<%BANNER%>

Identification of Novel Virulence Factors of Vibrio vulnificus by Using Alkaline Phosphatase Mutagenesis

Permanent Link: http://ufdc.ufl.edu/UFE0022676/00001

Material Information

Title: Identification of Novel Virulence Factors of Vibrio vulnificus by Using Alkaline Phosphatase Mutagenesis
Physical Description: 1 online resource (187 p.)
Language: english
Creator: Brown, Roslyn
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: bacterium, degp, fadr, fatty, infection, mouse, pathogen, pathogenesis, regulation, rsea, rseb, sigma, stress, vibrio, virulence, vulnificus
Immunology and Microbiology (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Vibrio vulnificus, a gram-negative estuarine bacterium, is capable of causing serious opportunistic infections - primary septicemia and wound infection. Our goal was to identify novel virulence factors, with specific focus on those that are exported into and beyond the cell membrane because most virulence factors are exported. We used PhoA mutagenesis, a powerful tool for mutating and identifying genes that encode exported products. We identified several PhoA mutants of V. vulnificus that were attenuated for virulence in a mouse model of disease and chose two for detailed analysis. FLA602 had a phoA insertion in the fadR gene. Although FadR is not an exported protein, the mutant was significantly attenuated for virulence. We deleted fadR from wild-type V. vulnificus and characterized the role of fadR in fatty acid metabolism, membrane integrity, growth, and virulence, and concluded that the major defect of the fadR mutant was insufficient synthesis of fatty acids. FLA609 had an in-frame fusion of phoA with rseB that encodes a negative regulator of sigma E (envelope stress sigma factor). FLA609 exhibited phase-variation between translucent and opaque morphologies and was attenuated for virulence in the s.c. mouse model of V. vulnificus infection. To investigate into the roles of rseB and the envelope stress response in V. vulnificus, defined mutations in sigma E-related genes (rseB, rseA, rpoE, and degP) were constructed for analysis of virulence, morphology, and stress resistance. Carbohydrate analyses suggested that the decreased opacity of rseB phase-variants may be due to decreased extracellular polysaccharides. As such, we may have identified a novel mechanism for control of colony opacity in V. vulnificus. Analysis of V. vulnificus rpoE and degP mutants showed that these genes are essential for responding to envelope stress in V. vulnificus, but they appeared to be dispensable for virulence. These studies identified two novel factors that have roles in virulence in V. vulnificus, FadR and RseB. This work was also the first to investigate the role of the envelope stress response in V. vulnificus. Further studies will further uncover the role of fatty acid metabolism in pathogenesis and the mechanism of phase variation in rseB mutants.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Roslyn Brown.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Gulig, Paul A.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-02-28

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022676:00001

Permanent Link: http://ufdc.ufl.edu/UFE0022676/00001

Material Information

Title: Identification of Novel Virulence Factors of Vibrio vulnificus by Using Alkaline Phosphatase Mutagenesis
Physical Description: 1 online resource (187 p.)
Language: english
Creator: Brown, Roslyn
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: bacterium, degp, fadr, fatty, infection, mouse, pathogen, pathogenesis, regulation, rsea, rseb, sigma, stress, vibrio, virulence, vulnificus
Immunology and Microbiology (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Vibrio vulnificus, a gram-negative estuarine bacterium, is capable of causing serious opportunistic infections - primary septicemia and wound infection. Our goal was to identify novel virulence factors, with specific focus on those that are exported into and beyond the cell membrane because most virulence factors are exported. We used PhoA mutagenesis, a powerful tool for mutating and identifying genes that encode exported products. We identified several PhoA mutants of V. vulnificus that were attenuated for virulence in a mouse model of disease and chose two for detailed analysis. FLA602 had a phoA insertion in the fadR gene. Although FadR is not an exported protein, the mutant was significantly attenuated for virulence. We deleted fadR from wild-type V. vulnificus and characterized the role of fadR in fatty acid metabolism, membrane integrity, growth, and virulence, and concluded that the major defect of the fadR mutant was insufficient synthesis of fatty acids. FLA609 had an in-frame fusion of phoA with rseB that encodes a negative regulator of sigma E (envelope stress sigma factor). FLA609 exhibited phase-variation between translucent and opaque morphologies and was attenuated for virulence in the s.c. mouse model of V. vulnificus infection. To investigate into the roles of rseB and the envelope stress response in V. vulnificus, defined mutations in sigma E-related genes (rseB, rseA, rpoE, and degP) were constructed for analysis of virulence, morphology, and stress resistance. Carbohydrate analyses suggested that the decreased opacity of rseB phase-variants may be due to decreased extracellular polysaccharides. As such, we may have identified a novel mechanism for control of colony opacity in V. vulnificus. Analysis of V. vulnificus rpoE and degP mutants showed that these genes are essential for responding to envelope stress in V. vulnificus, but they appeared to be dispensable for virulence. These studies identified two novel factors that have roles in virulence in V. vulnificus, FadR and RseB. This work was also the first to investigate the role of the envelope stress response in V. vulnificus. Further studies will further uncover the role of fatty acid metabolism in pathogenesis and the mechanism of phase variation in rseB mutants.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Roslyn Brown.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Gulig, Paul A.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-02-28

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022676:00001


This item has the following downloads:


Full Text

PAGE 1

IDENTIFICATION OF NOVEL VIRULENCE FACTORS OF Vibrio vulnificus BY USING ALKALINE PHOSPHATASE MUTAGENESIS By ROSLYN NATASHA BROWN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008 1

PAGE 2

2008 Roslyn Natasha Brown 2

PAGE 3

To my Mom and Dad, my consta nt sources of inspiration 3

PAGE 4

ACKNOWLEDGMENTS I thank my mentor, Paul A. Gulig for his gui dance, advice, and patie nt instruction during my graduate career. He has ta ught me invaluable lessons in sc ience and in life, and has gone above and beyond the call of duty in many instances I will forever cherish having him as a mentor and friend. My colleagues in the laboratory made it a joy to work there. I especially thank Jennifer Joseph, Patrick Thiaville, Juli o Martin, and Aaron Mittel for assistance with animal experiments. I have also had the pleasu re of assisting two undergraduates, Viviana Reyes and Fernando Donoso, in their scientific training. They have taught me more than I have taught them. I also acknowledge the guidance of my committee members, Drs. Anita Wright, Robert Burne, and Shouguang Jin. They always kept me thinking and asking why. The staff and administrators of the depart ment of Molecular Ge netics and Microbiology and the Office Graduate of Gra duate Education of the College of Medicine also deserve my deepest gratitude for their constant assistance. In particular, I thank Dr. Wayne Mc Cormack, Valerie Cloud-Driver and Susan Gardner from the Graduate Office, and Michelle Ramsey of the MGM Fiscal Office. I am especially thankful to the omniscient Joyce Conners, without whom I would have been a lost raft bobbing aimlessly (and behind schedule) in the vast sea of graduate student rules, regulations, d eadlines, and requirements. I am forever grateful to my parents for their never-ending support, love, prayers, and concern. My extended family has also been an in credible source of support. I also thank my beloved husband, Joseph, for his constant encourag ement, assistance, friendship, and love. He has been my biggest cheerleader and my closest friend during my graduate career. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS ............................................................................................................... 4LIST OF TABLES ...........................................................................................................................9LIST OF FIGURES .......................................................................................................................10ABSTRACT ...................................................................................................................... .............13 CHAPTER 1 INTRODUCTION ................................................................................................................ ..15Vibrio vulnificus ......................................................................................................................15Disease ....................................................................................................................................15Classification of V. vulnificus Strains .....................................................................................16Virulence Factors ....................................................................................................................19Capsule ....................................................................................................................... .....20Acquisition of Iron ..........................................................................................................2 0Flagella and Motility .......................................................................................................21Type IV Pili .....................................................................................................................21Toxins and Destructive Enzymes ....................................................................................22Focus of Investigation .............................................................................................................23Specific Aim 1. Develop and Implement a System for Identifying Secreted Virulence Factors of V. vulnificus ................................................................................24Specific Aim 2. Characterize a fadR Mutant Identified by PhoA Mutagenesis and Investigate the Role of FadR and Fa tty Acid Metabolism in Virulence of V. vulnificus ......................................................................................................................24Specific Aim 3. Characterize an rseB Mutant Identified by PhoA Mutagenesis and Investigate the Role of RseB and Aspects of the Extracytoplasmic Stress Response in Virulence of V. vulnificus ........................................................................252 MATERIALS AND METHODS ...........................................................................................26Standard Microbiological and An imal Infection Protocols ....................................................26Bacterial Strains, Media, and Chemicals .........................................................................26Use of pGTR201 for PhoA Mutagenesis .........................................................................27Measurement of Growth ..................................................................................................27Fatty Acid Analysis .........................................................................................................28Cerulenin MIC ................................................................................................................. 28Assays of Membrane Stress Sensitivity ..........................................................................28Serum Sensitivity .............................................................................................................29Motility ...................................................................................................................... ......29Biofilm Formation Assay ................................................................................................29Extraction of EPS ............................................................................................................30 5

PAGE 6

Analysis of Whole Cell Lysates or EPS Extracts by SDS-PAGE ...................................31Staining of Gels with Alcian Blue ...................................................................................31Staining of Gels with Stains-All ......................................................................................31Infection of Mice ............................................................................................................. 32Molecular Genetic Tech niques and Analyses .........................................................................32Construction of mini-Tn 5Km2phoA Mutagenesis Vector, pGTR201. ............................32USER Friendly Cloning...............................................................................................333-Way USER Deletion Cloning ......................................................................................34Chitin-Induced Transformation .......................................................................................35qRT-PCR ....................................................................................................................... ..35Statistical Analysis ..........................................................................................................363 CONSTRUCTION AND TESTING OF A PhoA MUTAGENESIS SYSTEM FOR Vibrio vulnificus AND IDENTIFICATION OF MUTANTS ................................................41Rationale for Study .................................................................................................................41Introduction .................................................................................................................. ...........41Results .....................................................................................................................................43Development of a System for Identifying Secreted Products of Vibrio vulnificus Using Alkaline Phosphatase (PhoA) Mutagenesis. .....................................................43Testing the PhoA Mutagenesis System in E. coli. ...........................................................43Testing the PhoA Mutagenesis System in V. vulnificus .................................................45Use of the PhoA Mutagenesis System to Identify Secreted Virulence Factors of V. vulnificus ..................................................................................................................47Significant Attenuation of Virulence in a cvpA Mutant of V. vulnificus. ........................51Discussion .................................................................................................................... ...........54Limitations of PhoA Mutagenesis. ..................................................................................54Suitability of pGTR201 for Identifying Genes Encoding Secreted Proteins of V. Vulnificus. ....................................................................................................................55Mutations that Caused Attenuated Virulenc e in the Subcutaneously Inoculated Iron Dextran-Treated Mouse Model of Disease. .................................................................554 THE ROLE OF FadR AND ASPECTS OF FATTY ACID METABOLISM IN VIRULENCE OF V. vulnificus ...............................................................................................64Rationale for Study .................................................................................................................64Introduction .................................................................................................................. ...........64Results .....................................................................................................................................65A V. vulnificus Mini-Tn5Km2phoA Insertion Mutant Is Severely Attenuated for Skin and Liver Infection in Iron Dextran-Treated Mice ..............................................65The FLA602 mini-Tn 5Km2phoA Insertion is in the fadR Gene .....................................66Confirmation of the fadR Phenotype ...............................................................................67In vitro Growth of the fadR Mutant .................................................................................68Complementation and Reversion of the fadR ::mini-Tn 5Km2phoA Mutation .................69Deletion of fadR from Wild-Type CMCP6 .....................................................................70Altered fatty acid content of the fadR mutant ...............................................................71Envelope Stress Sensitivity and Motility of the fadR Mutant .......................................72 6

PAGE 7

Supplementation with Unsaturated Fa tty Acid in Vitro and in Vivo ..............................74The Role of Fatty Acid U tilization in Infection with V. vulnificus .................................75Discussion .................................................................................................................... ...........78Identification of a V. vulnificus fadR Mutation that Attenuates Virulence .....................78Comparison of V. vulnificus FadR with Known FadR Properties ...................................78The Role of FadR in Infection of Mice ...........................................................................79The Roles of FadD and the Glyoxylate Enzymes in V. vulnificus ..................................81Conclusion and Future Directions ...................................................................................825 THE ROLE OF RseB AND THE EXTR ACYTOPLASMIC STRESS RESPONSE IN VIRULENCE OF V. vulnificus ...............................................................................................98Rationale for Study .................................................................................................................98Introduction .................................................................................................................. ...........98Envelope Stress Responses in Bacteria ...........................................................................98Function and Control of the E-Mediated ESR .............................................................100RseB, E, and Virulence in B acterial Pathogens ...........................................................101Results ...................................................................................................................................101Identification of an rseB ::mini-Tn 5Km2phoA Mutation in Vibrio vulnificus ..............101Virulence of RseB Mutants. ..........................................................................................102Deletion of rseB from Wild-Type CMCP6. ..................................................................105Characterization of RseB Phase Variants ......................................................................108Analysis of phase variation. ...................................................................................108CPS and carbohydrate expression. .........................................................................108rpoE expression in rseB mutants ............................................................................115Effects of rpoE overexpression in wild-type V. vulnificus .....................................116Role of the E-Mediated Extracytoplasmi c Stress Response in V. vulnificus ...............118Characterization of a V. vulnificus rpoE mutant ..................................................118Virulence of a V. vulnificus rpoE mutant ............................................................119Identification of V. vulnificus degP ........................................................................120Characterization of a V. vulnificus degP mutant ....................................................121Virulence of V. vulnificus degP ...........................................................................123Deletion of rseA from wild-type CMCP6 ..............................................................124Discussion .................................................................................................................... .........125Identification of a V. vulnificus rseB ::mini-Tn 5Km2phoA Mutation that Caused Altered Colony Morphology and Attenuated Virulence ............................................125EPS and the Altered Colony Morphology of V. vulnificus rseB Mutants ..................126Comparison Between the rseB -Tp Mutant and an Acapsular V. vulnificus Mutant ...128Possible Reasons for the Attenuated Virulence of rseB and rseA Mutants ..............129Regulation of E in V. vulnificus ...................................................................................129E and DegP are Important for Stress Resistance but not for Virulence in V. vulnificus ................................................................................................................130Future Directions ...........................................................................................................1316 PERSPECTIVES ................................................................................................................ ..171 7

PAGE 8

Host-Adapted Versus Accidental Pathogens: Redefining the Term Virulence Factor for Opportunistic Pathogens ..............................................................................................171Role of Metabolism in the Pathogenesis of V. vulnificus .....................................................172REFERENCE LIST .....................................................................................................................174BIOGRAPHICAL SKETCH .......................................................................................................187 8

PAGE 9

LIST OF TABLES Table page 2-1 Bacterial strains used in this study .....................................................................................3 72-2 Plasmids used in this study ............................................................................................... .382-3 Oligonucleotides us ed in this study ...................................................................................392-3 Continued ...........................................................................................................................403-1 Summary of mutants identif ied by PhoA mutagenesis of V. vulnificus. ...........................594-1 Expression of putative FadR-regulated genes in a V. vulnificus fadR mutant. ..................864-2 In vitro growth rate and virulence of V. vulnificus fadR ::mini-Tn 5Km2phoA compared to ptsI ::mini-Tn 5Km2phoA and wild-type. .......................................................874-3 Changes in fatty acid profile of fadR mutant compared to wild-type. ............................904-4 fadR mutant sensitivity to envelope stress ......................................................................914-5 Growth of aceAB and fadD mutants and wild-type CMCP6 in oleate .........................965-1 Sensitivity of rseB mutants to serum complement. .......................................................1525-2 Sensitivity of rseB mutants to envelope stress. .............................................................1535-3 Analysis of expression of rpoE and degP in rseB mutants by qRT-PCR .....................1555-4 Sensitivity of rpoE and degP mutants to envelope stresses. ............................................158 9

PAGE 10

LIST OF FIGURES Figure page 3-1 VectorNTI diagram showing important features of PhoA mutagenesis vector pGTR201............................................................................................................................57 3-2 Improvement of screen ing of PhoA mutants by a ddition of glucose to BCIPcontaining agar ...................................................................................................................58 3-3 Virulence of PhoA mutants. ...............................................................................................6 0 3-4 Robust growth of FLA608 on LB-N ag ar containing 0.2% (w t/vol) glucose. ..................61 3-5 Ability of FLA608 to transform growth of neighboring strains. .......................................62 3-6 Attenuated virulence of recreated VV1_1996 mutant .......................................................63 4-1 Skin and liver infection caused by wild-type and fadR ::mini-Tn 5Km2phoA V. vulnificus in iron dextran-treated mice ..........................................................................84 4-2 Alignments of FadR homologues. .....................................................................................85 4-3 Complementation and reversion of the fadR::mini-Tn 5km2phoA mutation ......................88 4-4 Complementation of fadR mutation ................................................................................89 4-5 Sensitivity of wild-type CMCP6 and fadR FLA614 to heat and cold. ............................92 4-6 Decreased motility of the fadR mutant ............................................................................93 4-7 Supplementation with oleate in vitro .................................................................................94 4-8 Oleate potentiates infection of mice with fadR V. vulnificus ..........................................95 4-9 Virulence of aceAB and fadD mutants of V. vulnificus ................................................97 5-1 Schematic showing V. vulnificus rpoErseABC operon. ...................................................133 5-2 Mixed morphology of rseB ::mini-Tn 5Km2phoA FLA609 ..............................................134 5-3 Preliminary mouse infection with FLA609 ( rseB ::mini-Tn 5Km2phoA ) .........................135 5-4. Mixed morphologies of b acteria harvested from mice infected with FLA609 ................136 5-5 Virulence of translucen t and opaque variants of rseB ::mini-Tn 5Km2phoA FLA609 in mice ..................................................................................................................................137 5-6 Virulence of the opaque variant of rseB ::mini-Tn 5Km2phoA FLA609-O .....................138 10

PAGE 11

5-7 Colony morphology of co mplemented FLA609-Tl .........................................................139 5-8 Complementation of rseB::mini-Tn 5Km2phoA FLA609-Tl. .........................................140 5-9 Virulence of rseB transparent, translucent, a nd opaque variants in mice .....................141 5-10 Variable virulence of FLA610-O .....................................................................................142 5-11 Highly attenuated virulence of FLA610-Tp ....................................................................143 5-12 Complementation of rseB mutation ..............................................................................144 5-13 Morphology of CPS mutant, FLA1009 compared to rseB variants and wild-type CMCP6 ......................................................................................................................... ...145 5-14 Ability of strains to bind Congo red ................................................................................146 5-15 Evidence of carbohydrate content vi sualized by Calcofluor binding. .............................147 5-16 Stains-all staining of whole cell lysates ...........................................................................148 5-17 Analysis of EPS extracts ..................................................................................................149 5-18 PCR to observe genetic rearra ngements in CPS transport genes .....................................150 5-19 Biofilm formation in rseB mutants ................................................................................151 5-20 Motility of rseB mutants ................................................................................................154 5-21 Effect of E overexpression on virulence of CMCP6 ......................................................156 5-22 Morphology of two E mutant isolates compared to wild-type ....................................157 5-23 Sensitivity of rpoE FLA1001 to heat ............................................................................159 5-24 Motility of rpoE mutant compared to wild-type. ...........................................................160 5-25 Virulence of rpoE FLA1001 at high and low inocula.. .................................................161 5-26 Morphology of rpoE and degP mutants on LB-N plates containing the fluorescent dye, Calcofluor. .................................................................................................................162 5-27 Sensitivity of degP FLA1002 to heat. ...........................................................................163 5-28 Motility of degP mutant compared to wild-type. ..........................................................164 5-29 Biofilm formation in degP and rpoE mutants.. ...........................................................165 5-30 Virulence of degP mutant compared to wild-type ........................................................166 11

PAGE 12

5-31 Mixed morphology of rseA FLA1012. ..........................................................................167 5-32 Binding of rseA rseB and wild-type to Congo red. .........................................................168 5-33 Binding of rseA rseB and wild-type to Calcofluor. ........................................................169 5-34 Mixed morphology of bacteria harvested from mice infected with rseA -Tl .................170 12

PAGE 13

Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy IDENTIFICATION OF NOVEL VIRULENCE FACTORS OF Vibrio vulnificus BY USING ALKALINE PHOSPHATASE MUTAGENESIS By Roslyn Natasha Brown August 2008 Chair: Paul A. Gulig Major : Medical Sciences Immunology and Microbiology Vibrio vulnificus a gram-negative estuarine bacterium, is capable of causing serious opportunistic infections primary septicemia and wound infection Our goal was to identify novel virulence factors, with specific focus on those that are exported into and beyond the cell membrane because most virulence factors are expo rted. We used PhoA mutagenesis, a powerful tool for mutating and identifying genes that en code exported products. We identified several PhoA mutants of V. vulnificus that were attenuated for virulen ce in a mouse model of disease and chose two for detailed analysis FLA602 had a phoA insertion in the fadR gene. Although FadR is not an exported protein, the mutant was significantly attenuated for virulence. We deleted fadR from wild-type V. vulnificus and characterized the role of fadR in fatty acid metabolism, membrane integrity, growth, and virulence, and conclude d that the major defect of the fadR mutant was insufficient synthesis of fatty acids. FLA609 had an in-frame fusion of phoA with rseB that encodes a negative regulator of E (envelope stress response sigma factor). FLA 609 exhibited phase-variati on between translucent and opaque morphologies and wa s attenuated for virulence in the s.c. mouse model of V. vulnificus infection. To investig ate into the roles of rseB and the envelope stress response in V. 13

PAGE 14

14 vulnificus, defined mutations in E-related genes ( rseB, rseA, rpoE, and degP ) were constructed for analysis of virulence, colony morphology, and stress resistance. Carbohydrate analyses suggested that the d ecreased opacity of rseB phase-variants may be due to decreased extracellular polysaccharides. As such, we may have identified a novel mechanism for control of colony opacity in V. vulnificus Analysis of V. vulnificus rpoE and degP mutants showed that these genes are essential for re sponding to envelope stress in V. vulnificus but they appeared to be dispensable for virulence. These studies identified two novel factor s that have roles in virulence in V. vulnificus, FadR and RseB. This work was also the first to investigate the role of the envelope stress response in V. vulnificus Further studies will further uncove r the role of fatty acid metabolism in pathogenesis and the mechan ism of phase variation in rseB mutants.

PAGE 15

CHAPTER 1 INTRODUCTION Vibrio vulnificus Vibrio vulnificus is the bacterium responsible for most of the seafood-related deaths reported in the United States (1). It is a gram-n egative, motile, curved ro d that naturally occurs in estuarine environments. It resides in areas with tropical to subtropical climates and proliferates when water temper atures are above 18C (1). V. vulnificus can be found free-living in the water column or associated with filter-feed ing molluscs such as clams and oysters. In the United States it is found mainly in the waters of the Gulf of Mexico, and during the warm summer months almost all of the oysters harves ted from these waters are contaminated with V. vulnificus (1). Disease V. vulnificus is capable of causing serious and ofte n fatal infections: primary septicemia and wound infection. Consumption of raw contaminated shellfish can lead to primary septicemia, a systemic illness that often lead s to death. Wound inf ections occur upon exposure of preexisting wounds to V. vulnificus or when a patient sustains a wound upon contact with seawater or seafood contaminated with V. vulnificus. In some cases, wound infection progresses to septicemia. A hallmark of both diseases is the extremely rapid replic ation of the bacteria leading to extensive tissue damage There are also reports that V. vulnificus can cause gastroenteritis, although a clea r causal relationship has not been established (2). Patients with septicemia present with fe ver, chills, hypotension, and characteristic secondary bullous skin lesions on the extremities (3-5). The skin lesions are filled with a hemorrhagic fluid, are usually conf ined to the subcutaneous (s.c.) regions, and lead to swelling and pain in the extremities. These lesions often become necrotic, a condition known as 15

PAGE 16

necrotizing fasciitis. Patients often require aggressive debridem ent of wounds or amputation of the affected limb (2,6,7). Death occurs in 60 to 75% of cases (1). The symptoms of V. vulnificus wound infection are similar to those mentioned for primary septicemia, but are usually less severe. Inflammation of the wound can advance to cellulitis and necrosis and septicemia often ensues. The mortality rate for V. vulnificus wound infection is 20 to 30% (1). V. vulnificus is susceptible to many antibiotics, but prompt treatment is important because the fatality rate increases rapidly as the time betw een initial infection and start of antibiotic treatment increases (8). Several conditions predispos e hosts to contracting V. vulnificus septicemia, especially immune compromise and chronic li ver diseases such as cirrosis, hepatitis, or liver cancer. Conditions resulting in elevated serum iron levels such as hemochromatosis and thalassemia also increase host susceptibility (9). Interestingly, despite levels of V. vulnificus contamination of oysters and susceptible individuals consumi ng raw oysters during the summer months, the number of reported cases of V. vulnificus infection remains low, with 30 to 50 cases annually in the U.S. (1,8,10). More research into the pred isposing conditions for th e disease and into the reasons for the rapid and highly de structive nature of the infections will likely lead to new preventive measures and treatment options. Classification of V. vulnificus Strains V. vulnificus can be classified by several means: biotypes, lipopolysaccharide (LPS) antigens, and most recently genetic se quences. There are three biotypes of V. vulnificus Biotype 1 is the main human pat hogen, biotype 2 is predominantly associated with fish and eels, and biotype 3 is an emerging human pathogen identif ied in Israel (11). While biotype 2 strains possess a single type of LPS, biotype 1 stra ins have heterogeneous LPS profiles (12). 16

PAGE 17

Various forms of molecular typing have b een applied to differentiate strains of V. vulnificus. Ryang, et al. (13) used pulsed-field gel electroph oresis (PFGE) and random amplified polymorphic DNA (RAPD) analysis to assess the molecular diversity of V. vulnificus strains isolated from clinical specimens in Korea. Their analysis revealed considerable genetic diversity among the clinical isolates studied. A similar anal ysis was done to evalua te genetic relationships among 62 V. vulnificus strains of different geographical and host origins using multilocus enzyme electrophoresis (MLEE), RAPD and sequence analyses of the recA and glnA genes (14). The MLEE analysis showed numerous genetic po lymorphisms among the loci examined (11 of 15 loci were polymorphic). MLEE and RAPD rev ealed two general subdi visions (divisions I and II) of the 62 strains studied, although the signif icance of the divisions was not clarified. Sequence polymorphism of the 16S rRNA gene was also identified as a possible means of classifying V. vulnificus Nilsson, et al. (15) noted 17-nucleotide differences throughout the sequence of the small subunit (16S ) rRNA gene that can divide V. vulnificus into two major groups designated types A and B. Inspection of the 16S rRNA genotype in 67 clinical and nonclinical strains of V. vulnificus confirmed that the majority ( 31 of 33) of nonclinical isolates were type A and (26 of 34) human clini cal strains were mo stly type B (15). Warner and Oliver (16) developed an RAPD method that identified a PCR amplicon that was present in clinically-derived (C) strains and was only occasionally present in strains of environmental (E) origin. This group later develope d a PCR method to distinguish between Ctype (correlates with clinical origin) and E-type (correlates with environmental origin) strains based on the previously-identified DNA polymorphism (17). Repetitive Extragenic Palindromic DNA PCR (rep-PCR), a genetic typing method that targets multiple genomic loci comprised of cons erved repetitive elements, also distinguished 17

PAGE 18

between strains of clinical versus environmental origin (18). In terestingly, this method showed greater genetic diversity than was revealed by other typing methods. Most recently, a 33-kb genomic island identified by Cohen, et al. (19) was associated with a virulent clade of V. vulnificus The genomic region, designated region XII, was identified by genome comparisons as one of three genomic regions that are unique to the V. vulnificus species compared to V. parahaemolyticus and V. cholerae The authors suggested that the genes located on region XII, an arylsulfatase gene cluster, a sulfate reduction sy stem, two chondroitinase genes, and an oligopeptide ABC transport system, may provide a selective advantage for survival in the environment or in the host. This is currently under investig ation in our laboratory. Furthermore, multi locus sequence typing (MLS T) was used to investigate DNA sequence polymorphisms of housekeeping genes as a mean s of exploring genetic relationships among V. vulnificus strains. Cohen, et al. (19) examined 67 strains from a globally distributed sample of environmental isolates and clini cal isolates derived from cases of human infection. MLST using six housekeeping genes divided most (63) of the isolates into tw o main lineages, and there was a higher proportion of clinical isolat es in lineage I than in lineage II. Also the lineage I strains were associated with the genomic island referred to above (19). Bi sharat, et al. (11) used MLST to genotype 159 isolates of V. vulnificus sampled from cases of human disease and from environmental sources around the wo rld. This analysis placed V. vulnificus strains into two groups, Clusters I and II. Clus ter II was composed mainly of strains sampled from infected humans, and Cluster I was comprised mainly of strains sampled from environmental sources. This suggested that not all populations of V. vulnificus are equally pathogenic to humans. Although these genetic methods sh ow clear associations between various genetic loci and clinical or environmental origin of V. vulnificus strains, care should be taken before making 18

PAGE 19

inferences that these markers indicate tr ue virulence potential. An analysis of V. vulnificus from market oysters and oyster isolates associated primary septicemia cases investigated genetic traits (presence of plasmids and a 460-bp RAPD amplic on) that could potentiall y distinguish the two sources (20). As part of that study, our laborat ory tested these strain s in a mouse model of disease and showed that nearly all V. vulnificus strains, regardless of th eir origin, were virulent, as defined as causing skin infection in subcutan eously (s.c.) inoculated, iron dextran-treated mice. Moreover, the genetic tests did not distin guish between fully virulent and less virulent strains or between clinical and environmental is olates. Recently, our laboratory performed a detailed analysis of virulence properties of 71 stra ins of clinical and environmental sources in the s.c. inoculated iron dextran-treated mouse model of infection (Gulig, et al., in preparation). While almost all strains could cause severe sk in infection, only a subs et of strains had the potential to cause liver infection and death. Notably, this higher virulence potential was significantly correlated with genotyp es derived from clinical isolat es. Thus, nearly all biotype 1 V. vulnificus strains have the ability to cause skin in fection, but the clinical-type clade is more proficient at causing systemic infect ion and death in our mouse model. Virulence Factors The hallmarks of diseases caused by V. vulnificus are the extreme destruction of host tissues and the rapid replication of the b acteria in the host. The means by which V. vulnificus achieves these characteristics remains unclear, despite a considerable amount of published research (reviewed by Gulig, et al ., (9)). Several putative virule nce factors have been proposed, including a metalloprotease, hemolysin/cyto lysin, lipopolysaccharide (LPS), capsular polysaccharide (CPS), iron acquisition, and various secreted toxins, but few of these have satisfied the Molecular version of Kochs postulates (21) for definitive proof of linkage with virulence. The known virulence factors incl ude the antiphagocytic capsule (22-24), iron 19

PAGE 20

sequestration via siderophores (25,26), flagel la and motility (27,28), type IV pili (29), the fibronectin-binding outer membrane protein OmpU (30), the HlyU prot ein that upregulates expression of some toxin genes (31), and the RtxA1 toxin (32-34). Capsule All virulent strains of V. vulnificus produce an extracellular polysaccharide capsule. To date, this remains the most important know n virulence factor of this bacterium Acapsular mutants, either naturally occu rring or constructed (23,24), are attenuated in mouse models of infection and susceptible to complement The capsule appears to be the major mechanism of resistance to host defenses, as it enables resistance to the opsonic and bactericidal activity of complement. The presence of capsule is relate d to colony morphology: en capsulated strains are opaque and unencapsulated (or under-capsulated) strains are translucent (23,24). Additionally, capsule biosynthesis undergoes reversible phase variation, with opaque strains switching to translucent and vice versa with frequencies in the range of 10-4 (24). Acquisition of Iron The fact that iron overload is the major risk factor for V vulnificus infection suggests that the pathogenesis of this organism is dependent on its ability to sequester excess iron from the host. V. vulnificus belongs to a small group of disease-causing organisms w hose virulence is enhanced by elevated iron levels (35). Key observations on the iron-requirement of V. vulnificus were made by the group of Wright, Simpson, and Oliver (25). The bacteria showed reduced growth in serum unless iron in the form of tran sferrin or hematin was added, and pretreatment with iron in a mouse model reduced the LD50 from 106 to 1 CFU. V. vulnificus produces ironscavenging siderophores and membrane receptor s that bind host iron-containing proteins including hemoglobin, methemoglobin, and hematin (36). Vulnibactin, a V. vulnificus siderophore, is crucial for viru lence, as a transposon mutant was unable to bind iron and was 20

PAGE 21

attenuated for virulence in a mouse model (26). Recently, some groups tried to determine if iron overload acts to increase host sus ceptibility (e.g., by i nhibiting phagocyte func tion) or increases the virulence of the pathogen (e.g., by increasi ng growth rate or regulating virulence gene expression). Two groups, includi ng our laboratory, have shown that the result of iron overload is to increase the replicatio n of the bacteria in the animal host (37,38). One investigator (35) suggested that hepcidin, a hepa tic antimicrobial protein invol ved in host iron regulation, is needed for an effective host response against V. vulnificus. Overall, it appears that the main role of iron scavenging by V. vulnificus is to facilitate growth of the bacteria in the host. Flagella and Motility There is an increasing body of evidence that fl agella are important fo r the pathogenesis of V. vulnificus Kim and Rhee (27) showed that a mutant in the flagellar basal body ( flgC gene) was defective in motility, adherence to HeLa (cervi cal cancer cell line) cells, and virulence, with a 104-fold increase in LD50 in orally inoculated suckling mice. A mutation in the gene encoding the flagellar hook protein ( flgE ) decreased virulence in i.p. inj ected iron dextran-treated mice, with a 103-fold increase in LD50 (28). Our laboratory isolated a mutant in a flagellar biosynthesis gene, fliP, by signature-tagged mutagenesis. The mutant was nonmotile and was attenuated in the s.c. inoculated, iron dextra n-treated mouse model of diseas e (Lang and Gulig, unpublished). Our laboratory further investigated the role of flagella in V. vulnificus by constructing deletions in the flagellin gene clusters, flaCDE and flaFBA (Tucker, et al. in preparation). The flaCDE gene cluster was essential for flagellar biosynthesis and system ic disease in an iron dextrantreated mouse model. Type IV Pili Paranjpye, et al. (39) showed that a mutation in pilD encoding a type IV prepilin peptidase/Nmethyltransferase, abolished expression of surface pili and resulted in a 100-fold 21

PAGE 22

increase in LD50 in an iron-treated, i.p. inoculated m ouse model of infection. However, the pilD mutation caused pleiotropic effects. Hemolysin/cytolysin, protease, and chitinase secretion were also ablated, so the virulence def ect could not be directly attribut ed to a lack of surface pili. Later Paranjpye and Strom (29) also constructed a mutation in the type IV pilin, PilA, that resulted in decreased adheren ce to HEp-2 cells, decreased biof ilm formation, and had a 10-fold increase in LD50 in iron-treated i.p. inoculated mice. The V. vulnificus pilABCD operon shares homology with pilus biosynthesis loci of other gram-negative pathogens, including Vibrio cholerae, Pseudomonas aeruginosa, and Aeromonas hydrophila (29). Toxins and Destructive Enzymes The hemolysin/cytolysin and metalloprotease of V. vulnificus have been the focus of much research, but their roles in virulence remain unc lear. While experiments involving injection of the purified toxins (one or bot h) into animals are able to reproduce some of the disease symptoms (40,41), mutations in the genes do not significantly attenuate virulence in animal models (42-44). Nevertheless, the cyto lysin has been shown to be expressed in vivo, (45,46), suggesting that the in vitro cytoxicity seen in many studies may be involved in infection. Of two other hemolysins subsequently identified, hlyIII has been mutated and shown to contribute to virulence in i.p. inoculated mice (47). The protease of V. vulnificus possesses collagenase, elastase, caseinase, and zinc metalloprotease activ ities (48-50). Like th e hemolysin, injection of the metalloprotease causes pathology, but knockout mutants show no attenuation in virulence (43,44). Even a hemolysin/cytolysin-protease double mutant showed no attenuation in i.p. infected mice (51). A periplasmic nuclease was identified and the gene was mutated (52). The mutant had a 10-fold increased ability to uptake DNA but showed no attenuation in i.p. inoculated mice. Other extracellular enzymes expressed by V. vulnificus of both clinical and environmental origins may 22

PAGE 23

be possible virulence factors but have not been examined for vi rulence in relevant models. These enzymes include a mucinase, protease, lipase, chondroitinase, hyaluronidase, DNase, and sulfatase (53). Most recently, the presence of toxins of the RTX (r epeat in t ox in) family in V. vulnificus has received attention (32,33,54). RtxA1 appears to be the major cytotoxicity factor of V. vulnificus, and mutants show reduced virulence in mice (32,33,54). Studies in our lab showed that toxicity in cell culture assa ys does not necessarily correlate with virulence in our iron treated mouse model, as the same mutant that was si gnificantly inhibited in cell lysis, monolayer destruction, and tight-junction de struction in cell culture, was only moderately attenuated in mouse infections (Starks, Bourdage, and Gulig, unpublished). To date, the polysaccharide capsule remains the most important virulence factor of V. vulnificus, with unencapsulated strains being avirulent. Many of the other virulence factors contribute to infection to a lesser degree, with some having onl y marginal (10-fold or lower) effects. The virulence factors discovered thus far do not fully explai n the rapid growth and destructive abilities of V. vulnificus. Clearly, this subject warrants continued investigation into the means by which this opportunistic pat hogen causes such severe diseases. Focus of Investigation The goal was to identify nove l virulence factors of V. vulnificus Because bacterial virulence factors are involved in host-pathogen interactions, they are generally localized to the bacterial cell surface or are secret ed into the extracellula r space to interact with host targets. In light of this, we used PhoA fusion-insertion mutagenesis to target genes encoding secreted products. This mutagenesis approach, coupled with screening of mutants in the s.c inoculated, iron dextran-treated mouse model of V. vulnificus disease (38,55), could le ad to the discovery of new virulence-associated genes. Novel virulenc e factors identified by these studies will provide 23

PAGE 24

insight into the pathogenic mechanisms of V. vulnificus and perhaps lead to new chemotherapeutic strategies. Specific Aim 1. Develop and Implement a Sy stem for Identifying Secreted Virulence Factors of V. vulnificus PhoA mutagenesis has been extensively used for mutating bacterial genes encoding secreted products. Initial at tempts at PhoA mutagenesis in our laboratory were based on Tn phoA (56) but this system did not work well. We recreated mini-Tn phoA (57) for use in V. vulnificus We tested the ability of the P hoA mutagenesis system to genera te in-frame PhoA fusions in E. coli and V. vulnificus generated a pool of blue (PhoA+) mutants, analyzed virulence via the s.c. inoculated iron dextran-treated mouse mode l, and further characterized mutants that had attenuated virulence to identif y potential virulence factors of V. vulnificus Specific Aim 2. Characterize a fadR Mutant Identified by PhoA Mutagenesis and Investigate the Role of FadR and Fa tty Acid Metabolism in Virulence of V. vulnificus Sequencing of genomic DNA from PhoA mutant FLA602 revealed a backward insertion of phoA into the fadR gene that encodes a cytoplasmic fatty acid metabolism regulatory protein. Although FLA602 did not represen t an in frame fusion to phoA it was highly attenuated for infection and so warranted further study. V. vulnificus replicates extremely rapidly in the fatty s.c. tissues during skin infecti on. We hypothesized that the bacter ia use host lipids and free fatty acids as a growth substrate during infection, and that the inability to effectively coordinate fatty acid metabolism rendered the FadRbacteria unfit for both local and systemic infection via the s.c. route of inoculati on. We characterized the fadR ::mini-Tn 5Km2phoA mutant to confirm the FadRphenotype, made a targeted deletion of fadR from wild-type V. vulnificus and performed various assays to determine the reason for the attenuation of the fadR mutant. To investigate the potential role of utilization of fatty acids during infection, we deleted fadD that encodes the fatty acyl-CoA synthetase and aceA and aceB that encode the glyoxylat e bypass pathway enzymes. 24

PAGE 25

25 FadD is needed to activate exogenou s fatty acids as they enter the cell, allowing them to enter the fatty acid degradation pathway, while AceA and AceB are required for growth in fatty acids as a sole carbon source. Together with the fadR mutant, these mutations gave us new insight into the role of fatty acid metabolism in V. vulnificus infection. Specific Aim 3. Characterize an rseB Mutant Identified by PhoA Mutagenesis and Investigate the Role of RseB and Aspects of the Extracytoplasmic Stress Response in Virulence of V. vulnificus PhoA mutant FLA609 had an in-frame fusion of phoA to the gene encoding RseB, a periplasmic negative regulator of sigma E (E) activity. The rseB mutant exhibited several interesting phenotypes includi ng phase-variation between translucent and opaque colony morphologies. RseB negatively regulates E, an alternative RNA polym erase sigma factor that controls an extensive regulon involved in respond ing to cell envelope stresses. This response, termed the envelope (or extracytoplasmic) stress response (ESR), is essen tial for maintaining the envelope integrity of gram-negative bacter ia under certain stress conditions. Because rseB is involved in the ESR, the FLA609 mutation opene d the door to determine the role of the Emediated ESR in V. vulnificus. We also investigated the possible reasons for the translucent morphology of RseB variants by comparison with an acapsular translucent mutant of V. vulnificus. These studies have uncovered a possible ro le for RseB in causing phase variation of capsular polysaccharide, and have been the first to investigate the extracytoplasmic stress response in V. vulnificus

PAGE 26

CHAPTER 2 MATERIALS AND METHODS Standard Microbiological and Animal Infection Protocols Bacterial Strains, Media, and Chemicals Bacterial strains are listed in Table 2-1. All strains were grown in Luria-Bertani broth containing 0.85% (wt/vol) NaCl (LB-N) or on LB-N plates containing 1. 5% (wt/vol) agar. When required, antibiotics were used as follows: rifampin, kanamycin, chloramphenicol, and tetracycline at 50 g/mL, 40 g/mL, 30 g/mL, and 12.5 g/mL, respectively, for E. coli and 50 g/mL, 300 g/mL, 3 g/mL, and 6.25 g/mL, respectively, for V. vulnificus. LB-N containing 105 U/L colistin and VVM (58) were used to select against E. coli during conjugations. M9 minimal salts (59) containing 0.2% (wt/vol) glucos e or 0.01 % (wt/vol) fatty acids were used to assess auxotrophy and the ability of strains to use fatty acids as sole carbon sources. Strains were stored at -80C in LB-N with 35% (vol/v ol) glycerol. For mouse infection experiments, static overnight st arter cultures of bacteria were grown in culture tubes at room temperature. Before infection, starter cultures were diluted 1:20 into pre-warmed LB-N and shaken at 37C until the optical density at 600nm (OD600) reached 0.4 to 0.6 (exponential-phase growth). The bacteria were diluted in phosphate-buffered saline (PBS) containing 0.01% (wt/vol) gelatin (BSG) (60) to the appropriate inoculum for inf ection. Further dilution in BSG and plating were used to confirm CFU/mL inoculated. Unless otherwise noted, components of growth media were from Difc o (Franklin Lakes, NJ), chemicals were from Sigma (St. Louis, MO), DNA extraction and purification kits were from Qiagen (Valencia, CA), molecular gene tics enzymes were from New England Biolabs (Ipswich, MA), and oligonucleotides were from IDT (Coralville, IA). 26

PAGE 27

Use of pGTR201 for PhoA Mutagenesis PhoA mutagenesis plasmid pGTR 201 was electroporated into E. coli S17-1 pir for conjugation into V. vulnificus FLA399 by filter mating. Transconjugant s were selected on LB-N agar plates containing 50 g/mL rifampin, 300 g/mL kanamycin, 40 g/mL 5-bromo-4-chloro3-indolyl phosphate (BCIP), and 0.2% (wt/vol) gl ucose. VVM-agar was used to confirm that colonies were V. vulnificus Colonies appearing bluer than FLA399 on BCIP-containing plates were considered PhoA+. After FLA600 and FLA601 were c onfirmed to have in-frame PhoA fusions, these strains were used as positive controls for blue color. Genomic DNA was extracted (DNeasy Blood & Tissue Kit, Qiagen) from PhoA+ strains for sequencing using primer phoA5rev at the University of Florida Interd isciplinary Center for Biotechnology Research DNA Sequencing Core Laboratory. Measurement of Growth For measurement of growth rate in rich brot h, static overnight cult ures were diluted 1:20 into prewarmed LB-N and grown shaking at 37 C to exponential phase. Exponential-phase cultures were diluted to approximately 105 CFU/mL in 20 mL of LB-N Aliquots were removed for measurement of OD600 immediately and every 30 min thereafter for up to five hours. To determine if strains could use glucose or fatty acids as a sole carbon source, static overnight cultures in M9-glucose were pelleted, wash ed once in M9 minimal medium without carbon source, and suspended in 10 mL of M9, M9 + 0.2% (wt/vol) glucos e, M9 + 0.01% (wt/vol) sodium decanoate (Sigma-Aldrich Inc., St. Louis, MO), or M9 + 0.01% (wt/vol) sodium oleate (Sigma-Aldrich). Turbidity afte r overnight incubation at room te mperature indicated growth in the specified medium, and dilutions were plated on LB-N to verify viable CFU. 27

PAGE 28

Fatty Acid Analysis Wild-type CMCP6 and fadR mutant FLA614 were grown to mid-exponential phase in 60 mL LB-N broth. Cultures were pelleted by cen trifugation at 8,000 g for 10 minutes at 23C, resuspended in 1 mL LBN in sterile 1.5 mL tube s and centrifuged again. Pellets were weighed and adjusted to approximately 50 mg. Fatty ac id analysis was done by gas chromatography of fatty acid methyl esters according to the Sherlo ck system (MIDI, Inc., Newark, DE) at the University of Florida Bacterial Identification & Fa tty Acid Analysis Laboratory. Four separate cultures of each strain were analyzed. Cerulenin MIC A modification of the method of Ca mpbell and Cronan (61) was used. Overnight cultures of FLA602, CMCP6, and FLA602 (pGTR349) were diluted 1:200 into LB-N. Cerulenin (MP Biomedicals, Inc.) was dissolved in 100% (vol/v ol) ethanol at 1 mg/mL an d was added to culture tubes containing 1 mL of the diluted bacterial cu ltures to achieve concen trations of 20, 15, 10, 5, 2.5, 2, and 1 g/mL. A tube with bacterial cultu re without cerulenin was used as a positive control for growth. The tubes were incubated shaking at 37 C overnight. The MIC was the concentration of cerulenin that resulted in no visible growth. Assays of Membrane Stress Sensitivity The minimum inhibitory concentrations of ethanol, sodium dodecyl sulfate (SDS), and polymyxin B were tested essentially was done for cerulenin (above). For ethanol, solutions of 6, 5, 4, 3, 2, and 1 % (vol/vol) were made using 95% ( vol/vol) ethanol in LB-N. Concentrations of SDS were 0.5, 0.45, 0.4, 0.35, 0.3, 0.25, 0.2, 0.15, and 0.1 (wt/vol) in LB-N. Polymyxin B concentrations were 350, 300, 250, 200, 150, and 100 (U/ml) in LB-N. Alternatively, disk diffusion tests were done to observe inhibition by membrane perturbing agents. Eight millimeter discs of Millipore Abso rbent Pads were aseptically placed over lawns 28

PAGE 29

of 107 bacteria plated on LB-N agar. Aliquots of i nhibitory agents were added to the discs as follows: 20 L of 0.1% SDS, 40 L of 100% ethanol or 40 L of 3% hydr ogen peroxide. Plates were incubated overnight at 37C, and the diameter of the zone of inhibition surrounding each filter was measured. At least three filters were used for each strain, and at least two separate assays performed. Serum Sensitivity Overnight starter cultures grow n standing at room temperatur e were diluted 1:20 in LB-N and grown to OD600 0.4 to 0.6. Culture volumes were adjusted to 108 CFU in 1mL and serially diluted to 103 CFU/mL for plating to quantify input CF U. An aliquot of 20 L from the 10-1 dilution (107 CFU/mL) was incubated with 180 L of unt reated or heat-inac tivated (30 min at 56C) rat serum for 2 h, then plated to quantify survival. The ratio of log transformed CFU in untreated versus heat-inactivated serum was reported. At least tw o separate assays performed. Motility Static overnight cultures were grown in 5 ml LB-N at room temperature. A sterile motility loop (straight wire) was dipped in the culture and used to stab a 0.3% low agar LB-N plate. Plates were incubated 15 to17 h at 37C, and mo tility was measured by the diameter of spread from the center of the plate. Each strain wa s assessed on triplicate plates in at least two independent experiments. Biofilm Formation Assay Strains were grown standing ove rnight on the bench in 5 mL LB-N and diluted 1:100 into fresh LB-N broth. Aliquots of 200 L were added to wells of a clear, op tical, non-tissue culturetreated 96-well plate (7 wells per strain). A se ries of wells with LB-N only was included as blanks. The plate was incubated for 48 h at 30 C without agitation. Media and planktonic cells were removed by pipetting. The wells were then washed once in PBS, and the plate was allowed 29

PAGE 30

to dry for several hours on the bench. Wells were stained with 200 L of 1% (wt/vol) crystal violet (CV) for 15 min, and the CV was removed by pipetting gently. The plate was then washed by gentle immersion in a dish containing dist illed water and blotted on paper towels until no more CV appeared on the towels (usually two to three washes). To solubilize and quantify crystal violet staining the biofilms, 200 L of 95% ethanol were added to each well and allowed to stand for 15 min. The contents of each well were removed to a fresh 96-well plate, and the absorbance was read at a wavele ngth of 630 nm on a plate reader. Extraction of EPS A modification of the method of Enos-Berlage and McCarter (62) was used. An entire petri plate was inoculated with a single bacter ial colony and incubated overnight at 37C. The lawn of bacteria was scraped from the plate and suspended in 5 ml of PBS with vigorous vortexing for one min. A 20 L aliquot was of this suspension was diluted and plated to quantify starting CFU. The remaining suspension was shak en at 200 rpm on a rotary shaker for 1.5 h at 37C. The vortexing and shaking process was re peated, and cells and de bris were removed by centrifugation at 10,000 g for 15 mi n at room temperature. Th e supernatant containing the EPS was removed to a 50 mL conical tube. RNaseA (50 g/mL), DNaseI (50 g/ml), and MgCl2 (10mM) were added, and the tube was inc ubated for 8 h at 37 C Proteinase K (200 g/mL) was added and incubated a further 17 h at 37C, and 750 L (1/15 of the total volume) were removed and extracted twice with phenol-chloroform using a Phase-lock Heavy Gel tube (Eppendorf, Westbury, NY). The sample was pr ecipitated with 2.5 volumes of 95% ethanol, centrifuged at 8,000 g for 30 min at 4C, and the supernatant was decanted. The pellet was washed with 3 mL of 95% ethano l, air dried, and suspended in 200 L of water. Samples were stored at -20C. 30

PAGE 31

Analysis of Whole Cell Lysates or EPS Extracts by SDS-PAGE Whole cell lysates were prepared from exponential-phase bacteria grown shaking in LB-N at 37C. An aliquot of each culture was diluted an d plated to confirm the starting CFU. A volume containing 108 CFU was centrifuged at 8,000 g for 5 min. Pellets were washed twice in PBS and suspended in 100 L Laemmli sample buffer (LSB). 10 L (107 CFU) of whole cell lysates or 25 L (1/120 of starting volume) of EPS extract diluted 1:1 with LSB were run on Criterion 420% Tris-HCl Gels (Bio-Rad) for 1.5 h at 100V. Staining of Gels wi th Alcian Blue After SDS-PAGE, gels were fixed overnight at room temperature in fresh Alcian blue fixing solution (30% ethanol + 10% acetic acid). Gels were stained with fresh Alcian Blue (0.2% (wt/vol) Alcian blue, 40% (vol/vol) ethanol, and 10% ( vol/vol) acetic acid) for 1 to 3 hours with gentle rocking and we re destained with a solution containing 40% ethanol and 10% acetic acid. Several changes of destain solution were used ov er 2 hours to remove background staining. Gels were scanned using an Epson 1670 scanner for image capture. Staining of Gels with Stains-All Staining was done according to the method of Ke lley and Parker (63). Before staining, gels were fixed overnight in 25% (vol/vol) isop ropanol in 3% (vol/vol) acetic acid to remove SDS. Gels were stained overnight in the dark with a working solution of stain that contained 10 mL of stock, 10 ml of formamide, 30 mL of isopropanol, 1 mL of 3M Tris-hydrochloride (pH 8.8), 0.1 ml of 2-mercaptoethanol, and deionized distilled water to 200 ml. The stock solution was made fresh and contained 10 mg of Stains-A ll in 10 ml of formamide. The gels were destained slowly with 15% isopropanol in 0.1 M Na2HPO4 (pH ~ 9.2) plus 200 L of 2mercaptoethanol per 400 ml of solution. Gels were scanned using an Epson 1670 scanner for image capture 31

PAGE 32

Infection of Mice We used the s.c. inoculated, iron-de xtran treated mouse model of Starks, et al. (38,55). Seven to 10-week-old female ICR mice (Harla n Sprague-Dawley, Indian apolis, IN) housed under specific-pathogen-free conditions were used. Mice were injected intraperitoneally with 250 g of iron dextran (Sigma-Aldrich) per gr am of body weight at least 45 min before inoculation. Inocula consisting of 0.1 mL of bacteria suspended in BSG were injected s.c. in the lower back. Mice were euthanized by carbon dioxide asphyxiation 15 to 22 hours postinoculation or when rectal temperatures dropped below 33 C, indicating that mice were moribund. After euthanization the skin was peeled back to reveal the s.c. lesions. Samples of skin lesions and liver were aseptically removed from mice, homogenized in 5 mL of BSG using glass tissue homogenizers, diluted, and plated on LB-N agar. Strains bearing expression plasmids were plated both nonselectively (LB-N) and selectively (LB-N with 6.25 g/mL tetracycline.) Samples were not taken from mice with no visibl e skin lesion, and minimum detectable CFU/g was used for these mice for statistical analysis. The minimum detectible CFU/g for skin was 104 CFU/g and for liver was 102.5 CFU/g. Molecular Genetic Techniques and Analyses Construction of mini-Tn 5Km2phoA Mutagenesis Vector, pGTR201. phoA was PCR-amplified from the vector pRT291 (64) using Taq polymerase and primers engineered to insert Not I (underlined) sites fla nking the product: PhoA3Not I-2 and TnPhoA5NotI-3. The PCR product was cloned into the pCR2.1-TOPO vector according to the TOPO TA Cloning Kit instructions (Invitr ogen) and electroporated into E. coli TOP10 for blue-white screening using 40 g/mL 5-Bromo-4-chloro-3-indolyl -D-galactoside (X-Gal) (SigmaAldrich). Plasmid mini-extracts (QIAprep Spin Miniprep Kit, Qiagen Inc., Valencia, CA) of the 32

PAGE 33

resulting clones were digested with Not I to excise the phoA fragment. The 1.3-kb phoA fragment was purified by gel-ex traction (QIAquick Gel Extraction Kit, Qiagen Inc.) and ligated to Not I-digested pUTmini-Tn 5Tag3 (65), a derivative of pUTmini-Tn 5Km2 (57). The ligated product was electroporated into E. coli EC100D for selection on LB-N plates containing 40 g/mL kanamycin Plasmid mini-extracts of the resulting clones were resolved on agarose gels to confirm correct size, and restriction digest of the plasmids further confirmed correct insertion of phoA. USER Friendly Cloning USER (Uracil-Specific Excision Reagent) Friendly Cloning (66-68) (New England Biolabs, Inc., Ipswich, MA) is a method for eas ily cloning PCR products. Briefly, a target was PCR amplified using primers designed with 8 additional nucleotides at their 5 ends including an internal deoxyuridine (dU). Th e PCR product was treated with the USER Enzyme, a mixture of uracil DNA glycosylase that removed the uracil residue and DNA glycosylase-lyase Endonuclease VIII that nicked the deoxyuridine-cont aining strand thereby releasing the terminal single-stranded DNA fragment and creating an 8-bp 3' overhang th at was complementary to the vector. The resulting PCR fr agment was flanked by 8-nucleot ide long 3 single-stranded extensions. A capture vector was created to have an 8-nucleotide l ong 3 single-stranded extension on each end when linearized. Each ex tension was complementary to one end of the PCR product, allowing for assembly into a reco mbinant molecule. Our use of USER Friendly cloning will be described in detail elsewh ere (Gulig, et al, in preparation). pGTR1129, a USER capture vect or constructed using the allelic exchange vector pCVD442 (69) as a backbone, was used to clone a USER PCR product according to the USER Friendly Cloning procedure. Briefly, 10 L of PCR product was incubated with 20 ng of USER vector and USER enzyme at 37 C for 15 minutes, followed by the addition of T4 DNA ligase 33

PAGE 34

and T4 DNA ligase buffer and incubation at room temperature for 15 minutes. The reaction was cleaned using a DNA Clean & Concentrator K it (Zymo Research Corporation, Orange, CA) and electroporated into EC100D pir+ for blue-white screening on LB-N plates containing chloramphenicol (30 g/mL) and X-Gal (40 g/mL). Plasmid mini-extracts of the resulting clones were resolved on an agarose gel to confir m the correct size. The clone was electroporated into E. coli S17-1 pir for conjugation into V. vulnificus FLA602 by filter-mating. Transconjugants were selected on LB-N plates containing 3 g/mL chloramphenicol and 105 U/L colistin. As an alternative to conjugations, th e clone could be directly used for chitin-induced transformation of V. vulnificus 3-Way USER Deletion Cloning To clone upstream and downstream sequences flanking the DNA sequence to be deleted, a modification of USER cloning was used. USER cap ture plasmids were prepared as above, and oligonucleotide primers were designed to pr oduce 500-bp PCR amplicons flanking the deletion target, with compatible ends for ligation into the USER vector. The ends of the upstream and downstream fragments that were to be joined together were desi gned to have compatible USER ends with a Sma I restriction enzyme site (Table 2-2). The PCR products were cleaned using a PCR purification kit (Qiagen), and 5 L each of upstream and downstream PCR product were combined with approximately 20 ng of USER vector and incubated for 15 min at 37 C with the USER enzyme. The remainder of the USER cl oning protocol (above) was followed exactly. To allow for selection of the deletion construct, an antibiotic resistance cassette was added. The deletion plasmid was digested with Sma I to linearize between th e upstream and downstream PCR products. A blunt antibio tic cassette (in most cases, aph encoding kanamycin resistance) was ligated into the Sma I site. Correct size of the resulting construct was conf irmed by resolving 34

PAGE 35

on an agarose gel, and PCR was used to confirm the presence of insert. The plasmid was then sequenced to verify co rrect DNA sequences. Chitin-Induced Transformation A modification of the method of Meibom, et al. (70) was used. Cultures were grown to exponential phase in LB-N at 37C. Bacteria were harvested by centrif ugation, pelleted, washed with an equal volume of diluted (0.75) seawater, and resuspende d in twice the original volume of 0.75 seawater. Two milliliter aliquots were added to wells of a 12-well tissue culture plate containing sterile pieces of crab she ll, and the plate was incubated without agitation at 30C for 18-24 hrs. The growth medium was removed from the pl ate and replaced with 2 mL of fresh 0.75 seawater, followed by 2 g of sheared or linearized DNA. After 24h growth at 30C, the crabshells were removed to 50 ml conical tubes containing 2 ml PBS and vortexed vigorously for 30 sec. The crabshell culture was diluted a nd plated on LB-N containing an appropriate antibiotic for selection of transformants. qRT-PCR The Qiagen RNeasy Mini Kit (Qiagen, Inc.) wa s used to isolate RNA from mid-log phase cultures (OD600 0.4 to 0.6) grown in LB-N. The RNA wa s treated with DNaseI using the Qiagen RNase-Free DNase Set during RNA extraction, and reverse-transc ribed using the iScript cDNA Synthesis Kit (Bio-Rad Laboratories, Inc.) To check for DNA contamination, purified RNA without reverse transcriptase was used as a negati ve control. A standard curve was plotted for each primer set using known quantities of purif ied PCR products from CMCP6 genomic DNA. iQ SYBR Green Supermix (Bio-Rad) was used for detection via the iQ Real-Time PCR Detection System (Bio-Rad). The PCR reactio ns were done in tripli cate, and PCR cycling conditions were 1 cycle at 95 C for 30 sec; 40 cycles at 95 C for 10 sec, 60C for 45 sec; 100 35

PAGE 36

cycles at 60C for 10 sec, and a 4C hold. Re lative expression was determined by calculating 2Ct using the 16S rRNA gene as an internal control. Statistical Analysis Students t -test was used to examine significant differences between pairs of means. Groups of more than two means were analyzed by ANOVA followed by Bonferronis post test to identify significant differe nces between the groups. 2 tests were used in mouse experiments to determine if the number of mice with detectable CFU was significantly changed in mutant versus wild-type infections. Statistical analyses were done using Microsoft Excel or GraphPad Prism5 software. Values were considered significant for P 0.05. All quantitative experiments were repeated at least once. 36

PAGE 37

Table 2-1. Bacterial stra ins used in this study Strain Relevant Characteristics Reference or Source E. coli TOP10 F mcrA ( mrr -hsdRMS -mcrBC) 80lacZ M15 lacX74 recA1 araD139 (ara-leu )7697 galU galK rpsL (Strr) endA1 nupG Invitrogen S17-1 pir pir lysogen; thi pro hsdR hsdM+ recA RP4-2 Tc::MuKm::Tn 7 (Tpr Smr) (71) EC100D pir + F mcrA (mrr -hsdR MSmcrBC) 80dlac Z M15 lacX74 recA1 endA1 araD139 (ara leu )7697 galU galK rpsL nupG pir+ (DHFR) Epicentre V. vulnificus CMCP6 Clinical isolate (31) FLA399 Spontaneous Rif r derivative of CMCP6 This study FLA600 FLA399 dctQ ::mini-Tn5Km2phoA This study FLA601 FLA399 ppiC ::mini-Tn5Km2phoA This study FLA602 FLA399 fadR ::mini-Tn 5Km2phoA This study FLA603 FLA399 cvpA ::mini-Tn 5Km2phoA This study FLA604 FLA399 with unsequenced mini-Tn 5Km2phoA insertion This study FLA605 FLA399 VV1_2399::mini-Tn5Km2phoA This study FLA606 FLA399 with unsequenced mini-Tn 5Km2phoA insertion This study FLA607 FLA399 VV1_0641::mini-Tn5Km2phoA This study FLA608 FLA399 ptsG ::mini-Tn 5Km2phoA This study FLA609 FLA399 rseB ::mini-Tn 5Km2phoA This study FLA610 CMCP6 rseB This study FLA611-R Spontaneous rugose isolate of FLA399 This study FLA612 FLA602 reverted to wild-type by allelic exchange This study FLA614 CMCP6 fadR This study FLA1000 CMCP6 ptsI ::mini-Tn5Km2phoA Donoso and Gulig, unpublished FLA1001 CMCP6 rpoE This study FLA1002 CMCP6 degP This study FLA1003 CMCP6 aceAB This study FLA1006 CMCP6 fadD This study CMIT232 MO6/24-O wzA ::Tn phoA (24) FLA1009 CMCP6 wzA ::Tn phoA via chitin transformation with genomic DNA from CMIT232 This study FLA1012 CMCP6 rseA This study FLA1013 CMCP6 cvpA ::mini-Tn 5Km2phoA via chitin transformation with genomic DNA from FLA603 This study 37

PAGE 38

Table 2-2. Plasmids used in this study Plasmid Relevant Characteristics Reference or source pRT291 IncP1, Tn phoA Km r Tc r source of 'phoA (64) pUTminiTn 5Tag3 mini-Tn 5Tag3 delivery vector; R6K ori, mob RP4 Ap r Kmr (65) pGTR201 pUTmini-Tn 5 Tag3 phoA ( phoA delivery vector ) This study pCR2.1 T/A cloning vector, lacZ multiple cloning site; Ap r Kmr Invitrogen pRK437 Expression vector, mob RK2, lacZ multiple cloning site, Tcr (72) pGTR2000 cvpA cloned into pRK437 for complementation of FLA603 and FLA1013 This study pGTR349 fadR cloned into pRK437 for complementation of FLA602 This study pCVD442 R6K ori-based suicide plasmid, mob RP4, sacB Ap r (69) pGTR1129 pCVD442 with lacZ from pUC19 with USER Friendly cloning oligonucleotide linker incorporated, cat Gulig, et al., in preparation pGTR2007 fadR USER-cloned into pGTR1129 for reversion This study pGTR2009 500 bp upstream and downstream of fadR USER-cloned into pGTR1129 for deletion of fadR This study pUC4K Km r derivative of pUC4 (73) pGTR2010 pGTR2009 with aph from pUC4K cloned into Sma I This study pGTR2017 500 bp upstream of aceA and downstream of aceB USERcloned into pGTR1129 for deletion of aceAB This study pGTR2018 500 bp upstream and downstream of fadD USER-cloned into pGTR1129 for deletion of fadD This study pGTR2019 pGTR2017 with aph from pUC4K cloned into Sma I This study pGTR2020 pGTR2018 with aph from pUC4K cloned into Sma I This study pGTR1160 pRK437 with USER Friendly cloning oligonucleotide linker incorporated Gulig, et al., in preparation pGTR2005 rseB cloned into pGTR1160 for complementation of FLA609 and FLA610 This study pGTR2006 500 bp upstream and downstream of rseB USER-cloned into pGTR1129 for deletion of rseB This study pGTR2008 rpoE cloned into pGTR1160 for expression from the lac promoter This study pGTR2011 500 bp upstream and downstream of rpoE USER-cloned into pGTR1129 for deletion of rpoE This study pGTR2013 pGTR2011 with aph from pUC4K cloned into Sma I This study pGTR2014 500 bp upstream and downstream of degP USER-cloned into pGTR1129 for deletion of degP This study pGTR2015 pGTR2014 with aph from pUC4K cloned into Sma I This study pGTR2021 500 bp upstream and downstream of rseA USER-cloned into pGTR1129 for deletion of rseA This study pGTR2022 pGTR2021 with aph from pUC4K cloned into Sma I This study 38

PAGE 39

Table 2-3. Oligonucleotides used in this study Oligonucleotide primer Sequence 5 3 a For cloning and deletion TnPhoA5-NotI-3 CGCGGCCGC CCTGTTCTGGAAAACCGGGCTGC PhoA3-NotI-2 GGCGGCCGC GGTTTTATTTCAGCCCCAGAGCGGC vv1_1996-5USER GGAGACAU GCTCAAATTTAAACCACTAAACCC vv1_1996-3USER CGGGAAAGU GACAGTAGCTGGCACCGCTAG vv12233-5-rbs CGGATCC TGAGTGCCATTCGACCCAAAAC vv12233-3 GGGATCC GTCAATTATTAGCTATTAGCAGTCG FadR5USER GGAGACAU GACGACTTCCAGATTCCGCAA FadR3USER GGGAAAGU CTATTAGCAGTCGTCTTCTGTG sacB5-2 AAGTTCCTGAATTCGATTCGTCC sacB3-2 CCTTTCGCTTGAGGTACAGCG delfadRup5 GGAGACAU CTTGCCAAGTTACTTCCCTTGAA delfadRup3 ACCCGGGU GGCTCTTTGCCTTAATGACCATT delfadRdn5 ACCCGGGU AGACGACTGCTAATAGCTAATAATT delfadRdn3 GGGAAAGU CCGATCCCGCGACCTTCTTG AceAUp5USER GGAGACAU GGTGACGCCAAAAATATGATGGA AceAUp3SmaI-USER ACCCGGGU TACCTTTCCATCTTTAGGTATTAAC AceBDn5SmaI-USER ACCCGGGU TCTTCTTCCTTTCATTTTGCTTTAC AceBDn3USER GGGAAAGU ATTTCACATAACTTTCTTAATGAAGC aceABinternal5 CCAGAGCTTCGATTTGCTGGC aceABinternal3 CAACTTCTTAACCGTACCGGG fadDup5USER GGAGACAU CCGTGCGCGACTCACTCTTAT fadDup3SmaI-USER ACCCGGGU AGCCGCGCAAAGTGTCACGAA fadDdn5SmaI-USER ACCCGGGU CGATGTTACTCCTCGTTTAAAGC fadDdn3USER GGGAAAGU CCGTAATCTCAGTAGTTTTGATGA rseB5USER GGAGACAU GGATCCGACTCATCTCATATTGCAGGTGA rseB3USER GGGAAAGU GGATCCTCACTTGGGTGACGGTGGCG delrseBup5USER GGAGACAU CAACAATGGTGAATTAGAATGGC delrseBup3USER-SmaI ACCCGGGU TTTCTTCATTCAATTTCCGATGTAG delrseBdn5USER-SmaI ACCCGGGU GCAATGATGACCGCACTCGC delrseBdn3USER GGGAAAGU GGCATTTCAGTTAAGTTGGCAGA SigmaE5rbs-USER GGAGACAU CCATATCTGAGATATGTGGAGCA SigmaE3rbs-USER GGGAAAGU TCACCATTGTTGTTACTGGTACTA delSigmaEup5USER GGAGACAU CAGCGATACCACCTTGAGCATA delSigmaEup3 SmaI-USER ACCCGGGU TCAACTGCTCGTTCATTCGAGC delSigmaEdn5 SmaI-USER ACCCGGGU TAGTACCAGTAACAACAATGGTGA delSigmaEdn3USER GGGAAAGU CTGGCTCAGCACTGCCGGC sigE out5 CCACCGGTCGCGAGCACGA sigE out3 CGCGTTCATGCTATGTTACGTG sigE internal5 GAGCGAGTTCAGAGTGGCGAT sigE internal3 CCGTGCTCGTGAGGCGGTG degPup5USER GGAGACAU GAGCAACAACGCCAAGAGCTG degPup3USER-SmaI ACCCGGGU ACAATGCAGTCAAAACAAGCAAAG degPdn5USER-SmaI ACCCGGGU CGAGCATACAAAGGGCAGTCC degPdn3USER GGGAAAGU AATGGCGATGTCAGTGCGACG rseAup5USER GGAGACAU CGAGCAGTTGACCGATCAAGTA rseAup3USER-SmaI ACCCGGGU TTCTAATTCACCATTGTTGTTACTG rseAdn5USER-SmaI ACCCGGGU CAGGTGAGCAATCTACATCGGA rseAdn3user GGGAAAGU GGCTTCACGTGCTCGGCCG 39

PAGE 40

40 Table 2-3. Continued Oligonucleotide primer Sequence 5 3 a CMCP6wza5 CTTAATTTAGACGATTTGGCTTAC CMCP6wza3 CGGCCACGTTTTAACGCGTAG CMCP6wzc3 GGAGATACTGAGGAGATAACAG For qRT-PCR fabA fwd TTTGACTGCCATTTCCCTGG fabA rev AACTGCCACATCGCATCCAA fabB fwd TGTTTGCAGGTGGTGGTGAA fabB rev TATTTGGTTGACAGTGCGCC fadB fwd TCGTTGCTTACGCAGCCAAA fadB rev AGCACGCGGTTCACAAAGAA fadD fwd TGATGCCAAACCTGCTGCAA fadD rev TCACAGCAATACAGCCAGCA 16S fwd TCGTCAGCTCGTGTTGTGAA 16S rev ACTCGCTGGCAAACAAGGAT degPfwd TGAGATGTCCGATGTTGCTCTGCT degPrev AACTTGTCGGAATCGGCGAGCTTA sigmaEfwd TGACTCTACGAGAGCTTGATGGCT sigmaErev ATACGCGAACGTACCGTTCCTACA aRestriction sites are underlined and US ER cloning sequences are bolded.

PAGE 41

CHAPTER 3 CONSTRUCTION AND TESTING OF A PHOA MUTAGENESIS SYSTEM FOR VIBRIO VULNIFICUS AND IDENTIFICATI ON OF MUTANTS Rationale for Study Because only a few virulence f actors had been identified for V. vulnificus, we proposed to study the pathogenesis of this bacterium with the goal of identifying and characterizing previously unknown virulence determinants. Specifically, we intended to focus on those factors that are exported into and beyond the cell memb rane of a pathogen, considering that most virulence factors are exported. PhoA mutagene sis was used because it is a powerful tool for identifying genes encoding exported products (5 6). Initial attempts at PhoA insertion mutagenesis in our lab were based on Tn phoA (56), but Tn phoA did not work well in our system. The failure of Tn phoA to consistently produce screenable phoA fusions in V. vulnificus led to the development of a new delivery system for phoA. We modified the Tn 5-based approach using a min-Tn 5 transposon delivery system that works well in V. vulnificus Introduction Alkaline phosphatase, the product of the phoA gene is a scavenger enzyme that cleaves phosphate groups from large molecules in the peri plasm for use in cytoplasmic reactions (74). Two factors make PhoA a useful t ool to researchers. The first fact or is its obligate transport out of the cytoplasm to the periplasm for the enzyme to be active. The enzyme must dimerize by disulfide bond formation, which only occurs in the oxidizing environment of the periplasm where disulfide isomerase enzymes reside. The sec ond factor is the availab ility of a simple assay for PhoA activity the chromogenic compound BCIP turns blue upon cleavage by alkaline phosphatase (75). E. coli alkaline phosphatase was originally used to study protein secretion (74) and membrane topology (56,76). TnphoA (56) based on the broad host range transposon Tn 5, was 41

PAGE 42

the first transposon-based delivery system for phoA The premise for PhoA mutagenesis is that a truncated alkaline phosphatase gene lacking its promoter a nd secretion signal sequence ( phoA ) could be inserted randomly into to genes. If phoA inserts in frame with a gene encoding an exported product, the secretion-deficient phoA can use the secretion signal of the disrupted gene for export beyond the cytoplasm where alkaline phospha tase becomes active. The availability of the chromogenic substrate BCIP made PhoA mutage nesis the tool of choice for mutating genes encoding secreted products, as bacteria could be easily screened for alka line phosphatase activity by observing their blue color on BCIP-containing media. Tn phoA has been widely used to detect genes that code for exported proteins in many gram-negative organisms, and modified systems have been created for gram-positive bacteria (77) This transposon-based mutagenesis approach showed particular promise in the study of bacterial pathogenesis as most pathogenic determinants are transported beyond the cytoplasm to locations where they can readily interact with the host. De Lorenzo and coworkers (57) created a series of mini-Tn 5 transposon derivatives for insertion mutagenesis, promoter probing, and chromosomal insertion of cloned DNA in gramnegative bacteria. One of these derivatives, mini-Tn 5phoA could be used for PhoA mutagenesis. Because we had previously had success using other mini-Tn 5 derivatives in our laboratory, we requested a plasmid containing mini-Tn 5phoA from this group. Because we were unable to obtain this plasmid, we essentia lly recreated it by inserting phoA from Tn phoA (56) into pUTmini-Tn 5Tag3 (65). 42

PAGE 43

Results Development of a System for Iden tifying Secreted Products of Vibrio vulnificus Using Alkaline Phosphatase (PhoA) Mutagenesis. A truncated alkaline phosphatase gene lacking its start codon and secretion signal sequence ( phoA ) was PCR amplified from Tn phoA using oligonucleotides Tn PhoA5-NotI-3 and PhoA3NotI-2 that were designed to incorporate flanking Not I sites to facilitate cloning of this fragment into suitable vectors. This phoA sequence would have to be insert ed in frame into a gene that encodes a secreted product to be expressed and active. The phoA gene fragment was cloned the conjugative suicide plasmid pUTmini-Tn 5Tag3 (65). It should be noted that mini-Tn 5Tag3 (65) is mini-Tn 5Km2 (57) with an additional oligonucleotide tag for use in signature-tagged mutagenesis. pUTmini-Tn 5Tag3 also contains the oriT of plasmid RP4 (71) for conjugal transfer, the pirdependent oriV of plasmid R6K (78), aph in the mini-transposon for selection, and a transposase that is unlinked to the transp oson to prevent continuing transposition in the host after the donor plasmid has been lost. After ligation of phoA into pUTmini-Tn 5Tag3, the correct phoA insert size and orientation were conf irmed by restriction analysis and the phoA insert was sequenced using a primer facing outward from the O (5) end of mini-transposon (phoA5Rev). The resulting plasmid construct, pUTmini-Tn 5Km2phoA was named pGTR201 and was used for PhoA mutagenesis in V. vulnificus (Figure 3-1 .) As indicated above, this is essentially the same construct as pUTmini-Tn 5phoA (57) that we requested but were unable to obtain Testing the PhoA Mutagenesis System in E. coli Before PhoA mutagenesis was attempted in V. vulnificus, which poses the potential screening problem of endogenous PhoA activity and had exhibited problems with Tn phoA we tested the ability of the pGTR201 to be mob ilized, transpose, and produce screenable fusion43

PAGE 44

insertion mutants in alka line phosphatase-negative E. coli DHB4 (76). A spontaneous rifampinresistant derivative of DHB4, UF L011, was isolated to facilitate selection after conjugations. pGTR201 was electroporated into E. coli S17-1 pir that produces the protein ( pir gene product) that permits replication from the R6K or igin and encodes RP4 c onjugative functions for mobilization of pGTR201. pG TR201 was mobilized into E. coli UFL011 by conjugation with S17-1pir(pGTR201). UFL011 does not encode pir, thereby preventing replication of pGTR201, and is kanamycin-sensitive, thus allowing for selection of successful transconjugants and transposition events on LB-N containing kana mycin and rifampin. Because the recipient UFL011 was also PhoA-, identification of phoA fusion-insertion mutants simply involved observing their distinct blue colo r on LB-N agar containing BCIP. After conjugating S17-1 pir(pGTR201) with UFL011, blue co lonies were observed at a frequency of about 1 in 500. Two blue transconjugan ts were selected to verify the insertion of the phoA in frame in a gene encoding a secreted protein. Genomic DNA was isolated from the two mutants, and the presence of the mini-trans poson was confirmed by PCR analysis using the same primers originally used to amplify the phoA fragment. To sequence the phoA fusion junction and to determine the site of the insert ion, a region of genomic DNA from the mutants encompassing the mini-Tn 5Km2phoA and some portion of the mutated gene was needed. Genomic DNA was digested with Kpn I which cuts once at the end of the mini-transposon and presumably again in the genome. Th e genomic fragments were cloned into Kpn I-digested pUC19 selecting for kanamycin resistance and se quenced using a primer facing outward from the O end of the mini-transposon (phoA5rev). Se quencing was done at the University of Florida ICBR DNA Sequenc ing Facility. 44

PAGE 45

Sequence analysis of the clones from the two mutants revealed phoA insertions in a single ORF. The first was mglB encoding a periplasmic galactose-bi nding transport prot ein involved in the uptake of galactose. The second was ompP encoding an outer membrane protease. Analysis of the DNA sequences revealed that phoA had indeed inserted in frame with each gene in the correct orientation for transcription. Because both blue colonies revealed correct phoA fusions, we were confident in the ability of pGTR201 to generate screenab le in-frame fusions to genes encoding secreted proteins. Testing the PhoA Mutagenesis System in V. vulnificus The encouraging result in E. coli prompted us to begin phoA mutagenesis of V. vulnificus. V. vulnificus FLA399, a spontaneous rifampin-resistant derivative of clinical isolate CMCP6, was used as the parent. The complete genom ic sequence of CMCP6 is known, allowing easy identification of mutated genes upon sequenc ing. pGTR201 was introduced into FLA399 by conjugation from S17-1 pir(pGTR201), and the transconjugants were selected on media containing kanamycin (300 g/mL), rifampin ( 50 g/mL) and BCIP (40 g/mL). Due to the endogenous PhoA activity of V. vulnificus it was difficult to identif y PhoA mutants based on blue color on BCIP-containing ag ar. Glucose had been used in our lab and others to help suppress the endogenous PhoA activity of V. vulnificus ((24) and Qiu Y, personal communication), although the mechanism of acti on is unknown. Plating the transconjugants on media containing BCIP and 0.2% (wt/vol) glucos e caused a reduced plating efficiency and decreased colony size. V. vulnificus does not grow well on glucose-containing media, possibly due to glucose shock, a term coined for the inhibitory effect glucose on V. cholerae (79). Nevertheless, the addition of glucose to BCIP-c ontaining media made it much easier to identify blue colonies, thus reducing the incidence of false positives (Figure 3-2). The frequency of blue colonies was in the range of 1 in 1,000. Blue transconjugants were single-colony passaged and 45

PAGE 46

tested on the V. vulnificus -specific medium VVM (58) to ensure that they were not spontaneous rifampin-resistant E. coli S17-1pir. Three blue V. vulnificus mutants, FLA600, FLA601, and FLA602, were chosen for initial study. The presence of the phoA insert was verified by PCR using oligonucleotide primers targeting phoA As a screen for virulence, FLA600, FLA601, and FLA602 were tested in an iron dextran-treated mouse model of V. vulnificus infection, as detailed in the Experimental Design section, at an inoculum of 1,000 CFU in two mice each. The minimum lethal dose for wild-type CMCP6 in mice is 300 CFU. This inoculum typically yields 108 CFU/g skin lesion and 104 to 105 CFU/g liver tissue at 20 h post inoculation. FLA600 and FLA601 displayed wildtype levels of local (skin) and systemic (liver) infection, with greater than 107.5 CFU/g skin and greater than 104 CFU/g liver (Figure 3-3). FLA602, howev er, was avirulent at this inoculum, and mice did not develop skin lesions or signs of illness, i.e., no decrease in body temperature, scruffy fur, or lethargy were observed. To dete rmine if the three mutants contained in-frame phoA fusions and to identify the pot entially important vi rulence gene interr upted in FLA602, the DNA sequence of the fusion junctions of the muta tions was needed. Difficulties in cloning the mutated portion of the genomic DNA into pU C19 for sequencing prompted attempts at sequencing directly from genomic DNA. This provided a simple means of identifying the mutated gene. BLAST searches against the CMCP6 genome were used to determine the identity of the mutated genes. PSORT (80) predictions were done to determine the predicted subcellular localizations of the gene products identified. The phoA insertion junction was analyzed using Vector NTI software (Invitrogen) to ensure that phoA was in the correct orientation relative to and in frame with the gene into which it inserted. 46

PAGE 47

FLA600 had a phoA insertion in frame with a gene encoding a TRAP-Type C4dicarboxylate transport system small permease component, with predicted inner membrane localization and designated DctQ by the ExPASy (Expert Protein Analysis System) proteomics server of the Swiss Institute of Bioi nformatics (Table 3-1). FLA601 had a phoA insertion in frame with a parvulin-like pe ptidyl-prolyl isomerase with predicted localization to the periplasmic space that was designated PptC by th e ExPASy server. The gene disrupted in FLA602 was a fatty acid metabolis m regulator with a predicted cy toplasmic localization. This was an unexpected result in that the phoA was not in frame, but was, in fact, oriented in the opposite direction to the reading frame of th e interrupted gene, making it impossible for phoA to have been expressed from the promoter of the gene FLA602 is discussed in detail in Chapter 4. The fact that FLA602 was blue on BCIP ag ar but did not contain an in-frame phoA fusion emphasized the need for careful screening of transconjugants and for comparison to mutants known to have in-frame phoA fusions. With the exception of FLA602, the sequencing of genomic DNA from PhoA mutants in both E. coli and V. vulnificus demonstrated that pGTR201 could function to deliver mini-Tn5Km2phoA into target genomes and that screening for blue transconjugants could identify phoA fusions in frame with gene s encoding secreted products. Use of the PhoA Mutagenesis System to Identify Secreted Virulence Factors of V. vulnificus Thirty-four more phoA insertion mutants were isolated from several conjugations. Blue color on BCIP was compared to the w ild-type and to confirmed in-frame phoA fusion mutants FLA600 and FLA601 as positive controls. A primary screen for virulence by s.c. inoculation of two mice at high inocula (1,000 CFU/ mouse) yielded six mutants with attenuated virulence, viz. FLA603, FLA604, FLA605, FLA606, FLA607, and FLA609. A seventh PhoA mutant, FLA608, had the very interesti ng phenotype of extremely robus t growth on glucose (discussed 47

PAGE 48

below), whereas the wild-type and other PhoA mutants grew very slowly as tiny colonies. FLA608 was not significantly attenuated for viru lence in the mouse model, but its remarkable phenotype warranted furt her investigation. A problem with sequencing directly from genomic DNA of the mutants prompted the return to the original approach of cloning the mutations into pUC19 for sequencing of the fusion junction. Eco RV that does not cut within mini-Tn 5Km2phoA and Kpn I that cuts once at the end of mini-Tn 5Km2phoA were chosen to excise the mutated region out of the mutant genomes for cloning into pUC19. The sequen ce of four of the seven mu tations revealed in-frame phoA insertions, one contained a backward-facing phoA, and the other two sequences had homology only to the pGTR201 vector (Table 3-1.). Surpri singly, it appeared that the entire plasmid had integrated into the genome of these two mutants. This could have occurred if more than one copy of pGTR201 was transferred into the same recipient and tran sposition of miniTn 5Km2phoA from one copy was followed by integrat ion of the entire second copy via homologous recombination between the mini-trans poson sequences. The presence the entire 8.4kb sequence of pGTR201 in the mutated gene comp licated attempts to clone the mutated region into pUC19 for sequencing, so the identity of the genes mutated in FLA604 and FLA606 was not revealed. It is noteworthy that FLA604 and FL A606 may be genetically identical because they were obtained from the same conj ugation (thus may be clones), had the same level of blue color on BCIP plates, and both presented with the unusual sequencing result. FLA605 had an in-frame fusion to VV1_2340 th at encodes a hypothetical protein. The start codon of this gene was 2 nt downstream of what appeared to be a large operon (at least eight genes) encoding prot eins involved in flagellar pilus (Flp pilus) assembly. The Flp proteins are homologous to the Tad (t ight ad herence) proteins that are widespread in bacterial and archeal 48

PAGE 49

species. Tad proteins were brought to the forefront by a study in Actinobacillus actinomycetemcomitans (81) showing that Tadmutants adhere poorly to surfaces, fail to form large autoaggregates, and lack long bundled fibrils. The Tad prot eins are thought to be involved in the secretion of factors required for tight adherence and may have roles in nonspecific adherence to surfaces and co lonization. Full or partial tad loci are found in various bacterial pathogens including Yersinia pestis, Pasturella multo cida, Haemophilus ducrei, Pseudomonas aeruginosa, and Bordetella pertussis. Recently, the published genome of Bdellovibrio bacteriovorus revealed Flp pilus assembly proteins that are likely to be invol ved in adherence to bacterial hosts before invasion (82). V. vulnificus FLA605 was capable of wild-type skin infection (107.5 CFU/g) but low systemic infection (102.4 CFU/g) compared to the wild type ( P = 0.0007) (Figure 3-3). The possible roles of the V. vulnificus Flp pilus assembly machinery in adherence and in secretion remain to be examined. Adherence mediated by type IV pili has been implicated as a virulence factor in V. vulnificus (29,39). Further investigat ions into the role of Flp pili in V. vulnificus may uncover a novel virulence mechanism. The gene disrupted by phoA in FLA607 encoded a cytoplasmic protein, the ATPase component of an ABC-type phosphate transport sy stem. This did not represent and in-frame phoA fusion, but the mutation resulted in decreased systemic, but not local, infection compared to the wild type (108.3 CFU/g skin; 103.1 CFU/g liver) (Figure 3-3). Because the gene mutated in FLA607 was in an ABC-type phosphate transport syst em, the mutant defect could be a failure to correctly transport phosphates into the cytoplasm. Interestingly, this may also provide an explanation for the blue color of FLA607 despit e the cytoplasmic fusion protein formed. If FLA607 was deficient for trans port of phosphates into the cytoplasm, the cell would suffer a phosphate deficit that could be overcome by upr egulating the native alkaline phosphatases to 49

PAGE 50

cleave phosphorylated substrates. The increased synthesis or activity of these phosphatases could lead to a deeper blue color on BCIP-containing plates. FLA608 had an in-frame phoA fusion with a gene encoding the glucose-specific IIBC component of the phosphoenolpyruv ate sugar phosphotransferase system (PTS). The gene product is predicted to be local ized to the inner membrane and is 66% identical to the E. coli protein PtsG, the "fused glucose-specific PT S enzymes: IIB component/IIC component". Although FLA608 was fully virulent in mice (Figure 3-3), it displaye d a very interesting growth phenotype; whereas V. vulnificus normally grows slowly and forms small colonies on media containing glucose, this mutant grew robustly on glucose-containing plates (Figure 3-4). This result suggested that while gluc ose is inhibitory to wild-type V. vulnificus the ptsG mutant had a defect in glucose uptake that relieved the growth inhibiti on. Although FLA608 grew like the wild type in LB-N broth, the mu tant could not grow in M9 mini mal salts containing glucose as a sole carbon source, confirming the glucose transport defect. Even more interestingly, the ptsG mutant had the ability to c onfer growth enhancement to the wild type and other strains growing on glucose-containing plat es (Figure 3-5). This growth enhancer was not diffusing through the agar, as placement of a physical barrier between the ptsG mutant and wild type on agar pl ates still allowed the growth phe nomenon to occur (Figure 3-5). We further determined that the factor res ponsible was probably a gas. Growth of the ptsG mutant on a glucose-containing plate facing the wild type gr owing on a separate glucosecontaining plate still conferred a growth advantage to the wild t ype. A control experiment in which two glucose-containing plates with the wild-type were grown in close proximity under similar conditions failed to repeat these findings. Similarly, when a slice of glucose-containing agar with the fully grown ptsG mutant was placed in a petri pl ate together with, but not touching, 50

PAGE 51

a slice of glucose-containing agar with the (growth inhibited) wild-type strain, the wild type began to grow robustly. The growth advantag e conferred upon strains grown near to FLA608 was not permanent, however, as strains returned to their original glucose-inhibited form when subcultured onto a fresh glucosecontaining plate in the absence of FLA608. Furthermore, the ability to confer robust growth on glucosecontaining media was not unique to FLA608, but stemmed from the fact that it grew robustly on glucose-contai ning media. When a wild-type strain was grown on LB-N agar plates and inc ubated facing another wild -type strain cultured on glucose-containing media, the glucose-grown bact eria grew robustly. It appeared that an unidentified gaseous si gnal emanating from V. vulnificus in high cell numbers conferred a significant growth advantage to cells gr owing under normally inhibitory conditions. Analysis of FLA608 revealed several interesti ng properties. The growth inhibition seen on glucose-containing agar plates could be relieved by a muta tion causing decreased glucose transport or by a yet unidentified gaseous signal emitted by high-density cultures. Also, because FLA608 was defective for growth in glucose as a sole carbon source but was fully virulent in mice, glucose may not be a major nutrien t used for growth in the host tissues. FLA609 had an in-frame fusion of phoA to a gene encoding RseB, a periplasmic negative regulator of E activity. FLA609 was attenuated for viru lence in the initial screen and had interesting morphological phenotypes that warra nted further investig ation. FLA609 will be discussed in detail in chapter 5. Significant Attenuation of Virulence in a cvpA Mutant of V. vulnificus FLA603 was significantly attenuated for both local (skin) and systemic (liver) infection in the s.c. inoculated iron dextrantreated mouse model of disease. At an inoculum of 1,000 CFU (3 times the wild-type MLD), FLA603 did not caus e skin lesions in mice. All mice remained healthy in appearance for the duration of the expe riment (22 h) and had temperatures in the range 51

PAGE 52

of healthy, uninfected mice (37C). At an inoculum of 105 CFU, the FLA603 mutant caused skin lesions in one of six of mice, with a mean bacterial yield of 103.1 CFU/g skin lesion (Figure 3-3). None of the liver samples had detectable bacteria, indicating an absence of systemic infection even at an inoculum of 105 CFU. CFU counts on plates from the liver samples were assigned the minimum detectable value of 1 for calculations of CFU/g liv er, yielding a mean of 102.5 CFU/g liver (Figure 3-3). FLA603 had an in-frame fusion of phoA with VV1_1966 encoding a putative bacteriocin production protein predicted to be localized to the inner membrane. This protein had 64% identity (83% similarity) to the E. coli protein CvpA that was reported to be essential for production of ColicinV from a ColV plasmi d, pColV-K30 (83). VV1_1966 had homology to proteins designated bacteriocin production protein from severa l bacterial species but lacked similarity to any other product. Because the only report of CvpA having an involvement in bacteriocin production did not show a direct link between this protein and colicinV production, it remains unclear if this is th e true function of the protein. Fath and coworkers (83) noted that the CvpA product, a membrane protein, may not be di rectly involved with ColV production but that the mutant protein could have interfered with Co lV export, leading to the drop in ColV activity seen in their experiments. Other possible roles of CvpA in E. coli were not studied, nor was it determined how the protein was involved with production of the bacteriocin. Because of the significant level of attenua ted virulence in FLA603, we wanted to characterize the mutation further. We first attempted to complement the mutation in trans. VV1_1996 was cloned into pRK437 for expression from the lac promoter, forming pGTR2000. However, introduction of pGTR2000 into FLA603 did not restore virulence. To rule out the possibility of a secondary mutation el sewhere in the genome, we moved the 52

PAGE 53

cvpA ::miniTn 5Km2phoA mutation into wild-type V. vulnificus CMCP6 by chitin-induced transformation as detailed in the Materials a nd Methods. Briefly, genomic DNA was extracted FLA603 and sheared by vigorous vortexing. Th e sheared genomic DNA was incubated with CMCP6 growing in seawater in the presence of chitin in the form of crab shell. Transformed bacteria were selected by plating on LB-N containing kanamycin (selects for aph in the minitransposon of FLA603 DNA). The presence of the mini-transposon and of the VV1_1996 mutation in the transformants was confirmed by P CR. A representative transformant was named FLA1013. The virulence propertie s for FLA1013 mirrored those of FLA603 (Figure 3-6). Even at an inoculum of 105 CFU, FLA1013 was significantly attenuated compared to wild-type. Skin CFU from mice inoculated with 105 CFU FLA1013 were nearly 100-f old lower than those from mice inoculated with 103 CFU of the wild type (106.3 and 108.1 CFU/g respectively, P = 0.025). Likewise, liver CFU recovered from mice inoculated with 105 CFU of FLA1013 were 1,000-fold lower than those of mice inoculated with 103 CFU of the wild type (102.2 and 105.3 CFU/g, respectively, P = 2 10-6). Although the chitin-recreated mu tant had an identical virulence phenotype to FLA603, complementation in trans did not restore wild -type virulence to FLA1013. It was possible that the phoA mutation may have caused dom inant-negative effects or have been polar on downstream genes. It will be necessary to construct a targeted deletion of VV1_1996 from wild-type CMCP6 to address this. Fath and coworkers (83) noted that the cvpA gene was encoded just upstream of purF (amidophosphoribosyltransferase) in E. coli and that the cvpA mutant was an adenine auxotroph, possibly due to polar effects on purF expression. VV1_1966 of V. vulnificus is also encoded upstream of a purF homolog, VV1_1967. We tested FLA1013 for auxotrophy by growing it in M9 minimal salts with glucose as a sole carbon source and found that it di d not grow, even after 53

PAGE 54

overnight incubation. Further tests will reveal if the mutant is specifically an adenine auxotroph. The severe attenuation of the cvpA mutants in V. vulnificus highlights this gene as a good candidate for further investiga tion as a novel virulence factor. Discussion Limitations of PhoA Mutagenesis. PhoA fusions are far supe rior to their precursor, lacZ fusions, for studying protein localization and for mutational analyses. Fusions of the cytosolic enzyme produced by lacZ, galactosidase, to exported pr oteins were screened by a loss of enzymatic activity on a chromogenic substrate, doubtless a daunting task. PhoA mutagenesis is not without its caveats, however. Because it is based on transposition, ther e is the possibility of polar mutations. Large extracytoplasmic structures such as flagella pili, secretion pathways, and cell wall and membrane components are very often encoded in vast operons, making polarity of insertions a definite concern, though not an impossible hu rdle. Also, there is a possibility of transdominant mutations if the fused gene encodes a member of a multi-step pathway or multi-subunit complex. Another concern is the fact that the expression of fused phoA is dependent on transcription from a promoter that is active under laboratory conditions. Any viru lence gene promoters that are inactive in vitro will not be identified by PhoA muta genesis. This possibility is not disconcerting, however, as many of the confirme d and putative virulence factors identified in V. vulnificus thus far are expressed in vitro Another limitation is that mutations in some genes encoding trans-membrane protei ns may not be identified if phoA fuses with a portion of the gene corresponding to a cytoplasmic or inner membrane domain, leading to an inactive PhoA enzyme. Lastly, E.coli alkaline phosphatase localized to the cytoplasm slowly acquires enzymatic activity in cells with suspended growth (66). Therefore, if an in-frame fusion of phoA to a gene encoding a cytoplasmic protein causes a significant growth defect the resulting mutant 54

PAGE 55

may appear to be PhoA+, a false positive. Despite these limitations, PhoA mutagenesis is a powerful tool for identifying and simultaneously mutating genes encoding secreted products. Suitability of pGTR201 for Identifying Genes Encoding Secreted Proteins of V. Vulnificus We constructed pGTR201, a modified P hoA mutagenesis plasmid, for use in V. vulnificus Seventy-five percent of blue co lonies whose insertion junctions we sequenced revealed in-frame fusions of phoA with a gene encoding a secreted product (T able 3-1). The fact that some of the mutants appearing blue on BCIP-containing media did not have in-frame phoA fusions was most likely due to difficulties in assessing blue colonies given that wild-type V. vulnificus has alkaline phosphatases that confer a backgr ound level of blue color on this medium. Nevertheless, we were satisfied that the majority of blue colonies represented mutations in secreted products. Thus, our mutagenesis system should be useful to others whose goal is to identify and mutate genes encoding secreted products in V. vulnificus and in other gram-negative organisms. Mutations that Caused Attenuated Virulen ce in the Subcutaneously Inoculated Iron Dextran-Treated Mouse Model of Disease. Of the 37 mutants we examined for virulence, five had decreased virulence compared to the wild-type (FLA604 and FLA606 were excluded from this analysis due to the unresolved nature of the mutations involved) representing an approximately 10% discovery rate. The fact that mutants had varying degrees of attenuated virulence provides further evidence that the virulence of V. vulnificus is multifactorial; no one mutation will likely ablate virulence completely. Mutations that attenuated virule nce were found in gene s belonging to various functional classes. These functions included metabolism (possible purine biosynthesis in FLA603), transport (ABC-type phosphate transp ort system, ATPase component in FLA607), transcriptional regulation (regul ation of fatty acid metabolism in FLA602), and stress responses (RseB, negative regulator of sigm a E activity in FLA609). There was also a mutation in a gene 55

PAGE 56

encoding a hypothetical protein, VV1_2340. Because the start codon of this open reading frame was only two nucleotides downstream of a series of at least eight gene s involved in Flp pilus biosynthesis, we believe that VV1_2340 may sh are Flp pilus-related functions such as intracellular trafficking and secretion or adherence. In summary, PhoA mutagenesis plasmid pGTR201 worked well in E. coli and in V. vulnificus and generated several trans poson mutations, some of which attenuated virulence in the iron dextran-treated mouse model of disease. With the exception of FLA602 and FLA609 that will be discussed in detail in subsequent chapters, the PhoA mutants described in this chapter were characterized only to a small degree. Thus, further studies on the gene s identified here will significantly improve our understanding of the mechanisms underl ying the severe infections caused by V. vulnificus 56

PAGE 57

pGTR201 aph bla transposase 'phoA O end RP4 oriT R6K ori I end pGTR201 aph bla transposase 'phoA O end RP4 oriT R6K ori I end Figure 3-1. VectorNTI diagram showing important features of PhoA mutagenesis vector pGTR201. The origin of replic ation (R6K ori), origin of transfer for conjugations (RP4 oriT), transposase, kanamycin-resistance marker ( aph ), ampicillin-resistance marker ( aph), truncated alkaline phosphatase gene ( phoA ), and the O and I end delineating the boundaries of the mini-transposon are indicated. 57

PAGE 58

Figure 3-2. Improvement of sc reening of PhoA mutants by addition of glucose to BCIPcontaining agar. A) Two transconjugant s (T18, T21) were plated on LB-N with BCIP. B) The same transconjugants were plated on LB-N with BCIP + 0.2% (wt/vol) glucose. While both strains appeared pa le blue when plated on BCIP agar, only T18 was blue when glucose was added to the plate. 58

PAGE 59

59 Table 3-1. Summary of mutants id entified by PhoA mutagenesis of V. vulnificus. PhoA Mutant Gene disrupted In-frame phoA fusion? Protein localization Virulence phenotype FLA600 TRAP-Type C4-dicar boxylate transport system, small permease component Yes Inner membrane Virulent FLA601 Parvulin-like peptidyl-prolyl isomerase Yes Periplasmic space Virulent FLA602 Fatty acid metabolism regulator (not an in-frame phoA fusion) No Cytoplasm Attenuated FLA603 Bacteriocin production protein Yes Inner membrane Attenuated FLA604 ND1 ND ND Attenuated FLA605 Hypothetical protein Yes Inner membrane Attenuated FLA606 ND ND ND Attenuated FLA607 ABC-type Phosphate Transport System, ATPase component No Cytoplasm Attenuated FLA608 PTS system, glucose-specific IIBC component Yes Inner membrane Virulent FLA609 Negative regulator of sigma E activity Yes Periplasm +/-2 1Not determined; cloned insertion junctions showed similarity only to pGTR201 plasmid 2Depended on colony morphology

PAGE 60

0 1 2 3 4 5 6 7 8 9 10 SKIN LIVER TEMPERATURELog CFU/g Tissue and Temp/10 FLA600 FLA601 FLA603 FLA604 FLA605 FLA606 FLA607 FLA608 FLA399a5/5 5/52/6 4/5 5/5 5/55/55/55/5 5/5 5/50/6 0/5 0/51/5 2/5 5/5 5/5*** *** ** ********** ** *** 0 1 2 3 4 5 6 7 8 9 10 SKIN LIVER TEMPERATURELog CFU/g Tissue and Temp/10 FLA600 FLA601 FLA603 FLA604 FLA605 FLA606 FLA607 FLA608 FLA399a5/5 5/52/6 4/5 5/5 5/55/55/55/5 5/5 5/50/6 0/5 0/51/5 2/5 5/5 5/5*** *** ** ********** ** *** 0 1 2 3 4 5 6 7 8 9 10 SKIN LIVER TEMPERATURELog CFU/g Tissue and Temp/10 FLA600 FLA601 FLA603 FLA604 FLA605 FLA606 FLA607 FLA608 FLA399a5/5 5/52/6 4/5 5/5 5/55/55/55/5 5/5 5/50/6 0/5 0/51/5 2/5 5/5 5/5 0 1 2 3 4 5 6 7 8 9 10 SKIN LIVER TEMPERATURELog CFU/g Tissue and Temp/10 FLA600 FLA601 FLA603 FLA604 FLA605 FLA606 FLA607 FLA608 FLA399a 0 1 2 3 4 5 6 7 8 9 10 SKIN LIVER TEMPERATURELog CFU/g Tissue and Temp/10 FLA600 FLA601 FLA603 FLA604 FLA605 FLA606 FLA607 FLA608 FLA399a5/5 5/52/6 4/5 5/5 5/55/55/55/5 5/5 5/50/6 0/5 0/51/5 2/5 5/5 5/5*** *** ** ********** ** *** Figure 3-3. Virulence of PhoA mutants. Mice were inoculated with 1,000 CFU of each strain, with the exception of FLA603a that was inoculated at 105 CFU. 14 to 22 h post-infection, mice were sacrificed, and samples of skin and liver were removed for quantification of the level of infection. Skin and liver CFU were enumerat ed and temperatures were recorded for each mouse. Fractions beneath the bars indicate the number of sa mples that yielded bacteria di vided by the total number of mice inoculated. *, P < 0.05; **, P < 0.01; *** P < 0.001 by ANOVA with Bonferronis po st test for CFU/g tissue or temperature in mutant infections compared to wild-type. P 0.02; P = 0.002; P = 0.0009 by 2 tests for number of samples yielding detectable bacteria. 60

PAGE 61

Figure 3-4. Robust growth of FLA608 on LB -N agar containing 0.2% (wt/vol) glucose. FLA608 was plated on LB-N agar containi ng BCIP and 0.2% glucose to suppress the endogenous PhoA activity of V. vulnificus. Two other PhoA mutants (T-44 and T-45) plated on the same medium are shown fo r comparison. Plates were routinely incubated overnight at 37C and then overnig ht at room temperature to allow full blue color to develop. FLA608 grew robustly on this medium and was not inhibited by glucose. 61

PAGE 62

Figure 3-5. Ability of FLA608 to transform growth of neighboring st rains. FLA608 was streaked on the same LB-N BCIP glucose plat e as three other PhoA mutants. A) The plate was incubated overni ght at 37 followed by an additional 24h at room temperature. B) The plate was incubated ove rnight at 37 followed by an additional 72 h at room temperature. After 4 days of co-incubation, a ll strains resembled FLA608. C) The presence of a physical barrier did not pr event the growth-enhancing effects of FLA608. Strains were streaked on LB-N agar containing BCIP and glucose and incubated overnight at 37 followed by an additional 24h at room temperature. The white bar is a plastic barrier separating the plate in to top and bottom halves. D) The same strains were streaked on LB-N agar containing BCIP and glucose and incubated overnight at 37 followed by an a dditional 48 h at room temperature. After 3 days of co-incubation, even the two strains that were separated from FLA608 by the plastic barrier (top half of plate) began to grow robus tly on the glucose-containing plate. 62

PAGE 63

63 0 1 2 3 4 5 6 7 8 9 SKIN LIVER TEMPERATURELog CFU/g tissue and Temp/10 FLA1013 10^ 3 FLA1013 10^ 5 CMCP6*1/5 5/5 5/5 0/5 0/55/5 103 105*** **** **** 103 0 1 2 3 4 5 6 7 8 9 SKIN LIVER TEMPERATURELog CFU/g tissue and Temp/10 FLA1013 10^ 3 FLA1013 10^ 5 CMCP6*1/5 5/5 5/5 0/5 0/55/5 103 105*** **** **** 103 Figure 3-6. Attenuated virulence of recreated VV1_1996 mutant. Mice were inoculated with 1,000 CFU of wild-type CMCP6 or with 1,000 CFU or 105 CFU of chitin-recreated VV1_1996 mutant, FLA1013. Fractions beneath the bars indicate the number of samples that yielded bacteria divided by th e total number of mice inoculated *, P = 0.025; ** P 10-5 for mutant skin and liver CFU or temperatures compared to wildtype by ANOVA with Bonferronis post test. P = 0.01; P = 0.002 by 2 tests comparing the numbers of samples yielding bacteria from mutant versus wild-type infections.

PAGE 64

CHAPTER 4 THE ROLE OF FADR AND ASPECTS OF FATTY ACID METABOLISM IN VIRULENCE OF V. VULNIFICUS Rationale for Study V. vulnificus fadR ::mini-Tn 5Km2phoA FLA602 was identified in our initial use of phoA mutagenesis. Colonies of FLA602 appeared bl uer than did the wild type on BCIP-containing media, but sequencing revealed that there was not an in-frame fusion of phoA to fadR In fact, fadR encodes a cytoplasmic protein, and the phoA fusion was backwards relative to the fadR gene. Despite the unexpected non-fusion result of FLA602, this mutant had a sufficiently attenuated phenotype in s.c. mouse infection to merit its continued ch aracterization. Further study of the fadR mutant led to the discovery of an impor tant link between fatty acid metabolism and virulence in V. vulnificus Introduction There is considerable publishe d work on the roles of FadR in fatty acid metabolism and the glyoxylate shunt in E. coli but far less is known about FadR in other bacteria. The first reported role for FadR in E. coli was negative regulation of th e fatty acid degradation ( fad ) genes (84). Nunn and associates (85) suggested a role for FadR in unsaturated fatty acid synthesis and noted that FabA enzyme ( -hydroxydecanoyl-thioester dehydrase) activity was decreased in the fadR mutant. Later, the fabB gene was also shown to be positively regulated by FadR in E. coli (61). E. coli FadR also activates iclR encoding the repressor of the glyoxylate shunt enzymes (86) and represses uspA encoding a universal st ress protein (87). In E. coli fadR mutants produce less unsaturated fatty acids than do wild-type stains but this was reported to be phenotypically asymptomatic (85). There have been no reports of FadR being essential for bacterial infection, and there is a scarcity of liter ature concerning the role of fatty acid metabolism during infection. 64

PAGE 65

Results A V. vulnificus Mini-Tn 5Km2phoA Insertion Mutant Is Severely Attenuated for Skin and Liver Infection in Iron Dextran-Treated Mice Colonies of FLA602 had a slightly bluer co lor than did the wild-type parent on BCIP plates. FLA602 also formed smaller colonies th an did the wild type on LB-N agar plates and grew with a longer doubling time than did the wild type in LB-N broth (see below). However, the slow growth of FLA602 was not due to auxotrophy, since it could grow in M9 minimal medium. In the initial screen for virulence at 1,000 CFU of FLA602, and in three subsequent infections with inocula ranging from 2,000 to 4,000 CFU, V. vulnificus FLA602 failed to cause skin lesions in mice over a 22 h course of in fection. The FLA602-inf ected mice remained healthy in appearance, and rectal temperatures were in the normal range of uninfected mice (37 C). To assess the extent of attenuation, we infected mice with increasing inocula of V. vulnificus FLA602. Inoculation of 104 CFU failed to cause skin lesions, mice remained healthy in appearance, and body temperatures were similar to thos e of uninfected mice throughout the 22 h course of infection. Skin lesions were observed onl y with inocula at or above 105 CFU, and liver infection was dete ctable only at an inoculum of 107 CFU (Figure 4-1). With an inoculum of 105 CFU of FLA602 only three of five mi ce had visible skin lesions and the mean number of bacteria recove red from the skin lesions (105.8 CFU/g skin tissue) was significantly lower than that of the wild type at an inoculum of 103 CFU (108 CFU/g skin tissue; P = 0.01). Similarly the number of bacteria recovered from liver samples of mice inoculated with 105 CFU of the mutant (102.4 CFU/g liver) was significantly lower than the wild type (104.7CFU/g, P = 0.0007). Even when 106 CFU of the mutant were inoculated, skin infection (106.1 CFU/g) and liver infection (102.5 CFU/g) were significantly lowe r than the wild type at a 103 CFU inoculum ( P = 0.04 and P = 0.0008, respectively). 107 CFU of FLA602 resulted in 65

PAGE 66

wild-type skin infection (107.8 CFU/g skin), but low systemic infection (102.6 CFU/g, P = 0.002 compared to the wild type). FLA602 caused fu ll wild-type levels of infection only when 108 CFU were inoculated. Mean temper atures of mice infected with 105, 106, or 107 CFU of FLA602 were also higher than those of mice inoculated with 103 CFU of the wild type, indicating that the mice inoculated with the mutant were not moribund. Additionally, the proportion of liver samples of mice that yielded bacteria from the mutant infections at 105, 106, and 107 CFU showed a statistically significan t decrease compared to the pro portion of liver samples of mice that yielded bacteria from the wild-type infection ( P = 0.002, P = 0.002, and P = 0.04, respectively, by 2 test) (Figure 4-1). The FLA602 mini-Tn 5Km2phoA Insertion is in the fadR Gene The blue color of colonies of FLA602 on LB-N agar containing BCIP and glucose suggested that the mini-Tn 5Km2phoA formed an in-frame gene fusion of a V. vulnificus gene to phoA However, DNA sequence analysis of genomic DNA from the mutant identified a backward insertion of phoA into V. vulnificus CMCP6 gene VV1_2233, annotated as a "fatty acid metabolism regulator". PSORT (80) predic tion based on the amino acid sequence suggested a cytoplasmic protein localization. The amino acid sequence contained the conserved domain FadR and had 52% identity to the well-studied E. coli FadR protein (Figure 4-2). As such we named the mutated gene fadR. The mini-Tn 5Km2phoA insertion occurred at nucleotide 346 of 840 in the fadR gene, meaning that 40% of the FadR protein was possibly translated. Because FadR is a cytoplasmic protein, we were puzzled by the blue color of colonies of FLA602 on BCIP plates. We did not resolve th e cause but noted that complementation and reversion of the mutation resulted in colonies th at were similar in color to the wild type (see below), so the blue color of the fadR ::mini-Tn 5Km2phoA mutant on BCIP plates was not likely 66

PAGE 67

due to a secondary mutation. Furthermore, as detailed below, a fadR mutation did not yield a similar blue color. Confirmation of the fadR Phenotype Although the gene interrupted by the mini-Tn5Km2phoA insertion possessed significant homology to fadR of E. coli, we further examined FLA602 to confirm the phenotype. The fatty acid synthase inhibitor cerulenin has been used as an indicator of the fadR mutant phenotype in E. coli (61) and to inhibit fa tty acid synthesis in V. vulnificus (88). Because FadR positively regulates genes involved in unsatur ated fatty acid biosynthesis in E. coli one aspect of the fadR mutant phenotype is decreased synthesis of unsatur ated fatty acids (85). This renders the mutant hypersensitive to cerulenin because the drug further de creases fatty acid synthesis. Similar to its E. coli counterpart, the V. vulnificus fadR ::mini-Tn 5Km2phoA mutant showed increased sensitivity to cerulenin compared to the wild -type parent. The mean minimum inhibitory concentration (MIC) for the mutant was 2.8 g/ml cer ulenin, in contrast to the wild-type parent with a mean MIC of 16.7 g/ml, an approximately 6-fold difference in MIC ( P = 0.001, by Students t test). Mean values are for 3 biological replicates. This was similar to the 10-fold increase in sensitivity reported for an E. coli fadR mutant examined by Campbell and Cronan (61). Increased cerulenin sensitivity in the V. vulnificus fadR mutant suggested decreased levels of the 3-ketoacyl-ACP synthase I enzyme (FabB), the main target of the inhibitor (89,90). To confirm this result and to assess some aspects of FadR-regulated gene expression in the mutant, we performed qRT-PCR using genes know n to be regulated by FadR in E. coli. In E. coli, FadR positively regulates at least tw o genes involved in unsaturat ed fatty acid biosynthesis, fabA and fabB and negatively regulates genes involved in -oxidation (recently reviewed in (91)). To 67

PAGE 68

determine if V. vulnificus fadR functioned similarly, we tested the relative expression of fabA and fabB as well as two genes related to -oxidation, fadB and fadD, in fadR ::mini-Tn 5Km2phoA mutant FLA602 compared to the wild type usin g qRT-PCR. All gene expression levels were normalized to 16S rRNA as an endogenous control. Consistent with the expected results for a fadR mutation, expression of fabA and fabB was lower, and fadB levels were higher in FLA602 compared to the wild type. fadD expression was not significan tly changed in FLA602 under the conditions tested (Table 4-1). In vitro Growth of the fadR Mutant Increased sensitivity to cerulenin and changes in fad and fab gene expression suggested that the fadR mutant may be impaired for fatty acid metabolism. E. coli fadR mutants are able to grow on the medium chain fatty acid decanoate as a sole carbon source, while wild-type cells cannot (92). In contra st, both wild-type and fadR mutant E. coli can use long chain fatty acids such as oleate as a carbon source. We tested growth of the V. vulnificus fadR ::miniTn 5Km2phoA mutant FLA602 and parental FLA399 usi ng decanoate and oleate as sole carbon sources. Both strains grew in both fatty acid s as sole energy sources. Unexpectedly, the V. vulnificus fadR mutant grew slower in rich LB-N br oth (doubling time 29 min) than did the wild type (18 min) (Table 4-2), a phenotype that was not reported for E. coli fadR mutants. To test if the slow in vitro growth of FLA602 could be respons ible for the observed attenuation of virulence, we compared FLA602 with a nother slow-growing mutant generated by phoA mutagenesis, FLA1000, that contains a ptsI ::mini-Tn 5Km2phoA mutation. FLA1000 had a doubling time in LB-N of 25 min, compared to 18 min for the wild type; however, FLA1000 retained full virulence in mice (Table 42). Therefore, the slow growth of fadR FLA602 did not necessarily explain its severe attenua tion in our mouse model of disease. 68

PAGE 69

Complementation and Reversion of the fadR ::mini-Tn 5Km2phoA Mutation To prove that the attenuated virulence phe notype of FLA602 was due to the knockout of the fadR gene by the transposon insertion and not a polar effect or se condary mutation, we introduced the wild-type fadR gene in trans into V. vulnificus FLA602 via plasmid pGTR349 (pRK437 containing the V. vulnificus fadR gene) Complementation of the fadR ::miniTn 5Km2phoA mutant restored the wild-type size to the colonies and reduced cerulenin sensitivity to wild-type levels (data not show n). The complemented strain l ooked identical to the wild type (pale blue) on BCIP plates, indicating that the bl ue color of FLA602 was somehow related to the genotype of fadR ::miniTn 5Km2phoA Complementation also resulted in a significant increase in virulence (Figure 4-3). At an inoculum of 1,000 CF U, skin lesions were observed in all mice ( P = 0.002 compared to fadR ::mini-Tn 5Km2phoA mutant by 2 test). These skin lesions yielded a mean of 7.7 0.53 log CFU/g, similar to a wild-t ype infection at this inoculum. Although high levels of skin infection were restored, liver CF U were observed in only one of five mice. With an inoculum of 104 CFU, however, both skin and liver infection were detected at wild-type levels 7.8 0.28 log CFU/g skin lesion and 4.8 1.6 CFU/g liver. Because complementation of the fadR ::mini-Tn 5Km2phoA mutant did not completely restore wild-type virulence, we reverted the mutation by allelic exchange with the wild-type fadR gene. This would preclude possi ble dominant-negative effects of the mutant FadR protein, in which the first 40% of the protein was intact. The truncated FadR pr otein would contain the critical N-terminal DNA-binding domain with resi dues that are predicted to form alpha-helices that interact during dimerizati on and DNA binding (93). Thus, the truncated FadR protein could potentially dimerize with wild-type FadR provided in trans thereby reducing the efficiency of the complementation. In contrast to the complementation, reversion of the mutation completely restored virulence in terms of skin and liver infe ction (Figure 4-3). Mean bacterial yields from 69

PAGE 70

skin (107.6 CFU/g) and liver (105.2 CFU/g) tissues of in fected mice were similar to those of the wild type (108.2 CFU/g skin; 104CFU/g liver) at the same inocul um (1,000 CFU). Therefore, we concluded that the virulence phenotypes of FLA602 were due to effects on the fadR gene. Deletion of fadR from Wild-Type CMCP6 To address the failure to fully complement the fadR ::mini-Tn 5Km2phoA mutation in trans, we deleted the fadR gene from wild-type V. vulnificus CMCP6 by using the 3-way USERSma I deletion method detailed in the Mate rials and Methods. DNA flanking the fadR gene was amplified using oligonucleotide pairs delf adR up5 delfadRup3, delfadR dn5 delfadR dn3. (Table 2-3) to create pGTR 2009 and pGTR2010 for deletion of fadR (Table 2-2) by via chitininduced transformation. The deletion mutant, FLA614, shared colony morphology characteristics with mini-Tn 5Km2phoA insertion mutant FLA602; colonies were small and slightly yellow compared to the wild type. FLA614 also grew slow er than the wild-type in rich broth (discussed below, see Fi gure 4-7) and was hypersensitive to cerulenin (not shown). Surprisingly, fadR FLA614 did not appear bluer than the wild type on BCIP-containing media. This suggested that the blue color seen for FLA602 was unique to the fadR ::mini-Tn 5Km2phoA mutation. Most notably, the deletion mutant was attenuated for virulence in the mouse model of disease. At an inoculum of 1,000 CFU, FLA614 did not cause skin lesions and mice remained healthy throughout the 22 h course of infection (Figure 4-4). Surprisingly, infections with increasing inocula showed that the deletion muta nt was not as highly attenuated as the miniTn 5Km2phoA insertion mutant, FLA602. An inoculum of 104 CFU of FLA614 was sufficient to cause skin lesions, although th e mean CFU/g recovered (105.8) was more than two logs lower than the mean CFU/g recovered from an infec tion with the wild-type parent CMCP6 at 1,000 CFU (108.4 CFU/g; P = 0.04). Liver CFU were recovered from only 1 of 5 mice inoculated with 104 CFU of FLA614 (mean CFU/g liver = 102.2). This was also significantly lower than wild70

PAGE 71

type recovery (104.2 CFU/g liver; P = 0.003), representing a 100-fold decrease in liver CFU for FLA614 with an inoculum that was 10-fold greater than that used for the wild-type infection. An inoculum of 106 CFU of FLA614 was sufficient to cause wild-type levels of skin and liver infection, with a mean recovery of 108.3 CFU/g skin and 105.6 CFU/g liver. Most important, complementation in trans with the fadR gene expressed on plasmid pGTR349 restored full wildtype virulence to fadR FLA614. 1,000 CFU of the complement ed deletion strain caused wildtype levels of skin infection (108.3 CFU/g) and liver infection (105.1 CFU/g) in mice (Figure 4-4). Because fadR FLA614 was phenotypically similar to fadR ::mini-Tn 5Km2phoA FLA602 and was fully complemented in trans we concluded that the fadR mutation was indeed responsible for the morphology, growth, and virulence phenot ypes seen in both mutants, and FLA614 was used as a definitive fadR mutant for subsequent experiments. Altered fatty acid content of the fadR mutant Given that the fadR mutations (insertion or deletion) cau sed hypersensitivity to cerulenin and evidence of decreased fab gene expression, we hypothesized that a fadR mutant should exhibit decreased synthesis of unsaturated fatty acids that could lead to an altered membrane lipid profile (essentially all fatty acids and lipids in bacteria are found in the cell envelope (94). Gas chromatographic analysis of fatt y acid methyl esters derived from fadR FLA614 and wildtype CMCP6 revealed a small (13%) but statistically significant ( P = 0.005) decrease in unsaturated fatty acids and a concurrent 12% increase in saturated fatty acids in the fadR mutant compared to the wild type ( P = 0.006) (Table 4-3). Taken together, the increase in cerulenin sensitivity, decrease in fab gene expression, and decrease in cellular unsaturated fatty acids suggested that fadR mutants are deficient in synthesi s of unsaturated fatty acids. 71

PAGE 72

Envelope Stress Sensitivity and Motility of the fadR Mutant Because FLA614 showed slightly lower membra ne unsaturated fatty acids than did the wild type, it was possible that the unbalanced fatty acid composition might render the mutant more susceptible to external stresses, possibly ex plaining the attenuated virulence. To test the susceptibility of the fadR mutant to envelope stresses we performed MIC experiments using chemicals that disrupt membrane integrity: etha nol, a well known membrane perturbant; SDS, a detergent that solubilizes the lipid bilayer; and polymyxin B that binds LPS and disrupts membrane integrity. The MICs of ethanol and of polymyxinB for the fadR mutant were similar to those of the wild type (Table 4-4). The fadR mutant was slightly more sensitive to SDS than was the wild type, with 0. 22% SDS causing growth inhibition of the mutant compared to 0.33% for the wild type (P = 0.02) (Table 4-4). In light of th e small magnitude of the change in SDS sensitivity and the fact that FLA614 was not hypersensitive to other membrane-perturbing agents, we do not believe that FLA614 was more sensitive than the w ild type to envelope stresses. We also tested sensitivities to heat and co ld as indications of membrane integrity. fadR mutant FLA614 and wild-type CMCP6 were grow n to exponential phase in LB-N broth, and serial dilutions were spotted onto LB-N plates and incubated at 37C, 42C, or 4C for 18h. FLA614 grew as well as CMCP6 at 37C and 42C (F igure 4-5), but neither strain showed signs of growth after 18h incubation at 4C (not shown). Further incubation of the 4C plates overnight at room temperature (23C) resulted in equal growth of both strains (Figure 4-5). Overall, the fadR mutant did not exhibit in creased sensitivity to envelope stresses compared to the wild type. As a further test of membra ne defects, we tested the fadR mutant for sensitivity to serum complement as an indicator of possible decreased resistance to host defenses. fadR mutant 72

PAGE 73

FLA614, wild-type CMCP6, and complement-sensitive E. coli MG1655 were incubated in intact or heat-inactivated 90% rat serum for 2 h, then plated to quantify survival. E. coli MG1655 was completely killed by incubation with intact rat serum. The ra tio of MG1655 that survived in intact versus heat-killed serum was log -6.6 0.4 (P = 10-5 compared to the wild type; n = 3 biological replicates). In contrast, neither the fadR mutant nor wild-type V. vulnificus was sensitive to complement-mediated killing. The ratio of CMCP6 that survived in intact versus heat-killed serum was log 0 0.1 and for FLA614 was log -0.2 0.2 ( P = 0.19 compared to the wild type, n = 3 biological rep licates). Thus, the fadR mutant was not more complementsensitive than the wild type. In addition to the possibility that loss of fadR could cause changes in envelope stress sensitivity, it was also possible that the fadR mutant could display altered motility. Motility is a complex function that requires the assembly and activity of the flagellum and motor apparatus across the cell envelope as well as ion gradie nts across the cell inner membrane to provide energy. Furthermore, flagella of the Vibrionaceae possess a sheath that appears to be an extension of the cell outer membra ne (reviewed in (95)). We, therefore, hypothesized that the altered fatty acid profile of the fadR envelope could lead to changes in motility. We tested motility of fadR FLA614 and wild-type CMCP6 by measuring the diameter of spread through 0.3% motility agar. FLA614 was only 45% as motile as the wild type ( P = 3-7). Motility was fully restored by complementation (Figur e 4-6). We noted that the slow-growing ptsI mutant FLA1000 was fully motile (not shown), i ndicating that slow growth was not likely the cause of the decreased motility seen in the fadR mutant. 73

PAGE 74

Supplementation with Unsaturated Fatty Acid in Vitro and in Vivo Deletion of V. vulnificus fadR resulted in decreas ed growth rates in vitro as well as decreased membrane unsaturated fatty acids (see above ). To test if the decreased growth rate of the V. vulnificus fadR mutant in vitro was related to decreased unsat urated fatty acid synthesis, we supplemented LB-N broth with an unsaturat ed fatty acid derivative, sodium oleate, and measured growth of fadR FLA614 and wild-type CMCP6. The addition of 0.005% (wt/vol) sodium oleate (half the concentration used as a sole carbon source) fac ilitated growth of the fadR mutant to wild-type levels (Figure 4-7). Interestingly, wild-type CMCP6 did not grow any better when oleate was added to the grow th medium, suggesting that CMCP6 was already achieving maximal growth rates and that the av ailability of fatty acids in LB-N was not a limiting factor for wild-type growth. The addition of 0.005% sodium oleate to LB-N plates also significantly increased the mean diameter of colonies formed by the fadR mutant. While the mean colony diameter of the fadR mutant grown on LB-N plates was 2.4 0.2 mm, the mean colony diameter for the mutant grown on LB-N plates supplemented with sodium oleate was 3.7 0.4 mm ( P = 4-4). Similar to the trend seen for wild-type CMCP6 grown in LB-N broth with oleate, the addition of oleate to LB-N plates did not significantly alter the mean colony diameter of wild-type CMCP6 (5.0 0 mm on LB-N; 5.2 0.4 mm on LB-N + oleate; P = 0.35). Because supplementation with oleate functiona lly complemented the growth defect of the fadR mutant in vitro, we inferred that the slow growth of the mutant in vitro was due to insufficient synthesis of unsaturat ed fatty acids. To test if a lack of unsaturat ed fatty acids was responsible fo r the attenuated virulence of the fadR mutant, mice were infected with FLA614 su spended in sodium oleate or in BSG as a control. We used a concentr ation of sodium oleat e (0.225% (wt/vol)) that was was nearly 5074

PAGE 75

fold higher than that used for in vitro supplemen tation to account for equili bration with the fluid phase of the mouse. Mice were also infected with an equal inocul um of wild-type CMCP6 suspended in BSG for comparison. Control mice i noculated with sodium oleate alone showed no skin pathology upon necropsy (not shown). The a ddition of oleate to the inoculum allowed an inoculum of 1,000 CFU of the fadR mutant to establish wild-type levels of skin infection (108.2 CFU/g skin tissue) and near wild-typ e levels of systemic infection (104.8 CFU/g liver tissue) compared to CMCP6 that had a mean of 108.4 CFU/g skin ( P = 0.35) and 105.8 CFU/g liver ( P = 0.01). In contrast, the fadR mutant without oleate supplementa tion established infection in only one of the 5 mice infected (Figure 4-8). We were surprised to find that one of the fadR inoculated mice developed disease in this experi ment; this was the only time in 4 separate experiments with 5 mice each that a mouse inoculated with FLA614 developed disease at an inoculum of 1,000 CFU. It was possible that this mouse had some unnoticed underlying predisposition. Given that the oleate supplementa tion significantly increase d the virulence of the fadR mutant in mice, we concluded that the major defect of the fadR mutant in vivo was insufficient synthesis of uns aturated fatty acids. The Role of Fatty Acid Utilization in Infection with V. vulnificus V. vulnificus replicates in the fatty s.c. tissues of iron dextran-treated mice with a very rapid doubling time of 15 to 28 minu tes (38,55). We hypothesized that V. vulnificus survives by metabolizing fatty acids or lipids freed from the s.c. tissues. Two pathways are specifically required for the utilization of fa tty acids in bacteria. The -oxidation pathway degrades fatty acids to acetyl-CoA units (96), and the glyoxylate cycle is require d when glycolytic substrates are absent or in low abundance (96). 75

PAGE 76

The glyoxylate pathway is essential for virule nce in some bacterial pathogens, most notably Mycobacterium tuberculosis (97-99) In E. coli, the metabolic and re gulatory enzymes of the glyoxylate pathway are expressed from a pol ycistronic transcript. The metabolic enzymes isocitrate lyase and malate synthase are encoded by aceA and aceB and the regulatory enzyme isocitrate dehydrogenase kina se/phosphatase is encoded by aceK (100). V. vulnificus CMCP6 has a single ortholog of E. coli aceA VV1_0449, annotated as isocit rate lyase. Immediately following VV1_0449 is VV1_0450, annotated as malate synthase A. We found no ortholog of E. coli aceK in the genomic sequence of CMCP6. There was a second V. vulnificus gene, VV2_1647, annotated as malate synthase G in CMCP 6. However, this second malate synthase was not a paralog of VV1_0450 (malate synthase A), and was not surrounded by any other genes with predicted functions in glyoxylate metabolism. As such, we targeted VV1_0449 and VV1_0450 for deletion as candidate glyoxylate pathway enzymes. Oligonucleotide primers were designed to delete aceA and aceB simultaneously from wild-type CMCP6 by allelic exchange and chitin-induced tr ansformation, as detailed in th e Materials and Methods. A representative aceAB mutant was named FLA1003. We firs t determined if FLA1003 was able to use a fatty acid as a sole carbon source. FLA1003 and wild-type CMCP6 were grown shaking overnight at 37C in M9 minimal salts with 0.01% (wt/vol) oleate, in M9 with 0.2% (wt/vol) glucose, or in M9 with no car bon/energy source. Cultures were plated after overnight growth to enumerate viable CFU. While CMCP6 grew in oleate and in glucose as sole carbon sources, FLA1003 did not grow better in oleate as a sole carbon source than in M9 salts alone (Table 45). This result suggested that FLA1003 did not have an intact glyoxylate pathway and hence could not use fatty acids as a sole carbon source. We tested the virulence of aceAB FLA1003 76

PAGE 77

in mice. Despite the inability to utilize fats as nutrients, FLA1003 retained full virulence (Figure 4-9). There have been numerous reports of roles for the fatty acid -oxidation pathway in virulence in bacteria (101-103). To investigate the ro le of fatty acid -oxidation in V. vulnificus, we tried to find a member of the pathway that, if deleted, would ablate the entire process. This would preclude deletion of each of the genes involved in -oxidation, viz., fadBA fadD fadE fadF fadG fadH fadI and fadJ A prime target for deletion was fadL encoding the transmembrane fatty acid transporte r (reviewed in(104)). Deletion of fadL should result in bacteria that cannot import exogenous long chain fatty acids (105) and would be a good way to study the fatty acid requirement for bacteria in vivo However, four separate ORFs encode fadL homologs in V. vulnificus : VV1_1971, VV1_1972, VV2_0249, and VV2_0881. The putative proteins encoded by these ORFs were 19 to 36% identical to E. coli FadL. Because of the possibility of functional redundanc y among these paralogous genes, we did not attempt to mutate V. vulnificus fadL. In E. coli, transport of fatty acids is coupled to activation by the inner membrane-bound fatty acyl coenzyme A (CoA) synthase, FACS, the product of the fadD gene (reviewed in(104)). fadD mutants of E. coli are unable to grow using fatty acid s of any chain length as a sole carbon source (106). V. vulnificus encodes a single fadD ortholog, VV1_0136. We deleted VV1_0136 from wild-type CMCP6 by allelic exchange via ch itin-induced transformation as detailed in the Materials and Methods. A representative fadD mutant was named FLA 1006. We first tested the function of V. vulnificus fadD by observing if FLA1006 could gr ow using a long-chain fatty acid as a sole carbon source. FLA1006 and wild-t ype CMCP6 were grown shaking overnight at 37C in M9 with 0.01% (wt/vol) oleate, in M9 with 0.2% (wt/vol) glucose, or in M9 with no 77

PAGE 78

carbon source. Cultures were plated after overn ight growth to enumerate viable CFU. Surprisingly, FLA1009 grew in oleate as a sole carbon source (Table 4-5). Therefore, it appeared that there was another mechanism fo r transport of long-chain fatty acids in V. vulnificus. We tested the virulence of FLA1006 in s.c. inoculated mice and, not surprisingly, it was as virulent as was the wild type (Figure 4-9). Discussion Identification of a V. vulnificus fadR Mutation that Attenuates Virulence We used alkaline phosphatase fusion-insertion mutagenesis to identify secreted virulence factors of V. vulnificus and isolated a mini-Tn 5Km2phoA insertion in fadR (fatty acid metabolism regulator) that encodes a cytoplasmic protein. Th e transposon insertion was backward relative to the fadR open reading frame; hence, there was no phoA fusion. We do not understand the increased blue color of colonies of the fadR ::mini-Tn 5Km2phoA mutant on BCIP-containing agar plates, but it should be noted that, due to the presence of other alka line phosphatase genes in the genome, V. vulnificus CMCP6 has a low level of background blue color on BCIP plates that confounds screening on this medium. fadR ::mini-Tn 5Km2phoA FLA602 was one of the first mutants we isolated using PhoA mutagenesis. We subsequently used FLA602 as a negative control for blue color because it did not represent an in frame phoA fusion. The fadR ::miniTn 5Km2phoA mutant was significantly attenuated for viru lence in s.c. inoculated iron-dextran treated mice (Figure 4-1), leading us to investigate the role of fadR and fatty acid metabolism in virulence of V. vulnificus Comparison of V. vulnificus FadR with Known FadR Properties Although the V. vulnificus fadR mutant was not a fatty acid a uxotroph, it grew slower than did the wild type in rich LB-N broth (Table 4-1, Figure 4-7) and produced smaller colonies. These phenotypes were not reported for E. coli fadR strains, indicati ng that mutation of fadR in 78

PAGE 79

V. vulnificus has more severe phenotypic conseq uences. We confirmed expected fadR phenotypes including cerulenin sensitivity and regulation of fatty acid-related genes (Table 4-2); moreover, the amplitudes of tr anscriptional regulation of fab and fad genes that we observed were within the range of those reported for E. coli (61,87) Therefore, we are confident that our annotation of fadR to VV1_2233 is correct. Both E. coli and V. vulnificus wild-type strains grow on oleate as a sole carbon source, and fadR mutants of both species can grow on decanoate as a sole carbon source. However, wild-type V. vulnificus, but not E. coli, can grow on decanoate as a sole carbon source. A similar growth trend to V. vulnificus was reported for Salmonella enterica serovar Typhimurium ( S. Typhimurium) (92). A recent study that compared various aspects of FadR proteins of several pathogens (107) noted that the V. cholerae FadR protein contained a 40-residue in sertion relative to the E. coli protein. The am ino acid sequence of V. vulnificus FadR is 85% identical to that of the V. cholerae sequence, and V. vulnificus FadR contains an identical 40-residue insertion relative to the E. coli protein (Figure 4-2 ). Because V. vulnificus FadR shares only 52% identity with the E. coli protein, there may be differences in protein function yet to be discovered. The Role of FadR in Infection of Mice To our knowledge, the results reported herein are the first to demonstrate a role for fadR in the ability of a pathogen to infect animal hosts. Most of the literature on the role of fatty acid metabolism in bacterial virulence focuses on degradation ( -oxidation). For example, fadB was identified in S. Typhimurium as a gene specifically induced during infection (108). Fatty acid degradative defects are no t likely the explanation for the attenuation of the V. vulnificus fadR mutant as our results support previous reports that fad genes are expressed at higher levels in fadR mutants compared to the wild type. To our knowledge, there has been no published work on the deleterious effects of constitutive fatty acid degradation during infection. Fatty acid 79

PAGE 80

synthesis was identified as important for acid to lerance and virulence of the cariogenic bacterium Streptococcus mutans (109,110). A fabM ( trans -2, cis -3-decenoyl-ACP isomerase) mutant of S. mutans grew slower than did the wild type in vitro and had several altered enzymatic activities including reduced glycolytic cap ability and altered glucose-PTS activity (109). FabM in grampositive bacteria functions similar to FabA of gram-negative bacteria; both enzymes can isomerize the trans -2-enoyl bond of a nascent fatty acyl chain to the cis -3 isomer during fatty acid biosynthesis (reviewed in(67). This isomerization is essen tial to divert the growing acyl chain into the unsaturated fatty acid synthetic pathway. Therefore, loss of this isomerase activity results in decreased unsaturated fatty acid synthesis. Since the V. vulnificus fadR mutant showed decreased expression of fabA (Table 4-2), we were not surprised that the mutant had some of the phenotypic characteristics reported for a fabM mutant slow growth and decreased virulence. It remains to be seen if the V. vulnificus fadR mutant also has the far-reaching enzymatic defects seen in the S. mutans fabM mutant. We expected that a fadR mutation might severely impair membrane integrity and function, but we found that envelope stress sensitivity was not increased in the fadR mutant (Table 4-4, Figure 4-5) and serum complement resistance was not impaired in the mutant. Therefore, it seemed that the altered membrane fatty acid profile of the fadR mutant (Table 4-3) did not result in fragile membranes. On the other ha nd, the mutant was defective in motility (Figure 46). The 55% decrease in motility of the mutant co mpared to the wild type was similar to that seen in a V. vulnificus strain lacking the flagellin genes flaCDE (Tucker, M.S., et al ., manuscript in preparation). This flagellar mutant showed decreased systemic inf ection but not decreased local infection. Even completely non-motile V. vulnificus mutants ( flaFBA flaCDE and motAB ) retain the capacity to cause local infec tion (Tucker, M.S., et al., manuscript in 80

PAGE 81

preparation), so the decrease in motility of the fadR mutant was not sufficient to explain the severe attenuation of virulence. While it can be argued that slow growth coul d contribute to decr eased virulence, we observed that another slow-growing mutant of V. vulnificus was capable of causing wild-type infection (Table 4-2). Therefore, slow growth in vitro is not necessarily a pr edictor of attenuated virulence. On the other hand, we noted that the growth rate of the V. vulnificus fadR mutant could be functionally complemented in vitro by adding the fatty acid oleate to rich broth (Figure 4-7), suggesting that the gr owth defect may be due to decreased unsaturated fatty acid synthesis. Likewise, inclusion of oleate in the inoculum caused a significan t increase in skin infection during infection of mice (Figure 48), indicating that a defect in fatty acid biosynthesis was the main factor causing decreased infectivity in the V. vulnificus fadR mutant. The Roles of FadD and the Glyoxylate Enzymes in V. vulnificus During s.c. infection wild-type V. vulnificus thrives in the lipid-rich s.c. environment of the host with a doubling time of 15 to 28 minutes in iron dextran-treated mice (38,55). The bacteria may use lipases and phospholipases to destroy ph ospholipid bilayer memb ranes and to release lipids from adipocytes. The bloodstream also contains free fatty acids that could potentially be used by systemically infecting bacteria. Krivan, et al. (111) showed that phosphatidylserine from mouse intestinal mucus can serve as a sole carbon and nitrogen source for Salmonella and E. coli. Bacteria also use fatty acid oxidation to break down inflammatory byproducts of the host during infection. For example, fadB of S. Typhimurium has been proposed to break down the proinflammatory arachidonic acid that is generated by specific phospholipases during phagocytosis (108). Also, ma ny long chain fatty acids produced by mammalian hosts are inhibitory to bacter ial pathogens (112). We hypothesized that V. vulnificus uses lipids as a nutrient source during infection and that a fadD mutant would be unabl e to take in exogenous 81

PAGE 82

fatty acids for use as an energy source. Because fadD FLA1006 retained the ability to grow in oleic acid as a sole carbon source (Table 4-5), we were unable to assess the function of fatty acid transport/activation for virulence in V. vulnificus. We also investigated the role of the glyoxylate pathway in V. vulnificus This pathway is required for growth of bacteria in fatty acids as a sole carbon source. Although the V. vulnificus aceAB mutant was unable to grow on oleate as a sole carbon source, it was fully virulent in mice (Fig ure 4-9). The presence of a second malate synthase in the CMCP6 genome suggests that there may be functional redundancy. It may also be possible that V. vulnificus does not use lipids or fatty acid s as a main source of nutrients in vivo. This would be an in triguing possibility, as a ptsG mutant that was unable to use glucose as a sole carbon source was also fully virulent in mice (Figure 3-1). Except for a few cases, little is known about the nutrient sources of bacterial pathogens during inf ection (113). What V. vulnificus uses as carbon sources during inf ection remains to be identified. Conclusion and Future Directions We hypothesize that V. vulnificus, normally a resident of sh ellfish and estuarine water, causes lethal opportunistic infections in hum ans by overwhelming the host innate immune defenses by rapidly dividing to reach a critical mass while destroying host tissues with various toxins and enzymes. Our results reported here underscore the importance of FadR for regulating unsaturated fatty acid biosynthesis during the infectious process of V. vulnificus. These findings are likely to extend to additional bacterial pathog ens. Because the bacterial fatty acid system differs from the mammalian system, this has long b een an attractive target for antibiotics (114). However, some compounds identified for this pu rpose have had the drawback of interactions with the mammalian fatty acid metabolic system For example, cerulenin interacts with eukaryotic fatty acid synthases (115). There is no FadR homol og in mammals. As such, FadR 82

PAGE 83

may be potential target for chemotherapy for V. vulnificus and other bacterial pathogens. It will also be interesting to investigate the role s of FabA and FabB, the enzymes required for unsaturated fatty acid synthesis, in V. vulnificus. If mutants in fabA and fadB are attenuated, this would further corroborate our evid ence that the defect of the fadR mutant was decreased fatty acid synthesis. Defined mutations in genes of the fad regulon will reveal the role of fatty acid oxidation in V. vulnificus These and other studies will, undoubtedly, extend the current understanding of what nutrients and metabolic pa thways are used by path ogens during infection. 83

PAGE 84

Figure 4-1. Skin and liver in fection caused by wild-type and fadR ::mini-Tn 5Km2phoA V. vulnificus in iron dextran-treated mice. Mice were inoculated s.c. with approximately 103 CFU of wild-type V. vulnificus FLA399 or 104, 105, 106, 107, or 108 CFU of fadR mutant FLA602. Up to 22 hours post-infection, mice were euthanized, and skin and liver were sampled for quantification of V. vulnificus Skin lesions were not observed with an inoculum of 104 CFU for the fadR mutant. Liver infection was only detected with an inoculum of 107 CFU. Fractions beneath the bars indicate the numbers of mice that yielded detectable numbers of bacteria from the skin or liver samples over the number of inoculated mice. Asterisks indicate statistical significance of CFU/g tissue or te mperature in mutant infections compared to wild-type by ANOVA with Bonferroni s Multiple Comparison Test (*, P 0.05; **, P 0.01; ***, P 0.001). Daggers indicate statis tical significance of numbers of samples yielding bacteria in the mutant infection compared to wild-type by 2 test (, P = 0.04; P = 0.002). The minimum detectable CF U for skin and liver samples are 103.0 and 102.5 CFU/g, respectively. 84

PAGE 85

V v MVIKAKSPAGFAEKYIIESIWNGRFPPGSILPAERELSELIGVTRTTLREVLQRLARDGW 60 V c MVIKAKSPAGFAEKYIIESIWNGRFPPGSILPAERELSELIGVTRTTLREVLQRLARDGW 60 Ec MVIKAQSPAGFAEEYIIESIWNNRFPPGTILPAERELSELIGVTRTTLREVLQRLARDGW 60 V v LTIQHGKPTKVNQFMETSGLHILDTLMTLDVDNATNIVEDLLAARTNISPIFMRYAFKVN 120 V c LTIQHGKPTKVNQFMETSGLHILDTLMTLDAENATSIVEDLLAARTNISPIFMRYAFKLN 120 Ec LTIQHGKPTKVNNFWETSGLNILETLARLDHESVPQLIDNLLSVRTNISTIFIRTAFRQH 120 V v KENSERTIKTVIDSCEQLVAAESWDAFLSSSPYADKIQQNVKEDNEKDEAKRQEILIAKT 180 V c KESAERIMINVIESCEALVNAPSWDAFIAASPYAEKIQQHVKEDSEKDELKRQEILIAKT 180 Ec PDKAQEVLATANEVADH----------------------------------------ADA 140 V v FNFYDYMLFQRLAFHSGNQIYGLIFNGLKKLYDRVGSFYFSNPASRELALKFYRQLLLTC 240 V c FNFYDYMLFQRLAFHSGNQIYGLIFNGLKKLYDRVGSYYFSNPQARELAMEFYRQLLAVC 240 Ec FAELDYNIFRGLAFASGNPIYGLILNGMKGLYTRIGRHYFANPEARSLALGFYHKLSALC 200 V v ESGQREQLPALIRQYGIESAMIWNEMKKQLPTNFTEDDC 279 V c QSGEREHLPQVIRQYGIASGHIWNQMKMTLPSNFTEDDC 279 Ec SEGAHDQVYETVRRYGHESGEIWHRMQKNLPGDLAIQGR 239 Figure 4-2. Alignments of FadR homol ogues. Amino acid sequence alignment of V. vulnificus FadR (Vv) and its homologues from V. cholerae (Vc), and E. coli (Ec). Unshaded amino acids are identical matches, light gr ay shading indicates conserved and semiconserved amino acid substitutions, and dark gray shading indicates unconserved changes in amino acid. The two Vibrio species share an identical 40-residue insertion in the middle of the FadR protein, relative to E. coli FadR. 85

PAGE 86

86 Table 4-1. Expression of puta tive FadR-regulated genes in a V. vulnificus fadR mutant. Gene Fold-change in mutant P value fabA -8.4 5.3 0.005 fabB -6.9 1.5 0.0002 fadD -1.6 2.1 0.1 fadB 4.2 2.4 0.04 qRT-PCR was used to measure fold-change as 2Ct using 16S rRNA as an endogenous control between fadR ::mini-Tn 5Km2phoA mutant V. vulnificus FLA602 and wild-type FLA399, mean standard deviation. P values are for paired Students t tests comparing normalized Ct between the wild-type and mutant strains. Three bi ological replicates were done for each gene.

PAGE 87

Table 4-2. In vitro growth rate and virulence of V. vulnificus fadR ::mini-Tn 5Km2phoA compared to ptsI ::mini-Tn 5Km2phoA and wild-type. Strain Genotype LB-N doubling time, min Log skin CFU Log liver CFU % of mice with detectable infection FLA602 fadR ::miniTn 5Km2 phoA 29 ND ND 0* FLA1000 ptsI ::miniTn5Km2 phoA 25 8.4 0.16 6.3 0.32 100 CMCP6 Wild-type 18 8.5 0.15 6.5 1.51 100 For virulence 5 iron dextran-treated mice we re inoculated s.c. with approximately 1,000 CFU, and infection was measured between 14 and 22 h later. ND not detected. 2 compared with FLA1000 and CMCP6, P = 0.002. 87

PAGE 88

Figure 4-3. Complementati on and reversion of the fadR ::mini-Tn 5Km2phoA mutation. Iron dextran-treated mice were infected s. c. with approximately 1,000 CFU of each V. vulnificus strain as detailed in the Material s and Methods. Enumeration of the complemented strain on non-sele ctive medium is shown. Th ere was not a statistically significant difference between CFU on selec tive and nonselective media. Because the fadR ::mini-Tn 5Km2phoA strain was avirulent at this in oculum and did not yield CFU, the minimum detectable values for skin and liver, 104 and 102.5 CFU/g, respectively, were assigned. Asterisks indicate sta tistical significance of CFU/g tissue or temperature in fadR ::mini-Tn 5Km2phoA infections compared to complemented, reverted, or wild-type infections by ANOVA with Bonferronis post test (*, P 0.006, **, P 0.001; ***, P 3 x 10-7.) Daggers indicate st atistical significance of numbers of samples yielding bacteria in the mutant skin infection compared to complemented, reverted, or wild-type strains and in the mutant liver infection compared to reverted, or wild-type strains by 2 test (, P 0.009). 88

PAGE 89

0 1 2 3 4 5 6 7 8 9 SKIN LIVER TEMPLog CFU/g Tissue and Temp/10 fadR fadR(fadR) WT 0/5 5/5 5/5 0/5 4/5 5/5*** ** ** Figure 4-4. Complementation of fadR mutation. Mice were inoculated with 1,000 CFU of fadR mutant FLA614, fadR -complemented strain FLA 614(pGTR349), or wild-type CMCP6. Enumeration of the complemented strain on nonselective medium is shown. There was not a statistically significant difference between CFU counts on selective and nonselective media. The fadR strain was avirulent at 1,000 CFU and was assigned the minimum detectable values. The minimum detectable CFU for skin and liver samples are 104 and 102.5 CFU/g, respectively. Fract ions beneath the bars indicate the proportion of samp les that yielded bacteria. Daggers indicate statistical significance of numbers of samples yielding bacteria in the mutant infection compared to wild-type or complemented strain by 2 test (, P 0.01; P = 0.002). *, P = 0.01; **, P = 4 10-3; ***, P 2 10-10 by ANOVA with Bonferronis post test for difference in mean CFU/g tissue or temperature of fadR mutant compared to wild-type or complemented strain. 89

PAGE 90

Table 4-3. Changes in fatty acid profile of fadR mutant compared to wild-type. Fatty acid type Relative abundance in sample % change in mutant P -value WT fadR Saturated 41.4 1.9 46.9 1.8 +12% 0.006 Unsaturated 57.8 2.5 51.1 1.8 -13% 0.005 Wild-type CMCP6 and fadR mutant FLA614 were grown to exponential phase in rich broth, pelleted, and washed. An equal wet mass of each strain was used for methyl esterification and gas chromatography-mass spectrometry. Relative abundance values represent the mean standard deviation of 4 i ndependent experiments. P values are for unpaired Students t tests comparing mutant and wild-type fatty acid abundance values. 90

PAGE 91

Table 4-4. fadR mutant sensitivity to envelope stress Envelope stress Wild-type MIC fadR MIC P -value Ethanol, % (vol/vol) 5 0.8 4 0.8 0.13 Polymyxin B, U/mL 192 38 61 217 0.41 SDS, % (wt/vol) 0.33 0.06 0.22 0.06 0.02 Wild-type CMCP6 and fadR mutant FLA614 were grown to e xponential phase in rich broth. The MICs for ethanol, SDS, and polymyxin B we re tested. Values represent the mean standard deviation of at leas t 4 independent experiments. P values are for unpaired Students t tests comparing mutant and wild-type MIC values. Except for a slight increase in sensitivity to SDS, the fadR mutant was not more sensitive to enve lope stresses than was the wild-type. 91

PAGE 92

WT fadR 37C WT fadR 42C -1 -3 -2 -4 -5 -6 WT fadR 4C/23C WT fadR 37C WT fadR 37C WT fadR 37C WT fadR 42C WT fadR 42C WT fadR 42C -1 -3 -2 -4 -5 -6 -1 -3 -2 -4 -5 -6 WT fadR 4C/23C WT fadR 4C/23C Figure 4-5. Sensitivity of wild-type CMCP6 and fadR FLA614 to heat and cold. Strains were grown to exponential phase at 37C, and serial dilutions were aliquoted onto LB-N plates and incubated at the temperatures indicated for 18 h. The plates incubated at 4C showed no visible growth after 18h a nd were photographed after an additional 18h incubation at room temperature. Images ar e representative of triplicate plates. FLA614 was not more sensitive to heat or cold than was the wild-type. 92

PAGE 93

0% 20% 40% 60% 80% 100% 120%Motility relative to wild-type fadR fadR(fadR) WT* ** Figure 4-6. Decreased motility of the fadR mutant. Static overnight cultures of FLA614 and CMCP6 were used to stab the center of 0.3% agar motility plates. The diameter of spread of bacteria was measured 15 h to 17 h later, and the mean of the wild-type was assigned a value of 100%. n = 6 plates per strain. At least 2 independent experiments were performed. *, P = 3 10-7; **, P = 3 10 -8 by ANOVA with Bonferronis multiple comparison test of fadR mutant compared to wild-type or complemented strain. 93

PAGE 94

* ** ** ** Figure 4-7. Supplementati on with oleate in vitro. 106 CFU of wild-type CMCP6 or fadR mutant FLA614 were inoculated into LB-N or LB-N + 0.005% (wt/vol) sodium oleate and grown shaking at 37 C. OD600 was measured every 30 minutes. Supplementation with oleate increased the growth rate and final yield of the fadR mutant to wild-type levels. Data shown are mean SD for two independent experiments. *, P < 0.01, **, P < 0.001 by Two-way ANOVA followed by Bonferronis post test for fadR mutant grown in LB-N compared with fadR mutant grown with olea te, wild-type grown in LB-N or wild-type grown with oleate, at each time-point. 94

PAGE 95

0 1 2 3 4 5 6 7 8 9 SKIN LIVER TEMPLog CFU/g tissue and temp/10 fadR fadR+oleate WT1/5 5/5 5/51/5 5/5 5/5* ** *** 0 1 2 3 4 5 6 7 8 9 SKIN LIVER TEMPLog CFU/g tissue and temp/10 fadR fadR+oleate WT1/5 5/5 5/51/5 5/5 5/5* ** ** *** # Figure 4-8. Oleate potentiate s infection of mice with fadR V. vulnificus Mice were inoculated s.c. with 1,000 CFU of fadR mutant FLA614 suspended in BSG + 0.225% (wt/vol) sodium oleate, FLA614 suspended in BSG alone, or wild-type CMCP6 suspended in BSG. Skin lesion and liver samples were plated to quantify infection. Daggers indicate statistical significance of numbers of samples yielding bacteria in the mutant infection compared to oleate-supplemented mutant or wild-type by 2 test (, P = 0.009). # P = 0.01 for mean CFU/g liver of fadR with oleate compared to WT. P = 0.01; **, P 0.003; ***, P 3 10-4 for mean CFU/g tissue or temperature of fadR mutant infections compared to wild-type or olea te-supplemented fadR by ANOVA with Bonferronis post test. 95

PAGE 96

Table 4-5. Growth of aceAB and fadD mutants and wild-type CMCP6 in oleate Final CFU/Input CFU M9 salts M9 + glucose M9 + oleate WT 13 75 78 aceAB 8 65 9 fadD 8 96 50 Strains were grown shaking overnight at 37C in the indicated media. Aliquots were plated on LB-N before and after overnight growth. The number of bacteria at 24 h was divided by the number of bacteria at the start of the experiment. None of the strains grew appreciably in M9 salts alone. The aceAB mutant did not grow better in M9 + oleate than in M9 salts. 96

PAGE 97

97 0 1 2 3 4 5 6 7 8 9 10 SKIN LIVERTEMPERATURELog CFU/g tissue and Temp/1 0 aceAB fadD WT Figure 4-9. Virulence of aceAB and fadD mutants of V. vulnificus. Mice were infected with 1,000 CFU of aceAB FLA1003, fadD FLA1006, or wild-type CMCP6. All mouse samples yielded bacteria. There was no st atistically significant difference between CFU/g or temperature values for the mu tants compared to the wild type.

PAGE 98

CHAPTER 5 THE ROLE OF RSEB AND THE EXTRAC YTOPLASMIC STRESS RESPONSE IN VIRULENCE OF V. VULNIFICUS Rationale for Study The use of PhoA mutagenesis to iden tify secreted virulence factors of V. vulnificus yielded a mutant, FLA609, that formed blue colonies on BCIP-containing media, indicating an in-frame fusion of phoA with a V. vulnificus gene encoding a secreted protein. Interestingly, FLA609 appeared less opaque than did the wild-type on LB-N plates and exhibited phase-variation between translucent and opaque morphologies. FL A609 was attenuated for virulence in the s.c. mouse model of V. vulnificus infection, indicating that the muta ted gene was possibly a virulence factor. Nucleotide sequence an alysis showed an in-frame phoA fusion to VV1_1561 encoding a protein designated as a Negative regulator of E activity. VV1_1561 was homologous to rseB a negative regulator of E (envelope stress response sigma factor) in E. coli (see below) To gain insight into the role of rseB and the envelope stress response in V. vulnificus, rseB mutant FLA609 was further characterized and defined mutations in E-related genes (rseB, rseA, rpoE, and degP ) were constructed for analysis of virule nce, colony morphology, and stress-associated phenotypes in the mutants. Introduction Envelope Stress Responses in Bacteria Proper folding and assembly of proteins in the bacterial cell enve lope depend on factors that are distinct from the cytoplasmi c heat shock chaperones GroEL/GroES and DnaK/DnaJ/GrpE (68). While misfolded cytoplasmi c proteins are sensed directly by the general heat shock transcription factor 32 (RpoH), misfolded periplasmic and outer membrane proteins are dealt with by four separate pathways: the CpxRA system, the BaeSR system, the phage shock protein regulon, and the E regulon (116-118). 98

PAGE 99

CpxRA is a classical two-component signal transduction system. CpxA, a membranespanning histidine kinase, senses the presence of misfolded polypeptides in the periplasm through its periplasmic sensory domain and relays signals via its cytoplasmic transmitter domain to the response regulator, CpxR (116). CpxR is a transcriptional regula tor that activates the expression of more than 100 genes whose products act on the bacterial enve lope (119). Some CpxAR-regulated gene pr oducts overlap with the E regulon (119,120), most notably the periplasmic protease, DegP. A second two-component signal transduction system, the BaeSR system, was recently reported to function as an envelope-stress pathwa y (121). The system is composed of BaeS, a sensor kinase, and BaeR, a re sponse regulator. The BaeSR sy stem overlaps with the CpxAR system in regulating expression of spheroplast protein Y ( spy gene product), a periplasmic protein whose expression is induced by spheroplast formation a nd various other envelope stress conditions (121). In addition, the BaeSR system upregulates the expression of several drug exporter genes (122,123). The phage shock protein (PSP) response was identified in E. coli due to the induction of the pspABCDE operon by protein IV (pIV), a secretin from filamentous phage f1 that forms a pore in the bacterial outer membrane (124). The psp operon is induced by several environmental stress conditions including protei n secretion defects and disruption of the proton motive force, and it is required for long-term stationary phase survival at alkaline pH (118). The overall function of the PSP system is proposed to be th e maintenance of the proton-motive force across the inner membrane (125). The extracytoplasmic function (ECF) family of alternative sigma fact ors consists of RNA polymerase sigma subunits that can associate with core RNA polymerase to generate an active 99

PAGE 100

complex that controls expression of genes enco ding proteins that func tion outside of the cytoplasm (118). ECF sigma factors have been identified in gram-negative, gram-positive, and acid-fast bacteria and function in responding to external stresses and in some cases, virulence (118). Function and Control of the E-Mediated ESR The best-studied ECF sigma factor is the E. coli E, encoded by the rpoE gene (126). E is an essential sigma factor in laboratory strains of E. coli (127), but is not essential in most bacteria (118). E controls an extensive regu lon of close to 100 genes in E. coli (118). Eregulated genes encode proteins involved re sponding to misfolded polypeptides (proteases, chaperones and folding enzymes), sigma factors ( rpoD rpoH rpoN ), enzymes involved in phospholipid or lipopolysaccharide (LPS) biogenesi s and/or modification, enzymes involved in primary metabolism, proteins with sensory or regulatory functions, a nd several proteins of unknown function (118). E also regulates the expression of the rpoErseABC operon encoding E and its regulators (118). The activity of E RNA polymerase is negatively regul ated by the membrane-spanning anti-sigma factor RseA and the periplasmic negative regulator RseB (68,128). In E. coli deletion of rseA leads to a 9-fold increase in transcription of E-dependent promoters, whereas deletion of rseB results in a 2-fold increase (128,129) suggesting that RseA is the more important regulator of E activity in E. coli. Under non-stress conditions, E is sequestered at the cytoplasmic face of the inner membrane by Rs eA. When inducing stress occurs, RseA is degraded by the successive act ion of several proteases: DegS, RseP (formerly known as YaeL), and ClpXP in a process known as regulated intramembrane proteoly sis that leads to the release of free E into the cytoplasm to bind to its cognate promoters and activate gene expression (118). RseB reinforces the interaction between RseA and E by binding RseA. RseB binding increases 100

PAGE 101

the affinity of RseA for E (130) and prevents sign al-independent proteolysis of RseA (131). In addition, Grigorova et al. (132) proposed that RseB could also function as a sensor of misfolded periplasmic proteins and that titra tion by unfolded proteins could be used as an alternative way to induce the E pathway. RseB, E, and Virulence in Bacterial Pathogens The only reports of an rseB homolog having a role in virulence involve mucB of P. aeruginosa. Alginate production in P. aeruginosa is controlled by AlgU/ E (133). Like E of E. coli E of P. aeruginosa is negatively controlled by the anti sigma factor, MucA (RseA homolog) and the periplasmic negative regulator MucB (RseB homolog). Mutations in MucA or MucB lead to the alginate overproducing, mucoid phenotype associated with the establishment of lethal respiratory infections in cystic fibrosis patients (134-137). In contrast, there are reports of the involvement of E in virulence of several bacterial pathogens including V. cholerae, S. enterica P. aeruginosa Haemophilus influenzae, and M. tuberculosis (118). In a recent review of the role of ESR systems in bacterial pathogens, Rowley, et al. (118) noted that the essentiality of a given ESR regulator or regulon member can vary between pathogens. Moreover, the importance of some ESR sigma factors can vary between stages of an infection cycle in a given pathogen. For example, AlgU (E) of P. aeruginosa is important for respiratory, but not systemic, infection (118). Thus, care should be taken when making conclusions about the role of these factors in pathogenesis models. Results Identification of an rseB ::mini-Tn 5Km2phoA Mutation in Vibrio vulnificus Nucleotide sequence analysis of PhoA mutant FLA609 showed an in-frame phoA fusion to VV1_1561 encoding a protein designated as a Negative regulator of E activity (Figure 5-1). Analysis of the amino acid sequence using PSORT (80) predicted with 93 .8% certainty that the 101

PAGE 102

protein was localized to the periplasm. BLAST searches showed that the VV1_1561 protein sequence had high levels of homology to the negative regulator of E from other Vibrio species: 72% identical to V. splendidus and 68% identical to V. cholerae Analysis of the genomic region surrounding VV1_1561 in V. vulnificus showed a possible operon including VV1_1559, designated as Similar to sigma24 VV1_1560, another Negative regulator of E activity, VV1_1561, and VV1_1562, Positive regulator of E activity (Figure. 5-1). This operon structure was identical to that of the well-studied E. coli rpoE operon, and VV1_1560 was 45% identical to the E. coli protein RseB, a periplas mic negative regulator of E activity. As such, we named VV1_1560 rseB. Virulence of RseB Mutants. As noted, the V. vulnificus rseB:: mini-Tn 5Km2phoA mutant, FLA609, appeared less opaque than wild-type on LB-N pl ates. This suggested that it c ould have been a capsule mutant, as acapsular mutants of V. vulnificus appear translucent (22). S ubculturing of FLA609 showed that it was not homogeneous, but yielded coloni es of translucent and opaque morphologies (Figure 5-2). The opaque colonies were initially thought to be contaminants, but further passaging of single translucent co lonies always yielded a mixture of translucent and opaque colonies (Figure 5-2). The frequency of obtai ning opaque variants from a translucent colony ranged from 20% to over 90%. It appeared th at FLA609 was undergoing high-frequency phasevariation from translucent to opaque forms. The initial test for vi rulence involved s.c. infection of three mice with approximately 1,000 CFU of the translucent variant of FLA609. Considerable switching in colony morphology apparently occurred while growing the culture for inoculation, because pla ting indicated that the inoculum yielded approximately 10% opaque coloni es. One of the three inoculated mice failed to develop skin lesions. Of the two that devel oped a skin lesion, bacteria were recovered from 102

PAGE 103

only one skin sample. None of the liver samples yielded detectable CFU. In the one mouse that yielded bacteria from the skin, a mixture of tr anslucent and opaque bacteria was observed on the plate, yielding 107.8 CFU/g skin (Figure 5-3). Because th e other two mice were below the limit of detection for skin CFU and were assigned a minimum CFU count of 1 on the plate, the mean skin CFU for the experiment was 105 CFU/g skin (Figure 5-3). Li kewise, the liver samples did not yield detectable bacteria and were each assi gned a minimum detectable CFU count of 1 for each plate, making the mean CFU/g liver 102.5. FLA609 was attenuated for virulence, with an approximately 3-log decrease in mean skin CF U and a 2-log decrease in mean liver CFU compared to wild-type (108 CFU/g skin and 104.7 CFU/g liver). Also, considerable switching to the opaque morphology occurred in vivo, suggesting th at there was selective pressure against the translucent form (Figure 5-4). In subsequent infections the translucent variant of rseB ::mini-Tn 5Km2phoA (FLA609-Tl) was not significantly attenuated fo r local (skin) infection but was attenuated for systemic (liver) infection compared to wild-type, with a nearly 2-log decrease compared to wild-type FLA399, P = 0.01 (Figure.5-5). In addition, bacteria harves ted from mice infected with FLA609-Tl were always enriched for opaque coloni es, indicating that th e opaque variant was more virulent than the translucent variant. In fact, translucent bacteria were almost never recovered from the liver samples of mice; apparently only those that switched to opaque were capable of systemic infection. The opaque variant of FLA609 (FLA609-O) showed wild-type levels of skin infection but variable levels of systemic infection. When three 3 separate infections with FLA610-O were combined for analysis ( n = 14 mice), the mean log CFU/g skin was 8.1 0.25 while the mean log CFU/g liver was 3.8 1.8, showing a large degree of variation (s tandard deviation was nearly 50% of the mean) (Figure 5-6). Moreove r, skin CFU were rec overed from 100% of mice 103

PAGE 104

inoculated with FLA609-O (14/14) while liver CFU were recovere d from only 50% of mice. In the 50% of mice that yielded liver CFU the mean recovery was near, or sometimes above, typical wild-type levels. For exampl e, three mice yielded over 106 CFU/g liver (Figure 5-3). To complete the molecular version of Koch s postulates (21), we expressed wild-type rseB on plasmid pGTR2005 in a translucent variant of FLA609 to try to complement the mutation. Complementation in trans did not restore wild-type col ony morphology to the translucent rseB variant (Figure 5-7). The comple mented strain continued to exhi bit phase variation. In fact, a third morphology arose; this new phase variant wa s even more translucen t than the translucent parental strain, and was designate d transparent (Tp) (Figure 5-8). Virulence of the translucent and opaque derivatives of the complemented st rain was tested in mice. The translucent complemented strain appeared to have wild-type virulence in mice, with a mean yield of 107.9 CFU/g skin and 104.6 CFU/g liver. Also, 4 of 4 mice yielde d skin CFU, and 3 of 4 mice yielded liver CFU. This was in contrast to the translucent rseB ::mini-Tn 5Km2phoA mutant that was recovered from less than 50% of mice infected (Figure 5-8). Mo st notably, the complemented translucent strain yielded translucent bacteria from the liver (30100% translucent), whereas the translucent rseB ::mini-Tn 5Km2phoA mutant usually underwent exte nsive phase variation in vivo resulting in nearly 100% opaque colonies whenever liver CFU were detected. Surprisingly, the opaque derivative of the complemented rseB ::mini-Tn 5Km2phoA mutant did not show wild-type virulence (Figure 5-8). While skin CFU were in the rang e expected for wild-type (108.2 CFU/g; 100% of samples yielded bacteria), the mean CFU/g liver recovered was 103.1, and only 33% of liver samples yielded detectable bacteria. Ther efore, the opaque form of the complemented rseB ::mini-Tn 5Km2phoA strain was not restored to wild-type virulence. 104

PAGE 105

Deletion of rseB from Wild-Type CMCP6. Because complementation did not restore wild-type morphology or virulence to FLA609, we needed to confirm that the phenotypes of FLA609 were indeed due to the mutation in rseB and not due to secondary mutations or polar effects. Thus, a targeted deletion of rseB was constructed in wild-type V. vulnificus CMCP6 as detailed in th e experimental procedures, creating FLA610. Deletion of rseB from CMCP6 recapitulated the phase-variable phenotype observed with the rseB ::mini-Tn 5Km2phoA mutant. While the original mutant appeared slightly less opaque than wild-type on LB -N agar plates, subculturing showed that the colony was composed of mixed morphologies with various leve ls of opacity. Some variants were as opaque as the wild-type and designated FLA610-O; some were slightly less op aque than wild-type (similar to FLA609-Tl) and designated FLA610Tl, and others were even less opaque and described as transparent, FLA610-Tp. Virulence of rseB FLA610 variants was tested in mice (Figure 5-9). The most translucent form of rseB FLA610-Tp, was avirulent in mice at an inoculum of 1,000; none of the five mice inoculated developed skin lesions and the mice remained healthy throughout the 20 h course of infection. The rseB translucent variant was rec overed from only 2 of 5 mice, yielding a mean of 105.3 CFU/g skin and 103.7 CFU/g liver. As seen with the translucent rseB ::mini-Tn 5Km2phoA mutant, a mixture of translucent and opaque bacteria was recovered from the tissue samples of mice infected with FLA610-Tl (20-50% O). The opaque variant, FLA610-O was also recovered from only 2 of 5 mice, yielding a mean of 105 CFU/g skin lesion and 103.2 CFU/g liver. As seen for the opaque variant of rseB ::mini-Tn 5Km2phoA subsequent infections showed that FLA610-O had variable levels of virulen ce: fully virulent in 100% of mice in some infections, but significantly attenuat ed for virulence in othe r infections (Figure 5105

PAGE 106

10). The mean CFU/g skin for a total of 20 mice inoculated with FLA610-O (four separate infections) was 106.4, nearly two logs lower than that seen in a typica l wild-type infection. The mean liver CFU/g for 20 mice infected with rseB FLA610-O was 104.5 which is in the range expected for wild-type infection. Therefore, the opaque variant of rseB showed an overall attenuation for skin infection, but not an appreciable decrease in liv er infection. We believe that rseB -O may have been attenuated for liver infection, but this wa s not reflected in the mean CFU/g liver possibly because the minimum detectible value of 102.5 CFU/g was used for nearly half of the mice, and this value may be an overestimation of the true minimum. Because the transparent variant of rseB was avirulent at an inoculum of 1,000 CFU (three times the wild-type minimum lethal dose), we want ed to determine the extent of the attenuation. We infected mice with 105 CFU of FLA610-Tp. For comparis on, a mutant of CMCP6 that was similar in morphology to FLA610-Tp due to a mutation in the wzb capsule transport gene (CMIT232) was also assessed for vi rulence at an inoculum of 105 CFU. Mice infected with 105 CFU of FLA610-Tp yielded 7.9 0.3 log CFU/g skin (significantly higher compared to CMIT232 ( P = 0.001); and significantly lower compared to CMCP6 (P = 0.009) and 2.4 0.3 log CFU/g liver (Figure 5-11). Li ver CFU were recovered from onl y one of five mice inoculated with 105 CFU of FLA610-Tp, while skin CFU were rec overed from all five mice. Mice infected with 105 CFU of CMIT232 yielded 5.1 1.1 log CFU/g skin ( P = 8 10-5 compared to CMCP6), and bacteria were recovered from only three of five skin samples. For CMIT232, none of the liver samples yielded bacteria, so the minimum detectible CFU value (1 CFU) was assigned to each plate, making the log CFU/g liver 2.3 0.2 ( P = 0.79 compared to FLA610-Tp; P = 0.001 compared to CMCP6). FLA610-Tp was as atte nuated as the non-encapsulated mutant CMIT232 for liver, but not skin, infection. However, FLA 610-Tp was still significantly attenuated for both 106

PAGE 107

skin and liver infection compared to wild-type, with 100 times th e wild-type inoculum resulting in a two-log decrease in mean liver CFU compared to wild-type. Complementation in trans with cloned wild-type rseB did not restore wild-type colony morphology to rseB variants: phase va riation continued to be observed. Virulence of the complemented rseB variants was tested in mice at an inoculum of 1,000 CFU (Figure 5-12). The complemented transparent variant was recovered from the skin lesion of one of five mice. The skin sample yielded 105.2 CFU/g, still well below the expect ed yield for a wild-type infection (approximately 108 CFU/g skin). All five mice infected with the complemented transparent variant remained healthy in appearance and had normal temperatures (mean 37.2 0.6C) throughout the 22 h course of infection. Thus, this rseB -Tp was not restored to wild-type virulence by complementation. The complemented translucent variant was recovered from four of five skin and liver samples, yielding 106.1 CFU/g skin and 102.4 CFU/g liver. The mean skin and liver CFU obtained were not statistically significantly different from those of the uncomplemented translucent mutant, so complementation of rseB did not significantly increase its virulence. The complemented opaque varian t was recovered from 100% of mouse skin and liver samples, yielding 108 CFU/g skin and 105.3 CFU/g liver. The mean skin and liver CFU for the complemented opaque variant showed a statis tically significant increase compared to the uncomplemented opaque rseB variant ( P = 0.012 for skin CFU and P = 0.008 for liver CFU). Taken together, complementation in trans with wild-type rseB did not restore wild-type colony morphology and did not uniformly re store wild-type virulence to rseB variants. However, rseB sufficiently mirrored rseB ::mini-Tn 5Km2phoA to allow confident assumption that the phenotypes of the mutants were due to a mutation in rseB. 107

PAGE 108

Characterization of RseB Phase Variants Analysis of phase variation. The apparent switching of the rseB mutants from translucent to opaque morphologies was reminiscent of translucent V. vulnificus CPS mutants that spontane ously emerge from opaque wild-type strains (24). Spont aneous CPS mutants arise in V. vulnificus at a rate of 10-4, and the reverse phase variation from translucent to opaque occurs at a rate of 9 x 10-3 (24). To assess the rates of phase variation in V. vulnificus rseB mutants compared to wild-type, we grew wildtype CMCP6 and rseB FLA610 variants to log phase from st atic overnight cult ures inoculated from a single colony. Aliquots were diluted and plated on LB-N ( n 6) plates per strain) for observation. Wild-type CMCP6 switched from an opaque to a translucent morphology at a rate of 0.004%. rseB -O switched to a translucent morphol ogy at a rate of 0.01%, more than 10 times as frequently as did the wild-type. The reverse switch, from rseB -Tl to O occurred at a rate of 0.092. Because the rates of phase-variation from translucent to opaque and vice versa in rseB FLA610 were nearly 100 times higher than those reported for spontaneous capsule mutants, the underlying mechanism of phase vari ation between translucent and opaque morphologies in rseB mutants may be different fr om that of CPS mutants. CPS and carbohydrate expression. Of great interest was finding the reason for the translucent appearance and high rates of phase variation of the rseB mutants. Capsule mutants of V. vulnificus have a translucent appearance due to decreased or absent CPS ( 22). Thus, a decrease in CPS expression could explain the colony morphology of th e translucent and transparent rseB variants. We obtained an acapsular transposon mutant of V. vulnificus MO6/24-O, CVD752, that has a polar transposon insertion in the wza gene located near the beginning of the CPS operon (138). Because the mutation in CVD752 had both a selectable antibiotic re sistance marker and a 108

PAGE 109

screenable phenotype, we were able to introduce the transposon insertion mutation into wild-type V. vulnificus CMCP6 by allelic exchange via chitin-induced transfor mation using genomic DNA from CVD752. Briefly, genomic DNA was extracted from C VD752 and sheared by vigorous vortexing. The genomic DNA was incubated with CM CP6 in the presence of sterile crab shell according to the chitin transformation protocol detailed in the Materials and Methods. Recombinants were selected on LB-N agar containing 100 g/mL kanamycin. All transformants were translucent like the acapsular MO6/24-O C VD752. A representative acapsular mutant of CMCP6 was named FLA1009 (Figure 5-13). FLA1009 appeared translucent compared to wildtype CMCP6, and was comparable in morphology to rseB FLA610-Tl, but was not quite as transparent as FLA610-Tp on LB-N agar (Figur e 5-13). The CPS mutation was confirmed using PCR with oligonucleo tides targeting the wza open reading frame (CMCP6wza5 and CMCP6wza3), and FLA1009 was used as a nega tive control for carbohydr ate binding and other functional tests of CPS expression. FLA609 ( rseB ::mini-Tn 5Km2phoA ) and FLA610 ( rseB ) were compared with wild-type CMCP6 and acapsular mutant FLA1009 in several test s for extracellular polysaccharides. If the translucent or transparent rseB variants behaved like the acapsular mutant in these tests, it would suggest that the rseB deletion somehow caused decreased CPS expression; otherwise, it was possible that the rseB variants had decreased opacity i ndependent of capsule changes (for example, changes in non-CPS extracellular polys accharides). A spontaneous rugose mutant of CMCP6, FLA611-R, that had a morphology resemb ling the rugose EPS described by Grau and Pettis (139) was also used in the ca rbohydrate analyses for comparison to rseB mutants. rseB mutants, wild-type CMCP6, acapsular FLA1009, and rugose FLA611-R were plated on LB-N agar containing the carbohydrate-bind ing dye Congo red (140) to observe changes in 109

PAGE 110

carbohydrate levels (Figur e 5-14). Wild-type, rseB -O, and rugose all appeared bright red on Congo red agar, indicating comparable levels of binding of the dye. The CPSmutant was slightly less red than the wild-type. rseB -Tl and rseB -Tp were light pink on Congo red agar, indicating a low level of binding. rseB -Tp was only slightly less red than rseB -Tl when streaked on Congo red agar. On this medium, non-opaque rseB strains (Tp and Tl) resembled each other more than they di d the acapsular mutant. The ability of the strains to bind the fluores cent stain, Calcofluor was also observed by plating. Calcofluor white stain is a non-specifi c fluorochrome that binds with cellulose, chitin, and other carbohydrate moieties. It has been extensively used in clinical mycology for the detection of fungal cell walls, a nd binding of Calcofluor (evide nced by increased fluorescence under UV light) is an indicator of ex tracellular carbohydrate content. rseB variants, the CPSmutant, rugose mutant, and wild-type were plated on LB-N agar containing calcofluor. Plates were incubated overnight at 37C and then at room temperature for 24 h, and images were captured before and after exposure to UV light (F igure 5-15). The tran slucent and transparent rseB variants did not appear to fluoresce in the presence of calcofluor exposed to UV light. The CPSstrain fluoresced, but at a lower le vel than was seen for the wild-type, rseB-O, or the rugose mutant. Therefore, as with Congo red, plating on calcofluor indicated that the nonopaque rseB variants were more similar to eac h other than to an acapsular mutant. Whole cell lysates of the rseB mutants and wild-type and acapsular controls were resolved by SDS-PAGE and stained with the catio nic carbocyanine dye 1-ethyl-2-[3-(1ethylnaphtho(1,2d)-thiazolin-2-ylidene)-2-methylpropenyl]-napht ho(1,2d)thiazolium bromide (Stains-All) that differentially stains various cellular compone nts (protein, carbohydrate, nucleic acid). A darkly staining pur ple region was visible in the translucent and opaque rseB variants 110

PAGE 111

and wild-type, but absent from the CPSmutant and the transparent rseB variant (Figure 5-16). It was inferred that this dark region represen ted CPS because it was absent in the acapsular mutant. Because the darkly staining regi on was also absent from the transparent rseB variant, it appeared that rseB -Tp mutant might also be acapsular. The translucent rseB variant showed a slightly decreased level of dark staining compared to wild-type or rseB -O (Figure 5-16), suggesting that the d ecreased opacity of rseB -Tl might be due to decreased CPS. To further confirm that the darkly staining region on the Stains-all gel was polysaccharide, we extracted extracellular polys accharides using the EPS extraction method of Enos-Berlage and McCarter (62). The extracted EPS preparations were resolved by SDS-PAGE and stained with Stains-All or with Alcian blue, a soluble copper-phthalocyanine dye that binds carbohydrates and has been used to demonstrate bacterial capsules, including V. vulnificus (141). Purified EPS extracts from rseB transparent (Tp), translucent (Tl), and opaque (O) variants, acapsular FLA1009 (CPS-), rugose (R), and wild-type CMCP 6 (WT) were observed alongside EPS extracts from two strains that appear more opaque than wild-type on LB-N agar, rpoE FLA1001 and degP FLA1002 (Figure 5-17). The degP and rpoE mutants will be discussed in detail in later sections. E PS extracts of the transparent rseB variant (Tp) did not stain appreciably with Alcian blue or wi th Stains-all, similar to the CPSmutant. rseBTl stained slightly better with Alcian blue and Stains-all than did rseB-Tp or the CPSmutant, but did not stain as well as did the wild-type. EPS from rpoE and degP that appear more opaque than wild-type on LB-N did not stain significantly more with Alcian blue or Stains-all compared to wild-type. EPS from the rugose strain bound Alcian blue and Stains-all mo re strongly than did EPS from any other strain, indi cating that the rugose strain may be overexpressing EPS or may contain a different EPS composition from the wild -type. Overall, analys is of EPS from the 111

PAGE 112

rseB variants compared to wild-type and an acapsu lar mutant suggested that the translucent and transparent variants may have d ecreased opacity due to loss of or decreased amounts of capsule. Chatzidaki-Livanis and associates (142) noted that genetic rearrangements in V. vulnificus CPS loci could result in loss of or decreased capsule expre ssion. They reported that distinct genotypes could be observed for translucen t bacteria based on a PCR method using oligonucleotides that spanne d the CPS transport genes wza wzb wzc One genotype did not amplify with the primers (designated TR3), indi cating a more extensive deletion than could be detected by this PCR. Another geno type had a 435-nt deletion of the wzb gene (designated TR2). A third genotype, TR1, had an intact C PS operon and represente d the only translucent genotype that exhibited a phase-var iable phenotype (142). To rule out the possibility that genetic polymorphisms in the capsule region were re sponsible for the translucent phenotypes of rseB mutants, PCR was performed as was done by Ch atzidaki-Livanis and coworkers (142) using oligonucleotide primers CMCP6wza5 and CMCP6wzc3 that were designed to target the CPS operon of V. vulnificus CMCP6. Genomic DNA templates from rseB mutants FLA610-Tp, -Tl, and -O all generated the same PCR amplicon as did wild-type CMCP6 (expected wild-type amplicon was 4.8 kb) indicating that no gross gene tic rearrangements had occurred in the group1 CPS transport genes (Figure 5-18). In contrast, FLA1009 with a Tn phoA insertion in wza did not amplify the wild-type amplicon due to the insertion of the 7.7-kb transposon. This result suggested that the tran slucent and transparent rseB variants could be TR1. The capsule of V. vulnificus inhibits biofilm formation, so acapsular mutants produce significantly more biofilm than do encapsulated strains (143). We compared the biofilm-forming ability of the transparent, translucent, and opaque rseB variants to wild-type CMCP6, acapsular FLA1009, and the spontaneous rugose mutant FLA611-R. The rugose mutant was expected to 112

PAGE 113

form very robust biofilms based on observati ons by Grau and coworkers (139) that rugose V. vulnificus form significantly more biofilm than wild -type. We hypothesized that the translucent and transparent variants of rseB would form more dense biofilms than the wild-type because we believed that the decrease in opacity of these varian ts was due to decreased CPS. To assess biofilm formation, strains were grown in LB-N in wells of a 96-well polystyrene plate at 30C for 48h without agitation. The wells were washed and stained with 1% (w t/vol) crystal violet. The crystal violet was dissolved in ethanol and removed to a fresh plate, and the absorbance was read on a plate reader at 630nm wavelength (Figure 5-19 A). It is important to note that V. vulnificus, in general, did not form very adherent biofilms, and bacteria were very easily washed away from the wells. This was particularly true for the rugose mutant (Rug) that often produced the most dense biofilms in plates (Figure 5-19 B), but that was often lost by washing before the biofilms could be stained and quantified. Thus, th e rugose mutant had the greatest variability in overall absorbance measurements (Figure 5-19 A) despite perhaps being the producer of the most dense biofilms. The transparent strain FL A610-Tp consistently form ed more substantial biofilms than did the wild-type and was similar in biofilm-forming ability to the acapsular mutant, FLA1009 (Figure 5-19 A). The translucent (FLA610-Tl) and opaque (FLA610-O) rseB mutants were not significantly different fr om wild-type CMCP6 in biofilm formation. The similarity in biofilm-forming ability between the transparent rseB mutant (FLA610-Tp) and the acapsular FLA1009 suggested that the transparent rseB mutant might express low amounts, if any, CPS. The observati on that the translucent rseB variant did not produce more biofilm than the wild-type was surprising in light of the fact that rseB -Tl appeared to express lower levels of CPS than did the wild-type. It was possible that while the translucent rseB variant expressed decreased CPS, there was still e nough present in the ce ll to inhibit significant biofilm formation. 113

PAGE 114

Capsule mutants of V. vulnificus are more sensitive to serum complement than are wildtype bacteria (144). As a fu rther test for encapsulation of rseB mutants, we assayed the sensitivity of opaque, tran slucent, and transparent rseB mutants, opaque wild-type CMCP6, and acapsular FLA1009 to 90% (vol/vol) normal rat serum. A serum-sensitive E. coli strain MG1655 was used as a positive control for co mplement-mediated killing. Strains were incubated for 2h at 37C in intact serum or in se rum that had been heat inactivated for 30 min at 56C. While wild-type CMCP6, FLA610-Tl, and FLA610-O survived as well in intact serum as in heat-inactivated serum, the capsule mutant FLA1009 and transparent FLA610-Tp were complement-sensitive ( Table 5-1). In this respect, FLA610-Tp behaved like an acapsular mutant. In contrast, FLA610-Tl did not mirro r the acapsular mutant. As with the biofilm formation assay, these results suggested that rseB -TL expressed some CPS despite its translucent morphology and that this amount of CPS was enough to confer wild-type characteristics to the translucent rseB variant. Stress resistance and other phenotypes of rseB variants. E that is negatively regulated by RseB is essential for resistance to environmental stresses in many bacteria (145). We hypothesized that rseB mutants would show altered re sponses to external stresses compared to wild-type, possibly due to dysregulation of the E-mediated envelope stress response. Disk diffusion assays were performed to assess the sensitivity of rseB mutants to membrane-perturbing agents. There were no st atistically significant differences between the rseB variants and wild-type in sensitivity to 10 0% (vol/vol) ethanol or to 3% (vol/vol) hydrogen peroxide (Table 5-2). We also tested the ability of wild-type CMCP6 and of the rseB CPS-, and rugose mutants to grow at 42C to see if there was a correlation between colony morphology and heat sensitivity. Neither the rseB variants nor the CPSmutant showed 114

PAGE 115

increased sensitivity to incubation at 42C compar ed to wild-type (not shown). Conversely, the rugose mutant showed a marked decrease in viabi lity when incubated at 42C. Therefore, there was not a correlation between colony opacity and h eat resistance. Rather, the presence of rugose EPS or some pleiotropic defect related to the rugose phenot ype appeared to cause heatsensitivity. Taken together, these results indicated that rseB is not essential for resistance to envelope stresses. We tested if the rseB mutation affected motility because the flagellar motor and accessory proteins are intimately associated with bacterial membranes (95); thus, any changes in membrane integrity or function may disrupt motility. None of the rseB variants had significant changes in the ability to swim through 0.3% (wt/vol) agar. Conversely, the acapsular mutant, FLA1009, was hyper-motile and showed a 23% increase in motility ( P = 2 10-4) compared to wild-type (Figure 5-20). In this respect, the rseB mutants were phenotypically different from an acapsular mutant. This finding, along with the distinctions seen on Congo red and Calcofluor agar plates, indicated only some of the rseB phenotypes could be due to decreased CPS; for example, the decreased binding of Congo red a nd Calcofluor and the decreased EPS staining with Alcian blue and Stains-all may be due to due to decreased CPS. However, only the transparent rseB variant showed similarity to a CPS mutant in other assays, including biofilm formation and sensitivity to serum complement. None of the rseB variants shared the hypermotile phenotype of the CPS mutant. Therefore, there were aspects of rseB variants that were not due to a lack of CPS. rpoE expression in rseB mutants RseB is a negative regulator of E in E. coli (116). To test if V. vulnificus rseB had a similar function, rpoE expression was measured in rseB mutants compared to wild-type 115

PAGE 116

CMCP6 by qRT-PCR using mRNA isol ated from exponential-phase cells grown in LB-N. RseB negatively regulates E in a post-translational manner in concert with RseA by tethering E to the inner membrane (116). However, changes in the amount E can be measured at the transcriptional level because rpoE auto regulates its own transc ription (146). Thus, in the absence of RseB, there should be elevated levels of free E in the cell and increased transcription of rpoE Consistent with the expected phenotype for rseB mutants, V. vulnificus rseB variants had significantly higher rpoE levels than the wild-type ( Table 5-3). rpoE levels were increased 24 4, 20 2, and 13 3-fold in FLA610-Tp, Tl, and O, respectively, compared to wild-type ( P 0.0001 for Ct values in each mutant compared to wild-type). This result was evidence that V. vulnificus rseB negatively regulates E. As a further test for negative regulation of E in rseB mutants, we tested expression of a potential E-regulated gene, VV1_0603, a homolog of degP. Expression of degP is positively regulated by E in E. coli and other bacteria (116). Given that rpoE levels were increased in the rseB mutants due to the lack of the negative regulator, VV1_0603 levels were expected to increase. We obs erved large increases (28 to 100-fold) in expression of VV1_0603 in the rseB variants compared to wild-t ype (Table 5-3). This finding was further evidence that V. vulnificus RseB negatively regulated E. Effects of rpoE overexpression in wild-type V. vulnificus Because we confirmed that V. vulnificus rseB mutants had increased levels of rpoE we hypothesized that the rseB phenotypes were due to E overexpression. Furthermore, we expected that overexpression of E in wild-type V. vulnificus would recapitulate rseB phenotypes. V. vulnificus rpoE (VV1_1559) was PCR amplified from CMCP6 genomic DNA using primers SigmaE5rbs-USER and SigmaE 3rbs-USER and cloned into pGTR1160 for expression from the lac promoter. The resulting plas mid, pGTR2008, was introduced into CMCP6 by conjugation. The rpoE -overexpressing strain, CMCP6(pGTR2008), appeared 116

PAGE 117

identical to wild-type CMCP6 on LB-N agar plates. There were no differences in carbohydrate binding observed by plating on Calcof luor or Congo red agar plates (not shown). These results suggested that the rseB mutant phenotypes might be independent of E or that the level of E expression in CMCP6(pGTR2008) was different from that achieved by rseB mutants. Despite the inability to recapitulate the rseB morphology by overexpressing rpoE in CMCP6, we tested the virulence of CMCP6(pGTR2008) in mice (Fi gure 5-21). Approximately 300 CFU of wildtype CMCP6 and rpoE -overexpressing CMCP6(pGTR2008) were inoculated into mice. Because any curing of plasmid pGTR2008 from the rpoE -overexpressing strain would result in potentially virulent wild-type bacteria, a lo w inoculum of CMCP6(pGTR2008) was used to preclude inoculation of a large number of wild-type bacteria. Indeed, plating of the inoculum of CMCP6(pGTR2008) non-selectively on LB-N showed an inoculum of 340 bacteria, whereas plating selectively LB-N with te tracycline showed that only half of the inoculum (170 bacteria) was Tetr (i.e., contained the plasmid). Nonetheless, the rpoE -overexpressing strain showed decreased systemic virulence compared to the w ild-type, with nearly 3 logs lower CFU/g liver than the wild-type (P = 0.03). Similarly, the temperatur es of mice infected with the rpoE overexpressing strain (37.1 2.4C) were higher than those of mice inf ected with wild-type bacteria (30.3 4.2), P = 0.01. Skin CFU for mice inoculated with rpoE -overexpressing or wildtype bacteria were similar (log 7.2 0.7 CFU/g, log 7.7 1.4 CF U/g, respectively). Thus, as with the rseB mutants, overexpression of rpoE in wild-type CMCP6 caused attenuated virulence. Still, these results were not suffici ent to prove that the al tered colony morphology and decreased virulence of rseB mutants were due to overexpression of rpoE We observed elevated levels of rpoE in rseB mutants and decreased virulence in an rpoE -overexpressing strain, but lacked a definitive link between the two observations. 117

PAGE 118

Role of the E-Mediated Extracytoplasmic Stress Response in V. vulnificus Characterization of a V. vulnificus rpoE mutant E is important for virulence and stress resistan ce in several bacterial pathogens including V. cholerae, S. enterica and P. aeruginosa (118). To probe the role of the E-mediated extracytoplasmic stress response in V. vulnificus, rpoE was deleted from wild-type CMCP6 by allelic exchange via chitin-induced transformation. Th ere was a distinct morphological difference between rpoE isolates and wild-type colonies on LB-N-agar; rpoE mutants had a whitish appearance that suggested they might be more heavily encapsulated than wild-type CMCP6 (Figure 5-22). A representative rpoE mutant was named FLA1001. Several reports indicate that rpoE mutants in gram-negative bacteria are more susceptible to environmental stresses (reviewed in (118)). V. vulnificus rpoE mutant FLA1001 was compared to wild-type CMCP6 for sensitivity to envelope stresses via disc diffusion assays ( Table 5-4). FLA1001 had a 15% increase in sensitivity to 100% (vol/vol) ethanol ( P = 0.006), a 29% increase in sensitivity to 3% (vol/vol) hydrogen peroxide ( P = 10-07), and a 25% increase in sensitivity to 1% (wt/vol) SDS (P = 3-07). These results indicated that rpoE is essential for resistance to envelope-perturbing stresses in V. vulnificus. Because E was identified as a heat shock regulator in E. coli we tested heat resistance of rpoE FLA1001 compared to wild-type by plating serial dilutions of exponential-phase bacteria on LB-N followed by 18 h incubation at 42C or 37C. The rpoE mutant showed an increase in heat sensitivity compared to wild-type ( Figure 5-23.) The increased sensitivity of FLA1001 to incubation at 42C was particularly evident when the 10-3, 10-4, and 10-5 dilutions of FLA1001 were compared to those of CMCP6. Conversely, there was no differen ce between wild-type CMCP6 and rpoE FLA1001 in plating efficiency when the strains were incubated for 18h at 4C followed by overnight incubation at room temperature, suggesting that the mutant wa s not hypersensitive to cold. The fact that 118

PAGE 119

FLA1001 was sensitive to heat and to envelope stresses but not to cold indicated that E was not needed for responding to all types of envelope stresses. Because the rpoE mutant showed significant increases in envelope stress sensitivit y, we tested the motility of FLA1001 to see if the rpoE phenotype extended to this important membrane-associated function. The rpoE mutant had a 20% decrease in motility ( P = 4 10-7) compared to wild-type ( Figure 5-24). Virulence of a V. vulnificus rpoE mutant As FLA1001 showed significantly increased susceptibility to envelope stresses compared to wild-type, this mutant was hypot hesized to be less fit to surviv e the harsh environment in the mouse host. At an inoculum of 300 CFU FLA1 001 showed a decrease in mean skin (106.1 CFU/g) and liver (103.2 CFU/g) infection compared to wi ld-type inoculated at 300 CFU (mean 107.7 CFU/g skin and 105.1 CFU/g liver) (Figure 5-25). This represented a nearly 100-fold decrease in both skin and liver CFU compared to wild-type. However, these values were not statistically significant by Students t test due to variability in th e data. At an inoculum of 2,000 CFU, FLA1001 exhibited full virulence compar ed to the wild-type infected at 3,000 CFU ( Figure 5-25). The mean recovery fr om FLA1001 infected at 2,000 CFU was 108.2 CFU/g skin and 106 CFU/g liver. It appears, therefore, that E does not have a major role in virulence in V. vulnificus. Overall, E was important for resistance to envelope stresses and for motility, but was dispensable for virulence of V. vulnificus in s.c. inoculated mice. The levels of E in V. vulnificus cells appeared to be tightly regulated, as any changes in rpoE expression ( rseB mutation, overexpression of cloned rpoE in CMCP6, or deletion of rpoE ) caused deleterious effects. While the possibility remained that E could be needed for virulence via another route of 119

PAGE 120

infection, it was clear that E was essential for surviving ex ternal stresses, and thus may represent an essential envir onmental survival factor. Identification of V. vulnificus degP An important mediator of the ESR is the pe riplasmic protease DegP, also known as HtrA and DO (106,126,147,148). DegP helps maintain the integrity of periplasmic and membrane proteins by binding and degrading mi sfolded proteins caused by envel ope stress. DegP also acts as a chaperone to direct proper folding of some e nvelope proteins (148,149). The lack of data on the role of the E-mediated ESR in the Vibrio family led us to investigate the role of DegP as a possible effector of the ESR in V. vulnificus To find the V. vulnificus homolog of DegP, BLAST search es were performed using DegP homologs from V. cholerae (protease DO), V. fischeri (DegP), V. parahemolyticus (protease DO), and E. coli (protease DO). The highest homology ( 56-84%) was consistently achieved by VV1_0603, the V. vulnificus protease DO. There was also significant homology to the gene immediately downstream, VV1_0604, degS. In all of the Vibrio species tested, degP was followed by degS. This indicated that VV1_0603 could have been a degP homolog. Interestingly, the genomic region surrounding degP in the Vibrio species was almost identical to the region surrounding another degP homolog, degQ, in E. coli. There was a chance that V. vulnificus DO might be DegQ and not DegP. This was an important distinction because E. coli degP is regulated by E but degQ is regulated only by 70, the housekeeping sigma factor (150). Thus degQ is not a member of the E regulon. Based on this discrepancy in the genome annotation databases and the fact that some bacterial genomes do not contain degP (151), we were careful to corr ectly annotate VV1_0603. To test if V. vulnificus DO could be regulated by E, a sequence homology search was performed for a promoter sequence upstr eam of VV1_0603 that may be recognized by E. The 120

PAGE 121

E. coli E promoter consensus sequence -10 T/C GGTCAAAA, -35 GGAACTTTT (152) was used as a guide. A putative E-responsive promoter sequen ce was located upstream of VV1_0603. More importantly, we observed signif icantly increased expression of VV1_0603 in rseB mutants compared to wild-type by qRT-P CR (Table 5-3), suggesting that VV1_0603 was regulated by E. As a final approach to determine if VV1_0603 encoded DegP, we performed a bioinformatic search for othe r serine proteases in the V. vulnificus genome that could be DegP. Six genes in the CMCP6 genome sequence were annotated as serine end oprotease. However, none of these had PDZ domains that are an essentia l feature of DegP proteins (150). In contrast, VV1_0603 had two C-terminal PDZ domains (ami no acids 257 to 346 and 368 to 444), based on the NCBI conserved domain database. Anot her gene, VV2_1530, was a pr edicted periplasmic protease that had a single PDZ domain, the PDZ domain of C-terminal processing-, tailspecific-, and tricorn proteases. However, De gP is not a tricorn protease and has two PDZ domains (150), so VV2_1530 does not likely encode a DegP homolog. Furthermore, the amino acid sequence of VV2_1530 has no significant similarity to E.coli DegP or to DO of V. vulnificus or V. cholerae. Instead, the putative protein encoded by VV2_1530 has significant similarity to a V. cholerae tricorn protease and so was rejected as a potential DegP homolog. Based on these sequence analyses and the f acts that VV1_0603 has a putative E-responsive promoter sequence and showed indirect eviden ce of positive regulation by E, viz. upregulation in rseB mutants that overexpressed E, we were confident that VV1_0603 encoded a DegP orthologue of V. vulnificus, and thus we named the gene degP Characterization of a V. vulnificus degP mutant VV1_0603 was deleted from wild-type CMCP 6 by allelic excha nge and chitintransformation, as described in the Materials and Methods. The degP mutant formed slightly 121

PAGE 122

smaller colonies than WT on LB-N agar plates and had a whitish color, similar to that of rpoE mutants (Figure 5-26). degP mutants in several of bacteria we re reported to be less resistant to oxidative and other envelope stresses than their w ild-type counterparts (reviewed in (150)). To see if this was true for V. vulnificus, sensitivity to 3% (vol /vol) hydrogen peroxide, 100% (vol/vol) ethanol, and 1% (wt/vol) SDS was te sted for FLA1002 and wild-type CMCP6 by disk diffusion assays (Table 5-4). degP mutant FLA1002 was 32% more sensitive to hydrogen peroxide ( P = 9 10-7) and 27% more sensitive to SDS ( P = 1 10-7) than was the wild-type. FLA1002 had no statistically signifi cant increase in sensitivity to 100% (vol/vol) ethanol. As DegP has roles in heat resistance in other bacteria (150), we tested the heat sensitivity of the V. vulnificus degP mutant compared to wild-type, essentially as was done for the rpoE mutant (Figure 5-27). The degP mutant showed a marked decrease in viability after incubation at 42C for 18h, compared to wild-type or to the degP mutant incubated at 37C. Taken together, these results indicated that DegP is essential for envelope stress resistance and, in particular, heat resistance, of V. vulnificus Because the degP mutant resembled the rpoE mutant in envelope sensitivity phenotypes, we tested if the degP mutant also was defective for motility. degP mutant FLA1002 and wild-type CMCP6 were stabbed into 0.3% agar mo tility plates, and the diameter of spread was measured after overnight incubation. FLA1002 had a 36% reduction in motility compared to wild-type, P = 4 10-5 by Students t test of mutant values compared to wild-type ( Figure 5-28). The increased opacity of the degP and rpoE mutants compared to wild-type suggested that they might be overexpressing CPS, as CPS e xpression is inversely correlated with opacity in V. vulnificus (22). As a test for increased exopolysaccharides, EPS extracts of the degP and rpoE mutants were resolved on SDS-PAGE gels and stained with Alcian Blue or with Stains122

PAGE 123

all as described for rseB mutants (Figure 5-17). We did not observe any marked increase in exopolysaccharides in the degP or rpoE mutants, compared to wild-type, using these methods. The possibility remained that the increased opacity of the degP and rpoE mutants was independent of exopolysaccharide expression. For example, proteins could be responsible for the changes in opacity, a phenome non that has been well documented for Neisserria gonorrhoeae (Opa proteins) (153,154). Alternatively, it was possi ble that the EPS staining methods used were not sensitive enough to detect an increase in EPS in the degP and rpoE mutants compared to wild-type. Additionally, degP FLA1001 and rpoE FLA1002 were observed for biofilm-forming ability with the expect ation that they would form poor biofilms if their increased opacity was due to increased CPS expression capsule mutants form better biofilms in V. vulnificus ((143) and Figure 5-29). Biofilm assays were performed as described for the rseB mutants. While the rpoE mutant did not show any d ecrease in biofilm formation compared to wild-type, the degP mutant had a 40% decrease in mean absorbance ( P = 0.0007) compared to wild-type (Figure 5-29). Virulence of V. vulnificus degP degP has been identified as a virulence factor for several bacterial pathogens including Yersinia enterocoliti ca, Yersinia pestis adherent and invasive E. coli, Brucella abortus Klebsiella pneumonia, Legionella pneumoph ila, Staphylococcus aureus, Streptococcus pyogenes, Streptococcus pneumoniae, and Streptococcus mutans (118,150,155). Moreover a degP -deficient Salmonella Typhimurium mutant is used as a vaccine strain (118). To test the virulence of the V. vulnificus degP mutant, mice were infected with 1,000 CFU of degP FLA1002 or with 1,000 CFU of wild-type CMCP6 (Figure 5-30). Mean skin (108.3) and liver (104.8) CFU/g for FLA1002 were almost identical to the wild-type (108.4CFU/g skin; 104.2 CFU/g liver). Thus, FLA1002 was not attenuated at this inoculum via the s.c. route of infection. 123

PAGE 124

Deletion of rseA from wild-type CMCP6 In E. coli RseA is the major negative regulator of E activity (116). We hypothesized that a V. vulnificus rseA mutant would have similar colony morphology and virulence phenotypes to the FLA610 rseB mutant. We deleted rseA from wild-type V. vulnificus CMCP6 by allelic exchange via chitin transformation as desc ribed in the Materials and Methods, using oligonucleotides rseAup5USER, rseAup3 USER-SmaI, rseAdn5USER-SmaI and rseAdn3USER to form plasmids pGTR2021 a nd pGTR2022. A representative confirmed mutant was named FLA1012. rseA FLA1012 exhibited phase vari ation and displayed mixed colony morphologies with varying degrees of opac ity. Some colonies were as opaque as the wild-type (designated rseA -O), some were more opaque than wild-type (designated rseA -vO), some were translucent, similar to rseB -Tl (designated rseA -Tl), and some were intermediate between translucent and opaque (designated rseA-Int) (Figure 5-31). Su rprisingly, we did not observe any rseA colonies that were transparent like rseB -Tp. As was seen for rseB variants, rseA colonies exhibited altere d binding of Congo red and Calcofluor, compared to wild-type (Figure 5-32 and Figure 5-33). rseA -Tl was poor at binding Congo red and appeared light pink on Congo red-agar, simila r to transparent and translucent rseB variants. rseA -Int bound Congo red better than did rseA -Tl, but not as well as did the wild-type or other opaque bacteria ( rseA -O, rseB-O). rseA-O bound Congo red at least as well as did the wild-type and rseB -O. Interestingly, rseA -vO, that appeared whitis h on LB-N agar, bound Congo red well, but retained a whiti sh haze (Figure 5-32). Binding of Calcofluor by rseA variants was similar to bi nding of Congo red (Figure 533). rseA -Tl did not fluoresce in the pres ence of Calcofluor, similar to rseB -Tp and rseB -Tl. rseA -Int fluoresced at a lower level than wild-t ype but higher than translucent variants. rseB -O and rseA -vO fluoresced to a higher extent than wild-type. These observations of 124

PAGE 125

altered Congo-red and Calcofluor binding s uggested altered exopolys accharides in the rseA mutant. We tested the ability of the translucent, intermediate, and opaque rseA variants to cause disease in iron dextran-treated mice via the s.c. route of infection with inocula of 1,000 CFU. The rseA translucent variant switched to opaque in vivo with high frequency; 13-84% (mean 57%) of bacteria recovered were opaque (Figure 5-34). The mean log skin and liver CFU/g for the translucent rseA variant were 8.0 0.2 and 4.8 2.5, respectively. The intermediate variant of rseA was fully virulent in mice and di d not switch to a more opaque form in vivo Mean log skin and liver CFU/g for the intermediate rseA variant were 7.8 0.4 and 6.0 0.6, respectively. Not su rprisingly, the opaque rseA variant was virulent in mice. Mean log skin and liver CFU/g for the opaque variant were 8.0 0 and 5.1 0, respectively. Overall, none of the rseA variants showed decreased virulence, excep t for the fact that the translucent variant switched to opaque in vivo suggesting that the translucent form was less fit for causing disease. The switch to a more opaque form in vivo was similar to the phenomenon seen for rseB mutant infections. Taken together, these results indicated that rseA and rseB likely have an overlapping or complementary function in V. vulnificus that is somehow rela ted to colony morphology through changes in surfa ce carbohydrate expression. Discussion Identification of a V. vulnificus rseB ::mini-Tn 5Km2phoA Mutation that Caused Altered Colony Morphology and Attenuated Virulence V. vulnificus rseB ::mini-Tn 5Km2phoA mutant FLA609 initially appeared less opaque than did the wild-type on LB-N agar and underwent high-frequency phase variation to a more opaque morphology (Figure 5-2). The translucent variant was attenuated for vi rulence in the s.c. inoculated iron dextran-treated mouse model of V. vulnificus infection and switched to a more 125

PAGE 126

opaque form in vivo (Figure 5-4, Figure 55). In contrast, the opaque variant was not consistently attenuated for viru lence and did not appear to unde rgo phase variation in the mouse host. Notably, it was possible that the opaque variant might have undergone phase variation in vivo but the translucents were selected against. Deletion of rseB from wild-type CMCP6 essentially recapitulated these phenotypes with th e addition of a transparent variant, FLA610-Tp, which was highly attenuated for virulence (Figure 5-9 and Figure 5-11). The only reports of an rseB homolog having a role in virulence involve mucB of P. aeruginosa. Alginate production in P. aeruginosa is controlled by AlgU/ E (133). Mutations in MucB, the periplasmic negative regulator of E, lead to the alginate-overproducing, mucoid phenotype associated with the establishment of lethal disease in cystic fibrosis patients (134-137). This scenario is opposite to that of V. vulnificus wherein the rseB mutation caused decreased expression of extracellular polysaccharides, and decreased virulence. To our knowledge, th is is the first report of an rseB mutation causing attenu ated virulence. EPS and the Altered Colony Morphology of V. vulnificus rseB Mutants V. vulnificus rseB ::mini-Tn 5Km2phoA and rseB mutants had a translucent colony phenotype that showed high freque ncies of phase variation to a more opaque form. Colony opacity has been correlated to CPS expression in V. vulnificus (24,138,156). However, Chatzidaki-Livanis, et al (142) noted that a genetic event that led to a ph ase-variable translucent phenotype (designated Tr1) did not corres pond to genetic rearrangements in the CPS biosynthesis operon. Rosche et al (157) observed that V. vulnificus can exhibit a morphology that is intermediate between the translucent and opaque forms (designated intermediate, Int). The Int morphotype was characterized by decreased expression of CPS, leading to the proposal that non-opaque strains that re vert to opaque actually have the Int morphotype which undergoes reinduction of CPS expressi on (157). While the rseB -Tl variant examined here had a level of 126

PAGE 127

opacity that was between that of rseB -O and rseB -Tp, we do not believe that rseB-Tl was the Int morphotype observed by Rosche, et al. (157) because the trend of phase variation they reported was very different from ours. Rosche and coworkers (157) observed that non-opaque (Int and translucent) colonies arose from opaque colonies at low to moderate rates of 2.2% to 26.2% but that Int strains yielded translucent colonies with higher frequency (75.1% to 98.8%) under the conditions tested. Conversely, we obs erved that the variation from translucent rseB variants to opaque occurred with higher fre quency than did any other conversion, and that the rseB -Tl variant infrequently became transparent (or more translucent). More importantly, while the translucent rseB variant FLA610-Tl was not as translucent as the acapsular mutant FLA1009 on LB-N agar (Figure 5-13), FLA 610-Tl was less red than was FLA1009 on Congo red agar (Figure 5-14) and fluor esced less than did FLA1009 on Ca lcofluor agar (Figure 5-15), indicating that the translucent rseB variant contained lo wer amounts of some exopolysaccharides than did th e acapsular mutant FLA1009. There is increasing evidence that V. vulnificus possesses extracellular polysaccharides in addition to CPS. A mutation in ntrC (transcriptional regulator) or in gmhD (sugar epimerase regulated by NtrC and by RpoN) caused significan tly decreased EPS but did not cause decreased colony opacity (158). Th is result suggested that V. vulnificus makes several kinds of extracellular polysaccharide. In terestingly, RpoN that regulat es expression of the sugar epimerase GmhD, is a member of the core E regulon (152). It wa s recently reported that increased c-di-GMP levels in V. vulnificus caused expression of an E PS that was distinct from CPS (159). Translucent strains of V. vulnificus with a mutation in the wzy CPS polymerase gene, the rmlC sugar epimerase gene, or the wzc tyrosine autokinase gene could all be restored to an opaque morphology via expression of a diguanylate cyclase gene, dcpA in trans Moreover, the 127

PAGE 128

non-CPS exopolysaccharides described by Kim, et al. (158) and by Nakhamchik, et al (159) appeared to be composed of aci dic carbohydrate moieties as they st ained with Stains-all and with Alcian Blue. It is possible, theref ore, that altered co lony morphology of the V. vulnificus rseB variants described here could be due decreased expression of EPS determinants that are distinct from CPS. This could explain the fact that rseB -Tp and rseB -Tl variants appeared different from the acapsular mutant on the carbohydr ate-binding media containing Congo red or Calcofluor (Figure 5-14 and Figur e 5-15). Of course, these differences in binding could also result from differences in amount of CPS expressed between the rseB and acapsular mutants. It will be necessary to an alyze the chemical composition of th e exopolysaccharides of the various strains to determine if the rseB mutants are deficient in CPS or another EPS. Comparison Between the rseB -Tp Mutant and an Acapsular V. vulnificus Mutant The translucent rseB and rseA variants were very similar: they showed similar binding of Congo red and Calcofluor (Figure 5-32 and Figure 5-33) and both switched to a more opaque form during infection of the mouse host (Figure 54 and Figure 5-34). This result suggested that a similar mechanism probably caused the translucent morphology observed in these two strains. rseB -Tp, however, was less opaque than rseB -Tl, rseA -Tl, and the translucent rseB::miniTn 5Km2phoA strain. In fact, rseB -Tp most resembled an acapsular mutant of V. vulnificus CMCP6 in staining of EPS with Alcian blue and Stains-all (Figure 5-16, Figure 5-17), increased biofilm-forming ability (Figure 5-19), increased serum sensitivity (Table 5-1), and highly attenuated virulence (Figure 5-11). Also similar to the acapsular mutant, the transparent rseB variant did not switch to a more opaque form in vivo With the exception of differences in morphology on LB-N agar (Figure 5-13) and in binding of Calcofluor and Congo red (Figure 515 and Figure 5-14), the major phenotypic difference we observed between the rseB -Tp mutant and the acapsular mutant was in motility. The acapsular mutant was hyper-motile compared to 128

PAGE 129

the wild type while the rseB -Tp mutant was not altered for mo tility (Figure 5-20). While there was no evidence of genetic rearra ngements in the CPS locus in rseB -Tp compared to the wild type (Figure 5-18), there still remains the possibility that rseB -Tp could be completely acapsular. Possible Reasons for the Attenuated Virulence of rseB and rseA Mutants There are two main possibilities regard ing the attenuated virulence of the V. vulnificus rseB and rseA mutants. Because the translucent va riants caused by both mutations showed decreased fitness in vivo and switched to a more opaque form, it is plausible that the attenuation was due to a decrease in exopolysa ccharides (EPS or CPS) that may be directly or indirectly due to the rseB and rseA mutations. The second possibility is that the attenuation was due solely to the deleterious effects of overexpression of E. CMCP6 expressing rpoE from pGTR2008 did not exhibit decreased op acity but was attenuated for virule nce at an inoculum of 300 CFU (Figure 5-21). When the opaque rseB mutant was examined for virulence in 20 mice (four separate infections of five mice) at an inoc ulum 1,000 CFU, 35% of the mouse skin and liver samples did not yield detectable CFU (Figure 5-10). Thus, even without being translucent, the rseB -O variant that overexpressed rpoE (Table 5-3) showed decrea sed virulence in mice. Regulation of E in V. vulnificus V. vulnificus rseB variants showed 13to 24-fold increases in rpoE expression compared to the wild type (Table 5-3). E. coli RseB was reported to be th e minor negative regulator of E; deletion of rseB resulted in a 2-fold increase in transcription from E-dependent promoters, whereas deletion of rseA led to a 9-fold increase (128,129). Conversely, inactivation of either MucA (RseA) or MucB (RseB) in P. aeruginosa can result in comparably large increases in AlgU activity (160,161), supporting majo r regulatory roles for both prot eins. Further studies will be needed to compare the level of rpoE expression in rseB versus rseA mutants of V. vulnificus. 129

PAGE 130

Although the regulation is post-tr anslational, the effects on rpoE can be measured at the transcriptional level because E autoregulates its expression. E and DegP are Important for Stress Resistance but not for Virulence in V. vulnificus Deletion of rpoE or degP from V. vulnificus CMCP6 resulted in a significantly altered morphology on LB-N agar (Figure 5-26). Both st rains appeared whiter (more opaque) than did the wild type. Both the rpoE and degP mutants also exhibited in creased fluorescence in the presence of Calcofluor, compared to the w ild type (Figure 5-26), suggesting altered EPS expression. Motility was decreased by 20% and 36% in rpoE mutant FLA1001 and in degP mutant FLA1002, respectively, compared to the wild type. Both mutants also exhibited increased sensitivity to membrane-perturbing agents (Table 5-2) and to heat (Figure 523 and Figure 5-27). The degP mutant was also deficient in biof ilm formation (Figure 5-29), possibly due to the reduced motility. Motility is essential for bi ofilm formation, and a positive relationship between these two functions has been shown for V. vulnificus (158). The rpoE mutant had decreased motility but did not show an overall decreas e in biofilm formation in our assay, although it was reduced compared to the wild-type in some biologi cal replicates of the assay. Despite the overall increase in envelope stress sensitivity in the rpoE mutant, it showed only a marginal decrease in virulence compared to the wild type at an inoculum of 300 CFU (Figure 5-25). The degP mutant was not attenuated compared to the wild type at an inoculum of 1,000 CFU. It is possible that the degP mutant may show attenuated virulence compared to the wild type if inoculated at a lower cell number (as for the rpoE mutant). Nevertheless, it appears that E and DegP have little importance for V. vulnificus to infect mice via the s.c. route, despite their importance for resisting exte rnal stresses. It is possible that E and DegP are essential for survival in the environment but not fo r causing disease, or that they may have a role 130

PAGE 131

in infection that could not be assessed by our an imal model. As previously noted, AlgU ( E) of P. aeruginosa is important for respiratory, but not syst emic, infection (118). Likewise, HtrA (DegP) of S. Typhimurium is essential for systemic, but not enteric, disease (118). Therefore, a role for E or DegP in V. vulnificus virulence may yet be discovere d via a different animal model or route of infection, e.g., oral inoculation for which V. vulnificus must survive the gut acidity. Future Directions This work has opened the door to several lin es of investigation. The reason behind the altered colony morphology of rseA and rseB mutants is at the forefront. Although we know that the phenotype is due to decreased exopolysacch arides, the nature of these carbohydrates is still unknown. Detailed chemical analyses wi ll be needed to identify the extracellular carbohydrates missing from non-opaque rseB and rseA mutants. Another pressing question is the mechanism of phase variation that results in the frequent switch from translucent to opaque rseB and rseA mutants. It will be interesting to observe the frequency of phase variation under different stress conditions to see if the swit ch is induced by extracellular stress. Primary evidence for this hypothesis comes from the fact that the translucent va riants switch to opaque with very high frequency during mous e infection. To determine if the rseB phenotypes are due to E overexpression, it will be necessary to construct a deletion of rseB in a rpoE mutant background. This is already underw ay in our laboratory. If the rpoE rseB mutant has a translucent phenotype that is capable of phase variation, this result will indicate that the rseB mutant phenotype is independent of E. Finally, an analysis of the E regulon in V. vulnificus by transcriptome or proteome analysis of the rpoE mutant compared to th e wild type will add to the overall understanding of the role and regulation of this system in V. vulnificus and may provide insight into a possible role for E in regulating colony opacity in V. vulnificus For example, wzb and wzc capsule biogenesis gene s are members of the E regulon of E. coli K-12 131

PAGE 132

(152). Also, because members of the E regulon are virulence f actors in many bacterial pathogens (118), identifying the V. vulnificus E regulon may reveal novel virulence factors for this pathogen. 132

PAGE 133

Mini-Tn 5Km2phoA insertion Potential signal sequence cleavage siteVV1_1561 rseB VV1_1562 rseC VV1_1560 rseA VV1_1559 rpoE Mini-Tn 5Km2phoA insertion Potential signal sequence cleavage siteVV1_1561 rseB VV1_1562 rseC VV1_1560 rseA VV1_1559 rpoE Figure 5-1. Schematic showing V. vulnificus rpoErseABC operon. Similar to the well-studied E. coli rpoErseABC operon, the V. vulnificus genome contains homologs of rpoE (VV1_1559), rseA (VV1_1560), rseB (VV1_1561), and rseC (VV1_1562) that appear to form an operon. The location of the phoA insertion in VV1_1561 ( rseB ) is shown (arrowhead). The location of the pot ential signal sequence cleavage site in RseB is indicated. Figure is approximately to scale. 133

PAGE 134

134 F L A 609 A FLA609-OFLA609-Tl CMCP6 B FLA609-Tl C Figure 5-2. Mixed morphology of rseB ::mini-Tn 5Km2phoA FLA609. A) A single translucent colony of FLA609 was grown standing overn ight, diluted and grown to log phase shaking at 37C, and an aliquot was diluted and plated on LB-N. A considerable proportion (>30%) switched to a more opa que morphology during the growth period. B) Subculturing of a single opaque (top le ft) or translucent (top right) colony of FLA609 compared to wild-type (bottom). C) Higher magnification of the translucent sector of (B) showing opaque variants (wh ite specs) scattered throughout the mainly translucent culture.

PAGE 135

0 1 2 3 4 5 6 7 8 9 SKIN LIVER TEMPERATURELog CFU/g Tissue and Temp/10 FLA609 FLA399* **1/3 5/5 0/3 5/5 Figure 5-3. Preliminary mous e infection with FLA609 ( rseB ::mini-Tn 5Km2phoA ). In the initial infection, 3 mice were infected with rseB ::mini-Tn 5Km2phoA FLA609-Tl at approximately 1,000 CFU per mouse. Plati ng of the inoculum revealed that it contained approximately 10% opaque variants. Wild-type FLA399 was inoculated at approximately 1,000 for comparison CFU. Fractions below bars indicate the proportion of mouse samples th at yielded bacteria. *, P = 0.03; **, P = 0.008 for log CFU/g tissue in mutant compared to wild-type by Students t test. 135

PAGE 136

Opaque Translucent Figure 5-4. Mixed morphologies of bacteria harvested from mice infected with FLA609-Tl. Bacteria harvested from tissues of mice in fected with FLA609-Tl were enriched for the opaque (O) variant (up to 90% O). 136

PAGE 137

0 1 2 3 4 5 6 7 8 9 10 SKIN LIVERTEMPERATURELog CFU/g Tissue and Temp/1 0 FLA609-Tl FLA609-O FLA3994 /5 5 /5 5/ 5 2 /5 4 /5 5/5 Figure 5-5. Virulence of transl ucent and opaque variants of rseB ::mini-Tn 5Km2phoA FLA609 in mice. Mice were infected with with 1,000 CFU of FLA609-Tl or O. Mice were also infected with 1,000 CFU of wild-type FLA399 for comparison. The translucent variant (FLA609-Tl) was attenuated for system ic (liver) infection compared to wildtype, although mean temper ature (an indication of sy stemic disease) was not significantly different from wild-type. *, P = 0.04 compared to WT. 137

PAGE 138

Figure 5-6. Virulence of the opaque variant of rseB ::mini-Tn 5Km2phoA FLA609-O. Scatter plot showing three separate mouse infections combined to illustrate trends. FLA609O consistently achieved high levels of skin infection but showed large variations in the level of liver infection. Mean and standard deviat ions are shown on graph. 138

PAGE 139

Figure 5-7. Colony morphology of complemented FLA609-Tl. Instead of restoring wild-type colony morphology, complementation with wild-type rseB resulted in continued phase variation from translucent to opa que morphologies. In addition, a third morphology arose (transparent, Tp), that was even less opa que than the translucent (Tl) variant. The plat e shown contains Congo red agar that emphasizes the morphological differences. 139

PAGE 140

0 1 2 3 4 5 6 7 8 9 10 SKIN LIVERTEMPERATURELog CFU/g tissue and temp/10 rseB::mini-Tn5Km2phoA-Tl Complemented Tl rseB::mini-Tn5Km2phoA-O Complemented O WT Figure 5-8. Complementation of rseB::mini-Tn 5Km2phoA FLA609-Tl. Mice were infected with 1,000 CFU of translucent or opaque complemented strains. Plate counts on nonselective medium are shown (there was no statistical difference between plate counts on selective and nons elective media). Data fo r prior infection with rseB ::miniTn 5Km2phoA mutants are included for comparison. Due to the large variations in the data, no statistically significant differences were observed between the complemented strains and mutants or complemented strains and wild-type. 140

PAGE 141

0 1 2 3 4 5 6 7 8 9 SKINLIVERTEMPERATURELog CFU/g tissue and Temp/10 FLA610-Tp FLA610-Tl FLA610-O0/5 2/5 2/50/5 2/5 2/5 Figure 5-9. Virulence of rseB transparent, translucent, and opaque variants in mice. Graph represents bacterial yield and temperatures re sulting from s.c. infection of 5 mice per strain at an inoculum of 1,000 CFU. 141

PAGE 142

Figure 5-10. Variable virulence of FLA610-O (similar to that seen for rseB ::mini-Tn 5Km2phoA FLA609-O) 142

PAGE 143

0 1 2 3 4 5 6 7 8 9 10 SKIN LIVERTEMPERATURE Log CFU/g tissue and Temp/1 0 FLA610-Tp CMIT232 CMCP65/5 3/5 6/6 1/5 0/5 5/6* *** ** ** Figure 5-11. Highly attenuated virulence of FLA610-Tp. Comparison of virulence of the transparent rseB variant (FLA610-Tp) with a nonencapsulated mutant of CMCP6 (CMIT232) at an inoculum of 105 CFU. An infection with wild-type CMCP6 inoculated at 1,000 CFU was included for comparison. *, P 0.009 ; **, P = 0.001; ***, P = 1 10-4. for mutants compared to wild-type. 143

PAGE 144

0/5 1/5 2/5 4/5 2/5 5/5 0/5 0/5 2/5 4/5 2/5 5/5 0 1 2 3 4 5 6 7 8 9 SKIN LIVERTEMPERATURELog CFU/g Tissue and Temp/10 rseB Tp Tp complemented rseB Tl Tl complemented rseB O O complemented* Figure 5-12. Complementation of rseB mutation. Mouse infection showing complemented rseB transparent, translucent, and opaque va riants. Plating on nonselective media is shown (there was no statistical difference between plate counts on selective and nonselective media). Data for prior infection with rseB mutants is included for comparison. *, P = 0.01; ** P = 0.008 by Bonferronis t test comparing mutant to respective complemented strain; = 0.017. 144

PAGE 145

rseBTl rseBTp CPSrseBO WT Rugose Figure 5-13. Morphology of C PS mutant, FLA1009 compared to rseB variants and wild-type CMCP6. Plating on LB-N is shown. The CPSmutant appeared translucent compared to wild-type, and as more translucent than rseB FLA610-Tl, but was not as translucent as rseB FLA610-Tp. 145

PAGE 146

rseBTl rseBTp CPSrseBO WT Rugose Figure 5-14. Ability of V. vulnificus strains to bind Congo red. rseB variants, acapsular FLA1009, rugose FLA611-R, and wild-type CM CP6 were plated on LB-N containing the carbohydrate-binding dye, Congo red. The plate was incubated at 37C and photographed. Image is representative of se veral replicate plates. Wild-type and rseB -O bound Congo red at about equal leve ls. The rugose strain bound Congo red at a level that was equal to or s lightly less than wild-type. The CPSmutant bound less than wild-type. rseB -Tp and rseB -Tl bound Congo red much less than wildtype, and were almost indistinguishable from each other when streaked on Congo red agar. 146

PAGE 147

White light White lightAB UV light UV light Figure 5-15. Evidence of carbohydrate content of V. vulnificus strains visualized by Calcofluor binding. Strains were plated on LB-N ag ar containing the fluorescent carbohydratebinding agent, Calcofluor. Plates were inc ubated overnight at 37 C and then at room temperature for 24h. Images were captured before (A) and after (B) exposure to UV light. The translucent and transparent rseB variants did not exhibit appreciable fluorescence in the presence of Calcof luor exposed to UV light. The CPSstrain fluoresced, although at lower levels th an those seen for the wild-type, rseB-O, or rugose strains. Images are representa tive of several replicate plates. 147

PAGE 148

CPS-WT TpTlO Possibly CPS250 kd 37 15 20 25 50 75 100 150 CPS-WT TpTlO Possibly CPS250 kd 37 15 20 25 50 75 100 150 Figure 5-16. Stains-a ll staining of whole cell lysates. 107 CFU of whole cell lysates of rseB variants, wild-type, and CPSmutant were resolved by SDS-PAGE on a 4-20% gradient gel and stained with Stains-all. A darkly stai ning purple region was visible in the translucent and opaque rseB variants and wild-type, but absent from the CPSmutant and the transparent rseB variant, FLA610-Tp. It was deduced that this dark band probably represented CPS. 148

PAGE 149

T p Tl O CPSR E de g P WTA Tp Tl O CPSR E degP WT B Figure 5-17. Analysis of EPS extr acts. Purified EPS extracts from rseB transparent (Tp), translucent (Tl), and opaque (O) variants, acapsular FLA1009 (CPS-), rugose (R), rpoE ( E), degP (degP) and wild-type CMCP6 (W T) cells were resolved on SDSPAGE gels and stained with alcian blue (A) or with Stains-all (B) to observe exopolysaccharides. The transparent rseB variant (Tp) did not stain appreciably with alcian blue or with St ains-all, similar to the CPSmutant. rseBTl stained slightly better with alcian blue and Stains-all than did rseB -Tp or the CPSmutant. rpoE and degP that appear more opaque than wild-type on LB-N, did not stain significantly more with alcian blue or Stains-all compared to wild-type. EPS from the rugose strain (R) appeared to bind alcian bl ue and Stains-all more strongly than did EPS from any other strain. indicating that this strain may be overexpressing EPS, or may contain a different EPS co mposition from the wild-type. 149

PAGE 150

Tp Tl 1-kb O CPSWT 4 5 3 6 Figure 5-18. PCR to observe gene tic rearrangements in CPS transport genes. PCR primers were designed to produce amplicons that spanned the transport region ( wza to wzc ) of the CPS operon as in Chatzidaki-Livanis et al. (142). Genomic DNA from rseB transparent (Tp), translucent (Tl), and opaque variants, an acapsular mutant bearing a Tn phoA insertion in wza (CPS-), and wild-type CMCP6 (WT) were used as templates. None of the rseB variants showed genetic rearrangements in the CPS locus compared to wild-type. Genomic DNA from the CPSmutant failed to amplify with these primers due to the 7.7-kb Tn phoA insert in the wza gene. 150

PAGE 151

A A TpTlO Rug WT Blank TpTlO Rug WT Blank Blank B. TpTlO Rug WT Blank TpTlO Rug WT Blank Blank B. Figure 5-19. Biofilm formation of V. vulnificus strains. Transparent, translucent, and opaque rseB variants (Tp, Tl, O) were compared to an acapsular mutant of CMCP6 (CPS-), a rugose mutant (rugose), and wild-type CMCP 6 (WT) in the ability to adhere to a polystyrene plate. (A) Data are from thr ee biological replicates, each consisting of six wells per strain. The transparent rseB variant was consistently better at biofilm formation than were the wild-type or the CPSmutant. The rugose mutant was highly variable in final absorban ce measurement due to loss of bound cells during washes, but usually formed better pre-wash biofilms th an did any other strain (B), and had an overall mean that was higher than that of the wild-type. (B) Photograph of polystyrene plate immediately after staining with crystal vi olet (prior to washes). *, P = 4 10-7; **, P = 3 10-9; ***, P = 2 10-11 for mutants compared to wild-type; P = 0.006 for rseB Tp compared to CPSby ANOVA with Bonferr onis post test. 151

PAGE 152

Table 5-1. Sensitivity of rseB mutants to serum complement. Strain Log CFU intact serum-log CFU heat inactivated serum CMCP6 -0.7 FLA610-Tp -5.0 FLA610-Tl -1.3 FLA610-O -1.0 FLA1009 -4.6 MG1655 -6.5 Wild-type CMCP6, rseB transparent, translucent, and opaque variants (FLA610-Tp, FLA610Tl, and FLA610-O), acapsular FL A1009, and complement-sensitive E. coli MC1655 were incubated in intact serum or in heat-inactivat ed serum for two hours. CFU were enumerated under both conditions to assess sensitivity to co mplement-mediated killing. The transparent rseB variant was as sensitive to complement as was the acapsular mutant. A similar trend was observed in a second biological replicate. 152

PAGE 153

Table 5-2. Sensitivity of rseB mutants to envelope stress. Strain Zone of Inhibition, cm Ethanol H2O2 rseB -Tp 0.77 0.06 0.73 0.05 rseB -Tl 0.78 0.04 0.69 0.04 rseB -O 0.76 0.11 0.68 0.06 WT 0.67 0.11 0.69 0.05 Disc diffusion tests were done to determine th e zone of inhibition for ethanol or hydrogen peroxide in rseB mutants compared to wild-type. n = 6 filters per strain per condition. No statistically significant di fferences were observed. 153

PAGE 154

Figure 5-20. Motility of rseB mutants. As an indicator of a possible cell envelope perturbance, the motility of the rseB mutants was compared to wild-type CMCP6 and CPSFLA1009. Motility was measured as diameter of spread from the center of a 0.3% (wt/vol) agar motility plate. n = six plates per strain. None of the rseB variants showed statistically significant differences in motility compared to wild-type. The CPSmutant was hyper motile compared to wild-type. Similar results were obtained in repeat experiments. *, P = 2 10-4 by ANOVA with Bonferronis post test for mutant values compared to wild-type. 154

PAGE 155

Table 5-3. Analysis of expression of rpoE and degP in rseB mutants by qRT-PCR Mean fold-change in gene expression Gene/Strain rseB -Tp rseB -Tl rseB -O rpoE (SigmaE) 24 4** 20 2** 13 3** degP (periplasmic protease) 70 30* 100 25* 28 6* Because rseB is a negative regulator of E, an rseB mutant should overexpress rpoE Gene expression analysis by qRT-PCR show ed increased expression of both rpoE and degP in all rseB mutant variants, confirming predictions for the rseB mutant phenotype. Values are fold-change (2Ct) relative to wild-type, normalized to 16S rRNA as an endogenous control. *, P 0.01; **, P 0.0001 for raw Ct values in each mutant compar ed to wild-type by Students t test. 155

PAGE 156

Figure 5-21. Effect of E overexpression on virulence of CMCP 6. Mice were infected with 300 CFU of wild-type CMCP6 or 300 CFU of CMCP6 overexpressing rpoE via plasmid pGTR2008. Plate counts on selec tive medium are shown for the rpoE -overexpressing strain. *, P = 0.03; **, P = 0.01 for CFU/g liver or temperatures of rpoE overexpressing strain compared to wild-type by Students t test. 156

PAGE 157

Figure 5-22. Morphology of two E ( rpoE ) mutant isolates compared to wild-type. Strains were streaked onto LB-N agar, incubate d at 37C overnight, and photographed. The E mutants appeared whiter (more opaque) than did the wild-type. 157

PAGE 158

Table 5-4. Sensitivity of rpoE and degP mutants to envelope stresses. Ethanol H2O2 SDS rpoE 115 6%* 129 6% ** 125 3% ** degP 106 3% 132 4% ** 127 4% ** Inhibition by 100% ethanol, 3% hydrogen peroxi de, or 1% SDS was determined by zone of inhibition tests. n 6 filters per experiment. Each experime nt repeated at least once. Values are % of wild-type inhibition. P = 0.006, ** P 9 x 10-7 by unpaired t test comparing mutant zones of inhibition to wild-type zones of inhibition for each chemical assayed. The degP and rpoE mutants were both sensitive to envel ope stresses compared to wild-type. 158

PAGE 159

Figure 5-23 Sensitivity of rpoE FLA1001 to heat. Serial dilutions of exponential-phase bacteria were spotted on plates and incubated at the indi cated temperatures for 18 h. Each experiment repeated at least once. The rpoE mutant was slightly more heatsensitive than was the wild-type. 159

PAGE 160

Figure 5-24. Motility of rpoE mutant compared to wild-t ype. Motility was measured as diameter of spread from the cente r of a 0.3% agar motility plate. n = six plates per strain. Similar results were obtai ned in repeat experiments. The rpoE mutant had a 20% decrease in motility compared to wild-type. *, P = 4 10-7 by Students t test of mutant values compared to wild-type. 160

PAGE 161

Figure 5-25. Virulence of rpoE FLA1001 at high and low inocula. Mice were infected with approximately 300 CFU of wild-type CMCP6 or rpoE FLA1001, or with 3000 CFU of CMCP6 or 2,000 CFU of FLA1001. Mutant and wild-type infecti ons at low or at high inocula were compared. There were no statistically significant differences between FLA1001 and CMCP6, except for mo use temperatures (an indication of systemic disease) when low inocula were administered. Overall, the rpoE mutant of V. vulnificus was not significantly attenuate d, compared to the wild-type. 161

PAGE 162

Figure 5-26. Morphology of rpoE and degP mutants on LB-N plates containing the fluorescent dye, Calcofluor. Plates were photographed under white (A) and UV (B) light after overnight incubation at 37C. The rpoE and degP mutants both appeared whiter (more opaque) than wild-type on LB-N agar, and fluoresced brighter than did the wild-type in the presence of Calcofluor, indicating elevated exopolysaccharides. 162

PAGE 163

Figure 5-27. Sensitivity of degP FLA1002 to heat. Serial dilutions of exponential-phase bacteria were spotted on plates and incubated at the indi cated temperatures for 18h. Each experiment repeated at least once. FLA1002 was heat-sensitive, compared to wild-type CMCP6. 163

PAGE 164

0.0 0.2 0.4 0.6 0.8 1.0 1.2 de g PW TRelative Motility* 0.0 0.2 0.4 0.6 0.8 1.0 1.2 de g PW TRelative Motility* Figure 5-28. Motility of degP mutant compared to wild-type. Motility was measured as diameter of spread from the center of a 0.3% (wt/vol) agar motility plate. n = six plates per strain. Similar results were obtained in repeat experiments. The degP mutant had a 36% reduction in mo tility compared to wild-type. *, P = 4 10-5 by Students t test of mutant values compared to the wild type. 164

PAGE 165

* ** ** Figure 5-29. Biofilm formation in degP and rpoE mutants. Biofilm assays were performed as described for rseB The degP mutant produced less biofilms than did the wild-type. *, P = 0.0007; **, P = 5 10-6 by ANOVA with Bonferronis post test for all wells compared to wild-type wells. 165

PAGE 166

5/5 5/5 5/5 5/5 Figure 5-30. Virulence of degP mutant compared to wild-type. Mice were infected with 1,000 CFU of degP FLA1002 or with wild-type CMCP6. The degP mutant showed full wild-type skin and liver in fection at this inoculum. 166

PAGE 167

rseATl rseAInt rseAO rseAvO WTA rseATl rseAInt rseAO rseAvO WTB Figure 5-31. Mixed morphology of rseA FLA1012. Phase-variants with different degrees of opacity were isolated from a single colony of rseA that originally appeared slightly less opaque than wild-type CMCP6. Plating on LB-N with Calcofluor is shown to emphasize the differences in opacity compared to wild-type. Images were captured before (A) and after (B) exposure to UV light. 167

PAGE 168

rseATl rseAInt rseAO rseAvO WT rseBO rseBTp rseBTl Figure 5-32. Binding of Congo red to rseA rseB and wild-type V. vulnificus. Strains were streaked onto LB-N agar containing th e red carbohydrate-binding dye, Congo red, and incubated at 37C overnight. rseA -Tl resembled rseB -Tl and Tp on this medium and showed low binding compared to wild-type. rseA -Int showed a level of Congo red binding that was greater than rseA -Tl but less than wild-type. rseA O and rseA -vO bound Congo red at least as well as the wild type. Overall, rseA had similar phenotypes to those seen for rseB variants on Congo red agar. 168

PAGE 169

A rseATl rseAInt rseAO rseAvO WT rseBO rseBTp rseBTl rseATl rseAInt rseAO rseAvO WT rseBO rseBTp rseBTl B rseATl rseAInt rseAO rseAvO WT rseBO rseBTp rseBTl rseATl rseAInt rseAO rseAvO WT rseBO rseBTp rseBTl Figure 5-33. Binding of rseA rseB and wild-type to Calcofluor. Strains were streaked onto LBN agar containing the fluorescent ca rbohydrate-binding dye, Calcofluor, and incubated at 37C overnight. Images were captured before (A) and after (B) exposure to UV light. rseA -Tl and rseA -Int resembled rseB -Tp and rseB -Tp before exposure to UV, but upon UV exposure, rseA -Int fluoresced at a higher level than did rseATl, rseB -Tp or rseB -Tl. rseA -Tl was still decreased for Calcofluor binding compared to wild-type. rseA -O and rseA -vO, live rseB -O, bound Calcofluor at least as well as wild-type. Overall, rseA variants followed similar trends in calcofluor biding to those seen for rseB variants. 169

PAGE 170

170 Figure 5-34. Mixed morphology of bacteria harvested from mice infected with rseA-Tl. Translucent and opaque bacteria were observed when skin and liver CFU were plated from mice infected with translucent variant of rseA Image is of an LB-N plate incubated at 37C overnight. As observed for rseB this indicated that there was selective pressure to swit ch to a more opaque form in vivo

PAGE 171

CHAPTER 6 PERSPECTIVES Host-Adapted Versus Accidental Pathogens : Redefining the Term Virulence Factor for Opportunistic Pathogens Stanley Falkow (21) laid out the rules that most microbi ologists follow when defining a bacterial virulence factor. These guidelines, th e Molecular Kochs postulates, state that a potential virulence factor should be found in all pathogenic strains of a species but be absent from non-pathogenic strains, the relevant gene should be expressed durin g infection, inactivation of the relevant gene should attenuate viru lence in an appropriate animal model, and reintroduction of the gene into the bacterium should restore viru lence in the animal model. Pallen and Wren (162) recently explored the c reative clash between genomic research and bacterial pathogenesis research. They pointed out that if the ru le that genes encoding virulence factors cannot be common to pathogenic and non-pat hogenic strains were to be strictly enforced, then some pathogens would completely lack virulence factors! However, by ignoring the postulate, many virulence factors turn up in non-pathogens, e.g., invasion genes yijP ibeB and ompA occur in E. coli K-12. Even so, these invasion genes ar e considered to be virulence factors (162) V. vulnificus naturally resides in oysters and shellf ish, causes lethal septicemia mainly in humans who have underlying health concerns, and can infect the skin on ly upon contact with an open wound (9). Also, V. vulnificus causes an acute disease proce ss that often leads to death, and humans are a dead-end host (9). Thus, V. vulnificus can be defined as an accidental pathogen as compared to host-adapted pathogens that have a close asso ciation with the host, often cause asymptomatic or self-limiting inf ections, and have a high metabolic dependency on the host (163). It is possible that gene products involve d in survival of V. vulnificus in the oyster environment or in the water column may inadve rtently cause damage to the host or promote 171

PAGE 172

bacterial invasiveness, survival, or multiplication. There is a large number of genes that are traditionally classified as virulence factors, e.g., adhesins, invasins, toxins, and secretion systems. Hilbi, et al (164) pointed out that some bacter ia have developed complex defense strategies in the course of their coevolution with environmental predators such as amoeba, including the secretion of toxins a nd the ability to avoid lysosomal k illing. It is interesting that the Flp pilus biogenesis operon iden tified in our PhoA mutagenesi s screen could be involved in adherence to algae or to oyster tissue in the estu arine environment and could inadvertently aid in colonization of the mammalian host. In addition to genes that directly interact with the host to promote disease, a separate class of genes influences bacterial f itness; any gene that provides a bacterium with a growth advantage could easily increase its virulenc e in a given host (163). This is highlighted by the fact that a small (13%) decrease in the ability to synthesize unsaturated fatty acids in vitro and a 40% decrease in growth rate in vitro correlated with a co mplete inability of a V. vulnificus fadR mutant to establish infection when three times the wild-type MLD were inoculated. Thus, a broader de finition of the term virulence factor should be used for V. vulnificus and similar pathogens to include factors that may be present in pathogenic and nonpathogenic strains as well as factors that may not interact directly with host cells to cause disease. Role of Metabolism in the Pathogenesis of V. vulnificus Three of the genes identified in our PhoA mu tagenesis screen encoded products that were implicated in bacterial metabolism: fadR that regulates fatty acid metabolism, cvpA that is part of the purF purine biosynth esis operon, and ptsG that encodes a glucose transporter of the Pts system that is involved in carbohydrate ca tabolism. Of these, mutations in cvpA and ptsG caused auxotrophy, but the ptsG mutant was fully virulent in mice while the cvpA mutant was highly attenuated for skin and liver infection. Conversely, the fadR mutation did not result in 172

PAGE 173

173 auxotrophy but caused significantly a ttenuated virulence. These fi ndings lead to questions about which metabolic pathways are important during b acterial infection, but th ere is a scarcity of information on this topic (165). Interestingly, it was recently reported that de novo nucleotide biosynthesis represents the most critical meta bolic function for bacterial growth in blood, suggesting that purine biosynthetic genes could be potential vaccine targets (166). Along similar lines, the role of iron acquisition in bact erial pathogens has been well documented (167). In V. vulnificus, as in other pathogens, iron-scavenging siderophores are used to acquire iron of the host during infection (25,26). It will be interesting to compare the metabolome of V. vulnificus growing in vitro and in vivo to determine which metabolic pathways are specifically required for the fulminant growth and massi ve tissue destructi on that characterize V. vulnificus infections. Recent advances in molecular typing and th e discovery of novel vi rulence factors for V. vulnificus have broadened our understanding of this pathogen. Further probing of the genes described herein coupled with the continued use of largescale mutagenesis strategies such as the one detailed here will shed new light on the complex lifestyle of V. vulnificus in the environment and in the host and will, undoubtedly, add to the general knowledge of the pathogenic strategies of bacteria.

PAGE 174

REFERENCE LIST 1. Hlady, W. G., and Klontz, K. C. 1996. The ep idemiology of Vibrio in fections in Florida, 1981-1993. J. Infect. Dis. 173:1176-1183. 2. Strom, M. S., and Paranjpye, R. N. 2000. Epidemiology and pathogenesis of Vibrio vulnificus Microbes Infect. 2 :177-188. 3. Vollberg, C. M., and Herrera, J. L. 1997. Vibrio vulnificus infection: an important cause of septicemia in patients with cirrhosis. South. Med. J. 90:1040-1042. 4. Kumamoto, K. S., and Vukich, D. J. 1998. Clinical infections of Vibrio vulnificus : a case report and review of the literature. J. Emerg. Med. 16:61-66. 5. Halow, K. D., Harner, R. C., and Fonten elle, L. J. 1996. Primary skin infections secondary to Vibrio vulnificus : the role of operative intervention. J. Am. Coll. Surg. 183:329-334. 6. Linkous, D. A., and Oliver, J. D. 1999. Pathogenesis of Vibrio vulnificus. FEMS Microbiol. Lett. 174 :207-214. 7. Blake, P. A., Merson, M. H., Weaver, R. E., Hollis, D. G., and Heublein, P. C. 1979. Disease caused by a marine Vibrio. Clin ical characteristics and epidemiology. N. Engl. J. Med. 300:1-5. 8. Hlady, W. G., Mullen, R. C., and Hopkin, R. S. 1993. Vibrio vulnificus from raw oysters. Leading cause of reported deaths fr om foodborne illness in Florida. J. Fla. Med. Assoc. 80:536-538. 9. Gulig, P. A., Bourdage, K. L., and Star ks, A. M. 2005. Molecular Pathogenesis of Vibrio vulnificus. J. Microbiol. 43:118-131. 10. Food and Nutrition Board Institute of Medicine 1991. Seafood Safety/Committee on Evaluation of the Safe ty of Fishery Products National Academy Pre ss. Washington, D.C. 11. Bisharat, N., Agmon, V., Finkelstein, R., Raz, R., Ben Dror, G., Lerner, L., Soboh, S., Colodner, R., Cameron, D. N., Wykstra, D. L ., Swerdlow, D. L., and Farmer, J. J., III 1999. Clinical, epidemiological, and microbiological features of Vibrio vulnificus biogroup 3 causing outbreaks of wound infection a nd bacteraemia in Israel. Israel Vibrio Study Group. Lancet 354 :1421-1424. 12. Martin, S. J., and Siebeling, R. J. 1991. Identification of Vibrio vulnificus O serovars with antilipopolysaccharide monoclonal antibody. J. Clin. Microbiol. 29 :1684-1688. 13. Ryang, D. W., Cho, S. W., Shin, M. G., Shin, J. H., and Suh, S. P. 1997. Molecular typing of Vibrio vulnificus isolated by random amplified polymorphic DNA (RAPD) analysis. Jpn. J. Med. Sci. Biol. 50 :113-121. 174

PAGE 175

14. Gutacker, M., Conza, N., Benagli, C., Pedro li, A., Bernasconi, M. V., Permin, L., Aznar, R., and Piffaretti, J. C. 2003. Population genetics of Vibrio vulnificus : identification of two divisions and a distin ct eel-pathogenic clone. Appl. Environ. Microbiol. 69:32033212. 15. Nilsson, W. B., Paranjype, R. N., DePaola, A., and Strom, M. S. 2003. Sequence polymorphism of the 16S rRNA gene of Vibrio vulnificus is a possible indi cator of strain virulence. J. Clin. Microbiol. 41:442-446. 16. Warner, J. M., and Oliver, J. D. 1999. Randomly amplified polymorphic DNA analysis of clinical and environmental isolates of Vibrio vulnificus and other vibrio species. Appl. Environ. Microbiol. 65 :1141-1144. 17. Rosche, T. M., Yano, Y., and Oliver, J. D. 2005. A rapid and simple PCR analysis indicates there ar e two subgroups of Vibrio vulnificus which correlate with clinical or environmental isolation. Microbiol. Immunol. 49:381-389. 18. Chatzidaki-Livanis, M., Hubbard, M. A., Gordon, K., Harwood, V. J., and Wright, A. C. 2006. Genetic distinctions among clini cal and environmental strains of Vibrio vulnificus Appl. Environ. Microbiol. 72:6136-6141. 19. Cohen, A. L., Oliver, J. D., DePaola, A., Feil, E. J., and Boyd, E. F. 2007. Emergence of a virulent clade of Vibrio vulnificus and correlation with the presence of a 33-kilobase genomic island. Appl. Environ. Microbiol. 73:5553-5565. 20. DePaola, A., Nordstrom, J. L., Dalsgaar d, A., Forslund, A., Oliver, J. D., Bates, T., Bourdage, K. L., and Gulig, P. A. 2003. Analysis of Vibrio vulnificus from market oysters and septicemia cases for virulence markers. Appl. Environ. Microbiol. 69:40064011. 21. Falkow, S. 1988. Molecular Koch's postula tes applied to microbial pathogenicity. Rev. Infect. Dis. 10:S274-S276. 22. Yoshida, S., Ogawa, M., and Mizuguchi, Y. 1985. Relation of capsular materials and colony opacity to virulence of Vibrio vulnificus. Infect. Immun. 47:446-451. 23. Simpson, L. M., White, V. K., Zane, S. F ., and Oliver, J. D. 1987. Correlation between virulence and colony morphology in Vibrio vulnificus Infect. Immun. 55:269-272. 24. Wright, A. C., Simpson, L. M., Oliver, J. D., and Morris, J. G., Jr. 1990. Phenotypic evaluation of acapsular transposon mutants of Vibrio vulnificus Infect. Immun. 58:17691773. 25. Wright, A. C., Simpson, L. M., and Oliver J. D. 1981. Role of iron in the pathogenesis of Vibrio vulnificus infections. Infect. Immun. 34:503-507. 26. Litwin, C. M., Rayback, T. W., and Ski nner, J. 1996. Role of catechol siderophore synthesis in Vibrio vulnificus virulence. Infect. Immun. 64:2834-2838. 175

PAGE 176

27. Kim, Y. R., and Rhee, J. H. 2003. Flagellar basal body flg operon as a virulence determinant of Vibrio vulnificus Biochem. Biophys. Res. Commun. 304:405-410. 28. Lee, J. H., Rho, J. B., Park, K. J., Kim, C. B., Han, Y. S., Choi, S. H., Lee, K. H., and Park, S. J. 2004. Role of flagellum and motility in pathogenesis of Vibrio vulnificus Infect. Immun. 72:4905-4910. 29. Paranjpye, R. N., and Strom, M. S. 2005. A Vibrio vulnificus type IV pilin contributes to biofilm formation, adherence to epit helial cells, and virulence. Infect. Immun. 73:14111422. 30. Goo, S. Y., Lee, H. J., Kim, W. H., Han, K. L., Park, D. K., Lee, H. J., Kim, S. M., Kim, K. S., Lee, K. H., and Park, S. J. 2006. Identification of OmpU of Vibrio vulnificus as a fibronectin-binding protein and its ro le in bacterial pathogenesis. Infect. Immun. 74:5586-5594. 31. Kim, Y. R., Lee, S. E., Kim, C. M., Kim, S. Y., Shin, E. K., Shin, D. H., Chung, S. S., Choy, H. E., Progulske-Fox, A., Hillman, J. D., Handfield, M., and Rhee, J. H. 2003. Characterization and pat hogenic significance of Vibrio vulnificus antigens preferentially expressed in septicemic patients. Infect. Immun. 71:5461-5471. 32. Liu, M., Alice, A. F., Naka, H., and Crosa, 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 Infect. Immun. 75:3282-3289. 33. Lee, J. H., Kim, M. W., Kim, B. S., Kim, S. M., Lee, B. C., Kim, T. S., and Choi, S. H. 2007. Identification and characterization of the Vibrio vulnificus rtxA essential for cytotoxicity in vitro and virulence in mice. J. Microbiol. 45 :146-152. 34. Bourdage,K., Starks,A., and Gulig,P. 2003. The RtxA toxin is a major cytotoxic factor but plays a minor role in virulence of Vibrio vulnificus Abstr. Annu. Meet. Am. Soc. Microbiol. B-58: (Abstr.) 35. Ashrafian, H. 2003. Hepcidin: the mi ssing link between hemochromatosis and infections. Infect. Immun. 71:6693-6700. 36. Helms, S. D., Oliver, J. D., and Travis, J. C. 1984. Role of heme compounds and haptoglobin in Vibrio vulnificus pathogenicity. Infect. Immun. 45:345-349. 37. Hor, L. I., Chang, Y. K., Chang, C. C., Le i, H. Y., and Ou, J. T. 2000. Mechanism of high susceptibility of iron-overloaded mouse to Vibrio vulnificus infection. Microbiol. Immunol. 44:871-878. 38. Starks, A. M., Bourdage, K. L., Thiaville, P. C., and Gulig, P. A. 2006. Use of a marker plasmid to examine growth and death of Vibrio vulnificus in infected mice. Mol. Microbiol. 61 :310-323. 176

PAGE 177

39. Paranjpye, R. N., Lara, J. C., Pepe, J. C., Pepe, C. M., and Strom, M. S. 1998. The type IV leader peptidase/ N -methyltransferase of Vibrio vulnificus controls factor s required for adherence to HEp-2 cells and viru lence in iron-overloaded mice. Infect. Immun. 66:5659-5668. 40. Kreger, A., and Lockwood, D. 1981. Detect ion of extracellular t oxin(s) produced by Vibrio vulnificus Infect. Immun. 33:583-590. 41. Gray, L. D., and Kreger, A. S. 1987. M ouse skin damage caused by cytolysin from Vibrio vulnificus and by V. vulnificus infection. J. Infect. Dis. 155:236-241. 42. Wright, A. C., and Morris, J. G., Jr. 1991. The extrace llular cytolysin of Vibrio vulnificus : inactivation and relationship to virulence in mice. Infect. Immun. 59:192-197. 43. Shao, C. P., and Hor, L. I. 2000. Metalloprotease is not essential for Vibrio vulnificus virulence in mice. Infect. Immun. 68:3569-3573. 44. Jeong, K. C., Jeong, H. S., Rhee, J. H., Lee, S. E., Chung, S. S., Starks, A. M., Escudero, G. M., Gulig, P. A., and Choi, S. H. 2000. Construction and phenotypic evaluation of a Vibrio vulnificus vvpE mutant for elastolytic protease. Infect. Immun. 68:5096-5106. 45. Gray, L. D., and Kreger, A. S. 1989. Detection of Vibrio vulnificus cytolysin in V. vulnificus infected mice. Toxicon 27:439-464. 46. Lee, Y. R., Park, K. H., Lin, Z. Z., Kho, Y. J., Park, J. W., Rho, H. W., Koo, B. S., Kim, H. R., Song, E. K., Yu, H. N., Han, M. K., Lee, S. O., Jhee, E. C., and Kim, J. S. 2004. A calcium-calmodulin antagonist blocks experimental Vibrio vulnificus cytolysin-induced lethality in an experimental mouse model. Infect. Immun. 72:6157-6159. 47. Chen, Y. C., Chang, M. C., Chuang, Y. C ., and Jeang, C. L. 2004. Characterization and virulence of hemolysin III from Vibrio vulnificus Curr. Microbiol. 49:175-179. 48. Smith, G. C., and Merkel, J. R. 1982. Collagenolytic activity of Vibrio vulnificus : potential contribution to its invasiveness. Infect. Immun. 35:1155-1156. 49. Kothary, M. H., and Kreger, A. S. 1985. Production and partial characterization of an elastolytic protease of Vibrio vulnificus Infect. Immun. 50:534-540. 50. Miyoshi, N., Shimizu, C., Miyoshi, S., and Shinoda, S. 1987. Purification and characterization of Vibrio vulnificus protease. Microbiol. Immunol. 31:13-25. 51. Fan, J. J., Shao, C. P., Ho, Y. C., Yu, C. K., and Hor, L. I. 2001. Isolation and characterization of a Vibrio vulnificus mutant deficient in both extracellular metalloprotease and cytolysin. Infect. Immun. 69:5943-5948. 52. Wu, S. I., Lo, S. K., Shao, C. P., Tsai, H. W., and Hor, L. I. 2001. Cloning and characterization of a pe riplasmic nuclease of Vibrio vulnificus and its role in preventing uptake of foreign DNA. Appl. Environ. Microbiol. 67:82-88. 177

PAGE 178

53. Oliver, J. D., Wear, J. E., Thomas, M. B., Warner, M., and Linder, K. 1986. Production of extracellular enzymes and cytotoxicity by Vibrio vulnificus. Diagn. Microbiol. Infect. Dis. 5:99-111. 54. Kim, Y. R., Lee, S. E., Kook, H., Yeom, J. A., Na, H. S., Kim, S. Y., Chung, S. S., Choy, H. E., and Rhee, J. H. 2008. Vibrio vulnificus RTX toxin kills host cells only after contact of the bacteria with host cells. Cell Microbiol. 10:848-862. 55. Starks, A. M., Schoeb, T. R., Tamplin, M. L., Parveen, S., Doyle, T. J., Bomeisl, P. E., Escudero, G. M., and Gulig, P. A. 2000. Pa thogenesis of infec tion by clinical and environmental strains of Vibrio vulnificus in iron dextran-treated mice. Infect. Immun. 68:5785-5793. 56. Manoil, C., and Beckwith, J. 1985. Tn phoA : A transposon probe for protein export signals. Proc. Natl. Acad. Sci. U. S. A. 82:8129-8133. 57. de Lorenzo, V., Herrero, M., Jakubzi k, U., and Timmis, K. N. 1990. Mini-Tn5 transposon derivatives for insertion mutage nesis, promoter pr obing, and chromosomal insertion of cloned DNA in gram-negative eubacteria. J. Bacteriol. 172 :6568-6572. 58. Cerda-Cuellar, M., Jofre, J., and Blanch, A. R. 2000. A selective medium and a specific probe for detection of Vibrio vulnificus Appl. Environ. Microbiol. 66:855-859. 59. Sambrook,J., Fritsch,E.F., and Maniatis,T. 1989. Molecular cloning: a laboratory manual Cold Spring Harbor laborato ry. Cold Spring Harbor, N.Y. 60. Provence,D.L., and Curtiss III,R. 1994. Gene transfer in gram-n egative bacteria. In Methods for general and molecular bacteriology P.Gerhardt, Murray,R.G.E., Wood,W.A., and Krieg,N.R., editors. Ameri can Society for Microbiology. Washington, D.C. 317-347. 61. Campbell, J. W., and Cronan, J. E., Jr. 2001. Escherichia coli FadR positively regulates transcription of the fabB fatty acid biosynthetic gene. J. Bacteriol. 183 :5982-5990. 62. Enos-Berlage, J. L., and McCarter, L. L. 2000. Relation of capsular polysaccharide production and colonial cell orga nization to colony morphology in Vibrio parahaemolyticus J. Bacteriol. 182 :5513-5520. 63. Kelley, J. T., and Parker, C. D. 1981. Iden tification and preliminar y characterization of Vibrio cholerae outer membrane proteins. J. Bacteriol. 145:1018-1024. 64. Taylor, R. K., Manoil, C., and Mekalanos, J. J. 1989. Broad-host-range vectors for delivery of TnphoA : Use in genetic analysis of secr eted virulence determinants of Vibrio cholerae. J. Bacteriol. 171 :1870-1878. 65. Lehoux, D. E., Sanchagrin, F., and Levesque R. C. 1999. Defined oligonucleotide tag pools and PCR screening in signature-tagge d mutagenesis of essential genes from bacteria. Biotechniques 26:473-480. 178

PAGE 179

66. Derman, A. I., a nd Beckwith, J. 1995. Escherichia coli alkaline phosphatase localized to the cytoplasm slowly acquires enzymatic ac tivity in cells whos e growth has been suspended: a caution for gene fusion studies. J. Bacteriol. 177 :3764-3770. 67. Marrakchi, H., Zhang, Y. M., and Roc k, C. O. 2002. Mechanistic diversity and regulation of Type II fatty acid synthesis. Biochem. Soc. Trans. 30:1050-1055. 68. Missiakas, D., and Raina, S. 1997. Protei n folding in the bacterial periplasm. J. Bacteriol. 179 :2465-2471. 69. Donnenberg, M. S., and Kaper, J. B. 1991. Construction of an eae deletion mutant of Enteropathogenic Escherichia coli by using a positive-selection suicide vector. Infect. Immun. 59:4310-4317. 70. Meibom, K. L., Blokesch, M., Dolganov, N. A., Wu, C. Y., and Schoolnik, G. K. 2005. Chitin induces natural competence in Vibrio cholerae. Science 310:1824-1827. 71. Simon, R., Priefer, U., and Phler, A. 1983. A broad host range mobilization system for in vivo genetic engineering: tr ansposon mutagenesis in gram negative bacteria. Bio/Technology 1:784-791. 72. Scott, H. N., Laible, P. D., and Hanson, D. K. 2003. Sequences of versatile broad-hostrange vectors of the RK2 family. Plasmid 50:74-79. 73. Taylor, L. A., and Rose, R. E. 1988. A correction in the nucleotide sequence of the Tn 903 kanamycin resistance determinant in pUC4K. Nucleic Acids Res. 16:358. 74. Hoffman, C. S., and Wright, A. 1985. Fusi ons of secreted prot eins to alkaline phosphatase: an approach for studying protein secretion. Proc. Natl. Acad. Sci. U. S. A 82:5107-5111. 75. Manoil, C., Mekalanos, J. J., and Beckwith, J. 1990. Alkaline phosphatase fusions: sensors of subce llular location. J. Bacteriol. 172 :515-518. 76. Boyd, D., Manoil, C., and Beckwith, J. 1987. Determinants of membrane protein topology. Proc. Natl. Acad. Sci. U. S. A 84:8525-8529. 77. Gibson, C. M., and Caparon, M. G. 2002. Al kaline phosphatase reporter transposon for identification of genes encoding secreted proteins in gram-positive microorganisms. Appl. Environ. Microbiol. 68:928-932. 78. Kolter, R., Inuzuka, M., and Helinski, D. R. 1978. Trans-complementation-dependent replication of a low molecular weight origin fragment from plasmid R6K. Cell 15:11991208. 79. Shiba, T., Hill, R. T., Straube, W. L., and Colwell, R. R. 1995. Decrease in culturability of Vibrio cholerae caused by glucose. Appl. Environ. Microbiol. 61:2583-2588. 179

PAGE 180

80. Gardy, J. L., Spencer, C., Wang, K., Ester, M., Tusnady, G. E., Simon, I., Hua, S., deFays, K., Lambert, C., Nakai, K., and Brinkman, F. S. 2003. PSORT-B: Improving protein subcellular loca lization prediction for Gram-negative bacteria. Nucleic Acids Res. 31:3613-3617. 81. Kachlany, S. C., Planet, P. J., Bhattacharjee, M. K., Kollia, E., DeSalle, R., Fine, D. H., and Figurski, D. H. 2000. Nonspecific adherence by Actinobacillus actinomycetemcomitans requires genes widespread in bacteria and archaea. J. Bacteriol. 182:6169-6176. 82. Rendulic, S., Jagtap, P., Rosinus, A., Eppi nger, M., Baar, C., Lanz, C., Keller, H., Lambert, C., Evans, K. J., Goesmann, A., Meyer, F., Sockett, R. E., and Schuster, S. C. 2004. A predator unmasked: life cycle of Bdellovibrio bacteriovorus from a genomic perspective. Science 303 :689-692. 83. Fath, M. J., Mahanty, H. K., and Ko lter, R. 1989. Characterization of a purF operon mutation which affects colicin V production. J. Bacteriol. 171 :3158-3161. 84. Simons, R. W., Egan, P. A., Chute, H. T., and Nunn, W. D. 1980. Regulation of fatty acid degradation in Escherichia coli : isolation and characteri zation of strains bearing insertion and temperature-se nsitive mutations in gene fadR J. Bacteriol. 142:621-632. 85. Nunn, W. D., Giffin, K., Clark, D., and Cronan, J. E., Jr. 1983. Role for fadR in unsaturated fatty acid biosynthesis in Escherichia coli J. Bacteriol. 154:554-560. 86. Gui, L., Sunnarborg, A., and LaPorte, D. C. 1996. Regulated expr ession of a repressor protein: FadR activates iclR J. Bacteriol. 178 :4704-4709. 87. Farewell, A., Diez, A. A., DiRusso, C. C., and Nystrom, T. 1996. Role of the Escherichia coli FadR regulator in stasis survival and growth phase-dependent expression of the uspA, fad, and fab genes. J. Bacteriol. 178:6443-6450. 88. Day, A. P., and Oliver, J. D. 2004. Change s in membrane fatty acid composition during entry of Vibrio vulnificus into the viable but nonculturable state. J. Microbiol. 42:69-73. 89. Buttke, T. M., and Ingram, L. O. 1978. Inhi bition of unsaturated fatty acid synthesis in Escherichia coli by the antibiotic cerulenin. Biochemistry 17 :5282-5286. 90. Price, A. C., Choi, K. H., Heath, R. J ., Li, Z., White, S. W., and Rock, C. O. 2001. Inhibition of beta-ketoacylacyl carrier protei n synthases by thiolactomycin and cerulenin. Structure and mechanism. J. Biol. Chem. 276:6551-6559. 91. Fujita, Y., Matsuoka, H., and Hirooka, K. 2007. Regulation of fatty acid metabolism in bacteria. Mol. Microbiol. 66:829-839. 92. Iram, S. H., and Cronan, J. E. 2006. The beta-oxidation systems of Escherichia coli and Salmonella enterica are not functionally equivalent. J. Bacteriol. 188 :599-608. 180

PAGE 181

93. van Aalten, D. M., DiRusso, C. C., Knudse n, J., and Wierenga, R. K. 2000. Crystal structure of FadR, a fatty acid-responsive transcription factor with a novel acyl coenzyme A-binding fold. EMBO J. 19:5167-5177. 94. DiRusso, C. C., and Nystrom, T. 1998. The fats of Escherichia coli during infancy and old age: regulation by global regulators, alarmones and lipid intermediates. Mol. Microbiol. 27 :1-8. 95. McCarter, L. L. 2001. Polar flag ellar motility of th e Vibrionaceae. Microbiol. Mol. Biol. Rev. 65:445-62, table. 96. DiRusso, C. C., Black, P. N., and Weimar J. D. 1999. Molecular inroads into the regulation and metabolism of fatty acids, lessons from bacteria. Prog. Lipid Res. 38:129197. 97. Vereecke, D., Cornelis, K., Temmerman, W ., Jaziri, M., Van Montagu, M., Holsters, M., and Goethals, K. 2002. Chromosomal lo cus that affects pathogenicity of Rhodococcus fascians J. Bacteriol. 184 :1112-1120. 98. Munoz-Elias, E. J., and McKinney, J. D. 2005. Mycobacterium tuberculosis isocitrate lyases 1 and 2 are jointly required for in vivo growth and virulence. Nat. Med. 11:638644. 99. Lorenz, M. C., and Fink, G. R. 2002. Li fe and death in a macrophage: role of the glyoxylate cycle in virulence. Eukaryot. Cell 1 :657-662. 100. Chung, T., Klumpp, D. J., and LaPorte, D. C. 1988. Glyoxylate bypass operon of Escherichia coli : cloning and determination of the functional map. J. Bacteriol. 170:386392. 101. Mahan, M. J., Slauch, J. M., and Mekalanos, J. J. 1993. Selection of bacterial virulence genes that are specifically induced in host tissues. Science 259 :686-688. 102. Spector, M. P., DiRusso, C. C., Pallen, M. J ., Garcia, d. P., Dougan, G., and Finlay, B. B. 1999. The medium-/long-chain fatty acyl-CoA dehydrogenase ( fadF ) gene of Salmonella typhimurium is a phase 1 starvation-stre ss response (SSR) locus. Microbiology 145 ( Pt 1) :15-31. 103. Lucas, R. L., Lostroh, C. P., DiRusso, C. C., Spector, M. P., Wanner, B. L., and Lee, C. A. 2000. Multiple factors independently regulate hilA and invasion gene expression in Salmonella enterica serovar Typhimurium. J. Bacteriol. 182 :1872-1882. 104. Black, P. N., and DiRusso, C. C. 2003. Transmembrane movement of exogenous longchain fatty acids: proteins, enzyme s, and vectorial esterification. Microbiol. Mol. Biol. Rev. 67:454-72, table. 181

PAGE 182

105. DiRusso, C. C., and Black, P. N. 1999. Longchain fatty acid transp ort in bacteria and yeast. Paradigms for defining the mechanism underlying this proteinmediated process. Mol. Cell Biochem. 192:41-52. 106. Dartigalongue, C., Missiakas, D., and Raina, S. 2001. Characterization of the Escherichia coli sigma E regulon. J. Biol. Chem. 276:20866-20875. 107. Iram, S. H., and Cronan, J. E. 2005. Unexp ected functional diversity among FadR fatty acid transcriptional regulatory proteins. J. Biol. Chem. 280:32148-32156. 108. Mahan, M. J., Tobias, J. W., Slauch, J. M., Hanna, P. C., and Collier, R. J. 1995. Antibiotic-based selection for bacterial ge nes that are specific ally induced during infection of a host. Proc. Natl. Acad. Sci. U. S. A. 92:669-673. 109. Fozo, E. M., and Quivey, R. G., Jr. 2004. The fabM gene product of Streptococcus mutans is responsible for the synthesis of monou nsaturated fatty acids and is necessary for survival at low pH. J. Bacteriol. 186:4152-4158. 110. Fozo, E. M., Scott-Anne, K., Koo, H., and Quivey, R. G., Jr. 2007. Role of unsaturated fatty acid biosynthesis in virulence of Streptococcus mutans Infect. Immun. 75:15371539. 111. Krivan, H. C., Franklin, D. P., Wan g, W., Laux, D. C., and Cohen, P. S. 1992. Phosphatidylserine found in intestinal mucu s serves as a sole source of carbon and nitrogen for salmonellae and Escherichia coli Infect. Immun. 60:3943-3946. 112. Hogan, J. S., Pankey, J. W., and Duthie, A. H. 1987. Growth responses of Staphylococcus aureus and Streptococcus agalactiae to Corynebacterium bovis metabolites J. Dairy Sci. 70:1294-1301. 113. Smith, H. 1990. Pathogenicity and the microbe in vivo J. Gen. Microbiol. 136:377-383. 114. Heath, R. J., White, S. W., and Rock, C. O. 2002. Inhibitors of fatty acid synthesis as antimicrobial chemotherapeutics. Appl. Microbiol. Biotechnol. 58:695-703. 115. Nomura, S., Horiuchi, T., Omura, S., and Hata, T. 1972. The action mechanism of cerulenin. I. Effect of cerulenin on sterol and fatty acid biosynthesis in yeast. J. Biochem. 71:783-796. 116. Ruiz, N., and Silhavy, T. J. 2005. Se nsing external stress: watchdogs of the Escherichia coli cell envelope. Curr. Opin. Microbiol. 8:122-126. 117. Endemann, H., and Model, P. 1995. Location of filamentous phage minor coat proteins in phage and in infected cells. Journal of Molecular Biology 250:496-506. 118. Rowley, G., Spector, M., Kormanec, J., a nd Roberts, M. 2006. Pushing the envelope: extracytoplasmic stress responses in bacterial pathogens. Nat. Rev. Microbiol. 4:383394. 182

PAGE 183

119. De Wulf, P., McGuire, A. M., Liu, X., and Lin, E. C. 2002. Genome-wide profiling of promoter recognition by the two-com ponent response regulator CpxR-P in Escherichia coli J. Biol. Chem. 277:26652-26661. 120. Raivio, T. L., and Silhavy, T. J. 2001. Periplasmic stress and ECF sigma factors. Annu. Rev. Microbiol. 55:591-624. 121. Raffa, R. G., and Raivio, T. L. 2002. A third envelope stress signal transduction pathway in Escherichia coli Mol. Microbiol. 45:1599-1611. 122. Nagakubo, S., Nishino, K., Hirata, T., and Yamaguchi, A. 2002. The putative response regulator BaeR stimulates multidrug resistance of Escherichia coli via a novel multidrug exporter system, MdtABC. J. Bacteriol. 184:4161-4167. 123. Hirakawa, H., Nishino, K., Hirata, T., a nd Yamaguchi, A. 2003. Comprehensive studies of drug resistance mediated by overexpression of response regulators of two-component signal transduction systems in Escherichia coli J. Bacteriol. 185:1851-1856. 124. Brissette, J. L., Russel, M., Weiner, L., a nd Model, P. 1990. Phage shock protein, a stress protein of Escherichia coli Proc. Natl. Acad. Sci. U. S. A 87:862-866. 125. Kleerebezem, M., Crielaard, W., and Tommassen, J. 1996. Involvement of stress protein PspA (phage shock protein A) of Escherichia coli in maintenance of the protonmotive force under stress conditions. EMBO J. 15:162-171. 126. Erickson, J. W., and Gross, C. A. 1989. Identification of the sigma E subunit of Escherichia coli RNA polymerase: a second alternate sigma factor involved in hightemperature gene expression. Genes Dev. 3:1462-1471. 127. De Las, P. A., Connolly, L., and Gross, C. A. 1997. SigmaE is an essential sigma factor in Escherichia coli J. Bacteriol. 179 :6862-6864. 128. De Las, P. A., Connolly, L., and Gross, C. A. 1997. The sigmaE-mediated response to extracytoplasmic stress in Escherichia coli is transduced by RseA and RseB, two negative regulators of sigmaE. Mol. Microbiol. 24:373-385. 129. Missiakas, D., Mayer, M. P., Lemaire, M., Georgopoulos, C., and Raina, S. 1997. Modulation of the Escherichia coli sigmaE (RpoE) heat-shoc k transcription-factor activity by the RseA, RseB and RseC proteins. Mol. Microbiol. 24:355-371. 130. Collinet, B., Yuzawa, H., Chen, T., Herrera, C., and Missiakas, D. 2000. RseB binding to the periplasmic domain of Rs eA modulates the RseA:sigmaE interaction in the cytoplasm and the availability of sigmaE RNA polymerase. J. Biol. Chem. 275:33898-33904. 131. Cezairliyan, B. O., and Sauer, R. T. 2007. Inhibition of regulated proteolysis by RseB. Proc. Natl. Acad. Sci. U. S. A 104:3771-3776. 183

PAGE 184

132. Grigorova, I. L., Chaba, R., Zhong, H. J., Alba, B. M., Rhodius, V., Herman, C., and Gross, C. A. 2004. Fine-tuning of the Escherichia coli sigmaE envelope stress response relies on multiple mechanisms to inhibit signal-independent proteolysis of the transmembrane anti-sigma factor, RseA. Genes Dev. 18:2686-2697. 133. Martin, D. W., Holloway, B. W., and Dere tic, V. 1993. Characte rization of a locus determining the mucoid status of Pseudomonas aeruginosa: AlgU shows sequence similarities with a Bacillus sigma factor. J. Bacteriol. 175:1153-1164. 134. Martin, D. W., Schurr, M. J., Mudd, M. H., Govan, J. R., Holloway, B. W., and Deretic, V. 1993. Mechanism of conversion to mucoidy in Pseudomonas aeruginosa infecting cystic fibrosis patients. Proc. Natl. Acad. Sci. U. S. A 90:8377-8381. 135. Martin, D. W., Schurr, M. J., Mudd, M. H., and Deretic, V. 1993. Differentiation of Pseudomonas aeruginosa into the alginate-producing form: inactivation of mucB causes conversion to mucoidy. Mol. Microbiol. 9:497-506. 136. Goldberg, J. B., Gorman, W. L., Flynn, J. L., and Ohman, D. E. 1993. A mutation in algN permits trans activation of alginate production by algT in Pseudomonas species J. Bacteriol. 175 :1303-1308. 137. Gilligan, P. H. 1991. Microbiology of airway disease in patients with cystic fibrosis. Clin. Microbiol. Rev. 4 :35-51. 138. Wright, A. C., Powell, J. L., Kaper, J. B ., and Morris, J. G., Jr. 2001. Identification of a group 1-like capsular polysaccharide operon for Vibrio vulnificus. Infect. Immun. 69:6893-6901. 139. Grau, B. L., Henk, M. C., and Pettis, G. S. 2005. High-frequency phase variation of Vibrio vulnificus 1003: isolation and char acterization of a rugose phenotypic variant. J. Bacteriol. 187 :2519-2525. 140. Novelli, A. 1953. New method of staining of bacterial capsules in films and sections. Experientia 9:34-35. 141. Nakhamchik, A., Wilde, C., and Rowe-Ma gnus, D. A. 2007. Identification of a Wzy polymerase required for group IV capsular polysaccharide and lipopolysaccharide biosynthesis in Vibrio vulnificus Infect. Immun. 75:5550-5558. 142. Chatzidaki-Livanis, M., Jones, M. K., and Wright, A. C. 2006. Genetic variation in the Vibrio vulnificus group 1 capsular polysaccharide operon. J. Bacteriol. 188 :1987-1998. 143. Joseph, L. A., and Wright, A. C. 2004. Expression of Vibrio vulnificus capsular polysaccharide inhibits biofilm formation. J. Bacteriol. 186 :889-893. 144. Shinoda, S., Kobayashi, M., Yamada, H., Yoshida, S., Ogawa, M., and Mizuguchi, Y. 1987. Inhibitory effect of capsular antigen of Vibrio vulnificus on bactericidal activity of human serum. Microbiol. Immunol. 31:393-401. 184

PAGE 185

145. Warne, S. R., Varley, J. M., Boulnois, G. J., and Norton, M. G. 1990. Identification and characterization of a gene that controls colony morphol ogy and auto-aggregation in Escherichia coli K12. J. Gen. Microbiol. 136:455-462. 146. Rouviere, P. E., De Las, P. A., Mecsas, J., Lu, C. Z., Rudd, K. E., and Gross, C. A. 1995. rpoE the gene encoding the second heat-s hock sigma factor, sigma E, in Escherichia coli EMBO J. 14:1032-1042. 147. Lipinska, B., Sharma, S., and Georgopoulos C. 1988. Sequence analysis and regulation of the htrA gene of Escherichia coli : a sigma 32-independent mechanism of heatinducible transcription. Nucleic Acids Res. 16:10053-10067. 148. Strauch, K. L., and Beckwith, J. 1988. An Escherichia coli mutation preventing degradation of abnormal periplasmic proteins. Proc. Natl. Acad. Sci. U. S. A 85:15761580. 149. Seol, J. H., Woo, S. K., Jung, E. M., Yoo, S. J., Lee, C. S., Kim, K. J., Tanaka, K., Ichihara, A., Ha, D. B., and Chung, C. H. 1991. Protease Do is essential for survival of Escherichia coli at high temperatures: its identity with the htrA gene product. Biochem. Biophys. Res. Commun. 176:730-736. 150. Pallen, M. J., and Wren, B. W. 1997. Th e HtrA family of serine proteases. Mol. Microbiol. 26 :209-221. 151. Kim, D. Y., and Kim, K. K. 2005. Structure and function of HtrA family proteins, the key players in protein quality control. J. Biochem. Mol. Biol. 38:266-274. 152. Rhodius, V. A., Suh, W. C., Nonaka, G., West, J., and Gross, C. A. 2006. Conserved and variable functions of the sigmaE st ress response in related genomes. PLoS. Biol. 4:e2. 153. Swanson, J. 1982. Colony opacity and protein II compositions of gonococci. Infect. Immun. 37:359-368. 154. Robertson, B. D., and Meyer, T. F. 1992. Genetic variation in pathogenic bacteria. Trends Genet. 8:422-427. 155. Ahn, S. J., Lemos, J. A., and Burne, R. A. 2005. Role of HtrA in growth and competence of Streptococcus mutans UA159. J. Bacteriol. 187 :3028-3038. 156. Wright, A. C., Powell, J. L., Tanner, M. K., Ensor, L. A., Karpas, A. B., Morris, J. G., Jr., and Sztein, M. B. 1999. Di fferential expression of Vibrio vulnificus capsular polysaccharide. Infect. Immun. 67:2250-2257. 157. Rosche, T. M., Smith, B., and Oliver, J. D. 2006. Evidence for an intermediate colony morphology of Vibrio vulnificus Appl. Environ. Microbiol. 72 :4356-4359. 158. Kim, H. S., Lee, M. A., Chun, S. J., Park, S. J., and Lee, K. H. 2007. Role of NtrC in biofilm formation via controlling expre ssion of the gene encoding an ADP-glycero185

PAGE 186

186 manno-heptose-6-epimerase in the pathogenic bacterium, Vibrio vulnificus Mol. Microbiol. 63 :559-574. 159. Nakhamchik, A., Wilde, C., and Rowe-Ma gnus, D. A. 2008. Cyclic-di-GMP regulates extracellular polysaccharid e production, biofilm formation, and rugose colony development by Vibrio vulnificus Appl. Environ. Microbiol. 74:4199-4209. 160. Schurr, M. J., Yu, H., Martinez-Salazar, J. M., Boucher, J. C., and Deretic, V. 1996. Control of AlgU, a member of the sigma E-like family of stress sigma factors, by the negative regulators MucA and MucB and Pseudomonas aeruginosa conversion to mucoidy in cystic fibrosis. J. Bacteriol. 178:4997-5004. 161. Rowen, D. W., and Deretic, V. 2000. Memb rane-to-cytosol redist ribution of ECF sigma factor AlgU and conversion to mucoidy in Pseudomonas aeruginosa isolates from cystic fibrosis patients. Mol. Microbiol. 36 :314-327. 162. Pallen, M. J., and Wren, B. W. 2007. Bacterial pathogenomics. Nature 449:835-842. 163. Field, D., Hood, D., and Moxon, R. 1999. Contribution of genomics to bacterial pathogenesis. Curr. Opin. Genet. Dev. 9:700-703. 164. Hilbi, H., Weber, S. S., Ragaz, C., Nyfeler, Y., and Urwyler, S. 2007. Environmental predators as models for bacterial pathogenesis. Environ. Microbiol. 9:563-575. 165. Smith, H. 2000. Questions about th e behaviour of bacterial pathogens in vivo Philos. Trans. R. Soc. Lond B Biol. Sci. 355 :551-564. 166. Samant, S., Lee, H., Ghassemi, M., Chen, J., Cook, J. L., Mankin, A. S., and Neyfakh, A. A. 2008. Nucleotide biosynthesis is critical for growth of bacteria in human blood. PLoS. Pathog. 4:e37. 167. Bullen, J. J., Spalding, P. B., War d, C. G., and Gutteridge, J. M. 1991. Hemochromatosis, iron and septicemia caused by Vibrio vulnificus Arch. Intern. Med. 151:1606-1609.

PAGE 187

BIOGRAPHICAL SKETCH Roslyn N. Brown is a native of the tropical island paradise of Trinidad and Tobago. After completing her high school education in Tr inidad, she attended Savannah State University in Savannah, Georgia. Although Roslyns underg raduate career was first funded by a tennis scholarship, she was later awarded Savannah State Universitys Presidential Scholarship. After graduating, Roslyn moved to Florida to pursue a graduate degree in the Interdisciplinary Program in Biomedical Sciences (IDP) as a Gra duate Alumni Fellow. During her tenure at the University of Florida she made many close fr iends, though none was as close as fellow IDP student, Joseph Brown, whom sh e married in October of 2007. Roslyn looks forward to a rewarding career as a microbiologist. 187