|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 THE RELATIONSHIP OF THE ViuB HYDROLASE TO IRON LIMI TED GROWTH AND VIRULENCE OF Vibrio vulnificus By RICK A. SWAIN A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIR EMENTS FOR THE DEGREE OF MASTERS OF SCIENCE UNIVERSITY OF FLORIDA 2011
2 2011 Rick A. Swain
3 To my Mom and Dad, who have never stopped support ing me through all my endeavors
4 ACKNOWLEDGMENTS I thank my family and friends who has made this p ossible throughout these years. Mom, Dad, Liz, and Sarah; you all have been my foundation and supported me regardless of what I went through and I am lucky to have you as family. To my friends and mentors, I could not be here without you. Marianne, G abe, D an, Danielle C, Mike, Katie, and Jorge. ; you have all been unbelievable friends and made my time in Florida even in the tough spots when it was difficult continue Melissa you took me under your wing and I will neve r forget all the guidance you gave me. You are the definition of a true friend and mentor. I thank Dr. Wright for being my advisor and giving me this opportunity to pursue this degree and expand my experiences To the rest of my committee, Dr. Schneider an d Dr. Gulig, I thank you for all the assistance in and out of the lab and pushing me to work my hardest. Finally, I would like to thank Dani. You were always there for me and never minded listening to the rants of a less than sane graduate student. I will always appreciate the foundation you were for me.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 12 2 LITERATURE REVIEW ................................ ................................ .......................... 14 Iron Response in Vibrio Vulnificus ................................ ................................ .......... 14 Iron and Virulence of V. vulnificus ................................ ................................ .... 14 Iron Acquisition Genes ................................ ................................ ..................... 15 Global Regulation of Iron Acquisition ................................ ................................ ...... 17 Bacterial Iron Regulation ................................ ................................ .................. 17 GacS/GacA Regulatory System ................................ ................................ ....... 18 GacS/GacA Regulation of Iron in Vibrios ................................ .......................... 18 Role of GacA/Gacs in Virulence of V. vulnificus ................................ ..................... 19 3 OBJECTIVES AND HYPOTHESIS ................................ ................................ ......... 21 4 MATERIALS AND METHOD S ................................ ................................ ................ 22 Bacterial Strains and Growth Conditions ................................ ................................ 22 Phenotypic Responses of V. vulnificus ................................ ................................ ... 23 Individual Gene Expression ................................ ................................ .................... 23 Sequence Comparisons of Iron Response Genes ................................ .................. 25 viuB Deletion Mutation and Complementation ................................ ........................ 25 Virulence Analysis ................................ ................................ ................................ .. 26 5 ROLE OF GACA IN IRON LIMITED GROWTH AND GENE EXPRESSION .......... 29 GacA Regulates Iron Limited Growth ................................ ................................ ..... 29 GacA Regulates Gene Expression of the Catechol Siderophore System ............... 31 Gene Expression of the Hydroxamate Siderophore System ................................ ... 32 6 COMPARATIVE GENETICS OF THE CATECHOL SIDEROPHORE SYSTEM ..... 35 Comparative Alignments of ViuB Sequences ................................ ......................... 35 Comparative Alignments of VuuA Sequences ................................ ........................ 36
6 Comparative Alignments of VenB ................................ ................................ ........... 40 Relationship of Sequence Types with Virulence ................................ ..................... 40 7 ROLE OF VIUB IN IRON ACQUISITION AND VIRULENCE ................................ .. 42 Role of ViuB under Iron Limiting Growth ................................ ................................ 42 Effect of viuB on Virulence in Mice ................................ ................................ ......... 43 Competi tive Virulence between Wild Type CMCP6 and the viuB Deletion Mutant ................................ ................................ ................................ .................. 44 Relationship of ViuB to Hydroxamate Gene Expression ................................ ......... 45 8 SUMMARY AND CONCLUSIONS ................................ ................................ .......... 48 The GacA Regulation of Iron Acquisition in V. vulnificus ................................ ........ 48 viuB and its Role in Iron Acquisition and Virulence ................................ ................. 49 APPENDIX : ADDITIONAL MATERIALS ................................ ................................ ....... 52 LIST OF REFERENCES ................................ ................................ ............................... 55 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 59
7 LIST OF TABLES Table page 4 1 V. vulnificus strains used in studies performed. ................................ .................. 22 4 2 Strains used for sequencing analysis. ................................ ................................ 27 6 1 Virulence comparison in the mouse model, compared to Type 1. Significant values indi cated in parentheses. ................................ ................................ ....... 41 7 1 Virulence comparison in the mouse model. A) Virulence assay in mice injected with iron. B) Virulence assay in mice without injected iron. .................. 43 7 2 Competitive virulence assay in the mouse model, using an mtlA deletion mutant as a surrogate for wild type strain. ................................ .......................... 44
8 LIST OF FIGURES Figure page 2 1 Iron acquisition gene arrangement in V. vulnificus ................................ ............. 15 2 2 GacS/GacA signal transduction pathway in proteobacteria ............................. 20 4 1 Primers for PCR and sequencing. Bold nucleotides indicate the USER specific sites for USER cloning. ................................ ................................ .......... 24 4 2 Diagram of viuB deletion mutation construction in V. vulnificus strain CMCP6 using USER cloning and chitin transformation. ................................ ................... 26 5 1 Growth yield under iron limiting conditions. A) Percent growth yield o f ................................ ................................ ... 30 5 2 Fold changes of catechol siderophore iron acquisition genes when the g acA deletion mutant is compared to wild type strain CMCP6 under iron replete and iron limiting conditions.. ................................ ................................ ............... 32 5 3 Fold changes of hydroxamate siderophore iron acquisition genes when the gacA deletion mutant is compared to wild type strain CMCP6 under iron replete and iron limiting conditions.. ................................ ................................ ... 33 6 1 Comparative alignments of deduced amino acids of ViuB (A), VuuA (B), and VenB (C) ................................ ................................ ................................ ............. 37 6 2 Deduced amino acid alignment of ViuB for several V. vulnificus strain s. ........... 38 6 3 Alignment of deduced amino acids of VuuA in V. vulnificus with corresponding region with high affinity to V. cholerae sequence highlighted.. .... 39 7 1 Percent growth yield under iron limitation cond Significant differences are noted for p<0.01 (*). ................................ ................. 42 7 2 Fold changes of hydroxamate siderophore iron acquisition genes when the viuB deletion mutant is compared to wild type strain CMCP6 under iron replete and iron limiting conditions.. ................................ ................................ ... 46 A 1 Role of GacA in the fold changes of gene transcripts in the catechol siderophore system under iron replete and iron limiting conditions .................... 52 A 2 Role of GacA in the fold changes of gene transcripts in the hydroxamate siderophore system under iron replete and iron limiting conditions. ................... 53
9 A 3 Role of ViuB in the fold changes of gene transcripts in the hydroxamate siderophore system under iron replete and iron limiting conditions. ................... 54
10 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science THE RELATIONSHIP OF THE ViuB HYDROLASE TO IRON LIMITED GROWTH AND VIRULENCE OF Vibrio vulnificus By Rick A. Swain December 2011 Chair: Anita C. Wright Major: Food Science and Human Nutrition Vibrio vulnificus is an opportunistic, ferrophilic human pathogen and the most common cause of reported, fatal seafood related bacterial infections in the U.S. This estu arine bacterium commonly infects individuals with iron overloading disorders, such as hemochromatosis. It was previously demonstrated that the GacS/GacA two component signal transduction system regulates iron acquisition and virulence in V. vulnificus ( Gauthier et al. 2010 ) M ultip le genes are responsible for iron acquisition and comprise the catechol and hydroxamate siderophore systems in V. vulnificus The catec hol siderophore system includes venB vuuA and viuB genes. The venB gene was previously shown to be required for virulen ce, and viuB was proposed as a virulence marker. D educed amino acid sequences were examined for all three genes, and the sequences of viuB segregated most clinical strains of V. vulnificus into a group that correlated with increased virulence in mice. Howe ver, no correlations were observed for venB or vuuA sequences Therefore, the specific role of the ViuB hydrolase in iron acquisition and virulence of V. vulnificus and its regulation by GacA, was examined. Mutational analysis confirmed GacA was required for increased expression of all three catechol genes under iron limiting conditions ; however, significant differences in gene
11 expression for the hydroxamate system were only observed under iron replete conditions. USER friendly cloning and chitin transform ation were used to construct a deletion mutation of viuB in order to determine its role in iron uptake and v irulence Compared to wild type, the viuB deletion mutant was significantly deficient in growth under iron limiting conditions However, virulence o f t he viuB mutant in mouse infections did not differ significantly from virulence observed in the wild type strain G ene expression of the hydroxamate system may compensate somewhat for loss of viuB function, as the viuB deletion mutant showed significant increases in transcript levels for two of three hydroxamate genes, as compared to the wild type In conclusion GacS/GacA differentially controls the expression of iron acquisition genes in the c atechol and hydroxamate systems of V. vulnificus Although all elic variation of the viuB gene correlated with virulence in mice and was required for growth under iron limiting conditions viuB does not appear to be a virulence factor for V. vulnificus
12 CHAPTER 1 INTRODUCTION Vibrio vulnificus is a n estuarine pathog en and the most common cause of fatal seafo od related bacterial illnesses in the United States ( CDC, 2008 ) This opportunistic bacterium is gram negative, halophilic, and causes systemic disease through consumption of raw oysters or through exposure o f wound infections to sea water ( Jones & Oliver, 2009 ) Previous findings showed a strong link between salinity and temperature on numbers of Vibrio vulnificus in oysters and sea water with the highest levels of this bacte rium in the warmer summer months ( CDC, 2008 ; Motes et al. 1998 ; Randa et al. 2004 ) O ptimal conditions for recovery of this bacterium ranged from 5 to 25 ppt for salinity and 13 to 22 o C for temperature ( Kaspar & Tamplin, 1993 ) Interestingly, this pathogen does not cause harm to the oysters, and the bivalve appears to be a primary ha bitat for V. vulnificus Oysters harvested from the Chesapeake Bay, showed approximately 80% of the mollusks contained the bacterium dependent on time of the year ( Wright et al. 1996 ) V. vulnificus disease is not spread by human to human transmission, and the bacterium is considered to be an opportunistic pa thogen in humans, as serious infections only occur in individuals with some type of underlying condition that compromises host defenses Transmission of disease through oyster consumption or wound infections can lead to septicemia and death in these indivi duals. Common conditions that are associated with increased proliferation of the pathogen in the host include hemochromatosis, diabetes, and alcoholic cirrhosis ( Blak e et al. 1979 ; Tacket et al. 1984 ) A common factor among these compromising conditions involves increased iron availability in the body ( Forouhi et al. 2007 ; Ganne Carrie et al. 2000 ;
13 Jehn et al. 2007 ) Hemochromatosis is a condition that is speci fically characterized by elevated serum iron levels: the hereditary form is primary hemochromatosis, and the develop mental form is termed secondary hemochromatosis ( Gan et al. 2011 ) Primary hemochromatosis is the most common form and involves mutation s in the HFE gene s wh ich when mutated cause the body to absorb excess iron This protein is a negative regulator that decreases the affinity of transferrin for free iron in the blood. Therefore, more iron is stored within the body as opposed to being lost through urine ( Gan et al. 2011 ) Iron is important to a variety of metabolic pathways for all aerobic bacteria. The correlation of hemochromatosis with V. vulnificus infections suggested an association between iron availability and the ability of this organism to cause disease. Furthermore, se veral studies experimentally demonstrated the important role of iron in pathogen i city and virulence of V. vulnificus in animal models ( Litwin et al. 1996 ; Okujo et al. 1996 ; Simpson & Oliver, 1983 ; Starks et al. 2000 ; Wright et al. 1981 ; Wright et al. 1986 ) The purpose of the study herein was to look at global regulation of iron acquisition and its relationship to the genes involved in the V v ulnif icus iron uptake siderophore systems in order to better understand their role in the virulence of V. vulnificus
14 CHAPTER 2 LITERATURE REVIEW Iron Response in Vibrio Vulnificus Iron and Virulence o f V v ulnificus Injection of exogenous iron during mouse infections cause s a large decrease in the 50% lethal dose of V. vulnificus indicating that iron is essential for virulence ( Wright et al. 1981 ) Additional research showed that the various iron acquisition genes contributed to the degree of virulence of V. vulnificus in mouse models. Mutation of venB produce d strains that were defective in tran s ferrin iron utilization and virulence and their phenotypes were r estored through complementation ( Litwin et al. 1996 ) In a more recent study ( Starks et al. 2000 ) the correlation between severity of disease in mice and the level of iron availability demonstrated that V. vulnificus w as a ferrophilic pathogen In this model of infection, mice were injected with iron dex t ran prior to subcutaneous inoculation with the pathogen, and both clinical and environmenta l isolates of V. vulnificus were compared for virulence potential The results showed that at higher inocula, the mice without iron dex t ran treatment typically lacked any symptoms of disease In fact, the mice required approximately one million fold increa se in inocula during the infection to replicate the same results as seen in the iron treated mice. Furthermore, t his observation was true for the majority of both cli nical and environmental strains and was attributed to the assumption that the iron injecti on compensates for any innate inability to proliferate within the host ( Starks et al. 2000 ) Not limited to these two examples, the contribution of iron acquisition on virulence of V. vulnificus in animal models has been studied and characterized in a variety of s tudies
15 ( Litwin et al. 1996 ; Okujo et al. 1996 ; Simpson & Oliver, 1983 ; Starks et al. 2000 ; Wright et al. 1981 ; Wright et al. 1986 ) Iron Acquisitio n Genes In V. vulnificus multiple genes are responsible for iron acquisition and these genes comprise both the catechol and hydroxamate siderophore systems. Siderophores are compounds that compete with t he host proteins for available free iron and are c apable of stripping iron away from the transferrins ( Kim et al. 2007b ; Litwin et al. 1996 ) The c atechol sider ophore iron acquisition system has been characterized in V. vulnificus and includes venB vuuA and viuB among other genes ( Webster & Litwin, 2000 ) The venB gene is responsible for the synthesis of the vulnibactin siderophore which is us ed in high affinity iron acquisition. The vuuA gene encodes a ferric vulnibactin receptor, and the expressed product is located on the outer membrane. In vestigations on a homologous system in Vibrio cholerae showed that once the siderophore iron complex is translocated through the inner membrane, the hydrolase product of the viuB gene cleaves the iron from the siderophore ( Butterton & Calderwood, 1994 ) These three iron acquisition genes are tightly linked on chromosome 2 in V. vulnificus as seen in Figure 2 1 (Crosa et al ., 2004). Also i ncluded in this gr ouping is the gene entE which encodes a protein responsible for peptide arylation Figure 2 1 Iron acquisition gene arrangement in V. vulnificus (Crosa, 2004)
16 The catechol iron acquisition genes are considered to be potenti al virulence factors because mutations in them often result in reduced virulence of the pathogen ( Litwin et al. 1996 ) S tudies have also examined viuB and suggested it may be a potential virulence marker, as original reports linked the presence of this gene with the clinical origin of the strain. Using a multiplex PCR assay for evaluation of clinical and environmental strains of V. vulnificus in shellfish the viuB gene was found in all clinical isolates, while only 24% of environmental strains were positive for the gene ( Panicker et al. 2004 ) In a later study, viuB positive clinical isolates survived longer in human serum than viuB negative strains (most environmental isolates) ( Bogard & Oliver, 2007 ) However it was later rev ealed that all strains of V. vulnificus actually contained the viuB gene, and the reported differences in absence or presence of the gene were the consequence of genetic polymorphisms at the primer sites used for PCR detection ( Bogard & Oliver, 2008 ) Other system s exist for iron acquisitio n in V. vulnificus but are much less characterized when compared to the catechol pathway and include hydroxamate ( Tanabe et al. 2005 ) and desferoxamine ( K im et al. 2007a ) systems The hydroxamate system is similar to the catechol system in that siderophores are released into the environment to bind iron with high affinity, but it differ s in that it uses a hydroxamic acid derived siderophore rather than a phenolate derived siderophore. In addition to this, they differ regarding he specific genes required for uptake and utilization of the iron. Meanwhile the desferoxamine system consists of just one gene, desA which encodes the receptor for the desferoxami ne iron complex; V. vulnificus can utilize the siderophore made from other sources, but does not synthesize it ( Kim et al. 2007a )
17 The hydroxamate siderophore system include s three genes The products of two of these genes are thought to be compo nents in the ABC transporter system used to release the aerobactin siderophore into the environment GenBank ID AAO07920.1 (ATP) is the ATPase component while AAO07921.2 (PP) is the periplasmic substrate binding protein. The characterized gene iutA (AAO07 924.2) is the aerobactin siderophore receptor. Global Regulation of Iron Acquisition Bacterial Iron Regu lation Iron related virulence factors are regulated by a combination of global regulatory systems that in turn are dependent upon the availability of i ron ( Lapouge et al. 2008 ) The f erric uptake regulation protein ( F ur) specifically manages i ron procurement through inhibition of iron acquisition genes under conditions of higher iron availability in V. vulnificus ( Litwin & Calderwood, 1993 ) During a test of global gene expression based on the status of iron, Alice et al. in 2008 showed that high levels of serum iron could indeed influence the expre ssion levels of many genes in V. vulnificus in mice. Based on their results, low iron concentration s influence d expression levels of genes required for general growth and survival, including outer membrane porins and enzymes for amino sugar biosynthesis. H owever, upon infection into a host, the bacterium causes disease through the expression of a variety of virulence factors including pili, RtxA1, and Group 1 capsular polysaccharide ( Alice et al. 2008 ; Liu et al. 2007 ) Although iron plays a part in virulence, it appears as though it is not the sole reactant in activation of virulence expression ( Alice et al. 2008 )
18 GacS /Gac A Regulatory System Another global regulatory system that has been related to iron acquisition in Vibrio fischeri is the GacS/GacA two component regulatory system ( Whistler & Ruby, 2003 ) GacA and GacS are common to all proteobacteria examined, including V. vulnificus ( Gauthier et al. 2010 ) GacS is the cognate sensor kinase and GacA is the response regulator ( W histler et al. 1998 ) The GacS/GacA regulatory system is known to act on a global scale within the cells through the small RNAs cs rB / c srC and also through the CsrA regulatory protein, as seen in Figure 2 2 (adapted from Lapouge et al. 2008. Gac/Rsm sign al transduction pathway of gamma proteobacteria: from RNA recognition to regulation of social behaviour. Molecular microbiology 67 241 253). Both csrB and csrC bind to the small regulatory protein CsrA, and thus inhibiting its function as a transcription factor, which in turn regulates various behaviors including biofilm, quorum sensing, virulence, chemotaxis, and motility in other Vibrio species ( Lapouge et al. 2 008 ; Whistler & Ruby, 2003 ) Gac S /Gac A Regulation of Iron i n Vibrios T he GacA/GacS pathway has been studied in relationship to iro n acquisition in V. fischeri ( Whistler & Ruby, 2003 ) In this s tudy siderophore production in gacA deletion mutants was examin ed using low iron media Mutants lacking GacA expression were defective in iron acquisition that required siderophores, and this phenotype was restored through complementation ( Whistler & Ruby, 2003 ) This result demonstrated that a closely related species V. fischeri regulates catechol siderophore iron acquis ition via the GacS/GacA two component system and suggested that a similar mechanism could potentially be present in V. vulnificus
19 Role of GacA/ Gacs in Virulence of V v ulnificus In a recent study ( Gauthier et al. 2010 ) transcript levels of various genes were measured in a gacA de letion mutant, and compared back to expression in the wild type strain Results showed that gacA deletion mutant s of V. vulnificus CMCP6 had decreased levels of vario us transcripts compared to wild type. T hese genes include: csrB1 csrB2 csrB3 csrC flaA rpoS and vvpE Altered phenotypes in the mutant included protease and cytotoxin expression, phase variation of capsular polysaccharide, biofilm formation, and virulence. The vvpE gene encodes protease activity in V. vulnificus and the gacA deletion mut ant showed significantly decreased levels of this activity and vvpE transcripton which was complemented in trans The authors concluded that GacA regulates protease expression in V. vulnificus Cytoxicity related to the RTX hemolysin was tested based on t he ability of V. vulnificus to destroy INT 407 monolayers in vitro The mutant showed decreased destruction of INT 407 cells in comparison to wild type ; however, expression of the corresponding rtxA gene was not significantly changed in mutant compared to the wild type strain Virulence in a mouse model was tested with and without the injection of exogenous iron The results showed iron overloaded mice exhibited no difference between V. vulnificus CMCP6 wild type strain and mutant for virulence. However wit hout iron loading conditions, the CMCP6 mutant was significant ly decrease d for virulence compared to the wild type strain These results were not replicated in V. vulnificus MO6 24, and demonstrate d that although gacA may contribute to virulence its role is dependent upon strain and iron availability. In summary, our current knowledge of the relationship of GacA to the V. vulnificus iron response includes the following associations: 1) t he ability to acquire iron is
20 required for virulence in V. vulnificus ( Starks et al ., 2000 ; Wright et al ., 1981 ) ; 2) mutation of the iron acquisition gene venB decreased virulence in infant mice compared to wild type response ( Litwin et al. 1996 ) ; 3) differences in the viuB alleles correlate with clinical origin in some strains ( Bogard & Oliver, 2008 ) ; 4) mutations in the global regulator GacA decrease virulence and iron acquisition in V. vulnificus ( Gauthier et al. 2010 ) ; and 5) iron acquisition in V. fischeri is regulated by GacA ( Whistler & Ruby, 2003 ) Therefore, I hypothesize it is likely that GacA is an essential component for iron regulation in V. vulnificus and that the role of GacA in virulence is related to its regulation of iron uptake. Figure 2 2. GacS/GacA signal transduction pathway in proteobacteria ( Lapouge et al. 2008 )
21 CHAPTER 3 OBJECTIVES AND HYPOT HESIS The overall objec tive of this research was to study the relationship of GacA and the i ron response of V. vulnificus by determining the effect o f a gacA mutation on expre ssion of specific iron acquisition genes and relating any observed effects to the virulence of this bac terium The gacA deletion mutant s were previously shown to be more sensitive to iron limiting growth conditions and less virulent than wild type strains in mice that were not iron loaded ( Gauthier et al. 2010 ) The role of GacA on the expression profile of genes related to iron upt ake in V. vulnificus was examined in the present study during exposure to vario us in vitro conditions with or without iron limitation F urther more, the relationship of GacA and iron was examined by looking at the function of the GacA regulated viuB gene in iron uptake and virulence of V. vulnificus The hypotheses for this project include the following : GacA regulates growth under conditions of iron limitation ( Gauthier et al. 2010 ) : therefore, GacA may control expression of genes ( venB, vuuA, viuB ) related to catechol siderophore u ptake. Genetic polymorphisms in V. vulnificus viuB correlated with clinical origin and increased virulence, suggesting it may be a virulence marker. Similar profiles may be common to other genes in the catechol siderophore system. If viuB is a virulence f actor deletion of this gene should alter the iron response and decrease virulence in the mutant when compared to the wild type strain. V. vulnificus iron response also involves the hydroxamate siderophore system, and the se genes involved (ATP, PP, iutA ) m ay also be regulated by GacA or could compensat e for loss of viuB
22 CHAPTER 4 MATERIALS AND METHOD S Bacterial Strains and Growth Conditions Strains of V. vulnificus described in Table 4 1, were stored as frozen stocks ( 70 o C) in Luria Burtani Broth with NaCl (LBN) and 50% glycerol and streaked onto LB N Agar (LA) for isolation overnight at 37 o C. LB N at pH of 7.5 was used for the iron replete conditions because it contains iron from the yeast extract (1%) and tryptone (1 %) For iron limiting conditions, the iron chelator dipyridyl (Acros Organics) was added to LB N at concentrations ranging from 100 to Table 4 1. V. vulnificus strains used in studies performed Strains Used Description Reference MO6 24/0 Clini cal, virulent wild type strain Wright et al. 2001 MO6 24/0 gacA deletion mutant with kanamycin inserted Gauthier et al. 2010 MO6 24/0 (pGacA) gacA mutant with complemented vector containing gacA in trans Gauthier et al. 2010 MO6 24/0 A (pGTR1160) gacA mutant with empty complemented vector Gauthier et al. 2010 CMCP6/0 Clinical, virulent wild type strain Kim et al. 2003 CMCP6/0 gacA deletion mutant with kanamycin inserted Gauthier et al. 2010 CMCP6/0 (pGacA) gacA mutant with complemented vector containing gacA in trans Gauthier et al. 2010 CMCP6/0 (pGTR1160) gacA mutant with empty complemented vector Gauthier et al. 2010 viuB viuB deletion mutant with kanamycin inserted This study viuB (pViuB ) viuB mutant with complemented vector containing viuB in trans This study viuB (pGTR1160) viuB mutant with empty complemented vector This study CMCP6/0 mtlA deletion mutant with kanamycin inserted Donated by Dr. Paul Gulig, University of F lorida mtlA (pMtlA) mtlA mutant with complemented vector containing mtlA in trans Donated by Dr. Paul Gulig, University of Florida mtlA (pGTR1160) mtlA mutant with empty complemented vector Donated by Dr. Paul Gulig, University of Florid a
23 Phenotypic Responses of V. vulnificus Individual colonies were inoculated into 250 mL flasks containing 35 mL of LB N The flasks were incubated at 80 rpm at 37 o C overnight in a C24 Incubator Shaker (New Brunswick Scientific). The absorbance at 600nm ( OD 600 ) was recorded at 24 hours using the SPECTRAmax PLUS 384 (Molecular Devices), and was adjusted to an A 600 of 0.30 in 1 mL PBS. Cells were then washed 2 times using LB N at 8,000xg, for 90 seconds with the Centrifuge 5415D from Eppendorf. The final p ellet was suspended in 1 mL LB N Washed V. vulnificus cells (710 ) w ere added to each 250 mL flask containing 35 mL LB N to achieve an inoculum of ap p roximately 2.0 x 10 6 CFU/ mL All flasks were incubated overnight at 37 o C, shaking at 80 rpm to achieve stat ionary phase After 18 24 hours, the absorbance at 600nm was recorded. Three biological replicates of each variant were used for the experiment, with one technical replicate for each sample Percent growth yield was calculated using the equation: % growth yield = ( OD 600 in L B:150 OD 600 in L B ) x 100 Individual Gene Expression Quantitative reverse transcription PCR ( q RT PCR ) was the primary method used to examine gene expression of viuB venB and vuuA catechol siderophore genes and ATP, PP (both previously described and de signated) and iutA hydroxamate genes in both the gacA and viuB deletion mutants compared to wild type V. vulnificus strain CMCP6 V ariations of CMCP6 described in Table 4 1 were grown in LB N under iron replete conditions and iron limiting conditions used dipyridyl at concentration s of 150 V. vulnificus caused a large inhibitory effect on the gacA deletion mutant s, but allowed the wild type to grow normally. Following overnight growth, RNA was extracted as previously
24 des cribed ( Gauthier et al. 2010 ) After isolation, the RNA was quantified using a Gene Spec and under went reverse transcription through the Invitrogen cDNA synthesis kit to a final amount of For qRT PCR, a 1/20 dilution of the cDNA was used in the Express SYBR Green qRT PCR k it by Invitrogen with primers ( Figure 4 1 ) designed through primer 3 software The thermocycler used for the Q PCR was the Smart Cycler II from Cepheid. The fold differences in transcript levels were calculate d from Ct values using the Ct method with 16S as the reference gene ( Livak & Schmittgen, 2001 ) Figure 4 1 Primers for PCR and sequencing Bold nucleotides indicate th e USER specific sites for USER cloning.
25 Sequence Comparisons of Iron Response Genes DNA from c linical and environmental strains of V. vulnificus (Table 4 2 ) w as used to sequence the venB vuuA and viuB genes P rimers ( Figure 4 1 ) were derived using the pri mer 3 software and the GenBank sequences for V. vulnificus CMCP6, YJ016 and available MO6 24/0 sequences provided by Paul Gulig. Primers were selected from sequence s that were conserved among all strains The isolation of the amplified DNA was performed us ing a PCR Clean Up kit ( Qiagen ) ICBR of the University of Florida provided sequencing S equence s were aligned using the MEGA program version 4 to generate a dendrogram to identify phlogroupings ( Tamura et al. 2007 ) viuB Deletion Mutation and Complementation To perform the deletion mutation of viuB in V. vulnificus strain CMCP6 (Type 1) USER friendly cloning and chitin transformation w ere used to remove viuB and substitute the gene for kanamycin resistance, according to the protocol of Gulig et al (2010). 500 base pair regions upstream and downstream of viuB were PCR amplified Figure 4 1 These PCR products were combined and directionally cloned into the USER ready broad host range vector pGTR1129. Included in the upstream and downstream primers was a SmaI site such that the kanamycin resistance marker could be inserted into the joined regions. This construct was electroporated into E scherichia coli EC100D to screen for successful ligations. The extracted pla smid was then digested with SmaI to open the vector. A kanamycin resistance cassette was then ligated into the USER vector and electroporated into E. coli again. The plasmid was extracted and 2 g of the vector was sheared and added to V. vulnificus CMCP6. This mixture was contained in a 12 well plate with cleaned crab shells in each well. After 24 hours the bacteria were
26 plated and screened for kanamycin resistance. PCR confirmation was used to verify the insertion along with sequence analysis. The complemented mutant (pViuB) was created by PCR amplifying the viuB gene from V. vulnificus CMCP6 with USER end primers and then cloned in trans using pGTR11 60, a USER end modified vector containing a m aker for tetracycline resistance. C onjugation was then used to transfer the plasmid from E. coli S17 to a V. vulnificus CMCP6 viuB deletion mutant. Meanwhile the plasmid control was the empty vector pGTR11 60 conjugated into the viuB mutant. Figure 4 2 Diagram of viuB deletion muta t ion construction in V. vulnificus strain CMCP6 using USER cloning and chitin transformation. Virulence Analysis V. vulnificus strains were tested for virulence potential using a previous ly described mouse model ( Starks et al. 2000 ) Female mice ( n =5) were injected sub cutaneously with 0.1 mL of bacteria in PBS This process used approximately 10 3 colony forming units (cfu) for iron treated mice and 10 6 cfu for mice without added iron. To iron tre at the mice, 250 g of iron dextran per gram of body weight was injected into the mice. Mice were euthanized using CO 2 asphyxiation and a rectal temperature of
27 Table 4 2. Strains used for sequencing analysis. 33 o C was the surrogate for death. Skin tissue at lesion si te and liver samples were excised, homogenized and diluted in PBS, and plated on LA Bacterial coun ts were calculated as log cfu per gram. In addition to comparison of virulence assays for individual strains competitive virulence assays were performed in the mouse model as well. An mtlA deletion mutant defective in mannitol utilization was used as a surrogate for wild type C o inoculation
28 studies were performed following the same procedure as before ; h owever, the mice were infected with twice the amount o f bacteria, using 10 6 of each strain per mouse. Skin tissue and liver samples were prepared as before but were plated onto LA supplemented with mannitol and phenol red as a pH indicator. On the se plates viuB mutants are yellow colonies due to their ability to ferment the mannitol, which causes a pH change and a corresponding color change from the phenol red. Meanwhile wild type surrogate mtlA mutants are pink due to their inability to utilize the mannitol as an energy source.
29 CHAPTER 5 R OLE OF GACA IN IRO N LIMITED GROWTH AND G ENE EXPRESSION GacA Regulates Iron Limited Growth As a first step in establishing a link between GacA and the phenotypic response to iron limitation V. vulnificus was grown under of iron limit ed and iron replete conditions Since LB N medi um contains iron, both wild type and gacA deletion mutant strains thrive as neither has to scavenge for available iron. However, once iron limiting conditions are introduced by the addition of dipyridyl the gacA deletion mutant was limited in growth compared to wild type strain indicating that the iron acquisition is regulated in part by GacA ( Gauthier et al. 2010 ) Mutants in two strains of V. vulnificus showed decreased growth under iron limiting conditions while under iron replete conditions growth yields of all strains were approximately equal (Figure 5 1). T riplicate experiments showed wild type strains had significantly greater growth yields under iron limitation (84.1% and 78.6 % in CMCP6 and MO6, respectively ) compared to the corresponding gacA deletion mutant s (8.7% and 14.6% in CMCP6 and MO6, respectively with p values 0.002 and 0.0002 respectively ). Thus, iron limitation imposed by application of a chelating agent demonstrated that the mutants were deficient in their ability to acquire iron for survival compared to wild type These result s support H ypothesis 1 that GacA regulates the phenotypic response to iron limitation in V. vulnificus The phenotype of m utant strains was restored when the gacA gene was introduced in trans on a plasmid vector (pGacA) and showed activity comparable to wild type. However, the vector (pGTR1129) controls showed conflicting results between the two strains. For CMCP6 (pGacA) growth yield was similar to the gacA deletion mutant while MO6 24 (pGacA) growth yield was more sim ilar to the wild type
30 strain. C omplement ation of the mutant strain should restore the phenotype, but the plasmid control should not ( Falkow, 198 8 ) Ultimately, t hese results support to the conclusion that GacA regulates growth of V. vulnificus under conditions of iron limitation, but vector effect s cannot be explained for the one strain A. B. Figure 5 1. Growth yield under iron limiting conditions. A) Percent growth yield of Significant differences are noted for p<0.001 ( *). Percent growth yield was calculated using the equation: % growth yield = (OD 600 600 in LB) x 100.
31 GacA Regulates Gene Expression o f t he Catechol Siderophore System The role of GacA was examined for its influence on the expression o f various genes related to iron acquisition. V. vulnificus CMCP6 Wild type strain, the gacA deletion mutant the complemented mutant, and plasmid control strain ( Table 4 1 ) were compared for gene expression by quantitative reverse transcription PCR (qRT PC R) in order to determine differences in gene expression for viuB venB and vuuA genes unde r iron replete (LBN) and iron limiting conditions (LBN with dipyridyl was added to a final concentration of 150 M) These catechol siderophore related genes were cho sen because they are important iron acquisition which is also important for virulence The fold changes for these genes were compared in the gacA deletion mutant relative to wild type expression Under iron replete conditions the gacA deletion mutant sho wed slightly increased transcript levels compared to wild type (Figure 5 2) while the complement ed mutant and plasmid control had decreased transcript levels under iron replete conditions (Appendix A Figure A 1) However, none of the genes demonstrated si gnificant changes in transcript levels for gacA deletion mutant vs. wild type strain Conversely, u nder iron limiting conditions, t he gacA mutant and the plasmid control showed significant decreases in gene expression compared to wild type strain for all t hree genes examined ( Figure 5 2, p values<0.001). Complementation restored viuB expression to levels that did not differ significantly from wild type (p=0.05). Although the complementation of mutations for venB, vuuA ( Appendix A, Figure A 1 ) still showed s ignificant decreases in transcript levels (p values 0.004 and 0.002 respectively), the phenotype was somewhat restored, as difference s were greatly reduced.
32 These results confirm the role of GacA regulation in the iron response of V. vulnificus under iron limiting conditions and are consistent with results reported for the related species of V. fischeri, whereby GacA regulates iron acquisition ( Whistler & Ruby, 2003 ) Based on these results we conclude that GacA contributes to regulation of iron acquisition genes in the catechol siderophore system in V. vulnific us Figure 5 2. Fold changes of catechol siderophore iron acquisition genes when the gacA deletion mutant is compared to wild type strain CMCP6 under iron replete and iron limiting conditions. Results are the average of 3 independent experiments with at least 2 technical replicates for each experiment. ( P values are indicated compared to wild type ) Gene Expression of t he Hydroxamate Siderophore System V. vulnificus has the ability to produce and utilize more than one type of siderophore. Therefore, it was hypothesized that gacA may also play a role in regulation of the hydroxamate system, and the transcript levels of several genes (ATP, PP, iutA ) related to the hydroxamate siderophore system were also analyzed. As seen in Figure 5 3, when compared to the wild type under iron replete conditions the gacA deletion mutant exhibited significant fold increases for ATP, PP, and to a lesser extent iutA (p values 0.001, 0.00003, and 0.04 respectively) When gacA was introduced in t rans the same genes showed significant fold decreases compared (0.001) (0.00002) (0.0008)
33 to wild type ( Appendix A, Figure A 2) ; p values 0.0003, 0.000004, and 0.000002 ); however, the plasmid control did not significantly differ from wild type. Under iron limiting conditions (Figu re 5 3), the gacA deletion mutant showed somewhat increased expression for ATP, PP, and iutA when compared to wild type (1.07, 1.59, and 1.34 fold, respectively), but differences were not significant. Interestingly, the complemented strain (3.00, 1.36, and 2.71 respectively) and the plasmid control ( 4.49 1.07, and 2.12 ) both exhibited decreased expression in the mutant compared to wild type, and significant difference were noted ( boldened ). Figure 5 3. Fold changes of hydro xamate siderophore iron acquisition genes when the gacA deletion mutant is compared to wild type strain CMCP6 under iron replete and iron limiting conditions. Results are the average of 3 independent experiments with at least 2 technical replicates for eac h experiment. (Significant differences are shown). The results for hydroxamate gene expression are difficult to interpret, but they indicate that G acA does not play a role in regulating the hydroxamate siderophore system under iron limiting conditions. Ins tead, the gacA deletion mutant had significant increases for two genes, ATP and iutA under iron replete conditions, possibly through de repression in the absence of the gacA gene expression. Clearly, GacA provides differential regulation for the two high a ffinity iron acquisition systems examined in V. (0.001) (0.00003)
34 vulnificus and more research will be required to sort out these pathways and the conditions that influence their regulation.
35 CHAPTER 6 COMPARATIVE GENETICS OF THE CATECHOL SIDE ROPHORE SYSTEM Comparative A l ignments of V iuB Sequences Prior research ( Bogard & Oliver, 2007 ; Panicker et al. 2004 ) showed the correlation of a particular viuB sequence with all strains from clinical origin while most oyster isolates differed from the c linically associated sequences and were negative for PCR assays for this gene These prior publications initially reported that only clinical isolates contained the gene, and concluded that viuB should be considered as a marker for virulence of V. vulnific us However, both groups later published errata showing their data w ere incomplete, and DNA sequence comparisons instead identified the sequence in all strai ns but with polymorphic sites. Thus, sequence differences were attributed as the reason that the or iginal PCR screening of the gene failed to detect its presence in environmental strains. These studies described above examined complete sequence from only a limited number of strains. To further investigate H ypothesis 2, that sequence differences related to genes for iron acquisition could predict virulence, viuB genes from 35 additional strains (Table 4 3), as well as other genes ( venB and vu uA ) related to the catechol siderophore system, were sequenced and their deduced amino acids were used to create phylogenetic trees. S equenc e analysis of viuB genes from the 35 V. vulnificus strains is shown in Figure 6 1A and segregated ViuB sequences into two phylogenetic types (phylotypes). ViuB Type 1 contained 80% of the clinical strains examined, while Type 2 was found in 93% of the environmental isolates examined. These data confirmed differences at the amino acid level and demonstrated divergent phylogenetic groups with excellent correlation between strain source and phylotype.
36 Alignments of ViuB identified 1 4 polymorphic amino acids sites (Figure 6 2) including sites with acidic residues aspartic (D) and glutamic acid s (E) that were found in Type1 strains CMCP6 and YJ105, but were substituted with glycine (G, neutral) residues in Type 2 strains 109, 141, B1, and 108. These acidic residues have been shown to contribute to iron binding in other related proteins ( Bailey et al. 1988 ) We hypothesize that these amino acid substitutions may alter the iron binding capacity and hence the function of ViuB, and future studies are planned. Comparativ e A lignments of VuuA Sequences The vuu A gene, encoding the outer membrane receptor for the vulnibactin iron complex, was also sequenced in multiple strains from various sources, and the deduced amino acid sequences were aligned. The corresponding phylogene tic tree is shown in Figure 6 1 B. Unlike ViuB, VuuA sequences did not segregate into two phylogroups that correlated with the strain source. Instead a multi branching tree with much greater distances (0.1 scale compared to 0.01 for ViuB) between branches w as observed, indicating greater diversity for this protein compared to ViuB. Interestingly, the V. cholera e sequence did not root this tree as seen with ViuB; instead, V. parahaemolyticus VuuA sequence was used as a root, although some V. vulnificus sequen ces showed greater similarity to those of V. parahaemolyticus compared to other sequences of V. vulnificus. The reason that V. cholerae sequence did not root the tree for V. vulnificus VuuA sequences was due to regions with highly conserved sequences betwe en the two species for this protein (Figure 6 3) The alignments showed a 320 amino acid insert that had 59% agreement with sequences from V. cholerae
37 A. B. C. Figure 6 1 Comparative alignments of deduced amino acids of ViuB (A ), VuuA (B), and VenB (C) using V. cholerae as a root for all three, and using V. parahaemolyticus as a root for VuuA. ViuB Type s are indicated by boxes. 1 2
38 Figure 6 2 Deduced amino acid alignment of ViuB for several V. vulnificus strains CMCP6 and YJ106 are Type 1 while 109, 141, B1, and 108 are Type 2. Three of the changes to acidi c residues a re noted in boxes.
39 Figure 6 3 Alignment of deduced amino acids of VuuA in V. vulnificus with corresponding region with high affinity to V. cholera e sequence highlighted. n amino acid with highly represents a change to an amino acid with highly dissimilar properties
40 Conversely, this region has the highest degree of va riability among some V. vulnificus strains ( ~ 68% agreement) when compared to the flanking regions ( ~98 % for upstream region and ~90% for downstream region ), indicating the insert was a recent acquisition. Comparative Alignments of VenB The final gene sequ enced was venB encoding an enzyme required for synthesis of the vulnibactin siderophore. Deduced amino acid sequences were aligned into a phylogenetic tree ( Figure 6 1 C). Approximately 93% of all environmental strains examined showed sequence that cluster ed into single branch with <0.01 divergence, while 81% the clinical isolates were distributed across other branches. Although the distinction between strains from different sources was not as clear as that seen with ViuB, these results also suggested a clo ser relationship of strains from clinical origin as compared to environmental isolates Relationship of Sequence Types with Virulence Litwin et al demonstrated that a V. vulnificus venB deletion mutant had reduced virulence compared to wild type in the i nfant mouse model (1996). As iron acquisition contribute s to virulence of V. vulnificus, the different phylotypes of the genes described above were examined for their relative virulence potential in a mouse model of disease ( Thiaville et al. 2011 ) It was shown that s trains with in ViuB Type 1 demonstrated increased virulence in mice when compared to strains without that sequence (p<0.01 ). As shown in Figure 6 4 significant differences were observed for the log CFU/g in the liver and for temperature (p values 0.008 and 0.004 respectively ( adapted by Thiaville et al. 2 011 Genotype is correlated with but does not predict virulence of Vibrio vulnificus biotype 1 in subcutaneously inoculated, iron dextran treated mice. Infection and
41 immunity 79 1194 1207 ) ) However, similar correlations were not observed for any other phylotypes and strain origin for either VuuA or VenB deduced amino acid sequences. Table 6 1. Virulence comparison in the mouse model, compared to Type 1. Significant values indicated in parentheses. Represents 1 experiment, with each strain inoculated int o mice (n=5) ( Thiaville et al. 2011 )
42 CHAPTER 7 ROLE OF VIUB IN IRON ACQUISITION AND VIRU LENCE Role of ViuB under Iron Limiting Growth Alt hough viuB is part of a cluster of genes that has been demonstrated to play a role in the acquisition of iron ( Litwin et al. 1996 ) the specific contribution of this gene to iron uptake has not been previously demonstrated. Therefore, a deletion mutant in V. vulnificus strain CMCP6 was constructed to determine the role of this gene in growth under iron limiting conditions. As seen in Figure 7 1, the viuB mutant and plasmid control had significantly lower % growth yields (17.8% and 18.1% respectively, with p values<0.0001) when compared to the wild type and complemented mutant (90.2% and 97.9% respectively). These results verified that viuB is required for growth under iron limiting conditions in V. vulnificus and that the phenotype could be complemented in trans Results are consistent with prior reports that identified viuB as a gene in the catechol siderophore pathway in related bacteria ( Butterton & Calderwood, 1994 ) Figure 7 1. Percent growth yield under iron limitation conditions at 1 3 Significant differences are noted for p<0. 01 (*). Percent growth yield was calculat ed using the equation: % growth yield = (OD 600 OD 600 in LB) x 100
43 Effect of viuB on V irulence in M ice The viuB deletion mutant of V. vulnificus was also used to study the role of this gene in virulence in mice. Using a previously describe d mouse model ( Starks et al. 2000 ) but with or without exogenous iron injections wild type strain CMCP6 and the viuB mutant were infected subcutaneously and the CFU/g in tissues that were harvested after 18 hours was compared in order to determines the relative levels of virulence between these strains. As seen in Table 7 1, t wo separate conditions were tested: mice (n=5) were inoculated A) with or B) without pretreatment with injected iron dextran. The results showed that n either the iron status of the host nor the absence of the viuB gene in V. vulnificus significantly a ltered the virulence of V. vulnificus Conditions in mice without iron pretreatment are presumed to be iron limiting as the untreated host will sequester iron in the form of transferrin and other iron binding compounds. This condition was where differences in virulence between mutant and wild type were expected, as the gacA mutant showed reduced virulence compared to wild type under similar conditions. Therefore, the no iron added condition was repeated in two additional experiments, and similar results wer e obtained. The experiment with added iron was only tested once, but no significant difference was seen between viuB mutant and wild type virulence under these conditions Table 7 1 Virulence comparison in the mouse model. A) Virulence assay in mice injec ted with iron. B) Virulence assay in mice without injected iron. Represents 1 experiment, but similar results were obtained on two additional experiments A.
44 B. Competit ive Virulence between Wild Type CMCP6 and the viuB Deletion Mutant In order to evalua te further the contribution of viuB to virulence, virulence comparisons were performed using a competitive in vivo analys is. In these competitive assays a V. vulnificus mutant defective in mannitol utilization ( mtlA ),donated by the lab of Dr. Paul Gulig, was used as a surrogate for wild type strain CMCP6 in order to mark the strain and easily distinguish it from the viuB mutant variant of the same strain (CMCP6) on the same medium. Equal inocula were prepared for each strain (approximately 1 x 10 6 CFU for each mutant in 100 ) and injected sub cutaneously into mice as previously described ( Gauthier et al. 2010 ) As seen in Table 7 2 there was no significant difference in virulence between the viuB mutant and wild type surrogate based on liver and skin samples (p values of 0.99 and 0.11 respectively) This result also supports the above evidence that viuB does not directly play a role in virulence. As with the previous mouse assay, these results represent data from one experiment but were similar to results from two additional experiments, which also did not show significant differences Tab l e 7 2. Competitive virulence assay in the mouse model, using an mtlA deletion mutant as a surrogate for wild type strain. Represents 1 experiment, but similar results were obtained on two additional experiments
45 Previous data ( Gauthier et al 2010 ) demonstrated that gacA did play a significant role in virulence depending on the strain examined and the host iron status, and the results herein revealed an association between gacA and expression of iron acquisition genes in the catechol sider ophore pathway, including the viuB gene. However, results in a mous e model clearly showed that the viuB product in the catechol pathway did not contribute to virulence under these conditions. These results argue against the use of this gene as a virulence marker for V. vulnificus ( Bo gard & Oliver, 2007 ; Panicker et al. 2004 ) The supposition that observed differences in phylogroups based on ViuB may be an underlying factor in evolution of more virulence strains is also not su pp orted. Thus, the results from this experiment show that although viuB Type 1 correlates with clinical origin, it does not predict or contribute to virulence in this model Relationship of Viu B to Hydroxamate Gene Expression Because virulence assays yielded no observed difference between the viuB deletion mutant and wild type in the mouse model, the specific role of iron acquisition in the virulence of V. vulnificus is still unclear. A prior report ( Litwin et al. 1996 ) showed small but significant reduction in virulence that was attributed to the loss of vulnibactin expr ession. These differences could be attributed to differences in animal models (infant vs. adult mice), but they might also be a consequence of the relative contribution of the two genes to the catechol pathway. It is also possible that alte rnative iron acq uisition pathways ( i.e. the hydroxamate pathway ) may compensate for the loss of viuB expression. Therefore, the effects of the viuB mutation on transcript levels of the genes contributing to the hydroxamate siderophore system were also examined. As seen i n Figure 7 2 the viuB deletion mutant had some changes (0.95, 1.99 and 0.86 fold increase) for transcript levels of ATP, PP, and iutA respectively, compared
46 to wild type; however, none of these values were significant under iron replete conditions. Signi ficant differences were also not observed for the complemented mutant or the plasmid control under these conditions. However, significant differences were seen between mutant and wild type strain f or iron limiting growth. The viuB deletion mutant exhibited fold increases of 5.56 (0.008), 2.01, and 6.84 (0.0001) for ATP, PP, and iutA transcript levels, respectively, compared to wild type. The plasmid control had similar fold increases of 4.99 ( p value 0.001), 2.88, and 4.06 ( p value 0.0005) when compared to wild type and the complemented mutant did not differ significantly from wild type for all three genes examined Figure 7 2 Fold changes of hydroxamate siderophore iron acquisition genes when the viuB deletion mutant is com pared to wild type strain CMCP6 under iron replete and iron limiting conditions. Results are the average of 3 independent experiments with at least 2 technical replicates for each experiment. (Significant differences are shown). Further work is needed to u nderstand the relative contribution of the catechol and hydroxamate pathways to survival of V. vulnificus under limiting conditions, but these data suggest that loss of one pathway may be circumvented by the increased activity of the alternate pathway for iron uptake. A compensatory pathway may also explain why (0.008) (0.0002)
47 little or no effect on virulence in observed when components of the catechol pathway are eliminated through genetic manipulation.
48 CHAPTER 8 SUMMARY AND CONCLUSI ONS The GacA Regul ation of Iron A cquisition in V. vulnificus Vibrio vulnificus is a halophilic, opportunistic bacterium known to cause systemic disease in individuals with compromised conditions that involve altered host iron status with excess physiologically available ir on Prior experimental research demonstrated an association between iron availability in the host and the relative virulence of V. vulnificus in mice, whereby increased host iron status through injection of exogenous iron caused greatly increased lethality of both clinical and environmental strains of the bacterium ( Litwin et al. 1996 ; Okujo et al. 1996 ; Simpson & Oliver, 1983 ; Starks et al. 2000 ; Wright et al. 1981 ; Wright et al. 1986 ) The LD 50 approached one bacterium in the first study. V vulnificus contains multiple high affinity iron acquisition system s, including the expression of both catechol and hydroxamate siderophores. In th e present stud y the catechol siderophore system was examined for its contribution to growth under iron li miting conditions and virulence, as well for its interactions with the hydroxamate system. Research in the related species, Vibrio fischeri demonstrated that the G acS/GacA two component signal transduction pathway could regulate iron acquisition ( Whistler & Ruby, 2003 ) and therefore the role GacS/A on regulation of iron acquisition genes of V. vulnificus was also examined A deletion mutation of the gacA gene of V. vulnificus significantly reduced growth yield under con ditions of iron limitation as compa red to wild type strain CMCP6. The catechol system was found to be under GacA control. Under iron limiting conditions, the gacA deletion mutant exhibited significant decreases of all three catechol genes tested compared t o wild type strain (p<0.001), and this activity was partially restored in the
49 complemented mutant, as the fold decreases in transcripts were significantly smaller compared to the gacA deletion mutant For the hydroxamate system, the gacA deletion mutant sh owed no significant decrease compared to wild type strain under iron limiting conditions, and expression actually increased during iron replete growth. These results demonstrate that the GacS/GacA system regulates the catechol siderophore system under iron limitation through up regulated expression of catechol related genes for iron acquisition However, GacA behaves differentially with respect to the hydroxamate system. Instead, loss of GacA expression did not alter regulat ion of the hydroxamate siderophor e system under iron limiting conditions. The incre ased gene expression under iron replete conditions could be explained through possible de repression in the absence of GacA. Clearly, GacA provides differential regulation for these two high affinity iron a cquisition systems in V. vulnificus viuB and its Role in Iron Acquisition and Virulence The viuB gene has been proposed as a potential virulence factor for V. vulnificus and was examined as possible marker to discriminate virulent vs. avirulent strains ( Bogard & Oliver, 2007 ; Panicker et al. 2004 ) Comparative phylogenetic analysis of clinical and environmental isolates confirmed correlation of viuB gene sequence with strain origin and further demonstrated that viuB sequence w as also associated with virulence in V. vulnificus in mice. However, analysis of the deduced amino acid sequence s for other catechol genes did not show similar assoc iations with clinical origins and virulence. VenB did show a branch clustered with all the environmental isolates studied, but it proved not to be significantly different from the rest of the strains. This demonstrates that although iron acquisition is important for virulence of V.
50 vulnificus the evolution of these particular genes in clinical and environmental isolates is not the sole cause for differences in virulence In order to definitively ascertain the role of the ViuB hydrolase and the catechol system in iron acquisition and virulence, a viuB deletion mutant was cons tructed and examin ed for altered phenotypes under various in vitro and in vivo conditions. The results demonstrated that viuB is required for growth under in vitro iron limiting conditions in V. vulnificus However, r egardless of host iron status or the presence or absence of the viuB gene, no significant differences were seen between the mutant and wil d type in virulence assessment. A competitive virulence assay yielded similar results, as nearly equal amounts of each strain was recovered from the liver and skin lesions fol lowing infections. H ypothesis 3 regarding that viuB was important to virulence was not supported by these results. These data seemed to conflict with previous research demonstrating that the venB gene did cause a significant decrease of virulence in V. vul nificus ( Litwin et al. 1996 ) as it might be expected for both g enes to contribute similarly regarding virulence However, V enB is responsible for siderophore biosynthesis and deleting this gene also prevents the function of downstream components of the pathway. Conversely the viuB deletion mutant s presumably synthes ize the siderophore, release it, and bring in the iron siderophore complex ; but then are either not able to release the iron from the siderophore at all, or release it with much lower efficiency In the iron limiting growth study, the viuB mutant showed gr eatly reduced growth yield but was still able to survive and grow somewhat In the mice the mutant was just as virulent as wild type, and demonstrated no competitive disadvantage. These results suggest that alternate
51 pathways may be deployed for iron acqui sition in vitro vs. virulence in an animal host The transcript levels of two of the hydroxamate related genes, ATP and iutA were significantly increased in the viuB mutant compared to wild type strain under iron limitin g conditions. This increase althou gh small was significant and could offer an explanation for the maintenance of the virulence phenotype in the viuB mutant. The hydroxamate siderophore system uses chemical reduction rather than a hydrolase to remove the iron from the siderophore, and thus this system cannot compensate through its own hydrolase R ather an alternative pathway most likely exists to explain this observation and is for future studies Finally, different animal models were used in the Litwin et al. study (2000) and may also exp lain contradictory finding for the role of the catechol iron uptake system in virulence of V. vulnificus This research demonstrated that although viuB is required for growth under iron limiting conditions, its role in virulence of V. vulnificus is not su pported It is likely that the multiple pathways available for high affinity iron acquisition, contribute differentially to survival in the host and other habitats. This research confirmed the complexity of the multiple iron acquisition systems in this spe cies, indicating co ordinate regulation by GacA and feedback between the systems as indicated by the effects of loss of viuB expression on hydroxamate system. Further research is needed to define the distinct roles of different iron acquisition systems an d the regulatory pathways that control their function. This research confirms the multi factorial nature of V. vulnificus virulence and supports the need for additional investigations in order to discover meaningful markers for virulence in this species.
52 A PPENDIX ADDITIONAL MATERIALS A. B. C. IRON REPLETE IRON LIMITING Figure A 1. Role of GacA in the fold changes of gene transcripts in the catechol siderophore system under iron replete and iron limiting conditi ons comparing a gacA deletion mutant (A), the complemented mutant (B), and the plasmid control (C) to wild type strain CMCP6. Iron replete conditions were Luria Burtani Broth with NaCl ( LBN ) and iron limiting were LBN + 150 M dipyridyl. Fold differences i n gene expression are shown and p values were calculated as described in Materials and Methods.
53 A. B C. IRON REPLETE IRON LIMITING Figure A 2 Role of GacA in the fold changes of gene transcripts in the hydroxamate sider ophore system under iron replete and iron limiting conditions comparing a gacA deletion mutant (A), the complemented mutant (B), and the plasmid control (C) to wild type strain CMCP6. Iron replete conditions were LBN and iron limiting were LBN + 150 M dip yridyl. Fold differences in gene expression are shown and p values were calculated as described in Materials and Methods.
54 A. B. C. IRON REPLETE IRON LIMITING Figure A 3 Role of ViuB in the fold changes of gene transcr ipts in the hydroxamate siderophore system under iron replete and iron limiting conditions comparing a viuB deletion mutant (A), the complemented mutant (B), and the plasmid control (C) to wild type strain CMCP6. Iron replete conditions were LBN and iron l imiting were LBN + 130 M dipyridyl. Fold differences in gene expression are shown and p values were calculated as described in Materials and Methods.
55 LIST OF REFERENCES Alice, A. F., Naka, H. & Crosa, J. H. (2008). Global Gene Exp ression as a Function of the Iron Status of the Bacterial Cell: Influence of Differentially Expressed Genes in the Virulence of the Human Pathogen Vibrio vulnificus. Infect Immun 76 4019 4037. Bailey, S., Evans, R. W., Garratt, R. C., Gorinsky, B., Hasnai n, S., Horsburgh, C., Jhoti, H., Lindley, P. F., Mydin, A. & other authors (1988). Molecular structure of serum transferrin at 3.3 A resolution. Biochemistry 27 5804 5812. Blake, P. A., Merson, M. H., Weaver, R. E., Hollis, D. G. & Heublein, P. C. (1979). Disease caused by a marine Vibrio. Clinical characteristics and epidemiology. N Engl J Med 300 Bogard, R. W. & Oliver, J. D. (2007). Role of Iron in Human Serum Resistance of the Clinical and Environmental Vibrio vulnificus Genotypes. Appl Environ Micr obiol 73 7501 7505. Bogard, R. W. & Oliver, J. D. (2008). Role of Iron in Human Serum Resistance of the Clinical and Environmental Vibrio vulnificus Genotypes. Appl Environ Microbiol 74 3322 Butterton, J. R. & Calderwood, S. B. (1994). Identification, cloning, and sequencing of a gene required for ferric vibriobactin utilization by Vibrio cholerae. J Bacteriol 176 5631 5638. CDC (2008). Surveillance for waterborne disease and outbreaks associated with recreational water use and other aquatic facility a ssociated health events -United States, 2005 2006. MMWR 57 1 29. Falkow, S. (1988). Molecular Koch's postulates applied to microbial pathogenicity. Reviews of infectious diseases 10 Suppl 2 S274 276. Forouhi, N. G., Harding, A. H., Allison, M., Sandhu, M S., Welch, A., Luben, R., Bingham, S., Khaw, K. T. & Wareham, N. J. (2007). Elevated serum ferritin levels predict new onset type 2 diabetes: results from the EPIC Norfolk prospective study. Diabetologia 50 949 956. Gan, E. K., Powell, L. W. & Olynyk, J. K. (2011). Natural history and management of HFE hemochromatosis. Seminars in liver disease 31 293 301. Ganne Carrie, N., Christidis, C., Chastang, C., Ziol, M., Chapel, F., Imbert Bismut, F., Trinchet, J. C., Guetti er, C. & Beaugrand, M. (2000). Liver iron is predictive of death in alcoholic cirrhosis: a multivariate study of 229 consecutive patients with alcoholic and/or hepatitis C virus cirrhosis: a prospective follow up study. Gut 46 277 282.
56 Gauthier, J. D., Jo nes, M. K., Thiaville, P., Joseph, J. L., Swain, R. A., Krediet, C. J., Gulig, P. A., Teplitski, M. & Wright, A. C. (2010). Role of GacA in virulence of Vibrio vulnificus. Microbiology 156 3722 3733. Jehn, M. L., Guallar, E., Clark, J. M., Couper, D., Dun can, B. B., Ballantyne, C. M., Hoogeveen, R. C., Harris, Z. L. & Pankow, J. S. (2007). A prospective study of plasma ferritin level and incident diabetes: the Atherosclerosis Risk in Communities (ARIC) Study. American journal of epidemiology 165 1047 1054 Jones, M. K. & Oliver, J. D. (2009). Vibrio vulnificus: Disease and Pathogenesis. Infection and immunity In press Kaspar, C. W. & Tamplin, M. L. (1993). Effects of temperature and salinity on the survival of Vibrio vulnificus in seawater and shellfish. Applied and environmental microbiology 59 2425 2429. Kim, C. M., Park, R. Y., Chun, H. J., Kim, S. Y., Rhee, J. H. & Shin, S. H. (2007a ). Vibrio vulnificus metalloprotease VvpE is essentially required for swarming. FEMS microbiology letters 269 170 179. Kim, C. M., Park, Y. J. & Shin, S. H. (2007b ). A widespread deferoxamine mediated iron uptake system in Vibrio vulnificus. The Journal of infectious diseases 196 1537 1545. Lapouge, K., Schubert, M., Allain, F. H. & Haas, D. (2008). Gac/Rsm signal transdu ction pathway of gamma proteobacteria: from RNA recognition to regulation of social behaviour. Molecular microbiology 67 241 253. Litwin, C. M. & Calderwood, S. B. (1993). Cloning and genetic analysis of the Vibrio vulnificus fur gene and construction of a fur mutant by in vivo marker exchange. J Bacteriol 175 706 715. Litwin, C. M., Rayback, T. W. & Skinner, J. (1996). Role of catechol siderophore synthesis in Vibrio vulnificus virulence. Infect Immun 64 2834 2838. Liu, M., Alice, A. F., Naka, H. & Cros a, J. H. (2007). The HlyU protein is a positive regulator of rtxA1, a gene responsible for cytotoxicity and virulence in the human pathogen Vibrio vulnificus. Infection and immunity 75 3282 3289. Livak, K. J. & Schmittgen, T. D. (2001). Analysis of relati ve gene expression data using real time quantitative PCR and the 2( Delta Delta C(T)) Method. Methods 25 402 408. Motes, M. L., DePaola, A., Cook, D. W., Veazey, J. E., Hunsucker, J. C., Garthright, W. E., Blodgett, R. J. & Chirtel, S. J. (1998). Influenc e of water temperature and salinity on Vibrio vulnificus in Northern Gulf and Atlantic Coast oysters (Crassostrea virginica). Applied and environmental microbiology 64 1459 1465.
57 Okujo, N., Akiyama, T., Miyoshi, S., Shinoda, S. & Yamamoto, S. (1996). Invo lvement of vulnibactin and exocellular protease in utilization of transferrin and lactoferrin bound iron by Vibrio vulnificus. Microbiology and immunology 40 595 598. Panicker, G., Vickery, M. & AK., B. (2004). Multiplex PCR detection of clinical and env ironmental strains of Vibrio vulnificus in shellfish. Canadian journal of microbiology 50 911 922. Randa, M. A., Polz, M. F. & Lim, E. (2004). Effects of temperature and salinity on Vibrio vulnificus population dynamics as assessed by quantitative PCR. Ap plied and environmental microbiology 70 5469 5476. Simpson, L. M. & Oliver, J. D. (1983). Siderophore production by Vibrio vulnificus. Infection and immunity 41 644 649. Starks, A. M., Schoeb, T. R., Tamplin, M. L., Parveen, S., Doyle, T. J., Bomeisl, P. E., Escudero, G. M. & Gulig, P. A. (2000). Pathogenesis of Infection by Clinical and Environmental Strains of Vibrio vulnificus in Iron Dextran Treated Mice. Infect Immun 68 5785 5793. Tacket, C. O., Brenner, F. & Blake, P. A. (1984). Clinical features a nd an epidemiological study of Vibrio vulnificus infections. The Journal of infectious diseases 149 558 561. Tamura, K., Dudley, J., Nei, M. & Kumar, S. (2007). MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Molecular biology and evolution 24 1596 1599. Tanabe, T., Naka, A., Aso, H., Nakao, H., Narimatsu, S., Inoue, Y., Ono, T. & Yamamoto, S. (2005). A novel aerobactin utilization cluster in Vibrio vulnificus with a gene involved in the transcription regulation of the iutA ho mologue. Microbiology and immunology 49 823 834. Thiaville, P. C., Bourdage, K. L., Wright, A. C., Farrell Evans, M., Garvan, C. W. & Gulig, P. A. (2011). Genotype is correlated with but does not predict virulence of Vibrio vulnificus biotype 1 in subcuta neously inoculated, iron dextran treated mice. Infection and immunity 79 1194 1207. Webster, A. C. D. & Litwin, C. M. (2000). Cloning and Characterization of vuuA, a Gene Encoding the Vibrio vulnificus Ferric Vulnibactin Receptor. Infect Immun 68 526 534 Whistler, C. A., Corbell, N. A., Sarniguet, A., Ream, W. & Loper, J. E. (1998). The two component regulators GacS and GacA influence accumulation of the stationary phase sigma factor sigmaS and the stress response in Pseudomonas fluorescens Pf 5. Journal of bacteriology 180 6635 6641.
58 Whistler, C. A. & Ruby, E. G. (2003). GacA regulates symbiotic colonization traits of Vibrio fischeri and facilitates a beneficial association with an animal host. Journal of bacteriology 185 7202 7212. Wright, A. C., Hill R. T., Johnson, J. A., Roghman, M. C., Colwell, R. R. & Morris, J. G., Jr. (1996). Distribution of Vibrio vulnificus in the Chesapeake Bay. Applied and environmental microbiology 62 717 724. Wright, A. C., Simpson, L. M. & Oliver, J. D. (1981). Role of iron in the pathogenesis of Vibrio vulnificus infections. Infection and immunity 34 503 507. Wright, A. C., Simpson, L. M., Richardson, K., Maneval, D. R., Oliver, J. D. & Morris, J. G. (1986 ). Siderophore production and outer membrane proteins of select ed Vibrio vulnificus strains under conditions of iron limitation. FEMS Microbiol Lett 35 255 260.
59 BIOGRAPHICAL SKETCH Rick Swain was born on September 9 th 1986, in Cape May, New Jersey. After graduating 3 rd in his high school class, he decided to att end Coastal Carolina University in Conway, South Carolina where he majored in Biology. After one year, he transferred to the Biotechnology Department of Rutgers University in New Brunswick, New Jersey to pursue a new field of study. During the remaining th ree years of his undergraduate experience, Rick excelled in his academics and undertook a minor in Biochemistry to supplement his education. In addition to academics, Rick also began research at the Waksman Institute located in Piscataway, New Jersey where he studied an unknown protein found in Arabidopsis thaliana in the lab of Dr. Todd Michael. After graduating magna cum laude with a B.S., he was offered a graduate position under the guidance of Dr. Anita Wright in the Department of Food Science and Human Nutrition at the University of Florida. In 2011, he graduated with his masters and plans on pursuing a career in the biotechnology and biomedical industry before potentially going back to earn a Ph.D.