<%BANNER%>

Interactions of Coral Associated Bacteria with Quorum Sensing and the Potential Contribution to Defense of the Holobiont...

Permanent Link: http://ufdc.ufl.edu/UFE0022780/00001

Material Information

Title: Interactions of Coral Associated Bacteria with Quorum Sensing and the Potential Contribution to Defense of the Holobiont against Pathogens
Physical Description: 1 online resource (99 p.)
Language: english
Creator: Halbig, Stephanie
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: associated, bacteria, coral, pdl100, quorum, sensing, swarming
Plant Molecular and Cellular Biology -- Dissertations, Academic -- UF
Genre: Plant Molecular and Cellular Biology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Bacteria have been observed in and on coral tissues since the early 20th century. However, it is only recently that investigation in to the potential role of associated bacteria in coral polyp health has begun to take place. One putative responsibility of coral associated bacteria may be as a host defense mechanisms against pathogen infection. Quorum sensing (QS) is the processes by which bacterial colonies regulate multicellular behavior with many pathogens using QS to control virulence factor expression. In this study, coral associated bacteria were identified that were capable of inhibiting QS in model reporter systems. Six coral isolates were identified using 16S rDNA sequence analysis and were found to be Planococcus spp., Photobacterium spp., Marinobacter salsuginis, Agrobacterium stellulatum, Vibrio spp., and Caryophanon spp. A QS agonist from Caryophanon spp. was partially purified from whole culture extracts using HPLC and analyzed with mass spectrometry. The ability of the six coral isolates to inhibit QS controlled surface spreading of the ubiquitous pathogen Serratia marcescens was investigated and supported the isolates interference with QS. In conjunction, the inherent QS system of the pathogen responsible for the White Pox disease of coral, S. marcescens PDL100, was also investigated. Two compounds present in organic extracts of PDL100 were found to stimulate a broad range QS reporter and were susceptible to inactivation by the lactonase enzyme AiiA. Multicellular behaviors of PDL100, including surface spreading and biofilm formation were evaluated in wild type and AiiA expressing strains of PDL100 as well. Overall, the data presented here confirms that coral associated bacteria are capable of inhibiting QS in model systems and potentially contribute to protection of the coral holobiont from pathogens.
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 Stephanie Halbig.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Folta, Kevin M.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-06-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022780:00001

Permanent Link: http://ufdc.ufl.edu/UFE0022780/00001

Material Information

Title: Interactions of Coral Associated Bacteria with Quorum Sensing and the Potential Contribution to Defense of the Holobiont against Pathogens
Physical Description: 1 online resource (99 p.)
Language: english
Creator: Halbig, Stephanie
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: associated, bacteria, coral, pdl100, quorum, sensing, swarming
Plant Molecular and Cellular Biology -- Dissertations, Academic -- UF
Genre: Plant Molecular and Cellular Biology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Bacteria have been observed in and on coral tissues since the early 20th century. However, it is only recently that investigation in to the potential role of associated bacteria in coral polyp health has begun to take place. One putative responsibility of coral associated bacteria may be as a host defense mechanisms against pathogen infection. Quorum sensing (QS) is the processes by which bacterial colonies regulate multicellular behavior with many pathogens using QS to control virulence factor expression. In this study, coral associated bacteria were identified that were capable of inhibiting QS in model reporter systems. Six coral isolates were identified using 16S rDNA sequence analysis and were found to be Planococcus spp., Photobacterium spp., Marinobacter salsuginis, Agrobacterium stellulatum, Vibrio spp., and Caryophanon spp. A QS agonist from Caryophanon spp. was partially purified from whole culture extracts using HPLC and analyzed with mass spectrometry. The ability of the six coral isolates to inhibit QS controlled surface spreading of the ubiquitous pathogen Serratia marcescens was investigated and supported the isolates interference with QS. In conjunction, the inherent QS system of the pathogen responsible for the White Pox disease of coral, S. marcescens PDL100, was also investigated. Two compounds present in organic extracts of PDL100 were found to stimulate a broad range QS reporter and were susceptible to inactivation by the lactonase enzyme AiiA. Multicellular behaviors of PDL100, including surface spreading and biofilm formation were evaluated in wild type and AiiA expressing strains of PDL100 as well. Overall, the data presented here confirms that coral associated bacteria are capable of inhibiting QS in model systems and potentially contribute to protection of the coral holobiont from pathogens.
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 Stephanie Halbig.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Folta, Kevin M.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-06-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022780:00001


This item has the following downloads:


Full Text

PAGE 1

1 INTERACTIONS OF CORAL ASSOCIATED BACTERIA WITH QUORUM SENSING AND THE POTENTIAL CONTRIBU TION TO DEFENSE OF TH E HOLOBIONT AGAINST PATHOGENS By STEPHANIE MARIE HALBIG A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Stephanie Marie Halbig

PAGE 3

3 To my husband, mother, father and sister

PAGE 4

4 ACKNOWLEDGMENTS First and forem ost, I thank my husband for holding me up through the hard bits and celebrating the good bits while constantly tole rating and supporting my completely random career. I also thank my sister for her help in navigating the complex waters of graduate school and for pushing me to be a better scientist than I thought I could be. Cory Krediet, and Dr. Mengsheng Gao contributed greatly to the process, and therefore su ccess, of my experiments and I acknowledge their skills and support. Thelma Madzima encouraged me throughout this degree and her support is greatly apprec iated. My committee has honored me with their wisdom and guidance through this journey and I thank them all for their eternal patience. The Plant Molecular and Cellular Biology Program gave me the opportunity and financial support to pursue this degree and I am grateful. Support fo r the supplies and stipe nd was provided by Mote Marine Laboratory-Protect Our R eefs grant (PI's: Teplitski, Horenstein, Krediet), Charles A. and A. Morrow Lindbergh Foundation award (PI's: Ritchie, Teplitski) and the State of Florida allocations to the Teplitski research program. The contribution of Kevin Folta to the success of this thesis is not quant ifiable, and I am forever in his dept. Finally, I would like to thank my parents for their never-ending s upport through everything that has led me to this achievement.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........9 LIST OF FIGURES.......................................................................................................................10 ABSTRACT...................................................................................................................................11 CHAP TER 1 INTRODUCTION..................................................................................................................13 1.1 Introduction ................................................................................................................... 13 1.1.1 Coral Reef Crisis ............................................................................................... 13 1.1.2 Coral Disease Occurrence in the W ider Caribbean........................................... 14 1.1.3 Potential Role of Cora l-Associated Bacteria..................................................... 16 1.1.3.1 Bacterial associations of other m arine invertebrates.......................... 19 1.1.4 Quorum Sensing and Quorum Sensing Disruption........................................... 21 1.1.4.1 QS in m ost gram-negative bacteria..................................................... 22 1.1.4.2 QS in gram -positive bacteria.............................................................. 23 1.1.4.3 The hybrid QS system in Vibrio harveyi ............................................24 1.1.4.4 QS disruption ......................................................................................24 1.1.5 QS and Virulence ..............................................................................................25 1.2 Hypothesis Tested and the Objectives of This Study....................................................28 2 MATERIALS AND METHODS........................................................................................... 31 2.1 Strains and Plasm ids Used in this Study....................................................................... 31 2.2 Media ............................................................................................................................32 2.3 Culture Conditions ........................................................................................................32 2.3 Quorum Sensing Reporter Assays................................................................................ 33 2.3.1 CV026 Assa y....................................................................................................33 2.3.2 Luminescence Based Assays.............................................................................33 2.4 Organic Solvent Extractions ......................................................................................... 34 2.5 Biofilm s.........................................................................................................................34 2.6 Thin Layer Chrom atography......................................................................................... 34 2.7 PCR for 16S rDNA ....................................................................................................... 34 2.8 Gel Purification ............................................................................................................. 35 2.9 Subcloning in to TOPO TA and Transfor mation in DH5 ...........................................35 2.10 Surface Spreading Experim ents.................................................................................... 36 2.11 Drop Collapse ...............................................................................................................36 2.12 Conjugation ...................................................................................................................37

PAGE 6

6 3 THE SCREENING OF CORAL BACTERIA L I SOLATES FOR QUORUM SENSING ANTAGONIST COMPOUNDS AND IDENTI FICATION OF CORAL BACTERIAL ISOLATES BY 16S RDNA SEQUENCE HOMOLOGY..................................................... 38 3.1 Introduction ................................................................................................................... 38 3.1.1 Assessing Q S with Chromobacterium violaceum Assays................................39 3.1.2 Luminescence-Based QS Reporters.................................................................. 39 3.1.3 Detecting QS-active Compounds with Thin Layer Chrom atography............... 40 3.1.4 Identification of Environm ental B acterial Isolates Using 16S rDNA Sequences..........................................................................................................40 3.1.5 Hypothesis Tested in This Experim ent............................................................. 41 3.2 Materials and Methods ..................................................................................................42 3.2.1 CV026 Screening of a Library of Coral-Associated Bacteria ........................... 42 3.2.2 PCR Amplification of 16S rDNA Seq uences of Coral Bacterial Isolates........42 3.2.3 Luminescent QS Reporter Assays..................................................................... 43 3.2.4 Cross-streak Assays with a Luminescent QS Reporter and the Coral Bacterial Isolates............................................................................................... 43 3.2.5 Liquid Culture Extraction of Coral Bacterial Isolates....................................... 43 3.2.6 Thin Layer Chrom atography (TLC).................................................................. 44 3.2.7 QS Reporter Overlay of TLC Plates to Detect QS Active Compounds from the Coral Bacterial Isolates ............................................................................... 44 3.3 Results ........................................................................................................................ ...45 3.3.1 Screening of Coral-Associated Bact eria for QS Disruption with CV026 and the Identification of Six Coral Bacterial Isolates Differentiated by 16S rDNA Sequences...............................................................................................45 3.3.2 Bioassays w ith Luminescent QS Reporters to Test the Coral Bacterial Isolates Range of QS Disruption..................................................................... 46 3.3.3 Coral Bacte rial Isolates Were Te sted for Extractable QS-active Compounds with TLC....................................................................................... 47 3.4 Discussion .....................................................................................................................48 3.4.1 Screening and Identification of Coral Bacterial Isolates ................................... 48 3.4.2 Coral Bacterial Isolates Can Interact with Different QS System s.................... 49 3.4.3 Investigatio n into the Presence of QS Activities in Organic Extracts of Coral Bacterial Isolates with TLC..................................................................... 50 4 CORAL ISOLATES DISRUPT NORMAL SURFACE SPREADING PHENOTYPES OF Serratia marcescens STRAINS ........................................................................................ 55 4.1 Introduction ................................................................................................................... 55 4.1.1 Surface Spreading S marcescens .......................................................................55 4.1.2 S. marcescens as a Tool for QS Screening ........................................................ 55 4.1.3 Hypothesis and Experi m ental Importance........................................................ 56 4.2 Materials and Methods ..................................................................................................57 4.3 Results ........................................................................................................................ ...57 4.3.1 Effects of the Coral Bacter ial Isolates on Swar ming of S. marcescens MG1...57 4.3.2 Effects of the Coral Bacterial Iso lates on the Swarming of S. marcescens MG44 when Complemented with AHL............................................................ 58

PAGE 7

7 4.3.3 Effect of the Coral Bacteria l Isolates on the Swarm ing of S. marcescens PL10 When Complemented by Surfactant........................................................ 59 4.4 Discussion .....................................................................................................................60 4.4.1 Planococcus spp. Effect on Surface Motility of S marcescens Strains............60 4.4.2 Photobacterium spp and Marinobacter salsugini s Have Sim ilar Effects on the Swarming of S. marcescens Strains............................................................ 60 4.4.4 Vibrio spp. Effect on Surface Motility of S. marc escens Strains...................... 61 4.4.5 Caryophanon spp. Effect on Surface Motility of S marcescens Strains.......... 61 5 PRELIMINARY STUDIES OF THE QS SYSTEM PRESENT IN THE CORAL PATHOGE N Serratia marcescens PDL100 AND POTENTIAL REGULATION OF MULTICELLULAR PHENOTYPES BY QS IN PDL100.................................................... 64 5.1 Introduction ................................................................................................................... 64 5.1.1 White Pox Disease of Acropora Corals in the Caribbean ................................. 64 5.1.2 Quenching Quorum Sensing............................................................................. 64 5.1.3 Conjugation ....................................................................................................... 65 5.1.4 Behaviors R egulated by QS in Other S. marcescens Strains Can Be Investigated in PDL00.......................................................................................66 5.1.4.1 Surface spreading................................................................................ 66 5.1.4.2 Biosurfactant production .................................................................... 66 5.1.4.3 Biofilm formation............................................................................... 67 5.1.5 Hypothesis .........................................................................................................67 5.2 Materials and Methods ..................................................................................................68 5.2.1 Conjugation ....................................................................................................... 68 5.2.2 16S rDNA Sequence Amplification by PCR .................................................... 68 5.2.3 PCR to Amplif y the aiiA Gene.........................................................................68 5.2.4 Digestion of 16S rDNA PCR Products with Hind III........................................ 69 5.2.5 Organic Extraction ............................................................................................ 69 5.2.6 Thin Layer Chrom atography............................................................................. 70 5.2.7 Drop Collapse Test ............................................................................................70 5.2.8 Swar ming with PDL100 WT, pDSK519, and pDSK-aiiA............................... 70 5.2.9 Biofilm Assay................................................................................................... 71 5.3 Results ........................................................................................................................ ...71 5.3.1 Introduction of the aiiA Gene into P DL100 using Conjugation....................... 71 5.3.2 Detection of QS Agonist Com pounds Present in Crude PDL100 Culture Extracts Using TLC..........................................................................................72 5.3.3 Effects on Surface Spreading of PD L100 after the Introductio n of the aiiA Gene 73 5.3.4 Effect of Coral Isolates o n Su rface Spreading of PDL100 WT, pDSK519, and pDSK-aiiA Strains......................................................................................74 5.3.5 Effect of Introduction of aiiA on Surfactant Production in PDL100 ................ 75 5.3.6 Effect of Introducing ai iA into PDL100 on Biofilm Formation....................... 76 5.4 Discussion .....................................................................................................................76 5.4.1 Conjugation and Confirm ation of th e Presence and Activity of the aiiA Gene in PDL100 pDSK-aiiA............................................................................76

PAGE 8

8 5.4.2 Effects of aiiA Introduction in to PDL100 on Surface Spreading and Surfactant P roduction........................................................................................ 78 5.4.3 Effect of Introducing ai iA into PDL100 on Biofilm Formation....................... 79 6 CONCLUSIONS AND FUTURE WORK ............................................................................. 86 6.1 Potential Im portance of Coral Bacterial Isolates.......................................................... 86 6.2 Identification of Coral Bacterial Is olates that Exhibit QS Antagonism ........................ 87 6.3 Potential Ro le for Coral Bacterial Isolates in the Defense of Corals from Pathogens......................................................................................................................88 6.4 Prelim inary Results Suggest a Limited Role of QS in Regulating VirulenceRelated Multi-Cellular Behaviors in S. marcescens PDL100....................................... 88 6.5 General Conclusions ..................................................................................................... 90 LIST OF REFERENCES...............................................................................................................91 BIOGRAPHICAL SKETCH.........................................................................................................99

PAGE 9

9 LIST OF TABLES Table page 2-1 Bacterial strains and plasmids used in this study............................................................... 31

PAGE 10

10 LIST OF FIGURES Figure page 1-1 Generalized diagrams of model QS system s present in different bacterial types. ........... 30 3-1 The CV026 indirect assay sc reen of the coral bacterial isolates for QS antagonism ........51 3-2 Indirect co-culture bioa ssay of three QS reporters a nd coral bacterial isolates. ................ 52 3-3 Cross streak assay with luxR QS reporter and coral bacterial isolates. ............................. 53 3-4 TLC of coral bacterial is olate organic extracts overlaid with the QS reporters CV026 and Agrobacterium tumefaciens NTL4. .............................................................................54 4-1 Coral bacterial isolates alter swarm ing phenotypes of Serratia marcescens strains......... 63 5-1 PCR amplification of the aiiA gene in PDL100................................................................. 81 5-2 16S rDNA PCR and subsequent Hind III d igest of PDL100 QS quenched strains and E. coli strain controls.........................................................................................................81 5-3 TLC of PDL100 WT, pDSK519, and pDSK-aiiA extracts. .............................................. 82 5-4 Spreading behavior of S erratia marcescens PDL100 WT, pDSK519, and pDSK-aiiA strains.................................................................................................................................83 5-5 Surface spreading of PDL100, pDSK519, and pDSK-aiiA when incubated in proximity to coral bacterial isolates................................................................................... 83 5-6 Drop collapse test of Serratia marcescens strains. ............................................................ 84 5-7 Biofilm formation of PDL100 in glass t ubes is affected by the introduction of the aiiA gene. ...........................................................................................................................85

PAGE 11

11 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science INTERACTIONS OF CORAL ASSOCIATED BACTERIA WITH QUORUM SENSING AND THE POTENTIAL CONTRIBU TION TO DEFENSE OF TH E HOLOBIONT AGAINST PATHOGENS By Stephanie Marie Halbig December 2008 Chair: Kevin Folta Major: Plant Molecular and Cellular Biology Bacteria have been observed in and on coral tissues since the early 20th century. However, it is only recently that investigation in to the po tential role of associated bacteria in coral polyp health has begun to take place. One putative respons ibility of coral associated bacteria may be as a host defense mechanisms against pathogen infection. Quorum sensing (QS) is the processes by which bacterial colonies regulate multicellular behavior with many pathogens using QS to control virulence factor expression. In this study, coral associ ated bacteria were identified that were capable of inhibiting QS in model reporter systems. Six cora l isolates were identified using 16S rDNA sequence analysis and were found to be Planococcus spp ., Photobacterium spp ., Marinobacter salsuginis Agrobacterium stellulatum Vibrio spp., and Caryophanon spp A QS agonist from Caryophanon spp. was partially purified from whol e culture extracts using HPLC and analyzed with mass spectrometry. The abilit y of the six coral isol ates to inhibit QS controlled surface spreading of the ubiquitous pathogen Serratia marcescens was investigated and supported the isolates interference with QS. In conjunction, the inherent QS system of the pathogen responsible for the White Pox disease of coral, S. marcescens PDL100, was also investigated. Two compounds present in organic extracts of PDL100 were found to stimulate a

PAGE 12

12 broad range QS reporter and were susceptible to inactivation by the lactonase enzyme AiiA. Multicellular behaviors of PDL 100, including surface spreading and biofilm formation were evaluated in wild type and AiiA expressing strains of PDL100 as well. Overall, the data presented here confirms that coral associated bacteria are capable of inhibiting QS in model systems and potentially contribute to protec tion of the coral holobiont from pathogens.

PAGE 13

13 CHAPTER 1 INTRODUCTION 1.1 Introduction 1.1.1 Coral Reef Crisis Coral re efs are the largest natural structur es produced by living organisms and provide economic value to the surrounding regions thoug h fishing, tourism, and coastal protection activities (Rosenberg et al. 2007b). These magnif icent structures are built by a group of corals referred to as scleractinian, or stony, corals. Scleractinian corals se crete a calcium carbonate skeleton that provides habitats for a variety of marine organisms, even after the coral colony dies. One genus of scleractinian corals is Acropora and two species of Acropora found in the Florida reefs are Acropora palmata (Elkhorn coral) and Acropora cervicornis (Staghorn coral). These two species are listed as endangered. A massive decline of the populations of A. palmata and A. cervicornis in the Atlantic ocean has occurred since 1980 (Miller et al. 2002b). While capable of sexual reproduction, stony corals ofte n rely heavily on asexua l reproduction (Miller et al. 2007), and few of the gametes produced by sexual reproduction settle to form new colonies (Quinn and Kojis 2005). While this mechanism allows for rapid recovery from physical disturbances such as storms, it allows for maintenance and spread of the species only if there are many healthy large stands of the coral availa ble for repopulation (Quinn and Kojis 2005, Miller et al. 2007). Compounding the low success of sexual reproducti on is the increasing occurrence of coral bleaching which causes the coral polyps to expel sy mbiotic algae. Bleaching greatly reduces the sexual reproductive capacities of corals (Rosenberg et al. 2007b). Bleac hing events coincide with the hottest months of the year and have been on the rise in frequency, intensity, and

PAGE 14

14 geographic distribution over the la st two decades (Sutherland and Ritchie 2004, Rosenberg et al. 2007b). All of these factors have lead to a decrease in Acropora coverage in the Florida Keys. Miller et. al. (2002b) analyzed the coverage of A. palmata and A. cervicornis at Looe Key reef and found that between 1983 and 2000 the reef had lost 93% of A. palmata coverage and 98% of A. cervicornis coverage. While some of this loss ma y have been due to hurricanes and boat groundings, the surveys taken before and after th ese events showed a preponderance of dead colonies predating those distur bances (Miller et al. 2002b). The majority of the loss of Acroporid corals must therefore be an effect of other factors. 1.1.2 Coral Disease Occurrence in the Wider Caribbean Microbes have been observed w ithin coral tis sue as far back as 1902 however, diseases have only been reported within th e last thirty years (Green an d Bruckner 2000, Rosenberg et al. 2007b). W hile the absence of disease reports may be due to the lack of access to SCUBA before the 1970s, it is still noteworthy that over the last three decades there has been a 30% decline in the worlds coral population as a result of disease (Rosenberg et al. 2007b). A detailed analysis of disease occurrence reports in the literature on coral reefs throughout the world by Green and Bruckner (2000) revealed that de spite the fact that coral reef s in the wider Caribbean only represent 8% of the worlds reefs, this region accounted for 66% of the overall reported disease occurrences (Spalding and Gr enfell 1997, Green and Bruckner 2000). Other disturbing information presented in this literature review included the fact that the greatest number of Black Band Disease (BBD) observations occurred on Flor ida reefs, as well as the third highest occurrence of White Band Diseas e (WBD) (Green and Bruckner 2000). Another emerging coral disease, Yellow-blotch Disease (YBD, formerly Ye llow-band Disease), was first identified in the Florida Keys (Reeves 1994, Green and Bruckner 2000).

PAGE 15

15 The increased density of disease within the Fl orida reef tract may be due to anthropogenic impacts (Richardson 1998). Over half the reefs in the region are at medium to high risk of anthropogenic interactions, such as snorkeling, coastal development, and pollution (Bryant et al. 1998, Green and Bruckner 2000). A strong correlati on exists between di sease occurrence and anthropogenic interactions, such that 97% of disease reports in the wider Caribbean were on reefs with medium to high levels of anthropoge nic interactions (Green and Bruckner 2000). Furthermore, new marine diseases are emerging fa ster than before and while this may be an artifact of the increase in reporting, it may also be a result of host range increases of terrestrial pathogens. The host range of pathogens can be influenced by alterations in the marine environment, such as increased temperature and human activities (Harvell et al. 1999). The White Pox disease of Acropora palmata originally documented in the Florida Keys in 1996, has been responsible for Acropora palmata losses on Florida reefs (Patterson et al. 2002, Sutherland and Ritchie 2004). Serratia marcescens PDL100 was tested, and by satisfying Kochs postulates, it met the requirement to be the causative agent of the White Pox disease (Patterson et al. 2002, Sutherland and Ritchie 200 4). PDL100 was the only White Pox lesionisolated bacterium that induced disease symptoms on healthy A. palmata fragments (Patterson et al. 2002). The sample size for th e re-infection experiments of A. palmata with PDL100 was small but does support the hypothesis that S. marcescens PDL100 causes White Pox disease (Patterson et al. 2002, Lesser et al. 2007). The reduction in Acropora palmata due to White Pox caused an average loss of gr eater than 70% of living A. palmata cover per year at sites studied over a four-year period (Patterson et al. 2002). Tissue loss due to the disease occurred at an alarming rate of 2.5cm2/ day (Patterson et al. 2002).

PAGE 16

16 The identification of S. marcescens, a ubiquitous pathogen of plants and animals, similar as the causative agent of White Pox Disease was the first documentation of a microbe associated with the human gut to also be associated with a coral disease (Patterson et al. 2002). There is evidence that human fecal microflora is concentrat ed on and in the mucus layers of corals in the Florida Keys (Patterson et al. 2002). How corals deal with disease, being morphologically simple and sessile organisms, has become an interesting and necessary area of research. 1.1.3 Potential Role of Coral-Associated Bacteria Corals possess an innate, or natural, immune syst em that is composed of physical barriers such as mucus, phagocyte cells that lyse invading microorganisms, and antimicrobial agents (Rosenberg et al. 2007b). Bacteria reside within the coral mucu s that may produce antimicrobial compounds that inhibit pathogen invasion (Ritch ie 2006, Rosenberg et al. 2007b). The bacterial ecological niches within a stony coral colony are the mucus layer, coral tissue, and the calcium carbonate skeleton (Rosenberg et al. 2007b). Each of these nich es harbors a unique microbial community in which specific bacterial species ha ve been enriched up to 1000-fold higher than the surrounding sea water (Rosenberg et al. 2007b). These bacterial communities are extremely diverse, yet often distinct between coral spec ies. A consortium of species allows for the potential of multiple bacterial activities that are beneficial to the coral (Klaus et al. 2007). Nitrogen cycling bacteria, or isol ates closely related to known nitr ogen fixing bacteria, have been identified in association with multiple coral species (Rohwer et al. 2002, Rosenberg et al. 2007b, Wegley et al. 2007). Available nitrogen is impor tant to a healthy cora l polyp to sustain the populations of unicellular algae resident within the polyp. Ni trogen needs to be brought into the holobiont either by feeding of the polyp or fixatio n by associated bacteria (Wegley et al. 2007). Coral mucus is an ideal environment for nitrogen fixation because it is high in available carbon

PAGE 17

17 and low in oxygen concentration (Rohwer et al. 2002). However, the vast diversity of the bacterial species residing on or in corals suggest s multiple functions of the bacterial consortium. An emerging concept regarding the role of bact eria in coral health is that coral may be protected from potential pathogens by its native microflora. This concept is based on the observations that coral microbial populations are different in stre ssed and diseased corals as compared to those in healthy corals (Ritchie and Smith 2004, Gil-Agudelo et al. 2006). Normal microflora of a coral may provide protection from pathogen infection through basic competition for space and nutrients, but may also provide secret ed antimicrobials (Klaus et al. 2007). While resident populations of bacteria are difficult to establish, since horizontal acquisition allows for constant changes, general trends in population dynamics have been assessed (Rohwer et al. 2002, Klaus et al. 2007). Coral species that are ge ographically close to each other do not possess similar microbial communities but corals of the same species that are geographically distant do (Rohwer et al. 2002). These data support the con cept that the microbial communities associated with corals are specific and ar e therefore selected. Ritchie (2 006) found that coral mucus was itself a selection agent because the number of bacteria isolated from surrounding sea water capable of growing on mucus was significantly lower than the number of bacteria from the same water sample capable of growing on artificial se awater media (Ritchie 2006). In addition, in an analysis of bacterial diversity from three diffe rent coral genus, Rohwer et. al. (2002) discovered that, in general, coral-associated microbes we re predominantly geneti cally dissimilar to known bacterial species; while the bacterial isolates from random seawat er and reef seawater samples contained bacteria that were chiefly genetically similar to know n species (Rohwer et al. 2002). This indicates that specific, and oftentimes rare microbes are enriched w ithin the corals due to

PAGE 18

18 the small percentage of unidentified bacteria from surrounding waters and the large percentage of unidentified species within the coral. Phylogenetic analyses of the bacteria from the Rohwer et al (2002) study indicated that many of the coral-associated bacteria were related to nitrogen fixing species as well as antimicrobial-producing bacteria (Rohwer et al. 2002). Koh (1997) extracted one hundred different coral species to asse s antimicrobial activiti es of stony corals (Koh 1997). The report found that the majority of the corals tested co ntained antimicrobial activ ity and that the coral species with the highest levels of anti-microbial activity had the lowest number of culturable bacteria associated with the sample (Koh 1997). However, the samples used in this study were mixtures of coral tissue, skeletal material, and coral mucus; theref ore, the precise origin of these activities could not be confirmed (Koh 1997). Current data on coral/bacteria l associations was used to form the coral Probiotic Hypothesis (Reshef et al. 2006). This hypothesis suggests that corals can combat disease and stress by shifting the proportions of species with in their associated microbial populations (Reshef et al. 2006). While this hypothe sis has not been tested dire ctly, a growing body of evidence supports this possibility. Re shef and colleagues (2006) invok ed the example of pathogeninduced bleaching of the coral Oculina patagonica as an example of potential adaptive coral immunity. O. patagonica bleaching was shown to be caused by the pathogen Vibrio shiloi in studies conducted form 1995 to 2002. However, after 2003 the strains of V. shiloi that had been maintained in storage could no longe r cause bleaching of fresh wild O. patagonica samples (Reshef et al. 2006). Another example of coral deve loped immunity is that of Aurantimonas coralicida a bacterium shown to induce White Plague on corals in the mid 1990s. This bacterium can no

PAGE 19

19 longer infect healthy corals of the Florida Keys where the initial outbreak of the disease occurred (Reshef et al. 2006). The reduction in reinstatement of the disease indicates that either the corals became immune to the disease, or that the pathogen became avirulent, the former being consistent with the Coral Probiotic Hypothesis. The hypothesis is also supported by findings that corals exposed to high intensity light prior to elevated surface temperatures are less likely to bleach, implying an environmentally-induced shif t in the microfloral makeup of the holobiont (Green and Bruckner 2000, Rosenberg and Ben-Ha im 2002, Rosenberg et al. 2007b). However, this observation may also be due to the fact that intense UV light kills off pathogens and stimulates photosynthesis in th e resident unicellular algae, th ereby increasing mucus production 1.1.3.1 Bacterial associations of other marine invertebrates There are a num ber of examples of bacterial associations with marine invertebrates. The colonization of the light organ of the squid Euprymna scolopes (Hawaiian bobtail squid) by luminescent Vibrio fischeri has been studied in detail. E. scolopes acquires its bacterial symbiont horizontally from the surrounding waters shor tly after hatching (Nyholm and McFall-Ngai 2004). This is similar to the hypothe sis that corals recruit bacteria for association from the water column (Ritchie 2006). Newly hatched squi d secrete mucus that selectively traps V. fischeri and the squid are colonized within hours of emergence (Nyholm and McFall-Ngai 2004). While the exact mechanism of selection is unknown, it has been shown that bacteria other than V. fischeri will not effectively colonize the light organ and that even V. fischeri mutants that do not luminesce are excluded from the light organ (Nyh olm and McFall-Ngai 2004). Work by Ritchie demonstrated that coral mucus is a selective agen t for bacteria from seawater samples (Ritchie 2006). Finally, a relationship betw een any organism and bacterium must be carefully maintained so that the bacteria do not overg row the organism, and it also must allow release of beneficial

PAGE 20

20 bacteria back into the environment to continue the relationships in ot her generations (if the bacteria is acquired horizontally). In the squid, every day approximately 95% of the V. fischeri population within the light organ are expelled while the squid is in the s ubstratum (Nyholm and McFall-Ngai 2004). This release keeps the bacterial numbers within the squid in check and also provides symbionts for any newly hatched offspring (Nyholm and McFallNgai 2004). Corals with mucus layers intermittently slough this coating along with the a ssociated microbial community. This may be a way for corals to maintain bacterial numbers or change populations of bacteria within the holobiont. The diverse groups of bacteria resident in sponges have also been heavily examined. Sponges use bacteria as a food source but also have bacteria that re side exclusively in extracellular spaces or within th e sponge cell nucleus (Hentschel et al. 2001). Some of the bacteria found associated with sponges are unique in that they have yet to be isolated from water samples, although this may be due to the magnitude of dilution in the ocean environment (Taylor et al. 2007). Distantly relate d sponges from distinct geographic locations often carry similar consortia of bacteria (Taylor et al. 2007). The large volume of sponge-specific bacterial DNA sequences, approximately 1500 (Taylor et al. 2007) aids in the inves tigation of community structure of sponge-associated microbes. This ty pe of resource is yet not available for similar studies in corals. Also, unlike coral/microbe associations (Ritchie and Smith 2004), sponge associations with commensal bacter ia are stable and resistant to external disturbances such as temperature, depth, and location (Taylor et al. 200 7). This stability may be due in part to the largely vertical transmission of sponge comm ensal bacteria through both maternal and/or paternal gametes (Taylor et al. 2007).

PAGE 21

21 There are some reports of shifts in the bacterial community struct ure in sponges over long periods of time. Anand et. al. (2006) f ound that when evaluated in 2006, there were approximately equal numbers of gram-negative and gram-positive bacteria in four sponge species tested and that the same sponges had predominantly hosted gram-negative bacteria when evaluated in 1997 (Anand et al. 2006). Sponges are the most prolific producers of novel natural compounds within the marine environment, with approximately two hundred new compounds reported each year (Taylor et al. 2007). This has made sponges, and their associated bacteria, a popular subj ect of investigations for new antimicrobial compounds. In the sponges of the genus Aplysina microbes can represent 40% of the total biomass of a given sponge (Hents chel et al. 2001). This large proportion of microbes is capable of producing brominated compounds, which are stro ng antimicrobials, and the brominated compounds can account for up to 13% of the total dry-weight of a sponge sample (Hentschel et al. 2001). Brominated compounds ha ve also been identifi ed in the red seaweed Delisea pulchra D. pulchra produces brominated furanones that inhibit bacter ial cell-to-cell communication (Manefield et al. 1999). Othe r screens of sponge-associated microbes for antimicrobial compounds have shown that antimic robial compound production is a characteristic conserved among many sponge spec ies and cognate microbial populations (Anand et al. 2006). 1.1.4 Quorum Sensing and Quorum Sensing Disruption Bacteria often m onitor their population de nsity through a cell-t o-cell communication system termed quorum sensing (QS) (Parsek and Greenberg 2000). QS allows for the coordination of gene expression that is necessary for continued success within an ecological niche, such as inside of a particular host (P arsek and Greenberg 2000). QS systems regulate a series of physiological and developmental res ponses such as motility, sporulation, biofilm formation, conjugation and competence (Miller and Ba ssler 2001). Specific to this work QS is

PAGE 22

22 important for antibiotic production and symbiosis. QS is critical to regulating virulence behaviors, since often only larg e numbers of bacteria can effec tively cause infection (Parsek and Greenberg 2000). While there are many variations, the basic mechanism of QS is composed of a signal molecule and a response regulator. Ther e are single-component QS regulatory systems, two-component QS regulatory systems, and hybrids of both singleand two-component systems. The complexity of QS networks, once thought to be straightforward, is only beginning to be understood. 1.1.4.1 QS in most gram-negative bacteria In m ost gram-negative bacteria, the QS signal molecule is an N -acyl-homoserine lactone (AHL) and is usually produced by a member of the LuxI autoinducer (AHL) synthase family (Taga and Bassler 2003). The receptor proteins in gram-negative bacteria are of the LuxR receptor protein family and also act as respons e regulators (Taga and Bassler 2003). LuxI and LuxR regulate luminescence in V. fischeri and were the first QS system to be characterized (Engebrecht et al. 1983, Fuqua et al 1994). AHL signaling is highly specific due to the nature of the signal molecule synthase. The synthase conn ects the acyl side chain to the lactone ring and will only recognize the specific acyl moiety required for the particular signal molecule (Taga and Bassler 2003). The signal and receptor are produced at basal levels at a ll times. AHLs with short acyl moieties are diffusible and move freel y through the membrane of bacterial cells while AHLs that have longer side-chains may be exported from the cell with the help of a transporter (Waters and Bassler 2005). At a certain threshol d of signal concentration (a point known as a quorum ), the receptor, located in the cytoplasm, bi nds the AHL and in most cases auto-regulates the production of the signal and the receptor, thus creating a positive feedback loop for rapid signal amplification. A general diagram of this process is illustrated in Fi gure 1-1A. Once the receptor protein binds the signal molecule it becomes competent for correct folding and possible

PAGE 23

23 dimerization (Taga and Bassler 2003, Waters an d Bassler 2005). The receptor TraR, from Agrobacterium tumefaciens was the first receptor shown to require the binding of the cognate AHL for proper protein folding (Miller and Bassl er 2001). Once the receptor is activated by binding the AHL signal molecule, it can regula te target genes involved in multi-cellular behaviors (Taga and Bassler 2003, Waters and Bassler 2005). 1.1.4.2 QS in gram-positive bacteria Gram -positive bacteria use a system that is similar in concept to gram-negative bacteria, yet contain different mechanistic components. Instead of sm all molecules like AHLs, grampositive bacteria utilize small peptides, called Autoinducing Peptides (AIP) as QS signals (summarized in Taga and Bassler 2003). These peptides are cleaved off of larger precursor peptides and modified before being exported ou t of the cell (Miller and Bassler 2001, Taga and Bassler 2003, Waters and Bassler 2005). Gr am-positive bacteria produce membrane-bound twocomponent sensor kinases with external signal recognition sites instead of cytoplasmic singlecomponent receptors as in AHL QS systems (Miller and Bassler 2001, Taga and Bassler 2003, Waters and Bassler 2005). The two-component sensor kinase binds the AIP signal in the external domain of the sensor kinase, a nd then activates the response regulator by phosphotransfer (Miller and Bassler 2001, Taga and Bassler 2003, Waters and Bassler 2005). The response regulator then di rectly participates in modul ation of gene expression. A generalized gram-positive QS mechanism is diagramed in Figure 1-1B. AIPs in gram-positive bacteria are highly specific, due to the fact that they are encoded in the genome, and therefore the signal can only be altered through mutation (T aga and Bassler 2003). QS, along with nutrient status, regulates the lifestyle switch in gram-positive bacter ia from a competent (able to participate in genetic transfer) lif estyle to the spore life strate gy. A benefit of depositing large numbers of spores in a niche is that when nutrient conditions improve, there is a large enough

PAGE 24

24 population to ensure success within the niche (Miller and Bassler 2001 ). An interesting aspect of gram-positive QS systems is that each strain/group of a species, such as Staphylococcus aureus, has a specific AIP that will activ ate its cognate QS system but w ill inhibit the related QS systems in all other strains (M iller and Bassler 2001). 1.1.4.3 The hybrid QS system in Vibrio harveyi Vibrio harveyi is a gram -negative bacteria that is closely related to V. fischeri however, the QS system of V. harveyi is a hybrid system between an AHL QS system and an AIP system (Waters and Bassler 2005). The signal molecules in V. harveyi are AHLs but the receptors are two-component sensor kinases (Waters and Bassler 2005). There are two QS circuits present in V. harveyi that channel information into one regulato ry pathway (fig 1-1C) (Waters and Bassler 2005). Current evidence suggests that a single AHL synthase produces all of the autoinducers present in V. harveyi (Waters and Bassler 2005). Th e AHL synthase identified in V. harveyi that produces 3-hydroxybutanoyl-AHL is also not a member of the LuxI family, but represents a novel type of AHL synthase (Miller and Bassler 2001). In contrast to V. fischeri, the response regulator controlling luminescence in V. harveyi represses expression of the luxCDABE operon at low cell densities, and the repression is removed at high cell density through dephosphorylation (Miller and Bassler 2001). The id entification of the hybrid QS system in V. harveyi represents a novel QS stra tegy; however, homologs of th e genes encoding components of the QS system in V. harveyi have been identified in V. fischeri although the function of these genes in V. fischeri is currently unknown (W aters and Bassler 2005). 1.1.4.4 QS disruption The disruption of QS by other organism s is a common theme in ecological interactions. There are many ways that the AHL systems of gram -negative bacteria can be inhibited, including destruction of the signal molecule through enzymatic activity and compounds that create

PAGE 25

25 competition for binding the receptor (Dong et al 2000, Taga and Bassler 2003, Frezza et al. 2006, Geske et al. 2007a). In fact, the commonality in diverse bacterial species of strategies for inhibition of AHL-mediated QS suggests that this mechanism is a conserved means for bacteria to compete within ecological ni ches (Taga and Bassler 2003). The production of enzymes, such as lactonases and acylases, represent evolved m echanisms that allow bacteria to suppress the communications networks of potential competito rs present in the environment (Leadbetter 2001, Taga and Bassler 2003). The enzyme AiiA from the gram positive bacterium Bacillus sp. 240B1, was one of the first AHL lactonases discovered (Dong et al. 2000). This enzyme hydrolyzes the lactone bond within the lactone ring and renders the AHL inactive, limiting signaling potential (Dong et al. 2000). Other enzymes, such as AiiD from Ralstonia spp., hydrolyze the AHL at the side-chain linkage thus releasing the acy l moiety from the ring (Taga and Bassler 2003). Competitive binding of AHL receptors is inferred through modeling of receptor/ligand interactions and through titration experiments usi ng synthetic ligands (Frezza et al. 2006, Geske et al. 2007b). Structure-function i nvestigations have shed light onto important amino acid residues within the receptor that form hydrogen bonds with the ligand that are responsible for receptor protein folding and act ivation (Reverchon et al. 2002, Smith et al. 2003, Frezza et al. 2006, Muh et al. 2006). While QS inhibition is widespread among bacteria, the mechanisms by which it takes place are still being elucidated. 1.1.5 QS and Virulence QS allows m embers of the same bacterial community to act toge ther to coordinate behaviors including, but not limited to, surface colonization, luminescence, the production/excretion of virulence factors, and establishing/maintaining biofilms (Waters and Bassler 2005). It has long been su spected that QS contributes to the regulation of virulenceor symbiotic-type interactions of bacteria with potential hosts (Dong et al. 2000, Parsek and

PAGE 26

26 Greenberg 2000, Dong et al. 2001, Miller and Bassl er 2001, Miller et al. 2002a, Hentzer et al. 2003, Bjarnsholt and Givskov 2007, Cegels ki et al. 2008, Liu et al. 2008 ). Enzymatic activities, such as nucleases, proteases, and chitinases, are key factors of pathogenesis and have been investigated for QS regulation, along with surface spreading and biofilm formation (Chernin et al. 1998, Howard et al. 2003, Van Houdt et al. 2007a, Liu et al. 2008). Investigations into the connection between QS and virulence span a range of hosts from plants to humans, since bacterial pathogens cause substantial monetary losses of agricultural products and the need for new antimicrobial th erapies for human diseases. For example, Burkholderia plantarii is a plant pathogen that causes seedli ng blight in rice (S olis et al. 2006). An AHL synthase mutant of B. plantarii showed a delay of seed ling blight symptoms as compared to wild typefrom seven days for the w ild type strain to twenty-eight days for the AHL synthase mutant (Solis et al. 2006). However, the genes regulated by the QS system were not identified in B. plantarii (Solis et al. 2006). In a related species, B. glumae, the production of the primary toxin, toxoflavin, is regulated by QS (Kim et al. 2004). Another plant pathogen, Pectobacterium atrosepticum also regulates virulence mechanis ms through QS (Liu et al. 2008). P. atrosepticum causes soft rot symptoms in a number of plants and has both blunt force and stealth methods of infection (Liu et al. 2008). The blunt force methods includes the production and export of plant cell wall-degrading enzymes while the stealth method consists of the production of compounds that antagonize plant defenses (Liu et al. 2008). In a microarray assessment of wild type and AHL synthase muta nts after infection of potato, it was found that approximately 26% of the genome was differentially regulated in an AHL synthase mutant of P. atrosepticum as compared to the wild type (Liu et al. 2008). The AHL synthase mutant also contained lower transcript levels of virulencerelated genes from both the blunt force and

PAGE 27

27 stealth methods of infection (L iu et al. 2008). However, this does not prove a direct regulation of virulence by QS in P. atrosepticum because there was no charac terization of interactions between the QS response regulator and the virulence-related genes (Liu et al. 2008). Therefore, the alteration of transcript levels of the viru lence related genes observed in the AHL synthase mutant may be the result of indirect inhibition. Vibrio cholerae is an important waterborne path ogen causing disease outbreaks in developing countries where water treatment is inadequate (Faruque et al. 1998). V. cholerae is closely related to V. harveyi and it was hypothesized by Mille r and colleagues (2002) that V. cholerae virulence may be regulated in a si milar fashion as luminescence is in V. harveyi (Miller et al. 2002a) An analysis of the V. cholerae genome revealed that all of the elements of one of the two QS systems from V. harveyi were present in V. cholerae and that all of the components of the response regulator system that is shared by the two QS systems in V. harveyi were also present in V. cholerae (Miller et al. 2002a). It was shown that a V. cholerae strain with a mutation in the response regulator, luxO, had reduced virulence in an infant mouse system (Miller et al. 2002a). Both cholera toxin produc tion and toxin-coregulated pilus production are eliminated by the mutation in luxO (Miller et al. 2002a). However, mutations in any of the three sensor kinases in V. cholera e did not significantly reduce virule nce in infant mice or alter the production of cholera toxin or to xin-coregulated pilus production (Miller et al. 2002a). This result was likely because in each of the sensor kinase mutation experiments, one QS sensor kinase and regulatory systems was left intact and was sufficient for full virulence (Miller et al. 2002a). One of the first levels of host infection is rapid colonization of suitable host surfaces. Bacteria employ behaviors such as swarming and sliding to spread across solid surfaces in a

PAGE 28

28 swift manner (Givskov et al. 1998, Fraser a nd Hughes 1999, Sharma and Anand 2002). After colonization, the formation of a complex colony structure referred to as a biofilm allows for pathogen persistence within the host (Costerton et al. 1995, Parsek and Greenberg 2000, Kjelleberg and Molin 2002, Lynch et al. 2002, Rice et al. 2005, Cegelski et al. 2008). Biofilms, defined as matrix-enclosed bacter ial populations that are adherent to each other and/or surfaces (Costerton et al. 1995), confer heightened re sistance to host immune systems and antibiotics within the bacterial community (Costerton et al. 1995). Both surface motility and biofilm formation have been shown to be coordinated by QS in Serratia marcescens and Pseudomonas aeruginosa (Costerton et al. 1995, Kjelleberg and Mo lin 2002, Labbate et al. 2004, Grimont and Grimont 2006). These two bacterial species repr esent important groups of multi-host pathogens. Pseudomonas aeruginosa is a human and plant pathogen that causes severe infections in human patients suffering from cystic fibrosis (K jelleberg and Molin 2002, Bjarnsholt and Givskov 2007). Serratia marcescens is a ubiquitous path ogen that causes disease in plants, animals, humans, and corals (Patterson et al. 2002, Bruton et al. 2003, Ku rz et al. 2003, Rice et al. 2005, Grimont and Grimont 2006). QS disruption has now become an area of interest for antimicrobial compounds, since this type of therapy would not place heavy selection pressure on pathogens, and is also a possible way to treat multi-drug-resistant bacteria (Bjarnsholt and Givskov 2007, Cegelski et al. 2008). 1.2 Hypothesis Tested and the Objectives of This Study The resea rch in this thesis focused on two ma in areas: (i) investigating the presence of QS inhibitory capabilities of bacterial isolates associated with the coral Acropora palmata and characterization of those activitie s; and (ii) identifying and charact erizing the QS system present in the coral pathogen S. marcescens PDL100 and the possible regulation by QS of virulencerelated phenotypes. The overall goal of this research was to begin to elucidate possible

PAGE 29

29 mechanisms for bio-control of PDL100 and es tablish a plausible role for the microbial consortium of A. palmata The central hypothesis of this stud y was that the bacterial isolates associated with A. palmata prevent infection of the coral by pathogens through quorum sensing disruption. To test this hypothesis, libraries of A. palmata associated bacteria were first assessed for QS inhibition. Those inhibiti on activities were probed further using QS reporter strains as well as swarming mobility test s. In addition, PDL100 was subjected to QS quenching and potentially QS-regulated phenotypes were evaluated. The results of this study indicate that the bacterial consortium associated with coral Acropora palmata may contribute quorum sensing antagonist compounds to the holobiont.

PAGE 30

30 Figure 1-1. Generalized diagrams of model QS systems present in different bacterial types. AC) Models of AHL-based QS systems, Au toinducing peptide-based (AIP-based) QS systems, and the system in V. harveyi, respectively. A B C

PAGE 31

31 CHAPTER 2 MATERIALS AND METHODS 2.1 Strains and Plasmids Used in this Study The bacterial strains and the plasm ids used in this study are summa rized in Table 2-1. When available, a reference regard ing the origin of information on th e strain or plasmid is listed. If no publicized information is available regardin g the strain or plasmid then the contact who either created or obtained the strain or plasmid is listed. Table 2-1. Bacterial strains and plasmids used in this study Strain Relevant Characteristics Reference/Source Serratia marcescens PDL100 Isolated from White Pox diseas ed corals (Pa tterson et al. 2002) Serratia marcescens MG1 Isolated from rotten cucumber, previously referred to as S. liquifacians swarms, AHL-producer (Eberl et al. 1996) Serratia marcescens MG44 AHL synthase mutant of MG1 (Eberl et al. 1996) Serratia marcescens PL10 Swarming null mutant of MG44 carrying a luxAB insertion into swrA (Lindum et al. 1998) Escherichia coli JM109 F traD36 pro A+B + lac Iq (lac Z)M15/ (lac-pro AB) gln V44 e14gy rA96 rec A 1 rel A1 end A1 thi hsd R17 (Winson et al. 1998) Escherichia coli DH5 F80lac Z M15 (lac ZYAarg F)U169 rec A1 end A1 hsd R17(rk-, mk+) phoA sup E44 thi -1 gyr A96 rel A1 Invitrogen Escherichia coli MT616 Helper strain for conjugation, pro82 rec A thi+ end A hsdR17 Teplitski, unpublished 34-D8 Planococcus spp. Isolated from mucus of Acropora palmata (Ritchie 2006) 34-E11 Photobacterium spp. Isolated from mucus of Acropora palmata (Ritchie 2006) 46-H6 Marinobacter salsugin is Isolated from zooxanthellae of A cropora palmata Ritchie, unpublished 47-G8 Agrobacterium stellulatum Isolated from zooxanthellae of Acropora palmata Ritchie, unpublished 52-B8 Vibrio spp. Isolated from mucus of Acropora palmata (Ritchie 2006) 52-E5 Caryophanon spp. Isolated from mucus of Acropora palmata (Ritchie 2006) Chromobacterium violaceum CV026 Chromobacterium violaceum AHL synthase mutant strain, responsive to C4-HSL, produces a purple pigment (McClean et al. 1997) Agrobacterium tumefaciens NLT4 AHL reporter, no AHL production, lacZ, blue pigment production (Shaw et al. 1997) Plasmids pSB401 luxR luxCDABE cassette driven by the luxI promoter, Tetr, QS Reporter (Winson et al. 1998) pSB536 ahyR luxCDABE cassette driven by the ahyI promoter, Ampr, QS Reporter (Lindsay and Ahmer 2005) pSB1075 lasR luxCDABE cassette driven by the lasI promoter, Ampr, QS Reporter (Winson et al. 1998) pDSK519 IncQ broad host range plasmid, Kanr (Gao et al. 2007) pDSK aiiA pDSK519 carrying the aiiA gene encoding the AiiA AHL lactonase from Bacillus spp 240B1 cloned into the Bam HIEco RI sites of pDSK519, aiiA expression driven by the Ptac promoter, Kanr (Gao et al. 2007) pTIM2442 luxCDABE cassette driven by the lambda phage promoter, Ampr, Control for QS Reporters Teplitski, unpublished pCR2.1 TOPO TA Cloning vector Ampr Invitrogen aAmpr, ampicillin resistance; Tetr, tetracycline resistance; Kanr, kanamycin resistance

PAGE 32

32 2.2 Media The following m edia were used in the ex ecution of the experiments that follow: GASW(A): 356mM NaCl, 8mM KCl, 40mM MgSO4, 20mM MgCl2H2O, 60 M K2HPO4, 7 M FeSO4, 33 M Tris, .05% Peptone, .2% Yeast Extract, 2% Glycerol, 1.5% Agar (if needed) AB SWARM: 17.2mM K2HPO4, 8mM NaH2PO4, 2mM KCl, .5% Casamino Acids, .4% glucose, 0.001%CaCl2, .03% MgSO4, 2.5x10-4% FeSO4, 18.7mM NH4Cl, 0.6% Agar Luria-Bertani (LB): 1% Triptone, 0.5% Yeas t Extract, 0.5% NaCl M9 Sucrose : 41mM Na2HPO4, 22mM KH2PO4, 8.55mM NaCl, 1.9mM NH4Cl, 1mm MgS04, 5.8mM Sucrose, 950nM Bioti n, 1.5% Agar (if needed) 2.3 Culture Conditions In this study all E. coli strains were cultured in/on LB media and grown at 37C unless otherwise stipulated. Serratia marcescens strains and Chromobacterium violaceum strains were cultured in/on LB media at 30C unless otherwise noted. All coral isolates were grown in/on GASW(A) media at 30C. When needed liquid cultures of Serratia marcescens and the coral isolates were buffered with 0.1M HEPES to a fi nal pH of pH 6.5. Stra ins carrying antibiotic resistance were cultured with the appropriate antibiotics unless otherw ise noted. All plate cultures had media supplemented with 1.5% agar unless otherwise indicated. Coral isolates were also cultured on gla ss fiber disks and in GASWA plugs. Disk cultures were made by pipetting 20 L of overnight coral bacteria isolate culture onto a 5mm glass fiber disks and placing the disks onto GASWA media plates. The plates were incubated until a ring of dense bacterial colony was obser ved around the disk. GASWA plug cultures were used for screening libraries (heterogenic mixes) of coral associated bacteria. Plugs of 100ul GASWA media were stab-inoculated with library isolates and grown for forty-eight hours at 30C before use in a CV026 indirect assay.

PAGE 33

33 2.3 Quorum Sensing Reporter Assays 2.3.1 CV026 Assay Chromobacterium violacein assays are a variation on McCl ean et al (1997). CV026 was cultu red overnight at 30C in 5mL LB liquid sh ake cultures. Two one-milliliter aliquots of culture were centrifuged and each one was resuspended in 50 L fresh LB. Twenty milliliters of LB with 1.5% agar was poured in to a Petri plate and allowed to solidify. Once the base was firm, the QS reporter overlay was prepared by adding the 2mL equivalent of the CV026 overnight culture to 25mL of LB with a final agar concen tration of 0.3% and a final C4-AHL concentration of 2 M. After brief vortexing, 5mL of the CV026 overlay mix was poured over the LB 1.5% agar base and allowed to set fo r 30 minutes. Glass fiber disks or GASWA plug cultures of coral bacterial isolates were set onto the overlay and plates were incubated at 30C overnight. 2.3.2 Luminescence Based Assays Escherich ia coli JM109 strains carrying ei ther pSB401, pSB107, or pSB536 (Winson et al. 1998, Lindsay and Ahmer 2005) were utilized as re al time indicators of QS interactions. All luminescent reporter strains were grown overnig ht at 37C and then diluted by removing the culture from the test tube and a dding 5mL of fresh LB with appropr iate antibiotics. This culture was incubated for forty-five minutes at 37C be fore removing the media and replacing it with fresh LB and antibiotics and incubated for an additional forty-five minutes. Reporter cultures were measured for absorbance at 600nm and d iluted to a final absorbance of 0.01. Sixty microliters of reporter culture was incubated with 40 L of overnight coral bacterial isolate culture in co-culture based assays.

PAGE 34

34 2.4 Organic Solvent Extractions For TLC experim ents, 500mL coral bacterial isolate cultures were grown in GASW media buffered with 0.5M HEPES pH 6.5 (final concentration 0.1M) at 30C for forty-eight hours. Cultures were extracted twice w ith an equal volume of ethyl acetate (Fisher Scientific), the organic phases were collected, and the solvent removed from the sample by rotary evaporation (Buchi). The remaining extract was dissolved in 400 L of ethyl acetate and stored at -20C. 2.5 Biofilms Serratia ma rcescens PDL100 (WT), S. marcescens PDL100 pDSK519 (Vector Control, VC), and S. marcescens PDL100 pDSK-aiiA (aiiA) strains we re grown for forty-eight hours in 5mL of LB supplemented with 10 g/mL of tetracycline (WT), or 10 g/mL tetracycline and 50 g/mL kanamycin (VC and aiiA) at 30C in a rotary shaker. Media was then carefully removed and 10mL of crystal violet solution wa s added incubated for fifteen minutes. Crystal violet was removed and the biofilms were washed gently with 10mL of distilled water three times. 2.6 Thin Layer Chromatography Organic ex tracts of the coral bacterial isolates or the S. marcescens strains were fractionated on TLC reverse phase C18 Silica plates (Whatman). Twenty-five microliters of each sample was spotted, 1 L at a time, 1cm from the bottom of th e 10cm square plate. Plates dried for thirty minutes and were developed in a glass chamber with ~30mL of 60% MeOH/40% H2O for forty-five minutes. The plates were then re moved from the chamber and allowed to dry in a fume hood for one hour. 2.7 PCR for 16S rDNA Colony PCR was used to am plify 16S rDNA sequ ences from bacterial strains. The two sets of PCR primers used for 16S rDNA amplif ication were: MT44 pA8F forward primer 5

PAGE 35

35 AGAGTTTGATCCTGGCTCAG 3 with MT45 pH 1542 reverse primer 5 AAGGAGGTGATCCAGCCGCA 3 (Pantos and Bythell 2006); and MT42 8F forward primer 5 AGAGTTTGATCCTGGCTCAG 3 with MT43 1489F reverse primer 5 TACCTTGTTACGACTTCA 3(Bruneel et al. 2006). The PCR amplification parameters included an initial denaturation at 94C for 5 min followed by thir ty-five cycles of a one minute denaturation at 94C, annealing for one and one-h alf minutes at 53C, an extension for three minutes at 72C, and concluded with a final extension for tem minutes at 72C. All reactions used standard Taq polymerase from New England Biolabs and the standard 10X buffer supplied with the enzyme. 2.8 Gel Purification PCR products were purified from electrophores is gels (0.9% agar) using the illustra DNA and Gel Band Purification Kit (GE H ealthcare). DNA was eluted in 50 L of DNA Grade Water (Fisher Scientific). 2.9 Subcloning in to TOPO TA and Transformation in DH5 Subcloning of the 16S rDNA PCR p roduct was performed with the Original TOPO TA Cloning Kit from Invitrogen with the pCR2.1 vect or. Ligation reactions were conducted as per the manufacturers instructions. Ligations were transformed in to chemically competent DH5 For transformations, 80 L of competent DH5 was thawed on ice for thirty minutes and then added to 5 L of ligation mix. The transformation mix incubated on ice for a further forty-five minutes before being heat shocked at 42C for th irty seconds and then returned to ice for two minutes. After the two minute incubation on ice, the transformed cells recovered in 1mL of NZY+ media for one hour at 37C in a rotary sh aker. When recovery was complete, the cells were centrifuged for one minute at 10,000 rcf and th e majority of the supernatant was removed. The cells were resuspended in the remaining me dia and plated on to LB plates containing 1.5%

PAGE 36

36 agar supplemented with 50 g/mL kanamycin and 40 g/mL x-gal. Plates were incubated overnight at 37C. White colonies were selected from the tran sformation plates and grown overnight in 5mL of liquid LB supplemented with 50 g/mL kanamycin. The plasmid DNA was extracted using the QIAprep Spin Miniprep Kit from Qiage n. Transformation and ligation success were assessed with PCR using either the M13 primers from the Invitrogen cloning kit or with the BA1111 reverse primer for the 3 end of the lacZ gene (5 ATTAATGCAGCTGGACGACAGGTT 3) coupled with the BA184-lacZ forward primer that binds the 5 end of the lacZ gene (5GAT GTGCTGCAAGGCGATTAAGTTG 3). PCR cycling conditions were the same as those outlined in 2.8. PCR Products were resolved by electrophoresis on a 0.9% agarose/TAE gel. 2.10 Surface Spreading Experiments Surface spreading was tested using an assay m odified from Eberl et. al. (1999). Disk cultures of coral bacterial isolat es were prepared by pipetting 20 L of overnight coral bacterial isolate culture onto a glass fiber disk and in cubating the disk on GASWA media overnight at 30C. Disks were transferred to 20mL plates of solidified AB SWARM media, when required plates were supplemented with C4-HSL (final concentration 2 M). Coral bacterial isolate disk cultures were grown until the colony had grown beyond the perimeter of the disk. S. marcescens strains were grown in 5mL liqui d shake cultures without antibiotics overnight at 30C and then spotted 1cm away from the isolat e disk culture. The plates were incubated at 30C with hourly monitoring of surface spreading. 2.11 Drop Collapse Drop collapse tests were a variation on the m e thod described by Lindum et. al. (1998). For this experiment, 40 L of liquid culture was pipetted on to a Petri plate lid and the lid was tilted

PAGE 37

37 to ~90. Presence of a surface tension-loweri ng compound was visualized by the collapse and run of the drop after tilting. 2.12 Conjugation Conjugation was used to m ove genetic material into the S. marcescens PDL100 strain. Five milliliter cultures of the donor, recipient, and helper strains were grown overnight in the presence of appropriate antibiotics. On day two of the experiment, 1mL of the cultures was pelleted at 10,000 rcf and washed four times with liquid LB. The aliquot was used to start fresh 5mL cultures in LB. The fresh cultures were incubated at the appropr iate temperature on the rotary shaker for two and one-half hours. Af ter two and one-half hours, the donor and helper strains were removed from the shaker and inc ubated statically for thirty minutes at the appropriate temperature. The r ecipient strain was allowed to continue incubating in the shaker for those thirty minutes. Mating mixes were pr epared with ratios of 1:1, 1:10, and 10:1 (donor: recipient), with the addition of 100 L of the helper strain. The final volumes of the three mating mixes were 500 L for the 1:1mix, and 320 L for the 1:10 and 10:1 mixes. Mating mixes were pipetted onto a .45 m cellulose nitrate membrane (Whatman) in a vacuum funnel/flask and the liquid was removed from the culture via vacuum. The filters were then moved aseptically to LB plates containing 1.5% agar and incubated at th e appropriate temperatur e overnight. On day three of the experiment, the filters were asepti cally cut into small pi eces, moved into 5mL LB with the selective antibio tics, and incubated for one hour at the recipient strain temperature. One milliliter aliquots of the transformed cells were centrifuged for one minute at 10,000rcf and the majority of the supernatant was re moved. The remaining media (~50-100 L) was used to resuspend the cells before being spread onto selective antibiotic plates and grown overnight at the correct temperature for the recipient strain. Parental strains were al so plated on selective antibiotics as a control.

PAGE 38

38 CHAPTER 3 THE SCREENING OF CORAL BACTERIA L I SOLATES FOR QUORUM SENSING ANTAGONIST COMPOUNDS AND IDENTI FICATION OF CORAL BACTERIAL ISOLATES BY 16S RDNA SEQUENCE HOMOLOGY 3.1 Introduction There are numerous mutualistic relationshi ps between eukaryotic hosts and bacteria. Examples of these relationships include plants that obtain nitrogen from b acteria, bacteria that aid in food digestion or energy acquisition in animals, and marinesponge associated bacteria that produce antimicrobial compounds. Bacteria ha ve also been observed to reside within coral tissues for over 100 years (Green and Bruckner 2000) however the biological functions of coral associated bacteria are still uncertain. Reef building (scleractinian) corals provide several different niches for bacteria in cluding coral tissues, a mucous layer, and the calcium carbonate skeleton (Rosenberg et al. 2007a, Rosenberg et al. 2007b). Investigations into potential functions of coral associated ba cteria with regards to the cora l holobiont have indicated that associated bacteria may fix nitrogen, decom pose chitin, or produce antimicrobial compounds (Rohwer et al. 2002). The production of anti-mi crobial compounds by scleractinian corals has been investigated, but the source of the activ e compounds has not been elucidated (Koh 1997). Anti-microbial compounds do not have to be germicidal in order to be effective. In fact, purely lethal activities in crease selection pressure on potential pathogens, thus creating resistant strains. Bacteria have developed a cell-to-c ell communication system based on diffusible small molecules by which population density can be monitored and regulate community-level behaviors. This process is te rmed quorum sensing (QS) and a llows for coordinated behaviors, developmental habits or physiologi cal processes within a population of unicellular organisms. At the basis is expression of genes, incl uding those required for pathogenic or symbiotic behaviors within a host (Miller and Bassler 2001, Taga and Bassler 2003, Waters and Bassler

PAGE 39

39 2005). Interference with QS by other organisms is an effective way to reduce the virulence of pathogens without congruently implementing selec tion pressure (Manefield et al. 1999, Dong et al. 2000, Rasmussen and Givskov 2006, Bjarnsholt and Givskov 2007). 3.1.1 Assessing QS with Chromobacteriu m violaceum Assays The understanding of host/microbe interactions has been aided in recent years by the study of QS. Detection methods have been de veloped to aid in the identification of organisms/compounds that influence QS. Chromobacterium violaceum is a bacterium that has a defined QS system and produces a purple pigment, called violacein. This pigment is induced by short side-chain acyl-homoserine lactones (AHLs) and is antagonized by long side-chain AHLs. A strain of C. violaceum has been engineered with a mutation in the AHL synthase gene (McClean et al. 1997). This stra in, CV026, has been widely used in direct and indirect assays when researchers to test for a gonist or antagonist capabilities, respectively, of a sample. Production of purple pigment in a direct CV026 assay indicates the presence of an agonist compound such as short side-chain AHLs. Indire ct CV026 assays are used to detect antagonist compounds, such as long side-chain AHLs, th rough the inhibition of pigment production. The range of AHLs with side-chains of C4 to C8 in length stimulate CV026 violacein production, although sensitivity of CV026 to each AHL varies. 3.1.2 Luminescence-Based QS Reporters Another group of QS reporters has been developed based on the luxCDABE (lux cassette) luminescence operon from Vibrio fischeri (Winson et al. 1998, Lindsay and Ahmer 2005). Plasmids that carry a full-length AHL receptor and a truncated AHL synthase promoter from V. fischeri ( luxR PluxI respectively), Pseudomonas aeruginosa ( lasR PlasI respectively), and Aeromonas hydrophila ( ahyR PahyI respectively) were constructed such that the luxCDABE genes from V. fischeri are controlled by these systems (Swift et al. 1997, Winson et al. 1998,

PAGE 40

40 Lynch et al. 2002, Kirke et al. 2004 ). Indirect and direct assa ys conducted with these Lux reporters allow the range of QS interactions of a sample to be accurately determined. 3.1.3 Detecting QS-active Compounds wi th Thin Layer Chromatography Thin layer chromatography (TLC) has been used extensively to identify QS-active compounds in sample extracts (McClean et al. 1997, Shaw et al. 1997, Gao et al. 2007). The immobile phase, the TLC plate, is chosen based on the type of separation desired. Previous work has shown that QS active compounds, incl uding AHLs, can be separated based on hydrophobicity (McClean et al. 1997, Shaw et al. 1997). The active compounds can be detected by overlaying the TLC plate with a QS reporter such as visu al markers associated with Agrobacterium tumefaciens NTL4 or Chromobacterium violaceum CV026 (McClean et al. 1997, Shaw et al. 1997). TLC overlays are semi-solid media mixtures that are seeded with liquid culture of a reporter bacterium and supplemented w ith appropriate chemicals such as an AHL for CV026, or x-gal for A. tumefaciens NTL4, in order to visualiz e the active compound migration. 3.1.4 Identification of Environmental Bacterial Isolates Using 16S rDNA Sequences A challenge of investigating m icrobes in highly specific ecological niches is that nichespecific environmental bacterial isolates have not been well characterized or perhaps even identified. It is therefore imperative to identi fy occupants of such environments in order to understand the ecological community in its entire ty. The most rapid and accurate method to identify unknown bacterial isolates is th rough genotyping based on the sequence of the candidates 16S rDNA. The 16S rDNA gene se quence that encodes the 1540 RNA nucleotides that encode the small subunit of the ribosome. Ribosomes are responsible for translation and proper production of proteins, and therefore the 16S rDNA seque nce has not been extensively altered over evolutionary time. Small changes in less c onserved regions of the 16S rDNA sequence occasionally occur and are an excellent means to establish bacterial identity or

PAGE 41

41 uniqueness. The16S gene allows for amplif ication and subsequent identification of environmental bacterial isolates from the high ly conserved regions, providing almost universal amplification primer sites. The adjacent slightly altered regions of the gene provide information that allow for the determination of phylogene tic relationships (Ra ppe and Giovannoni 2003). However, 16S rDNA sequence identification is no t perfect. Bias in current gene sequence databases is a result of the under representation of bacteria that ar e difficult to culture (Rappe and Giovannoni 2003). The procedures by which 16S rDNA sequences are generated and evaluated can also produce discrepancies in phylogenies. This is due to the reliance on PCR to amplify 16S rDNA sequences and PCR can introduce poin t mutations. In addition, computational algorithms vary in how phylogenies are produc ed and can lead to disagreement of the phylogenetic placements of environmental bacter ial isolates. Theref ore, 16S rDNA sequence based identification should be supported with other methods such as membrane fatty acid analysis or carbon sour ce utilization assays. 3.1.5 Hypothesis Tested in This Experiment The persis tent presence of bacteria in association with corals is consistent with the idea that coral associated bacteria perfor m a function within the coral holobi ont. In this experiment coral bacterial isolates from a library provided by Dr. Kim Ritchie were identified using 16S rDNA sequences and tested for the disruption of QS in model QS reporter systems. The hypothesis tested was that the coral bacterial isolates w ould produce compounds that in terfered with QS of model QS reporter systems. The ability to inhibit QS in model sy stems may be interpreted as the coral bacterial isolates providi ng the holobiont with antimicrobial compounds. This hypothesis was tested with luminescent QS reporter plasmids hosted in E. coli and with thin layer chromatography of organic extracts of coral bacterial isolate cultures, coupled with QS reporter overlays. These experiments provide information regarding the identity of the coral bacterial

PAGE 42

42 isolates, the coral bacterial isolates range of QS activities, and the extractability of the QS active compounds from the coral bacterial is olates using organic solvents. 3.2 Materials and Methods 3.2.1 CV026 Screening of a Library of Coral-Associated Bacteria Coral bac terial isolates were obtained from libraries provi ded by Dr. Kim Ritchie of the Mote Marine Laboratory. Isolates were stab-inoculated into 100 L of solidified Glycerol Artificial Seawater Agar (GASWA) in 96-well micr otiter plates and grown overnight at 30C. CV026 reporter was grown overnight in a 5mL LB shake culture at 30C. One-milliliter aliquots of the reporter culture were centr ifuged to collect the cells. The reporter cells were concentrated into ~50 L of culture supernatant. Twenty millilite rs of LB supplemented with 1.5% agar was poured into standard Petri plats a nd allowed to solidify. Once this base was firm, a soft overlay was prepared by adding a 2mL equivalent of reporter culture to a 50m L tube, along with 25mL of 0.3% agar LB and C4-AHL (final concentration of 0.2 g/mL) and mixed by vortexing briefly (AHL was dried in the tube prior to the addition of other component s). The seeded soft agar was poured over the solid base and allowed to solidif y for thirty minutes. Coral bacterial isolate culture plugs were moved into the soft agar and incubated at 30C overnight. Inhibition of QS is visualized as a ring of pigment inhibition, but not a reduction in repor ter growth, around the plug. 3.2.2 PCR Amplification of 16S rDNA Seque nces of Coral Bacterial Isolates 16S rDNA s equences were used to identify of th e coral bacterial isolates selected based on the QS inhibition screen. 16S rDNA sequences were amplified w ith PCR using the primers and cycling parameters detailed in Chapter 2.8. On ce amplified, these sequen ces were sent to the University of Florida Cancer and Genetics Institute Core Sequenci ng facility for Sanger

PAGE 43

43 sequencing. Resulting sequences were analyzed by NCBI BLAST against the nr database to obtain putative identifications. 3.2.3 Luminescent QS Reporter Assays Co-culture assays were conducted with Lux ca ssette reporters from Wi nson et. al. (1998) and Lindsay and Ahmer (2005) and the coral bacteria l isolates. Indirect co -culture assays were composed of 40 L of overnight isolat e culture added to 60 L of reporter culture induced with the appropriate acyl-homoserine lactone (AHL) in a black microtiter plate. AHL final concentrations in co-culture were 0.5 M for JM109 pSB401 and 1 M for JM109 pSB536 and JM109 pSB1075. Co-cultures were incubated static ally for s6 h at 37C, the optimal growing temperature for the E. coli reporter strains. Plates were read for luminescence with the VICTOR3 plate reader from Perkin-Elmer (Waltham, MA). 3.2.4 Cross-streak Assays with a Lumines cent QS Reporter and the Coral Bacterial Isolates For cross-streak assays, 20 L of overnight coral bacterial iso late culture was spot/run onto 10mL GASWA plates and allowed to incubate for 48 h at 30C. The streak colony was excised and moved to a 20mL 0.6% agar LB pl ate. Luminescent QS reporters were grown overnight and 10 L of each reporter was spot/run perpen dicular to the isolate streak. Plates were incubated at 37C for 5 h and photographed with a ChemiPro camera from Roper Scientific (Tucson, AZ). 3.2.5 Liquid Culture Extraction of Coral Bacterial Isolates Five-hundred m illiliter culture of the cora l bacterial isolates, buffered with 100 mM HEPES (pH 6.5) was grown in a shaking incuba tor at 30C for 48 h. Whole cultures were extracted twice with an equal vol ume of ethyl acetate in a separa tory funnel. The organic phases were collected and concentrated in a rotary evap orator from Buchi (New Castle, DE). Once all

PAGE 44

44 solvent had been removed, the ex tracts were dissolved in 400 L of ethyl acetate and stored at 20C. 3.2.6 Thin Layer Chromatography (TLC) Twenty-five m icroliters of each coral bacterial isolat e extract, an equiva lent of 30mL of whole culture, was spotted 1 L at a time 1cm from the edge of a C18 silica reverse phase TLC plates (Whatman). Plates were allowed to dry for 30 minutes and were developed with a mobile phase of 60% methanol (MeOH)/40% water until th e front of the field was 5 mm from the top of the plate. Plates were removed and dried for 1 h. Plates were then overlaid with one of two QS reporters. 3.2.7 QS Reporter Overlay of TLC Plates to Detect QS Active Compounds from the Coral Bacte rial Isolates Five milliliters of overnight cultures of either Agrobacterium tumefaciens NLT4 or Chromobacterium violaceum CV026 were diluted in to 35mL of either M9 liquid media or LB liquid media respectively. QS reporter cu ltures were incubated at 30C for 5 h ( A. tumefaciens) or 3.5 h (CV026). A. tumefaciens overlays were prepared by combining 75mL of M9 media supplemented with 1.5% agar with 25mL of M9 liquid media. Forty milliliters of bacterial culture supplemented with x-ga l (final concen tration of 8.6 g/mL) was added to the M9 media mixture. Approximately 100mL of this mix was p oured gently over the TLC plate in the lid of a 10 cm diameter Petri plate. This was allowed to solidify for 1 h before being covered with the bottom of the Petri plate and in cubated at 30C overnight. CV026 overlays were prepared in a similar manner as A. tumefaciens overlays except the CV026 s eeded overlay was supplemented with C4-AHL (final concentration 5 M). CV026 overlays were also conducted in 10cm Petri plate lids, except that a base of 1.5% Agar supplemented LB wa s poured and solidified prior to the application of the TLC plate. A portion of the solidified base, the size of the TLC plate, was

PAGE 45

45 removed and the plate was placed into the void Approximately 80mL of CV026 seeded media was poured over the TLC plate. CV026 cannot pr oduce pigment in the absence of oxygen and therefore a reduction in the overlay depth wa s required to ensure proper oxygenation. 3.3 Results 3.3.1 Screening of Coral-Associated Bacteria for QS Disruption with CV026 and the Identification of Six Coral Bacterial Iso lates Differentiated by 16S rDNA Sequences Three hundred and sixty-nine is olates from the libraries provided by K. Ritchie were screened with indirect CV026 assays to test the hypothesis that coral-asso ciated bacteria produce QS antagonist compounds (fig. 3-1). The recovery of six isolates that produced a zone of pigment inhibition in lawns of CV026 of at least 5 mm in diameter indicates that the approach is useful for detection of bacteria l isolates capable of QS disrup tion. The 16S rDNA sequence for each of the six coral bacterial isolates was amplified and the amplicons were compared to GenBank sequences using NCBI BLAST against the nr database. Based on comparison to sequences in the NCBI database, the six isol ates were consistent with sequences from Planococcus spp ., Caryophanon spp ., Marinobacter salsuginis Vibrio spp ., Photobacterium spp., and Agrobacterium stellulatum Planococcus spp ., Caryophanon spp ., and Marinobacter salsuginis are gram-positive bacteria while Vibrio spp ., Photobacterium spp ., and Agrobacterium stellulatum are gram-negative bacteria. Two of the coral bacterial isolates, Marinobacter salsuginis and Agrobacterium stellulatum were associated with zooxanthellae while the remaining four coral bacterial is olates were isolated from coral mucus. The only one of these isolates to be previously tested for identification was isolate 34-D11, Photobacterium spp., and was identified as Photobacterium mandapamensis by K. Ritchie. However, no species name could be ascribed to this isol ate based on the 16S rDNA sequence amplified in this experiment.

PAGE 46

46 3.3.2 Bioassays with Luminescent QS Reporters to Test the Coral Bacterial Isolates Range of QS Disruption The purpose of these bioassays was to exam in e the range of QS sy stems that the coral bacterial isolates could disrupt A reduction in the luminescen ce of a reporter indicates a disruption of that QS system. Due to the dyna mic nature of the marine environment, it was expected that the coral bacterial isolates would inte ract with a cariety of QS systems. Each coral bacterial isolate showed a vary ing amount of inhibiti on and as shown in Fig. 3-2, a large amount of deviation was present in the results. The graph represents nine total replications, three technical replications each of three biological replications. Vibrio spp. showed strong inhibition of all the reporters including the constitutive reporter Plambda luxCDABE that may indicate a general metabolic inhibition and not a QS specific inhi bition. When the vari ation was evaluated for significance with two-tailed ttests, the only significant (P<0.05) results were co-cultures with the coral bacterial isolates and the LuxIR re porter system. However, co-cultures with Agrobacterium stellulatum,and all of the luminescent reporters were statsitally non-reproducible under these experimental conditions. The exte nt of deviation was low in the technical replications. but when biological replications were averaged toge ther the deviation became more pronounced. This is most likely due to the large amount of variables within the biological systems of the coral bacterial is olates and physiological data on the coral bacterial isolates is limited. The cross-streak assay was develope d in order to remove the confounding factors present in liquid co-culture experiments. The LuxR reporter was chosen for these experiments because of the reproducibility indicated by th e preliminary liquid co-culture experiments. Figure 3-3 is a compilation of th e cross-streak assay images. Planococcus spp ., Photobacterium spp., and Marinobacter salsuginis showed inhibition, above that of the GASWA media control, at the head of the indirect assay streak (f ig.3-3 F, E, and D re spectively). None of

PAGE 47

47 the isolates exhibited any stimulation activities in the direct assay streaks of the LuxR reporter. The slight amount of luminescence observed in th e direct assay streaks is consistently on the portion of the streak that is proximal to the induced LuxR streak culture. Therefore, the stimulation of luminescence is likely due to AHL diffusion from the indirect assay streak culture. This result suggests that the coral bacterial isolates produ ce compounds that are active in multiple QS systems. It also supports the hypot hesis that the coral bacterial isolates produce compounds that antagonize m odel QS reporter systems. 3.3.3 Coral Bacterial Isolates Were Tested for Extractable QS-active Compounds with TLC Organic solvent extraction and TLC were us ed to test if th e QS active compounds produced by the coral bacterial isolates were ex tractable sma ll molecules. If a QS active compound from the coral bacterial isolates is a sm all molecule, as opposed to an enzyme, then it should be extractable with organic solven ts. An overlay of the TLC plate with Agrobacterium tumefaciens NTL4 pZLR4 (Shaw et al. 1997), an AHL synthase mutant that has a lacZ fusion to a QS regulated promoter, was used for broad-ra nge detection of QS ac tive compounds present in extracts of the coral bacterial isoaltes. Chromobacterium violaceum CV026 (McClean et al. 1997), which is also an AHL synthase mutant, produces purple pigment when QS is activated and was used for indirect assays to detect QS antagonists. Figure 3-4C shows a TLC plate with the equivalent of 30mL of culture of each isolate developed with 60% MeOH/40% water and overlaid with the Agrobacterium tumefaciens NTL4 reporter. Standards of C8-AHL and C6-AHL were also included as positive controls for the bioassay (fig. 3-4C). All of the gram-negative strains appear to be producing compounds that are capable of stimulating the A. tumefaciens NTL4 reporter. However, none of the gram -positive bacteria produce compounds that can activate this reporter. Figures 34A and B show similar TLC plates with an indirect and a direct

PAGE 48

48 assay CV026 overlay, respectively. Planococcus spp. and Caryophanon spp. produce compounds that inhibit the CV026 indi rect assay (fig. 3-4A), while Photobacterium spp. and Vibrio spp. produce compounds that stimulate the reporter (fig. 3-4B). This data indicates that the QS active compounds produced by the coral bact erial isolates are extractable with organic solvents; consistent with the notion that the QS active compounds may be small molecules. 3.4 Discussion 3.4.1 Screening and Identification of Coral Bacterial Isolates Two of the coral bacterial isolates identified in this screen, Marinobacter salsuginis and Vibrio spp., are related to bacteria th at have been previously described in other marine invertebrate-bacterial associations. Marinobacter spp. was identified in a screen of spongeassociated bacteria for antimicrobial compound production (Anand et al. 2006) and in a screen for QS activities in marine snow aggregates (Gra m et al. 2002). Large aggregates of bacteria can be seen in organ-like structures of Aiptasia pallida an anemone, and have been tentatively assigned to the Vibrio genus based on cell membrane fatty acid analysis (Palincs ar et al. 1989). The identification of gram-positive bacteria us ing the screening method here is interesting because the assay was designed against the gr am-negative QS system. Gram-positive QS systems utilize compounds other than the acyl-homo serine lactones as signals, and thus are not expected to interact with the gr am-negative signal receptors. There have been recent discoveries of a universal autoinducer molecule, A I-2 (Waters and Bassler 2005). Although not chemically characterized, it does appear that many bacterial species rely on this elusive compound (Waters and Bassler 2005) and simila r principles may underlie the unexpected response.

PAGE 49

49 3.4.2 Coral Bacterial Isolates Can Interact with Different QS Systems Co-cultu re assays were a preliminary way of testing the coral bacterial isolates for a the range of potential QS disruption cap abilities. However, statistical analysis revealed that while the variation within the experimental system is lo w, the variation present in the biological system is high as the level of deviation increased when the three biological repl icates were analyzed with two-tailed independent t-test s. Measures were taken to atte mpt to control for this variation such as controlling the starting cel l density of the coral bacterial isolates in the assay, ensuring a homogenous culture environment, maintaining pure cultures, and careful consideration of the time course of the experiments. However, none of these measures removed the biological variation. This persistent deviation is most likely due to the gap in knowledge regarding standardization of the physiological conditions that are required to produce the QS active compounds. The QS inhibition activities from the cora l bacterial isolates were subsequently investigated with the LuxIR reporter system in cross streak assays. The LuxIR QS reporter was created based on the QS system from V. fischeri and is responsive to 3-oxo-C6-AHL (Winson et al. 1998). The cross streak assays proved repeatable and reliable. The reproducible inhibition seen with Planococcus spp ., Photobacterium spp ., and Marinobacter salsuginis may be a product of the isolates being grown on a solid su rface instead of in a liquid culture, as gene regulation in bacteria can be a ffect by association with a surface (Eberl et al. 1999, Fraser and Hughes 1999, Kjelleberg and Molin 2002, Lynch et al. 2002). The cross streak assay shows promise for furthe r testing of these isolates to gain insight into the range and mechanism of the coral bacteria l isolates QS interaction activities. The range of activity can be tested by changing the QS report ers used in the cross st reak assay to reporters based on other QS systems such as the LasIR system, responsive to 3-oxo-C12-AHL, and the

PAGE 50

50 AhyIR system, responsive to C4-AHL. The mechanism of inhi bition by the coral bacterial isolates can be tested with QS reporters that carry mutations in the different domains of the receptor protein. These domains include the signal molecule-binding domain, the receptor dimerization domain, and the DNA binding domain (Miller and Bassler 2 001, Taga and Bassler 2003, Waters and Bassler 2005). These mutants would test at what point in a signal transduction cascade a coral bacterial isol ate interferes with QS. 3.4.3 Investigation into the Presence of QS Ac tivities in Organ ic Extracts of Coral Bacterial Isolates with TLC The TLC analysis with the A. tumefaciens reporter showed that the three gram-negative coral bacterial isolates produce compounds that st imulate this broad range AHL reporter. The CV026 assay confirmed the presen ce of agonist compounds in Photobacterium spp. and Vibrio spp. Antagonist compounds were also detected with the CV026 overlay, but only from the gram-positive Planococcus spp. and Caryophanon spp. The finding that organic extracts from the gram-negative isolates were not able to antagonize violacein pr oduction in CV026 but that the plug-cultures of the coral bacterial isolates did inhibit production of the pigment suggest that these activities are not solubl e in ethyl acetate or that when the compounds are highly concentrated there is an inverse effect co mpared to that of the intact colony.

PAGE 51

51 Figure 3-1. The CV026 indirect assa y screen of the coral bacteria l isolates for QS antagonism. The six coral bacterial isolates that produ ced the largest zones of pigment inhibition of CV026 were selected for further study. Agrobacterium stellulatum Vibrio spp., Caryophanon spp ., the GASW liquid media negative control, and the 3-oxo-C12-AHL (ODDHL) positive control were conducted on glass fiber disks. Marinobacter salsuginis, Photobacterium spp ., and Planococcus spp were conducted with GASWA plug cultures.

PAGE 52

52 0 50 100 150 200 250Plan. Phot. Mar. Agro. Vib. Cary.% Luminescence of Non-Cocultured Reporter Lambda Lux LuxR AhyR LasR Figure 3-2. Indirect co-culture bi oassay of three QS reporters and coral bacterial isolates. Plan: Planococcus spp .; Phot: Photobacterium spp .; Mar: Marinobacter salsuginis ; Agro: Agrobacterium stellulatum ; Vib: Vibrio spp.; Cary: Caryophanon spp. The average of nine total replications expressed as percentage of reporters luminescence when reporter is cultured alone and induced with appropriate AHL. Graph represents the average of three independent co-culture ex periments with three replications of each coral isolate within each inde pendent co-culture experiment Statistical evaluation of the results of the three in dependent experiments reveal ed non-significant (P<0.05) differences in effect of the coral bacter ial isolates on reporter luminescence. The lambda-lux E. coli reporter that is constitutively lumi nescent was co-cultured with the coral bacterial isolates as a control for general metabolic interference. Induced reporters that were not co-cul tured were included in the assay to create a baseline and as a control for proper reporte r function. Non-induced report ers were also included as a control for proper reporter function but were not included in the production of the graph.

PAGE 53

53 Figure 3-3. Cross streak assay with luxR QS reporter and cora l bacterial isolates. A-H: Caryophanon spp ., Vibrio spp., Agrobacterium stellulatum Marinobacter salsuginis, Photobacterium spp., Planococcus spp ., GASWA (control), and reporters incubated alone respectively (horizontal streaks). The ve rtical streaks from left to right are the negative control E. coli DH5 The direct assay with the LuxR reporter without cognate AHL, the indirect assay consisti ng of the LuxR reporter with 3-oxo-C6-AHL (final concentration of 6 M), and control for general metabolic inhibition with the constitutive reporter lambda-Lux. Images are representative of two independent experiments. Top images are photographs with light exposure a nd bottom images are the same plates without light. Images are after 5hrs of incubation at 37C and dark images were exposed for 4min.

PAGE 54

54 Figure 3-4. TLC of coral bacteria l isolate organic extr acts overlaid with the QS reporters CV026 and Agrobacterium tumefaciens NTL4. A-C) CV026 indirect assay, CV026 direct assay, and Agrobacterium tumefaciens NTL4 overlays respectively of C18 Reverse Phase silica TLC plates loaded with the equi valent of 30mL of coral bacteria isolate culture. 1-6 are Planococcus spp. Photobacterium spp., Marinobacter salsuginis, Agrobacterium stellulatum Vibrio spp., and Caryophanon spp. respectively. In A &B, AHL control (C) is 10 L of 10mg/mL 3-oxo-C12-AHL and in C AHL controls (C) are 1 L of 20 M C8-AHL (bottom spot) and 15 L of 100 M C6-AHL (top spot). TLC was developed in 60% MeOH : 40% Water. TLC was conducted twice with the same culture extract for the Agrobacterium reporter and twice with independent culture extracts for the CV026 assay. Sim ilar compound patterns were apparent in each of these replications.

PAGE 55

55 CHAPTER 4 CORAL ISOLATES DISRUPT NORMAL SU RFACE SPREADING PHENOTYPES OF Serratia ma rcescens STRAINS 4.1 Introduction 4.1.1 Surface Spreading S marcescens The ability of a pathogen to spread across surfaces is crucial to the initialization of an infection. S erratia marcescens is a ubiquitous pathogen that can infect plants, mammals, and insects (Grimont and Grimont 2006). Investigati ons into the virulence mechanisms present in S. marcescens have been undertaken (Givskov et al. 1998, Lindum et al. 1998). The surface spreading capabilities in multiple S. marcescens strains have been shown to be regulated by quorum sensing (QS) (Givskov et al. 1998, Lindum et al. 1998, Horng et al. 2002). Two surface spreading phenotypes have been characterized in different S. marcescens strains, swarming and sliding. Swarming motility involves the differentia tion of cells into swarm cells coordinated with production of a surfactant that allows the swarm cells to move easily over a surface (Givskov et al. 1998). Sliding motility is solely dependant on the production of a surfactant allowing the colony to spread out in finger-like projections (Hor ng et al. 2002). Bioassays that utilize spreading behaviors are implemented to test for QS inhibition because they can be manipulated and are easily visu alized (Rasmusse n et al. 2000). 4.1.2 S. marcescens as a Tool for QS Screening Molecular study of QS i n S. marcescens has been aided by the development of mutant strains that are deficient in traits that contri bute to surface spreading mo tility. The strain of S. marcescens responsible for the bacterial soft rot in cucumbers, MG1, relies on the swarmingtype motility to rapidly colonize surfaces (Givsk ov et al. 1998). MG1 has been mutagenized to produce the AHL synthase mutant MG44 (Givs kov et al. 1998), and MG44 was subsequently mutagenized to produce the su rfactant production mutant PL10 (Lindum et al. 1998). The S.

PAGE 56

56 marcescens PL10 strain was produced by transposon muta genesis. The transposon contained the luxAB genes, encoding the subunits for luciferase, allowing QS to be monitored via light production when n-decanal is exogenously applied to the system(Lindum et al. 1998). PL10 is a powerful tool for studying QS disruption because swarming can be restored by the addition of surfactant to determine if a re duction in swarming, motility is due to interference with the flhDC (flagellar master operon) genes responsible fo r swarm cell differentiation as opposed to QS regulated surfactant production. The use of MG1 and MG44 strain s allow for the identification of QS disruption activitie s in a sample and PL10 enables investigation in to the mechanism of that disruption (Rasmussen et al. 2000). The large amount of information available on the QS regulated spreading systems in S. marcescens strains makes this species an ideal tool for molecular investigations of QS antagonism produced by environmen tal samples (Manefield et al. 1999). 4.1.3 Hypothesis and Experimental Importance In this experim ent, the ability of the six coral bacterial isolates to disrupt swarming mediated surface spreading of S. marcescens was tested. It was hypot hesized that the coral bacterial isolates would inhibit swarming of S. marcescens through antagonism of QS. The ability to interfere w ith surface spreading of the model pathogen S. marcescens may suggest that the coral bacterial isolates provide QS anta gonist compounds to the holobiont, possibly to prevent pathogen infection. This hypothesis was tested with swarming assays using three S. marcescens strains to determine if antagonism of sw arming was due to QS disruption or was an effect of general metabolic repression. This e xperiment provides informa tion about the potential for the coral bacterial isolates to use QS antagoni sm against a behavior involved in virulence of a known pathogen.

PAGE 57

57 4.2 Materials and Methods Swar ming experiments were conducted based on Givskov et. al. (1998) Coral bacterial isolates were grown as disk cultures on GASWA media for 48 h prior to use in the swarming experiments. Disk cultures of each isolate where then aseptically moved on to AB SWARM media and allowed to grow for 24 h. Ten mi croliters of an overnight culture of a S. marcescens strain was spotted 1cm away from the isolate disk. In assays where S. marcescens MG44 was used, butryl-acyl hom oserine lactone (C4-AHL) was added to the AB SWARM media to a final concentration of 2 M. Plates were then incubated face-up at 30C for 8 h or 16 (MG44 and MG1 respectively) and surface spreadi ng was recorded with photography. Complementation of PL10 swarming was achie ved by extracting surfactant from swarming plate cultures of MG1 with ethy l acetate extraction based on the methods of Matsuyama et. al. (1995). Extracts were dried by rotary evaporation and re-suspended in 95% ethyl alcohol. Ten microliters of extract was spotted onto a glass fiber disk and allowed to dry overnight. Disks where placed on top of the AB SWARM media and 5 L of overnight PL10 culture was placed on the center of the disk. Plates were incubated at 30C for 16 h. 4.3 Results 4.3.1 Effects of the Coral Bacter ial Isolates on Sw arming of S. marcescens MG1 Swarming assays with wild type S. marcescens MG1 were used to determine if the coral bacterial isolates would inhibit surface spreading. An alteration to the spreading behavior of S. marcescens MG1 would potentially indicate that the co ral bacterial isolates were capable of disrupting this QS regulated behavior. Based on the results of the cross-streak assays it was expected that the coral bacterial isolat es would inhibit su rface spreading in S. marcescens MG1. When incubated with Planococcus spp., Photobacterium spp., Marinobacter salsuginis or Caryophanon spp. the swarming of the wild type (WT) strain of S. marcescens, MG1 was

PAGE 58

58 inhibited. Marinobacter salsuginis Photobacterium spp. and Caryophanon spp all show general inhibition of swarming across the en tire MG1 colony (Fig. 4-1 A3,4&7) while Planococcus spp. only inhibits swarming proximal to the MG1 colony (Fig 4-1 A2). Agrobacterium stellulatum has no effect on the swarming of MG1 as shown in Figure 4-1 A5. The promotion of swarming of MG1 by Vibrio spp. is interesting because Vibrio spp. inhibited the C4-AHL-controlled pigment production in CV026 during the initial library screens. The QS system of MG1 also includes C4-AHL; however, C6-AHL is also capable of inducing swarming in S. marcescens (Givskov et al. 1998). The TLC of Vibrio spp. extracts suggests the production of QS agonists by this coral ba cterial isolate and may explain the swarming promotion seen in MG1. 4.3.2 Effects of the Coral Bacterial Isolates on the S warming of S. marcescens MG44 when Complemented with AHL The AHL synthase mutant of S. marcescens MG1, MG44, was used to te st if the lack of an AHL synthase had an effect on the swarming inhi bition exerted by the coral bacterial isolates during indirect assays. Due to the ability to cont rol the concentration of AHL in indirect assays with MG44, it was expected that th e effect of the coral bacteria l isolates on swarming of MG44 would be similar to those obs erved with MG1. Swarming phenot ypes of MG44 incubated in the presence of the coral isolates is altered from those of MG1. Marinobacter salsuginis and Photobacterium spp. show strong proximal swarming inhibition, with a zone of avoidance between the coral isolate disk and the MG44 colony (Fig 4-1 B3 and B4) but not overall inhibition as seen in the wild type strain MG1. Planococcus spp. continues to show proximal inhibition of swarming but not to the same extent as with the wild type MG1 strain (Fig. 4-1 B2). Caryophanon spp. (Fig. 4-1 B7) shows mild proximal inhibi tion in MG44 in contrast to the wild type in which Caryophanon spp. strongly inhibited swar ming across the entire S. marcescens

PAGE 59

59 colony. Vibrio spp. (Fig. 4-1 B6) and Agrobacterium stellulatum (Fig. 4-1 B5) maintain similar effects on MG44 as on MG1 with promotion of swarming and no effect on swarming, respectively. The results support the hypothesis that the coral bacter ial isolates that were capable of inhibiting swarming in WT would also inhi bit swarming in the AHL synthase mutant MG44 in indirect assays. However, the swarming patte rn of MG44 is different then MG1, and therefore the extent of inhibition by some of the coral bacterial isolates were diminished in MG44 as compared to MG1. This effect may be due to the addition of excess AHL to the experimental system that may titrate out the antagonist capabilities of the coral bacterial isolates. 4.3.3 Effect of the Coral Bacterial Isolates on the Swarming of S. marcescens PL10 When Compleme nted by Surfactant PL10 is a mutant created by transposon mutagenesis of MG44 (Lindum et al. 1998) that has both a mutation in the auto-inducer gene (f rom MG44) and a transposon insertion into the QS regulator target gene swrA The swrA gene is involved in su rfactant production and this insertion makes PL10 incapable of swarming (Lindu m et al. 1998). Swarming can be restored to PL10 by addition of surfactant, since the strain still maintains a w ild type copy of the flhDC operon required for swarm cell differentiation (L indum et al. 1998). Swarming assays with the surfactant mutant of S. marcescens, PL10 were employed to determine if the coral bacterial isolates antagonism of swarming in S. marcescens was due to the disruption of QS. If swarming is impaired in complemented PL10, it indicates that swarm cell differentiation was inhibited. Based on the cross-streak assays it was expected that the coral bact erial isolates would not inhibit swarming of the PL10 strain complemented exogenous ly with surfactant. When incubated with any of the coral isolates other than Vibrio spp., PL10 swarms normally in the presence of surfactant. Vibrio spp. promotes swarming in PL10 in a fa shion similar to that of MG44 and

PAGE 60

60 MG1 (Fig. 4-1C). This data supports the hypothesis that the coral bacterial isolates disrupted the swarming of MG1 and MG44 through QS antagonism. 4.4 Discussion 4.4.1 Planococcu s spp. Effect on Surface Motility of S. marcescens Strains Planococcus spp. was capable of inhibiting swar ming proximally in all of the S. marcescens strains tested. Surface spreadi ng of MG1 was strongly inhibited by Planococcus spp, and mildly inhibited the spread of MG44 on AHL-s upplemented media. This suggests that the inhibition of spreading had been diminished in this experiment, as compared the wild type experiment. One explanation may be that the excess amount of cognate AHL in the supplemented media may have titrat ed out the antagonist effect of Planococcus spp Planococcus spp. did not have an effect on th e swarming of complemented PL10. This may suggest that the disrupt ion of swarming in the S. marcescens strains MG1 and MG44 was due to QS antagonism. 4.4.2 Photobacterium spp an d Marinobacter salsugini s Have Similar Effects on the Swarming of S. marcescens Strains The ability of Photobacterium spp. and Marinobacter salsugin is to disrupt surface motility was first tested with MG1 and MG44. Swarming of MG1was strongly inhibited by Photobacterium spp and Marinobacter salsuginis (Fig 4-1 A3 and A4 respectively). However, when the mutant MG44 was incubated with Photobacterium spp. or Marinobacter salsuginis a large zone of avoidance appeared proximal to the isolate culture disk (Fig. 54-1 B3 and B4 respectively). This pattern may indicate the presence of a diffusib le repellent compound produced by Photobacterium spp and/or Marinobacter salsuginis. Similar zones have been observed in another swarming bacterium, Pseudomonas aeruginosa (Caiazza et al. 2005, Tremblay et al. 2007). P. aeruginosa controls the swarming colony by producing compounds

PAGE 61

61 that repel other arms of the colony within a z one of diffusion (Caiazza et al. 2005, Tremblay et al. 2007). A pattern of avoidance similar to that of Photobacterium spp and Marinobacter salsuginis is observed when the repellent compound, a rhamnolipid, was purified and applied to a disk next to a culture of swarming P. aeruginosa (Tremblay et al. 2007). A titration test would be an experimental way to test for the pres ence of a repellent com pound produced by the coral isolates. The presence of a re pellent would be suggested if, when more cognate AHL were added to the system, there were no change in sp reading inhibition. If the inhibition exerted by Photobacterium spp. or Marinobacter salsuginis is due to a QS antagonist compound, then adding more cognate AHL to the system should overcome the inhibition. 4.4.4 Vibrio spp. Effect on S urface Motility of S. marcescens Strains Vibrio spp. was the only isolate to show promoti on of swarming in the strains MG1 and MG44. This isolate was also shown to pr oduce compounds capable of stimulating two QS reporters when culture extracts were separated with thin layer chromatography (see Chapter 3). This data taken together may suggest th at the QS agonist compounds present in Vibrio spp are the cause of the swarming promotion in experiments with MG1 and MG44. However, Vibrio spp. promoted swarming in PL10, a swarming mutant that is insensitive to AHL concentration. The results with Vibrio spp and PL10 suggest that swarming promotion by Vibrio spp. is due to factors other then QS agonists. 4.4.5 Caryophanon spp. Effec t on Surface Motility of S. marcescens Strains Caryophanon spp. inhibited spreading of MG 1 colonies but only mildly inhibited spreading of MG44 in indirect assays. This re sult may be due to the amount of AHL in the experimental system. In an ecologi cal scenario, a compound produced by Caryophanon spp may be capable of sufficient antagonism of QS syst ems, but when excess AHL is present in to an experimental system, the inhibition is lost. Caryophanon spp. did not inhibit spreading of strain

PAGE 62

62 PL10; lending support to the hypothesis, that swar m inhibition is due to QS antagonism and not disruption of cellular differentiation.

PAGE 63

63 Figure 4-1. Coral bacterial isol ates alter swarming phenotypes of Serratia marcescens strains. A) Wild Type Serratia strain MG1 B) AHL synthase mutant S. marcescens MG44 C) Surfactant deficient mutant S. marcescens PL10. A&B 2-7 and C3-8: Planococcus spp., Photobacterium spp., Marinobacter salsuginis, Agrobacterium stellulatum Vi brio spp. and Caryophanon spp respectively. Coral isolates were pre-grown on glass fiber disks for 24 h on GASWA me dia and then incubated on AB SWARM media for 24 h prior to plate inocul ation with 5ul of diluted overnight S. marcescens at a distance of 1cm from the isolate dis k. Swarming was monitored at 30C for 8 h for MG44 and 24 h for PL10 and MG1. Media for the MG44 experiments was supplemented with C4-HSL to a final concentration of 2 M. PL10 cultures were spotted onto a disk that was infused w ith surfactant extracte d from swarming WT cultures. All photos for each S. marcescens strain are from the same time point. Images in A&B are representative of two i ndependent culture experiments of both the isolate culture and the S. marcescens culture, with each experiment containing two replications of each isolate and S. marcescens combination. The c ontrols for A and B were also repeated two times in each of the two independent experiments. Images in C are representative of one experiment c ontaining two replications of each isolate incubated with PL10 complemented with surfactant as well as two replications of the controls (PL10 alone, C1, and PL10+surfactant, C2)

PAGE 64

64 CHAPTER 5 PRELIMINARY STUDIES OF THE QS SYSTEM PRESENT IN THE CORAL PATHOGEN Serratia ma rcescens PDL100 AND POTENTIAL REGULATION OF MULTICELLULAR PHENOTYPES BY QS IN PDL100 5.1 Introduction 5.1.1 White Pox Disease of Acropora Corals in the Caribbean White pox is a disease of corals caused by the Ser ratia marcescens strain PDL100 (Patterson et al. 2002). A major epidemic of the disease White Pox destroyed 70% of Acropora palmata on an annual average between 1998 and 2002. While S. marcescens is a ubiquitous pathogen, PDL100 is the first identification of this species causing symptoms in coral (Patterson et al. 2002). Symptoms include white circular lesions on the coral head indicative of tissue necrosis (Patterson et al. 2002, Sutherland and Ritchi e 2004). The spread of the disease has been calculated at up to 2.5 cm/day (Patterson et al. 2 002). White Pox can result in the death of the entire coral colony in the span of a single year (Patterson et al. 2002, Sutherland and Ritchie 2004). This disease fits the pattern of a pa thogen-induced disease, because coral heads neighboring infected colonies ar e most susceptible and infecti on with the PDL100 strain has been shown to cause disease symptoms on hea lthy coral fragments in laboratory experiments (Patterson et al. 2002). 5.1.2 Quenching Quorum Sensing The inactiva tion of AHLs from bacterial colonies is referred to as QS quenching (Dong et al. 2001). QS quenching can be accomplished through at least two mechansisms, including a lactonase that hydrolyzes the lactone ring or an acylase that removes the acyl side chain of the AHL (Taga and Bassler 2003) A lactonase from Bacillus spp ., AiiA, was identified by Dong and colleagues in 2000 and was shown to reduce virulence of the plant pathogen Erwinia carotovora (Dong et al. 2000). The applications of this technology are far reaching and of

PAGE 65

65 utmost interest due to the simplicity of incor porating a single gene into bacterial strains, generating new tools that may mitigate substantia l economic or environmental losses (Dong et al. 2001). Introduction of the aiiA gene allows for investigation into the contribution of QS in regulating bacterial phenotypes such as motility, biofilm formation, and enzyme production. In characterized strains of Serratia marcescens QS controls swarming, surfactant production, and chitinase production, and it contribu tes to biofilm formation and maintenence (Eberl et al. 1996, Horng et al. 2002, Christensen et al. 2003, Labbate et al. 2004, Rice et al. 2005, Queck et al. 2006, Van Houdt et al. 2007b). The QS system that may be present in S. marcescens PDL100, and any multi-cellular behaviors that may be regulated by that system, are presently unknown. Insight into QS systems in PDL100 may be gained by introducing the aiiA gene into PDL100 through the process of conjugation. 5.1.3 Conjugation Conjugation is a prom iscuous act in bacteria that facilita tes the exchange of genetic material. Conjugation allows bacteria to adapt to environments and overcome selective pressure, driving rapid bacterial evolution (de la Cruz and Davies 2000). This process is initiated by the donor cell, the cell where the plasmid to be tran sferred resides. The donor cell produces a pilus that attaches to a recognition si te on the recipient cell (Curtiss 1969). The pilus then contracts and pulls the donor and recipient cells in contact. Once in cont act with each other, a pore complex forms between the two bacteria. While the pore complex is forming, plasmid replication begins on the donor cell pl asmid at the origin of transfer (oriT). As the replication of the plasmid continues, the lagging strand is pushed through the pore comple x into the recipient cell. The lagging strand is replic ated inside the recipi ent cell and the leading strand is replicated within the donor cell. Once rep lication of both strands is comple te, the DNA is nicked at the oriT sites and the plasmids se parate. Shortly after the comp letion of replication the pore

PAGE 66

66 complex collapses and the cells separate (Wilkins and Lanka 1993, Lanka and Wilkins 1995, Firth et al. 1996). Conjugation ca n be manipulated to introduce gene s of interest into bacteria that are resistant to other labor atory methods of genetic transf ormation, such as electroporation or chemical induced competency, thus making conjugation an important tool for the study of bacterial genetics. 5.1.4 Behaviors Regulated by QS in Other S. marcescens Strains Can Be I nvestigated in PDL00 5.1.4.1 Surface spreading Surface spreading is on e of the proce sses that contribute to infection. S. marcescens MG1 swarming is positively regulated by QS through the stimulation of surfactant production (Givskov et al. 1998). In S. marcescens strain SS-1 QS negatively regulates the production of surfactant and therefore represse s sliding motility (Horng et al. 2002). The variation in surface spreading mechanisms, the potential means of regulation, and the potential implications of surface spreading to a serious coral disease increases interest in de termining if QS has a role in modulating this behavior in PDL100. If surface sp reading in PDL100 is controlled by QS, then a QS quenched strain would exhib it an altered phenotype compared to wild type and any QS inhibitory effects on PDL100 by the coral isolates may be altered. 5.1.4.2 Biosurfactant production Biosurfactan t production is required for bot h swarming and sliding motilities (Givskov et al. 1998, Horng et al. 2002). Surfactant production can be visualized by the drop collapse test (Lindum et al. 1998). Lindum et. al. used this method to show that QS regulates surfactant production in S. marcescens strain MG1 (Lindum et al. 1998). Surfactant production is also regulated, albeit negatively, in S. marcescens SS-1 by QS (Horng et al. 2002). If PDL100

PAGE 67

67 produces a surfactant then it may be required fo r surface spreading, and thus be important for virulence of this pathogen. 5.1.4.3 Biofilm formation One of the bacterial lifestyles that is key to sustained infec tion is biofilm formation. QS regulation contributes to biofilm formation and maintenance in some S. marcescens strains (Kjelleberg and Molin 2002, Labbate et al. 2004, Rice et al. 2005). Biofilms are almost ubiquitous in liquid environments and may form on biotic or abiotic surf aces (Costerton et al. 1995, Kjelleberg and Molin 2002). Biofilms are comp lex community structures that can range in depth between a few micrometers to millimeters (Costerton et al. 1995). Bacterial cells will aggregate to form mushroom-like structures wi th channels of fluid between them and each microcolony may have a distinct gene expression profile (Costerton et al. 1995, Kjelleberg and Molin 2002). The precise role of QS in controll ing the formation of biofilms is debated, but evidence exists that supports a contribution of QS to proper biofilm formation and maintenance (Kjelleberg and Molin 2002, Labbate et al. 2004, Rice et al. 2005). 5.1.5 Hypothesis The purpose of these experim ents was to i nvestigate QS regulati on of multi-cellular behaviors in S. marcescens PDL100 using a QS quenching appr oach. I hypothesized that the incorporation of the aiiA gene into PDL100 would affect surface spreading, surfactant production, and biofilm formation. This hypothe sis was tested with swarming assays, drop collapse tests, and biofilm assays. If the introduction of the aiiA gene did affect these behaviors, it would indicate a role for QS in regulati ng virulence in the pathogen PDL100. These experiments provide information about the regulation of virulen ce related behaviors in PDL100.

PAGE 68

68 5.2 Materials and Methods 5.2.1 Conjugation Conjugation was used to m ove the aiiA gene carried on the pDSK519 plasmid into the PDL100 strain. The protocol for this experiment is detailed in 2.13. The pDSK519 plasmid carries kanamycin resistance and the empty plasmi d was also moved into PDL100 as a control. In this experiment the Recipien t was PDL100, the Donor was either E. coli DH10B pDSK-aiiA or E. coli DH5 pDSK519, and E. coli MT616 was used as the help er strain. Selection for successful transformants was done on tetracyclin e (10mg/mL) and kanamycin (50 mg/mL). The tetracycline resistance comes from th e PDL100 strain and the other three E. coli strains were sensitive to this antibiotic. 5.2.2 16S rDNA Sequence Amplification by PCR To confirm the identity of the transformants as PDL100 and not E. coli, the 16S rDNA sequences were amplified from single col onies of the PDL100 WT PDL100 pDSK519 and PDL100 pDSK-aiiA strains as well as from the E. coli DH10B pDSK519 and DH5 pDSK-aiiA strains. This PCR was conducted as outlined in Chapter 2 with the MT44/MT45 primers. Cycling parameters and reaction mix were the same as those indicated in 2.8. 5.2.3 PCR to Amplify the ai iA Gene Colony PCR was performed with aiiA -specific primers on the colonies that resulted from conjugation to confirm the presence of the aiiA gene. The primers used were aiiA2L2 (forward) with the sequence TGTATGTTGGATCATTCGTCTGT and aiiA148R1 (reverse) with the sequence CAGGACCGGATTTTTCTGTC. These primers were designed against the aiiA gene from Bacillus spp 240B1. The reaction mixture contained the same final concentrations of reagents as listed for 16S rDNA PCR and the sa me cycling parameters (2.8). Products were

PAGE 69

69 electrophoresed on a 0.9% agarose/TAE gel stained with ethidium bromide and photographed with the GelDoc EQ imager (BioRad). 5.2.4 Digestion of 16S rDNA PCR Products with Hind III The 16S rDNA PCR products from the five st rains were run on a 0.9% agarose/TAE gel with ethidium bromide and eluted using the illustra DNA and Gel Band Purification Kit (GE Healthcare). Eluted DNA was then subj ected to diagnostic digestion with the HindIII restriction enzyme (Fisher Scientific). The digestion mix consisted of 1 L of HindIII enzyme, 1 L of the appropriate 10X buffer, and 8 L of DNA. This reaction was in cubated for 1.5 h at 37C. The reactions were separated on anot her 0.9% agarose/TAE gel contai ning ethidium bromide. The resulting pattern of product sizes was diagnostic, as the Hind III enzyme does not cut the PDL100 16S rDNA amplicon but cleaves the E. coli 16S rDNA sequence once, producing two products. 5.2.5 Organic Extraction PDL100 WT, pDSK519, and pDSK-aiiA strains were grown in 500m L of LB to a final OD600 of 1.7 and were buffered to pH6.0 with mor pholinepropanesulfonic ac id (MES) at a final concentration of 10mM. Cultures were divided in half and extracted two times with half volumes of ethyl acetate. The ex tracts were concentrated with the rotary evaporator listed in 2.4. One extract of each strain representing 250mL of culture was treated with 100 L of 0.1M ammonium hydroxide (NH4OH) in water and the second extract aliquot from each strain was left untreated as a control. The NH4OH-treated extracts incubated at room temperature for one hour. After the incubation, the samples were extracted twice with eq ual volumes of ethyl acetate, vortexed briefly, and centrifuge d at 12,000 rcf for one minute. Organic phases were moved to clean tubes and concentrated in the LABCONCO CentriVap Concentrator spin-vac for thirty minutes at 42C. Both the treated and untreated concentrated extracts were dissolved in 200 L of ethyl acetate and stored at -20 C. This is a modification of the protocol from Gao et al (2005).

PAGE 70

70 5.2.6 Thin Layer Chromatography Twenty-five m icroliters (the equivalent of 31mL of culture ) of treated and untreated extracts was spotted on to aC18 reverse phase TLC plate (Whatman). The plate was dried for one hour in a fume hood and developed for forty-fi ve minutes in 60% MeOH: 40 % water. The developed plate was dried for one hour and overlaid with the Agrobacterium tumefaciens NLT4 reporter as described in 3.2.7. 5.2.7 Drop Collapse Test To test the hypothesis that the a iiA gene would influence surface spreading, a drop collapse test was performed. This test would provide indication of surfactant production in the PDL100 WT (wild type), pDSK519 (vector control), and pDSK-aiiA ( aiiA carrying strain). Five milliliter cultures of each strain were grown for 48 h in LB supplemented with appropriate antibiotics at 30C. Forty microliters of each strain was place d by pipetting onto a Petri plate lid and the lid was tilted to 90 to allow the drops to run. Migr ation of the drop down the vertical plate would be indicative of surfactant productio n. For statistical purposes, the results were as run or no run. Three biological replicati ons containing fifteen technical re plicates each were conducted. The number of run results was analyzed using a two-tailed independent t-test. This is a variation of the protocol desc ribed in Lindum et al (1998). 5.2.8 Swarming with PDL100 WT, pDSK519, and pDSK-aiiA Swar ming experiments were conducted as described in 5.2 except that the strains used were the S. marcescens PDL100 strains, the distance between the PDL100 strain and the isolate disk was 1.5cm, and the duration of incubation wa s 17 h. Plates were photographed using a BioRad ChemiDoc XRS imager.

PAGE 71

71 5.2.9 Biofilm Assay For this experim ent, 5mL cultures of the PDL100 strains were grown in LB with appropriate antibiotics for 48 h at 30C. Cu lture was removed by gentle pouring. Remaining biofilms were stained using crystal violet and inc ubated at room temperature for fifteen minutes. The crystal violet was poured o ff and the biofilms were washed with 10mL of water three times before the tubes were photographed. 5.3 Results 5.3.1 Introduction of the aiiA Gene into PDL100 us ing Conjugation The initial hypothesis was that the aiiA gene, borne on the pDSK519 plasmid, could be mobilized into PDL100 using conjugation and then faithfully detected in the resulting colonies with a combination of PCR and diagnostic digestions. The introduction of the aiiA -carrying plasmid and the empty vector into the PDL100 stra in was verified with se quence amplification of both the aiiA gene and the 16S rDNA sequence that was subsequently digested with HindIII. As shown in Figure 5-1, the PDL100 stra in transformed with the pDSKaiiA plasmid tested positive for the aiiA gene. The DH10B aiiA strain was included as a positive control and the DH5 pDSK519 strain was included as a negative control. The PDL 100 pDSK519 strain and the WT both tested negative for the aiiA gene. Figure 5-2A shows the PCR products visualized on an agarose gel stained with ethidium bromide. A ll PCR products appeared to be the correct size (1.5kb). After purification of the DNA out of the gel matrix the DNA was digested with HindIII. Figure 5-2B depicts the agarose gel of these di gestions stained with ethidium bromide. HindIII does not cut the 16S gene of S. marcescens but does cut the 16S rDNA sequence of E. coli once producing two products from the single amplic on. All of the PDL100 sequences remained undigested by HindIII and all of the E. coli sequences were digested in to the appropriate size. These results support the hypothesis that PDL100 may be transformed by conjugation and that

PAGE 72

72 low-resolution analysis of the 16S rDNA sequence may serve as a diagnostic tool for confirming transformation. 5.3.2 Detection of QS Agonist Compounds P resent in Crude PDL100 Culture Extracts Using TLC The hypothesis tested with this experim ent was that the QS-active compounds present in crude PDL100 culture extracts that were capable of stimulating the Agrobacterium tumefaciens NTL4 QS reporter would be inactiv ated by both the presence of the aiiA gene and by base hydrolysis. For this experiment 500mL culture s of the WT, pDSK519 an d pDSK-aiiA strains were extracted with ethyl acetate and run on TLC plates as desc ribed in 3.2.6 and 3.2.7. To test if reduction in active compounds present was due to hydrolysis of a lactone linkage, samples were treated with a strong base. Strong bases chemically induce the hydrolysis of a lactone linkage in the presence of water. Both treate d and untreated extracts were spotted on a TLC plate and developed in 60% MeOH/ 40% water. Figure 5-3 presents the results of the TLC plate after A. tumefaciens NTL4 overlay. The lanes with asteri sks indicate extract samples that were hydrolyzed prior to TLC. The PDL100 WT stra in appears to have two compounds, although not well separated, that can induce act ivity in the reporter. In th e hydrolyzed WT sample, roughly inline with the C6-AHL standard, the compound towards the bottom of the field was mildly inactivated by the base, while the activity of the compound traveling the farthest in the field was completely inactivated (Fig. 5-3). The PDL100 pDSK519 strain had a sim ilar pattern of active compounds as WT, except that the base-hydrol yzed extract still c ontained both active compounds. The two compounds derived from the treated sample from PDL100 pDSK519 shifted farther up the field and separated more clearly, suggesting altera tion to those compounds. However, the compounds were still capable of act ivating the reporter to a similar level as the untreated extract.

PAGE 73

73 Finally, the PDL100 pDSK-aiiA untreated extract exhibite d a lower level of active compounds than both the WT and the vector contro l strain with the farthest migrating compound being completely removed from the PDL100 pDSK-a iiA strain (Fig. 5-3). However, overall the compounds in the aiiA strain did not migrate as far as the WT and vector control. This may be due to an exaggerated oiliness of the PDL100 pDSK-aiiA extract as compared to the other two strains. Again, there is a reduction in the con centration of the compounds present in the PDL100 pDSK-aiiA base-hydrolyzed extract, and they are shif ted comparably to the vector control strain. However, the level of activity is similar to the untreated sample. The control compounds, C6 and C8-AHL, were completely inactivated by the NH4OH treatment. The results of this experiment suggest that the compou nds that stimulate the Agrobacterium tumefaciens NTL4 reporter present in PDL100 are not susceptible to base hydrolysis. However, this result may be due to the short period of incubation with the ba se and further optimization of this experiment may produce different results. 5.3.3 Effects on Surface Spreading of PDL100 after the Introduction of the a iiA Gene The hypothesis tested with this experiment was that surface spreading of S. marcescens PDL100 would be affected by the introduction of the aiiA gene. Surface spreading is a behavior in other S. marcescens strains that has been shown to be regulated by QS (Givskov et al. 1998, Horng et al. 2002). Therefore, the effects of aiiA on surface spreading in PDL100 were tested. PDL100 spreads in a sunburst pattern, as show n in Figure 5-4A. The introduction of pDSK519 (Fig. 5-4B) or pDSK-aiiA (Fig. 5-4C) showed no effect on spreading behavior of PDL100. However, there is variation in all strains in re gards to timing of swarming, diameter, and pattern. This variation may be due to po ssible inconsistencies in media vi scosity since it has been shown that altering the level of agar in the SWARM me dia can inhibit or promote spreading behavior

PAGE 74

74 (Rice et al. 2005). The results of this e xperiment do not support the hypothesis that the introduction of the aiiA gene would affect surface spreading in S. marcescens PDL100. 5.3.4 Effect of Coral Isolates on Surface Sp reading of PDL100 WT, pDSK 519, and pDSKaiiA Strains The hypothesis tested with this experime nt was that the introduction of the aiiA gene into PDL100 would alter the effect of the coral bacterial isolates on surface spreading of PDL100. The effect on surface spreading of the PDL100 strains by the coral bacterial isolates was investigated due to the drastic eff ects on surface spreading of the model S. marcescens strains. If surface spreading observed in the wild type strain is different than the quorum-quenched strain, then this evidence would support the hypothesis. Planococcus spp ., Photobacterium spp ., Marinobacter salsuginis and Agrobacterium stellulatum all inhibited surface spreading of PDL100 WT over the entire S. marcescens colony (Fig. 5-5). There was no effect on surface spreading of PDL100 WT by Caryophanon spp and Vibrio spp. promoted spreading of this strain (Fig. 5-5). Swarming experiments with the QS quenched PDL100 and the vector control PDL100 strains were conducted in the presence of the coral bacteria l isolates to investigate any effect that the presence of a lactonase gene may have on the coral bacteria l isolates inhibition of surface spreading. The vector control strain had similar results as the WT strain, except that Caryophanon spp appeared to inhibit PDL100 pDSK519 sp reading (Fig. 5-5). However, this could be an artifact of the variation observed in all PDL100 strains spreading as addressed previously. The PDL100 pDSK-aiiA strain was aff ected by the coral isolates in ways similar to WT except that the promotion of surface spreading exerted by Vibrio spp. was more pronounced and accelerated, with spreading evident after 13 h of incubation, whereas the WT and vector control strain only exhibited promotion between 16 and 17 h of incubation (Fig. 5-5). Overall, the data from this experiment do not support the hypothesis that introducing the aiiA gene into S.

PAGE 75

75 marcescens PDL100 alters the affect of the coral bacterial isolates on the surface spreading of PDL100. 5.3.5 Effect of Introduction of aiiA on Surfactant P roduction in PDL100 To test the hypothesi s that intr oducing the aiiA gene into PDL100 would alter surfactant production in the pathogen, drop collapse tests were employed. Surfactant production has been shown to be regulated by QS in the S marcescens strains MG1 and SS-1 (Givskov et al. 1998, Lindum et al. 1998, Horng et al. 2002). Therefore, this trait was also tested in the PDL100 QS quenched strains to ascertain if QS was invol ved in regulating this action in PDL100. The presence of a surfactant in PDL 100 WT was indicated by the results of a drop collapse test with MG1, MG44, and PL10. MG1 and PL10 liquid cultu res do not produce surfactant and therefore drops of culture do not collapse and run. MG44 liquid cultures supplemented with C4-AHL will collapse and run since the QS signal molecule is added above the biologi cal threshold required for the gene activation. As shown in Figur e 5-6A, PDL100 runs down the Petri plate lid, indicating the presence of a surface tension-lowering substance such as a surfactant. There was a large amount of variation when th e QS quenched strain and the vect or control strain were tested alongside WT. Therefore, three i ndividual colonies were used to start three biological replicate liquid cultures. These cultures were than tested fifteen times for collapse-and-run activity. The results were scored as run or fail to run as shown in an ex ample plate in Figure 5-6B. The total number of run results from each strain wa s subjected to two-tailed independent t-tests. Statistical analysis revealed that there was no significant difference between WT and the vector control strain (p=.171). There was a si gnificant difference between the WT and aiiA strains (p=.005); however, there was no statistically si gnificant difference between the vector control and the aiiA strain (p=.633). The introduction of the aiiA gene into PDL100 did not significantly

PAGE 76

76 alter the surfactant production of this pathogen, at least as evaluated by drop collapse tests, and therefore does not suppo rt the hypothesis tested. 5.3.6 Effect of Introducing a iiA into PDL100 on Biofilm Formation The hypothesis tested with this e xperiment was that introducing the aiiA gene into PDL100 would affect biofilm formation. It was discovere d that after 48 h of incubation at 30C, liquid cultures of PDL100 WT, pdSK519, and pDSK-aiiA exhibited diffe rent biofilm morphologies (Fig. 5-7). In glass culture tubes PDL100 WT (Fig. 5-7A&D) and pDSK519 (Fig. 5-7 B&E) formed similar biofilms, consisting of a dense ring at the interface of the culture and the air along with a slightly developed film between the surface of the cult ure and a depth of ~5mm. The PDL100 pDSK-aiiA biofilm did not have the pro nounced ring at the interface of the culture. Furthermore, the film between the interface and a depth of ~5mm was more dense than that of the WT or the vector control (Fig. 5-7C&F). This experiment was also car ried out in polystyrene and polypropylene microtiter plates as well as on glass slides but was only reproducible in glass tubes. Two other culture media were also tested but the biofilms produced by the three strains were comparable. Biofilm formation of PDL 100 was altered after th e introduction of the aiiA gene as compared to the WT and vector control strains. This result s upports the hypothesis that introduction of the lactonase gene aiiA into PDL100 affects biofilm formation. However, surfaces and culture conditions also appear to influence biofilm formation in all PDL100 strains tested. 5.4 Discussion 5.4.1 Conjugation and Confirmation of the Presence and Activity of the aiiA Gene in PDL100 pDSK-aiiA Three m ethods of transforming S. marcescens PDL100 were attempted by members of this lab: electroporation, chemical transformation, and conjugation. The only method that proved

PAGE 77

77 effective was conjugation. Therefore, conjugation was used to introduce the aiiA gene from Bacillus spp ., borne on the pDSK519 plasmid, into PD L100. One of the major hurdles of conjugation is confirming the identity of the colo nies recovered from the conjugation processes. This is because the antib iotic resistance from the donor plasmid can be transferred to the helper strain and in some cases, due to the nature of the mating mix, donor or he lper cells can grow on the selective plates if they are in close contact to the recipient cells. This created the need to devise methods to confirm colony genotype with a means other than selective antibiotics. One option is the amplification of the 16S rDNA gene sequence and restricti on digest analysis. However, this process may not be applicable for environmental samples if there is no information in sequence databases with which to compare the sample. Therefore, a rapid and inexpensive method was developed here to c onfirm the identity of strains produced by conjugation. The method involves amplificatio n of the 16S rDNA and subsequent diagnostic restriction enzyme digestion. This method pr oved effective, considering the high level of similarity of the 16S sequences of S. marcescens and E. coli. The HindIII enzyme will cut the E. coli 16S gene once, but the S. marcescens 16S gene does not possess a HindIII recognition site. This diagnostic digest allowed us to confirm that the colonies recovered on the selective antibiotic plates were S. marcescens PDL100 and not E. coli, but did not confirm the presence of the aiiA gene in the PDL100 pDSK-aiiA strain. In order to detect whether the aiiA gene had been successfu lly moved into PDL100, PCR amplification with aiiA specific primers was performed. This experiment confirmed the presence of the aiiA gene within the PDL100 pDSK-aiiA strain. Howeve r, the presence of the aiiA gene did not necessarily indicate an activ e AiiA enzyme. The activity of AiiA was investigated with organic extrac tion and TLC. Preliminary results showed that there were two

PAGE 78

78 compounds present in PDL100 WT th at were able to activate the A. tumefaciens NTL4 reporter and that the concentration of these compounds appeared to be diminished in the PDL100 pDSKaiiA strain. Yet the fact that these compounds were able to ac tivate the reporter did not confirm that these compounds were AHLs. To investig ate the presence of a lactone linkage in compounds from the organic extrac ts of the PDL100 strains, base hydrolysis was conducted with NH4OH to open up a lactone bond in a similar fashi on as AiiA. This treatment inactivates AHL compounds (Gao et al. 2005). The TLC with the hydrolyzed extracts revealed that in the WT strain the majority of the activity of the two compounds had been removed. The treatment also reduced the level of activity in the vector control strain, PDL100 pDSK519, and also shifted the compounds farther up the TLC fiel d indicating alteration of the compounds present. In the PDL100 pDSK-aiiA strain active co mpounds were not as concentrated as in the WT and vector control strains. In addition, th ese compounds did not travel as fa r in the field as the compounds from the other two strains. This is mo st likely due to an oily nature of the aiiA extract. All cultures were streaked to purity before extract ion so the possibility of contamination is negligible. These results cannot confirm the activity of AiiA sin ce the untreated ex tracts of the vector control and aiiA strains did not respond to ba se hydrolysis like the WT. 5.4.2 Effects of aiiA Introduction in to P DL100 on Surface Spreading and Surfactant Production The effect of aiiA introduction on PDL100 surface sp reading and surfactant production was tested, due to the importance of surface sp reading in pathogenic ity and the role of surfactants in surface spreading behavior of other S. marcescens strains (Givskov et al. 1998, Lindum et al. 1998, Eberl et al. 1999, Horng et al. 2002, Van Houdt et al. 2007a). The incorporation of the aiiA gene into PDL100 produced a statistically significant difference in surfactant production, measured by drop collapse of the PDL100 WT and PDL100 pDSK-aiiA

PAGE 79

79 strains. However, there was no significant diffe rence between the vector control strain PDL100 pDSK519 and the QS quenched strain PDL100 pDSK -aiiA. This indicates that the difference seen between the WT and QS quenched strain may be an artifact of the introduction of the vector alone. The vector control strain and the WT strain did not have st atistically significant differences in drop collapse results Some of this variation may be due to a lack of knowledge about the growth habits and nature of the surface-tension reducing compound(s) produced by PDL100, and suggest that a more sensitive method for surfactant production be implemented in future experiments. In line with the results of the drop collapse test, there was no significant difference between the surface spreading beha viors of the WT, vector contro l, and QS quenched strains. Only slight differences were obs erved between the effects of the coral bacterial isolates on the WT, vector control, and aiiA strains. The most interesti ng of these differences was that Vibrio spp. promoted spreading more rapidly in the aiiA strain than in the other two. Other experiments regarding the QS activities present in the coral bact erial isolates found that Vibrio spp. produces at least two QS agonist compounds detectable by C. violaceum (Fig 3-3). In addition, the Vibrio spp. was able to stimulate swarming in the MG1, MG44, and PL10 strains. 5.4.3 Effect of Introducing a iiA into PDL100 on Biofilm Formation Incubation for 48 h at 30C was required to produce biofilms in liquid cultures of the PDL100 strains in glass culture t ubes. The biofilms of the WT and vector control strains were very similar, while the PDL100 pDSK-aiiA strain had a deeper and denser biofilm. For further investigation of this phenomenon, the strains were cultured in sm all volumes in two types of microtiter plates and stained. The stain was solubilized and the concentration of the stain was detected with the multi-channel plate reader. Ho wever, this experiment showed no statistical difference between the three PDL100 strains. Th e biofilms of PDL100 strains were also grown

PAGE 80

80 on glass slides in 30mL of culture. While there were visible di fferences in the biofilms, the staining produced no significant differences in stai n uptake. All of these results indicate that while QS may have a role in biofilm developm ent, the type of surf ace appears to be more important in regulating biofilm formation. This finding is in agreement with other studies of biofilm formation in S. marcescens. Those studies suggest that while QS contributes to biofilm formation and maintenance, environmental cues, su ch as nutrient status and surface type, have a stronger influence over the regul ation of this behavior in S. marcescens strains (Labbate et al. 2004, Rice et al. 2005, Labbate et al. 2007).

PAGE 81

81 Figure 5-1. PCR amplification of the aiiA gene in PDL100. PCR products from PDL100 WT (WT), PDL100 pDSK519 (519), and PDL100 pDSK-aiiA (aiiA) visualized on agarose gel stained with ethidium bromide. E. coli Donor strains were included as controls for the vector control pDSK519 (E519) and the aiiA plasmid pDSK-aiiA (EaiiA). Figure 5-2. 16S rDNA PCR and subsequent Hind III digest of PDL100 QS quenched strains and E. coli strain controls. A ) 16S rDNA PCR products of the PDL100 WT (WT), pDSK519 (519), and pDSK-aiiA (aiiA) strains with E.coli donor strains as controls for pDSK519 (E519) and pDSK-a iiA (EaiiA). B) One hour HindIII digests of purified 16S rDNA PCR products. The ladder is a1kB ladder from Fisher Scientific. B A

PAGE 82

82 Figure 5-3. TLC of PDL100 WT pDSK519, and pDSK-aiiA extr acts. Organic extracts, equivalent to ~30mL culture, of the three S. marcescens PDL100 strains were spotted onto the TLC plate and developed in 60% Me OH: 40% water. Standards of C6-AHL and C8-AHL were included to the right. Plate was dried and overlaid with the A. tumefaciens reporter in x-gal supplemented media. Asterisk (*) indicates a sample that was treated with NH4OH prior to TLC run and C* is the same AHL compounds at the same concentrations as the control la ne treated with NH4OH. Figure represents one extraction of each PDL100 strain, contro ls were treated the same as the PDL100 culture extract during the hydrolysis treatment.

PAGE 83

83 Planococcus spp. Photobacterium s pp M arinobacter salsu g inis Ag robacterium stellulatum Vibrio s pp Car y o p hanon s pp Figure 5-4. Spreading behavior of Serratia marcescens PDL100 WT, pDSK519, and pDSK-aiiA strains. A) WT PDL100 B) PDL100 pDSK519 (empty vector control) C) PDL100 pDSK-aiiA. Images were taken after 17 h of incubation at 30C of an initial inoculum of 5 L diluted overnight liquid culture on AB SWARM media. Images are representative of three replications from the same culture. Figure 5-5. Surface spreading of PDL100, pD SK519, and pDSK-aiiA when incubated in proximity to coral bacterial isolates. PDL100 strains were spotted on the right of the coral isolate disk cultures and incubated for 17 h at 30C. Scale bars represent 7.5mm. All plates were incubated for the same period before photographing. Two replications of each isolat e/PDL100 strain combination from the original cultures were included in this experiment.

PAGE 84

84 Figure 5-6. Drop collapse test of Serratia marcescens strains. A) Drop collapse tests performed after 24 h incubation of MG1, MG44 supplemented with 2 M C4-AHL, and PL10 cultures and 48 h of incubation for PDL100 WT. B) is an example of drop collapse test after 48 h of incubation of PDL100 WT, PDL100 pDSK519, and PDL100 pDSKaiiA. All cultures incubated at 30C. Forty microliters of each strain was deposited on to the Petri plate lid at the same starting position and the plate ti lted to ~90. Plate was returned to a horizontal position and photographed. Collapse and run of the droplet indicates the presence of a su rface tension reducing compound such as a biosurfactant. The experiment in A) wa s conducted three times with droplets coming from the same culture each time. Figure B) is a sample plate of one of the three independent experiments conducted that contai ned fifteen replications of each strain culture. Statistical analysis by independent two-tailed t-test revealed no significant difference (P<.05) in the number of run results between the v ector control (519) and the QS quenched strain (aiiA). A B

PAGE 85

85 Figure 5-7. Biofilm formation of PDL100 in gla ss tubes is affected by the introduction of the aiiA gene. A) PDL100 WT in LB, B) PDL100 pDSK519 in LB, C) PDL100 pDKaiiA in LB, D-F) crystal violet staining of the biofilms in A-C. Five milliliter cultures grew for 48 h at 30C in LB media. Biofilm s were stained with crystal violet for 15 m and washed three times with water be fore being photographed. PDL100 pDSKaiiA has a more developed biofilm than the WT and the vector control. This experiment was conducted two times with i ndependent cultures with highly similar results.

PAGE 86

86 CHAPTER 6 CONCLUSIONS AND FUTURE WORK 6.1 Potential Importance of Coral Bacterial Isolates The coral holobiont is a com plex micro-ecosystem consisting of unicellular algae, bacteria and the coral. Understanding how this delicate and intricate ecosystem is created and maintained may aid efforts to preserve and restore coral reef s, as well as illuminate fundamental questions in symbiont biology. A comprehensive inventory of the coral holobiont constituents may provide additional means to estimate and possibly avoid detrimental effects of anthropogenic and natural perturbations. In addition, understanding the normal microflora of coral will aid in the investigations of disease outbreak s and perhaps inform the formula tion of preventative measures. Coral reefs represent a large oppor tunity for economies of the su rrounding regions, as coral reefs are primary draws for tourism and fishing. Most importantly, healthy coral reefs are v ital to the overall environmental security of the planet. For these reasons the study presented here was undertaken to test the hypothesis that the bacteria found associat ed with corals inhibit quorum sensing (QS), thereby protecting the coral from pathogen infection. The potential role of QS in virulencerelated behaviors of the coral pathogen S. marcescens PDL100 was also investigated, starting with th e hypothesis that community behavior in PDL100 would be regulated by QS in a manner similar to other characterized S. marcescens strains. This assessment is significant because S. marcescens PDL100 is responsible for the major disease of Acroporid corals, White Pox. White Pox is responsible for de vastating losses of Acropora palmata and Acropora cervicornis in the Florida Keys National Marine Sanctuary. Acropora palmata and Acropora cervicornis are listed endangered species. Due to the dramatic loss of the corals due to disease, information regarding vi rulence of Acroporid pathogens is of utmost importance.

PAGE 87

87 6.2 Identification of Coral Bacterial Isolates that Exhibit QS Antagonism Six isolates that exhibited QS antagonism were identified by screening a collection of bacteria associated with corals. The isolates were Planococcus spp ., Photobacterium spp ., Marinobacter salsuginis Agrobacterium stellulatum Vibrio spp., and Caryophanon spp. While the majority of these bacterial species have not previously been identified as associated with corals, Marinobacter spp. Vibrio spp., and Photobacterium spp. have described interactions with other marine invertebrates. Many Photobacterium spp and Vibrio spp. have been found in symbiosis with various squid specie s (Nishiguchi and Nair 2003), and Marinobacter spp. was identified in a screening of sponge-specific bacteria for isolates producing antimicrobial compounds (Anand et al. 2006). Further testing of the coral bacterial isolates suggested the capability of antagonizing QS in mu ltiple model QS reporter systems. Vibrio spp., Photobacterium spp ., and Agrobacterium stellulatum culture extracts contained compounds that stimulated the broad range QS reporter Agrobacterium tumefaciens NTL4 These three coral bacterial isolates are gram-negative bacteria and therefore may use AHL-mediated QS, which would account for the agonist compounds found in the extracts. Planococcus spp. and Caryophanon spp. culture extracts containe d compounds that inhibite d QS-regulated violacein production of the short-side chain AHL reporter Chromobacterium violaceum CV026. Planococcus spp. and Caryophanon spp. are both gram-positive bacteria and may use cyclic dipeptides for QS. There is an emerging interest in cyclic dipeptides as QS antagonists in gramnegative bacteria as they have b een shown to inhibit AHL QS circ uits (Holden et al. 1999, Miller and Bassler 2001, Waters and Bassl er 2005). This evidence suppor ts the hypothesis that coral bacterial isolates are capable of QS anta gonism and suggests that the normal bacterial consortium contributes to the overall health of the coral.

PAGE 88

88 6.3 Potential Role for Coral Bacterial Isolates in the Defense of Corals from Pathogens Virulence related behaviors are often regul ated by population density to allow for successful infection (Parsek and Greenberg 2000 ). Surface spreading is a bacterial community behavior that is often associated with hos t infection (Fraser and Hughes 1999, Sharma and Anand 2002). Surface spreading in the ubiquitous pathogen Serratia marcescens has been shown to be partially regulated by QS (Mats uyama et al. 1995, Eberl et al. 1996, Givskov et al. 1998, Lindum et al. 1998, Eberl et al. 1999, Van Houdt et al. 2007a). The ability of the coral bacterial isolates to disrupt surface spreading in S. marcescens was tested with the S. marcescens strain MG1. Four of the six coral bacterial isolates were capable of altering the surface spreading of MG1 and related mutant strains. These results, that th e coral isolates did not inhibit surface spreading of a complemented surfactan t mutant, support the hypothesis that the inhibition of surface spreading is due to QS antagonism. The marine algae Delisea pulchra produces QS antagonistic halogenated furanones that act as anti-fouling agents and have been shown to inhibit su rface spreading in S. marcescens MG1 (Rasmussen et al. 2000). The coral bacterial isolates may provide QS antagonists to the coral holobiont for a similar purpose and the data from the surface spreading experiments supports this conclusion. 6.4 Preliminary Results Suggest a Limited Role of QS in Regulating Virulence-Related Multi-Cellu lar Behaviors in S. marcescens PDL100 Serratia marcescens PDL100 is the causative agent of the disease White Pox that affects Acropora palmata Since White Pox is a driving force in the decline of Acroporid corals in the Florida Keys National Marine Sanctuary, it is important to understand the regulation of virulence-related behaviors in this pathogen. The limited number of test s performed under this study provides only a preliminary description of the role of QS regulation. Tests of virulence related behaviors of surface spreading and biof ilm formation through the introduction of the aiiA

PAGE 89

89 gene suggest that QS has a limited role in re gulating these behaviors in PDL100. While these results are not definitive, they do suggest that the regulation of these be haviors is complex in PDL100. It is also possible that the assays impl emented may be insuffici ent to detect subtle effects or effects that vary wide ly based on other cultural metrics not considered in experimental design. The resultsthat four of the six coral bact erial isolates inhibited surface spreading in the wild type strain of PDL100 and that surface spreading of PDL100 was not affected by the introduction of a quorum-quenching gene, aiiA suggest that either AHLs do not regulate this behavior or that the aiiA gene is not functiona l in PDL100. These two possibilities are supported further by the result that the introduction of the aiiA lactonase gene into PDL100 had little affect on two compounds present in PDL100 culture extrac ts capable of stimulating the broad-range QS reporter Agrobacterium tumefaciens NTL4. One way to test for an active AiiA enzyme would be to add an AHL that is li kely not produced by PDL100, such as C16-AHL, to wild type, vector control, and aiiA -carrying strains of PD L100 prior to organic ex traction. A functional AiiA enzyme could be inferred if, after TLC separa tion of the extracts, ther e is a reduction in the amount of C16-AHL, detected with the A. tumefaciens NTL4 reporter, in the aiiA PDL100 strain as compared to wild type. In order to unde rstand the potential QS sy stems present in PDL100, the agonist compounds present in the crude culture extracts should be characterized further. The biofilm assays of the PDL100 strains suggested th at while QS may contribute to the regulation of biofilm formation, nutrient status and the type of surface are more influentia l. Surface spreading and biofilm formation should be tested for a lteration by addition of pure AHLs or other QS signals such as cyclic dipeptides to determine if the behaviors are responsive to an increase in signal.

PAGE 90

90 6.5 General Conclusions Overall, this study has only begun to dissect the QS interference capab ilities of the coral bacterial isolates. W hile QS active compounds ar e present in these bact eria, this is not novel since bacteria would generally benefit from bein g able to disrupt QS. However, the indication that these isolates inhibit surface spreading of the ubiquitous pathogen S. marcescens is highly significant, although the role of QS in surface spreading of the coral pathogen S. marcescens PDL100 is still unclear. Furthermore, the identificati on of QS active molecules in PDL100 and the effects of the introduction of the aiiA gene have opened up a new avenue of research. By understanding the QS system present in PDL100, and the virulence related behaviors that it may control, directed methods of treatment and prevention of this pathogen may be developed. Through more diligent and defined experimentation such as that outlin ed above, the virulence mechanisms of this important pathogen may be elucidated and potential control measures developed. The results of this study provide some support to the hypothesis th at the bacteria closely associated with the coral Acropora palmata confer protection to the coral through QS disruption.

PAGE 91

91 LIST OF REFERENCES Anand, T. P., A. W Bhat, Y. S. Shouche, U. Roy, J. Siddharth, and S. P. Sarma. 2006. Antimicrobial activity of marine bacteria asso ciated with sponges from the waters off the coast of South East India. Microbiol Res 161:252-262. Bjarnsholt, T., and M. Givskov. 2007. Quorum-sensi ng blockade as a strategy for enhancing host defences against bacterial pathogens. Philos Trans R Soc Lond B Biol Sci 362:12131222. Bruneel, O., R. Duran, C. Casiot, F. Elbaz-P oulichet, and J. C. Personne. 2006. Diversity of microorganisms in Fe-As-rich acid mine dr ainage waters of Carnoules, France. Appl Environ Microbiol 72 :551-556. Bruton, B. D., F. MItchell, J. Fletcher, S. D. Pair, A. Wayadande, U. Melcher, J. Brady, B. Bextine, and T. W. Popham. 2003. Serratia marcescens, a Phloem-colonizing, Squash Bug-Transmitted Bacterium: Causal Agent of Cucurbit Yellow Vine Disease. Plant Disease 87:937-944. Bryant, D., L. Burne, J. McManus, and M. Sp alding. 1998. Reefs at Risk: A Map-based Indicator of Threats to the World's Coral Reefs. WCMC, Cambridge. Caiazza, N. C., R. M. Shanks, and G. A. O'Toole. 2005. Rhamnolipids modulate swarming motility patterns of Pseudomonas aeruginosa. J Bacteriol 187 :7351-7361. Cegelski, L., G. R. Marshall, G. R. Eldridge and S. J. Hultgren. 2008. The biology and future prospects of antivirulence th erapies. Nat Rev Microbiol 6:17-27. Chernin, L. S., M. K. Winson, J. M. Thompson, S. Haran, B. W. Bycroft, I. Chet, P. Williams, and G. S. Stewart. 1998. Chitinolytic activity in Chromobacterium violaceum: substrate analysis and regulation by quorum sensing. J Bacteriol 180:4435-4441. Christensen, A. B., K. Riedel, L. Eberl, L. R. Flodgaard, S. Molin, L. Gram, and M. Givskov. 2003. Quorum-sensing-directed protein expres sion in Serratia proteamaculans B5a. Microbiology 149:471-483. Costerton, J. W., Z. Lewandowski, D. E. Caldwe ll, D. R. Korber, and H. M. Lappin-Scott. 1995. Microbial biofilms. Annu Rev Microbiol 49:711-745. Curtiss, R., 3rd. 1969. Bacterial conjugation. Annu Rev Microbiol 23:69-136. de la Cruz, F., and J. Davies. 2000. Horizontal gene transfer and the origin of species: lessons from bacteria. Trends Microbiol 8:128-133. Dong, Y. H., L. H. Wang, J. L. Xu, H. B. Zhang, X. F. Zhang, and L. H. Zhang. 2001. Quenching quorum-sensing-dependent bacterial infection by an N-acyl homoserine lactonase. Nature 411 :813-817.

PAGE 92

92 Dong, Y. H., J. L. Xu, X. Z. Li, and L. H. Zhang. 2000. AiiA, an enzyme that inactivates the acylhomoserine lactone quorum-sensing signal and attenuates the vi rulence of Erwinia carotovora. Proc Natl Acad Sci U S A 97:3526-3531. Eberl, L., S. Molin, and M. Givskov. 1999. Surf ace motility of serratia liquefaciens MG1. J Bacteriol 181 :1703-1712. Eberl, L., M. K. Winson, C. Sternberg, G. S. Stewar t, G. Christiansen, S. R. Chhabra, B. Bycroft, P. Williams, S. Molin, and M. Givskov. 1996. Involvement of N-acyl-L-hormoserine lactone autoinducers in controlling the multi cellular behaviour of Serratia liquefaciens. Mol Microbiol 20:127-136. Engebrecht, J., K. Nealson, and M. Silverma n. 1983. Bacterial bioluminescence: isolation and genetic analysis of functions from Vibrio fischeri. Cell 32:773-781. Faruque, S. M., M. J. Albert, and J. J. Me kalanos. 1998. Epidemiology, genetics, and ecology of toxigenic Vibrio cholerae Microbiol Mol Biol Rev 62:1301-1314. Firth, N., K. Ippen-Ihler, and R. Skurray. 1996. Structure and function of the F factor and mechanism of conjugation. Pages 2377-2401 in F. Neidhardt, editor. Escherichia coli and Salmonella: Cellular and Molecular Bi ology. ASM Press, Washington, D.C. Fraser, G. M., and C. Hughes. 1999. Swarming motility. Curr Opin Microbiol 2:630-635. Frezza, M., S. Castang, J. Estephane, L. Soulere, C. Deshayes, B. Chantegrel, W. Nasser, Y. Queneau, S. Reverchon, and A. Doutheau. 2006. Synthesis and biologi cal evaluation of homoserine lactone derived ur eas as antagonists of bacter ial quorum sensing. Bioorg Med Chem 14:4781-4791. Fuqua, W. C., S. C. Winans, and E. P. Greenbe rg. 1994. Quorum sensing in bacteria: the LuxRLuxI family of cell density-responsive transcriptional regulators. J Bacteriol 176:269275. Gao, M., H. Chen, A. Eberhard, M. R. Gronquist, J. B. Robinson, M. Connolly, M. Teplitski, B. G. Rolfe, and W. D. Bauer. 2007. Effects of AiiA-Mediated Qurorum Quenching in Sinorhizobium meliloti on Quorum-Sensing Signals, Pr oteome Patterns, and Symbiotic Interactions. Molecular Plant-Microbe Interactions 20:843-856. Gao, M., H. Chen, A. Eberhard, M. R. Gronquist, J. B. Robinson, B. G. Rolfe, and W. D. Bauer. 2005. sinIand expR-Dependent Quorum Sensin g in Sinorhizobium meliloti. J Bacteriol 187:7931-7944. Geske, G. D., J. C. O'Neill, and H. E. Blackwell. 2007a. N-phenylacetanoyl-L-homoserine lactones can strongly antagonize or superagonize quorum sensing in Vibrio fischeri. ACS Chem Biol 2:315-319.

PAGE 93

93 Geske, G. D., J. C. O'Neill, D. M. Miller, M. E. Mattmann, and H. E. Blackwell. 2007b. Modulation of bacterial quorum sensing with sy nthetic ligands: systematic evaluation of N-acylated homoserine lactones in multip le species and new insights into their mechanisms of action. J Am Chem Soc 129:13613-13625. Gil-Agudelo, D. L., C. Myers, G. W. Smit h, and K. Kim. 2006. Changes in the microbial communities associates with Gorgonia ventalina during aspergillosis infection. Marine Ecology Progress Series 69:89-94. Givskov, M., J. Ostling, L. Eberl, P. W. Lindum, A. B. Christensen, G. Christiansen, S. Molin, and S. Kjelleberg. 1998. Two separate regula tory systems particip ate in control of swarming motility of Serratia liquefaciens MG1. J Bacteriol 180 :742-745. Gram, L., H. P. Grossart, A. Schlingloff, and T. Kiorboe. 2002. Possible quorum sensing in marine snow bacteria: production of acylated homoserine lactones by Roseobacter strains isolated from marine snow Appl Environ Microbiol 68:4111-4116. Green, E. P., and A. W. Bruckner. 2000. The sign ificance of coral diseas e epizootiology for coral reef conservation. Biol ogical Conservation 96 :347-361. Grimont, F., and P. A. D. Grimont. 2006. The Genus Serratia. Pages 219-244 The Prokaryotes. Springer New York, New York. Harvell, C. D., K. Kim, J. M. Burkholder, R. R. Colwell, P. R. Epstein, D. J. Grimes, E. E. Hofmann, E. K. Lipp, A. D. Osterhaus, R. M. Overstreet, J. W. Porter, G. W. Smith, and G. R. Vasta. 1999. Emerging marine diseases-climate links and anthropogenic factors. Science 285 :1505-1510. Hentschel, U., M. Schmid, M. Wagner, L. Fiesel er, C. Gernert, and J. Hacker. 2001. Isolation and phylogenetic analysis of bacteria with antimicrobial activities from the Mediterranean sponges Aplysina aerophoba and Aplysina cavernicola. FEMS Microbiol Ecol 35:305-312. Hentzer, M., L. Eberl, J. Niel sen, and M. Givskov. 2003. Quorum sensing : a novel target for the treatment of biofilm infections. BioDrugs 17:241-250. Holden, M. T., S. Ram Chhabra, R. de Nys, P. St ead, N. J. Bainton, P. J. Hill, M. Manefield, N. Kumar, M. Labatte, D. England, S. Rice, M. Givskov, G. P. Salmond, G. S. Stewart, B. W. Bycroft, S. Kjelleberg, and P. Williams. 1999. Quorum-s ensing cross talk: isolation and chemical characterization of cyclic dipeptides from Pseudomonas aeruginosa and other gram-negative ba cteria. Mol Microbiol 33:1254-1266. Horng, Y. T., S. C. Deng, M. Daykin, P. C. Soo, J. R. Wei, K. T. Luh, S. W. Ho, S. Swift, H. C. Lai, and P. Williams. 2002. The LuxR family protein SpnR functions as a negative regulator of N-acylhomoserine lactonedependent quorum sensing in Serratia marcescens. Mol Microbiol 45:1655-1671.

PAGE 94

94 Howard, M. B., N. A. Ekborg, R. M. Wein er, and S. W. Hutcheson. 2003. Detection and characterization of chitinases and other chitin-modifying enzymes. J Ind Microbiol Biotechnol 30:627-635. Kim, J., J. G. Kim, Y. Kang, J. Y. Jang, G. J. Jog, J. Y. Lim, S. Kim, H. Suga, T. Nagamatsu, and I. Hwang. 2004. Quorum sensing and the Ly sR-type transcriptional activator ToxR regulate toxoflavin biosynthesis and transp ort in Burkholderia gl umae. Mol Microbiol 54:921-934. Kirke, D. F., S. Swift, M. J. Lynch, and P. Williams. 2004. The Aeromonas hydrophila LuxR homologue AhyR regulates the N-acyl homoserine lactone s ynthase, AhyI positively and negatively in a growth phase-depend ent manner. FEMS Microbiol Lett 241 :109-117. Kjelleberg, S., and S. Molin. 2002. Is there a ro le for quorum sensing signals in bacterial biofilms? Curr Opin Microbiol 5:254-258. Klaus, J. S., I. Janse, J. M. Heikoop, R. A. Sanford, and B. W. Fouke. 2007. Coral microbial communities, zooxanthellae and mucus along gradients of seawater depth and coastal pollution. Environ Microbiol 9:1291-1305. Koh, E. 1997. Do Scleractinian Corals Engage in Chemical Warfare Agai nst Microbes? Journal of Chemical Ecology 23:379-398. Kurz, C. L., S. Chauvet, E. Andres, M. Aurouze, I. Vallet, G. P. Michel, M. Uh, J. Celli, A. Filloux, S. De Bentzmann, I. Steinmetz, J. A. Hoffmann, B. B. Finlay, J. P. Gorvel, D. Ferrandon, and J. J. Ewbank. 2003. Virulence factors of the human opportunistic pathogen Serratia marcescens identif ied by in vivo screening. Embo J 22:1451-1460. Labbate, M., S. Y. Queck, K. S. Koh, S. A. Rice, M. Givskov, and S. Kjelleberg. 2004. Quorum sensing-controlled biofilm development in Serratia liquefaciens MG1. J Bacteriol 186:692-698. Labbate, M., H. Zhu, L. Thung, R. Bandara, M. R. Larsen, M. D. Willcox, M. Givskov, S. A. Rice, and S. Kjelleberg. 2007. Quorum-sen sing regulation of adhesion in Serratia marcescens MG1 is surface dependent. J Bacteriol 189:2702-2711. Lanka, E., and B. Wilkins. 1995. DNA processing reactions in bacteria l conjugation. Annual Review of Biochemistry 64:141-169. Leadbetter, J. R. 2001. Plant microbiology. Quieting the raucous crowd. Nature 411:748-749. Lesser, M. P., J. Bythell, R. D. Gates, R. W. Johnstone, and O. Hoegh-Guldberg. 2007. Are infectious diseases really killing corals? Al ternative interpretations of the experimental and ecological data. Journal of Experimental Biology and Ecology 346:36-44. Lindsay, A., and B. M. Ahmer. 2005. Effect of sdiA on biosensors of N-acylhomoserine lactones. J Bacteriol 187 :5054-5058.

PAGE 95

95 Lindum, P. W., U. Anthoni, C. Christopherse n, L. Eberl, S. Molin, and M. Givskov. 1998. NAcyl-L-homoserine lactone autoinducers control production of an extracellular lipopeptide biosurfactant required for swar ming motility of Serratia liquefaciens MG1. J Bacteriol 180 :6384-6388. Liu, H., S. J. Coulthurst, L. Pritchard, P. E. He dley, M. Ravensdale, S. Humphris, T. Burr, G. Takle, M. B. Brurberg, P. R. Birch, G. P. Salmond, and I. K. Toth. 2008. Quorum sensing coordinates brute force and stealth modes of infecti on in the plant pathogen Pectobacterium atrosepticum. PLoS Pathog 4:e1000093. Lynch, M. J., S. Swift, D. F. Kirke, C. W. Keevil, C. E. Dodd, and P. Williams. 2002. The regulation of biofilm development by quorum sensing in Aeromonas hydrophila. Environ Microbiol 4 :18-28. Manefield, M., R. de Nys, N. Kumar, R. Rea d, M. Givskov, P. Steinberg, and S. Kjelleberg. 1999. Evidence that halogenated furanones fr om Delisea pulchra inhibit acylated homoserine lactone (AHL)-mediated gene e xpression by displacing the AHL signal from its receptor protein. Microbiology 145 ( Pt 2):283-291. Matsuyama, T., A. Bhasin, and R. M. Harshe y. 1995. Mutational analysis of flagellumindependent surface spreading of Serrati a marcescens 274 on a low-agar medium. J Bacteriol 177 :987-991. McClean, K., M. Winson, L. Fish, A. Taylor, S. R. Chhabra, M. Camara, M. Daykin, J. H. Lamb, S. Swift, B. W. Bycroft, G. Stewart, and P. Williams. 1997. Quorum Sensing and Chromobacterium violaceum: exploitation of violacein production and inhibition for the detection of N-acylhomoserine lactones. Microbiology 143:3703-3711. Miller, M., I. Baums, and D. WIlliams. 2007. Visu al discernment of sexual recruits is not feasible for Acropora palmata. Marine Ecology Progress Series 335:227-231. Miller, M. B., and B. L. Bassler. 2001. Quorum sensing in bacteria. Annu Rev Microbiol 55:165199. Miller, M. B., K. Skorupski, D. H. Lenz, R. K. Taylor, and B. L. Bassler. 2002a. Parallel quorum sensing systems converge to regulate virulence in Vibrio cholerae. Cell 110 :303-314. Miller, M. W., A. S. Bourque, and J. A. B ohnsack. 2002b. An analysis of the loss of acroporid corals at Looe Key, Florida, USA: 1983-2000. Coral Reefs 21 :179-182. Muh, U., B. J. Hare, B. A. Duerkop, M. Schuster, B. L. Hanzelka, R. Heim, E. R. Olson, and E. P. Greenberg. 2006. A structurally unrelate d mimic of a Pseudomonas aeruginosa acylhomoserine lactone quorum-sensing si gnal. Proc Natl Acad Sci U S A 103:16948-16952. Nishiguchi, M. K., and V. S. Nair. 2003. Evol ution of symbiosis in the Vibrionaceae: a combined approach using molecules a nd physiology. Int J Syst Evol Microbiol 53:20192026.

PAGE 96

96 Nyholm, S. V., and M. J. McFall-Ngai. 2004. The winnowing: establishing the squid-vibrio symbiosis. Nat Rev Microbiol 2:632-642. Palincsar, E. E., W. R. Jones, J. S. Palincsar, M. A. Glogowski, and J. L. Mastro. 1989. Bacterial aggregates within the epidermis of the s ea anemone Aiptasia pallida. The Biological Bulletin 177 :130-140. Pantos, O., and J. Bythell. 2006. Bacterial comm unity structure associated with white band disease in the elkhorn coral Ac ropora palmata determined using culture-independant 16s rRNA techniques. Diseases of Aquatic Organisms 69:79-88. Parsek, M. R., and E. P. Gr eenberg. 2000. Acyl-homoserine lactone quorum sensing in gramnegative bacteria: a signaling mechanism involve d in associations with higher organisms. Proc Natl Acad Sci U S A 97:8789-8793. Patterson, K. L., J. W. Porter, K. B. Ritchie, S. W. Polson, E. Mueller, E. C. Peters, D. L. Santavy, and G. W. Smith. 2002. The etiology of white pox, a lethal disease of the Caribbean elkhorn coral, Acropora pa lmata. Proc Natl Acad Sci U S A 99:8725-8730. Queck, S. Y., M. Weitere, A. M. Moreno, S. A. Rice, and S. Kjelleberg. 2006. The role of quorum sensing mediated developmental traits in the resistance of Serratia marcescens biofilms against protozoan grazing. Environ Microbiol 8 :1017-1025. Quinn, N. J., and B. L. Kojis. 2005. Patterns of se xual recruitment of acroporid coral populations on the west fore reef at Discove ry Bay, Jamaica. Rev Biol Trop 53 Suppl 1:83-90. Rappe, M. S., and S. J. Giovannoni. 2003. Th e uncultured microbial majority. Annu Rev Microbiol 57 :369-394. Rasmussen, T. B., and M. Givskov. 2006. Quorum-s ensing inhibitors as anti-pathogenic drugs. International Journal of Medical Microbiology 296 :149-161. Rasmussen, T. B., M. Manefield, J. B. Anderse n, L. Eberl, U. Anthoni, C. Christophersen, P. Steinberg, S. Kjelleberg, and M. Givs kov. 2000. How Delisea pulchra furanones affect quorum sensing and swarming motility in Serratia liquefaciens MG1. Microbiology 146 Pt 12:3237-3244. Reeves, L. 1994. Newly discovered: yellow-band di sease strikes Keys reefs. Underwater USA 11:16. Reshef, L., O. Koren, Y. Loya, I. Zilber-R osenberg, and E. Rosenberg. 2006. The Coral Probiotic Hypothesis. Environmental Microbiology 8:2068-2073. Reverchon, S., B. Chantegrel, C. Deshayes, A. Doutheau, and N. Cotte-Pattat. 2002. New synthetic analogues of N-acyl homoserine la ctones as agonists or antagonists of transcriptional regula tors involved in bact erial quorum sensing. Bioorg Med Chem Lett 12:1153-1157.

PAGE 97

97 Rice, S. A., K. S. Koh, S. Y. Queck, M. Labbate K. W. Lam, and S. Kjelleberg. 2005. Biofilm formation and sloughing in Serratia marces cens are controlled by quorum sensing and nutrient cues. J Bacteriol 187:3477-3485. Richardson, L. L. 1998. Coral diseases: what is really known? Trends in Ecology and Conservation 13:438-443. Ritchie, K. 2006. Regulation of microbial populations by coral surface mucus and mucusassociated bacteria. Mari ne Ecology Progress Series 322:1-14. Ritchie, K., and G. Smith. 2004. Microbial Co mmunities of Coral Surface Mucopolysaccharide Layers. Pages 259-264 in E. Rosenberg and Y. Loya, editors. Coral health and disease. Springer-Verlag, New York. Rohwer, F., V. Seguritan, A. Farooq, and N. Kn owlton. 2002. Diversity and distribution of coralassociated bacteria. Mari ne Ecology Progress Series 243:1-10. Rosenberg, E., and Y. Ben-Haim 2002. Microbial diseases of corals and global warming. Environmental Microbiology 4:318-326. Rosenberg, E., C. Kellogg, and F. Rohw er. 2007a. Coral Microbiology. Oceanography 20:146154. Rosenberg, E., O. Koren, L. Reshef, R. Efr ony, and I. Zilber-Rosenberg. 2007b. The role of microorganisms in coral health, dise ase and evolution. Nature Reviews 5:355-362. Sharma, M., and S. K. Anand. 2002. Swarming: A coordinated bacterial activity. Current Science 83 :707-715. Shaw, P. D., G. Ping, S. L. Daly, C. Cha, J. E. Cronan, Jr., K. L. Rinehart, and S. K. Farrand. 1997. Detecting and characterizing N-acyl-homos erine lactone signal molecules by thinlayer chromatography. Proc Natl Acad Sci U S A 94:6036-6041. Smith, K. M., Y. Bu, and H. Suga. 2003. Library screening for synthetic agonists and antagonists of a Pseudomonas aeruginosa autoinducer. Chem Biol 10:563-571. Solis, R., I. Bertani, G. Degr assi, G. Devescovi, and V. Vent uri. 2006. Involvement of quorum sensing and RpoS in rice seedling blight caused by Burkholderia plantarii. FEMS Microbiol Lett 259 :106-112. Spalding, M., and A. Grenfell. 1997. New estimates of global and regional coral reef areas. Coral Reefs 16:225-230. Sutherland, K. P., and K. B. Ritchie. 2004. White Pox Disease of the Caribbean Elkhorn Coral, Acropora palmata. Pages 289-300 in E. Rosenberg and Y. Loya editors. Coral health and disease. Springer-Verlag, New York.

PAGE 98

98 Swift, S., A. V. Karlyshev, L. Fish, E. L. Durant, M. K. Winson, S. R. Chhabra, P. Williams, S. Macintyre, and G. S. Stewart. 1997. Quor um sensing in Aeromonas hydrophila and Aeromonas salmonicida: identification of the LuxRI homologs AhyRI and AsaRI and their cognate N-acylhomoserine lactone signal molecules. J Bacteriol 179 :5271-5281. Taga, M. E., and B. L. Bassler. 2003. Chemical communication among bact eria. Proc Natl Acad Sci U S A 100 Suppl 2:14549-14554. Taylor, M. W., R. Radax, D. Steger, and M. Wagner. 2007. Sponge-associated microorganisms: evolution, ecology, and biot echnological potential. Microbiol Mol Biol Rev 71:295-347. Tremblay, J., A. P. Richardson, F. Lepine, and E. Deziel. 2007. Self-produced extracellular stimuli modulate the Pseudomonas aerugi nosa swarming motility behaviour. Environ Microbiol 9 :2622-2630. Van Houdt, R., M. Givskov, and C. W. Michiels. 2007a. Quorum sensing in Serratia. FEMS Microbiol Rev 31:407-424. Van Houdt, R., P. Moons, A. Aertsen, A. Jansen, K. Vanoirbeek, M. Daykin, P. Williams, and C. W. Michiels. 2007b. Characterization of a l uxI/luxR-type quorum sensing system and Nacyl-homoserine lactone-dependent regulat ion of exo-enzyme and antibacterial component production in Serratia plymuthica RVH1. Res Microbiol 158 :150-158. Waters, C. M., and B. L. Bassler. 2005. Quorum sensing: cell-to-cell communication in bacteria. Annu Rev Cell Dev Biol 21:319-346. Wegley, L., R. Edwards, B. Rodriguez-Br ito, H. Liu, and F. Rohwer. 2007. Metagenomic analysis of the microbial community associated with the coral Porites astreoides. Environ Microbiol 9 :2707-2719. Wilkins, B., and E. Lanka. 1993. DNA processing and replication duri ng plasmid transfer between Gram-negative bacteria. Pages 105-136 in D. Clewell, editor. Bacterial Conjugation. Plenum Publishing, New York. Winson, M. K., S. Swift, L. Fish, J. P. Throup, F. Jorgensen, S. R. Chhabra, B. W. Bycroft, P. Williams, and G. S. Stewart. 1998. Construction and analysis of luxCDABE-based plasmid sensors for investigating N-acyl hom oserine lactone-mediated quorum sensing. FEMS Microbiol Lett 163 :185-192.

PAGE 99

99 BIOGRAPHICAL SKETCH Stephanie H albig is the second child of Karl and Jackie Halb ig and has one older sister, Kari. Ms. Halbig was raised in Naperville, a s uburb of Chicago, and attended Naperville North High School. She graduated high school in 2000 and enrolled as an undergraduate student at the University of Illinois in Urbana. Ms. Halbig participated in functional genomics research in maize in Dr. Stephen Mooses lab during her underg raduate education. Duri ng her final year at the University of Illinois, Stephanie completed in an internship at the University of Cape Town in South Africa where she aided the study of genes involved in drought tolerance of the native plant Xerophyta viscosa Baker under the supervision of Dr. Jennifer Thomson. She completed a Bachelor of Science degree in the field of crop science: biot echnology and molecular biology in 2004. After completion of her B.S., she became a plant transformation technician at the University of Nebraska and worked on improve ment of the transformation efficiency of soybeans under the supervision of Dr. Tom Clem ente. In 2006, Stephanie was accepted into the Plant Molecular and Cellular Biology Program at the University of Florida in Gainesville and completed her Master of Science degree in 2008.