walker A motif (pDcdV S52P , pDcdV S52W , pDcdV S52K , and pDcdV K55A ) to introduce st eric bulk and interfere with hydrogen bonding and a single variant with a double substitution in the Walker B motif (pDcdV D162A,Q163A ). Two variants were Lysates collected from E. coli expressing DcdV incubated with or without 7.5 mM dTTP and either 75 mM cytidine, 7.5 mM dCMP, or 7.5 mM dCTP. The evolution of NH 4 + resulting from substrate deamination was detected by measuring the solution ABS 630 reaction in microtiter plates. The relative deaminase activity was calculated by dividing the ABS 630 of the +dTTP reaction by the no dTTP control reaction for each lysate. Data represent the mean and standard deviat ion of three biological replicate lysates. constructed in the CDA active site; a double substitution of both C411A and C414A (pDcdV C411A,C414A ) to abrogate Zn 2+ binding, and a E384A substitution (pDcdV E384A ) to inhibit the deprotonation of water required for the hydrolytic deamination of cytosine. Surprisingly, only ectopic expression of pDcdV K55A was capable of inducing mild filamentation while the remaining seven variants, irrespective of domain or feature, failed to induce filamentation in E. coli ( ) . Global destabilization of DcdV resulting from these disparate substitutions is unlikely to explain the loss of DcdV induced filamentation in all cases as 6 x His - tagged constructs of the DCD variants main tain solubility and similar abundance to WT 6 x His - tagged DcdV (data not shown). Further investigation of the Walker A variant, K55A, will help determine whether the NK domain activity has been inhibited entirely or if it remains mildly active. Taken toge ther, these results indicate the f unctions performed by both the NK and CDA domain are required for DcdV induced filamentation. 3.5.6: Enzymes belonging to the CDA family have a high degree of specificity for their substrates [92] . To determine the substrate preference of DcdV we expressed affinity - tagged dcdV and the CDA active site variant dcdV E384A cloned under the T7 promote r on a high - copy plasmid in BL21(DE3) E. coli . Soluble lysates from each s train were incubated with a battery of amine containing nucleic acid substrates and monitored for the evolution of NH 4 + using a colorimetric assay in microtiter - plates. Of the 12 substrates tested, lysates containing DcdV produced ammonium when incubated w ith dCMP and dCTP, which was not detected in lysates containing DcdV E384A ( ). Both strains showed similar levels of cytosine deamination indicating the presence of endogenous E. coli CDA activity ( ) . Interestingly, we found that deamination of dCMP and dCTP by DcdV lysates was not inhibited by the addition of dTTP, a common negative allosteric regulator of DCD enzymes ( ). Together, these results demonstrate that DcdV is a DCD that specifically catalyzes the deamination of both dCMP and dCTP to produce dUMP and dUTP, respectively, and this activity was resilient in the presence of excess dTTP. ung E. coli maintaining an empty vector plasmid (Vector ) or P tac - inducible dcdV plasmid (pDcdV) or P tac - inducible dcdV inactive variant plasmid (DcdV E384A ) grown in the presence of 100 µM IPTG for 8 h. Cell length distribution for Vector are filled in white, pDcdV in light gray, and pDcdV E384A in dark gray. Ce ll length distributions represent ~1700 - 3000 cells measured per strain. Letters represent statistical significance across strains (p <0.05) determined by Two - - hoc multiple comparisons test. 3.5.7: The accumulation of dUTP is problematic for most organisms as DNA polymerases poorly discriminate between dUTP and dTTP [59] . This leads to the erroneous incorporation of dU into replicating strands of DNA, which is normally recognized and removed by the base excision repair (BER) pathway [93] . Excessive incorporation of dU in DNA can lead to over stimulation of BER resulting in double strand breaks and ultimately cell death. In the absence of uracil - DNA glycosylase (Ung or UDG), which initiates BER by cleaving uracil bases from From top to bottom: The in vivo activity of DcdV increases the cellular abundance of dUMP and dUTP leading to the miss incorporation of dU into genomic DNA following replication. In the ab sence of ung, genomic dU cannot be efficiently removed. The in vitro addition of purified UDG and ApeI to isolated genomic DNA leads to the hydrolytic cleavage of uracil and the phosphodiester backbone at the resulting abasic site. The level of DNA fragmen tation following agarose gel electrophoresis is indicative of the relative abundance of genomic dU. U = uracil, squares represent generic non - uracil bases, dotted lines represent in vivo activity, hatched line depict in vitro processes. DNA, a cell is un able to recognize and efficiently remove dU from its genome [94] . Therefore, the accumulation of genomic dU in a background can be used to infer the r elative abundance of intracellular dUTP. To test if DcdV increases the pool of available dUTP and its ung E. coli maintaining an empty vector plasmid (Vector), P tac - inducible dcdV plasmid (pDcdV), or P tac - inducible dcdV DCD active site mutant (pDcdV E384A ) grown in the prese nce of 100 µM IPTG for 8 h. A, B, and C represent genomic DNA isolated from three biological replicates for each strain. - and + indicate genomic DNA was incubated in the absence or presence of purified UDG and ApeI, respectively. incorporation into the genome, we expressed pDcdV, pDcdV E384A , and Vector in a ung strain of E. coli , collected genomic DNA, and measured cell morphologies after 8 h. Interestingly, the E. coli mutant expressing pDcdV was elongated, showing that Ung dependent removal of uracils in the DNA was not necessary for the filamentation phenotype ( ). As observed before, the DcdV E384A variant was unable to induce filamentation ( ). The isolated DNA was incubated with purified UDG to remo ve incorporated uracil bases and the endonuclease ApeI to cleave the DNA at abasic sites followed by agarose gel electrophoresis to investigate DNA fragmentation ( ). Isolated DNA from expressing pDcdV treated with UDG and ApeI was highly fragmented relative to the Vector control strain, indicating a dramatic increase in the abundance of genomic dU resulting from DcdV activity ( ). Surprisingly, despite being unable to catalyze the formation of dUTP ( ), the pDcdV E384A ung genomic DNA also had increased incorporation of dU in its genome relative to the Vector strain, although it ins erted less uracil than expression of WT DcdV ( ). Taken together these results demonstrate that DcdV activity increases the abundance of dUTP in the cell by two mechanisms; DCD activity and a second pathwa y originating from the NK - domain yet to be elucidated. While the relative comparison of genomic dU through digestions and gel electrophoresis is an indirect method for analyzing nucleotide abundance it is reasonable to expect more quantitative methods will reveal an imbalance in nucleotide pools, promoting dUTP formation, that coincide with contributions from each domain. 3.6: Characterization of the ~3 6 genes encoded within the El Tor V. cholerae VSP - 1 and 2 genomic islands is still in its infancy an d uncovering their contributions to bacterial fitness may help elucidate the origin and persistence of the seventh cholera pandemic. While the general architecture and composition of VSP - 1 and 2 are mostly preserved in Vibrio sp., it is clear from our corr elogy study that gene clusters from these islands are widely distributed across the bacterial phyla. In support of this method, our study accurately identified a correlog composed of the VSP - 1 antiphage CBASS system ( capV - dncV - vc0180 - vc0181 ) ( ) . Interestingly, this bioinformatic survey also revealed dncV , whose production of cGAMP is critical for CBASS induced abortive replication, is frequently found in genomes with the previously uncharacterized gene vc01 75 ( dcdV ) ( ) . Here, we showed that DcdV contains a functional DCD domain that catalyzes the deamination of deoxycytidine nucleotides and a putative NK - like domain. In total, DcdV activity appears to distor t deoxynucleotide pool homeostasis which manifests in a filamentous cell morphology. This abnormal morphology is abrogated by a novel post - translational mechanism by difV , a genetic component within a 222 NT region immediately dcdV locus in VSP - 1 ( A) , though the details of this process remain to be fully understood. The deamination of dCTP is canonically performed by non - zinc dependent enzymes [58] making the dual substrate repertoire of dCMP and dCTP in DcdV a rare trait. The only other DCD reported to work on both dCMP and dCTP is biDCD from chlorovirus PBCV - 1 [51] . PBCV - 1 has a double stranded DNA genome with a ~40% G + C content while its microalgae host, Chlorella NC64A, has a genomic content of ~67% G + C. During infection, biDCD, along with other viral peptides, rapidly converts host dC nucleic acids pools into dTTP thus facilitating replication of the A + T rich viral genome. Like many DCD enzymes, biD CD is allosterically regulated by dCTP and dTTP that activate and inactivate the deaminase, respectively, providing a means to fine - tune the pool of available dNTPs [91, 92] . Interestingly, DcdV does not appear to have maintained the allosteric nucleotide binding site, suggesting it is unlikely to modulate steady state nucleotide concentrations. In support of this, addition of excess dTTP to cel l lysates did not alter the catalytic activity of DcdV towards dCMP or dCTP in trial experiments ( ). Though the biochemical complexity of cell lysates may obscure potential regulation, our results suggest DcdV deamination of dC nucleic acids proceeds unencumbered by a feedback mechanism meant to preserve dC substrates for DNA replication. In lieu of a conserved deoxynucleotide allosteric site, DcdV is regulated post - translationally by DifV. Regardless of w hether DifV is a peptide or a regulatory sRNA, to our knowledge this type of regulation is unique among the CDA - family. The spacing, orientation, and relationship of difV and dcdV are vaguely reminiscent of Type 2 and Type 3 [95] Toxin - Antitoxin (TA) systems found across the bacterial phyla. These TA types utilize a small peptide or sRNA antitoxin to allosterically inactivate a co - transcribed toxin peptide and degradation of the antitoxin results in the liberation of the toxin ( reviewed in [96] ). Utilization of a Type 2 or 3 TA system - like regulation would allow for the targeted deployment of DcdV catalytic activity under specific conditions by the degradation of DifV. Additionally, the utility of deaminase domain c ontaining proteins as potent toxins has been illustrated by their inclusion in bacterial polymorphic toxin systems used for extracellular kin selection [97, 98] . However, unlike these systems the genomic architecture and lack of a clear translocation mechanism suggest that dcdV and difV are not utilized extrac ellularly. Interestingly, under our experimental conditions the presence of genomic difV is sufficient to inhibit ectopically expressed dcdV , while loss of difV does not result in filamentation from genomic dcdV activity. This observation implores further experimentation to understand the transcription and biological conditions under which difV and dcdV are expressed and active. Furthermore, outstanding questions remain as to the chemical makeup of DifV and the mechanism of action utilized to inhibit one or both DcdV domain activities. Cell filamentation is a hallmark of thiamine - less death (TLD), observed in prokaryotes and eukaryotes, which arises from a sudden loss of thiamine during robust cellular growth [99] . Insufficient dTTP substrate during DNA replication leads to the disintegration of replication forks and the accumulation of unresolved ssDNA that cause genomic instability and cell death [100] . Interestingly, this phenomenon is not limited to dTTP as dGTP starvation elicits a similar response in E.coli and is also hypothesized to occur when other deoxyn ucleotide substrates become disproportionately scarce [101] . It has also been sh own that even modest changes in The catalytic activity of DcdV (dark gray) leads to an accumulation of dUTP/dUMP. Two hypothetical and c omplementary pathways for this activity to contribute to phage defense are; DncV dependent (blue) and DncV independent (green). Major questions/assumptions in these pathways are marked with a question mark (red). The DncV dependent pathway increases the sy nthesis of deoxyribonucleotides, depicted here as the biosynthesis of dTMP by thymidylate synthase (TS) which utilize 5,10 - MTHF. This reduces the available 5 - MTHF and de - represses DncV activity. Liberated DncV produces cGAMP to activate CapV and initiate a bortive replication. The DncV independent pathway involves the corruption of replicating DNA phage genomes by misincorporation of dUTP substrates, thus targeting these genomes for scrutiny by endogenous DNA repair mechanism. A possible mechanism for the ac tivation of DcdV is the phage dependent inhibition of DifV (purple). dNTP pools can dramatically increase mutation rates [102] and an abundance of dUTP substrat es can lead to erroneous incorporation of genomic dU [59] . In the case of DcdV, it is conceivable the observed filamentat ion phenotype is a consequence of a TLD - like corruption of deoxynucleotide pools. Our evidence suggests the disparate activities encoded in the DCD and NK domains have been evolutionarily fused to enhance the formation and persistence of dUTP ( ) to the detriment of both dCTP and dTTP pools , respectively ( ) . Inactivating mutations in conserved catalytic features of either domain are sufficient to inhibit D cdV induced cell filamentation ( ) suggesting the native pyrimidine biosynthesis architecture is robust enough to handle a single perturbation in one deoxynucleotide pool, but not a simultaneous assault on both. We speculate that the NK domain catalyzes phosphotransfer between nucleotide substrates promoting formation of dUTP, which then increases the incorporation of dU into the genome in the absence of the DCD domain ( ). Future experiments using more quantitative methods to measure deoxynucleotide pools will be critical Overall , our evidence suggests that DcdV and DifV function in concert to prepare a bacterium for rapid adaptation to a specific cellular condition that warrants a dramatic change in Given that dcdV co - occurs with dncV , of the CBASS system, and phage defense systems often co - occur in defense islands [44, 57] , we hypothesize DcdV and DifV compose a novel phage defense mechanism. Based on our results, DcdV activity could facilitate p hage defense in two ways; increasing the mutational frequency and incorporation of dU in replicating viral genomes (DncV Independent) and de - repressing DncV cGAMP synthesis (DncV Dependent) ( ) . Even moder ate destabilization of dNTP pools has profound mutagenic consequences [102] and the mis incorporation of dU into rapidly replicating viral genomes could flag them for scrutiny and processing by native DNA repair mechanisms. To this point the phage of Bacillus subtilis deploys the protein p56 to inhibit host UNG activity thus preventing BER from targeting replicating viral genomes that have incorporated dU [103] . Synthesis of cGAMP by DncV has been shown to be allosterically inhibited by certain species of folates in cluding 5 - methyltetrahydrofolic acid (5 - MTHF) [23] . The two precursors of 5 - MTHF, 10 - formyl tetrahydrofolate and 5,10 - MTHF, are utilized for the de novo biosynthesis of purines and dTMP, respectively [50] . In the face of DcdV activity a native response to restore dNTP homeostasis could rapidly deplete folates leading to the de - repression of DncV cGAMP synthesis and initiate abortive replication ( ) . These pathways are not mutually exclusive and their combined effects could interfere with viral genome replication while catalyzing the DncV initiated abortive replication process. Further experiments are required to understand the relationship between dncV and dcdV and how the activity of one may influence the other, including de - repression by folates or DifV, respectively. Chapter 4: 4.1: Since 1817, pathogenic strains of V. cholerae have caused seven pandemics of the diarrheal disease cholera [3] . The seventh pande mic began in 1961 and continues to be perpetuated by strains of the El Tor biotype. Despite their genetic similarity and a shared ancestor, the El Tor biotype displaced the previous classical pandemic biotype in both environmental and clinical reservoirs. islands VSP - 1 and 2 in the mid - 20 th [3] . When considering the role of the VSP islands in El Tor V. cholerae three questions remain to be answered: what biological functions do they encode, what is their utility to bacterial fitness, and where did they come from? Through the work pr esented in this thesis I have endeavored to help answer these questions. 4.1.1: Following the identification of DncV in 2012, the cdN cGAMP remained an orphan second messenger for six years. Our discovery of the phospholipase CapV in 2018, de scribed in Chapter 2, and its allosteric regulation by cGAMP represented the first cGAMP signaling network described in bacteria [18] . By connecting the activity of DncV to CapV we f acilitated the identification of other discrete signaling networks across the bacterial phyla involving the dncV - like family of nucleotydil tranferases (CD - NTases or SMODS) [27, 28] . The work by Whiteley et al. drove a paradigm shift in the field of cdN signaling by expanding the repertoire of cdN second messengers and demonstrating the c atalytic flexibility of the CD - NTase family. In short order the laboratory of Rotem Sorek connected the cGAMP activation of V. cholerae CapV, and an orthologous E. coli system, with a novel phage defense system they named CBASS [43] . As the field of bacterial cdNs has rapidly e xpanded in the past two years, there remain a number of critical questions related to regulation and deployment of capV - dncV - vc0180 - vc0181 in El Tor V. cholerae and CBASS systems generally, and I will briefly discuss these topics. 4.1.2: Prediction of the possible gene network between dncV and the putative deoxycytidylate deaminase vc0175 , now called dcdV, may expand our understanding of CBASS systems and their connection to genes not encoded in a shared operon ( Supplemental File 3 , ) . DCD enzymes typically play a role in the de novo biosynthesis of dTMP by providing the necessary dUMP building blocks ( ). DcdV appears to be an abnormal DCD enzyme in its ability to perpetually deaminate dC substrates in spite of excess dTTP ( ). Interestingly, the dcdV , may have replaced the need for dTTP inhibition and may be important for the appropriate deployment of DcdV. Additionally, fusion of the DCD and NK domains fo und in DcdV is a novel configuration for CDAs, and their shared role in the likely production of dUTP is deleterious for the bacterium in which it is expressed ( , ) . The activity of DcdV is not completely dependent on cGAMP, as the enzyme is active in the absence of DncV. Whet her or not DcdV/DifV is influence d by cGAMP, or this system itself influences DncV activity, remains to be addressed. I will briefly address this and other outstanding experimental questions in the arena of DncV, DcdV, and DifV in a later section. 4.1.3: Thus far, five VSP island genes have been experimentally analyzed ( dncV, vspR, capV, dcdV, and difV ) [9, 18] and the work presented in this thesis has made fundamental contributions to understanding four of them. D espite these efforts, ~31 ORFs across the VSP islands remain unstudied. Our collaborators in the laboratory of Eva Top at the University of Idaho developed a bioinformatic tool, Correlogy, to predict conserved gene pathways within the VSP islands by identi fying the co - occurrence of individual genes across bacterial genomes ( , , Supplemental File , Supplemental File 4 ) . This analysis accurately identified two known VSP island biological networks; the VSP - 1 encoded CBASS system [43] and the three gene operon vc0490 - vc0491 - vc0492 encoded in VSP - 2 not discussed in this thesis but under investigation in the Waters lab. Correlogy also predicted numerous additional gene networks within t he VSP islands that will help to prioritize the further characterization of the function of these genes. Validation of these predicted gene networks may help to uncover the origins of the VSP islands and illuminate the evolutionary pathways that converged to generate them in the El Tor biotype. Additionally, Correlogy will be publicly available and this tool can be used to explore potential gene networks in any collection of bacterial genes. 4.1.4: Arguably, one of the mo st interesting questions that still remains to be answered is; why the El Tor biotype? For reasons yet to be elucidated, VSP - 1 and 2 were exclusively acquired by the progenitor strains of the El Tor biotype. Recognizing that the evolution of seventh pandem ic El Tor V. cholerae and its acquisition of the VSP islands occurred in the presence of classical strains begs the question; what was special about the progenitor El Tor strains that led to acquisition of these islands when the classical strains did not? By combining the biological functions encoded within the VSP islands with genomic comparison between the biotypes we will likely gain insight into this question. 4.2: 4.2.1: 4.2.1.1: A key feature of CBASS systems yet to be identified is the mechanism by which they are appropriately deployed. For efficient phage defense it is likely that all the necessary protein machinery (CD - NTase, effector, etc.) must be translated prior to phage insult. In the case of El Tor V. cholerae , capV and dncV expression is regulated by QS suggesting that at high - cell density both proteins are present in the cytoplasm at concentrations sufficient to kill the cell. In support of this hypothesis, ectopic expression of dncV lea ds to cell death in El Tor by activating endogenously produced CapV ( Chapter 2 ). A mechanism must therefore be established that prevents endogenously produced DncV from inappropriately activating CapV. In the case of HORMA CBASS, the formation of the activ e HORMA:CdnC complex requires the presence of an unidentified short peptide sequence of unknown origin that is abundant during infection [46] . While DncV activity is regulated by a variety of folates species, including 5 - MTHF, it is unclear why DncV has evolved to be regulated by them [23] . Hypotheses regarding the regulation of DncV by post translational - modification by VC0180 an d a connection to DcdV activity are briefly explored later. Finally, in order not to inappropriately kill their host, CBASS systems must be finely tuned to recognize an invading phage. Understanding this regulation is critical to understanding the utility of CBASS systems and will likely reveal mechanisms phage exploit to neutralize them. 4.2.1.2: The activity of CapV and other CBASS effectors is controlled through allosteric interactions with specific cyclic oligo - nucleotides [18, 28, 46] . CBASS systems are frequently found within mobile genetic elements [43] passed through horizontal gene transfer events where they are likely to encounter hosts with preexisting nucleotide second messenger systems. Theref ore, CBASS effector allosteric sites must be exquisitely insulated from interference In the case of CapV, V. cholerae contains an extensive c - di - GMP signaling network which can range from low nM to low µM intracellular concentrations of c - di - GMP [104] . We have shown in vitro that the phospholipase activity of CapV is specifically regulated by cGAMP and n ot c - di - GMP or c - di - AMP ( Chapter 2 ). In the case of E. coli ECOR31, two homologous CapV and DncV systems exist where each phospholipase specifically recognizes only the cdN synthesized by its partner dncV - like synthase ( ) [28] . The extensive catalogue of cyclic - oligonucleotides synthesized from the common catalytic motif found in CD - NTases ( ) [28] suggests the allosteric sites which recognize them may also share common features. Sequence al ignment and structural studies of these effectors will help to uncover ligand binding motifs utilized by CBASS systems. Additionally, these studies may help to predict effectors not encoded within CBASS operons that are still able to respond to them. 4.2.1.3: Connecting the catalytic activities of DncV and CapV to abortive replication represented the monumental breakthrough in our understanding of the VSP - 1 [43] . The architecture of the El Tor CBASS system includes vc0180 and vc0181 and the intact homologous system from E . coli provided phage immunity to a heterologous host for 6 of the 10 phage studied [43] . VC0180 is a two domain peptide composed of a putative E1 (ubiquitin - activating enzyme) and E2 (ubiquitin - conjugating enzyme) while VC0181 is a JAB domain (deubiquitinase). Infections of E . coli that expressed mutations in conserved features of the E1/E2 and JAB domains while maintaining wild type capV and dncV reduced phage protection to a single phage [43] . These results suggest that this putative ubiquitination system is central in the function of dncV and ca pV, but its activity and role in this process have not been studied. Ubiquitination in bacteria is not recognized as a method of post translational modification and understanding its role in phage defense would greatly expand this field of study [105] . In the canonical eukaryotic ubiquitination process the E1 domain adenylates the C - terminal carb oxylate of the small ~7 kDa protein ubiquitin (reviewed in [106] ). Adenylated ubiquitin is then transferred to a conserved cysteine residue located on the E1. This transfer step releases AMP during the formation of a labile thioester between the C - terminus of ubiquitin and the E 1 cysteine. Ubiquitin is then transferred to a conserved cysteine residue in an E2 enzyme. Finally, ubiquitin is transferred to specific lysine residues located on target proteins, creating an isopeptide bond. This process is often facilitated by a third e nzyme called E3 that does not typically covalently interact with the ubiquitin, although V. cholerae does not encode any proteins that appear to have an E3 domain. Ubiquitination frequently marks proteins for degradation, but in other cases ubiquitination serves as a post - translational modification to regulate the activity of the target protein. The isopeptidase activity of deubiquinating enzymes, like JAB domains, cleaves ubiquitin from ubiquitinated proteins, thus removing the post - translational modificat ion. Ubiquitin and ubiquitin - like proteins are often small ~ 7 kDa proteins, and canonically they contain a C - terminal GG motif. Analysis of the region within and surrounding the El Tor CBASS does not reveal any such small peptide . The dichotomous putative activities of VC0180 and VC0181 are reminiscent of the HORMA CBASS system where the activity of the Trp - 13 prevents formation of the HORMA:CdnD complex in the absence of a unknown peptide, hypothesized to be brought by invading phage or produced in respon se to infection [46] . Therefore, we hypothesize the ubiquitin - like protein that is utilized by VC0 180 and VC0181 is likely to be available under similar circumstances. Ser and Lys amino acid substitution of conserved Cys residues in E2 enzymes have been previously used to capture ubiquitin modification covalently linked to the ligase [107, 108] . We have constructed tagged variants of conserved E1/E2 Cys residues found in VC0180, which can be used in future ex periments to capture the ubiquitin - like protein modifier. By expressing these variants in a variety of cellular conditions, including phage infection, it may be possible to identify the ubiquitin - like peptide and determine its origin. Additionally, charact erization of this protein modifier will facilitate the identification of proteins targeted for modification by VC0180 to understand how this activity contributes to phage immunity. Interestingly, our Correlogy analysis found that DncV and VC0180 have an MR S of 0.501 ( , Supplemental File ), which is much greater than the 0. 045 threshold suggested for a gene network determined by Price & Kim [64] . cGAS, the eukaryotic homolog of DncV, is regulated by ubiquitination at a number of specific Lys residues (reviewed [109] ). A complicated hierarchy of ubiquitination and SUMOylation, a related post - translation modification, is used to coordinate cGAS stability as well as regulate the synthesis of cGAMP. These lines of evidence sugge st DncV is a likely target of VC0180 directed post - translation modification and structural comparisons of DncV and cGAS may reveal candidate residues involved in this process. 4.2.2: - 4.2.2.1: The in vitro activity of the DCD deaminase domain of DcdV was shown to deaminate both dCMP and dCTP substrates ( ), making it the second DCD enzyme described with this catalytic repertoire [51] . Furthermore, addition of equimolar dTTP to dCMP or dCTP activity ( ). While these data are reliably reproducible, the chemical complexity of cell lysates Millimolar concentrations of substrate are required in these reactions to produce a robust change in ABS 630 with the H 4 . By using purified proteins and more sensitive techniques to measure nucleotide substrates and products, such as thin - later chromatography [51] or LC - MS/MS [110] , detailed kinetic analysis and allosteric regulation can be determined in the absence of the lysate milieu. These experiments will he lp to clarify the precise function of the DCD domain in DcdV and aid in understanding its role in cell biology. 4.2.2.2: - - phosphate from an ATP donor to a deoxy/nucleotide monophosphate substrate with the help of a Mg 2+ cofactor, although many exceptions exist [86] . Despite showing no deaminase activity in cell l ysates ( ) the expression of the DCD domain variant DcdVE384A was still able to enhance the misincorporation of dU into the genome of E. coli ( ). Th is result suggests that the activity of the NK domain is likely to produce dUTP. In vitro experiments using purified DcdV and DcdV E384A in the presence of deoxy/nucleotide substrates and in vivo analysis of intracellular nucleotide concentrations using LC/ MS/MS will reveal whether the NK domain of DcdV is synthesizing dUTP. 4.2.2.3: While Correlogy predicted a shared gene network between dncV and dcdV there is currently no information about the conservation of difV or its association with either gene. As difV was previously unannotated, it was not included in the initial bioinformatic analysis and subsequent BLAST searches have revealed no homologs to sequences found within the orf1, where difV resides ( A) . Fur ther characterization of the precise location and coding sequence of difV will aid in the understanding of its conservation. Analysis of some dcdV homolog s has revealed a handful of case s where a small ORF dcdV homolog bu t both nucleotide and amino acid alignments have failed to reveal any conservation between them. Future experiments involving dcdV homologous will help to co mmon features and activities do these regulators share. 4.2.2.4: While much remains to be understood about the DcdV and DifV, a clear connection between these two gene products has been established. On the other hand, the connection between DncV and DcdV predicted by Correlogy remains mysterious. From the data presented in Chapter 3, it is clear that DcdV activity is unlikely to be allosterically activated by cGAMP. Additionally, DcdV and DncV are not encoded in a shared operon ( ). Both these features make DncV - DcdV unique from the CBASS CD - NTase effector relationships described to date [43, 46] , suggesting an alternative mechanism is responsible for the frequency with which they co - occur. DifV may be the key to bridging the divide between these two enzymes and further experiments will be performed to test the stabilit y and expression of DifV in the context of phage infection and cGAMP abundance. An alternative hypothesis for the predicted DcdV/DncV network could arise from the enzymatic activities that intersect at folate. The enzymatic activity of DncV is repressed b y certain folate species, including 5 - MTHF [23] . Two precursors to 5 - MTHF are 5,10 - MTHF and 10 - MTHF which are used for the denovo biosynthesis of purines and dTMP, respectively [50] . When active, DcdV is likely to alter deoxynucleotide pools which could lead to the consumption of 5,10 - MTHF and 10 - MTHF, which in turn depletes available 5 - MTHF. In this scenario, DcdV activity could se rve as a mechanism to de - repress DncV and induce activation of CBASS. Future studies measuring intracellular concentrations of nucleotides, cGAMP, and folates, using LC/MS - MS, will reveal if DncV and DcdV are linked through nucleotide metabolism. This regu latory mechanism could be easily deployed for the activation of other CBASS CD - NTases as a wide variety of nucleotide metabolizing enzymes could achieve similar results as DcdV. These experiments will help shed light on whether or not the primary role of V SP - 1 is to facilitate phage defense, and if not, what alternative benefits have been bestowed up on the El Tor biotype. 1. Ali M, Nelson AR, Lopez AL, Sack D A (2015) Updated Global Burden of Cholera in Endemic Countries. 1 13. doi: 10.1371/journal.pntd.0003832 2. Ryan ET (2011) The cholera pandemic, still with us after half a century: Time to rethink. PLoS Negl Trop Dis 5:1 2. doi: 10.1371/journal.pntd.00010 03 3. Hu D, Liu B, Feng L, et al (2016) Origins of the current seventh cholera pandemic. Proc Natl Acad Sci U S A 113:E7730 E7739. doi: 10.1073/pnas.1608732113 4. Dziejman M, Balon E, Boyd D, et al (2002) Comparative genomic analysis of Vibrio cholerae : Genes that correlate with cholera endemic and pandemic disease. Proc Natl Acad Sci U S A 99:1556 1561. doi: 10.1073/pnas.042667999 5. - II is a 26·9 kb genomic island present in Vibrio cholerae El Tor and O139 serogroup isolates that shows homology to a 43·4 kb genomic island in V. vulnificus. Microbiology 150:4053 4063. doi: 10.1099/mic.0.27172 - 0 6. Nguyen TH, Pham TD, Higa N, et al (2018) Analysis of Vibrio seventh pandemi c island II and novel genomic islands in relation to attachment sequences among a wide variety of Vibrio cholerae strains. 150 157. doi: 10.1111/1348 - 0421.12570 7. Taviani E, Grim CJ, Choi J, et al (2010) comparative genomic analysis. 16961:130 137. doi: 10.1111/j.1574 - 6968.2010.02008.x 8. Nusrin S, Gil AI, Bhuiyan NA, et al (2009) Peruvian Vibrio cholerae O1 El Tor strains possess a distinct region in the Vibrio seventh pandemic island - II that differentiates them from the prototype seventh pandemic El Tor strains. J Med Microbiol 58:342 354. doi: 10.1099/jmm.0.005397 - 0 9. 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. d oi: 10.1016/j.cell.2012.01.053 10. - di - GMP signaling. Curr Opin Microbiol 15:140 146. doi: 10.1016/j.mib.2011.12.008 11. Corrigan RM, Gründling A (2013) Cyclic di - AMP: Another second me ssenger enters the fray. Nat Rev Microbiol 11:513 524. doi: 10.1038/nrmicro3069 12. Jenal U, Reinders A, Lori C (2017) Cyclic di - GMP: Second messenger extraordinaire. Nat Rev Microbiol 15:271 284. doi: 10.1038/nrmicro.2016.190 13. Römling U, Galperin M Y, Gomelsky M (2013) Cyclic di - GMP: the first 25 years of a universal bacterial second messenger. Microbiol Mol Biol Rev 77:1 52. doi: 10.1128/MMBR.00043 - 12 14. Ross P, Weinhouse H, Aloni Y, et al (1987) Regulation of cellulose synthesis in Acetobacter x ylinum by cyclic diguanylic acid. Nature 325:279 81. 15. 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 Cel l 30:167 178. doi: 10.1016/j.molcel.2008.02.020 16. Kuchta K, Knizewski L, Wyrwicz LS, et al (2009) Comprehensive classification of nucleotidyltransferase fold proteins: Identification of novel families and their representatives in human. Nucleic Acids R es 37:7701 7714. doi: 10.1093/nar/gkp854 17. Hallberg ZF, Wang XC, Wright TA, et al (2016) Hybrid promiscuous (Hypr) GGDEF enzymes produce cyclic AMP - - cGAMP). Proc Natl Acad Sci 201515287. doi: 10.1073/pnas.1515287113 18. Severin GB, Ramliden MS, Hawver LA, et al (2018) Direct activ ation of a phospholipase by cyclic GMP - AMP in El Tor Vibrio cholerae. Proc Natl Acad Sci U S A 115:E6048 E6055. doi: 10.1073/pnas.1801233115 19. proteins for comparative genome analysis. Nucleic Acids Res 35:2006 2008. doi: 10.1093/nar/gkl949 20. Kazi MI, Conrado AR, Mey AR, et al (2016) ToxR Antagonizes H - NS Regulation of Horizontally Acquired Genes to Drive Host Colonization. PLoS Pathog 12:1 26. doi: 10.1371/journal. ppat.1005570 21. Van Kessel JC, Rutherford ST, Shao Y, et al (2013) Individual and combined roles of the master regulators apha and luxr in control of the Vibrio harveyi quorum - sensing regulon. J Bacteriol 195:436 443. doi: 10.1128/JB.01998 - 12 22. Li YH , Tian X (2012) Quorum sensing and bacterial social interactions in biofilms. Sensors 12:2519 2538. doi: 10.3390/s120302519 23. Zhu D, Wang L, Shang G, et al (2014) Structural Biochemistry of a Vibrio cholerae Dinucleotide Cyclase Reveals Cyclase Activit y Regulation by Folates. Mol Cell 55:931 937. doi: 10.1016/j.molcel.2014.08.001 24. Nelson JW, Sudarsan N, Phillips GE, et al (2015) Control of bacterial exoelectrogenesis by c - AMP - GMP. Proc Natl Acad Sci 112:5389 94. doi: 10.1073/pnas.1419264112 25. H allberg ZF, Chan CH, Wright TA, et al (2019) Structure and mechanism of a hypr GGDEF enzyme that activates cGAMP signaling to control extracellular metal respiration. Elife 8:1 36. doi: 10.7554/eLife.43959 26. Li F, Cimdins A, Rohde M, et al (2019) DncV Synthesizes Cyclic GMP - AMP and Regulates Biofilm Formation and Motility in Escherichia coli ECOR31. MBio 10:e02492 - 18. doi: 10.1128/mBio.02492 - 18 27. Burroughs AM, Zhang D, Schäffer DE, et al (2015) Comparative genomic analyses reveal a vast, novel netwo rk of nucleotide - centric systems in biological conflicts, immunity and signaling. Nucleic Acids Res 43:10633 10654. doi: 10.1093/nar/gkv1267 28. Whiteley AT, Eaglesham JB, de Oliveira Mann CC, et al (2019) Bacterial cGAS - like enzymes synthesize diverse n ucleotide signals. Nature 567:194 199. doi: 10.1038/s41586 - 019 - 0953 - 5 29. Nelson JW, Breaker RR (2017) The lost language of the RNA World. Sci Signal 10:1 11. doi: 10.1126/scisignal.aam8812 30. Margolis SR, Wilson SC, Vance RE (2017) Evolutionary Origi ns of cGAS - STING Signaling. Trends Immunol 38:733 743. doi: 10.1016/j.it.2017.03.004 31. Chen Q, Sun L, Chen ZJ (2016) Regulation and function of the cGAS - STING pathway of cytosolic DNA sensing. Nat Immunol 17:1142 1149. doi: 10.1038/ni.3558 32. Tan X, Sun L, Chen J, Chen ZJ (2018) Detection of Microbial Infections Through Innate Immune Sensing of Nucleic Acids. Annu Rev Microbiol 72:447 478. doi: 10.1146/annurev - micro - 102215 - 095605 33. Sun L, Wu J, Du F, et al (2013) Cyclic GMP - AMP Synthase Is a Cyto solic DNA Sensor That Activates the Type I Interferon Pathway. Science (80 - ) 339:786 791. doi: 10.1126/science.1232458 34. Gao D, Wu J, Wu YT, et al (2013) Cyclic GMP - AMP synthase is an innate immune sensor of HIV and other retroviruses. Science (80 - ) 341:903 906. doi: 10.1126/science.1240933 35. West AP, Khoury - Hanold W, Staron M, et al (2015) Mitochondrial DNA stress primes the antiviral innate immune response. Nature 520:553 557. doi: 10.1038/nature14156 36. Kranzusch PJ, Lee ASY, Wilson SC, et a l (2014) Structure - guided reprogramming of human cgas dinucleotide linkage specificity. Cell 158:1011 1021. doi: 10.1016/j.cell.2014.07.028 37. Zhang X, Shi H, Wu J, et al (2013) Cyclic GMP - AMP containing mixed Phosphodiester linkages is an endogenous hi gh - affinity ligand for STING. Mol Cell 51:226 235. doi: 10.1016/j.molcel.2013.05.022 38. Woodward JJ, Lavarone AT, Portnoy DA (2010) C - di - AMP secreted by intracellular Listeria monocytogenes activates a host type I interferon response. Science (80 - ) 328 :1703 1705. doi: 10.1126/science.1189801 39. Burdette DL, Monroe KM, Sotelo - Troha K, et al (2011) STING is a direct innate immune sensor of cyclic di - GMP. Nature 478:515 518. doi: 10.1038/nature10429 40. McFarland AP, Luo S, Ahmed - Qadri F, et al (2017) Sensing of Bacterial Cyclic Dinucleotides by the Oxidoreductase RECON Promotes NF - Proinflammatory Antibacterial State. Immunity 46:433 445. doi: 10.1016/j.immuni.2017.02.014 41. Gao P, Ascano M, Zillinger T, et al (2013) Struct ure - function analysis of STING 762. doi: 10.1016/j.cell.2013.07.023 42. McFarland AP, Burke TP, Carletti AA, et al (2018) RECON - dependent inflammation in hepatocytes enha nces listeria monocytogenes cell - to - cell spread. MBio 9:1 15. doi: 10.1128/mBio.00526 - 18 43. Cohen D, Melamed S, Millman A, et al (2019) Cyclic GMP AMP signalling protects bacteria against viral infection. Nature 574:691 695. doi: 10.1038/s41586 - 019 - 1605 - 5 44. Makarova KS, Wolf YI, Snir S, Koonin E V. (2011) Defense Islands in Bacterial and Archaeal Genomes and Prediction of Novel Defense Systems. J Bacteriol 193:6039 6056. doi: 10.1128/JB.05535 - 11 45. Schulman BA, Wade Harper J (2009) Ubiquitin - like protein activation by E1 enzymes: The apex for downstream signalling pathways. Nat Rev Mol Cell Biol 10:319 331. doi: 10.1038/nrm2673 46. Ye Q, Lau RK, Mathews IT, et al (2020) HORMA Domain Proteins and a Trip13 - like ATPase Regulate Bacterial cGAS - like E nzymes to Mediate Bacteriophage Immunity. Mol Cell 77:709 - 722.e7. doi: 10.1016/j.molcel.2019.12.009 47. Seed KD (2015) Battling Phages: How Bacteria Defend against Viral Attack. PLoS Pathog 11:1 5. doi: 10.1371/journal.ppat.1004847 48. Hampton HG, Wats on BNJ, Fineran PC (2020) The arms race between bacteria and their phage foes. Nature 577:327 336. doi: 10.1038/s41586 - 019 - 1894 - 8 49. Samson JE, Magadán AH, Sabri M, Moineau S (2013) Revenge of the phages: Defeating bacterial defences. Nat Rev Microbiol 11:675 687. doi: 10.1038/nrmicro3096 50. Rev 34:278 343. doi: 10.1128/mmbr.34.3.278 - 343.1970 51. Zhang Y, Maley F, Maley GF, et al (2007) Chloroviruses Encode a Bifunctio nal dCMP - dCTP Deaminase That Produces Two Key Intermediates in dTTP Formation. J Virol 81:7662 7671. doi: 10.1128/jvi.00186 - 07 52. Stavrou S, Ross SR (2015) APOBEC3 Proteins in Viral Immunity. J Immunol 195:4565 4570. doi: 10.4049/jimmunol.1501504 53. K oning FA, Newman ENC, Kim E - Y, et al (2009) Defining APOBEC3 Expression Patterns in Human Tissues and Hematopoietic Cell Subsets. J Virol 83:9474 9485. doi: 10.1128/jvi.01089 - 09 54. Son MS, Taylor RK (2011) Genetic Screens and Biochemical Assays to Chara cterize 17. doi: 10.1002/9780471729259.mc06a02s22 55. Beyhan S, Tischler AD, Camilli A, Yildiz FH (2006) Differences in gene expression between the classical and El Tor biotypes of Vibrio cholerae O1. Infect Immun 74:3633 3642. doi: 10.1128/IAI.01750 - 05 56. Taviani E, Grim CJ, Choi J, et al (2010) Discovery of novel Vibrio cholerae VSP - II genomic islands using comparative genomic analysis. FEMS Microbiol Lett 308:130 137. doi: 10.1111/j.1574 - 6968.201 0.02008.x 57. Doron S, Melamed S, Ofir G, et al (2018) Systematic discovery of antiphage defense systems in the microbial pangenome. Science (80 - ) 359:0 12. doi: 10.1126/science.aar4120 58. Johansson E, Fanø M, Bynck JH, et al (2005) Structures of dCT P deaminase from Escherichia coli with bound substrate and product: Reaction mechanism and determinants of mono - and bifunctionality for a family of enzymes. J Biol Chem 280:3051 3059. doi: 10.1074/jbc.M409534200 59. Vértessy BG, Tóth J (2009) Keeping ur acil out of DNA: physiological role, structure and catalytic mechanism of dUTPases. Acc Chem Res 42:97 106. doi: 10.1021/ar800114w 60. Graziani S, Xia Y, Gurnon JR, et al (2004) Functional analysis of FAD - dependent thymidylate synthase ThyX from Parameci um bursaria chlorella virus - 1. J Biol Chem 279:54340 54347. doi: 10.1074/jbc.M409121200 61. Zhang Y, Moriyama H, Homma K, Van Etten JL (2005) Chlorella Virus - Encoded Deoxyuridine Triphosphatases Exhibit Different Temperature Optima. J Virol 79:9945 9953. doi: 10.1128/jvi.79.15.9945 - 9953.2005 62. Chiu CS, Ruettinger T, Flanegan JB, Greenburg GR (1977) Role of deoxycytidylate deaminase in deoxyribonucleotide synthesis in bacteriophage T4 DNA replication. J Biol Chem 252:8603 8608. 63. Miller ES, Kutter E , Mosig G, et al (2003) Bacteriophage T4 Genome. Microbiol Mol Biol Rev 67:86 156. doi: 10.1128/mmbr.67.1.86 - 156.2003 64. Kim PJ, Price ND (2011) Genetic co - occurrence network across sequenced microbes. PLoS Comput Biol. doi: 10.1371/journal.pcbi.1002340 65. Batushansky A, Toubiana D, Fait A (2016) Correlation - Based Network Generation, Visualization, and Analysis as a Powerful Tool in Biological Studies: A Case Study in Cancer Cell Metabolism. Biomed Res Int. doi: 10.1155/2016/8313272 66. Skorupski K, Taylor RK (1996) Positive selection vectors for allelic exchange. Gene 169:47 52. doi: 10.1016/0378 - 1119(95)00793 - 8 67. Edelheit O, Hanukoglu A, Hanukoglu I (2009) Simple and efficient site - directed mutagenesis using two single - primer reactions in paral lel to generate mutants for protein structure - function studies. BMC Biotechnol 9:1 8. doi: 10.1186/1472 - 6750 - 9 - 61 68. Kunkel TA (1985) Rapid and efficient site - specific mutagenesis without phenotypic selection. Proc Natl Acad Sci U S A 82:488 492. doi: 1 0.1073/pnas.82.2.488 69. 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. doi: 10.1128/iai.6 4.7.2853 - 2856.1996 70. Dunn AK, Millikan DS, Adin DM, et al (2006) and pES213 - Derived Tools for Analyzing Symbiotic. Society 72:802 810. doi: 10.1128/AEM.72.1.802 71. Bose JL, Rosenberg CS, Stabb E V. (2008) Effects of luxCDABEG induction in Vibrio fis cheri: Enhancement of symbiotic colonization and conditional attenuation of growth in culture. Arch Microbiol 190:169 183. doi: 10.1007/s00203 - 008 - 0387 - 1 72. Fürste JP, Pansegrau W, Frank R, et al (1986) Molecular cloning of the plasmid RP4 primase regio n in a multi - host - range tacP expression vector. Gene 48:119 131. doi: 10.1016/0378 - 1119(86)90358 - 6 73. Vanbleu E, Marchal K, Vanderleyden J (2004) Genetic and physical map of the pLAFR1 vector. DNA Seq - J DNA Seq Mapp 15:225 227. doi: 10.1080/10425170410 001723949 74. Fernandez NL, Nhu NT, Hsueh BY, et al (2020) VIBRIO CHOLERAE ADAPTS TO SESSILE AND MOTILE LIFESTYLES BY CYCLIC DI - GMP REGULATION OF CELL SHAPE. bioRxiv 2020.04.15.043257. doi: 10.1101/2020.04.15.043257 75. Dong H, Liu Y, Zu X, et al (2015 ) An enzymatic assay for high - throughput screening of cytidine - producing microbial strains. PLoS One 10:1 14. doi: 10.1371/journal.pone.0121612 76. Ngo TT, Phan APH, Yam CF, Lenhoff HM (1982) Interference in Determination of Ammonia with the Hypochlorite - Alkaline Phenol Method of Berthelot. Anal Chem 54:46 49. doi: 10.1021/ac00238a015 77. Schmidt TT, Sharma S, Reyes GX, et al (2020) Inactivation of folylpolyglutamate synthetase Met7 results in genome instability driven by an increased dUTP/dTTP ratio. N ucleic Acids Res 48:264 277. doi: 10.1093/nar/gkz1006 78. Mandlik A, Livny J, Robins WP, et al (2011) RNA - seq - based monitoring of infection - linked changes in vibrio cholerae gene expression. Cell Host Microbe 10:165 174. doi: 10.1016/j.chom.2011.07.007 79. Krin E, Pierlé SA, Sismeiro O, et al (2018) Expansion of the SOS regulon of Vibrio cholerae through extensive transcriptome analysis and experimental validation. BMC Genomics 19:1 18. doi: 10.1186/s12864 - 018 - 4716 - 8 80. Papenfort K, Förstner KU, Cong J - P, et al (2015) Differential RNA - seq of Vibrio cholerae identifies the VqmR small RNA as a regulator of biofilm formation. Proc Natl Acad Sci 112:E766 E775. doi: 10.1073/pnas.1500203112 81. Raabe CA, Hoe CH, Randau G, et al (2011) The rocks and shallo ws of deep RNA sequencing: Examples in the Vibrio cholerae RNome. Rna 17:1357 1366. doi: 10.1261/rna.2682311 82. Liu JM, Livny J, Lawrence MS, et al (2009) Experimental discovery of sRNAs in Vibrio cholerae by direct cloning, 5S/ tRNA depletion and paral lel sequencing. Nucleic Acids Res 37:1 10. doi: 10.1093/nar/gkp080 83. Updegrove TB, Zhang A, Storz G (2016) Hfq: The flexible RNA matchmaker. Curr Opin Microbiol 30:133 138. doi: 10.1016/j.mib.2016.02.003 84. Lenz DH, Mok KC, Lilley BN, et al (2004) T he small RNA chaperone Hfq and multiple small RNAs control quorum sensing in Vibrio harveyi and Vibrio cholerae. Cell 118:69 82. doi: 10.1016/j.cell.2004.06.009 85. Kelley LA, Mezulis S, Yates CM, et al (2015) The Phyre2 web portal for protein modeling, prediction and analysis. Nat Protoc 10:845 858. doi: 10.1038/nprot.2015.053 86. Leipe DD, Koonin E V., Aravind L (2003) Evolution and classification of P - loop kinases and related proteins. J Mol Biol 333:781 815. doi: 10.1016/j.jmb.2003.08.040 87. Walk er JE, Saraste M, Runswick MJ, Gay NJ (1982) Distantly related sequences in the alpha - and beta - subunits of ATP synthase, myosin, kinases and other ATP - requiring enzymes and a common nucleotide binding fold. EMBO J 1:945 951. doi: 10.1002/j.1460 - 2075.1982. tb01276.x 88. Reizer J, Buskirk S, Reizer A, et al ubiquitous deaminase families. Protein Sci 3:853 856. doi: 10.1002/pro.5560030515 89. Sharma S, Patnaik SK, Thomas Taggart R, et al (2015) APOBEC3A cytidine deaminase induces RNA editing in monocytes and mac rophages. Nat Commun. doi: 10.1038/ncomms7881 90. Cohen R, Wolfenden R (1971) Cytidine deaminase from Escherichia coli. Purification, properties and inhibition by the potential transition state analog 3,4,5,6 - tetrahydrouridine. J Biol Chem 246:7561 5. 9 1. - Deoxycytidylate Deaminase and Its Complex with Substrate Analog and Allosteric Regulator dCTP·Mg2 +. J Mol Biol 377:220 231. doi: 10.1016/j.jmb.2007.12.064 92. Marx A, Alian A (2015) The First crystal structure of a dTTP - bound deoxycytidylate deaminase validates and details the allosteric - inhibitor binding site. J Biol Chem 290:682 690. doi: 10.1074/jbc.M114.617720 93. Krokan HE, Bjoras M (2013) Base Excision Repair. Cold Spring Harb Perspect Biol 5:a012583 a012583. doi: 10.1101/cshperspect.a012583 94. Tye BK, Chien J, Lehman IR, et al (1978) Uracil incorporation: A source of pulse - labelled DNA fragments in the replication of the Escherichia coli chromosome. Proc Nat l Acad Sci U S A 75:233 237. doi: 10.1073/pnas.75.1.233 95. Blower TR, Short FL, Rao F, et al (2012) Identification and classification of bacterial Type III toxin - antitoxin systems encoded in chromosomal and plasmid genomes. Nucleic Acids Res 40:6158 617 3. doi: 10.1093/nar/gks231 96. Kang SM, Kim DH, Jin C, Lee BJ (2018) A systematic overview of type II and III toxin - antitoxin systems with a focus on druggability. Toxins (Basel). doi: 10.3390/toxins10120515 97. Jamet A, Nassif X (2015) New Players in in. MBio 6:1 8. doi: 10.1128/mBio.00285 - 15.Copyright 98. Iyer LM, Zhang D, Rogozin IB, Aravind L (2011) Evolution of the deaminase fold and multiple origins of eukaryotic editing and mutagenic nucleic acid deam inases from bacterial toxin systems. Nucleic Acids Res 39:9473 9497. doi: 10.1093/nar/gkr691 99. Ahmad SI, Kirk SH, Eisenstark A (1998) Thymine Metabolism and Thymineless Death in Prokaryotes and Eukaryotes. Annu Rev Microbiol 52:591 625. doi: 10.1146/an nurev.micro.52.1.591 100. Guzmán EC, Martín CM (2015) Thymineless death, at the origin. Front Microbiol 6:1 7. doi: 10.3389/fmicb.2015.00499 101. Itsko M, Schaaper RM (2014) dGTP Starvation in Escherichia coli Provides New Insights into the Thymineless - Death Phenomenon. PLoS Genet. doi: 10.1371/journal.pgen.1004310 102. Schaaper RM, Mathews CK (2013) Mutational consequences of dNTP pool imbalances in E. coli. DNA Repair (Amst) 12:73 79. doi: 10.1016/j.dnarep.2012.10.011 103. Serrano - Heras G, Bravo A replication impairment caused by uracil excision activity of uracil - DNA glycosylase. Proc Natl Acad Sci U S A 105:19044 19049. doi: 10.1073/pnas.0808797105 104. Koestler BJ, Waters CM (2013) Explo ring environmental control of cyclic di - gmp signaling in vibrio cholerae by using the ex vivo lysate cyclic di - GMP assay(TELCA). Appl Environ Microbiol 79:5233 5241. doi: 10.1128/AEM.01596 - 13 105. Iyer LM, Burroughs AM, Aravind L (2006) The prokaryotic a ntecedents of the ubiquitin - signaling system and the early evolution of ubiquitin - - grasp domains. Genome Biol 7:1 23. doi: 10.1186/gb - 2006 - 7 - 7 - r60 106. Pickart CM, Eddins MJ (2004) Ubiquitin: Structures, functions, mechanisms. Biochim Biophys Acta - Mol Cell Res 1695:55 72. doi: 10.1016/j.bbamcr.2004.09.019 107. Streich FC, Lima CD (2018) Strategies to trap enzyme - substrate complexes that mimic michaelis intermediates during E3 - mediated ubiquitin - like protein ligation. Methods Mol Biol 1844:169 19 6. doi: 10.1007/978 - 1 - 4939 - 8706 - 1_12 108. Bard JAM, Martin A (2018) The Ubiquitin Proteasome System. Methods Mol Biol. doi: 10.1007/978 - 1 - 4939 - 8706 - 1 109. Saeed AFUH, Ruan X, Guan H, et al (2020) Regulation of cGAS - Mediated Immune Responses and Immunot herapy. Adv Sci. doi: 10.1002/advs.201902599 110. Chen P, Liu Z, Liu S, et al (2009) A LC - MS/MS Method for the Analysis of Intracellular Nucleoside Triphosphate Levels. Pharm Res 26:1504 1515. doi: 10.1007/s11095 - 009 - 9863 - 9