ABSTRACT Title of Document: QUORUM SENSING IN BACTERIA ASSOCIATED WITH MARINE SPONGES MYCALE LAXISSIMA AND IRCINIA STROBILINA Jindong Zan, Doctor of Philosophy, 2013 Directed By: Dr. Russell T. Hill, Professor, University of Maryland Center for Environmental Science Sponges can form close associations with microbes that in some cases comprise up to 30% of the biomass of the sponge, which we hypothesized would provide an ideal environment for quorum sensing. I isolated 420 bacterial strains from two marine sponge species and screened these isolates for acyl-homoserine lactone (AHL) production. Results showed that isolates from the Silicibacter- Ruegeria (SR) subclade of the Roseobacter group are the dominant AHL producers. Production of these signaling compounds was consistently observed in isolates obtained from different sponge individuals during different seasons. The SR-type strain Ruegeria sp. KLH11 was isolated from tissue of the sponge Mycale laxissima. Chemical analysis of the AHLs produced by Ruegeria sp. KLH11 showed them to be predominantly composed of a mixture of long chains AHLs with 3-OH substitutions. Two pairs of luxR and luxI homologues and one solo luxI homologue were identified and designated as ssaRI, ssbRI and sscI (sponge-associated symbiont locus A, B and C, luxR or luxI homologue). SsaI directs synthesis of predominantly 3-oxo-AHLs whereas SsbI and SscI specify 3- OH-AHL derivatives. Wild type Ruegeria sp. KLH11 cultures are dominated by SsbI or SscI-specified AHLs. Mutation of either ssaR or ssaI results in loss of swimming motility, flagellar production and flagellin synthesis whereas mutation of ssbR or ssbI had no effect on these characteristics and no detectable phenotype. In wild type cultures, flagella are produced only in late stage growth. The non-essential phosphorelay cckA-chpT-ctrA system acts downstream of ssaRI to control flagellar motility. Mutants of ssaI and ssaR showed increased biofilm formation while mutants of ssbI and ssbR were not affected in biofilm formation, and this is not due solely to the loss of motility. The results showed the presence of AHL molecules similar to those specified by SsaI in sponge tissues and that the ssaI gene is actively expressed in situ, revealed by RT-PCR. We have established Ruegeria sp. KLH11 as a model to study the complex symbiotic relationships between sponges and microbes. QUORUM SENSING IN BACTERIA ASSOCIATED WITH MARINE SPONGES MYCALE LAXISSIMA AND IRCINIA STROBILINA By Jindong Zan Dissertation submitted to the Faculty of the Graduate School of the University of Maryland, College Park, in partial fulfillment of the requirements for the degree of Doctor of Philosophy 2013 Advisory Committee: Professor Russell T. Hill, Chair Professor Robert Belas Associate Professor Feng Chen Professor Clay Fuqua Professor Jacques Ravel ? Copyright by Jindong Zan 2013 ii DEDICATION I dedicate my thesis to my wife Yue Liu, Ph.D, mother Jianfang Han, and father Shengfa Zan for their endless love and support. iii ACKNOWLEDGMENTS I would like to thank my wife, my parents and my parents-in law for their support and love. Their encouragement, understanding and high expectations have been always my motivation to work through many difficult stages in my education and also my personal life. It has been tremendously productive to work in the laboratory of Dr. Russell T. Hill over the past five years, where I have learnt how to be a good person, a good scientist and how to speak good English. I am grateful to my advisor, Dr. Russell T. Hill for accepting me into his laboratory at a very difficult time. I was fortunate to work with him since he provided guidance and gave me the freedom to think independently and make important decisions in research on my own. His high expectations were my strong motivation to work hard as his graduate student. He also provided me with all of the resources required for my research. I am also very grateful that he taught me how to speak and write well in English. I would like to thank my committee members: Drs. Robert Belas, Feng Chen, Clay Fuqua and Jacques Ravel for their support throughout my research. Special thanks to Dr. Chen for financial support in the first year of my Ph.D. study, Dr. Fuqua for hosting my visit to his laboratory where I learned many important and useful genetic techniques and also to Dr. Ravel for all the bioinformatic and sequencing help. I extend my thanks to all the past and current members in the Hill Laboratory that I have interacted with, especially to Mr. Matthew Anderson, who helped me iv settle down when I first joined the laboratory, and Dr. Naglaa Mohamed, who taught me many important laboratory techniques and has been a good friend. I would also like to give my thanks to Jeanette Davis for all her support and encouragement over the past five years. I thank Dr. Leah C. Blasiak for helping edit my writing many times and Ryan J. Powell for many useful scientific conversations. Jan Vicente is specially acknowledged for his help in sponge collection. I am also grateful to Dr. Michael Hibbing and Miss. Jing Xu from Dr. Fuqua?s laboratory, who taught me how to do biofilm assays and real-time PCR. I thank the National Science Foundation Microbial Observatories Program (MCB-0238515), the Microbial Interactions and Processes Program (MCB- 0703467) and BIO/IOS Program (IOS-0919728) for funding. v Statement of Contribution Elisha M. Cicirelli identified the ssaRI and ssbRI systems and Okhee Choi identified the sscI gene. Elisha M. Cicirelli, Stephanie Kroll and Okhee Choi created the strains and plasmids that start with EC, SK and OKC, respectively. Also, they performed most of the experiments with crude organic extract of KLH11 cultures. Jason E. Heindl performed the flagellar stain experiment and provided three expression constructs that start with JEH, described in Chapter 4. All these four colleagues are from Department of Biology, Indiana University. Hiruy Sibhatu, Charis L. Uhlson, Christina L. Wysoczyski, Robert C. Murphy, and Mair. E.A. Churchill performed all the mass spectrometric experiments. Christina L. Wysoczynski performed the organic extraction from sponge tissues. All these colleagues are from University of Colorado, Denver. vi Table of Contents Dedication!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!..ii Acknowledgments!!!!!!!!!!!!!!!!!!!!!!!!!!...iii Statement of Contribution!!!!!!!!!!!!!!!!!!!!!!!!v Tables of Contents!!!!!!!!!!!!!!!!!!!!!!!!!!..vi List of Tables!!!!!!!!!!!!!!!!!!!!!!!!!!!!...xiii List of Figures!!!!!!!!!!!!!!!!!!!!!!!!!!!!.xiv Chapter 1. Introduction and literature review ........................................................1 1.1. Marine sponges.....................................................................................2 1.1.1 Sponge body structure and reproduction.........................................2 1.1.2 Sponge taxonomy............................................................................4 1.1.3 Genomes of marine sponges...........................................................5 1.2. Marine sponge microbiology .................................................................6 1.3. Quorum sensing ..................................................................................12 1.3.1 General model of QS in Proteobacteria .........................................13 1.3.2 Case studies of QS: V. fischeri and P. aeruginosa ........................18 1.3.3 Quorum sensing in Roseobacters .................................................21 1.3.4 Novel AHL molecules ....................................................................27 1.3.5 AHL-mediated QS in non-Proteobacteria ......................................29 1.3.6 QS in V. harveyi and V. cholerae: LuxM, CqsS and LuxS-mediated pathways ......................................................................................................32 1.4. AHL-mediated inter-kingdom signaling: Ulva-AHL interaction ............35 1.5. Two-component systems ....................................................................37 vii 1.6. Focus and objectives...........................................................................41 Chapter 2. A complex LuxR-LuxI type quorum sensing network in a roseobacterial marine sponge symbiont activates flagellar motility and inhibits biofilm formation ..................................................................................................43 2.1. Abstract ...............................................................................................44 2.2. Introduction..........................................................................................45 2.3. Experimental procedures ....................................................................47 2.3.1 Reagents, strains, plasmids, and growth conditions......................47 2.3.2 Preparation of AHL samples and analysis by RP-HPLC and ESI mass spectrometry .......................................................................................49 2.3.3 Qualitative analysis and estimation of AHL quantities ...................50 2.3.4 Genomic library screen of KLH11 QS genes.................................51 2.3.5 Directed mutation, lacZ fusions and complementation ..................52 2.3.6 Plasmid-borne promoter fusions and expression plasmids ...........54 2.3.7 Preparation of log phase cell concentrates and "-galactosidase assays ..........................................................................................................55 2.3.8 AHL detection using an ultrasensitive A. tumefaciens reporter for KLH11 derivatives ........................................................................................56 2.3.9 Motility assay, flagellar stain and immunodetection of flagellin .....57 2.3.10 RNA extraction and quantitative reverse transcription PCR (qRT- PCR) analysis...............................................................................................59 2.3.11 Biofilm assays..............................................................................59 2.3.12 Statistical analysis .......................................................................60 viii 2.4. Results.................................................................................................68 2.4.1 AHL synthesis and genetic isolation of luxI and luxR homologues from KLH11 ..................................................................................................68 2.4.2 SsaI and SsbI synthesize long chain?length AHLs when expressed in E. coli........................................................................................................75 2.4.3 Mutational analysis of ssaRI and ssbRI in KLH11 .........................75 2.4.4 Ectopic expression of AHL synthases in a KLH11 QS mutant ......76 2.4.5 Expression of ssaI is stimulated in response to KLH11 AHLs .......82 2.4.6 SsaR activates expression of its cognate AHL synthase gene ssaI .....................................................................................................................83 2.4.7 Conserved sequences upstream of ssaI are required for activation by SsaR........................................................................................................88 2.4.8 SsaRI controls swimming motility and flagellar biosynthesis.........91 2.4.9 SsaRI mutants exhibit increased biofilm formation........................96 2.5. Discussion ...........................................................................................99 Chapter 3. A LuxI ?solo? synthesizes long chain acyl homoserine lactones and is involved in a reciprocal regulatory quorum sensing system in marine sponge symbiont Ruegeria sp. KLH11...........................................................................108 3.1. Abstract .............................................................................................109 3.2. Introduction........................................................................................110 3.3. Experimental procedures ..................................................................112 3.3.1 Bacterial strains, oligonucleotides and growth conditions ...........112 ix 3.3.2 Plasmid construction for null mutation, expression of sscI and lacZ- fusion..........................................................................................................112 3.3.3 Preparation of AHL samples and analysis by RP-HPLC and ESI mass spectrometry .....................................................................................114 3.3.4 Qualitative analysis and estimation of AHL quantities .................114 3.3.5 Preparation of log phase cell concentrates and "-galactosidase assays ........................................................................................................114 3.3.6 Motility assay ...............................................................................114 3.4. Results...............................................................................................118 3.4.1 Identification of sscI .....................................................................118 3.4.2 sscI encodes a protein for synthesis of long chain length AHLs and its expression is not stimulated by the KLH11 AHLs..................................119 3.4.3 sscI is positively regulated by ssaI and ssbI ................................124 3.4.4 SscI-derived AHL stimulates SsaR-dependent activation of ssaI 126 3.4.5 The sscI mutant shows reduced swimming motility. ....................127 3.4.6 KLH11 contains a novel enzyme responsible for the synthesis of p- HSL-like molecule. .....................................................................................128 3.5. Discussion .........................................................................................131 Chapter 4. The cckA-chpT-ctrA phosphorelay system is regulated by quorum sensing and controls flagellar motility in the marine sponge symbiont Ruegeria sp. KLH11..........................................................................................................135 4.1. Abstract .............................................................................................136 4.2. Introduction........................................................................................137 x 4.3. Experimental procedures ..................................................................141 4.3.1 Strains, growth conditions and plasmid transformation ...............141 4.3.2 Deletion of ssaR and generation of cckA, chpT and ctrA null mutants.......................................................................................................141 4.3.3 Cloning of phosphorelay components and promoter fusion constructs ...................................................................................................144 4.3.4 Evaluation of flagellar-based motility and presence of flagella ....146 4.3.5 Quantification of phosphorelay component promoter activity ......147 4.3.6 Analysis of KLH11 CtrA-dependent gene expression.................148 4.3.7 Multiple sequence alignment and phylogenetic analysis of the ctrA gene ...........................................................................................................149 4.4. Results...............................................................................................154 4.4.1 The KLH11 cckA, chpT and ctrA genes are non-essential and control flagellar motility...............................................................................154 4.4.2 CtrA regulates motility-related gene expression but not cell cycle- related genes. ............................................................................................161 4.4.3 CtrA autoregulates its own transcription but not that of the cckA gene ...........................................................................................................163 4.4.4 Cross complementation between KLH11 and A. tumefaciens homologues................................................................................................164 4.4.5 The SsaRI quorum sensing system regulates the transcription of ctrA, chpT and cckA genes ........................................................................167 4.4.6 SsaRI regulate ctrA, chpT and cckA expression indirectly ..........172 xi 4.4.7 Ectopic expression of ctrA restores motility to the QS deletion mutant ........................................................................................................173 4.5. Discussion .........................................................................................178 Chapter 5. Isolation and screening of AHL- and AI-2-producing bacteria and development of Ruegeria sp. KLH11 as a model to study bacterial colonization of sponges.............................................................................................................186 5.1. Abstract .............................................................................................187 5.2. Introduction........................................................................................189 5.3. Experimental procedures ..................................................................193 5.3.1 Sponge collection, bacterial isolation and identification, and AHL screening....................................................................................................193 5.3.2 luxS gene amplification from the vibrios ......................................194 5.3.3 Measurement of AI-2 activity .......................................................195 5.3.4 Organic extraction of sponge tissues and TLC overlay assay.....196 5.3.5 RNA extraction and RT-PCR from sponge tissue........................197 5.3.6 Colonization of sponge cell aggregates.......................................198 5.3.7 Colonization of whole sponges ....................................................199 5.4. Results...............................................................................................201 5.4.1 Screening of AHL-producing bacteria from marine sponges .......201 5.4.2 luxS genes from Vibrios in sponges ............................................208 5.4.3 Sponge tissues contain AHLs and detectable levels of ssaI transcripts...................................................................................................210 5.4.4 Colonization of sponge cell aggregates.......................................213 xii 5.4.5 Colonization of whole sponges ....................................................215 5.5. Discussion .........................................................................................217 Chapter 6. Conclusions and future directions ...................................................224 Appendix 1: Genome sequence of Ruegeria sp. Strain KLH11, an N- acylhomoserine lactone-producing bacterium isolated from the marine sponge Mycale laxissima!!!!!!!!!!!!!!!!!!!!!!!!!!..238 Literature cited...!!..!!!!!!!!!!!!!!!!!!!!!!!!.243 xiii List of Tables Table 2.1. Strains and plasmids used in Chapter 2.............................................61 Table 2.2. Primers used in Chapter 2..................................................................65 Table 2.3. AHLs: Retention times, identification and relative abundances based on tandem mass spectrometry. ...........................................................................81 Table 2.4. QS regulator expression in KLH11 null mutants. ...............................83 Table 2.5. Expression of KLH11 PssaI and PssbI promoters in an AHL- host ........85 Table 2.6. Cross-regulation experiments for SsaR and SsbR in an AHL- host. ...86 Table 3.1. Strains and plasmids used in Chapter 3...........................................116 Table 3.2 .SsaR-dependent activation of ssaI was stimulated by SscI-derived AHL....................................................................................................................127 Table 4.1. Strains and plasmids used in Chapter 4...........................................150 Table 4.2. Primers used in Chapter 4................................................................152 Table 4.3. Quantification of motility-related gene expression by qRT-PCR. .....162 Table 4.4. Quantification of ftsZ and ccrM expression by qRT-PCR.................162 Table 4.5. Regulation of cckA by ctrA. ..............................................................164 Table 4.6. Complementation of PcckA, PctrA expression by exogenous AHL.......171 Table 4.7. Expression of KLH11 PcckA, PchpT and PctrA promoters in an AHL- host. ...........................................................................................................................173 xiv List of Figures Figure 1.1. Schematic display of a sponge body structure...................................3 Figure 1.2. Phyogenetic distribution of sponge-associated bacteria. ....................8 Figure 1.3. A general mode of QS in Proteobacteria. .........................................14 Figure 1.4. Organization and mechanism of two-component systems................39 Figure 2.1. RP-TLC analysis of AHLs from KLH11 and QS mutants. .................70 Figure 2.2. Chemical analysis of purified samples from KLH11 derivatives........71 Figure 2.3. Comparative analysis of AHLs. ........................................................72 Figure 2.4. Gene maps of KLH11 ssaR/ssaI and ssbR/ssbI loci.........................74 Figure 2.5. Mass spectrometric analysis of plasmid-expressed SsaI and SsbI- directed AHLs from E. coli and KLH11. ...............................................................79 Figure 2.6. Sequence alignment of AHL synthases. ...........................................80 Figure 2.7. Activation of ssaI in response to synthetic AHLs. .............................87 Figure 2.8. Deletion analysis of the ssaI promoter. .............................................90 Figure 2.9. Regulation of swimming motility and flagellar biosynthesis by the SsaRI QS system. ...............................................................................................94 Figure 2.10. Growth curves of KLH11 and different QS mutants. .......................95 Figure 2.11. Flagellar staining of KLH11 quorum sensing mutants.....................96 Figure 2.12. Increased biofilm formation in ssaI and ssaR mutants....................98 Figure 2.13. A model for the complex regulatory control of QS circuits in KLH11. ...........................................................................................................................107 Figure 3.1. Gene maps of KLH11 sscI locus. ....................................................119 xv Figure 3.2. Mass spectrometry analysis of purified samples from triple mutant of #ssaI #ssbI sscI? (A) and from sscI expressed in E. coli MC4100(B). .............121 Figure 3.3. Alignment of SscI amino acid sequences to other AHL synthases. 123 Figure 3.4. "-galactosidase activity of lacZ transcriptional fusion with sscI in different mutant backgrounds. ...........................................................................125 Figure 3.5 sscI affects KLH11 swimming motility moderately. ..........................128 Figure 3.6. "-galactosidase assay of the expression of rpaI-lacZ fusion. .........130 Figure 4.1. The cckA-chpT-ctrA pathway controls motility ................................155 Figure 4.2. Alignment of KLH11 CtrA amino acid sequence to selected CtrA homologues. ......................................................................................................157 Figure 4.3. Growth curves of wild-type KLH11 (EC1) and derivatives. .............159 Figure 4.4. Detection of flagella and flagellin. ...................................................160 Figure 4.5. Comparative analysis of KLH11 CtrA ............................................166 Figure 4.6. Regulation of cckA, chpT and ctrA gene expression by the ssaRI system. ..............................................................................................................170 Figure 4.7. The cckA-chpT-ctrA phosphorely system is required for the ssaRI system to control motility ...................................................................................175 Figure 4.8. Suppression of motility defects in #ssaI and #ssaR mutants by CtrA. ...........................................................................................................................176 Figure 4.9.The cckA and chpT genes are required for the function of CtrA ......177 Figure 4.10. A tentative model for the ssaRI to cckA-chpT-ctrA regulatory circuit to control KLH11 flagellar motility. .....................................................................185 xvi Figure 5.1. Phylogenetic tree using neighbor-joining method of 16S rRNA genes (ca. 1300 bp) from SR AHL-producing bacteria. ...............................................205 Figure 5.2. Phylogenetic tree using neighbor-joining method of 16S rRNA genes (ca. 700 bp) from all the non-SR AHL producing bacteria.................................207 .Figure 5.3 Phylogenetic tree using neighbor-joining method based on the predicted 96 aa residues encoded by luxS genes from Vibrio isolates. ............209 Figure 5.4. AI-2 activities of all Vibrio isolates from sponges detected by reporter strainV. harveyi TL-26. ......................................................................................210 Figure 5.5. Detection of ssaI gene expression and AHLs in sponge tissue. .....212 Figure 5.6. Bacterial colonization of M. laxissima sponge cells at different timepoints after inoculation (0 h, 6 h, and 24 h). ...............................................214 Figure 5.7. Bacterial colonization of M. laxissima whole sponges. ...................216 1 Chapter 1. Introduction and literature review 2 1.1. Marine sponges 1.1.1 Sponge body structure and reproduction Sponges belong to the phylum Porifera and are considered to be the most ancient clade of the metazoans, with a fossil record dating back about 580 million years (Vacelet and Donadey, 1977; Taylor et al., 2007). Sponges are the morphologically simplest metazoans and have very simple body structures. In general, sponges have three distinct layers in their body plans as shown in Figure 1: the outer layer is the pinacoderm, the inner layer is the choanoderm, consisting of flagellated chambers and the middle region is the mesohyl (Hentschel et al., 2012). Sponges are filter feeders and they obtain their nutrition by filtering the surrounding water through small holes called ostia on their surface (Fig. 1). After passage through flagellated chambers, the filtered water is almost sterile on discharge and is pumped out through the main water channel termed the osculum. The size, number and shape of the ostia vary from one species to another (Taylor et al., 2007). 3 Figure 1.1. Schematic display of a sponge body structure. Adapted from Taylor et al. (2007) with permission. The arrows indicate the direction of the water flow. Sponges are able to reproduce asexually or sexually (Levi, 1957). There are generally three different mechanisms of asexual reproduction: 1) Budding. A set of cells can form on the external portion of the sponge body. This set of cells can develop into a complete sponge and breaks off from the original sponge and eventually becomes an independent individual. 2) Gemmule formation. This process is similar to budding but differs in that the gemmule is formed within the sponge and has a protective layer around it, which can help it survive even in harsh environments. 3) Regeneration. This refers to the process in which a sponge is broken into different pieces that are able to develop into new sponges. Sponges have both female and male reproductive organs, but they very rarely self-fertilize. A sponge produces sperm in the mesohyl layer. The sperm will 4 develop to the gamete stage before they are pumped into the surrounding water through the sponge osculum. Water currents can bring these gametes into the neighboring sponges through ostia where the sperm is able to contact the eggs produced by a different individual. The fertilized eggs can be either released immediately or undergo limited development in the mother sponge, to be relased as larvae into the surrounding water, which can settle onto the solid surface and proceed to develop into independent sponges (Bergquist, 1978). 1.1.2 Sponge taxonomy Sponges have a wide global distribution, can be found in almost all aquatic habitats and often are the dominant members of benthic communities (Taylor et al., 2007; Hentschel et al., 2012). There are about 8,500 described species in the phylum Porifera with an estimated total species number of 15,000 (Hooper and van Soest, 2002; van Soest et al., 2012). The majority of sponges are from the oceans although some sponge species live in fresh water. The phylum Porifera consists of four classes: Demospongiae, Homoscleromorpha, Hexactinellida and Calcarea. The class Demospongiae is the largest and it has 12 orders and roughly 7,000 species (about 83% of all the sponge species) (van Soest et al., 2012). The class Homoscleromorpha has the smallest number of representatives with only one order and 87 species (roughly 1% of all the sponge species). Members of the class Calcarea have spicules made of calcium carbonate in the form of calcite or aragonite while members of Hexactinellida are often referred to as glass sponges because they have a skeleton made of four to six-pointed siliceous spicules. An online database, World Porifera Database 5 (WPD) (http://www.marinespecies.org), categorizes the known sponge species and is organized in a fully hierarchical system to promote the stability of sponge names, which provides a useful resource for sponge taxonomy (van Soest et al., 2012). Traditionally, sponge taxonomy relies on characterizing the skeleton, mainly focusing on the size, form and location of spicules. However, this is challenging because of the high phenotypic plasticity due to the lack of basic organ and tissue differentiation (van Soest et al., 2012). Recently, identification of sponges by molecular methods has proved to be very useful in sponge taxonomy. The Sponge Barcoding Project (SBP) was initiated aiming to cover all sponge taxa, including the four classes described above and also sponges from both marine and freshwater environments. Phylogenetic markers such as the 28S rRNA gene, the mitochondrial cytochrome oxidase subunit 1 (CO1) gene and flanking regions can serve as suitable tools for identifying sponges. However, there are limitations when applying DNA barcoding to taxonomically understudied groups and this approach can fail to reveal new species that may clearly be morphologically different especially if only a single gene is used in the phylogenetic analysis (Meyer and Paulay, 2005; Hickerson et al., 2006). 1.1.3 Genomes of marine sponges To better understand the evolution and physiology of marine sponges, a group of scientists chose the Great Barrier Reef demosponge Amphimedon queenslandica as the model sponge for which to obtain a draft genome sequence (Srivastava et al., 2010). They elegantly used the embryos and larvae 6 from a single mother for genomic DNA extraction to minimize DNA contamination of microbial origin from the microbes associated with adult sponges. Sanger shotgun sequencing was utilized to obtain about 9-fold coverage of the genome. Their results predict that the Amphimedon queenslandica genome encodes about 30,000 coding sequences (CDS), similar to that of the human genome. About 18,000 of these CDS have identifiable homologues in other organisms. Overall, the genome shows remarkable similarity in terms of content, structure and organization to other animal genomes (Srivastava et al., 2010). With genome sequence available, now it is possible to answer some fundamental questions such as what makes an animal, including some of the genetic characteristics of the last common ancestor of multicellular life (Srivastava et al., 2010). It is technically challenging to sequence marine sponge genomes, because marine sponges harbor many millions of microbes in their tissues. It is critical to avoid microbial contamination in order to produce accurate sponge genome sequences. 1.2. Marine sponge microbiology Marine sponges can harbor highly diverse and abundant microbial communities and in some cases up to 30%-40% of the sponge biomass is derived from these microbes, including bacteria, archaea and single cell eukaryotes (Vacelet, 1975; Webster and Taylor, 2012). Sponges can be categorized into either High Microbial Abundance (HMA) or Low Microbial Abundance (LMA) sponges based on the abundance of associated microorganisms (Hentschel et al., 2006). In the early days, researchers used 7 electron microscopy (EM) to study the associated microorganisms. For example, Vacelet and Donadey (1977) observed a high abundance of microorganisms in sponge tissues by EM. With the advent of molecular techniques, such as PCR, Denaturing Gel Gradient Electrophoresis (DGGE) and 16S rRNA gene analysis, as well as next-generation high-throughput sequencing, the amazing diversity of bacteria associated with marine sponges is being revealed. A relatively comprehensive survey of the available 16S rRNA gene sequences deposited in GenBank that were obtained from bacteria associated with marine sponges conducted (roughly 11,000) by Webster and Taylor (2012) shows bacterial sequences mainly belong to 16 different phyla as shown in Figure 1.2. The top three most abundant phyla are Proteobacteria, Actinobacteria and Chloroflexi. It is certain that more and more sequences belonging to different phyla will be recovered as the bacterial communities associated with more and more sponges are analyzed by high?throughout sequencing. Of note, the Department of Energy of the United States currently is funding a collaborative project named the Earth Microbiome Project (EMP) (http://www.earthmicrobiome.org/), which aims to characterize the global microbial taxonomic and functional diversity that are likely beneficial to the planet and humans and a large proportion of the sequencing efforts will be directed towards microbial communities associated with marine sponges. 8 Figure 1.2. Phyogenetic distribution of sponge-associated bacteria. Adapted from Webster and Taylor (2012) with permission. A total of 11,284 16S rRNA gene sequences were obtained from GenBank in September 2010. Bacterial phyla that have fewer than 10 sequences obtained from sponges were excluded for clarity, including: Chlamydiae, Deferribacteres, Deinococcus-Thermus, Epsilonproteobacteria, Fusobacteria, Lentisphaerae, OP11, Tenericutes, and WS3. Hentschel et al. (2002) first defined sponge-specific bacterial clusters: ?the sequences of each cluster are more closely related to each other than to a sequence from non-sponge sources?. Many different sponge-specific clusters have been reported since then. A candidate phylum Poribacteria, in which members showed <75% identity on the16S rRNA gene sequence to known bacteria phyla and were found only in sponges was proposed (Fieseler et al., 2004). Thus, the existence of sponge-specific bacteria became a paradigm in 9 this field (Taylor, et al., 2007). However, with the development of next- generation sequencing techniques, deep sequencing enabled a better understanding of these sponge-specific clusters. Webster et al. (2010) first reported the recovery of sequences corresponding to sponge-specific clusters in samples from surrounding seawater at extremely low abundance by using 454- pyrosequencing methods. Taylor et al. (2012) conducted a comprehensive study in which 173 previously described sponge specific clusters were compared to over 12 million different 16S rRNA sequences from different ecological environments and results show that 77 of these clusters (ca. 44%) can be recovered from seawater or sediments although in general at very low abundance, suggesting the role of a ?seed bank? of bacteria present at low abundance in the surrounding environment for colonizing the sponges and thus providing strong evidence for horizontal acquisition of these sponge symbionts. Furthermore, the candidate phylum Poribacteria can also be found in a variety of non-sponge environments (Taylor et al., 2012). Taken together, it is questionable if sponge-specific clusters indeed exist; it seems more likely that this paradigm reflected the inadequacy of earlier sequencing approaches to detect bacteria present in the surrounding environment at very low abundance. One of the central questions in sponge microbiology is how the sponges acquire sponge-specific symbionts (Taylor et al., 2007). Two modes of bacterial transmission mechanisms have been proposed: vertical and horizontal. Vertical transmission refers to the transfer of these symbionts between generations through larvae. The evidence for vertical transmission of sponge-associated 10 bacteria first came from electron microscopy studies (de Caralt et al., 2007). Many subsequent studies utilized fluorescence in situ hybridization (FISH) to examine the distribution of bacteria in sponge larvae and found strong evidence of vertical transmission (Enticknap et al., 2006; Sharp et al., 2007). In addition, studies that compared the bacterial compositions between the adult sponges and the larvae or embryos showed overlap in the bacterial communities in these two life stages, suggesting vertical transmission (Schmitt et al., 2008; Lee et al., 2009). Horizontal transmission refers to the acquisition of bacteria from the surrounding seawater or other non-sponge environments. Many of the previously characterized sponge-specific clusters now have been found in the surrounding seawater or even sediments at extremely low abundance (Taylor et al., 2012), which strongly suggest that these bacteria can be acquired horizontally. Even for the bacteria for which vertical transmission has been experimentally shown, it is difficult to exclude the possibility that they can also be horizontally recruited. A central question in the field of sponge microbiology is ?What are the functions of the bacterial symbionts associated with sponges?? The vast majority of microorganisms in the water that pass through sponges are trapped and digested and the water pumped out from the main exhalent water channel, the osculum, is nearly sterile (Taylor et al., 2007). Symbiotic relationships between sponges and microorganisms are considered to contribute to the health and nutrition of sponge. However, little evidence exists showing the contribution of symbiotic microbes to sponge health or survival (Webster and Blackall, 2009). 11 Exceptions are the translocation of photosynthate from cyanobacteria to the host sponge and a decrease in the health status of sponge associated with loss of cyanobacteria (Wilkinson and Fay, 1979). It is worth pointing out that much circumstantial evidence points toward the significance of the symbionts to the sponge heath (Webster and Blackall, 2009). One good example is the role of symbionts in nitrogen fixation, a process whereby microorganisms convert atmospheric dinitrogen gas into biologically usable ammonium. Sponges that live in the nutrient-poor coral reef environments are a reasonable ecological niche for nitrogen-fixing bacteria because there could be an unbalanced carbon and nitrogen ratio, which is caused by photosynthetically derived carbohydrates provided to sponges since these compounds are rich in carbon but scarce in nitrogen (Wilkinson, 1983). Mohamed et al. (2008a) determined the $15N values in two shallow water marine sponges: Mycale laxissima and Ircinia strobilina, and found a low value in I. strobilina, consistent with this sponge obtaining a substantial part of its fixed nitrogen through biological nitrogen fixation. Furthermore, the expression of nifH genes was examined in these two sponges by using reverse transcription- PCR (RT-PCR) and results showed that the nifH genes exclusively from cyanobacteria are actively expressed (Mohamed et al., 2008a). Sponges are sessile organisms and need effective defense mechanisms against predation, fouling and disease and thus it is possible that the associated microbial community may be involved in host defense (Taylor et al., 2007). For example, Hentschel et al. (2001) found that 27 out of 238 bacterial strains 12 isolated from the Mediterranean sponges Aplysina aerophoba and Aplysina cavernicola showed antibacterial activities and they also observed an interesting pattern: Gram-positive bacteria tended to show inhibitory effects on Gram- positive isolates whereas Gram-negative bacteria inhibited Gram-negative strains. Thakur et al. (2004) found that strains from the genus Bacillus with a close association with the host sponge Ircinia fusca, showed antimicrobial activity against fouling bacteria. It is clear that our knowledge about the diversity of the microbes that live together with marine sponges is improving at a rapid pace; however our understanding of the functions of these bacteria and the nature of the symbioses is lagging behind. Major efforts need to be directed towards these aspects of bacteria-sponge symbiosis to further advance the development of this field. 1.3. Quorum sensing Bacteria are unicellular organisms and they can exist in almost all ecological niches and play vital roles in the ecosystem. However, bacteria can also coordinate their group behavior through a process termed quorum sensing (QS), which enables bacteria to sense and perceive their population density through the use of diffusible signals (Fuqua et al., 1994). About 40 years ago, seminal research on the control of bioluminescence production in Vibrio fischeri (previously named Photobacterium fischeri) found that bacteria can coordinate their group behavior by communicating with each other via diffusible chemical molecules (Nealson et al., 1970). It was unclear what the molecule was at that time. In the 1980s, the chemical was discovered to be N-3-oxohexanoyl-L- 13 homoserine lactone (3-oxo-C6-HSL) (Eberhard et al., 1981). Generally, the signal molecules are called N-acylhomoserine lactones (AHLs). The enzyme that is responsible for the synthesis of 3-oxo-C6-HSL in V. fischeri is called LuxI and the receptor that can respond to this molecule is called LuxR. Since its discovery, the LuxI-LuxR type QS pathway has been reported in more than 100 different bacterial species (Ahlgren et al., 2011), the vast majority of which belong to the ecologically diverse and abundant Proteobacteria. In the following sections, I review the progress of our understanding of bacterial quorum sensing. 1.3.1 General model of QS in Proteobacteria The canonical LuxI-LuxR QS pathway was first dissected in the marine animal symbiont V. fischeri, a gammaproteobacterium (Engebrecht et al., 1983). In the simplest model, the gene called luxI encodes the enzyme that is responsible for the synthesis of AHL molecules. The reason that the gene is called luxI is because in V. fischeri the QS pathway controls the luminescence (lux) system (Fuqua et al., 2001). At low cell density, the concentration of the AHL is low and it can quickly diffuse across the cell membranes. As the cell density increases, the AHL molecules accumulate. Once the cell density and thus the concentration of the AHL reach a certain threshold, the AHL diffuses back into the cell and bind to its cognate receptor LuxR. The complex of LuxR and AHL turns on or turns off a certain set of genes and thus coordinates the group behavior (Fig.1.3) (Miller and Bassler, 2001). The set of phenotypes that are typically controlled by this process includes motility, biofilm formation, 14 bioluminescence, and production of antibiotics or secreted products such as virulence factors (Fuqua and Greenberg, 2002). Figure 1.3. A general mode of QS in Proteobacteria. The I gene represents the luxI homologue. R represents the AHL receptor LuxR protein. The light blue solid circle represents the LuxI enzyme while the dark blue dots represent the AHL molecules. Stalked arrows indicate the transcription of the genes and the dotted line with arrow shows the positive feedback by the complex of the LuxR receptor and AHLs on the AHL synthase gene. The solid curved line with arrow depicts the function of R complex on the target genes. The left corner shows the basic structure of a typical AHL molecule. N can vary from 4 to 18. R can be oxo, OH or H. The figure is kindly provided by Dr. Clay Fuqua. The majority of AHL synthases that have been discovered so far belong to the LuxI family, although a few other types of enzymes that can synthesize AHL have been reported in different species, such as LuxM, VanM and AinS in a few Vibrio 15 species (Churchill and Chen, 2011). The focus of this section is on the LuxI-type enzyme. The LuxM will be reviewed in Section 1.3.6 together with LuxS, the autoinducer-2 synthase since both of these two enzymes are best characterized in Vibrio species (Ng and Bassler, 2009). The substrates for LuxI-type enzyme are S-adenosylmethionine (SAM) and fatty acyl precursors conjugated to the acyl carrier protein (ACP). The methionine in SAM provides the homoserine moiety, which is conjugated to the acyl chain donated by acyl-ACP (Mor? et al., 1996; Schaefer et al., 1996). It is worth mentioning that in a few species LuxI-type enzymes use CoA- rather than ACP-linked substrates for acyl-HSL synthesis (Lindemann et al., 2011). The LuxI-type enzyme is roughly 180-230 amino acids (aa) in length with some exceptions (such as the Sil1 in R. pomeroyi DSS-3 and SsaI in Ruegeria sp KLH11 that are both about 280 aa in length) (Moran et al., 2004; Zan et al., 2012). The LuxI-type protein can be divided into two regions: the N-terminal region and the C-terminal region. The N-terminal region, responsible for the binding of the converted substrate SAM and catalysis, is the most conserved part of the enzyme and has eight invariable residues, Arg24, Phe29, Trp35, Asp46, Asp49, Arg70, Glu100 and Arg103 (the numbering is based on the LuxI in V. fischeri) (Churchill and Chen, 2011). In contrast, there is very low conservation in the C-terminal region, which is involved in the recognition of the acyl chain, the highly variable part of the acyl-ACP substrate (Churchill and Chen, 2011). In most cases, the luxI gene is genetically linked with a cognate luxR gene, encoding the AHL receptor that is present in the cytoplasm (Subramoni and 16 Venturi, 2009). However, not all AHL receptors are in the cytoplasm. The exception is the membrane bound AHL receptor LuxN in V. harveyi, which is discussed in Section 1.3.6. Typically, LuxR-type AHL receptors have two domains. The N-terminal domain contains the AHL-binding sites and the C- terminal domain encodes the DNA binding motifs. These two domains are linked together via a conformationally flexible linker. The N-terminal domain inhibits the function of the DNA-binding domain when the receptor is in the apo-form, free of AHL (Schuster and Greenberg, 2008). Binding to the cognate AHL can relieve this inhibition and enable the LuxR receptor to multimerize, which has been shown in several different bacterial species (Choi and Greenberg, 1992; Zhu and Winans, 2001; Kiratisin et al., 2002). Binding to AHL can also enhance the resistance to protease-mediated degradation of LuxR (Zhu et al., 2001). The binding of AHL to its cognate receptor is a reversible process for LuxR in V. fischeri while it is an irreversible process for LasR in Pseudomonas aeruginosa or for TraR in Agrobacterium tumefaciens. The C-terminal domain of LuxR-type proteins has a Helix-Turn-Helix motif (HTH) and is able to recognize a conserved palindromic sequence element upstream of target genes (Stevens and Greenberg, 1999). This DNA element is 18-22 bp in length and is named the lux- type box, commonly found close to the regulated promoters. These sequences share considerable similarity with the LuxR target sequences upstream of the V. fischeri lux operon (the 20 bp sequence of which is ACCTGTAGGATCGTACAGGT), which is centered 42.5 bp upstream from the luxI transcriptional start site and -61.5 bp upstream of the luxI translational start 17 site (Devine et al., 1989; Egland and Greenberg, 1999). Both half sites of the box are required for LuxR activation and the position of the box is also critical (Egland and Greenberg, 1999). A later study showed that six critical nucleotides define the minimal sequences of a lux box (Antunes et al., 2008). In the plant- pathogen A. tumefaciens, the AHL receptor TraR is a transcriptional activator and regulates the conjugal transfer of the octopine type Ti plasmid at high cell density (Hwang et al., 1995). In total, there are seven TraR-dependent promoters and four tra boxes that have been identified (Fuqua et al., 1994; Fuqua and Winans, 1996). These four tra boxes are 18 bp in length and share high sequence similarity. TraR regulates its own expression but does not contain any recognizable tra boxes in its promoter region (Fuqua and Winans, 1996). Several members of this LuxR family are able to fold, dimerize, bind to DNA, and regulate transcription in the absence of AHLs and are antagonized by their cognate AHLs (Tsai and Winans, 2010). AHLs disrupt the complexes between some of these proteins and their DNA binding sites in vitro (Tsai and Winans, 2010). All such proteins are fairly closely related and form a monophyletic clade. The best-characterized representative is the EsaR from the plant pathogen Pantoea stewartii (formerly Erwinia stewartii). This QS system regulates exopolysaccharide (EPS) production and EsaR and EsaI mutants display completely opposite regulatory effects on EPS production. The EsaR mutant overexpressed the EPS production whereas the EsaI mutant decreased its production (von Bodman and Farrand, 1995; von Bodman et al., 1998). It was later discovered that apo-EsaR is able to bind to the promoter of the EPS 18 synthesis gene rcsA to repress its expression (Minogue et al., 2005). At high cell density, the AHL disrupts this binding and thus derepresses the rcsA expression. It is unclear whether or not the active apo-EsaR can bind to AHL. Interestingly, the mechanisms of how AHLs inactivate this type of LuxR proteins are unclear (Tsai and Winans, 2010). 1.3.2 Case studies of QS: V. fischeri and P. aeruginosa 1.3.2.1 V. fischeri V. fischeri ES114, is the light organ symbiont of the Hawaiian bobtail squid, Euprymna scolopes, which can be maintained and reproduced under laboratory conditions (Stabb et al., 2008). The luxI and luxR genes are adjacently arranged but transcribed separately. luxI is the first gene in the lux operon which contains another six genes (luxCDABEG) and LuxI synthesizes 3-oxo-C6-HSL. The LuxR protein complexed with 3-oxo-C6-HSL binds to the lux box described in Section 1.3.1 to create a positive feedback on the expression of the luxI gene. V. fischeri also has a luxS gene that can produce AI-2 and a homologue to luxM in V. harveyi described below in Section 1.3.6: the ainS gene that can produce 3- octanoyl-HSL (C8-HSL) (Kuo et al., 1994; Gilson et al., 1995; Hanzelka et al., 1999). For the symbiosis, the squid selectively acquires V. fischeri horizontally from the surrounding water and the light production follows a diel cycle. At night, the V. fischeri grows to high cell density and turns on the lux operon resulting in light production within the light organ that harbors the symbiotic bacteria, which can benefit the animal by matching the intensity of the moon above to protect 19 them from predators by a process termed counter-illumination camouflage (McFall-Ngai and Ruby, 1991). At dawn, the squid releases ~ 95% of the bacteria from the light organ to maintain a low cell density. The squid hide themselves in the sand during the day, a situation in which light production is not required. The remaining bacteria grow during the day to reach a high cell density again in the mid-afternoon. The pattern repeats daily (Nyholm and McFall-Ngai, 2004). Paradoxically, the LuxI signal plays a critical role in the symbiosis while it has little effect on luminescence in culture, demonstrating the fact that artifacts under laboratory conditions can seriously obscure our views. 1.3.2.2 P. aeruginosa P. aeruginosa is a ubiquitous and versatile gammaproteobacterium. It is an opportunistic pathogen and the dominant microorganism in chronic lung infection of cystic fibrosis patient (Lyczak et al., 2000). There are two sets of luxI-luxR pathways in P. aeruginosa: the lasIR and the rhlIR systems. LasI produces 3- oxo-dodecanoyl-HSL (3-oxo-C12-HSL) and RhlI produces 3-butyryl-HSL (C4- HSL), which binds to their cognate LuxR type receptors LasR and RhlR, respectively (Pearson et al., 1994; Pearson et al., 1995). There is also a third distinct QS system in P. aeruginosa that involves the regulator MvfR and the Pseudomonas quinolone signal (PQS) (Mashburn and Whiteley, 2005). Both of the first two LuxI-LuxR systems have the typical positive feedback on the expression of the AHL synthase gene via the cognate receptor with AHLs. This positive feedback can lead to dramatic increase of the expression of the genes regulated by QS, such as the virulence genes, once a critical threshold of the 20 cell-density is reached (Pesci et al., 1997). A transcriptomic study showed that the QS systems control ca. 6% of the total P. aeruginosa genes, including genes that encode the main virulence factors and antibiotics, such as rhamnolipid, elastase and pyocyanin (Schuster et al., 2003). It is generally considered that there is a regulatory hierarchy existing in this QS system. LasR initiates the regulatory system and then subsequently activates the transcription of rhlR and 3-oxo-C12-HSL can also block C4-HSL from binding to RhlR (Latifi et al., 1996; Pearson et al., 1997; Pesci et al., 1997). However, results from a series of studies have challenged this view. Dekimpe and D?ziel (2009) show that the activity of the rhl system is only delayed in lasR mutants into stationary phase and rhlR still shows significant expression in the lasR mutant although at lower levels compared to that in wild type culture. Furthermore, they show that the rhI system can also affect the factors that were previously thought to be controlled solely by the las system, including the lasI gene. Limmer et al. (2009) showed that RhlR plays a more complex role in pathogenesis than previously thought and RhlR and LasR mutants show opposite phenotypes in pathogenicity in a Drosophila melanogaster infection model, suggesting the absence of the positive regulatory hierarchy previously proposed (Pesci et al., 1997). However, it is still possible that under certain environmental conditions the regulatory hierarchy still plays a critical role in P. aeruginosa pathogenicity. Besides the two sets of LuxI-LuxR QS circuits, P. aeruginosa also has a third LuxR homologue, named QscR. QscR is not linked to any luxI homologue, thus 21 is called a LuxR solo (Fuqua, 2006). QscR can respond to LasI derived 3-oxo- C12-HSL and regulates a set of genes distinct from those regulated by lasIR- rhlIR systems (Lequette et al., 2006). LuxR solos have subsequently been very commonly found and are present in many different bacterial genomes, some of which even have multiple LuxR solos (Subramoni and Venturi, 2009). Thus far, investigations suggest that these LuxR solos can have diverse roles in intra- and inter-species communication and some may even be involved in inter-domain singaling (Subramoni and Venturi, 2009). 1.3.3 Quorum sensing in Roseobacters The Roseobacter clade is a group of bacteria that belong to the Alphaproteobacteria and its members share greater than 89% 16S rRNA gene sequence identity (Buchan et al., 2005). It includes about 17 different genera (Buchan et al., 2005; Wagner-D?bler et al., 2005). Members of this clade have a diverse and broad ecological distribution but are exclusively restricted to marine or hypersaline environments (Buchan et al., 2005, Moran et al., 2007). Generally, roseobacters have been characterized as ecological generalists and exhibit different lifestyle strategies, including heterotrophy, photoheterotrophy and autotrophy (Moran et al., 2004; Moran et al., 2007). Roseobacters are numerically abundant and these bacteria are estimated to account for about 20- 30% of the bacterial 16S rRNA genes in ocean surface waters. Furthermore, many members are found to be associated with marine invertebrates (such as marine sponges or corals), marine algae, dinoflagellates and seagrasses (Taylor et al., 2004; Mohamed et al., 2008c; Slightom and Buchan, 2009). Ruegeria 22 pomeroyi DSS-3 (previously known as Silicibacter pomeroyi DSS-3) was the first member in this clade to have its genome sequenced (Moran et al., 2004) and currently 42 bacterial strains from this group have been sequenced (see www.roseobase.org for details). The number of genome sequences available for Roseobacter strains as well as other bacteria is increasing rapidly as the price of sequencing is plummeting. An important ecological function of members in this clade is in global sulfur cycling through metabolizing dimethylsulfoniopropionate (DMSP), an organic sulfur compound produced by marine phytoplankton (Moran et al., 2012). The pathways and enzymes participating in the demethylation or cleavage pathways are comprehensively reviewed in Moran et al. (2012). Marine environments that contain high nutrient concentration would be reasonably expected to host bacteria that can produce AHLs as bacteria could reach sufficiently high cell density for quorum sensing. These environments include marine snow, marine invertebrates and marine algae. Gram et al. (2002) first screened bacteria isolated from marine snow for AHL production and showed that three roseobacterial isolates from marine snow are able to produce AHL molecules detected by an A. tumefaciens AHL biological reporter system. Marine snow is made up of organic and inorganic particles and is rich in energy and nutrients. Bacteria can colonize marine snow and produce exoenzymes to utilize the nutrient sources available in the aggregated organic particles (Gram et al., 2002). Taylor et al. (2004) found a bacterial strain from the genus Ruegeria isolated from a marine sponge that produced AHLs detected by reporter systems utilizing Chromobacterium violaceum CV026 and A. tumefaciens A136. In the 23 same study, organic extracts from 27 of 37 different marine invertebrates including marine sponges, corals, ascidians and bryozoans were showed to stimulate the short-chain AHL reporter (CV026), indicating the presence of short- chain AHLs in these invertebrates. The AHLs responsible for induction of the reporter systems in these studies by Gram et al. (2002) and Taylor et al. (2004) were not characterized. Wagner-D?bler et al. (2005) screened 102 marine bacterial strains isolated from different marine environments for AHL production and found that 33 roseobacterial strains were able to significantly increase the fluorescence of their AHL reporter system, suggesting AHL production. The majority of these AHL- producing roseobacterial strains were isolated from either marine dinoflagellates or picoplankton. Mohamed et al. (2008c) found that isolates from the Silicibacter- Ruegeria (SR) subgroup of the Roseobacter clade are the dominant AHL producers from the cultivated isolates from two marine sponges, M. laxissima and I. strobilina. These marine sponge roseobacterial symbionts produce a mixture of short and long-chain length AHLs revealed by Thin-Layer Chromatography (TLC) coupled with an AHL biological reporter assay. However, the AHL profile of Ruegeria sp. KLH11 chosen as a representative of this group was analyzed in detail by mass spectrometry (MS) and the AHLs were predominantly 12-16 carbons whereas the short chain length AHLs detected in bioassyas were below the limits of sensitivity of the MS approach (Zan et al., 2012). 24 Among the 42 sequenced roseobacterial genomes deposited in the Roseobase (www.roseobase.org), a genomic resource for marine roseobacters, 33 of them encode luxR-luxI type homologues, including species from almost all of the 17 genera in the Roseobacter clade. Of note, the Hill laboratory has recently sequenced 17 SR bacterial strains isolated from marine sponges that can produce AHLs (Zan, Hill, Fuqua and Ravel, work in progress) detected by the A. tumefaciens KYC55 reporter (Zhu et al., 2003). Once the annotation of these genomes is finished, substantial additional information on the LuxI-LuxR type QS pathways in these sponge symbionts will become available. In contrast to the growing body of knowledge about the genomes of these AHL producers, very few studies have been performed to examine the physiological and genetic characteristics of the QS pathway in these species. Efforts were first directed towards the free-living marine bacterium R. pomeroyi DSS-3 (Moran et al., 2004). Its genome encodes two LuxI-LuxR pathways, designated as SilI1/SiIR1 and SilI2/SiIR2. Both of the AHL synthases were expressed in E. coli and the AHL profile of SilI1 is very similar to that of the wild type culture of R. pomeroyi DSS-3 whereas SilI2 synthesizes different AHLs (Moran et al., 2004). Furthermore, the SilR1 increases the expression of the silI1 expression in response to addition of exogenous AHL when examined in A. tumefaciens NTL4 (E. Cicirelli and C. Fuqua, unpublished). However, it is unclear what phenotypes these QS pathways in R. pomeroyi DSS-3 can regulate. In recent years, novel non-AHL type QS molecules have been identified at a steady pace. For example, diffusible signal factor (DSF) cis-decenoic acid is a 25 QS molecule in Xanthomonas spp. (Ryan and Dow, 2008). Arguably, many antibiotics probably function as signaling molecules since the concentration for these molecules required to be effective as antibiotics has never been detected in nature (Surette and Davies, 2008). One good example of a novel non-AHL type QS molecule in Roseobacter is the sulfur-containing compound, tropodithietic acid (TDA) produced in Silicibacter sp. TM1040 discovered by the Belas laboratory (Geng et al., 2008). Silicibacter sp. TM1040 is an essential symbiont for its dinoflagellate host Pfiesteria piscicida (Miller and Belas, 2006). Surprisingly, Silicibacter sp. TM1040, closely related to R. pomeroyi DSS-3 and Ruegeria sp. KLH11, does not produce any detectable AHLs or encode any luxI or luxM homologues although it encodes four luxR homologues in the genome (Cicirelli et al., 2008), which might be able to respond to exogenous AHLs produced by other bacteria. Interestingly, its motility has been shown to be important in the beginning phase of the symbiosis. TM1040 forms a biofilm on the host surface when it associates with the host dinoflagellate (Miller and Belas, 2006). Using transposon mediated mutagenesis, 12 genes were identified as critical for TDA biosynthesis and six of these genes are located on a megaplasmid (Geng et al., 2008). TDA shows a growth phase-dependent production in static conditions with only very low levels of production under shaking conditions (Geng and Belas, 2010). Furthermore, an electrophoresis mobility shift assay showed that the LysR type regulator TdaA can bind to the promoter of one of the TDA synthesis genes tadC, within the same operon as tdaD and tdaE, although this binding is not stimulated by TDA in vitro (Geng and 26 Belas, 2011). However, addition of TDA to Silicibacter sp. TM1040 culture increases the expression of TDA synthesis genes most likely via TdaA (Geng et al., 2011). This type of positive feedback indeed resembles that of the expression of luxI homologues via LuxR protein complexed with AHLs. Furthermore, TDA-like molecules are produced by several different roseobacterial strains (Geng et al., 2010). Taken together, this provides convincing evidence supporting the role of TDA as a QS molecule. The involvement of TDA in the regulation of motility and biofilm formation and thus the symbiosis with the dinoflagellate warrants further investigation. Notably, the sponge symbiont Pseudovibrio sp. JE062, which is found associated with several different sponges (Enticknap et al., 2006) also produces TDA (Geng et al., 2010) and the involvement of TDA-mediated QS in the symbiosis of Roseobacters with sponge hosts is also an interesting area for further exploration. A regulatory link between LuxI-LuxR QS and TDA has been established in Phaeobacter gallaeciensis DSM 17395 (Berger et al., 2011). The luxI homologue pgaI in this strain synthesizes 3-hydroxydecanoyl-HSL (3-OH-C10- HSL) and the cognate luxR homologue is called pgaR. There seems to be a minor effect of PgaR on the expression of pgaI. Insertion mutants of pgaI and pgaR cannot synthesize TDA and also lose the production of a yellow-brown pigment. The essential transcriptional regulator for TDA biosynthesis TdaA is under positive transcriptional control by the pgaIR system, in contrast to the constitutive expression of TdaA in Silicibacter sp. TM1040. The TDA itself is an autoinducer in DSM 17395 because addition of TDA can restore the expression 27 of the TDA synthesis genes and the production of the pigment in the pgaI insertion mutant background. Furthermore, it was suggested that the function of TDA as an autoinducer requires the presence of the regulator PgaR (Berger et al., 2011). However, experimental evidence on the effect of TDA addition on the biosynthesis genes and pigment production in the pgaR mutant background is missing. Berger et al. (2011) hypothesize that the pgaIR QS system is located relatively high in the regulatory hierarchy and exerts its effects potentially through the regulator TdaA on the TDA synthesis genes and pigment production. However, it is unclear if pgaIR QS system directly or indirectly regulates the TDA system. 1.3.4 Novel AHL molecules The length of the fatty acid chains in AHLs varies from 4 carbons to 18 carbons, although chain lengths with odd numbers of carbons are less common. The oxidation status of the third carbon can vary from fully oxidized to fully reduced. Furthermore, some species have one or two double bonds somewhere in the chain. All these factors produce the diversity of AHL molecules that can be employed by different bacterial species (Fuqua and Greenberg, 2002). The Proteobacteria contains about 1,534 bacterial species (http://www.earthlife.net/prokaryotes/proteo.html). It is intuitive to hypothesize that AHLs with novel structures different from those described above are likely to exist in some species. The anoxygenic phototrophic soil bacterium Rhodopseudomonas palustris CGA009 gave the first surprise (Schaefer et al., 2008). Efforts to detect AHL activity by using traditional biological assays in this 28 species failed although a luxI-homologue rpaI is encoded in the genome. However, organic extracts of cultures of this strain grown in media with the substrate p-coumarate, a major plant aromatic monomer of lignin polymers, was able to trigger the rpaI-lacZ expression via RpaR, which can directly bind to the rpaI promoter to activate its expression (Hirakawa et al., 2011). The active molecule was identified as p-coumaroyl-HSL (pC-HSL) showing typical growth- phase dependent production (Schaefer et al., 2008). The pC-HSL like molecule can also be detected in Bradyrhizobium sp. and R. pomeroyi DSS-3 when grown in media supplemented with p-coumarate. The sponge symbiont Ruegeria sp. KLH11 has also been shown to be able to trigger RpaR-dependent expression of rpaI-lacZ. Notably, this activation is independent of the three LuxI homologues in the genome (See Chapter 3), indicating that a novel type of enzyme(s) is responsible for the synthesis. Phaeobacter gallaeciensis BS107 (also known as DSM 17395), a member of the Roseobacter clade, responds to the presence of p-coumarate produced by the microalga Emiliania huxleyi to control the production of algaecides and thus converts itself into an opportunistic pathogen of the host microalga (Seyedsayamdost et al., 2011). The pC-HSL like molecule in stem-nodulating photosynthetic Bradyrhizobium ORS278 was confirmed to be cinnamoyl-HSL, which lacks a hydroxyl group on the aromatic ring compared to pC-HSL. BraI synthesizes this molecule, although it does not require an exogenous source of cinnamate. BraI-BraR comprises the QS pathway in this strain (Algren et al., 2011). Interestingly, this strain produces a very low level of cinnamoyl-HSL in culture and BraR can respond to picomolar 29 concentrations, although BraR can also respond to other non-cognate AHL in the range of nanomolar to micromolar concentrations (Algren et al., 2011). In a soybean symbiont Bradyrhizobium japonicum USDA110, the LuxI homologue BjaI is closely related to RpaI and BraI that synthesize aryl-HSL. Surprisingly, BjaI does not synthesize aryl-HSL but rather isovaleryl-HSL (IV-HSL), a branched-chain fatty AHL. Its cognate receptor BjaR can also respond to picomolar concentrations of IV-HSL, similar to that of BraR (Lindemann et al., 2011). Overall, the discoveries of these three novel signaling molecules are important reminders of the diversity of molecules that can be synthesized by LuxI-type enzymes and also greatly extend the range of possibilities for AHL quorum sensing. Meanwhile it also raises the question of how many novel molecules have been overlooked in nature and reminds us that it is necessary to develop new reporter systems and new methods to detect potential novel signal molecules. 1.3.5 AHL-mediated QS in non-Proteobacteria 1.3.5.1 Gloeothece PCC6909 Gloeothece PCC6909 belongs to the phylum Cyanobacteria and can form biofilms and microcolonies. A protective layer of multilaminated sheath that forms around the colonies is considered to function as a diffusion barrier against toxic compounds and can also provide a conducive environment in which signal molecules can accumulate and thus trigger a quorum-sensing process (Meeks et al., 1978; Sharif et al., 2008). Using the AHL biological reporter strain A. 30 tumefaciens NTL4 (pZLR4) coupled with a TLC assay, Sharif et al. (2008) were able to show that this cyanobacterial strain produces an AHL-like molecule, which was confirmed to be N-octanoyl homoserine lactone (C8-HSL) by mass spectrometry. The production of the C8-AHL molecule displays a very typical pattern of autoinduction. Addition of this molecule to early-growth stage cells of Gloeothece PCC6909 can significantly affect about 15 proteins revealed by two- dimensional gel electrophoresis. These include the proteins RuBisCO, glutamate synthase, and chorismate synthase (Sharif et al., 2008). However, the molecular mechanism of the QS pathway is unknown. Whether this cyanobacterial strain contains the canonical LuxI-LuxR pathway remains elusive. This represents the first detailed characterization of AHL production in a non-Proteobacteria strain. 1.3.5.2. Cytophaga-Flavobacterium-Bacteroides Wagner-D?bler et al. (2005) first showed that a Flavobacterium sp. strain was able to trigger the response of an AHL-biological reporter strain although they were not able to identify the potential AHL-like molecule in this strain. Subsequently, the presence of AHL molecules was confirmed in two strains belonging to the genus Bacteroides, both of which contain a mixture of several different types of AHLs (Huang et al., 2008). Romero et al. (2010) reported the discovery of C4-HSL in the fish pathogen Tenacibaculum maritimum, a member of the Cytophaga-Flavobacterium-Bacteroides (CFB) phylum. Taken together with the finding of AHLs in the cyanobacterium Gloeothece, this suggests that AHL-mediated QS is not restricted only to the Proteobacteria. As more bacterial isolates from different phyla are screened, it is possible that an even broader 31 distribution of AHL-type QS will be found in a wide diversity of bacteria. However, genetic dissection of these QS pathways is still required to provide convincing data for the presence of LuxI-LuxR circuits in these non- Proteobacteria isolates. 1.3.5.3 AHLs in a methanogenic archaeon Until recently, AHL type QS pathways were thought to exist only in bacteria, more specifically in gram-negative bacteria, although a putative homoserine lactone-like molecule was previously detected in the haloalkaliphilic archaeon Natronococcus occultus (Paggi et al., 2003). A recent remarkable study by scientists from China showed the production of carboxylated AHLs by a LuxI homologue in the methanogenic archaeon Methanosaeta harundinacea 6Ac (Zhang et al., 2012). They observed a cell density dependent morphology change and then proceeded to probe the genome using bacterial luxI homologues. A CHASE 4 domain in the putative protein predicted to encode a multi-sensor signal transduction histidine kinase showed ca. 40% identity to the LuxI homologue AhlI in Erwinia chrysanthemi. The homologue was renamed as filI, showed a growth stage-dependent expression and was shown in vitro to be able to synthesize a new class of carboxylated AHLs. A few other methanogens likely also have filI homologues. The possible luxR homologue in this methanogen named as filR shows very low identity to bacterial luxR (Zhang et al., 2012). The QS mechanism of this novel fill-filR pathway deserves further investigation. 32 1.3.6 QS in V. harveyi and V. cholerae: LuxM, CqsS and LuxS- mediated pathways As described above, in a typical LuxI-LuxR QS pathway, LuxR is present in the cytoplasm. However, a separate class of membrane-bound AHL receptors has been identified in V. harveyi, a free-living marine bacterium but also an important pathogen of marine organisms (Austin and Zhang, 2006). The AHL in this case is synthesized by LuxM homologues. V. harveyi produces and responds to 3-OH-C4-HSL. The enzyme LuxM shows no homology to the LuxI- type enzyme but both can catalyze similar reactions (Bassler et al., 1993; Bassler et al., 1994). LuxM presumably uses SAM and acyl-ACPs or acyl-CoA substrate to synthesize the AHL while most LuxI homologues cannot use acyl-CoA as their substrate. LuxN is the cognate receptor of LuxM-derived AHLs (Freeman et al., 2000). It is a membrane bound protein with dual functions. At low-cell density, it does not bind to its AHL and functions as a histidine kinase while at high-cell density it can function as a phosphatase when it is bound to AHLs (Wei et al., 2012). The way that information is conveyed is the same as that of CqsS and LuxQ, receptors of CAI-1 and AI-2, and will be described in the following paragraph. In addition to AHL-based QS pathways, V. harveyi has another two sets of QS pathways: CAI-1 and Autoinducer-2 mediated, respectively. The CAI- 1 was first reported in V. cholerae, the etiological agent of the disease cholera, and proved to be 3-hydroxytridecan-4-one. The enzyme that synthesizes it is called CqsA and uses SAM and acyl-CoA as substrates. V. harveyi also has a 33 CqsA homologue. The cognate receptor for this type of molecule is the membrane-bound CqsS (Higgins et al., 2007; Wei et al., 2012). Autoinducer-2 (AI-2) is another well-known molecular cue and is generally considered to be an interspecies signal. Both Gram-positive and Gram-negative bacteria encode the luxS gene in their genomes. The molecular pathway has been extensively analyzed in V. harveyi and V. cholerae, where it is involved in regulation of bioluminescence and virulence-associated traits (Miller et al., 2002; Henke and Bassler, 2004; Lenz et al., 2004). The activated methyl cycle is a crucial metabolic pathway to recycle homocysteine from the major methyl donor S-adenosyl methionine. LuxS, a S-ribosylhomocysteinase, catalyzes part of the cycle and functions to convert S-ribosylhomocysteine to homocysteine; meanwhile, it can also, as a side reaction, synthesize 4,5-dihydroxy-2,3- pentanedione, the precursor of AI-2. 4,5-dihydroxy-2,3-pentanedione can spontaneously give rise to several furanone derivatives, collectively referred to as AI-2, which can freely diffuse across the cell membrane. The active AI-2 in V. harveyi requires the binding of metal boron, the only known case of a biological role for boron. However, in E. coli and Salmonella typhimurium, AI-2 signal does not require binding to boron to function probably due to limitation of this element in terrestrial environments (Ng and Bassler, 2009; references herein). The mechanisms for the three different pathways in V. harveyi and the two different pathways in V. cholerae are similar and involve a phosphoryl group transfer-based signal relay system. At low cell density, the membrane bound receptors LuxN, LuxPQ and CqsA for V. harveyi and LuxPQ and CqsA for V. 34 cholerae cannot bind to their cognate signal molecules and thus function as histidine kinases to phosphorylate the protein LuxU and then transfer the phosphate group to the protein LuxO. The phosphorylated LuxO activates transcription of genes encoding five small regulatory RNAs called Qrr1-5 in V. harveyi and four in V. cholerae called Qrr1-4. Together with the RNA chaperone Hfq, the Qrr1-5 can bind to the mRNA of the master quorum-sensing regulator LuxR in V. harveyi and Qrr1-4 can bind to HapR, a homologue of LuxR (not homologous to AHL-responsive LuxR) in V. cholerae, which destabilizes the LuxR /HapR mRNA transcript. The phosphate group flow from LuxU to LuxO is reversed at high cell density when these membrane bound receptors bind to their cognate signals and function as phosphatases. Non-phosphorylated LuxO is inactive and thus no small RNAs are transcribed, which allows LuxR/HapR to be translated (Lenz et al., 2004; Tu and Bassler, 2007; Tu et al., 2008). LuxR can regulate bioluminescence, the genes encoding type III secretion apparatus and metalloproteases (Henke and Bassler, 2004; Pompeani et al., 2008). In V. cholerae, HapR activates the expression of Hap protease at low cell-density while it inhibits biofilm formation and virulence factor production at high cell- density (Zhu and Mekalanos, 2003; Hammer and Bassler, 2003). However, it is worth pointing out that the dual roles of LuxS in metabolism and in AI-2 formation have led to controversy regarding its function in species other than V. harveyi or V. cholerae (Rezzonico and Duffy, 2008). 35 1.4. AHL-mediated inter-kingdom signaling: Ulva-AHL interaction The molecular mechanisms of AHL-mediated communication have been examined extensively in many bacterial species, including both intra- and inter- species communication (Fuqua and Greenberg, 2002; Ng and Bassler, 2009). Moreover, convincing data have revealed that some eukaryotes can also sense and respond to AHLs. Ulva intestinalis is a green intertidal biofouling macroalga in the division Chlorophyta and contributes to the biofouling of man-made surfaces worldwide (Joint et al., 2000). U. intestinalis can reproduce zoospores which are released from the thallus with timing often related to tidal cycles. Once these swimming zoospores find a favorable site for attachment, the zoospores firmly attach onto the surface permanently by secreting an adhesive glycoprotein and proceed to develop into mature individuals. However, if the surface is not satisfactory, zoospores will detach and swim to find another surface. The presence of bacterial biofilms on the surface is shown to be a determining factor in the choice of settlement or detachment for the spores. Pioneering work conducted by Joint et al. (2000) investigated the relationship between bacterial biofilms and the attachment of Ulva zoospores on glass slides and found a positive relationship between the number of zoospores that attach onto the surface and the number of bacteria present. A follow up study showed that AHLs in the fish pathogen V. anguillarum are involved in zoospore settlement (Joint et al., 2002). V. anguillarum has vanMN and vanIR QS pathways. VanM synthesizes 3-C6-HSL and 3-OH-C6-HSL while VanI produces 3-oxo-C10-HSL 36 (Milton et al., 2001; Milton, 2006). The biofilm of wild type V. anguillarum can effectively attract the spores while biofilms of vanM or vanIM mutants showed a substantial decrease in zoospore settlement (Joint et al., 2002). It would be also very interesting to test how these AHL receptor mutants (vanN or vanR) affect the settlement. Furthermore, biofilms of the wild type V. anguillarum strain expressing the Bacillus lactonase AiiA that can degrade AHLs by opening the lactone ring showed decreasing abilities to attract spores (Joint et al., 2002). To exclude the possibility that it is the known or unknown AHL-mediated phenotypes that participate in this settlement process, the authors provided more convincing data by expressing these AHL synthases in E. coli and showing an increase of zoospore settlement (Joint et al., 2002). Provision of functional pure AHL molecules repeated the pattern obtained by expressing the gene heterologously (Joint et al., 2002). Image analysis showed that zoospores indirectly bind to the bacterial microcolonies where AHL concentration is higher, which argues that the topology of microcolonies might play a role in the attachment process (Tait et al., 2005). However killing the biofilm with UV treatment or the treatment of antibiotic chloramphenicol that can kill the cells but does not affect the physical properties, significantly affected the zoospore attachment, which suggests that the topology of the microcolonies is not important in the attachment process (Tait et al., 2005). A control that adds the AHL back to these dead biofilms to test if it can restore the attachment would be a useful experiment to provide additional evidence to underscore the direct link between the AHL and attachment. An array of elegant experiments was performed by Wheeler et al. (2006) to understand the 37 mechanism of the effect of AHLs on the settlement. AHLs were shown to act as a chemoattractant for Ulva zoospores through chemokinesis. Zoospores can accumulate around an AHL point source while the swimming speed dropped dramatically as they approach closely to the AHL source. The decrease in swimming speed was also confirmed in the presence of V. anguillarum biofilm producing AHL while not in the presence of an AHL production deficient biofilm (Wheeler et al., 2006). However, it is unclear what molecular pathway is used by the zoospores to sense the presence of AHLs. To test whether the observed inter-kingdom signaling happens in situ, Tait et al. (2009) successfully showed the presence of AHLs (3-C8-HSL and 3-C10-HSL) on the surfaces of pebbles recovered from intertidal rock-pools where zoospores can colonize. The concentrations of these AHLs were roughly 600-pmol cm-2. The bacterial composition of these surface associated communities was examined by 16S rRNA gene clone library analysis. The dominant members are Alphaproteobacteria and Bacteroidetes. More importantly, some of the environmental isolates from the original habitats are able to produce AHLs. Zoospores show enhanced settlement towards biofilms of bacteria that can produce AHLs while not to those that do not produce AHLs, providing convincing data that this interaction between bacteria and algae occurs in situ (Tait et al., 2009) 1.5. Two-component systems Two-component systems (TCS) are arguably the most dominant means that allow bacteria to sense and respond to their environment. A typical TCS consists 38 of a membrane-bound histidine kinase (HK) that senses the signal and conveys the signal input to a cognate response regulator (RR) though phosphorylation (Fig. 1.4) (Capra and Laub, 2012). In its simple version, the HK is membrane bound and typically catalyzes an autophosphorylation reaction on the conserved histidine residues. The phosphoryl group is transferred to a conserved aspartate (D) on a cognate response regulator. In a more complicated version, the HK has an additional domain called the receiver domain and thus is called a hybrid histidine kinase. The phosphoryl group is still transferred to a conserved aspartate residue in the receiver domain first but in the same HK protein. In turn, the same group is transferred to the histidine residue in the intermediate protein called a histidine phosphotrasferase and ultimately transferred to the conserved aspartate residue in the cognate response regulator. This complicated version is also called a phosphorelay system (Capra and Laub, 2012). The phosphorylation on the aspartate residue of the response regulators would induce conformational changes, which can lead to homodimerization of the receiver domain and eventually cause transcriptional changes (Carroll et al., 2009). The response regulator has an input receiver domain and an output domain. There are many different types of output domain but most frequently these are DNA binding domains (Laub and Goulian, 2007). The two described versions of TCS are illustrated in Figure 1.4. When no signal is available, the HK functions as the phosphatase and the flow of the phosphoryl group is reversed and the response regulator is in an inactive status (Capra and Laub, 2012). 39 Figure 1.4. Organization and mechanism of two-component systems. Adapted from Capra and Laub (2012) with permission. DHp=dimerization and histidine phosphotransfer, CA=catalytic and ATP binding. The name of each domain represented by the ellipse with different colors is shown next to it except for the purple one (most frequently a DNA binding domain). Almost all the sequenced bacterial genomes encode two-component systems. The number of signaling proteins in these sequenced genomes roughly 40 scales with the genome size and also is related to the diversity of environments in which organisms live (Capra and Laub, 2012). How a single bacterial genome evolves to encode many TCSs and how the specificity of each system is determined have been extensively reviewed (Laub and Guolian, 2007; Capra and Laub, 2012). The two-component system mediated signal pathways have been implicated in the responses of bacteria to a variety of environmental signals and stimuli, such as nutrients, cellular redox state, quorum signals and others (Laub and Guolian, 2007). The cckA (histidine kinase)-chpT (phosphotransferase)-ctrA (response regulator) phosphorelay system is a good example that illustrates how the signal pathway works (Quon et al., 1996; Jacobs et al., 2003). It has been extensively examined in the bacterial development model Caulobacter crescentus, in which it coordinates DNA replication, cell division and polar morphogenesis and the response regulator CtrA is considered to be the cell cycle master regulator (Laub et al., 2002; Iniesta et al., 2006). The role of this system in regulating cell cycle has been implicated in several species while in some other bacteria species its role in regulating in cell cycle is obscure (Mercer et al., 2010; Greene et al., 2012). The example of LuxM, LuxPQ and CqsS in V. harveyi described above in Section 1.3.6 is another good example demonstrating the mechanisms of a two-component system. The difference is when these membrane-bound receptors bind to their cognate signals; it functions as a phosphatase not a kinase (Ng and Bassler, 2009). 41 1.6. Focus and objectives Marine sponges can be favorable environments for bacteria where nutrient concentration can be high and bacteria can grow dense enough to allow bacterial QS processes to be important. My research dissects the molecular pathways of QS in a representative sponge symbiont and reveals the phenotypes that can be controlled by QS and in turn the exact molecular mechanisms for that control. Furthermore, a major part of my research effort is directed towards establishing an experimental model for studying the ecological role of QS in the symbiosis between bacteria-sponge. To fulfill these goals, my research objectives are: I. Dissection of the QS pathways in the sponge symbiont Ruegeria sp. KLH11. II. Understanding of the molecular pathways for flagellar swimming motility control by QS. III. Establishment of the connection between laboratory understanding of QS and the role of QS in the sponge host. IV. Development of M. laxissima/KLH11 as a model system for the study of bacterial colonization of sponges. The results from my research will reveal for the very first time the architecture of the multiple QS pathways in a sponge symbiont that is also a member of the ecologically abundant and significant Roseobacter clade. Insights are also provided into the LuxI solo in bacteria. LuxR solos have been studied in many different bacterial species whereas only in a very few species has a LuxI solo been reported. Furthermore, the efforts devoted to establish an experimental 42 model to study the role of QS in the bacteria-sponge symbiosis provide a primary model for the future to better understand the complex symbioses existing in nature. 43 Chapter 2. A complex LuxR-LuxI type quorum sensing network in a roseobacterial marine sponge symbiont activates flagellar motility and inhibits biofilm formation 44 2.1. Abstract Bacteria isolated from marine sponges, including the Silicibacter-Ruegeria (SR) subgroup of the Roseobacter clade, produce N-acylhomoserine lactone (AHL) quorum sensing signal molecules. This study is the first detailed analysis of AHL quorum sensing in sponge-associated bacteria, specifically Ruegeria sp. KLH11, from the sponge Mycale laxissima. Two pairs of luxR and luxI homologues were identified and designated ssaRI, and ssbRI (sponge- associated symbiont locus A, and B, luxR or luxI homologue). SsaI produced predominantly long-chain 3-oxo-AHLs and SsbI specified 3-OH-AHLs. Addition of exogenous AHLs to KLH11 increased the expression of ssaI but not ssaR, ssbI or ssbR, and genetic analyses revealed a complex interconnected arrangement between SsaRI and SsbRI systems. Interestingly, flagellar motility was abolished in the ssaI and ssaR mutants, with the flagellar biosynthesis genes under strict SsaRI control, and active motility occurring only at high culture density. Conversely, ssaI and ssaR mutants formed more robust biofilms than wild type KLH11. The coordination of motility and biofilm formation by AHL signaling could contribute to the decision between motility and sessility and that it also may facilitate acclimation to different environments including the sponge host. 45 2.2. Introduction Over the past decade, culture-based and culture-independent techniques have frequently and consistently identified alpha-proteobacteria among the diverse bacteria from marine environments (Buchan et al., 2005), including in association with marine sponges (Webster and Hill, 2001; Mohamed et al., 2008c). More specifically, the Roseobacter clade, is estimated to account for 20- 30% of the bacterial 16S rRNA in the ocean surface waters (Buchan et al., 2005). Although a significant number of roseobacters are apparently free-living, many have been found associated with eukaryotic hosts, such as dinoflagellates and sponges (Miller and Belas, 2006; Mohamed et al., 2008b; Slightom and Buchan, 2009), and in some cases, the symbiotic bacteria are essential for host survival. When nutrients are plentiful, roseobacters can associate with particulate organic matter or algal particles to form aggregates (Fenchel, 2001). For roseobacters the switch from a free-living state to a multicellular aggregate may involve chemical signaling (Gram et al., 2002). Marine sponges harbor complex and diverse bacterial communities, which in some cases is estimated as 30-40% of the sponge biomass (Vacelet, 1975; Vacelet and Donadey, 1977; Hentschel et al., 2006). Although some bacteria serve as nutrients, there is evidence for stable symbionts, including several that are vertically transmitted (Enticknap et al., 2006, Sharp et al., 2007; Schmitt et al., 2008; Lee et al., 2009). Within the densely colonized sponge, there is also ample opportunity for both inter-species and intra-species signaling (Zan et al., 2011b). 46 Microbial signaling is common among dense microbial populations, such as those found within sponges. Quorum sensing (QS) allows bacteria to sense and perceive their population density through the use of diffusible signals (Fuqua et al., 1994). The acyl-homoserine lactones (AHLs), that are widespread among Proteobacteria, are most often synthesized by enzymes of the LuxI family (Churchill and Chen, 2011). AHLs have been identified in greater than a hundred different species and are among the best-studied signaling molecules in Proteobacteria (Ahlgren et al., 2011). QS regulates a variety of cellular processes including bioluminescence, conjugal transfer, symbioses, virulence, and biofilm formation (Fuqua and Greenberg, 2002; Daniels et al., 2004). Various sponge-associated roseobacters have been shown to synthesize AHLs (Taylor et al.,2004; Mohamed et al., 2008c). Ruegeria sp. KLH11, a sponge-associated member of the Silicibacter? Ruegeria subgroup of Roseobacter clade, produces at least six different AHLs detected using AHL-responsive biosensors (Mohamed et al., 2008c). Examination of the QS circuits in KLH11 provides a window into the complex interbacterial signaling that is potentially at play within sponges. In this study, we report (i) the types and relative amounts of KLH11 AHLs (ii) isolation and genetic analyses of multiple genetically-linked luxR and luxI homologues, their cognate AHLs, and their quorum sensing networks, (iii) the control of flagellar swimming motility and biofilm formation by QS. This study represents the first detailed chemical and genetic analyses of QS in the Roseobacter clade. 47 2.3. Experimental procedures 2.3.1 Reagents, strains, plasmids, and growth conditions All strains and plasmids used in this study are listed in Table 2.1 and all the primers used are listed in Table 2.2. Antibiotics were obtained from Sigma Chemical Co. (St. Louis, MO). Standards used for AHL analysis; D3C6-HSL (N- [(3S)-tetrahydro-2-oxo-3-furanyl]-hexanamide-6,6,6-D3 %99% deuterated product), 3-oxo-C14-HSL, 3-oxo-C14:1 &7cis-(L)-HSL and 3-oxo-C16:1 &11cis- (L)-HSL were purchased from Cayman Chemical (Ann Arbor, Michigan). 3-OH- C12-HSL, 3-OH-C13-HSL and 3-OH-C14-HSL were synthesized as described previously (Gould et al., 2006). Reagents used for sample derivatization and mass spectrometry analysis were: methoxyamine hydrochloride (MP Biomedicals, Solon, Ohio), bis (trimethylsily) tri-fluroacetamide (Supelco, Bellefonte, PA). HPLC grade acetonitrile, HPLC grade methanol, and sodium acetate trihydrate were purchased from Fisher Scientific (Fair Lawn, New Jersey). The solid phase extraction cartridges were Strata-x 33u polymeric reversed phase 60 mg ml-1 from Phenomenex (Torrance, CA) or Sep-Pak plus, silica cartridges (Milford, MA). DNA manipulations were performed by standard technique (Sambrook et al., 1989) and restriction enzymes and PhusionTM High-Fidelity DNA Polymerase were obtained from New England Biolabs (Ipswich, MA). Oligonucleotides were obtained from Integrated DNA Technologies (Coralville, IA) and DNA sequencing was performed on an ABI3700 automated sequencer. Sequence analysis was 48 performed with Vector NTI Advance 10 (Invitrogen Corp., Carlsbad, CA). KLH11 and KLH11-EC1 derivatives were grown in Marine Broth 2216 (MB2216) (BD, Franklin Lakes, NJ) at 28?C. E. coli strains were grown at 37?C in Luria-Bertani (LB) broth. A. tumefaciens strains were grown in AT minimal medium supplemented with 0.5% glucose and 15 mM (NH4)2SO4 (ATGN) (Temp? et al., 1977). Antibiotics were used at the following concentrations (?g ml-1): (i) E. coli (ampicillin, Ap, 100; gentamicin, Gm, 25; kanamycin, Km, 25; spectinomycin, Sp, 100); (ii) KLH11 (Km 100; Rifampicin, Rif, 200; Gm 25; Sp100) (iii), A. tumefaciens (Gm 300; Sp 200). Plasmids were introduced into KLH11 using either electroporation or conjugation, into E. coli using standard methods of transformation or electroporation (Sambrook et al., 1989) and into A. tumefaciens using a standard electroporation method (Mersereau et al., 1990). The methods to make electrocompetent cells of KLH11 were similar to that described by Piekarski et al. (2009). Briefly, mid-log phase (OD600~ 0.5-0.6) of KLH11 was harvested by centrifugation at 5000 x g for 7 min. The supernatant was discarded and the cell pellets were gently washed with 10% ice-cold glycerol followed by centrifuged at 5000 x g for 7 min. The wash step was repeated twice. The cells were eventually resuspended in 30% ice-cold glycerol and stored at -80?C. 49 2.3.2 Preparation of AHL samples and analysis by RP-HPLC and ESI mass spectrometry KLH11 derivatives were grown in MB 2216 with appropriate antibiotics at 28?C to stationary phase (OD600 ~2.0) in the presence of 5 g l-1 of Amberlite XAD 16 resin for 36 h. E. coli MC4100 expressing ssaI or ssbI was grown in LB at 37?C with 5 g l-1 of Amberlite XAD 16 resin to an OD600 of 0.6-0.8 and expression was induced 24 h by addition of 1 mM IPTG. Cells and resin were separated by centrifugation and extracted with 50 ml methanol and dried to 2 ml. 3 nanomoles of D3-C6-HSL was added to each sample as an internal standard and a volume of 0.2 ml was purified using solid phase extraction methods as described previously (Gould et al., 2006). The purified sample was dried down and resuspended in 38 'l solvent A (8.3 mM acetic acid-NH4 pH 5.7) and 2 'l solvent B (methanol). This solution was injected on to a 50 x 3.00 mm 2.6' C18 Kinetex Phenomenex column. A mobile phase gradient was generated from 5% B to 65% B in 5 min, then B was increased to 95% in 15 min and held for 8 min at a flow rate of 250 'l min-1. The HPLC system was interfaced to the electrospray source of a triple quadrupole mass spectrometer (Sciex API2000, PE Sciex, Thornhill, Ontario, CA). Precursor ion-scanning experiments were performed in positive-ion mode with the third quadrupole set to monitor m/z 102.1 and the first quadrupole set to scan a mass range of 170 to 700 over 9 s. The collision cell and instrument parameters were as follows: ion spray voltage of 4200 V, declustering potential of 50 V and collision energy of 25 V with nitrogen as the collision gas. 50 2.3.3 Qualitative analysis and estimation of AHL quantities The identification of AHL molecular species was carried out at the first level using the HPLC retention times for the specific ions that gave rise to m/z 102.1 (precursor ion scanning) corresponding to the available AHL reference standards. In all cases, co-injection of the authentic reference standard gave rise to an increase in the single HPLC peak corresponding to the correct AHL. For those AHLs for which reference material was not available, the observed precursor ion of m/z 102.1 was used to determine the [M+NH4]+ and therefore the molecular weight of a putative AHL. If this molecular weight corresponded to a saturated or monounsaturated AHL, then the retention time was compared to predicted retention time for this molecular species based on retention times of the reference AHLs. If the molecular weight corresponded to addition of an oxygen atom (keto or hydroxyl substituent, 14 and 16 u higher) to the saturated HSL series, then derivatization by trimethylsilylation (Clay and Murphy, 1979) and methoximation (Maclouf et al., 1987) and reanalysis was performed to confirm the presence of an oxidized AHL molecular species. An increase in 72 u or 29 u in the observed [M+NH4]+ would correspond to the formation of a trimethylsilyl ether or methoxylamine derivative of a hydroxylated-AHL or keto- AHL, respectively. The relative amount of each AHL species was estimated based on the ratio of the abundance of the transition [M+NH4]+ to m/z 102.1 divided by the abundance of the ion transition derived from the added internal standard (m/z 220.3 ? >102.1), as previously described (Gould et al., 2006). In separate experiments 51 this resulted in a linear relationship between the abundance ion ratios for the precursor ions of m/z 102.1 and quantity of AHL reference standards. 2.3.4 Genomic library screen of KLH11 QS genes KLH11 genomic DNA was obtained using the BactozolTM DNA Isolation Kit from Molecular Research Center Inc. (Cincinnati, OH). The DNA was independently digested to completion by restriction enzymes HindIII and SalI and the fragments were ligated into expression vector pBBR1-MCS5 (Kovach et al., 1995) followed by electroporation into E. coli Electro-Ten Blue competent cells. Cells were plated onto LB plates with gentamycin selection and incubated at 37?C overnight. Plasmids were extracted from colonies pooled from a large number of plates. The mixed plasmids preparations from the HindIII and SalI libraries were independently electroporated en masse into A. tumefaciens NTL4 (Zhu et al., 1998). Transformants were plated on ATGN media plus appropriate antibiotics and X-gal (40 'g ml-1). Blue colonies (harboring putative AHL synthases) were chosen for further analysis after growth at 28?C for 2-3 days. The plasmids pECH100, pECH101, and pECS102 were identified as AHL+ transformants in the KLH11 genomic library screen and were used as the template for PCR amplification of ssaI, ssbI, and ssaR. To PCR amplify ssbR, the oligonucleotide specific for the 3? end of ssbR was designed from de novo sequence obtained from pECH101, isolated from the HindIII genomic library. However, only 62 bp of the ssbR sequence was carried on pECH101, due to a HindIII site at this position. R. pomeroyi DSS-3 has silR2 homologous to ssbR (Moran et al., 2004) and based on this sequence a primer designed to the gene 52 presumptively flanking ssbR was generated, and used to PCR amplify the complete ssbR sequence from KLH11. 2.3.5 Directed mutation, lacZ fusions and complementation For Campbell-type, recombinational mutagenesis internal gene fragments were generated and used to disrupt target genes (Kalogeraki and Winans, 1997). An internal portion of the ssaI gene was amplified from the pECH100 template using primers designated as 1 and 2 of ssaI. The partial ssaI fragment was gel purified and cloned into pGEM?-T Easy (Promega, Madison, WI), creating pEC103, which was confirmed through sequencing. For recombinational mutagenesis, pEC103 was digested with EcoRI and KpnI, and the resulting ssaI fragment was ligated to a similarly digested R6K replicon, the pVIK112 suicide vector (Kalogeraki and Winans, 1997), creating pEC113. pEC113 was conjugated into KLH11 and transconjugants resistant to kanamycin (Km) were selected and confirmed by sequencing. The ssbI null mutant, designated as EC3, the ssaR null mutant EC4, and the ssbR null mutant EC5 were created similarly to EC2, each using the primers 1 and 2 of these genes and confirmed by sequencing. These plasmid insertions also create lacZ transcriptional fusions in the genes they disrupt. To study the effect of AHL on the luxR-type genes, a fragment of ssaR gene, about 500 bp ending at the stop codon was PCR amplified using primers ssaRintact F and ssaRintact R. The PCR amplicon was cloned into pGEM?-T Easy (Promega, Madison, WI) first and then subcloned into pVIK112, creating pJZ001. The pJZ001 was conjugated into KLH11 and transconjugants were selected and confirmed as described above. Thus, the 53 lacZ was fused into the 3? end of the ssaR gene, keeping ssaR intact. The same approach was used to fuse lacZ into the 3? end of the ssbR gene while retaining ssbR intact. To generate in-frame deletions, a standard approach was utilized (Merritt et al., 2007). For the ssaI gene, about 600 bp upstream of ssaI gene was PCR- amplified using primers ssaI D1and ssaI D2 and approximately 400 bp including 206 bp of ssaI encoding sequences and about 200 bp downstream of ssaI was PCR-amplified using primers ssaI D3 and ssaI D4 (Table 2.2). Primers ssaI D2 and ssaI D3 had complementary sequence at the 5? end to allow Splicing by Overlapping Extension (SOEing), as described previously (Merritt et al., 2007). In brief, these two fragments were gel purified, followed by PCR amplification with primers ssaI D1 and ssaI D4 using an equal amount of the two fragments as templates. The PCR product of this amplification was gel purified and digested using restriction enzymes SpeI and SphI. The digested PCR product was ligated into the sacB counter-selectable vector pNPTS138 (Hibbing and Fuqua, 2011), which was digested with the same combination of restriction enzymes. Derivatives of pNPTS138 were conjugated into Ruegeria sp. KLH11. The selection of the first crossover was performed by plating transconjugants onto MA 2216 plates with Rif and Km. The colonies that grew on this selective medium, but not on the same plates supplemented with 5% (w/v) sucrose were chosen and subcultured in MB 2216 without Km to allow for excision of the integrated plasmid, followed by plating on 5% sucrose MA2216 plates without Km. Sucrose resistant (SucR) KmS colonies were selected. Deletion of the 54 targeted region was confirmed by PCR. Deletion of ssbI was performed in the same way using primers ssbI D1-ssbI D4. The double deletion of ssaI and ssbI was performed by deleting ssbI in the ssaI deletion strain using the same method as described above. Controlled expression constructs of ssaI, ssaR, ssbI and ssbR were generated by PCR amplification of the coding regions of each gene using primers designated as 3 and 4 for each specific gene and genomic DNA of KLH11 as template (Table 2.2). The E. coli lacZ ribosomal binding site was engineered into the 5? primer of each gene. The PCR products were digested with the appropriate restriction enzyme(s) and ligated into the vector pBBR1-MCS5 (Kovach et al., 1995). The insert carried by each construct was confirmed by sequencing. 2.3.6 Plasmid-borne promoter fusions and expression plasmids The intergenic region upstream of the ssaI coding sequence and downstream of ssaR contains the presumptive ssaI promoter region and was PCR amplified from pECS102 using primers ssaI P1 and ssaI P4. The upstream and downstream oligos anneal 182 bp upstream, and 3 bp downstream of the ssaI translational start site, respectively. In addition, several plasmid-borne ssaI promoter deletions were also generated using PCR amplification to truncate the 5? sequences, at positions 79 bp and 63 bp upstream of the ssaI translational start site. These PCR products were cloned into pCR?2.1-TOPO? and their content was confirmed by DNA sequencing. The pCR?2.1-TOPO? derivatives were digested with EcoRI and PstI, and the resulting fragments were ligated with 55 equivalently digested pRA301 vector (Akakura and Winans, 2002). Similarly, the ssbI promoter and translational start site were amplified from KLH11 genomic DNA using primers ssbI P1 and ssbI P4 to generate an amplicon that extends 229 bp upstream and 3 bp downstream of the ssbI predicted translational start site. This amplicon was subsequently used to generate a plasmid-borne PssbI- lacZ fusion on pRA301. Plasmids Plac-ssaR (pEC112) and Plac-ssbR (pEC123) were introduced into the AHL- A. tumefaciens NTL4 in combination with the compatible PssaI-lacZ and PssbI-lacZ plasmids to examine gene regulation patterns in a heterologous AHL- host. 2.3.7 Preparation of log phase cell concentrates and !- galactosidase assays "-galactosidase assays for both A. tumefaciens and KLH11 derivatives were performed as described previously (Miller, 1972). Mid-log phase A. tumefaciens cultures were diluted at 1:100 dilution to an OD600~0.01 and were supplemented with different concentration of 3-oxo-C16:1 #11-HSL. The cell suspension was thoroughly mixed and then aliquoted into 15 ml test tubes, incubated at 28?C for 24 h to an OD600~0.4. Mid-log phase cultures were measured for OD600 and frozen at -20?C and used for subsequent "-galactosidase assays. Cultures of Ruegeria sp. KLH11 were prepared in a similar way. KLH11 was grown in MB 2216 supplemented with Km as required overnight and was diluted at 1:100 dilution to an OD600~0.01. 3-OH-C14-HSL and 3-oxo-C16:1 #11-HSL were added at the concentration of 20 'M and 2 'M, respectively. Mid-log phase of 56 KLH11 cultures was sampled and "-galactosidase assays were performed immediately. 2.3.8 AHL detection using an ultrasensitive A. tumefaciens reporter for KLH11 derivatives AHLs were extracted from KLH11 cultures using dichloromethane, fractionated by reverse phase TLC, and detected using an ultrasensitive AHL bioreporter derived from A. tumefaciens (Zhu et al., 2003, Mohamed et al., 2008c). Five ml MB 2216 cultures from KLH11 were grown to an OD600 of 1.5 ? 2.0, followed by extraction with an equal volume of dichloromethane. Culture pH was monitored and was within the range of 7.6 ? 0.2 at the time of harvesting (sterile Marine Broth 2216 is pH 7.6). Following centrifugation, the organic phase was removed and allowed to evaporate in a fume hood. Extracts were concentrated 1000 fold, normalized to an OD600 of 1.5 and resuspended in a final volume of approximately 5 'l of acidified (0.01%) ethyl acetate and loaded onto a C18 RP- TLC plate (Mallinckrodt Baker, Phillipsburg, NJ, USA). TLC plates were developed in a 60% methanol water mobile phase, dried, and overlaid with 100 ml of 0.6% ATGN media supplemented with 40 'g/ml X-gal and 1 ml of an OD600 = 12.0 suspension of the highly sensitive A. tumefaciens AHL reporter (A. tumefaciens KYC55 [pJZ372][pJZ384][pJZ410]) as previously described (Zhu et al., 2003). TLC plate overlays were placed in a sealed container and incubated at 28?C for 16-48 hrs. 57 2.3.9 Motility assay, flagellar stain and immunodetection of flagellin Bacterial swimming assays were performed using MB2216 with 0.25 % (w/v) agar. No antibiotics were added to the medium. Plates were inoculated at the center with freshly isolated colonies. 3-oxo-C16:1 #11-HSL was added to MB 2216 agar to 2 'M. Plates were placed in an air-tight container with a beaker containing 15 ml of K2SO4 to maintain constant humidity, and incubated 8 days at 28?C. Flagellar stains were performed based on methods described by Mayfield and Inniss (1977) and were viewed by phase contrast microscopy. For immunoblotting with anti-flagellin antibodies 5 ml MB2216 cultures (wild type, ssaI? and ssaR?), were grown to late stationary phase. Culture volumes were normalized to an OD600 of 0.6, and portions of these were centrifuged at 716 x g for 12 min to gently separate cells and supernatant. Protein from the whole culture and supernatant fractions was precipitated with an equal volume of 100% trichloroacetic acid. Following vortexing and 15 min incubation on ice, samples were centrifuged (8,765 x g; 20 min) and washed with 1 ml of acetone. The whole culture, pellet, and supernatant fractions were resuspended in 1X SDS lysis buffer, boiled 10 min, and used in subsequent immunoblotting analysis. Immunoblotting with anti-flagellin polyclonal antibodies raised against C. crescentus whole flagella (a gift from Y.V. Brun) was performed using a standard technique. Samples were separated by 15% SDS-PAGE at 100 V for 80-90 min, and transferred to the membrane (Osmonics, Westborough, MA) at 30 V for 40 58 min using a semi-dry electrotransfer system. The membrane was blocked overnight at 4?C in Blotto (1x TBS-T [Tris Buffered Saline], 1% Tween 20) and 5% dried milk). The polyclonal antibody was diluted 1:20,000 in 4 ml of Blotto and incubated with the membrane for 45 min on a rocking shaker at room temperature. The membrane was then washed three times for 5 min in 1X TBS- T. Secondary antibody solution at 1:20,000 in 4 ml of Blotto was incubated with the membrane and rocked 45 min at room temp. The blot was rinsed three times for 5 min with 1X TBS-T, followed by 2 times in 1X TBS. Chemoluminescent substrate, Supersignal West Pico Chemoluminescent Substrate (Pierce, Rockford, IL) was used as the detection reagent. Equal parts of the Luminol/Enhancer and Stable Peroxide Buffer were combined and pooled on top of the membrane for ~5 min. Excess detection reagent was removed by blotting on filter paper, and signal was detected by exposure to Kodak BioMax film (Kodak). To analyze flagellin biosynthesis across the KLH11 growth curve, samples were taken at a range of times between mid-exponential phase through late stationary phase. 100 ml cultures were grown at 28?C and samples were taken for EC1 (wt KLH11) at OD600 of 0.5, 1.3, 1.8, and 2.2; strain EC2 (ssaI?,) at OD600 of 0.8, 1.3, 1.7, and 2.4. All samples were processed and analyzed by western blotting and culture samples were also viewed under phase contrast microscopy to evaluate swimming at these time points. 59 2.3.10 RNA extraction and quantitative reverse transcription PCR (qRT-PCR) analysis Expression levels of fliC and flaA were measured using qRT-PCR with specific primers (Table 2.2). KLH11 and derivatives were grown in MB2216 to an OD600 of 1.8 and 1 ml RNA Protect Bacterial Reagent (Qiagen, Valencia, CA) was added to 0.5 ml culture. The mixture was kept at room temperature for 5 min and was centrifuged at 5000 x g for 10 min. The cell pellet was stored in at - 20?C for RNA extraction. Total RNA was isolated using an RNeasy miniprep kit (Qiagen, Valencia, CA), with genomic DNA removed by TURBO DNase (Ambion, Austin, TX), and cDNA synthesized using qScriptTM cDNA SuperMix according to the manufacturer?s instructions (Quanta BioSciences, Gaithersburg, MD). The RT-PCR was carried with PerfeCTaTM SYBR? Green FastMixTM Low ROX. Reactions were performed on an Mx3000P qPCR system (Stratagene, Santa Clara, CA) using the following cycling parameters: 2 min at 95?C for initial denaturation, 40 cycles for 10 s cycles at 95?C, and 30 s for primer annealing, and primer extension at 60?C. Melt curves were performed to confirm the specificity of primers and the absence of primer dimers. Expression levels of fliC and flaA genes were normalized to the KLH11 vegetative sigma factor 70 gene (rpoD). 2.3.11 Biofilm assays A standard coverslip biofilm assay was used to evaluate the impact of KLH11 QS on biofilm formation (Tomlinson et al., 2010). Briefly, overnight cultures of different strains were inoculated at OD600~0.06 in 3 ml MB 2216 in UV-sterilized 60 12-well polystyrene plates and biofilms were grown on sterile PVC coverslips suspended vertically in the wells. The 12-well plates were incubated statically at 28?C over three days. At specific time points, 200 'l of culture was removed to measure the turbidity. The adherent biomass was stained with 0.1% (w/v) crystal violet (CV) solution for 5-10 min and then rinsed gently with DI water to remove loosely attached cells. CV adsorbed to the biofilms was extracted with 33% acetic acid and its absorbance at 600 nm (A600) was measured. These values were normalized by the OD600 of planktonic growth. To examine inhibition of biofilm formation, wt KLH11 was grown in MB2216 at 28?C with shaking (200 rpm) to late stationary phase (55?60 h). Bacterial cells were removed from culture volumes of 300 ml by 2 rounds of centrifugation at 5000 x g for 10 min. The supernatant was filtered through a 0.22-'m filter and stored at -80 ?C until ready for use. For the biofilm assays to which culture fluids were added the #ssaI mutant was grown in MB2216 overnight and then diluted to an OD600 ~ 0.06. 2X MB2216 was diluted with the appropriate amounts of culture fluids and water to ensure that there was always at least a 1X concentration of nutrient in the initial biofilm inoculum. Biofilm formation was detected at 48 h as described above. 2.3.12 Statistical analysis Unpaired Student?s t test was used to calculate P value in all chapters. 61 Table 2.1. Strains and plasmids used in Chapter 2. Bacteria/Plasmids Relevant featurea Reference E. coli Electro-Ten Blue Standard alpha-complementation strain Stratagene E. coli DH5!/"pir Strain for propagating R6K suicide plasmids Lab collection E. coli DH5!/"pir Strain for propagating R6K suicide plasmids Lab collection E.coli XL-1 Blue Standard alpha-complementation strain Lab collection E. coli TOP 10 F? Standard alpha-complementation strain, lacIQ Qiagen E. coli MC4100 K-12 derivative, #lacZ Lab collection A. tumefaciens NTL4 Ti plasmidless derivative, nopaline chromosomal background (Zhu et al., 1998) A.tumefaciens KYC55 Ti plasmidless derivative, octopine chromosomal background (Zhu et al., 2003) Ruegeria pomeroyi DSS-3 wild type (Moran et al., 2004) KLH11 wild type (Mohamed et al., 2008c) KLH11-EC1 RifR This study KLH11-EC2 ssaI-lacZ, null ssaI, RifR, KmR This study KLH11-SK01 $ssaI, RifR This study KLH11-EC3 ssbI-lacZ, null ssbI, RifR, KmR This study KLH11-SK02 $ssaI $ssbI, RifR This study KLH11-EC4 ssaR-lacZ, null ssaR, RifR, KmR This study KLH11-EC5 ssbR-lacZ, null ssbR, RifR, KmR This study KLH11-JZ1 ssaR-lacZ, wild type ssaR, RifR, KmR This study KLH11-JZ2 ssbR-lacZ, wild type ssbR, RifR, KmR This study KLH11-OKC9 fliC-lacZ, null fliC, RifR, KmR This study 62 pCR?2.1-TOPO? PCR fragment cloning vector, ApR, KmR Invitrogen pBBR1-MCS2 Plac expression vector, KmR (Kovach et al., 1995) pBBR1-MCS5 Plac expression vector, GmR (Kovach et al., 1995) pGEM?T-Easy PCR fragment cloning vector, ApR Promega pVIK112 R6K-based lacZ transcriptional fusion, integration vector, KmR (Kalogeraki and Winans, 1997) pRA301 lacZ translational fusion vector (Akakura and Winans, 2002) pJZ372 PtraI-lacZ translation fusion, TcR (Zhu et al., 2003) pJZ384 PT7-traR, SpR (Zhu et al., 2003) pJZ410 T7 polymerase expressing plasmid (Zhu et al., 2003) pECH100 pBBR1-MCS5 derivative, 3 kb HindIII fragment containing ssaI and truncated ssaR, GmR This study pECH101 pBBR1-MCS5 derivative, 2.8 kb HindIII fragment containing ssbI and truncated ssbR, GmR This study pECS102 pBBR1-MCS5 derivative, 3.2 kb Sal I fragment containing ssaI and ssaR, GmR This study pEC103 pGEM?T-Easy derivative, carrying truncated ssaI fragment, ApR This study pEC104 pGEM?T-Easy derivative, carrying truncated ssbI fragment, ApR This study pEC105 pGEM?T-Easy derivative, carrying truncated ssaR fragment, ApR This study pEC106 pGEM?T-Easy derivative, carrying full length PCR-amplified ssbR fragment, ApR This study pEC107 pVIK112 derivative carrying truncated ssaR gene from pEC105, KmR This study pEC108 pBBR1-MCS5 derivative carrying full length Plac-ssaI, from pEC111, GmR This study 63 pEC109 pBBR1-MCS5 derivation carrying full length Plac-ssbI, from pEC110, GmR This study pEC110 pCR?2.1-TOPO? derivative carrying full length PCR-amplified Plac-ssbI, KmR This study pEC111 pCR?2.1-TOPO? derivative carrying full length PCR-amplified Plac-ssaI, KmR This study pEC112 pBBR1-MCS5 derivative, carrying full length Plac-ssaR from pEC106, GmR This study pEC113 pVIK112 derivation carrying truncated ssaI gene from pEC103, KmR This study pEC114 pCR?2.1-TOPO? derivative, carrying PCR amplified PssaI, Ap/KmR This study pEC115 pVIK112 derivative carrying truncated ssbI gene from pEC104, KmR This study pEC116 pRA301 derivation, PssaI-lacZ, SpR This study pEC117 pGEM?T-Easy derivative, carrying PssbR and ssbR ApR This study pEC118 pGEM?T-Easy derivative, carrying PssbI and ssbI ApR This study pEC119 pGEM?T-Easy derivative, carrying PssbR and ssbR, ApR This study pEC120 pGEM?T-Easy derivative, carrying full length Plac-ssbR, ApR This study pEC121 pRA301 derivative, PssbI-lacZ, SpR This study pEC122 pVIK112 derivative carrying truncated ssbR from pEC119, KmR This study pEC123 pBBR1-MCS5 derivative, carrying full length Plac-ssbR, from pEC120, GmR This study pEC124 pRA301 derivative, PssaI-lacZ, 5? promoter deletion, 79 bp with lux type box, SpR This study 64 pEC127 pRA301 derivative, PssaI-lacZ, 5? promoter deletion, 63 bp fragment lacks lux type box, SpR This study pJZ001 pVIK112 derivative, ssaR gene with 5? truncation, to retain wt ssaR, KmR This study pJZ002 pVIK112 derivative, ssbR gene with 5? truncation, to retain wt ssbR, KmR This study pOKC9 pVIK112 derivative, fliC gene with 3? truncation, KmR This study aAp=ampicillin, Gm=gentamicin, Km=kanamycin, Rif=rifampicin, Sp=spectinomycin. Tc=tetracycline. 65 Table 2.2. Primers used in Chapter 2. Primer name Sequence b(5?-3?) Restriction Enzyme ssaI D1 ACTAGTCTATGGTGACGACTGGAAG SpeI ssaI D2 GAATTCGTCAGTCAGTCAGTTTCCCGTAATATTGGCTT NA ssaI D3 TGACTGACTGACGAATTCAGGCTGGCGAACTCAAGCCTG NA ssaI D4 GCATGCGACTACATTGTCGAGCTG SphI ssbI D1 ACTAGTGCAATCAGGGTTATTCGATC SpeI ssbI D2 GAATTCGTCAGTCAGTCACAACATGATTGTTCCCCTTGT NA ssbI D3 TGACTGACTGACGAATTCGCCTGACCTTGGTGGAAATTG NA ssbI D4 GCATGCGATACGGTGAATGGTCGTTGC SphI ssaI 1 cggGAATTCATGTTCGAACTGCGCGCTCGGG EcoRI ssaI 2 gccGGTACCATCGCAGGGACCTTGCCCATC KpnI ssaI 3 ggcCTCGAGCTGAAACAGGAAACAGCTATGATTTTGGTAGTTGATG XhoI ssaI 4 ggcGAATTCGGGTCAGGCCTCATGAGCAAAAGC EcoRI ssaR 1 cgcGAATTCTCAGCACCCTCCCCGAACAGG EcoRI ssaR 2 cgcGGTACCCGGCCCATTGCAAAATCTC KpnI ssaR 3 gggCTCGAGGTGAAACAGGAAACAGCTATGGATATTGTTGATCTCAGC XhoI ssaR 4 gggGAATTCGGCTTAACCTGGGTAGATTAGCCC EcoRI ssbI 1 cgcGAATTCGGATGAGCTGCATAAATTTCCGG EcoRI ssbI 2 gccGGTACCAACGGAAATCCGGTCTCGCCCG KpnI ssbI 3 ggcCTCGAGCTGAAACAGGAAACAGCTATGTTGCGTTATTTATATGCG G XhoI ssbI 4 ggcGAATTCGGGTCAGGCGGAAAGCGCAAACCG EcoRI ssbR 1 ggcGAATTCTCGGTTCAGCTGCGCGATCGG EcoRI ssbR 2 cgcGGTACCTTCCGGTTGCCGGTTCTGGGC KpnI ssbR 3 gcgCTCGAGGTGAAACAGGAAACAGCTATGAGGCTTGCGCGCCCGCG XhoI ssbR 4 gggGAATTCGGCTTAAACGACTATTAATCCTCTGC EcoRI fliC 1 cgcgaattcAAGTCGGTCAACATGAACCTG EcoRI 66 fliC 2 gccggtaccGTTGTCACGATCAAGCGAGGA KpnI fliCRT1 CGCAGAACCTGTCGACCGGT NA fliCRT2 GGTATCGCCTGCGGCCAATGT NA flaART1 GCACCGATGCATATGCGCAAGCT NA flaART2 TGAGCGATTGCAGCCGGGTT NA rpoDRT1 GACGCCTATCGCGGCCGT NA rpoDRT2 GCCGACCTGCGCCATATCGT NA ssaI P1 ccgGAATTCTGCACTAACCACACCTCAGGCCG EcoRI ssaI P2 gcgGAATTCTACGGGAAACCCCCAATAGATTCG EcoRI ssaI P3 gggGAATTCTAGATTCGCTGTGAAATCCGAG EcoRI ssaI P4 ccgCTGCAGAATCATGTTAACCCCCTTCG PstI ssbI P1 cggGAATTCTATAGCCGGGCACAGGTGGCGC EcoRI ssbI P2 cccCTGCAGCAACATGATTGTTCCCCTTGTCGT PstI aP1 CGGCACCATTCATGGCCATGT NA aP2 CCATTCGTCCCGACTGCAGC NA aP3 ATGATTTTGGTAGTTGATG NA aP4 GTCGCATAGGACACCGAGTT NA bP1 CCAATATGGCTTCACGACCT NA bP2 AATAACCCTGATTGCCCACA NA bP3 GGTAAACGAAGATGGCGAAG NA bP4 GAGCCGATCATGCGATAAAT NA ssaRintactF GAATTCCAAGGCCTGCATCTGATCG EcoRI ssaRintactR GGTACCTTAACCTGGGTAGATTAGCCC KpnI ssbRintactF GAATTCTATCACCGCATTGATCCGG EcoRI ssbRintactR GGTACCTTAAACGACTATTAATCCTCTGCTG KpnI 67 b Engineered restriction sequences are underlined. Complementary sequences for PCR-SOEing are shown in bold and are also underlined. Start and stop codons are in bold. E. coli lacZ ribosomal binding sites are in italics and bold. NA=not applicable. 68 2.4. Results 2.4.1 AHL synthesis and genetic isolation of luxI and luxR homologues from KLH11 The sponge symbiont Ruegeria sp. KLH11 has a very complex profile of AHLs as evaluated by bioassays and fractionation by thin layer chromatography (Fig. 2.1; Mohamed et al., 2008c). This bioassay is highly sensitive, but has a bias towards short chain AHLs similar to the A. tumefaciens cognate 3-N-oxo- octanoyl-L-homoserine lactone (3-oxo-C8-HSL) (Zhu et al., 1998). A less biased semi-quantitative chemical analysis of the KLH11 whole culture organic extracts was performed using high performance liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS; Gould et al., 2006). The dominant AHLs were found to be hydroxylated forms of tetradecanoyl (C14) species, saturated (3-OH- C14-HSL) and unsaturated (3-OH-C14:1-HSL), and hydroxylated dodecanoyl species (3-OH-C12-HSL) (Fig. 2.2A, Fig. 2.3A, Table 2.3). The shorter chain AHLs detectable in bioassays (Fig. 2.1), were not observed with this chemical analysis suggesting that they are in low relative abundance. A genetic approach was employed to isolate KLH11 AHL synthase genes. Two plasmid-borne gene libraries, with KLH11 genomic DNA digested to completion with HindIII or SalI, both with cleavage products averaging 4 kb, were ligated into the expression (Plac) vector pBBR1-MCS5 (Kovach et al., 1995) and then transformed en masse into an A. tumefaciens strain that does not 69 synthesize AHLs (Mohamed et al., 2008c). This AHL sensitive strain responds to introduction of plasmid-borne AHL synthase genes and expresses an AHL- dependent lacZ fusion. A pool of several thousand transformants was screened on selective medium with no exogenous AHL and X-gal. Of the 16 blue colonies isolated, several produced a diffusible activity, which induced lacZ expression in closely adjacent colonies after extended incubation. From these presumptive AHL+ transformants, insert fragments were sequenced, identifying two separate loci with similarity to luxI from V. fischeri, each adjacent and in tandem arrangement downstream of a luxR homologue (Fig. 2.4A and B). We designated these genes ssaRI and ssbRI (Sponge-associated symbiont locus A and locus B, respectively). The draft genome of KLH11 was completed during the course of this study (Zan et al., 2011b) and 100% identical loci were identified; thus we refer to these Genbank Accession numbers for ssaRI, ssbRI and flanking genes (Fig. 2.4). The KLH11 ssaR-ssaI and ssbR-ssbI genes are highly similar to the silR1- silI1 and silR-silI2 genes from R. pomeroyi DSS-3 (Moran et al., 2004). SsaI is 71% identical to SilI1, and each has an unusual C-terminal extension (~60 aa) relative to other LuxI-type proteins. SsaR and SilR1 share 79% identity. SsbI shares 82% identity with SilI2 and SsbR is 74% identical to SilR2. Thus, it appears that SsaRI and SilRI1, and SsbRI and SilRI2, are orthologous, whereas the SsaRI and SsbRI systems are paralogous. After completing the KLH11 genome sequence, a third presumptive AHL synthase was identified and named sscI. Although sscI had escaped our genetic 70 screen, it shares 81% identity to SsbI on the amino acid level, is of similar size and is likely to be the result of a recent gene duplication (See Chapter 3). Figure 2.1. RP-TLC analysis of AHLs from KLH11 and QS mutants. TLC plates were overlaid with the A. tumefaciens ultrasensitive AHL reporter strain (Zhu et al. 2003). Mixtures of synthetic 3-oxo- and fully reduced AHL standards were run on each plate in lanes 1 and 2 (labeled on plate). (A) KLH11 SsaRI mutants. KLH11, ssaI?, and ssaR? mutants and the plasmid- complemented mutants are labeled. (B) KLH11 SsbRI mutants. KLH11, ssbI? and ssbR? mutants, and the plasmid-complemented mutants are labeled. AHL standard concentrations are: Fully reduced, C4, 1 mM; C6, 500 !M; C8, 50 nM; 71 C10, 125 !M. 3-oxo derivatives, 3-oxo-C6, 50 nM; 3-oxo-C8, 42 nM; 3-oxo-C12, 68 !M. Figure provided by Elisha M. Cicirelli and Clay Fuqua. Figure 2.2. Chemical analysis of purified samples from KLH11 derivatives. The products of reverse-phase chromatographic separation of AHLs extracted and purified from wild type KLH11 and mutants were examined using the precursor ion-scanning mode (transitions were monitored for precursor [M + H]+ - > m/z 102.1). The peaks in the chromatograms (A) wt KLH11, (B) KLH11 ssaI?, 72 (C) KLH11 ssaR?, (D) KLH11 ssbI?, and (E) KLH11 ssbR? are labeled with lowercase lettering and include species defined in Table 2.3 and the AHLs noted. Figures are provided by Mair Churchill. Figure 2.3. Comparative analysis of AHLs. The relative amounts of the known AHLs in (A and C) KLH11 derivatives and (B) E. coli MC4100 are shown in bar graphs, with the peak label above each set of bars. The plot shows the ratio of the area of the transition for each AHL to m/z 102.1 to the same transition for the internal standard D3-C6-HSL. Analysis of AHL samples was performed in 73 positive-ion mode with the third quadrupole set to monitor m/z 102.1 as described in Experimental Procedures. The relative amounts of the known AHLs in each strain are shown in the bar graph, and the data were analyzed as in Figures 2.2 and 2.5. (e), (f), (g), and (h) referring to detection of 3-OH-C12-HSL, 3-OH-C13-HSL/3-OH-C14:1-HSL, 3-OH-C14-HSL/3-oxoC14-HSL, and 3-oxo- C16-HSL/3-OH-C16:1-HSL respectively. In panels B and C, the SsaI y-axis is on the left and the SsbI y axis (with an asterisk) is on the right. Figures are provided by Clay Fuqua. 74 Figure 2.4. Gene maps of KLH11 ssaR/ssaI and ssbR/ssbI loci. Arrows represent genes. A) SsaR (ZP_05123091) and SsaI (ZP_05123801) and are predicted to be 233 and 284 amino acids, respectively. Genbank accession numbers are ZP_05124568 for transglycosylase and ZP_05123238 for trigger factor. Primers used to test whether ssaRI are in the same operon are indicated as P1, P2, P3 and P4 (see Table 2.2, aP1, aP2, aP3, aP4). Lanes 1-4 were PCR results using primers P1 and P2, lanes 5-8 used primers P3 and P4, and lanes 9-12 used primers P1 and P4. Lanes 1, 5 and 9 used cDNA as template, lanes 2, 6 and 10 used genomic DNA as template, lanes 3, 7 and 11 used RNA as templates. Lanes 4, 8 and 12 were negative controls comprising the primer sets. B) SsbR (ZP_05123460) SsbI (ZP_05121795) are predicted to be 239 and 212 amino acids. Genbank accession numbers are ZP_05124465 for crotonyl CoA reductase and ZP_05122236 for helicase. Primers used to test whether 75 ssbRI are in the same operon are indicated as P1, P2, P3 and P4 (see Table 2.2, primers bP1, bP2, bP3, bP4). The lanes are organized the same as in panel A. 2.4.2 SsaI and SsbI synthesize long chain?length AHLs when expressed in E. coli Each presumptive AHL synthase gene was expressed from the Plac promoter carried on a low copy number plasmid in E. coli MC4100 for mass spectrometry analysis. E. coli does not produce AHLs and thus those produced will generally reflect the intrinsic specificity of the enzyme in the E. coli background (Fig. 2.5, Table 2.3). Trimethylsilylation and methoximation of these samples (Clay and Murphy, 1979; Maclouf et al., 1987) revealed that SsaI produces 3-oxo-AHLs, the most abundant species of which is 3-oxo-C14-HSL (Table 2.3, Fig. 2.3B), but also including a C16 derivative, that was shown by methoximation to be a 3-oxo- C16 derivative. No 3-oxo-HSLs were observed for E. coli expressing SsbI, but rather 3-hydroxy-HSLs were identified, with predominant 3-OH-C14-HSL, 3-OH- C14:1-HSL and 3-OH-C13-HSL (Figs. 2.3B and 2.5B, Table 2.3). For both AHL synthases, putative AHL derivatives were also detected, but their identity was not confirmed due to a lack of reference standards (Table 2.3). Overall, SsaI and SsbI drive the synthesis of long chain (lc) AHLs that differ in their modification at the 3-position of the acyl chain, 3-oxo and 3-OH, respectively. 2.4.3 Mutational analysis of ssaRI and ssbRI in KLH11 Campbell-type plasmid insertions were generated in the ssaI, ssaR, ssbI and ssbR genes in KLH11 using pVIK112, which generates lacZ transcriptional 76 fusions to the disrupted gene (Kalogeraki and Winans, 1997). Mutation of ssaI resulted in complete loss of AHLs using the biosensor assay (Fig. 2.1 A). Consistently, mass spectrometry analyses of organic extracts from the ssaI mutant reveal a dramatic decrease in AHL abundance, although trace levels of 3- OH-C12-HSL were observed (Fig. 2.2B, Fig. 2.3A, Table 2.3). The loss of ssaR (Fig. 2.2C) did not significantly alter the pattern of AHLs observed in the wt KLH11, consistent with the bioassays (Fig. 2.1A). In contrast, chemical analysis of the ssbI? mutant surprisingly revealed an overall increase in AHL levels (Fig. 2.1B, Fig. 2.2 D), but also a shift in the spectrum of AHLs produced, including the presence of C16 derivatives (3-oxo-C16-HSL or 3-OH-C16:1-HSL, low levels precluded their distinction) as major species, but also hydroxylated derivatives, 3- OH-C14-HSL, 3-OH-C14:1-HSL, and 3-OH-C12-HSL (Fig. 2.3A, Table 2.3). Paradoxically, this indicates that although SsbI is clearly capable of driving AHL synthesis when expressed in E. coli, its presence in Ruegeria sp. KLH11 significantly repressed overall AHL production, dictating the range of AHLs synthesized. The loss of SsbR resulted in increased abundance of the AHL signals (Table 2.3, Fig. 2.2E), similar to the ssbI? mutant, although this was not clear from the bioassays (Fig. 2.1B). 2.4.4 Ectopic expression of AHL synthases in a KLH11 QS mutant A double mutant with in-frame deletions of both ssaI and ssbI was generated in KLH11. In contrast to the ssaI? mutant, this double mutant surprisingly retained low-level synthesis of several AHLs (Fig. 2.5C), suggesting the 77 presence of an additional unidentified AHL synthase. Indeed, we now know that a third LuxI- gene is still present in this strain and is likely to be responsible (See Chapter 3). The introduction of Plac-ssaI or Plac-ssbI plasmids into the "ssaI "ssbI mutant (Fig. 2.5D and 2.5E) was consistent with the E. coli experiments in that both enzymes drove synthesis of several different long chain signals. SsaI directed synthesis of C16 derivatives 3-oxo-C16-HSL (or possibly 3-OH-C16:1- HSL, again too low level to distinguish) and SsbI directed synthesis of 3-OH- C14:1-HSL and 3-OH-C14-HSL (Fig. 2.3C, Table 2.3). The mutant expressing ssaI produced longer chain length AHLs with greater hydrophobicity than it did with ssbI, but as in E. coli the Plac-ssbI plasmid resulted in much greater overall amounts of AHLs. Alignment of SsaI and SsbI sequences with several other LuxI homologues (Fig. 2.6) shows good conservation of their N-terminal halves (Watson et al., 2002; Gould et al., 2004). Previous studies have found that a threonine residue at position 143 (LuxI numbering) correlates well with the production of 3-oxo- HSLs (Watson et al. 2002). Interestingly, SsaI has a threonine at the equivalent position (SsaI 145), whereas SsbI has a glycine (SsbI 136) at this position (Fig. 2.6), consistent with the observed AHL profiles (Table 2.3). The SsaI C-terminal half is significantly longer than typical for LuxI-type proteins. Additionally, SsbI and SsaI vary considerably in a conserved sequence block (126-157, LuxI numbering), as well as more C-terminal to this block. This region of the enzyme is important for acyl-chain recognition and both SsaI and SsbI clearly deviate 78 here from other better-studied AHL synthases (Watson et al., 2002; Gould et al., 2004). 79 Figure 2.5. Mass spectrometric analysis of plasmid-expressed SsaI and SsbI-directed AHLs from E. coli and KLH11. Cultures of E. coli MC4100 and a KLH11 derivative with in-frame deletions of ssaI and ssbI expressing plasmid- borne ssaI or ssbI were extracted and subjected to reverse-phase chromatographic separation prior to tandem MS analysis using the precursor ion- 80 scanning mode (transitions were monitored for precursor [M + H]+-> m/z 102.1) for (A) E.coli MC4100 + pSsaI (Plac-ssaI). (B) E.coli MC4100 + pSsbI (Plac-ssbI) (C) KLH11 "ssaI "ssbI, (D) KLH11 "ssaI "ssbI + pSsaI (Plac-ssaI), (E) KLH11 "ssaI "ssbI + pSsbI (Plac-ssbI). The peaks in the chromatograms are labeled with lowercase lettering and include species defined in Table 2.3 and the AHLs noted. Figures are provided by Mair Churchill. Figure 2.6. Sequence alignment of AHL synthases. The grey shaded regions are the most conserved sequence blocks within the AHL synthase family. Residues are colored red to indicate acidic or hydrophilic, blue for basic, and orange for other. Shaded residues are absolutely conserved and the boxed residues are the most similar regions within the family. The black inverted triangle indicates the threonine at 143 of LuxI from V. fischeri. The numbers on top of the sequences refer to the numbering of LuxI and the number 284 indicates the length of SsaI. Figure is provided by Mair Churchill. 81 Table 2.3. AHLs: Retention times, identification and relative abundances based on tandem mass spectrometry. 82 2.4.5 Expression of ssaI is stimulated in response to KLH11 AHLs LuxI-type genes often are regulated in response to their own cognate AHL(s) (Fuqua and Greenberg, 2002). Cultures of each of the KLH11 Campbell insertion mutations for ssaI and ssbI which created lacZ fusions to the disrupted genes were assayed for #-galactosidase activity grown in the presence and in the absence of exogenous KLH11 culture extracts. KLH11 has no significant endogenous #-galactosidase activity (Miller Unit <1). The ssaI-lacZ fusion exhibited close to 16-fold induction (Table 2.4) when cultures were incubated in the presence of KLH11 extracts (2.5% v/v). Given that the dominant SsaI- directed are C16-HSLs, we also tested ssaI expression of the mutant with synthetic 3-oxo-C16:1 "11-HSL. In the presence of the synthetic AHL at 2 !M, the ssaI-lacZ fusion was induced roughly 40 fold (P <0.001, Table 2.4). In contrast, the expression of the ssbI-lacZ fusion was low and was not increased by addition of culture extracts or by 20 !M 3-OH-C14-HSL (P>0.05), the dominant long chain hydroxylated AHL produced via SsbI. However, the ssaI- lacZ fusion was induced roughly 30% (P<0.01) by addition of 20 !M 3-OH-C14- HSL, suggesting limited cross-recognition of this AHL. The ssbI-lacZ fusion was not activated by addition of 2 !M 3-oxo-c16:1 "11-HSL. We also created similar Campbell insertion mutants of the ssaR and ssbR genes. For each gene one derivative was created in which the plasmid was integrated to generate the lacZ fusion with the wild type coding sequence intact (JZ1 and JZ2), and a second derivative in which the gene was disrupted by 83 integration of the plasmid (EC4 and EC5). Although ssaR was expressed more strongly than ssbR (based on #-galactosidase activity), there was no effect of crude culture extracts or synthetic AHLs on the expression of these genes in the wild type or the null mutant background (Table 2.4). Disruption of the ssaR gene expression resulted in an approximately 4-fold decrease of ssaR expression, but this was independent of AHL. Table 2.4. QS regulator expression in KLH11 null mutants. ! -galactosidase activity1 Mutants Genotype No extract 2.5% No AHL +AHL2 EC2 ssaI-lacZ, null 6 (1.7) 95 (10.6) 6.3 (0.7) 213.3 (9.2) EC3 ssbI-lacZ, null 1 (<1) <1 (0.1) 1.1 (0.3) 1.1 (<0.1) JZ1 ssaR-lacZ, WT 107.1(0.8) 105.0(4.4) 93.7 (8.3) 99.3 (2.8) EC4 ssaR-lacZ, null 23 (3) 19 (1.4) 26.7 1.3) 28.6 (2.4) JZ2 ssbR-lacZ,WT 11.6 (2.8) 10.8 (3.1) 12.0 (.5) 12.6 (0.5) EC5 ssbR-lacZ, null 15 (<1) 10 (4.6) 27.6 (4.5) 26.0 (4.2) 1Specific activity in Miller units, averages of assays in triplicate (standard deviation). 2 2 "M 3-oxo-c16:1 !11-HSL was added for ssaRI and 20 "M 3-OH-C14-HSL was added for ssbRI. Elisha M. Cicirelli contributed the data for EC2, EC3, EC4 and EC5 with organic extract. 2.4.6 SsaR activates expression of its cognate AHL synthase gene ssaI It was unclear whether AHL-activated ssaI expression also required the SsaR protein and whether the Ssa system might directly influence ssbI expression. 84 Plasmid-borne copies of each luxR-type protein paired with compatible plasmids carrying either a PssaI-lacZ fusion or a PssbI-lacZ fusion, were introduced into the AHL-, plasmidless derivative A. tumefaciens NTL4. Cultures of these A. tumefaciens derivatives were grown in the presence or absence of 2 !M 3-oxo- C16:1 "11-HSL. The presence of ssaR activates ssaI in the absence of exogenous AHLs (~7-fold induction; P<0.001) (Table 2.5). This SsaR-dependent activation was stimulated a total of 30-fold (P<0.001) by the addition of 2 !M 3- oxo-C16:1 "11-HSL (Table 2.5). Several C14-HSLs were also tested, but only weakly influenced ssaI expression (Fig. 2.7). Activation of the ssaI-lacZ fusion by SsaR exhibited a dose-dependent response to 3-oxo-C16:1 "11 -HSL that paralleled the response to crude culture extracts (Fig. 2.7). In contrast, the PssbI- lacZ (pEC121) was not activated by Plac-ssbR (pEC123) irrespective of the presence of synthetic AHLs (<1 Miller Unit, P>0.05) or crude KLH11 extracts (Table 2.5, E. Cicirelli and C. Fuqua, unpublished.). Likewise, SsaR failed to activate ssbI and SsbR failed to activate ssaI in the presence of culture extracts and synthetic AHLs (Table 2.6). 85 Table 2.5. Expression of KLH11 PssaI and PssbI promoters in an AHL- host1 ! -galactosidase activity2 Expression plasmid Fusion plasmid No AHL +AHL3 Vector (pBBR1-MCS5) ssaI-lacZ (pEC116) 51 (5) 52(5) Plac-ssaR (pEC112) ssaI-lacZ (pEC116) 342(15) 1435(67) Vector (pBBR1-MCS5) ssbI-lacZ (pEC121) <1 (<0.1) <1 (<0.1) Plac-ssbR (pEC123) ssbI-lacZ (pEC121) <1 (<0.1) <1 (<0.1) 1All strains derived from Ti-plasmidless A. tumefaciens NTL4. 2Specific activity in Miller Units, averages of assays in triplicate (standard deviation). 3 2 !M 3-oxo-C16:1 "11-HSL was added for ssaI and 20 !M 3-OH-C14-HSL was added for ssbI. 86 Table 2.6. Cross-regulation experiments for SsaR and SsbR in an AHL- host1. ! -galactosidase activity2 Expression plasmid Fusion plasmid No AHL +AHL3 Vector (pBBR1-MCS5) ssbI-lacZ (pEC121) 0.7 (0.1) 0.8 (0.1) Plac-ssaR (pEC112) ssbI-lacZ (pEC121) 0.9 (0.1) 0.9 (0.2) Vector (pBBR1-MCS5) ssaI-lacZ (pEC112) 57.9 (0.9) 58.5 (3.6) Plac-ssbR (pEC123) ssaI-lacZ (pEC112) 50.6 (3.1) 44.3 (2.6) 1All strains derived from Ti-plasmidless A. tumefaciens NTL4. 2Specific activities in Miller Units, averages of assays in triplicate (standard deviation). 3 2 !M 3-oxo-C16:1 "11-HSL was added for Plac-ssaR and 20 !M 3-OH-C14- HSL was added for ssbI. 87 Figure 2.7. Activation of ssaI in response to synthetic AHLs. (A) Activation of ssaI-lacZ fusion with synthetic long chain AHLs. KLH11 ssaI- carrying the integrated ssaI-lacZ fusion. The final concentration of 3-OH-C14-HSL used was 20 ?M and the other 3 AHLs were 2 ?M. (B) Dose responsive activation of ssaI- lacZ by SsaR (Plac-ssaR) in A. tumefaciens NTL4 background. Different concentrations of the crude organic extract of KLH11 cultures (%, v/v) and 3-oxo- C16:1 "11-HSL (nM) were added at the time of culture inoculation. Values are averages of assays performed in triplicate and error bars are standard deviations. Elisha M. Cicirelli provided the dose response curve with extract. 88 2.4.7 Conserved sequences upstream of ssaI are required for activation by SsaR LuxR homologs often recognize conserved sequence elements, called lux- type boxes located upstream of target promoters, including those of luxI homologues (Devine et al., 1989; Egland and Greenberg, 1999). The ssaI gene is downstream of the ssaR gene in a tandem arrangement, with an intergenic region of 118 bp but reverse transcription (RT)-PCR assays demonstrated that ssaI and ssaR are not in the same operon (Fig. 2.4A; similarly, ssbI and ssbR are also in separate operons, Fig. 2.4B). Inspection of the sequence upstream of ssaI revealed no inverted repeats and no motifs with primary sequence similarity to bona fide lux-type boxes. Comparison of this region between ssaR and ssaI and the homologous region between silR1 and silI1 from R. pomeroyi DSS-3 (Moran et al., 2004) revealed only 60% identity, except for a 19 bp fully conserved segment (TACGGGAAACCCCCAATAG), located 60 bp upstream of the ssaI start codon (Fig. 2.8A). Although it is not an inverted repeat and shares limited primary sequence similarity with known lux-type boxes, we reasoned the sequence might be a regulatory element given its appropriate size and location, and tentatively designated this a ssa box. Deletions were generated in the AHL and SsaR-responsive PssaI-lacZ plasmid (pEC116), one to just upstream of the ssa box (pEC124) and a larger deletion (pEC127) that almost completely removes the element (Fig. 2.8B). In A. tumefaciens NTL4 the deletion construct retaining the ssa box (pEC124) was inducible by ssaR and 3-oxo-C16:1 "11-HSL to the same extent as the plasmid with the complete intergenic region. Deletion 89 of the ssa box abolished this induction, although it increased basal expression levels (Fig. 2.8C). These results reveal a role for the presumptive ssa box in AHL-dependent activation of the ssaI gene. 90 Figure 2.8. Deletion analysis of the ssaI promoter. (A) Putative promoter elements are indicated in boldface and putative -35 and -10 regions are indicated below the sequences; +1 indicates the predicted transcription start site. Presumptive Ssa box is underlined. (B) The presumptive ssa box is indicated within the rectangle. Plasmid name and size of insert are indicated adjacent to the translational start site. For the plasmid pEC116, it also included 103 bp upstream of the Ssa box. Lower case sequence indicates an EcoRI restriction site. (C) #-galactosidase activity of the three deletion constructs fused with the lacZ reporter. All strains derived from Ti-plasmidless A. tumefaciens NTL4. Bars represent the average of three biological replicates and the error bars are 91 standard deviation. The filled asterisks show the statistical significance between samples, with and without 3-oxo-C16:1 !11-HSL, in the strains that have SsaR (P<0.001). The unfilled asterisk shows the statistical significance of the basal expression levels between pEC127 and each of pEC116 and pEC124 (P<0.001). The result presented is a representative of several independent experiments each with three biological replicate. Elisha M. Cicirelli provided A and B. 2.4.8 SsaRI controls swimming motility and flagellar biosynthesis For several different bacteria, QS regulates bacterial motility and flagellar synthesis (Kim et al., 2007; Ng and Bassler, 2009). KLH11 swims under laboratory conditions but does not exhibit swarming (E. Cicirelli and C. Fuqua, unpublished). All four KLH11 mutants (!ssaI, ssbI, ssaR?, and ssbR?) were tested for their motility on MB 2216 swim agar plates (Fig. 2.9A). The !ssaI and ssaR? mutants did not migrate from the site of inoculation, whereas the ssbI- and ssbR? mutants migrated through the motility agar similar to wild type. The swimming deficiency of the !ssaI mutant was fully complemented with a plasmid- borne copy of the gene, and the ssaR?, mutant was partially complemented by similar provision of ssaR. Addition of 2 "M 3-oxo-c16:1 !11-HSL into swim agar can partially restore motility in the !ssaI strain, albeit less efficiently than through complementation (Fig. 2.9 A). As expected, motility was not restored in the ssaR? mutant in the media with 2 "M 3-oxo-C16:1 !11-HSL (E. Cicirelli and C. Fuqua, unpublished). The observed differences in migration through swim agar 92 were not likely due to growth effects as the ssaI? and ssaR? mutants grow at the same rate as wild type (Fig. 2.10). In order to visualize the presence of flagella, the wild type and mutant strains were observed by phase contrast microscopy using both wet mounts and flagellar stains. Interestingly, early stage wild type cultures did not have flagella, and no swimming cells were observed, whereas in late stage cultures flagella were clearly assembled (Fig. 2.11) and cells were visibly motile. Both the ssaI? and the ssaR?mutants lacked flagella and were never observed to swim, irrespective of culture stage. The cells also appeared to clump more readily. The ssbI? and ssbR? mutants were as motile as wild type and had abundant flagella in late stage cultures (Fig. 2.11). The presence of flagellar proteins in KLH11 cultures was examined using immunoblotting with antisera raised against whole flagella from Caulobacter crescentus, a related alpha-proteobacterium. In western blots from KLH11 late stage cultures, supernatants contained a protein of approximately 43 kDa (Fig. 2.9B) that was also present from pelleted cells. This protein matches the predicted 41.5 kDa size of the only flagellin homologue in KLH11, fliC gene product (Zan et al., 2011a)(See Appendix 1). A site-specific disruption of the KLH11 fliC homologue abolished swimming motility (unpublished results) and caused loss of the 43 kDa protein, the same presumptive flagellin protein was also absent from the "ssaI mutant (Fig. 2.9). To determine when flagellar biosynthesis occurs during culture growth, samples were harvested along the growth curve at four different time points; mid- 93 exponential, early stationary, mid-stationary, and late stationary phase from the parent strain and the ssaI- mutant. At an OD600 of 0.5, the parent strain had no detectable flagellin in either the whole cell or supernatant fraction, but as the culture density increased (OD600 >1.3), flagellin was detected in the whole culture fractions and weakly in the culture supernatant and continued to increase as the culture grew (Fig. 2.9C). Flagellin was never detected in samples of the ssaI?, mutant, irrespective of culture density (E. Cicirelli and C. Fuqua, unpublished). To determine if the SsaRI system regulates the transcription of the flagellin gene (fliC), quantitative RT- PCR (qRT-PCR) was used to measure fliC expression. Late stage cultures grown to an OD600 at which KLH11 produces visible flagella have approximately 2 orders of magnitude higher fliC expression in wt KLH11 compared to the !ssaI mutant (Fig. 6D, P<0.001). The Plac-ssaI plasmid (pEC108) complemented fliC expression compared to the !ssaI mutant (P<0.05), although not to full wild type levels (Fig. 2.9D). Silicibacter sp. TM1040 is well studied for motility (Belas et al., 2009), and flaA is a required motor- associated protein in a putative Class II flagellar operon. Examination of the KLH11 flaA homologue by qRT-PCR revealed a 10-fold decrease of flaA expression in the ssaI deletion mutant (P<0.001). A plasmid-borne ssaI copy was able to partially restore flaA expression (Fig. 2.9D, P<0.05). These results suggest that an intact SsaRI system is required for swimming through regulation of expression of flagellar genes. 94 Figure 2.9. Regulation of swimming motility and flagellar biosynthesis by the SsaRI QS system. (A) Swimming motility assays for QS mutants on 0.25% Marine Agar 2216 after 8 days at 28?C. 2 !M 3-oxo-C16:1 "11?HSL was added for the !ssaI mutant and pEC108 (Plac-ssaI) and pEC112 (Plac-ssaR) were used to complement !ssaI and ssaR?, mutants, respectively. (B) Antiserum raised against whole flagella from C. crescentus also recognizes KLH11 flagella. Estimated size of KLH11 flagellin is 43 kDa. (C) Flagellin synthesis during late culture stages requires ssaI. Wild type cultures were harvested at various time points along the growth curve. WC, whole culture; S, supernatant. (D) qRT-PCR results of genes fliC and flaA in wt, the !ssaI strain and complementation strains. Error bars are the standard deviations and results are representative of 95 two independent experiments with triplicates. Okhee Choi provided Fig. 2.9 B and Elisha M. Cicirelli provided Fig. 2.9C Figure 2.10. Growth curves of KLH11 and different QS mutants. Average OD600 of 5 ml cultures in MB2216. Cultures were grown in triplicate. Figure was maded based on the data provided by Elisha M. Cicirelli. 96 Figure 2.11. Flagellar staining of KLH11 quorum sensing mutants. Stained cells from late stage cultures were viewed under phase contrast microscopy with 100X lens. Wild type (EC1), ssaI? (EC2), ssbi? (EC3), ssaR? (EC4) and ssbR? (EC5). Red arrows indicate stained flagella. Figure is provided by Elisha M. Cicirelli and Clay Fuqua. 2.4.9 SsaRI mutants exhibit increased biofilm formation QS can also regulate bacterial biofilm formation (Hammer and Bassler, 2003; Shrout et al., 2006). Given the importance of the SsaRI system to swimming motility, we questioned if QS might influence KLH11 biofilm formation. A static coverslip biofilm assay was performed for the ssa and ssb mutants compared to the wild type. Crystal violet stained biofilms on PVC coverslips were solubilized in 33% acetic acid, and the absorbance at 600 nm (A600) was measured. Relative to wild type, the "ssaI and ssaR? mutants clearly have increased biofilm 97 formation, most pronounced by 48 h post-inoculation (Fig. 2.12A, P<0.01). It was plausible that increased biofilm formation was due to the loss of motility in these mutants, limiting emigration of bacteria from biofilms. Indeed, biofilm formation in the non-motile fliC?, mutant was also modestly increased compared to wild type (Fig. 2.12A, P<0.01). However, biofilm formation in the "ssaI fliC?, mutant was even stronger than the non-motile fliC mutant (P<0.05). This suggests that the increased biofilm formation in the "ssaI mutant is not entirely due to the lack of motility. The ssbI and ssbR mutants formed biofilms that were indistinguishable from wild type KLH11 (Fig. 2.12A, P>0.05). Addition of KLH11 late stage culture fluids (40% v/v) into biofilm assays with the "ssaI mutant reduced biofilm formation to 60% the level of untreated cultures (P<0.05) whereas planktonic growth is unaffected (Fig. 2.12B, P>0.05). This inhibition effect is not due to nutrient depletion because 1X MB2216 was provided in addition to the nutrients remaining in the supernatant. The inhibition is also not due to pH changes as there was <0.1 pH unit difference between normal MB2216 and MB2216 conditioned with wt KLH11 culture fluid. 98 Figure 2.12. Increased biofilm formation in ssaI and ssaR mutants. (A) Standard 48 hour coverslip biofilm assays measuring A600 of solubilized crystal violet normalized to the culture density (OD600) for the different KLH11 strains were performed. Filled asterisks indicate statistical significance between wt KLH11 and indicated strains (P<0.01). Open asterisk indicates statistical significance between fliC- and "ssaI fliC- mutants (P<0.05). Error bars are standard deviation of three biological replicates. (B) Inhibition of biofilm formation for DssaI mutant by wt KLH11 culture fluids. Different % (v/v) wt KLH11 culture fluids were added into "ssaI culture at the time of inoculation. Solubilized crystal violet stain of adherent biomass (A600) and the optical density of the cultures (OD600) were measured as a function of percent culture fluid addition. Data were normalized to the cultures with no added wt KLH11 culture fluids. 99 2.5. Discussion There is limited understanding of QS mechanisms in the diverse and abundant bacteria living in the marine environment (Cicirelli et al., 2008). Our studies on sponge symbionts initiated with the use of AHL responsive bioassays (Mohamed et al., 2008c). We have developed Ruegeria sp. KLH11, a member of the Silicibacter?Ruegeria group of the Roseobacter clade, as a quorum sensing model. KLH11 AHLs are readily detectable using the A. tumefaciens bioreporter, ranging from non-polar, long chain AHLs to more polar, short chain AHLs (Fig. 2.1; Mohamed et al., 2008c). However, even the broadly responsive A. tumefaciens system is biased towards AHLs structurally similar to its cognate 3- oxo-C8-HSL, and often cannot distinguish between related AHLs. We therefore employed mass spectrometric analysis to provide unbiased AHL identification that also provides information on relative abundance for at least a subset of AHLs. This approach detected AHLs with acyl chains greater than C12 from KLH11, but AHLs with shorter acyl chains revealed in bioassays were below the effective detection threshold for our technique. This finding highlights the strengths and weakness of each detection system; the AHL-responsive bioassays are extremely sensitive, but difficult to quantitate, whereas direct chemical detection can identify and quantitate AHLs, but at lower sensitivity. In the end, the combined approach provided the best insights into this complex system. KLH11 AHLs were predominantly 12-16 carbons, some with unsaturated bonds, with either 3-OH or 3-oxo substituents. Although SsaI-directed oxo-AHLs 100 were not detectable in wild type KLH11 extracts by the mass spectrometric approach (Fig. 2.2; Table 3.1), they are clearly present in the TLC assays (Fig. 2.1). The relatively abundant 3-OH-C14-HSL and 3-OH-C14:1-HSL likely originate from SsbI or the closely related SscI. The hydroxy-, oxo- and double bonded character of the long chain AHLs would be beneficial for maintaining their solubility, and longer acyl chains are more stable in moderately alkaline marine environments (Riebesell et al., 2000; Yates et al., 2002), consistent with other studies from marine systems (Wagner-D?bler et al., 2005). The pattern of AHLs specified by SsaI and SsbI, 3-oxo-HSLs and 3-OH- HSLs, respectively, vary in their chain length depending on the bacterial species in which they are expressed. For example, in E. coli harboring the Plac-ssaI plasmid, 3-oxo-C14-HSL was the most abundant AHL, and 3-oxo-C16 ?HSL was present at 15% (Table 2.3). This relationship reversed when the same plasmid was expressed in the KLH11"ssaI"ssbI mutant, resulting in high relative levels of 3-oxo-C16-HSL, with 3-oxo-C14 at less than 4% of its level in E. coli. Likewise, E. coli harboring the Plac-ssbI plasmid resulted in high levels of 3-OH-C12-HSL and detectable 3-OH-C13-HSL among the other OH-AHLs, whereas these AHLs were much less abundant when the same plasmid was expressed in the KLH11"ssaI "ssbI mutant (Table 2.3). This shift toward increased acyl-chain lengths in KLH11 relative to those in E. coli may be explained by the differences in acyl-ACP substrate pools available in the different strains. Indeed, this has been observed in cases of AHL synthase over-expression in E. coli, where unusual AHLs including those with odd-chain lengths have been observed 101 (Gould et al., 2006). Differences in growth temperature may also influence the levels and distribution of AHLs (Yates et al., 2002). However, expression in E. coli enables analysis of the intrinsic specificity of each AHL synthase and was important in some cases for AHL identification. Bacteria that have multiple AHL-based QS systems can organize these systems in interconnected regulatory networks (Atkinson et al., 2008). The results of our AHL chemistry, genetic analysis, and gene expression experiments reveal a complex network of signal production and regulation in KLH11 (Fig. 2.13). The ssaI null mutant loses the majority of detectable AHLs, as evaluated by bioassays and mass spectrometry (Figs 2.1and 2. 2, Table 2.3). The AHLs produced by wild type KLH11 are dominated by OH-AHLs, those synthesized by SsbI (and perhaps SscI), but the ssaI mutant phenotype suggests that it influences production of these AHLs. Surprisingly, and in contrast to the ssaI null phenotype, mutation of ssbI increases the overall AHL activity. Mass spectrometry reveals a large increase in a 3-oxo-C16-HSL, correlated with SsaI activity, which is lost in the "ssaI "ssbI mutant (Table 2.3). These findings suggest that SsbI exerts a suppressive effect on the Ssa system, and this is relieved by mutation of ssbI (and perhaps also ssbR). Several lines of evidence suggest that the connections between the Ssa and Ssb pathways are not directly through transcriptional control. Neither system appears to directly influence expression of the other AHL synthase genes. In fact, although SsbI-directed AHLs are lost in the ssaI? mutant, the ssaR? mutant does not affect their level, indicating that SsbI control is independent of SsaR. In 102 contrast, both SsaR and the SsaI-specified AHLs are required to autoregulate ssaI expression, activate motility and inhibit biofilm formation. It is plausible that an unidentified LuxR-type protein regulates the Ssb system in response to SsaI- produced AHLs, similar to the P. aeruginosa QscR protein, a LuxR-type transcription factor that responds to AHLs produced via the LasI-LasR system to control its target genes (Fuqua, 2006; Lequette et al., 2006). The genome sequences of KLH11 and R. pomeroyi DSS-3 reveal several additional solo-type LuxR-type proteins that might be functioning in response to SsaI-directed AHLs to influence the Ssb pathway. Inhibition of Ssa activity by the Ssb system is potentially due to competition for common substrates. Our findings demonstrate that both AHL synthases catalyze production of long chain AHLs and may be in competition for long chain acyl-acyl carrier protein conjugates (Churchill and Chen, 2011). In P. aeruginosa, AHLs derived from LasI can directly block the activation of another LuxR homologue RhlR by its own cognate AHL (Pesci et al., 1997). It was plausible that Ssb-derived AHLs may have had an inhibitory effect on SsaI activity, however addition of 3-OH-C14-HSL to KLH11 resulted in a modest activation of ssaI expression, and thus this is not likely. Our experiments implicate a non-symmetrical sequence element immediately upstream of ssaI that is required for SsaR activation (Fig. 2.8C). LuxR homologs have been shown to bind to symmetric (Zhang et al., 2002) as well as asymmetric R-boxes (Schuster et al., 2004). In contrast, there is no indication of AHL-responsive autoregulation for ssbI (Table 2.4) nor when SsbR and the ssbI- 103 lacZ fusion are tested in A. tumefaciens (Table 2.6). The ssbI gene joins a small list of LuxI homologues that are not positively autoregulated (von Bodman and Farrand, 1995). A growing number of bacteria including Rhizobium etli, Serratia liquefaciens and P. aeruginosa are recognized to control motility via quorum sensing, including swarming, twitching and swimming (Eberl et al., 1996; Daniels et al., 2004; Shrout et al., 2006). Many of these systems are inhibited by AHLs, whereas a smaller number are activated. In S. meliloti QS inhibits swimming motility, whereas for the #-proteobacteria Burkholderia glumae QS activates flagellar biosynthesis and thereby swimming and swarming motility (Kim et al., 2007). In KLH11 flagellar biosynthesis is under tight, positive SsaRI control (Fig. 2.9). Despite their low abundance, the SsaI-derived oxo-AHLs are clearly important to activate swimming in late stage cultures, mutation of ssaI leads to loss of swimming motility, and provision of AHLs alone can rescue the mutant?s swimming deficiency. Clearly, the SsaI-directed AHLs function below our mass spectrometry detection threshold. Some AHL QS systems are tuned to exceptionally low AHL levels, such as the ExpR system from S. meliloti (Pellock et al., 2002). Reconstructing SsaR-dependent gene regulation in A. tumefaciens suggests that as little as 10 nM 3-oxo-C16:1 "11-HSL is saturating (Fig.2.8). Our observations suggest that the influence of the SsaRI system on motility may be needed to limit aggregation. Interestingly, although the KLH11 and R. pomeroyi DSS-3 genomes encode flagellar motility functions, neither have genes for chemotaxis, including Che regulatory proteins and methyl-dependent 104 chemotaxis proteins (Moran et al., 2004; Zan et al., 2011a)(See Appendix 1). For KLH11, and perhaps R. pomeroyi DSS-3, motility may function to promote dispersal from aggregates instead of chemotaxis, employing QS control to provide a population density response. A similar function was proposed for QS in the photosynthetic microbe Rhodobacter sphaeroides (Puskas et al., 1997). How does SsaR control flagellar assembly and function in KLH11? Both flaA, and fliC, class II and class III flagellar genes, respectively, require SsaI for significant expression (Fig. 2.9D), which suggests that in KLH11 QS controls motility at an early step in flagellar gene expression. In most flagellated bacteria there is a primary regulator that initiates expression of the flagellar gene cascade (Macnab, 1996). Although it is conceivable that SsaR could be this master regulator in KLH11, it is more likely that it controls expression of another regulator. In E. coli and several other bacteria FlhDC proteins serve as the primary regulators of flagellar assembly (Soutourina and Bertin, 2003), but there are no FlhDC homologues in the KLH11 or R. pomeroyi DSS-3 genomes (Moran et al., 2004; Zan et al., 2011a)(See Appendix 1).. For C. crescentus, the essential cell cycle master regulator CtrA initiates flagellar assembly (Muir and Gober, 2004). KLH11 encodes a ctrA homologue, and in its relative Silicibacter sp. TM1040, this gene is also required for flagellar activity (Belas et al., 2009). It is well established that bacterial motility can have a profound impact on surface adherent biofilm formation (O?Toole and Kolter, 1998; Merritt et al., 2007). In P. aeruginosa, QS promotes biofilm maturation and QS mutants can attach but do not differentiate into mature biofilm structures (Davies et al., 1998). 105 In contrast, in V. cholerae QS inhibits biofilm formation by decreasing the expression of exopolysaccharide (EPS) genes, and may also promote dispersal from biofilms and in late infection stages (Hammer and Bassler, 2003; Waters et al., 2008; Krasteva et al., 2010). Accumulation of bacteria on a surface is the net sum of attachment, growth, and emigration. Decreased motility may reduce biofilm formation by limiting initial surface contact, but it also reduces migration from biofilms, thereby increasing biofilm formation. For KLH11, the ssaI and ssaR mutants are non- motile, and have increased biofilm formation (Fig. 2.12A), which is clearly due to QS as KLH11 supernatants antagonize biofilm formation in a dose-dependent manner (Fig.2.12B). The "ssaI mutant has a much more pronounced deficiency than the equally non-motile fliC (flagellin) mutant and the "ssaI fliC? double mutant has a biofilm phenotype indistinguishable from "ssaI itself. These findings suggest that the increased biofilm formation in ssaI and ssaR mutants is not only due to a loss of motility. Any additional relevant SsaRI target(s) remains to be determined but might include genes involved in surfactant synthesis or modulation of internal signaling molecules such as cyclic di-guanosine monophosphate (c-di-GMP) (Davey et al., 2003; Hengge, 2009). Searches of the KLH11 genome for matches to the ssa box identified here have failed to yield potential targets. Our study has revealed the coordinated regulation of motility and biofilm formation by QS in a roseobacterial sponge symbiont. From the larger ecological perspective, the regulatory pattern we have observed and the dispersal model 106 we have proposed makes sense. In nature, sponges actively pump large volumes of the surrounding seawater. Microbial symbionts obtained from seawater at this stage likely do not require flagella to be introduced into the sponge host. Once KLH11-type bacteria colonize the sponge and are provided a nutrient rich environment, they begin to grow to high density, perhaps even beginning to aggregate. QS activation of motility and adherence inhibition may facilitate dispersal from these crowded and potentially limiting microenvironments, and to promote more uniform colonization of the host tissue or even stimulate release back into the water column. By coordinating motility and biofilm formation, motile KLH11 cells can readily escape from their own aggregates. Experiments examining the colonization and distribution of KLH11 in live sponges, and tracking marked KLH11 QS mutants, may be the most direct approach to test these ecological hypotheses, and provide further insights into to SsaRI and SsbRI systems. 107 Figure 2.13. A model for the complex regulatory control of QS circuits in KLH11. The ssaRI, ssbR, and sscI genes are drawn according to scale while fliC and flaA are not. The black dots represent the AHLs synthesized by SsaI, mainly long chain (lc) 3-oxo-HSL. The grey dots represent the AHLs synthesized by SsbI, mainly long chain (lc) 3-OH-HSL. The lines with bars indicate inhibition while the arrows indicate activation. Squiggly lines indicate translation of genes or products of enzyme action. Dr. Clay Fuqua and I together made this figure. 108 Chapter 3. A LuxI ?solo? synthesizes long chain acyl homoserine lactones and is involved in a reciprocal regulatory quorum sensing system in marine sponge symbiont Ruegeria sp. KLH11 109 3.1. Abstract Marine sponges harbor abundant and diverse bacterial communities, providing an ideal environment for bacterial cell-density dependent cell-cell signaling, termed quorum sensing. Ruegeria sp. KLH11, isolated from the marine sponge Mycale laxissima, produces mainly long chain acyl-homoserine lactones (AHLs) and has been developed as a quorum-sensing model for roseobacterial sponge symbionts. Two pairs of luxR/I homologues were identified by genetic screening and were designated ssaRI and ssbRI (sponge- associated symbiont locus A or B, luxRI homologue) (Chapter 2). However, a double deletion mutant of the AHL synthase genes ssaI ssbI produced lower amounts of AHLs, suggesting that a third (or more) AHL synthase gene must be present. In this study, annotation of the KLH11 genome identified a third AHL synthase gene, named sscI. A triple mutant of the three AHL synthase genes showed no AHL production detectable by mass spectrometry, indicating that there are no additional AHL synthases in KLH11. Interestingly, sscI does not have a cognate LuxR homologue present in an adjacent locus and thus sscI is an AHL synthase solo. Expression of sscI in E. coli MC4100 shows that it produces mainly long chain 3-OH-AHLs, dominated by 3-OH-C14-HSL and 3-OH-C14:1- HSL. A genetic study showed that both ssaI and ssbI positively regulate sscI and that this regulation is indirect. SscI-specified AHL stimulated ssaI expression and has a moderate effect on the KLH11 swimming motility phenotype. Furthermore, KLH11 likely encodes a novel enzyme responsible for the synthesis of an aryl- HSL, p-coumaroyl-HSL (pC-HSL). 110 3.2. Introduction In many bacteria, a luxR-type gene without a cognate luxI-type gene has been discovered and these luxR-type genes are termed luxR solos. These solos occur both in bacteria that have a complete QS system and bacteria that do not (Subramoni and Venturi, 2009). LuxR solos can regulate expression of a set of genes by binding to AHL produced by other luxI homologues in the bacterium such as in the case of QscR in Pseudomonas aeruginosa or by binding to AHLs produced by other bacteria (Lequette et al., 2006). Furthermore, LuxR-type proteins can regulate gene expression independent of ligand binding (Subramoni and Venturi, 2009). Although progress has been made in understanding LuxR solos, there is little information about LuxI solos. These are proteins encoded by luxI homologues without cognate luxR homologoues. Our previous studies have genetically identified two pairs of genes encoding LuxI-LuxR QS systems, SsaRI and SsbRI, in the marine sponge symbiont Ruegeria sp. KLH11 (Chapter 2). However, the double deletion mutant of both AHL synthase genes is still able to produce some AHLs, such as 3-OH-C14-HSL, indicating that a third or more AHL synthases must be present. By mining the genome sequence of KLH11, we were able to identify a third luxI-type gene, sscI (sponge-associated symbiont locus C, luxI homologue), which does not have a cognate luxR homologue. Here, we report the chemical profile of AHLs synthesized by SscI and the involvement of sscI in the complex regulatory QS system in KLH11. Furthermore, we also try to test whether KLH11 can produce the novel p-coumaroyl-HSL (pC-HSL) molecule that 111 was originally discovered in R. palustris, which has a ring structure in the fatty acid chain and requires the substrate p-courmarate produced by plants for the synthesis by the LuxI homologue RpaI (Schaefer et al., 2008). 112 3.3. Experimental procedures 3.3.1 Bacterial strains, oligonucleotides and growth conditions Bacterial strains and plasmids used in this study are listed in Table 3.1 and oligonucleotides were obtained from Integrated DNA Technologies (Coralville, IA). Unless stated otherwise E. coli strains were grown in Luria-Bertani (LB) broth at 37?C with aeration, Ruegeria sp. KLH11 strains were grown in Marine Broth 2216 at 28?C (MB2216) (BD, Franklin Lakes, NJ) and A. tumefaciens strains were grown in AT minimal salt medium supplemented with 0.5% glucose and 15 mM (NH4)2SO4 (ATGN) (Temp? et al., 1977). Antibiotics were used at the following final concentrations (?g ml-1): (i) E. coli (ampicillin, Ap, 100; gentamicin, Gm, 25; kanamycin, Km, 25; spectinomycin, Sp, 100), (ii) KLH11 (Km, 100; rifampicin, Rif, 200; Gm 25, Sp, 100) (iii), A. tumefaciens (Gm, 300; Sp, 200). 3.3.2 Plasmid construction for null mutation, expression of sscI and lacZ-fusion The method used to construct the sscI Campbell insertion mutant was similar to that described in Chapter 2. Briefly, an internal fragment of the sscI gene was amplified using forward primer: 5?-GAATCCATGTTTCGCGATCGAGCAGAT-3? (the EcoRI recognition site is underlined) and reverse primer: 5?- GGTACCTCTTGATACTCCCGCTC-3? (the KpnI recognition site is underlined). The PCR amplicon was gel-purified and cloned into pCR 2.1-TOPO vector (Invitrogen, Grand Island, NY) to create pOKC1 and the insert was confirmed by 113 sequencing. For recombinational mutagenesis, pOKC1 was digested with EcoRI and KpnI, and the resulting sscI fragment was ligated to a similarly digested R6K replicon, the pVIK112 suicide vector (Kalogeraki and Winans, 1997), creating pOKC2. pOKC2 was conjugated into KLH11 and transconjugants resistant to kanamycin (Km) were selected and confirmed by sequencing. To construct double and triple AHL synthase gene mutants, pOCK2 was conjugated into $ssaI, $ssbI and $ssaI$ssbI strains, respectively, and the transconjugants were selected and confirmed as described for the sscI single mutant. A controlled expression construct of sscI was generated by PCR amplification of the coding regions using the forward primer: TCTAGACTGAAACAGGAAACAGCTATGCTCCGTTATGTTTTTGCA (the XbaI recognition site is underlined, the stop codon TGA and the start codon ATG are in bold and the E. coli lacZ ribosomal biding sites are in italics) and the reverse primer: CTCGAGTCAAGCGGTTCTTTGAAACTT (the stop codon is in bold and the XhoI recognition site is underlined). The PCR products were ligated into pCR?2.1-TOPO? vector (Invitrogen) to created pOKC3 and confirmed by sequencing. pOKC3 was digested by XbaI and XhoI and the insert was subcloned into the vector pSRKTc (Khan et al., 2008) to create pOKC4. The insert carried by the construct was confirmed by sequencing. The forward primer: 5? GAATTCGCCGAGATGAACTGTTCAAAGAAC-3? (the EcoRI recognition site is underlined) and the reverse primer: GGATCCGAGCATTTTTAACCTCTTGTTCAC (the BamHI recognition site is underlined) annealing 255 bp upstream and 3 bp downstream of the sscI 114 translational start site were used to amplify its promoter. The PCR products were cloned into pCR2.1-TOPO vector and the inserts confirmed by DNA sequencing. The pCR2.1-TOPO derivatives were digested with EcoRI and PstI and the resulting fragments were ligated with pRA301 vector digested with the same restriction enzymes (Akakura and Winans, 2002) to create pOKC8. 3.3.3 Preparation of AHL samples and analysis by RP-HPLC and ESI mass spectrometry The method used was as described in Chapter 2 Section 2.3.2. 3.3.4 Qualitative analysis and estimation of AHL quantities The method used was as described in Chapter Section 2.3.3. 3.3.5 Preparation of log phase cell concentrates and !- galactosidase assays The method used was as described in Chapter 2 Section 2.3.7. The only difference is that p-courmarate was added to MB2216 at the concentration of 0.5 mM when the supernatant of KLH11 and derivatives were assayed for pC-HSL- like activity using R. palustris CGA814 (rpaI-lacZ) as the reporter strain (Schaefer et al., 2008). 3.3.6 Motility assay Bacterial swim assays were performed using MB2216 with 0.25 % (w/v) agar. Plates were inoculated at the center with freshly isolated KLH11 colonies. KLH11 crude organic extract (0.5% v/v) was added to MB 2216 agar. Plates 115 were placed in an air-tight container with a beaker containing 15 ml of K2SO4 to maintain constant humidity, and incubated 5-7 days at 28?C. Photos were taken by using a Nikon D90 camera. 116 Table 3.1. Strains and plasmids used in Chapter 3. Bacteria/Plasmids Relevant featurea Reference E. coli DH5!/"pir Strain for propagating R6K suicide plasmids Lab collection E. coli S17-1/"pir IncP conjugal donor (Kalogeraki and Winans, 1997) E.coli XL-1 Blue Standard alpha-complementation strain Lab collection E. coli MC4100 K-12 derivative, #lacZ (Casadaban, 1976) A. tumefaciens NTL4 Ti plasmidless derivative, nopaline chromosomal background (Zhu et al., 1998) KLH11 wild type (Mohamed et al., 2008c) KLH11-EC1 RifR Zan et al., 2012 KLH11-EC2 ssaI-lacZ, RifR, KmR Zan et al., 2012 KLH11-SK01 $ssaI, RifR Zan et al., 2012 KLH11-EC3 ssbI-lacZ, RifR, KmR Zan et al., 2012 KLH11-SK02 $ssaI $ssbI, RifR Zan et al., 2012 KLH11-OKC2 sscI-lacZ, RifR, KmR This study KLH11-OKC3 $ssbI, RifR This study KLH11-OKC5 $ssbI sscI-lacZ, RifR, KmR This study KLH11-OKC6 $ssaI $ssbI ssc-lacZ, RifR, KmR This study KLH11-OKC7 $ssaI sscI-lacZ, RifR, KmR This study CGA814 Rhodopseudomonas palustris; rpaI-lacZ, KmR Schaefer et al., 2008 pCR2.1-TOPO PCR fragment cloning vector, Ap/KmR Invitrogen pBBR1-MCS5 BHR Plac expression vector, GmR (Kovach et al., 1995) pSRKTc BHR expression vector containing lac promoter and lacIq, TcR (Khan et al., 2008) pVIK112 R6K-based lacZ transcriptional fusion, KmR (Kalogeraki and Winans, 1997) pRA301 BHR lacZ translational fusion vector (Akakura and Winans, 2002) 117 pEC108 pBBR1-MCS5 derivative carrying full length Plac-ssaI, GmR Zan et al., 2012 pEC109 pBBR1-MCS5 derivation carrying full length Plac-ssbI, from pEC110, GmR Zan et al., 2012 pEC112 pBBR1-MCS5 derivative, carrying full length Plac-ssaR, from pEC106, GmR Zan et al., 2012 pEC116 pRA301 derivation, PssaI-lacZ, SpR Zan et al., 2012 pEC121 pRA301 derivative, PssbI-lacZ, Sp/SmR Zan et al., 2012 pEC123 pBBR1-MCS5 derivative, carrying full length Plac-ssbR, GmR This study pOKC1 pCR?2.1-TOPO?, carring internal fragment of sscI, KmR This study pOKC2 pVIK112 derivative, carrying internal fragment of sscI. KmR This study pOKC3 pCR2.1-TOPO?, carrying full length of sscI, KmR This study pOKC4 pSRKTC derivative, carrying full length Plac- sscI, TcR This study pOKC8 pRA301 derivation, PsscI-lacZ, SpR This study aAp=ampicillin, Gm=gentamicin, Km=kanamycin, Rif=rifampicin, Sp=spectinomycin. Tc=tetracycline. 118 3.4. Results 3.4.1 Identification of sscI Previous studies showed that the KLH11 strain deleted for both ssaI and ssbI still produced small amounts of AHL signals (Zan et al., 2012). This result indicated that a third or more AHL synthases must be present in KLH11. Analysis of the complete genome sequence of strain KLH11 revealed a third AHL synthase gene, designated sscI (sponge-symbiont associated locus C, LuxI homologue). It has 214 amino acids (aa) and does not have a cognate luxR gene. The upstream and downstream genes are predicted to be a putative transposase and a hypothetical protein, respectively (Fig. 3.1). Mass spectrometry analysis of the AHL profile of "ssaI "ssbI sscI? mutant showed no AHL production (Fig. 3.2A), confirming that there are likely three and not more AHL synthase genes in KLH11. 119 . Figure 3.1. Gene maps of KLH11 sscI locus. Arrows represent genes. SscI is predicted to be 214 amino acids. The putative transposase is predicated to be 287 amino acids. The scale bar represents 1 kb. 3.4.2 sscI encodes a protein for synthesis of long chain length AHLs and its expression is not stimulated by the KLH11 AHLs The AHL profile given by sscI expression in E. coli MC4100 was determined by the Churchill Laboratory and is shown in Fig. 3.2B. AHLs were confirmed by mass fragmentation to m/z = 102.1, retention time, as well as trimethylsilylation and methoximation of the samples (Clay and Murphy, 1979; Maclouf et al., 1987) indicated that sscI encodes production of 3-hydroxy-HSLs, dominated by 3-OH- C14-HSL and 3-OH-C14:1-HSL. Comparison of the AHLs encoded by sscI to those previously identified for ssaI and ssbI (Zan et al., 2012), showed that the profile of AHL production is strikingly similar to that specified by SsbI (Fig. 3.2C). Sequence alignment showed that SscI shares 81% identity with SsbI but shares only 32% identity with SsaI on the amino acid level (Fig. 3.3), which is 120 consistent with the similar profiles of AHL production derived from ssbI and sscI. The eight well-conserved amino acid residues identified in other LuxI homologues (Churchill and Chen, 2011) can also be identified in SscI (Fig. 3.3). Taken together, ssbI and sscI are presumably homologues while sscI and ssaI are paralogues. A Campbell insertion of sscI created a transcriptional lacZ fusion to the sscI gene. We were able to test whether addition of KLH11 AHLs would be able to increase the expression of sscI. Results showed the sscI was expressed (ca. 300 Miller unit) but did not respond to addition of crude organic extract of KLH11 culture (P>0.05) (Fig.3.4). We did a similar experiment by adding 3-OH-C14- HSL to the final concentration of 20 !M, and no induction for sscI expression was observed either (Miller units were 65.9 ? 3.9 with no AHL and 67.5 ? 3 7.4 with AHL). 121 Figure 3.2. Mass spectrometry analysis of purified samples from triple mutant of !ssaI !ssbI sscI? (A) and from SscI expressed in E. coli MC4100 (B). The products of reverse-phase chromatographic separation of AHLs extracted and purified from different strains were examined using the precursor ion-scanning mode (transitions were monitored for precursor -> m/z 102.1) for MC4100 + Plac-sscI. (C) Comparative analysis of AHLs. The relative amounts of the known AHLs in SscI compared to SsaI and SsbI from Chapter 2 shown in a bar graph. The plot shows the ratio of the area of the transition for each AHL to 122 m/z 102.1 to the same transition for the internal standard D3-C6-HSL. The SsbI and SscI y-axis is on the left and the SsaI y-axis is on the right. Analysis of AHL samples was performed in positive-ion mode with Q3 set to monitor m/z 102.1 as described in Experimental procedures in Chapter 2. The identities of each peak indicated by letters are indicated in Table 2.3. This figure is provided by Mair Churchill. 123 Figure 3.3. Alignment of SscI amino acid sequences to other AHL synthases. The grey shaded regions are the most conserved sequence blocks within the AHL synthase family. Residues are colored red to indicate acidic or hydrophilic, blue for basic, and orange for other. Shaded residues are absolutely conserved and the boxed residues are the most similar regions within the family. The numbers on top of the sequences refer to the numbering of LuxI and the number 284 indicates the length of SsaI. This figure is provided by Mair Churchill 124 3.4.3 sscI is positively regulated by ssaI and ssbI Our previous study showed that the QS pathways in KLH11 are interconnected. When ssaI is knocked out, the AHL production is severely decreased (Zan et al., 2012). The sscI expression level was detected in different QS gene mutant backgrounds. #-galactosidase assay results showed that there was about 3-fold decrease in the expression of sscI in the "ssaI strain or "ssbI strain compared to that of the wild type but there was no additive effect in the "ssaI "ssbI strain on sscI expression (Fig. 3.4). However, the addition of KLH11 crude organic extracts or provision of plasmid-borne SsaI or SsbI did not complement sscI expression at all. Even provision of SsaR and SsbR with KLH11 AHLs did not give complementation of the expression of sscI (Fig. 3.4). The plasmid pEC112, carrying a copy of SsaR, and plasmid pOKC8, carrying PsscI-lacZ, were transformed into the heterologous host A. tumefaciens NTL4. Different amounts of KLH11 AHLs were added but the sscI expression was not affected. Likewise, when SsbR was provided instead of SsaR, the expression of sscI was not affected either (O. Choi and C. Fuqua, unpublished). 125 Figure 3.4. !-galactosidase activity of lacZ transcriptional fusion with sscI in different mutant backgrounds. Plasmids carrying ssaI (pEC108), ssbI (pEC109), ssaR (pEC112) and ssbR (pEC123) were conjugated into the "ssaI "ssbI sscI? mutant, respectively, to try to restore sscI expression. Error bars stand for the standard deviation of triplicates. The result presented is a representative of several independent experiments with three biological replicates. Figure is provided by Okhee Choi. 126 3.4.4 SscI-derived AHL stimulates SsaR-dependent activation of ssaI To test whether SscI-derived AHLs can regulate ssaI expression, a plasmid- borne copy of ssaR paired with a compatible plasmid carrying the PssaI-lacZ fusion were introduced into an AHL-, plasmidless derivative of A. tumefaciens NTL4 to reconstruct the regulatory system. Cultures of the A. tumefaciens derivatives were grown with 2.5% (v/v) organic extract of SscI- derived AHL in A. tumefaciens NTL4 host with no culture amendments. The expression of ssaI was monitored by #-galactosidase assay. Results showed that SscI-derived AHLs increased ssaI expression about 4-fold compared to the negative control (P<0.01) (Table 3.2). This stimulation of ssaI expression is dependent on SsaR because sscI-derived AHLs did not increase ssaI expression when only the vector pBBR1-MCS5 was present (Table 3.2). 127 Table 3.2. SsaR-dependent activation of ssaI was stimulated by SscI- derived AHL in an AHL- host1 !-galactosidase Sp. Act.2 Expression plasmid Fusion plasmid +2.5%3 No extract pBBR1-MCS5 ssaI-lacZ (pEC116) 85(1) 99(6) pEC112 (Plac-ssaR) ssaI-lacZ (pEC116) 2335(75) 644(25) 1 All strains derived from Ti-plasmidless A. tumefaciens NTL4. 2Specific activities in Miller Units, averages of assays in triplicate (standard deviation). 3 2.5% (v/v) organic extract of A. tumefaciens NTL4 carrying Plac-sscI (pOKC4). 3.4.5 The sscI mutant shows reduced swimming motility. The ssaRI system positively controls KLH11 swimming motility and biosynthesis of flagella while the ssbRI system does not (Zan et al., 2012). We were interested in detecting whether sscI affects swimming motility. The sscI null mutant was tested for motility on MB2216 supplemented with 0.25% agar (w/v). There was ca. 20% decrease in the diameter of zones compared to wild type KLH11 (P<0.05) and plasmid-borne sscI was able to complement the swimming defects (Fig. 3.5). 128 Figure 3.5 SscI affects KLH11 swimming motility moderately. Swimming motility assays on MB 2216 supplemented with 0.25% (w/v) agar after a week at 28?C. This represents one biological replicate of each strain of one experiment with three biological replicates. The diameter of swimming ring was measured. The results presented are representative of several independent experiments each with three biological replicates. Figure is provided by Clay Fuqua. 3.4.6 KLH11 contains a novel enzyme responsible for the synthesis of p-HSL-like molecule. We used R. palustris CGA814 that has rpaI-lacZ fusion as the reporter strain to detect whether KLH11 produces the novel molecule pC-HSL (Schaefer et al., 129 2008). Results showed that mutation of any of the three luxI genes did not affect the ability of the culture supernatant to increase the expression of the pC-HSL synthesis gene rpaI, of which the expression is stimulated by pC-HSL (Schaefer, et al., 2008). Furthermore, the supernatant from the double mutant "ssbI sscI- and triple mutant "ssaI "ssbI sscI- can still stimulate the expression of rpaI (Fig.3.6). 130 Figure 3.6. !-galactosidase assay of the expression of rpaI-lacZ fusion. R. palustris CGA814 was used as the reporter strains. WT = wild type KLH11 (EC1). Supernatant from CGA814 was used as negative control and pC-HSL was used as a positive control. All the KLH11 derived strains were grown in MB2216 supplemented with 0.5 mM p-coumarate. Bars represent the average of three biological replicates and the error bars stand for the standard deviation of triplicates. Figure is provided by Okhee Choi. 131 3.5. Discussion Two pairs of LuxI-LuxR type QS genes (ssaRI and ssbRI) were previously identified in the sponge-associated bacterium Ruegeria sp. KLH11, a representative of the Silicibacter-Ruegeria group in the Roseobacter clade. Here we described the finding of sscI, a solo of luxI homologue that was not detected by the genetic screening. We reason that this could be due to several factors: 1) not enough colonies were screened; 2) the restriction enzymes used to cut the genomic DNA to prepare the screen library did not have recognition sites in the regions surrounding sscI, noting that the average size of fragments ligated into the pBBR1-MCS5 vector was 4 kb; or 3) sscI with its native promoters is not expressed in the heterologous A. tumefaciens system. It is highly likely that the genome annotation in which sscI was detected caught the only missing LuxI- type AHL synthase gene not found by the genetic screen because the triple knock out mutant of the three AHL synthase genes in KLH11 resulted in a mutant in which AHLs are undetectable by mass spectrometry. Typically, luxI homologues are arranged in tandem with a cognate luxR homologue. However, in this case a luxI solo, sscI, without a cognate luxR homologue was identified in KLH11. This arrangement has been reported in a roseobacterial strain that is a symbiont of the alga Dinoroseobacter shibae DFL12T. This strain also has two sets of LuxI-LuxR QS systems and one luxI-solo, named luxI3 (Wagner-D?bler et al., 2010). However, it is unknown whether LuxI3 synthesizes AHLs. 132 SscI shares 81% identity to SsbI on the amino acid level and also encodes a putative protein of similar size (214 aa versus 212 aa). These two genes result in synthesis of very similar AHL profiles when expressed in E. coli (Fig. 3.2C). Furthermore, both ssbI and sscI lack a LuxI-type box in their promoter regions and do not respond to exogenously added KLH11 AHLs. Taken together, this suggests that ssbI and sscI are homologues. Similarly, the dominance in the sscI-derived AHL profile of long chain length AHLs is consistent with the types of AHLs seen in a large number of marine bacteria isolated from a variety of different sources, which might reflect adaptation to the marine environment (Wagner-D?bler et al., 2005). A transcriptional fusion of lacZ to sscI allowed monitoring its expression in different genetic backgrounds and studying how other QS genes regulate sscI. Generally, bacteria that have multiple QS systems tend to have a complex regulatory network (Fuqua and Greenberg, 2002). Here, results show that sscI is transcriptionally positively regulated by ssaI and ssbI. Also, both the ssaI and ssbI regulation on sscI are indirect, because in the A. tumefaciens NTL4 background SsaR or SsbR do not affect sscI transcription by addition of exogenous KLH11 AHLs. One unexplained finding is that the decrease of sscI expression in "ssaI, "ssbI or "ssaI "ssbI mutant strains cannot be complemented with crude extract of KLH1 or provision of SsaI or SsbI. Our results also show that sscI-derived AHLs provide a feedback on ssaI expression via ssaR (Table 4.2). We know that ssbI expression is not affected by addition of exogenous AHLs (Zan et al., 2012). In this study we did not detect 133 an effect of sscI-derived AHLs on sscI expression. Many LuxI-LuxR systems tend to stay inactive when in the inactive state despite environmental perturbations and vice versa (Tsai and Winans, 2011). In the case of the ssaRI system, this results from the positive autoregulation of ssaI by SsaR?AHL complexes. sscI contributes to this positive regulation because SsaR can also respond to AHL synthesized by SscI, especially given the fact that the main AHL molecule detected in the wild type KLH11 culture is 3-OH-C14-HSL and 3-OH- C14:1?HSL (Zan et al., 2012). Moreover, the moderate effect of sscI on swimming motility is probably due to this contribution of sscI to ssaI expression because the ssaRI system positively controls KLH11 swimming motility. LuxR solos are quite commonly found in many bacterial genomes. They could be used for interspecies signaling or for ?eavesdropping? on other bacteria (Subramoni and Venturi, 2009). But what is the function of the luxI solo in the natural environment? Are they commonly distributed in bacteria? What is the evolutionary history of the luxI solo? All these questions remain to be answered. Recent years, several novel types of AHL molecules have been reported (Schaefer et al., 2008, Ahlgren et al., 2011; Lindemann et al., 2011). The pC- HSL has a ring structure in the fatty acid chain and was first discovered in R. palustris (Schaefer et al., 2008). A close relative of KLH11, R. pomeroyi DSS-3 was also found to produce this type of molecule. We used #-galactosidase assay to test whether KLH11 produces similar molecules that can stimulate the expression of rpaI. Surprisingly, our results show that KLH11 can also produce similar molecules and this was independent of any of the three known LuxI-type 134 enzymes but we do not know what the identity of the molecule(s) in the KLH11 culture is that can stimulate the expression of the rpaI-lacZ reporter, suggesting the existence of novel enzyme(s) in KLH11 responsible for the synthesis. Research also suggests that Phaeobacter gallaeciensis BS107 (also known as DSM 17395) can respond to the presence of p-coumarate produced by the microalga Emiliania huxleyi potentially via pC-HSL (Seyedsayamdost et al., 2012). Furthermore, several different roseobacterial species can produce the novel quorum-sensing molecule TDA originally described in Silicibacter sp. TM1040 (Geng and Belas, 2010). Taken together, this suggests that Roseobacters represent an underexplored resource for discovery of novel quorum sensing molecules. 135 Chapter 4. The cckA-chpT-ctrA phosphorelay system is regulated by quorum sensing and controls flagellar motility in the marine sponge symbiont Ruegeria sp. KLH11 136 4.1. Abstract Bacteria respond to their environment via signal transduction pathways, often two-component type systems that function through phosphotransfer to control expression of specific genes. Phosphorelays are derived from two-component systems but are comprised of additional components. The essential cckA-chpT- ctrA phosphorelay in Caulobacter crescentus has been well studied and is important in orchestrating the cell cycle, polar development and flagellar biogenesis. Although cckA, chpT and ctrA homologues are widespread among the Alphaproteobacteria, little is known about their function in the large and ecologically significant Roseobacter clade of the Rhodobacterales. In this study the cckA-chpT-ctrA system of the marine sponge symbiont Ruegeria sp. KLH11 was investigated. Our results reveal that the cckA, chpT and ctrA genes positively control flagellar biosynthesis. In contrast to C. crescentus, the cckA, chpT and ctrA genes in Ruegeria sp. KLH11 are non-essential and do not affect bacterial growth. Gene fusion and transcript analyses provide evidence for ctrA autoregulation and the control of motility-related genes. In KLH11, flagellar motility is controlled by the SsaRI system and acylhomoserine lactone (AHL) quorum sensing. SsaR and long chain AHLs are required for cckA, chpT and ctrA gene expression, providing a regulatory link between flagellar locomotion and population density in KLH11. 137 4.2. Introduction Roseobacters represent an abundant and important marine bacterial group in the Alphaproteobacteria. Members from this group can mediate key biogeochemical processes and account for up to 30% of bacterioplankton cells in some coastal environments (Gonz?lez and Moran, 1997). Many Roseobacters have been experimentally shown to exhibit flagellar motility, an important trait for their physical associations with eukaryotic cell surfaces or organic particles (Slightom and Buchan, 2009). For example, in Silicibacter sp. TM1040 flagellar mutants are defective in attaching to and forming biofilms on the dinoflagellates with which this bacterium is associated (Miller and Belas, 2006). Alphaproteobacteria, including Roseobacters are also found in association with marine sponges (Mohamed et al., 2008b). Ruegeria sp. KLH11 is a sponge symbiont within the roseobacterial Silicibacter-Ruegeria subgroup which is consistently and specifically isolated from soft-bodied marine sponges such as species of Mycale and Ircinia (Mohamed et al., 2008c). KLH11 has been developed as a model for studying the interactions of bacteria with sponge hosts. Two-component type phosphorelay signal transduction pathways, comprised of two or more proteins, are some of the most prevalent means by which bacteria sense, respond, and adapt to changes in their environment or their intracellular state. In their simplest form two-component systems consist of a sensor histidine kinase and a cognate response regulator, through which phosphotransfer controls the regulatory output (Laub and Goulian, 2007). More complex systems can have multiple components, and individual regulators can have multiple 138 phosphotransfer activities. In the alphaproteobacterial developmental model system Caulobacter crescentus the response regulator CtrA acts to control the cell cycle and is essential for viability. CtrA is phosphorylated on a conserved asparate residue (D51), via a phosphorelay pathway that initiates with the histidine kinase CckA. When active, CckA undergoes an intramolecular phosphotransfer between a conserved histidine and an aspartate in its receiver domain at the carboxy terminal end of the protein. Active CckA subsequently phosphorylates the ChpT histidine phosphotransferase (Hpt). ChpT can transfer phosphate to either of two response regulators, CpdR or CtrA (Quon et al., 1996; Biondi et al., 2006). CpdR normally inhibits CtrA, but is inactive when it is phosphorylated. Phospho-CtrA, relieved of CpdR inhibition, is an active transcriptional regulator that controls about 26% (144/553) of the genes involved in cell cycle progression and also controls flagellar motility in C. crescentus (Laub et al., 2000). Members of the Rhodobacterales such as KLH11 encode cckA, chpT, and ctrA homologues (Zan et al., 2011a)(See Appendix 1), but generally no cpdR homologue (Brilli et al., 2010). In Silicibacter sp. TM1040, a relative of Ruegeria sp. KLH11, the cckA, chpT and ctrA genes are required for flagellar motility, but in contrast to C. crescentus these genes are non-essential (Belas et al., 2009). Bacterial flagellar motility displays a critical role in many microbial processes, such as chemotaxis, colonization of hosts, and biofilm formation (Smith and Hoover, 2009). The biosynthesis of flagella is an ordered process that requires the coordinated and temporal regulation of many different genes via a very 139 complex regulatory hierarchy. For bacteria in which flagellar assembly has been well studied there is generally a primary regulator that initiates expression of the flagellar gene hierarchy and is referred as the master regulator. Several different types of master regulators, including CtrA from C. crescentus, have been identified. FlhDC is the most extensively studied master regulator in both Escherichia coli and Salmonella typhimurium (Kutsukake et al., 1990; Liu and Matsumura, 1994). FleQ and FlrA are the master regulators in Pseudomonas aeruginosa and Vibrio cholerae (Arora et al., 1997; Klose and Mekalanos, 1998), respectively. Although flagellar motility is common among the Roseobacters (Slightom and Buchan, 2009), little is known about its regulation. We recently reported that the sponge symbiont Ruegeria sp. KLH11 utilizes two distinct but interconnected quorum sensing (QS) systems, with the LuxR- LuxI homologues SsaRI and SsbRI, that rely upon an overlapping set of long chain acylhomoserine lactone (AHL) signal molecules (Zan et al., 2012). Many bacteria use intercellular signals such as AHLs to monitor their population density and accordingly regulate the expression of specific gene sets in crowded conditions. The SsaRI system is required for flagellar assembly and flagellar gene expression in KLH11 whereas the SsbRI system has no influence on motility. KLH11 specifically switches into a motile phase at high population densities, and this requires SsaRI (Zan et al., 2012). Although it is possible that SsaR functions as the primary regulator of motility, it is more likely that it controls expression of a downstream regulator specific for the flagellar genes. For example, in Burkholderia glumae the tofRI QS system regulates the expression 140 of flhDC, which in turn directly controls motility (Kim et al., 2007). Although the FlhDC and FleQ/FlrA homologues are not present in Ruegeria pomeroyi DSS-3 or in KLH11 genome sequences (Moran et al., 2004; Zan et al., 2011a)(See Appendix 1), it is possible that the cckA-chpT-ctrA pathway acts in this capacity. In this study we examined whether 1) cckA, chpT and ctrA genes are essential for the viability of KLH11; 2) they can control flagellar motility; 3) they are influenced by the SsaRI system. Our results show clearly that cckA, chpT and ctrA are non-essential, are tightly regulated by QS, and act downstream of QS in controlling flagellar motility. 141 4.3. Experimental procedures 4.3.1 Strains, growth conditions and plasmid transformation Bacterial strains and plasmids used in this study are listed in Table 4.1. Ruegeria sp. KLH11 and KLH11-EC1 derivatives were grown in Marine Broth 2216 (MB2216) (BD, Franklin Lakes, NJ) at 28?C. E. coli strains were grown at 37?C in Luria-Bertani (LB) broth. A. tumefaciens strains were grown in AT minimal medium supplemented with 0.5% glucose and 15 mM (NH4)2SO4 (ATGN) (Temp? et al., 1977). Antibiotics were used at the following concentrations (?g ml-1): (i) E. coli (gentamicin, Gm, 25; kanamycin, Km, 50; spectinomycin, Sp, 100), (ii) KLH11 (Km, 100; rifampicin, Rif, 200; Gm, 25; Sp, 100) (iii), A. tumefaciens (Gm, 300; Sp, 200). Plasmids were introduced into KLH11 and derivatives using either electroporation or conjugation (Zan et al., 2012) and into E. coli using standard methods of transformation and into A. tumefaciens using a standard electroporation method (Mersereau et al., 1990). 4.3.2 Deletion of ssaR and generation of cckA, chpT and ctrA null mutants DNA manipulations were performed using standard techniques or per manufacturers? specifications (Sambrook et al., 1989). Restriction enzymes and PhusionTM High-Fidelity DNA Polymerase were obtained from New England Biolabs (Ipswich, MA). Oligonucleotides, listed in Table 4.2, were obtained from Integrated DNA Technologies (Coralville, IA). DNA sequencing was performed 142 on an ABI3700 automated sequencer by the BioAnalytical Services Laboratory at the Institute of Marine and Environmental Technology (Baltimore, MD). To generate an in-frame, markerless deletion of the ssaR gene, splicing by overlap extension (SOE) polymerase chain reaction (PCR) was used (Warrens et al., 1997). An approximately 500-bp fragment upstream of and including the first three codons of the ssaR coding sequence was amplified using primers ssaR D1 and ssaR D2. An approximately 500-bp fragment downstream of and including the ssaR stop codon was amplified using primers ssaR D3 and ssaR D4. Primers ssaR D2 and ssaR D3 were designed to contain an 18 bp complementary sequence at the 5? end (the ?overlap?) to facilitate the SOEing reaction (Merritt et al., 2007). Following initial amplification the two fragments were gel purified and used as template in a second round of PCR with primers ssaR D1 and ssaR D4 generating an approximately 1 kb SOE fragment containing a fusion of the upstream and downstream regions of the ssaR locus. Primers ssaR D1 and ssaR D4 were designed to allow direct cloning using the In-Fusion Clone Kit. The final SOE fragment was gel purified and cloned into the sacB counter-selectable vector pNPTS138 that had been previously digested with EcoRI. The resulting plasmid, pJZ014, was confirmed by sequencing and conjugated into Ruegeria sp. KLH11-EC1 (RifR). The suicide vector pNPTS138 is a ColE1 plasmid carrying kanamycin resistance and is unable to replicate in Ruegeria sp. KLH11 (Zan et al., 2012). Transconjugants were plated onto Marine Agar 2216 (MA2216) (BD, Franklin Lakes, NJ) plates supplemented with both Rif and Km to select for RifR KmR plasmid integrants. Presumptive 143 integrants were tested for sucrose sensitivity, verifying introduction of the sacB counter-selectable marker on pNPTS138, by plating on MA2216 plates supplemented with Rif, Km, and 5% (w/v) sucrose. RifR KmR SucS colonies were subcultured in MB2216 without Km and plated on 5% sucrose MA2216 plates without Km to select for SucR allelic replacement candidates. Candidates were verified to be KmS by patching onto MA2216 plates supplemented with Km. Deletion of the targeted ssaR locus was confirmed by PCR using primers ssaR D1 and ssaR D4 and the "ssaR strain was designated JZ03. Null mutations in the Ruegeria sp. KLH11 cckA, chpT, and ctrA homologues were generated using Campbell-type recombinational mutagenesis. Internal gene fragments were generated by PCR using primers cckA P1/cckA P2, chpT P1/chpT P2, and ctrA P1/ctrA P2, respectively, using KLH11 genomic DNA as template. The partial cckA, chpT, and ctrA fragments were cloned directly into the pCR2.1-TOPO vector (Invitrogen, Grand Island, NY) and then subcloned into pVIK112, a suicide vector with an R6K conditional replication origin (Kalogeraki and Winans, 1997), creating plasmids pJZ003 (truncated at codon 513), pJZ004 (truncated at codon 160), and pJZ005 (truncated at codon 173), respectively. These plasmids were then conjugated into KLH11-EC1. Presumptive KmR transconjugants were selected and confirmed by PCR amplification using the primer 3 designated for each of the three genes that is located upstream of the recombined fragments and the primer 112R that is located downstream of the KpnI recognition site on the plasmid pVIK112 (Kalogeraki and Winans, 1997) and the amplicons were sequenced. The cckA-, chpT-, and ctrA- mutants were 144 designated JZ04, JZ05, and JZ06, respectively. To create strains JZ07-JZ12 plasmids pJZ003, pJZ004, and pJZ005 were conjugated into "ssaI strain (SK01) and "ssaR strain (JZ03), respectively. The KmR recombinants were selected and confirmed as for strains JZ04-JZ06. A transcriptional fusion of E. coli lacZ immediately downstream of the KLH11 cckA homologue at its native genomic location was generated by PCR amplifying a 3? fragment of the cckA gene, ending at the stop codon, using primers cckAintactF and cckAintactR. The PCR amplicon was cloned into pCR2.1-TOPO and then subcloned into pVIK112, creating pJZ012. This plasmid was conjugated into KLH11 and transconjugants were selected and confirmed as described above. Campbell-type recombination results in lacZ fused to the 3? end of the native cckA locus, keeping the cckA gene-coding region intact. 4.3.3 Cloning of phosphorelay components and promoter fusion constructs Complementation constructs of Ruegeria sp. KLH11 homologues of cckA (pJZ006), chpT (pJZ007), and ctrA (pJZ008) were generated by PCR amplification of the coding regions of each gene using primers designated as P3 and P4 for each specific gene and KLH11 genomic DNA as template. An E. coli lacZ ribosomal binding site was engineered into the 5? primer of each gene to allow for efficient translation. PCR products were cloned directly into the broad- host range vector pSRKGm that had been previously cut with SpeI (Khan et al., 2008) using the In-Fusion HD directional cloning system (Clontech, Mountain View, CA). The resulting expression plasmids carry each gene under the control 145 of an IPTG-inducible Plac promoter. The insert carried by each construct was confirmed by sequencing. The expression construct for the A. tumefaciens cckA homologue was created in the pSRKGm plasmid as described (Kim et al, submitted). Expression constructs for the A. tumefaciens chpT and ctrA homologues were generated by PCR amplification with Phusion High-Fidelity DNA polymerase using purified wild-type A. tumefaciens C58 genomic DNA as template. Primers JEH48 and JEH53 were used to amplify the chpT locus and JEH50 and JEH54 were used for the ctrA locus (Table 4.2). Amplicons were cloned into vector pGEM-T Easy and sequenced. Each gene was then sub-cloned into pSRKGm using engineered NdeI and NheI restriction sites. Fusions of the probable promoter regions for the KLH11 homologues of cckA, chpT, and ctrA to a promoterless E. coli lacZ !-galactosidase gene were created in plasmid pRA301 (Akakura and Winans, 2002). The intergenic region upstream of the cckA coding sequence was PCR amplified using primers cckA P5 and cckA P6. The upstream and downstream primers anneal 145 upstream, and 69 bp downstream, of the predicted cckA translational start site, respectively. The PCR product was fused with pCR2.1-TOPO and the insert was confirmed by DNA sequencing. The pCR2.1-TOPO derivative was digested with EcoRI and PstI, and the resulting fragment was ligated into similarly digested pRA301 creating pJZ009 which was confirmed by sequencing. Similarly, the intergenic regions upstream of the chpT and ctrA coding sequence were PCR amplified using primers chpT P5 and chpT P6 or ctrA P5 and ctrA P6, fused with pCR2.1- 146 TOPO and then subcloned into pRA301, creating plasmids pJZ010 or pJZ011, respectively. Plasmids pEC112 (Plac-ssaR) and either pJZ009 (PcckA-lacZ), or pJZ010 (PchpT-lacZ), or pJZ011 (PctrA-lacZ) were electroporated into A. tumefaciens NTL4. Plasmids pBBR1-MCS5 (Kovach et al., 1995) and either pJZ009, pJZ010 or pJZ011 were electroporated into A. tumefaciens NTL4 to serve as negative controls. 4.3.4 Evaluation of flagellar-based motility and presence of flagella Bacterial swimming motility assays were performed using MB2216 with 0.25% (w/v) agar supplemented with 200 !M IPTG. Swim plates were inoculated with mid-log phase cultures of the relevant KLH11 strains using an inoculation needle. Plates were wrapped tightly with plastic film and incubated at 28?C. Swim ring diameters were measured and pictures taken after 8 days with a Nikon D90 digital camera. Relative levels of flagellin in the wildtype, cckA-, chpT-, and ctrA- KLH11 strains were determined from culture supernatants followed by immunoblotting. Flagellin was enriched as described (Kanbe et al., 2007). Strains were grown to mid-log phase in MB2216, supplemented with antibiotics when necessary, and then back-diluted to an OD600 ~ 0.01 in 3 ml MB2216. Samples were collected at stationary phase and OD600 was measured. The samples were vigorously vortexed for 30 sec and then centrifuged (5 min, 10,000 x g) at 4?C. The resulting supernatant was transferred to a new centrifuge tube and polyethylene glycol added to a final concentration of 2%. Following vortexing and 100 min 147 incubation on ice, the mixtures were centrifuged (15 min, 17,400 x g). The resulting precipitate was resuspended in 100 #l 1 X SDS lysis buffer and boiled at 100?C for 5-10 min. The denatured samples were separated on a 15% SDS- PAGE gel at 90 V for 4 h and then were transferred to a nitrocellulose membrane (Amersham Biosciences, Seattle, WA). Immunoblotting was performed with polyclonal antibody raised against whole flagella from C. crescentus (a gift from the laboratory of Y.V. Brun) at a dilution of 1:20,000 as described by Zan et al. (2012). Staining of flagella on intact cells used a two-component stain modified from Mayfield and Inniss (1977). The first component contained equal volumes of saturated AlK(SO4)2 !12H2O and 5% phenol in 10% tannic acid while the second component contained 12% crystal violet in 100% ethanol. Ten ml of a 10:1 mixture of the two components was applied to the edge of a coverslip on a 3 ml wet mount for each strain. Flagella were observed within 5 min of staining on a Zeiss Axioskop 40 microscope equipped with an AxioCam MRm monochrome digital camera using a 100X oil immersion objective and bright field illumination. 4.3.5 Quantification of phosphorelay component promoter activity Promoter activities were quantified using lacZ translational and transcriptional fusions as indicated. !-galactosidase specific activity was measured as described previously, expressed in Miller Units, using o-nitrophenyl-b-galactoside (ONPG) as substrate (Zan et al., 2012). Ruegeria sp. KLH11 was grown in MB2216 supplemented with antibiotics as required overnight. Cultures were 148 diluted approximately 100-fold to obtain an OD600 ~0.01 in 3 ml MB2216 without antibiotics and incubated at 28?C. Mid-log phase KLH11 cultures were sampled and assayed for !-galactosidase activity immediately. Similarly, mid-log phase cultures of A. tumefaciens strain NTL4 were diluted at 1:100 dilution to an OD600~ 0.01 in 3 ml ATGN media and incubated at 28?C with shaking at 200 rpm to an OD600 ~ 0.4. Mid-log phase cultures were measured for OD600 and frozen at - 80?C and used for subsequent !-galactosidase assays. Exogenous AHL was added to each culture where indicated to a final concentration of 2 #M 3-oxo- C16:1 $11cis-(L)-HSL was purchased from Cayman Chemical (Ann Arbor, Michigan). 4.3.6 Analysis of KLH11 CtrA-dependent gene expression Expression of motility- and cell cycle-related genes was measured using qRT- PCR with specific primers (Table S5). KLH11 and derivatives were grown in MB2216 to stationary phase and 0.5 ml culture was collected and stored in 1 ml RNAprotect BacteriaReagent (Qiagen, Valencia, CA). The mixtures were centrifuged (10 min, 5100 x g) and the cell pellets were stored at -80?C for subsequent RNA extraction. Total RNA was isolated using an RNeasy miniprep kit (Qiagen, Valencia, CA), with genomic DNA removed by TURBO DNase (Ambion, Austin, TX), per manufacturers? supplied protocols. cDNA was synthesized using qScript cDNA SuperMix according to the manufacturer?s instructions (Quanta BioSciences, Gaithersburg, MD). RT-PCR was performed with Power SYBR Green Master Mix (Invitrogen, Grand Island, NY) on an ABI 7500 Fast Real-Time PCR system using the following cycling parameters: 2 min 149 at 95?C for initial denaturation, 40 cycles consisting of 10 s at 95?C, and 1 min at 60?C for primer annealing and extension. Melt curves were performed to confirm the specificity of primers and the absence of primer dimers. Expression levels were normalized to the housekeeping rpoD gene encoding s70. 4.3.7 Multiple sequence alignment and phylogenetic analysis of the ctrA gene Sequences of ctrA homologues from selected Alphaproteobacteria were downloaded from GenBank and aligned using ClustalW2 (http://www.ebi.ac.uk/Tools/msa/clustalw2/). The phylogenetic tree was constructed using software MEGA 4.0 (http://www.megasoftware.net/). BOXSHADE was used to determine the degree of residue shading (www.ch.embnet.org/software/BOX_form.html). 150 Table 4.1. Strains and plasmids used in Chapter 4 Bacteria/Plasmids Relevant featurea Reference E. coli TOP 10 F? Standard alpha-complementation strain, lacIQ Qiagen E. coli DH5!/"pir Strain for propagating R6K suicide plasmids Lab collection E. coli S17-1/"pir IncP conjugal donor (Kalogeraki and Winans, 1997) A. tumefaciens NTL4 Ti plasmidless derivative, nopaline chromosomal background (Zhu et al., 1998) KLH11 wild type (Mohamed et al., 2008c) KLH11-EC1 RifR (Zan et al., 2012b) KLH11-SK01 #ssaI, RifR (Zan et al., 2012b) KLH11-OKC8 fliC::pOKC12, KmR (Zan et al., 2012b) KLH11-JZ03 #ssaR, RifR This study KLH11-JZ04 cckA::pJZ003, RifR, KmR This study KLH11-JZ05 chpT::pJZ004, RifR, KmR This study KLH11-JZ06 ctrA::pJZ005, RifR, KmR This study KLH11-JZ07 #ssaI cckA::pJZ003, RifR, KmR This study KLH11-JZ08 #ssaI chpT::pJZ004, RifR, KmR This study KLH11-JZ09 #ssaI ctrA::pJZ005, RifR, KmR This study KLH11-JZ10 #ssaR cckA::pJZ003, RifR, KmR This study KLH11-JZ11 #ssaR chpT::pJZ004, RifR, KmR This study KLH11-JZ12 #ssaR ctrA::pJZ005, RifR, KmR This study KLH11-JZ13 cckA::pJZ012, wild type cckA, RifR, KmR This study pCR 2.1-TOPO? PCR fragment cloning vector, Ap/KmR Invitrogen pBBR1-MCS5 Plac expression vector, GmR (Kovach et al., 1995)1995) pNPTS138 colE1 origin, sacB, KmR gift of M. Alley pSRKGm pBBR1MCS-5-derived expression vector containing lac promoter lacIq, GmR (Khan et al., 2008) pVIK112 R6K-based lacZ transcriptional fusion, integration vector, KmR (Kalogeraki and Winans, 1997) pRA301 lacZ translational fusion vector (Akakura and Winans, 2002) 151 pEC108 pBBR1-MCS5 derivative carrying full length Plac-ssaI, GmR (Zan et al., 2012) pEC112 pBBR1-MCS5 derivative, carrying full length Plac-ssaR, GmR (Zan et al., 2012) pOKC12 pVIK112 derivative carrying truncated fliC, KmR Choi and Fuqua, unpublished pJZ003 pVIK112 derivative carrying truncated cckA gene, KmR This study pJZ004 pVIK112 derivative carrying truncated chpT gene, KmR This study pJZ005 pVIK112 derivative carrying truncated ctrA gene, KmR This study pJZ006 pSRKGm derivative carrying full length Plac-cckA, GmR This study pJZ007 pSRKGm derivative carrying full length Plac-chpT, GmR This study pJZ008 pSRKGm derivative carrying full length Plac-ctrA, GmR This study pJZ009 pRA301 derivative, PcckA-lacZ, SpR This study pJZ010 pRA301 derivative, PchpT-lacZ, SpR This study pJZ011 pRA301 derivative, PctrA-lacZ, SpR This study pJZ012 pVIK112 derivative, cckA gene with 5? truncation, to retain wt cckA, KmR This study pJZ014 pNPTS138 carrying ssaR deletion fragment, KmR This study pJEH010 pSRKGm derivative carrying full length Plac-cckA of A. tumefaciens, GmR Kim et al., submitted pJEH027 pSRKGm derivative carrying full length Plac-chpT of A. tumefaciens, GmR This study pJEH028 pSRKGm derivative carrying full length Plac-ctrA of A. tumefaciens, GmR This study aAp=ampicillin, Gm=gentamicin, Km=kanamycin, Rif=rifampicin, Sp=spectinomycin. 152 Table 4.2. Primers used in Chapter 4. Primer name Sequencesb (5?-3?) Restriction enzyme cckA P1 GAATTCCGCGCAAGGACAAAGAGATA EcoRI cckA P2 GGTACCCGACAAGGTTCATCAACACC KpnI cckA P3 CCGCTCTAGAACTAGTGTGAAACAGGAAACAGCTATGTCCAGTGTT TCTGAATC SpeI cckA P4 CGGGGGATCCACTAGTCTAGTTGAGTTGCTGGAAC SpeI cckA P5 GAATTCCGGAACCGATGGATTTTACA EcoRI cckA P6 CTGCAGCGACACATAGCGGCCGTGGTC PstI chpT P1 GAATTCCGCATCAGACGTCAACCTT EcoRI chpT P2 GGTACCCTTTCCAGAGGCCGCTATC KpnI chpT P3 CCGCTCTAGAACTAGTGTGAAACAGGAAACAGCTATGCAGCAGGA GGTACGCATG SpeI chpT P4 CGGGGGATCCACTAGTCTAAAAGCGCAGCGTTACAC SpeI chpT P5 GAATTCGGATCAAAGACCGGCATCAG EcoRI chpT P6 CTGCAGCTGCATCTGCCGCAGGAGCCG PstI ctrA P1 GAATTCGATGCTGACACATGCCAATC EcoRI ctrA P2 GGTACCATCTCTTTCGTCAGCGTGGT KpnI ctrA P3 CCGCTCTAGAACTAGTGTGAAACAGGAAACAGCTATGCGAATACT TCTCGTCGA SpeI ctrA P4 CGGGGGATCCACTAGTTCAGGCGCCGACCGCCA SpeI ctrA P5 GAATTCGGCGGAACATGGCGTCGA EcoRI ctrA P6 CTGCAGTCGCATTCAACTGCTCCAAT PstI ssaR D1 CTGGATCCACGAATTCGGTAAACCGCCCCTATTACGG EcoRI ssaR D2 AAGCTTGGTACCGAATTCAATATCCATCGGTAACGACCA NA ssaR D3 GAATTCGGTACCAAGCTTCCAGGTTAAAACCAAAACTCC NA ssaR D4 CGAAGCTAGCGAATTCGTCGCATAGGACACCGAGTTC EcoRI JEH48 gaagaaCATATGACGAGCAAACTCAATATCACGC NdeI JEH53 gaagaaGCTAGCTCACTCGGCAACCGTCTTTGC NheI 153 JEH50 gaagaaCATATGCGGGTTCTACTGATTGAAGACGA NdeI JEH54 gaagaaGCTAGCTCAGGCGGTTTCGAGGAA NheI rpoDRT1 GACGCCTATCGCGGCCGT NA rpoDRT2 GCCGACCTGCGCCATATCGT NA RT-FliFF GCGCGGTGTTGCCTATGAGAT NA RT-FliFR GATGCCGCGAATCGCTGG T NA RT-FlhAF CGGGCTTCTGATCACGCTCCT NA RT-FlhAR GCGTTGTCAGTGGCACCTTGT NA RT-FlgBF TACGCAATGGCAACCCATGCT NA RT-FlgBR GCGTCCGAAATGCCATGCAGAT NA RT-motBF GTGACGGCCATGATGGCGTT NA RT-motBR CCTTGTGCATCCACGCCTGT NA RT-FlgJF GCTTTGGTTCCACAACAGCTAA NA RT-FlgJR CTGTTGCTGCACGAGGAAAG NA fliCRT1 CGCAGAACCTGTCGACCGGT NA fliCRT2 GATGCCGCGAATCGCTGGT NA RT-ftsZF GCAGCTGGACGGCGTTGAAT NA RT-ftsZR CCGCCAGATGATCCACGATCTGT NA RT-ccrMF GTCGACGCGGTCGATGATCACT NA RT-ccrMR GTTCGACTTGCGCCACACAACAT NA cckAintactF GAATTCACAGGTCTGGGTCTGTCCAC EcoRI cckAintactR GGTACCCTAGTTGAGTTGCTGGAAC KpnI 112R GGCTGCAGGTCGACCATGGTC NA b Engineered restriction sequences are underlined. Complementary sequences for PCR-SOEing are shown in bold and are also underlined. Start and stop codons are in bold. E. coli lacZ ribosomal binding sites are in italics and bold. Protection nucleotides are in lower case. NA= not applicable. 154 4.4. Results 4.4.1 The KLH11 cckA, chpT and ctrA genes are non-essential and control flagellar motility Annotation of the KLH11 genome revealed that KLH11 has homologues to each of the cckA, chpT and ctrA genes (Zan et al., 2011a)(See Appendix 1). The putative KLH11 cckA gene (GenBank No. ZP_05124558) encodes a 763 amino acid (aa) protein which shares 48% identity at 52% coverage over its C-terminus (367-763) to the cckA gene in C. crescentus. The N terminus of KLH11 CckA (1- 366 aa) has no similarity to that of the Caulobacter CckA and had two transmembrane regions predicted by http://www.sbc.su.se/~miklos/DAS/. Domain scans using http://www.ebi.ac.uk/Tools/pfa/iprscan/ suggest that the KLH11 CckA protein has a sensory box (273-383 aa), a HisKA domain (394-457 aa), HATPase_c domain (500-620 aa) and REC domain (645-758 aa) (Fig. 4.1A). The domain organization of KLH11 CcKA is very similar to that of C. crescentus CckA (Jacobs et al., 1999). Furthermore, the histidine residue at position 402 and the aspartate residue at position 697 correspond to the conserved phosphorylation sites, histidine 322 and aspartate 623 of Caulobacter CckA. The ChpT Hpt homologue in KLH11 (GenBank No. ZP_05124304) encodes 204 aa and shares 34% identity at 58% coverage (13-131 aa) with C. crescentus ChpT, including a histidine at position 24 corresponding to the conserved histidine at position 61 of Caulobacter ChpT (Biondi et al., 2006). It has a 155 hypothetical domain DUF2328 (30-204 aa) conserved in bacteria. The CtrA homologue from KLH11 (GenBank No. ZP_05124475) encodes 238 aa and shares 74% identity across its length with that of C. crescentus. It has a receiver domain in the N-terminus (3-112 aa) and a DNA binding domain (145-221 aa) in the C-terminus (Fig. 4.1A). The phosphorylation site aspartate in KLH11 is also conserved compared to other ctrA homologues (Fig. 4.2). Figure 4.1. The cckA-chpT-ctrA pathway controls motility. A) Diagram of the predicted CckA protein, ChpT protein and CtrA protein. The N-terminus is shown at the left and the C-terminus is shown at the right. TM = transmembrane domain, HisKA = histidine kinase A dimerization/phosphoacceptor domain, HATPase_c = histidine kinase-like ATPase domain, REC = signal receiver domain. DUF2328 is a Pfam domain with unknown function. DBD= DNA binding 156 domain. The length of line was drawn according to scale. The partial promoter regions of the cckA and ctrA genes are shown on top of the lines. CtrA full recognition site (TTAAN7TTAAC) is in bold and both the CtrA half recognition site (TTAACCAT) and the region that has one mismatch are in grey. The start codon is boxed. B). Swimming motility assays. Wild-type KLH11 and derivatives were inoculated on MB2216 (supplemented with 0.25% agar) swim agar plates for 8 days at 28?C. 200 #M IPTG was added to the media. The results are representative of several independent experiments each with three biological replicates. The bar represents 2 cm. C). CtrA autoregulates its own expression. Plac-ctrA plasmid (pJZ008) was conjugated into the ctrA- mutant (JZ06) and the expression of ctrA-lacZ was monitored by !-galactosidase assay. The pSRKGm was conjugated into the ctrA- mutant as a negative control. Representative results of several independent experiments each with three biological replicates are presented. Values are averages of assays performed in triplicate and error bars are standard deviations. 157 Figure 4.2. Alignment of KLH11 CtrA amino acid sequence to selected CtrA homologues. The degree of shading is determined by using software BOXSHADE. The helix?turn?helix DNA-binding motif is boxed with a dashed line. The conserved Asparate residue is indicated with an asterisk above. Amino acid numbers for each CtrA protein are shown on the left. The GenBank accession numbers for sequences used in this alignment are shown in Fig. 4.5. To test whether cckA, chpT and ctrA genes are essential, we attempted to generate Campbell insertions to disrupt the KLH11 cckA, chpT and ctrA genes using the pVIK112 suicide plasmid (Kalogeraki and Winans, 1997) carrying truncated, internal fragments of each of the three genes (nt 1031-1537 of the cckA gene, nt 24-478 of the chpT gene, nt 55-518 of the ctrA gene). Presumptive kanamycin resistant (KmR) recombinants were readily isolated for all cckA, chpT and ctrA genes and the integration of the mutagenic plasmids was confirmed by PCR amplification and sequencing. Strikingly, in contrast to their essential role in C. crescentus, growth curves of the cckA-, chpT- and ctrA- null 158 mutants were similar to that of wild type KLH11 (Fig. 4.3). Taken together, these results show conclusively that the cckA, chpT and ctrA genes in Ruegeria sp. KLH11 are non-essential and do not affect bacterial growth under laboratory conditions. We tested the cckA-, chpT- and ctrA- null mutants on MB2216 (supplemented with 0.25% agar) swim plates and found that these three null mutants cannot migrate from the inoculation site, unlike the wild-type Ruegeria sp. KLH11 that demonstrates motility under these test conditions (Fig. 4.1B). Provision of plasmid-borne cckA, chpT and ctrA genes expressed from the lac promoter (Plac- cckA, pJZ006; Plac-chpT, pJZ007; Plac-ctrA, pJZ008) was able to partially restore motility in the corresponding cckA-, chpT- and ctrA- null mutants in the presence of 200 !M isopropyl-b-D-1-thiogalactopyranoside (IPTG) to induce Plac. Microscopic examination of these liquid cultures also revealed no detectable motility for these three mutants (data not shown). Results of a flagellar stain of stationary cultures showed that these three mutants did not synthesize any flagella, in contrast to the wild type (Fig. 4.4A). Antiserum raised against whole flagella from C. crescentus, a related alpha-proteobacterium, was able to recognize KLH11 flagellin protein encoded by the fliC gene, of approximately 41.5 kDa in size (Zan et al., 2011a; Zan et al., 2012)(See Appendix 1) and was used in the western blot assay. Results showed that none of the three null mutants produced any detectable flagellin protein (Fig. 4.4B). 159 Figure 4.3. Growth curves of wild-type KLH11 (EC1) and derivatives. Values are average (standard deviation) of triplicate samples. Growth curves in (A) and (B) were examined separately. Strains were inoculated in 50 ml MB2216 in a 250 ml flask. At indicated time point, 200 !l was sampled and OD600 was measured on a SpectrMax M5 microplate reader (Molecular Devices, Sunnyvale, CA). 160 Figure 4.4. Detection of flagella and flagellin. A) Flagellar stain of wild type KLH11, cckA- , chpT- and ctrA- null mutants. Stained cells from late stage cultures were viewed under phase contrast microscopy with 100X lens. Wild type (EC1), cckA- ( JZ04), chpT- (JZ05), ctrA- (JZ06), red arrows indicate stained flagella. The bar indicates 10 !m. B) Detection of flagellin in wild type KLH11, cckA- , chpT- and ctrA- null mutants. Antibody raised against C. crescentus whole flagella was used to probe for flagellin. Samples were collected at stationary phase. Flagellin was extracted from 3 ml late stage culture culture from each of the 4 strains with similar OD600. The extraction was dissolved in 100 !l 1X sample buffer and boiled for 5 min. 30 !l was loaded onto each lane. Estimated size of KLH11 flagellin is 43 kDa. Figure 4.4A provided by Jason E. Heindl. 161 4.4.2 CtrA regulates motility-related gene expression but not cell cycle-related genes. CtrA regulates expression of a wide range of genes involved in different cellular processes in several bacterial species (Laub et al., 2002; Mercer et al., 2010). We used quantitative reverse transcription-PCR (qRT-PCR) to detect the expression differences of motility-related genes between wild type KLH11 and ctrA- mutant. Five genes: motB, fliL, flgB, flgJ and fliG, which are the first genes in their predicted motility-related operons, and the flhA gene, which is the second gene in its operon (the presumptive first gene is not homologous to any known motility genes), were chosen for analysis. The flagellin gene (fliC) was also selected for testing. All the predicted motility-related genes we tested were significantly decreased in the ctrA- mutant, ranging from 9- to 93- fold differences between wild type KLH11 and the ctrA- mutant (Table 4.3). Provision of plasmid- borne KLH11 CtrA (Plac-ctrA) into the ctrA- mutant restored the expression levels almost to those in wild type KLH11. We similarly tested CtrA regulation of the cell cycle related genes ftsZ (GenBank No. ZP_05121748) and ccrM (GenBank No. ZP_05124520), orthologues of which are CtrA-controlled in C. crescentus. It is clear that under the conditions we examined, CtrA does not regulate the expression of the ftsZ or ccrM genes (Table 4.4). 162 Table 4.3. Quantification of motility-related gene expression by qRT-PCR. Gene name Putative gene class Wild-type 1,2 ctrA-1 Plasmid- borne CtrA1,2 fliL 2 456 (91) 12 (3) 535 (255) fliF 2 323 (65) 5 (1) 299 (75) flgB 2 746 (219) 8 (3) 602 (176) flhA 2 9 (3) 1 (1) 7 (1) flgJ 3 154 (28) 10 (1) 174 (67) fliC 3 93 (28) 7 (<1) 51 (16) motB 3 285 (56) 16 (2) 323 (103) 1Value relative to the rpoD gene. Average of three biological replicates (standard deviation). The values are multiplied by 1000. 2All P values are < 0.05 when compared the indicated column to the ctrA- column. Table 4.4. Quantification of ftsZ and ccrM expression by qRT-PCR Gene name Wild-type 1 ctrA- 1 ftsZ 9.9 (3.9) 10.9 (3.2) ccrM 1.3 (0.4) 1.5 (0.3) 1 Value relative to the housekeeping gene rpoD. Mean (standard deviation). The results presented are representative of two independent experiments each with three biological replicates. 163 4.4.3 CtrA autoregulates its own transcription but not that of the cckA gene KLH11 CtrA has an identical amino acid sequences in the putative helix-turn- helix DNA sequence recognition region to that of C. crescentus (Fig. 4.2). The DNA sequences with which this CtrA protein interacts, have been well characterized as TTAA-N7-TTAAC (full site) and TTAACCAT (half-site) in C. crescentus. However, it is clear that the CtrA protein can also bind to more degenerate sequences that appear to share only the TTAA sequences (Laub et al., 2002). Examination of the sequences upstream of the predicted ctrA translation start site revealed a putative half site (62 bp upstream of the predicted translational start, Fig. 4.1A) and thus we tested whether CtrA autoregulates its own expression. The plasmid integration used to disrupt the ctrA gene (pJZ005 derived from pVIK112) simultaneously generates a transcriptional fusion to the disrupted gene (Kalogeraki and Winans, 1997). The Plac-ctrA plasmid (pJZ008) and a vector control (pSRKGm) were conjugated in parallel into strain JZ06 (ctrA- lacZ). Under the 200 !M IPTG induction of the Plac-ctrA plasmid, there was a statistically significant, yet modest ~50% increase of ctrA expression (P<0.05) compared to the vector control (Fig. 4.1C). Inspection of the cckA upstream sequences for CtrA binding sites identified one putative CtrA full recognition site (62 bp upstream of the predicted translational start) and one half site (51 bp upstream of the predicted translational start), although these two sites overlap (Fig. 4.1A). We used a similar approach to that described above to test whether CtrA affects cckA expression. However, 164 we reasoned that the cckA gene might be required to generate the phosphorylated CtrA capable of regulating the cckA promoter. Therefore instead of using the strain JZ04 with a disrupted cckA gene fused to lacZ on the integrated plasmid, we created strain JZ13 in which the wild type cckA gene is retained, but transcriptionally fused to lacZ carried on the integrated plasmid (see Experimental procedures). Introduction of the Plac-ctrA plasmid into JZ13 did not alter cckA expression (Table 4.5). Inspection of the chpT upstream region for the CtrA binding sites did not identify sequences similar to either the full site or the half site. Table 4.5. Regulation of cckA by ctrA. Strain, Genotype Expression plasmid !-galactosidase Sp. Act1 cckA-lacZ, WT cckA (JZ13) Vector (pSRKGm) 14.5 (0.9) cckA-lacZ, WT cckA (JZ13) Plac-ctrA (pJZ008) 13.0 (0.1) 1 Specific activity in Miller units, averages of assays in triplicate (standard deviation) and representative results of two independent experiments each with three biological replicate. IPTG concentration was 200 !M. 4.4.4 Cross complementation between KLH11 and A. tumefaciens homologues Phylogenetic analysis showed that the ctrA gene of KLH11 falls into the non- essential group (Fig. 4.5A) of the two proposed by Greene et al. (2012). In contrast, the ctrA homologue in A. tumefaciens is within the predicted essential 165 group, and disruption of this gene is not possible unless a second copy of ctrA is also provided (J.E. Heindl and C. Fuqua, unpublished). We introduced an expression plasmid that carries the full-length ctrA gene from A. tumefaciens expressed from the Plac promoter (pJEH028) into the Ruegeria sp. KLH11 ctrA- mutant (JZ06) and determined if this A. tumefaciens CtrA protein can restore motility. Strikingly, provision of the A. tumefaciens CtrA can restore motility in the ctrA- mutant to the same extent as the KLH11 ctrA gene (P>0.05) (Fig. 4.5B). Similarly, the plasmids that carry the full-length cckA gene (pJEH010) and chpT gene (pJEH027) of A. tumefaciens were introduced into the KLH11 cckA- (JZ04) and chpT- (JZ05) mutants, respectively, and also partially restored motility at levels slightly lower than the KLH11 cckA and chpT genes can (P<0.05) (Fig. 4.5B). However, the A. tumefaciens cckA plasmid failed to restore the motility in KLH11 cckA- mutant roughly in 30% of our experiments suggesting there may be additional variables that we are not currently controlling (data not shown). 166 Figure 4.5. Comparative analysis of KLH11 CtrA. A) Phylogenetic analyses of CtrA from members of Alphaproteobacteria. CtrA sequences from bacterial species in which CtrA has been studied were chosen for phylogenetic anaylsis. The sequences used were A. tumefaciens C58 (GenBank Accession No. NP_355385), B. abortus (AAL86376), C. crescentus (NP_421829), E. chaffeensis (YP_507798), Magnetospirillum magneticum AMB-1 (YP_419992), Rhodopseudomonas palustris (NP_946978), Rhodobacter capsulatus (AAF13177), Rhodospirillum centenum (YP_002297962), Ruegeria sp. KLH11 (ZP_05124475), Silicibacter sp. TM1040 (YP_613394) and Sinorhizobium meliloti (NP_386824). The star indicates the divergence between organisms in 167 which CtrA is essential or implied to be essential and in which CtrA does affect viability, which was originally proposed by Green et al. (2012). a=motility-related genes are enriched in putative CtrA binding sites (Brilli et al., 2010); b=unable to obtain a ctrA deletion mutant without providing an extra copy of the ctrA gene (Heindl and Fuqua, unpublished results; Mercer, et al., 2010); c=gene target (ccrM) of CtrA is essential (Robertson et al., 2000). The scale bar indicates the number of amino acid substitutions per site. B) Cross complementation of motility between KLH11 and A. tumefaciens homologues. Wild-type KLH11 (EC1) and derivatives were inoculated on MB2216 (supplemented with 0.25% agar) swim agar plates for about 8 days at 28?C. 200 !M IPTG was added to the media. The diameter of the swim ring was measured. Parentheses indicate from which species the relevant homologue is used (At stands for A. tumefaciens). Values are averages of assays performed in triplicate and error bars are standard deviations. 4.4.5 The SsaRI quorum sensing system regulates the transcription of ctrA, chpT and cckA genes In KLH11, the QS circuit ssaRI controls flagellar motility (Zan et al., 2012). We therefore tested whether ssaRI regulates the expression of the ctrA, chpT and cckA genes. Campbell-type insertions in the ctrA, chpT and cckA genes using the suicide vector pVIK112 with internal fragments of each gene created null mutants and simultaneously generated lacZ transcriptional fusions to the disrupted gene (Kalogeraki and Winans, 1997). We used "-galactosidase assays to compare the expression of ctrA, chpT and cckA genes in #ssaI and 168 !ssaR deletion mutants, respectively, from cultures grown to an OD600 ~ 0.6. Expression of the ctrA-lacZ fusion was decreased approximately 25-fold in both the !ssaI and !ssaR mutants (Fig. 4.6A; P<0.01). The chpT-lacZ (Fig. 4.6B) and cckA-lacZ (Fig. 4.6C) fusions were also decreased significantly for the !ssaI and !ssaR mutants, but less dramatically for chpT (2 fold for both mutants; both with P<0.05) and 2-6 fold for cckA (!ssaI, 2-fold, P<0.05; !ssaR, 6-fold, P<0.05). Ectopic expression of plasmid-borne Plac-ssaI and Plac-ssaR restored the expression of cckA, chpT and ctrA genes in the corresponding ssaI and ssaR mutants to levels closer to wild type. AHL synthase gene mutants can usually be rescued by exogenous addition of the appropriate AHL. The KLH11 ssaI mutant motility defect can be partially restored with exogenous addition of synthetic 3-oxo-C16:1 !11-HSL, an AHL similar to that specified by SsaI (Zan et al., 2012). We therefore tested whether this AHL could rescue ctrA, chpT and cckA expression in the corresponding mutants. Surprisingly, addition of this AHL failed to restore the expression of the ctrA, chpT or cckA lacZ fusions in the !ssaI mutant (P>0.05, Fig. 4.6A-C). In all of these Campbell insertions, generation of the fusion also disrupts the gene. To examine QS-dependent expression of these genes in an otherwise wild type background, we introduced the following plasmid-borne fusions into the !ssaI strain: PctrA-lacZ (pJZ011), PchpT-lacZ (pJZ010) and PcckA-lacZ (pJZ009). The expression of these lacZ fusions was monitored by "-galactosidase assays in the presence and absence of 2 #M 3-oxo-C16:1 !11-HSL. A 2.5-fold increase for 169 PctrA-lacZ (P<0.05) and 4-fold increase for PcckA-lacZ (P<0.05) was observed in the presence of 2 !M 3-oxo-C16:1 "11-HSL (Table 4.6). 170 Figure 4.6. Regulation of cckA, chpT and ctrA gene expression by the ssaRI system. Results of !-galactosidase assays in detecting the expression of ctrA?lacZ (A), chpT-lacZ (B) and cckA-lacZ (C) in "ssaI and "ssaR mutants. Plasmids Plac-ssaI (pEC108) and Plac-ssaR (pEC112) were conjugated into the "ssaI and "ssaR mutants, respectively, to restore the expression of the ctrA, chpT and cckA genes. 2 #M 3-oxo-C16:1 "11-HSL was added into "ssaI ctrA- lacZ, "ssaI chpT-lacZ and "ssaI cckA-lacZ strains, respectively. Filled asterisks indicated statistically significant differences between the indicated strain and 171 wild-type quorum sensing strain. Unfilled asterisks indicated statistically significant differences between the quorum sensing complemented strains and quorum-sensing mutants for the expression of the ctrA, chpT and cckA genes. Representative results of several independent experiments each with three biological replicates are presented. Addition of exogenous AHLs failed to restore expression of ctrA, chpT and cckA in !ssaI backgroud (P>0.05). Values are averages of assays performed in triplicate and error bars are standard deviations. Table 4.6. Complementation of PcckA, PctrA expression by exogenous AHL. Wild type1 !ssaI1 Fusions No AHL +AHL2 No AHL +AHL2 PcckA-lacZ (pJZ009) 48 (2.3) 48.5 (2.6) 1.6 (0.2) 8 (0.3)3 PctrA-lacZ (pJZ011) 182.6(15.7) 182.6 (15.6) 2.9 (0.6) 7.1 (0.4)3 a #-Galactosidase activity was expressed in Miller units. b 3-oxo-C16:1 "11-HSL (2 !M) was added. c P <0.05 when compared to the expression level in "ssaI strain without AHL. 172 4.4.6 SsaRI regulate ctrA, chpT and cckA expression indirectly The gene expression experiments in KLH11 did not allow us to distinguish direct or indirect QS regulation of the CckA-ChpT-CtrA pathway. We therefore electroporated plasmids carrying PctrA-lacZ (pJZ011) and Plac-ssaR (pEC112) into the AHL- A. tumefaciens NTL4 (Ti-plasmidless) derivative to test whether the QS- dependent expression of ctrA was due to SsaR-dependent activation of the ctrA promoter. In this same background, SsaR and 3-oxo-C16:1 !11-HSL strongly activate the expression of the PssaI promoter (Zan et al, 2012). A. tumefaciens NTL4 harboring PctrA-lacZ (pJZ011) plus a vector (pBBR1-MCS5) was used as a negative control. Expression of the PctrA-lacZ fusion was unaffected by addition of 2 "M 3-oxo-C16:1 !11-HSL (Table 4.7). These results indicate that SsaR indirectly regulates the expression of PctrA-lacZ and that an intermediary regulator(s) must exist. We used the same approach to test the regulation of chpT (PchpT-lacZ, pJZ010) and cckA (PcckA-lacZ, pJZ009) by SsaR with 2 "M 3- oxo-C16:1 !11-HSL. These findings suggest that SsaR and 3-oxo-C16:1 !11- HSL do not directly activate the expression of ctrA, chpT and cckA genes (Table 4.7). 173 Table 4.7. Expression of KLH11 PcckA, PchpT and PctrA promoters in an AHL- host1 . !-Galactosidase Sp. Act.2 Expression plasmid Fusion No AHL +AHL3 Vector (pBBR1-MCS5) cckA-lacZ (pJZ009) 299.4 (38.0) 326.9 (33.4) Plac-ssaR(pEC112) cckA-lacZ (pJZ009) 295.0 (26.6) 328.7 (12.4) Vector (pBBR1-MCS5) chpT-lacZ (pJZ010) 33.4 (1.2) 32.5 (1.2) Plac-ssaR(pEC112) chpT-lacZ (pJZ010) 24.3 (0.5) 23.3 (1.9) Vector (pBBR1-MCS5) ctrA-lacZ (pJZ011) 123.8 (20.2) 139.4 (2.8) Plac-ssaR(pEC112) ctrA-lacZ (pJ011) 138.6 (18.6) 137.7 (10.1) 1 All strains derived from Ti- plasmidless A. tumefaciens NTL4. 2 Specific activity in Miller units, averages of assays in triplicate (standard deviation) and representative results of two independent experiments each with three biological replicates. 3 3-oxo-C16:1 !11-HSL (2 "M ) was added. 4.4.7 Ectopic expression of ctrA restores motility to the QS deletion mutant Provision of either the plasmid-borne Plac-ssaI (pEC108) or Plac-ssaR (pEC112) to the corresponding mutants (!ssaI cckA-, !ssaI chpT,- !ssaI ctrA- and !ssaR cckA-, !ssaR chpT-, !ssaR ctrA-) does not restore motility (Fig. 4.7), although they did restore nearly wild type expression levels for each lacZ fusion (Fig. 4.6). This is due to the disruption of the targeted gene by the Campbell 174 insertions. Consistent with our previous studies however (Zan et al., 2012) the Plac-ssaI (pEC108) or Plac-ssaR (pEC112) plasmids effectively complement motility in the "ssaI and "ssaR mutants, respectively (Fig. 4.8). This suggests that cckA, chpT and ctrA are required for motility and act downstream of the ssaRI system. Accordingly, IPTG-induced expression of the Plac-ctrA (pJZ008) in "ssaI and "ssaR did however restore motility (Fig. 4.8). Controls with the vector alone did not correct the motility defect in any of these derivatives (data not shown). Interestingly, the Plac-ctrA plasmid could not restore motility in the !ssaI cckA-, "ssaI chpT-, "ssaR cckA- or "ssaR chpT- mutants, respectively (Fig. 4.9). Thus, cckA and chpT genes are required for the suppression of the !ssaI or !ssaR mutant phenotypes by CtrA. 175 Figure 4.7. The cckA-chpT-ctrA phosphorely system is required for the ssaRI system to control motility. Plac-ssaI (pEC108) was conjugated into "ssaI cckA-, "ssaI cphT- and "ssaI ctrA - double mutants and Plac-ssaR (pEC112) was conjugated into "ssaR cckA-, "ssaR cphT- and "ssaR ctrA - double mutants, respectively. Strains were inoculated on MB2216 (supplemented with 0.25% agar) swim agar plates for about 8 days at 28?C. The results were representatives of several independent experiments each with three biological replicates. 176 Figure 4.8. Suppression of motility defects in !ssaI and !ssaR mutants by CtrA. Plac-ctrA plasmid (pJZ008) was conjugated into !ssaI and !ssaR mutants, respectively. The conjugants were selected and inoculated on swim agar plates for 8 days at 28 ?C. 200 "M IPTG was added to the media. The !ssaI mutant complemented with Plac-ssaI (pEC108) and the !ssaR mutant with Plac-ssaR (pEC112) were used as positive controls. Wild type KLH11 (EC1) was used as a positive and the !ssaI and !ssaR strains were used as negative control. The results were representatives of several independent experiments each with three biological replicates. 177 Figure 4.9.The cckA and chpT genes are required for the function of CtrA. Plac-ctrA was conjugated into "ssaI cckA-, "ssaI chpT-, "ssaR cckA-, and "ssaR chpT-, respectively. The conjugants were selected and inoculated for swim motility assay as described above. The results were representatives of several independent experiments each with three biological replicates. 178 4.5. Discussion The cckA-chpT-ctrA phosphorelay system has been well characterized in C. crescentus in which the expression levels of 144 genes are affected due to the loss of the ctrA gene (Laub et al., 2000). However, little is known about this phosphorelay system in the highly abundant marine Roseobacter clade. In several alphaproteobacterial systems, cckA and ctrA are essential, although little is known about the essentiality of the chpT gene other than in C. crescentus, R. capsulatus and Silicibacter sp. TM1040 (Biondi et al., 2006; Belas et al., 2009; Mercer et al., 2012). However, our results clearly reveal that all the cckA-, chpT- and ctrA- mutants have growth rates nearly identical to that of wild type KLH11, demonstrating that they are non-essential under laboratory conditions. This observation is consistent with the function of cckA and ctrA in R. capsulatus (Lang and Beatty, 2000), Silicibacter sp. TM1040 (Belas et al., 2009), and R. centenum (Bird and MacKrell, 2011). The ctrA gene is not essential in Magnetospirillum magneticum AMB-1 (Greene et al., 2012) but the essentiality of cckA has not been examined. Moreover, qRT-PCR revealed that KLH11 CtrA did not regulate the expression of the ftsZ or ccrM genes in KLH11 (Table 4.4). In C. crescentus CtrA directly regulates the expression of genes involved in cell division such as ftsZ, encoding the primary division protein, and ccrM, a DNA methyltransferase that modifies sequences at the replication origin to coordinate the timing of genome replication with the cell division cycle (Laub et al., 2002). Taken together, the non-essentiality of ctrA and the lack of regulation for ftsZ and 179 ccrM indicate that CtrA probably does not play a role in cell cycle control in KLH11. Our findings provide evidence that CckA, ChpT and CtrA activate swimming motility and control the biosynthesis of flagella (Figs 4.1 and 4.4), similar to a portion of their roles in C. crescentus, although the intimate relationship between the cell cycle and flagellation confounds this role (Quon et al., 1996, Laub et al., 2000). Furthermore, qRT-PCR revealed ctrA to regulate all the motility-related genes that were checked. We searched for both full and half ctrA binding sites using the motifs as determined from Laub et al. (2002) in the upstream regions of these motility-related genes, but no sequences with high similarity were identified. This is similar to the situation in R. capsulatus, in which expression of motility related genes is decreased in a "ctrA strain but none of these genes have clear ctrA binding motifs (Mercer et al., 2010). Identification of presumptive CtrA recognition sites in the upstream of ctrA suggested feedback on its own transcription. Our results indeed showed the positive feedback on the ctrA expression. This is consistent with findings in C. crescentus, although in that bacterium CtrA negatively regulates one promoter (P1) and positively regulates a second (P2) (Domian et al., 1999). It is unclear whether or not the KLH11 ctrA gene has multiple promoters. In S. meliloti, CtrA~P can bind proximal to its two ctrA promoters, presumably regulating the transcription of the ctrA gene (Barnett et al., 2001). Similarly, in the obligatory intracellular pathogen Ehrlichia chaffeensis, the CtrA protein can also bind proximal to its own promoter (Cheng et al., 2011). In contrast, for R. capsulatus, 180 ctrA-binding site was identified in the ctrA promoter region (Lang and Beatty, 2000, but CtrA does not affect its own transcription (Leung et al., 2013), suggesting that this autoregulatory loop is not conserved in all bacteria with CtrA. A search of the promoter region of the cckA gene also identified presumptive ctrA binding sites (Fig. 4.1A), suggesting that CtrA could potentially regulate the expression of the cckA gene. However, provision of ctrA in the KLH11 derivative with the integrated cckA-lacZ that maintains an intact copy of cckA (JZ13) does not affect cckA gene expression (Table 4.6). This is similar to R. capsulatus, in which the expression of cckA gene is not affected in a "ctrA strain, although interestingly, loss of the ctrA gene leads to a decrease of CckA protein levels (Mercer et al., 2010). We do not know whether the loss of the ctrA gene would affect the amount of CckA protein in KLH11. This also emphasizes that the presence of upstream sequences with similarity to CtrA binding sites does not necessarily mean that the associated gene (in this example cckA) is regulated by CtrA. Taken together, this likely reflects the limits on current understanding of what comprises a CtrA binding site outside of C. crescentus. Greene et al. (2012) proposed two groups of CtrA in the Alphaproteobacteria: in one group ctrA is essential and in the other it is non-essential, but in both groups it exerts control over motility. KLH11 ctrA clearly falls into the non- essential group by sequence comparisons (Fig. 4.5A). Interestingly, although the cckA-chpT-ctrA pathway is essential in A. tumefaciens (Kim et al., submitted), plasmid-borne expression of each of these A. tumefaciens genes can cross- complement the corresponding mutants in KLH11 for their impact on motility. 181 This cross-complementation suggests that the functionality of this pathway is well conserved and its role in controlling motility is ancestral among the Alphaproteobacteria. These proteins have retained their basic activities, even though the influence of this pathway can be expanded to include essential functions. It remains unclear whether the pathway?s essentiality is derived or ancestral among the Alphaproteobacteria. The CckA protein from A. tumefaciens shows inconsistent complementation in the KLH11 cckA- mutant, which hints at an additional signal(s) that may impact the activity of the A. tumefaciens CckA protein. Our results clearly show that the cckA-chpT-ctrA phosphorelay system is indirectly transcriptionally regulated by the SsaRI quorum sensing circuit. In C. crescentus, the transcription of the cckA gene is cell cycle dependent, but not affected by CtrA (Laub et al., 2000; Laub et al., 2002). Moreover, the level of the CckA protein is constant during the cell cycle whereas the phosphorylation of CckA is subject to temporal and spatial regulation (Jacobs et al., 1999; Jacobs et al., 2003). We do not know what signal(s) it is to which the CckA protein responds. However, it is plausible that the CckA protein may sense population density-associated signals and thus it can coordinate the activation of motility with the cell density. Of note, the "ssaR mutant exhibits a more profound deficiency in the cckA expression than the "ssaI mutant (Fig. 4.6C). One explanation is that SsaR is able to respond to the AHL levels synthesized in the !ssaI mutant, in which the ssbRI system remains genetically intact (Zan et al., 2012). Meanwhile, the activity of CtrA, a key factor in driving the cell cycle, is 182 tightly regulated at the levels of transcription, phosphorylation, degradation, and protein-protein interaction (Gora et al., 2010). On the transcriptional level, the C. crescentus ctrA gene is activated by the two-component type response regulator GcrA (Holtzendorff et al., 2004). In R. capsulatus, it was found that the LuxR- holomogue GtaR indirectly represses the transcription of ctrA while the AHL synthesized by GtaI derepresses its transcription (Leung et al., 2013). However, it was unclear whether QS affects the transcription of cckA and chpT genes in this bacterium. Interestingly, we can complement the expression of ctrA, chpT and cckA?lacZ fusions as Campbell insertions in the "ssaI background by providing the ssaI gene in trans but were not able to restore their expression by addition of exogenous synthetic 3-oxo-C16:1 "11-HSL (Figs. 4.6A-C). It is known that addition of AHL into "ssaI is able to partially restore motility (Zan et al., 2012) and data in this study clearly show that CtrA acts downstream of the ssaRI system to control flagellar assembly and motility. We reason that two factors can contribute to this observation: 1) The long chain AHL we added might not be able to partition into the cell from exterior efficiently due to its hydrophobicity. It has been shown that the long chain AHLs preferentially associate with the cell rather than being released extracellularly and that AHLs that partition into the cell membrane may not function as signals (Schaefer et al., 2002). Addition of the same AHL can stimulate the expression of the ssaI gene (Zan et al., 2012); however, the stimulatory effect of AHL on ssaI might not be propagated onto the cckA-chpT-ctrA effectively because of the indirect regulatory link between ssaRI 183 and cckA-chpT-ctrA. 2) There might be some positive feedback on the ssaRI system by this pathway. In the Campbell insertion, the gene is disrupted. It is possible that an intact copy of this pathway is required for optimal expression. Indeed, the significant, yet weak activation of the plasmid borne fusions by the addition of AHL in the "ssaI strain in which cckA-chpT-ctrA are intact supports this speculation (Table 4.6). Ectopic expression of ctrA in either the "ssaI or "ssaR mutant can restore motility. However, failure to restore motility in either "ssaI cckA- or "ssaR cckA- by provision of ctrA suggests that phosphorylation by CckA is required for the function of CtrA, although it is possible that CckA can also phosphorylate other regulator (s), which is also required for motility. Furthermore, chpT is also required for the function of CtrA since providing ctrA into "ssaI chpT- or "ssaR chpT- mutants does not restore motility. It was plausible that high levels of CtrA expression from the Plac-ctrA plasmid might mask the absence of the upstream components of the pathway, CckA and ChpT, via phosphorylation-independent CtrA activity. C. crescentus CtrA can bind to some sites in a phosphorylation- independent manner in vivo (Spencer et al., 2009) and several different response regulators exhibit phosphorylation-independent activity (Ma et al., 1998; Sch?r et al., 2005). CtrA might crosstalk with other phospho-donors such as other kinases or acetyl phosphate. However, it is clear that the intact CckA-ChpT phosphorelay is required for CtrA motility control. Taken together, our data support the model shown in Figure 4.10. SsaRI acts upstream of the cckA-chpT- ctrA phosphorelay system and indirectly regulates the transcription of all the 184 three genes, most dramatically through ctrA expression. CckA and ChpT are required via presumptive phosphotransfer to CtrA, which positively feeds back on its own expression, and controls flagellar assembly and motility. CtrA could thus be the potential flagellar master regulator in KLH11. This is similar to the model reported in the rice pathogen B. glumae, in which the tofRI QS pathway controls the regulator qsmR which in turn directly controls the flagellar master regulator flhDC (Kim et al., 2007). However, the presumptive regulator that links SsaRI to ctrA remains to be identified in KLH11. Our data also suggests that there may be feedback from the CckA-ChpT-CtrA pathway on the SsaRI system. Roseobacters are a key marine bacterial group with biogeochemical relevance and are also commonly found as symbionts of marine invertebrates, including sponges. Roseobacters are often highly abundant in phytoplankton blooms or near macroalgae and associated with organic particles (Slightom and Buchan, 2009). Motility is likely to be critical in many of these interactions. Our previous work contributed to understanding the role of quorum sensing in activating motility specifically at high cell density in the sponge-associated Roseobacter, Ruegeria sp. KLH11 (Zan et al. 2012), raising the possibility that this is a widespread mechanism in Roseobacters. In KLH11, flagellar motility is controlled by the SsaRI system and AHL quorum sensing. Surprisingly, the KLH11 genome and that of its relative R. pomeroyi DSS-3 lack any recognizable chemotaxis genes (Moran et al., 2004; Zan et al., 2011a)(See Appendix 1). Therefore QS regulation of motility is not simply augmenting the process, but appears to be its primary control mechanism. SsaR and long chain AHLs are 185 required for cckA, chpT and ctrA gene expression, revealing at least a portion of this central control pathway, although additional environmental signals may also function through the CckA-ChpT-CtrA cascade. The work reported here provides a discrete regulatory link between flagellar locomotion and population density in KLH11. Figure 4.10. A tentative model for the ssaRI to cckA-chpT-ctrA regulatory circuit to control KLH11 flagellar motility. The solid lines with the arrows indicate activation. The dashed line with arrows indicates the potential phosphate flow from CckA to CtrA via ChpT. The curved lines with arrows around CtrA and SsaRI indicate positive feedback loops. The ?X? indicates the unknown regulator (s). The thicker line from X to ctrA indicates stronger regulatory effect. 186 Chapter 5. Isolation and screening of AHL- and AI-2- producing bacteria and development of Ruegeria sp. KLH11 as a model to study bacterial colonization of sponges 187 5.1. Abstract Previous studies have revealed that the sponge symbionts belonging to the roseobacterial Silicibacter-Ruegeria (SR) subgroup can produce acylhomoserine lactone (AHL) signal molecules but only a small number of isolates was screened (Mohamed et al., 2008c). To confirm and extend this finding, I screened 420 bacterial isolates from M. laxissima and I. strobilina collected in October 2008, March 2010 and June 2011 for AHL production. The AHL-producing bacteria were identified by 16S rRNA gene sequence analysis. Results showed that 44% of these isolates are able to produce AHLs and 77% belongs to the SR subgroup of Roseobacter clade. Moreover, these SR isolates were isolated consistently from these two sponges, suggesting the potential importance of these bacteria to their marine sponge hosts. In addition, degenerate primers designed based on known luxS sequences from the genus Vibrio were able to amplify the luxS genes from all the Vibrio strains. A biological reporter V. harveyi TL-26 was used to detect the activity of AI-2 (synthesized by LuxS) and results showed that all the Vibrio isolates were able to turn on light production in the TL-26 biological reporter. Results of reverse transcription (RT)-PCR showed that the AHL synthase gene ssaI is actively expressed in the sponge microbial community and the results of TLC assay coupled with AHL biological reporter assay revealed the presence of AHL molecules in the sponge tissues. Genetically modified KLH11 derivatives were used to investigate the role of QS in the process of bacterial colonization of the sponge M. laxissima. An interesting pattern was observed 188 that the ssaR mutant seems to better colonize the sponge compared with the wildtype. 189 5.2. Introduction Quorum sensing can fine tune gene expression in responding to cell density and is commonly found in symbiotic or pathogenic bacteria associated with eukaryotic hosts (Gurich and Gonz?lez, 2009). It was first described in the non- pathogenic marine luminescent symbiotic bacterium Vibrio fischeri (Nealson et al., 1970). Current QS research mainly focuses on bacterial pathogens, such as V. cholerae, P. aeruginosa and A. tumefaciens. Cicirelli et al. (2008) suggested that Roseobacters actually might be the dominant AHL producers in the marine environment. For example, the AHL-producing bacteria isolated from the marine sponges M. laxissima and I. strobilina mainly belong to the Silicibacter-Ruegeria subgroup of the Roseobacter clade; furthermore, sponges harbor a higher proportion of AHL producers in their culturable bacterial communities compared with the surrounding seawater, although only a small set of bacterial isolates are tested (Mohamed et al., 2008c). Marine sponges harbor a high abundance of bacteria in their tissue and thus provide an ideal environment for bacterial quorum sensing. However, prior to this study it was unknown whether the association of AHL-producing bacteria with these sponges is consistent and whether AHL molecules are present in the sponge tissues. Autoinducer-2 (AI-2) is another well-known molecular cue in bacterial signaling and has been extensively analyzed in V. harveyi and V. cholerae, where it is involved in regulation of bioluminescence and virulence-associated traits (Ng and Bassler, 2009). The activated methyl cycle is a crucial metabolic pathway to recycle homocysteine from the major methyl donor S-adenosyl 190 methionine (SAM). LuxS, a S-ribosylhomocysteinase, catalyzes part of the cycle and functions to convert S-ribosylhomocysteine to homocysteine; meanwhile, it can also, as a side reaction, synthesize 4,5 dihydroxy-2, 3-pentanedione (DPD), the precursor of AI-2. DPD can spontaneously give rise to several furanone derivatives, collectively referred to as AI-2. The dual roles of luxS in metabolism and in AI-2 formation have led to controversy regarding its function in quorum sensing (Rezzonico and Duffy, 2008). The luxS gene is broadly distributed in gram-negative and gram-positive bacteria, although it is absent in all of the Alphaproteobacteria. Thus, it is interesting to study the luxS genes in the bacteria associated with marine sponges, especially Vibrio sp., since they represent the best-studied models. Symbiosis refers to two or more different organisms living closely together and is a widespread phenomenon. Within the wide range of different symbioses, the microbe-eukaryote symbioses have received tremendous attention. Sponge- microbe symbiosis has been developed as a primary model for studying marine symbiosis (Taylor et al., 2007). Marine sponges harbor highly abundant and diverse microorganisms that are different from these of the surrounding planktonic communities. Significant progress has been made in understanding the composition and even the dynamics of the sponge communities (Taylor et al., 2007; Webster and Taylor, 2012). However, fundamental questions remain unanswered, such as: How is the symbiosis established and maintained? How do the host and microbe select and recognize each other? Does quorum sensing play roles in the symbiotic relationships? Insights have been provided in 191 several other symbiotic systems. For example, one well-established model is the V. fischeri-squid symbiosis. Genetic studies have shown that flagellar motility is required for bacteria to colonize the light organ and nonmotilte strains failed to colonize the light organ (Graf et al., 1994). The bioluminescence controlled by QS is also critical for the maintenance of symbiosis (Visick et al., 2000). Furthermore, it has also been shown that a single bacterial regulatory gene can determine the host specificity (Mandel et al., 2009). On the other hand, inhibition of motility by quorum sensing is essential for effective nodule invasion by the alfalfa symbiont Sinorhizobium meliloti and the AHL synthase SinI and the receptor ExpR seem to play opposite roles in the nodule invasion process (Gurich and Gonz?lez, 2009). In KLH11, motility is under positive control of ssaRI QS system although KLH11 does not encode the chemotaxis system in its genome (Zan et al., 2011a)(See Appendix 1). Importantly, a suite of genetic tools can be applied to KLH11, which allow us to readily manipulate the genome and ask important questions (See Chapters 2 - 4). Therefore, we developed KLH11 as a model strain to study whether QS in KLH11 plays a role in colonization of its sponge host by this bacterium. In this study, I aimed at establishing that SR isolates are the important players for AHL production in marine sponges by screening a large number of bacterial strains isolated from sponges collected over several different years and then tried to connect our laboratory findings with the native environment by showing the presence of AHL molecules in the sponge tissues and expression of the AHL 192 synthase ssaI gene in situ within sponges. Finally efforts were directed towards building a model for studying the bacterial colonization process. 193 5.3. Experimental procedures 5.3.1 Sponge collection, bacterial isolation and identification, and AHL screening Three individuals of each of the marine sponges M. laxissima and I. strobilina were collected by SCUBA diving at Conch Reef, Key Largo, FL, USA at a depth of ca. 20 m in late October 2008, late March 2010 and June 2011. Water samples were collected within 1 m of the sponges at a depth of ca. 20 m in a sterile 20-L container. The water salinity was measured to be 36 ppt using a portable refractometer at all three times of collection and the water temperature was ca. 28?C, 24?C and 30?C in October 2008, March 2010 and June 2011, respectively. Sponges were rinsed with sterile artificial seawater three times to remove transiently associated bacteria and then processed for isolation of bacteria. Sponge tissue (1 cm3) was ground in artificial seawater using a sterile mortal and pestle and 10-fold serial dilutions were plated on Marine Agar 2216 (BD Biosciences, Franklin Lakes, NJ). Plates were incubated at 30?C for 1 week. Approximately 80 colonies were randomly picked and screened for AHL production. The method used for AHL screening was described by Mohamed et al. (2008c). Briefly, four strains were arrayed onto one MA2216 plates and were incubated at 30?C for a week. The reporter A. tumefaciens KYC55 was grown to mid-exponential phase in ATGN medium. Cells were harvested by centrifugation at 5000 x g for 10 min and the pellet was washed and resuspended in 30% glycerol to adjust the OD600 to ~ 12.0. Then one ml of the concentrated reporter 194 was added into 100 ml of ATGN with 0.6% agar (W/V) and 20 !g ml-1 X-gal. Twenty -thirty ml of this mixture was overlaid with MA2216 plates and were incubated at 30?C for 24-48 h. Serial dilutions of water samples were processed similarly for bacterial isolation and screening of isolates for AHL production. The AHL producers were identified by 16S rRNA gene sequencing. Bacterial genomic DNA from AHL-producing bacteria was extracted from isolates using the UltraClean Microbial DNA isolation kit (MO BIO Laboratories, Inc. Carlsbad, CA) following the manufacturer?s manual. Almost full-length 16S rRNA gene fragments were PCR-amplified using universal primers 27F and1492R as described by Enticknap et al. (2006). PCR products were sequenced using an ABI PRISM 3130xl genetic analyzer and primers 27F and 1492R. Sequences were assembled using online software CAP3 (http:// pbil.univ-lyon1.fr/cap3.php) with a manual check. Chimeric sequences were identified by using the CHECK_CHIMERA program of the Ribosomal Database Project (Maidak et al., 1999) and the sequences were analyzed initially by using the BLASTn tool at the National Center for Biotechnology Information website (NCBI). Partial 16S rRNA gene sequences of all the AHL positive bacteria were aligned using Clustal X 2.0.12 (http://www. clustal.org/) and a phylogenetic tree was constructed using software MEGA 4.0 (http://www.megasoftware.net/). 5.3.2 luxS gene amplification from the vibrios Degenerate primers based on an alignment of available Vibrio luxS gene sequences were designed for amplification of luxS gene fragments using online primer design software consensus-degenerate hybrid oligonucleotide primers 195 (Rose et al., 2003). The primers were designated VluxsF (5? TGCTGGACTCCTTCACCGTNGAYCAYAC-3?) and VluxsR (5?- TGCATGGCGGCGGTNCCRCAYTGTT-3?). PCR mixtures consisted of 50 !l containing two units Platinum Taq polymerase (Invitrogen Life Technologies, Carlsbad, CA), 1X PCR Buffer, 2 mM MgCl2, 200 mM dNTPs (Fermentas, Glen Burnie, MD), 0.2 !M of each primer and 10?20 ng of genomic DNA or distilled water as a negative control. PCR cycling conditions for Vibrio luxS gene amplification consisted of 94?C for 5 min followed by 30 cycles of 1 min at 94?C, 1 min at 52?C, and 1 min at 72?C. A final 10 min extension step was done at 72?C. PCR reactions were performed in a PTC-200 cycling system (Bio-Rad, Hercules, CA). The PCR products with expected size ~ 400 bp were purified using QIAquick gel extraction kit (Qiagen, Valencia, CA) and sequenced as described above for 16S rRNA gene fragments using primer VluxsF. 5.3.3 Measurement of AI-2 activity AI-2 activity was detected by using the reporter strain V. harveyi TL-26 generously provided by Drs. Long and Bassler, in which all three signaling pathways are disrupted. Receptor genes for AHL and CAI-1 are mutated as well as the luxS gene. Thus, TL-26 can respond only to exogenously added AI-2 (not AHLs or CAI) and cannot produce AI-2 (Long and Bassler; personal communication). All the Vibrio isolates and V. harveyi TL-26 were grown overnight in Luria?Bertani medium with 2% NaCl. All the tested strains were grown to similar OD600. Culture supernatants of test strains were filter sterilized (0.22 !m) and 60 !l of sterile supernatant was added to 140 !l AB medium 196 (Greenberg et al., 1979) containing V. harveyi TL-26 (inoculated 1:5000 from an overnight culture). V. mimicus ATCC 33653 was used as a positive control. The culture was incubated at 30?C with shaking at 180 rpm for 16 h and light production was measured by using a FLUOstar OPTIMA fluorescence microplate reader (BMGlabtech, Cary, NC). The induction of luminescence by each tested supernatant was expressed relative to that in a negative control comprising sterile Luria?Bertani medium with 2% NaCl instead of culture supernatant, similar to the method described by Zhu and Mekalanos (2003). 5.3.4 Organic extraction of sponge tissues and TLC overlay assay AHLs were extracted from sponge tissues using a modified Bligh Dyer procedure (Bligh and Dyer, 1959). In brief, frozen tissues were ground to powder in liquid nitrogen. Ten grams of ground sponge were re-suspended in 10 ml of methanol. The mixture was homogenized by vigorous vortexing and then was left overnight at room temperature. Five ml of chloroform was added to the mixture followed by sonication four times at full power. To completely homogenize the mixture, 5 ml of chloroform and 2.5 ml water were added after sonication. This mixture was further vigorously vortexed and shaken followed by centrifugation at 2,000 rpm for 1 min. The bottom layer was carefully transferred to a glass test tube using 9? disposable glass Pasteur pipettes. To increase the extraction efficiency, 5 ml of chloroform was added to the top layer followed by vigorous vortex for several times and then centrifugation at 2,000 rpm for 1 min. Again, the bottom layer was transferred to the test tubes and was evaporated to 197 complete dryness. The residue was resuspended in 6 ml of 1:1 isooctane: ethyl ether for solid phase extraction, as described previously (Gould et al., 2006). Organic extract from ca. 10.0 g tissues of each of three sponge individuals were dissolved in 40 !l methanol, and 20 !l was loaded onto a C18 RP-TLC plate (Mallinckrodt Baker, Phillipsburg, NJ). AHLs were detected using the A. tumefaciens ultrasensitive reporter (Zhu et al., 2003) as described in Chapter 2 Section 2.3.8. 5.3.5 RNA extraction and RT-PCR from sponge tissue. Freshly collected sponge samples were rinsed three times with sterile artificial seawater (ASW) and were stored in RNAlater solution (Qiagen, Valencia, CA) immediately for later RNA extraction. Total RNA was extracted from sponge samples using the TissueLyser system (Qiagen, Valencia, CA) and an RNeasy Mini spin column provided by the AllPrep DNA/RNA mini kit (Qiagen, Valencia, CA) per manufacturer?s instruction. RNAase-free DNAase (Qiagen, Valencia, CA) was added to RNAeasy mini columns in order to totally remove any DNA. Reverse transcription (RT) reactions were performed by using ssaI specific primer RtaR (below) and ThermoScript RT-PCR system (Invitrogen, Grand Island, NY). The conditions recommended by the manufacturer were used for the reverse transcription reaction. Primers set RtaF (5?- AAGTACTTGACGAAATGTTCG AACTG-3?) and RtaR (5?- GGTCGATCACGGTAATGATGTCTTC -3?) were used to amplify the ssaI gene from M. laxissima sponge cDNA. RNA samples without the RT step were used as negative controls to test for contaminating DNA. 198 5.3.6 Colonization of sponge cell aggregates A derivative of wild type KLH11 resistant to the antibiotic rifampcin (Rif) was selected and was further modified to have resistance to the antibiotic tetracycline (Tc) by conjugating the pSRKTc plasmid into it (Khan et al., 2008). The ssaI insertional mutant EC2 has a kanamycin (Km) resistance marker carried on the plasmid pVIK112 (Kalogeraki and Winans, 1997) and the plasmid pJZ402 (From S.C. Winans, Cornell University) that carries a spectinomycin (Sp) resistance marker was electroporated into EC2. These strains were used to study colonization of sponge cell aggregates. Roughly 1 cm3 of sponge tissue from sponges that were maintained in tanks at the Aquaculture Research Center (ARC) in the Institute of Marine and Environmental Technology (IMET) was ground in 50 ml artificial seawater (Instant Ocean, Aquarium Systems, Pearl City, HI) and squeezed through a 100 !m mesh net. Two ml of the sponge cell homogenates were distributed into small petri dishes (60mm X 15 mm) and these cells spontaneously aggregated. Wild type KLH11 (labeled with Rif and Tc resistant markers) and the ssaI mutant (labeled with Km and Sp resistant markers) were grown up to stationary phase and inoculated into these cell aggregates at 106 cells.ml-1; furthermore, cells of these two strains were mixed in a 1:1 ratio, also at a concentration of 106 cells ml-1 for each strain. This bacterial cell mixture was used in competition experiments. Artificial seawater was used as a negative control. The sponge cell aggregates were sampled at three different time points after inoculation (0 h, 6 h, 24 h) and were filtered through 5 !m membranes (Waterman, Boca Raton, FL). The cells that were retained on 199 the membranes were plated on plates with Rif (200 !g ml-1) + Tc (20 !g ml-1) and plates with Km (100 !g ml-1) +Sp (!g ml-1). The plates were incubated at 30?C for 4 days and the number of colonies on each plate was counted. 5.3.7 Colonization of whole sponges Nine individual of M. laxissima (each ca. 6 cm in diameter) were collected off Key Largo, FL on November 14th 2012 and were shipped to IMET overnight. Sponges were maintained in tanks filled with circulating artificial seawater (ASW, 80 L per tank) at ARC facility located in IMET (described in detail in Mohamed et al., 2008b). The water temperature was ca. 21?C. After acclimation to the laboratory environment for ca. 10 days, these nine sponges were transferred to three new tanks, each containing three randomly chosen sponge individuals. The ssaR insertional mutant EC4 has a kanamycin (Km) resistance marker carried on the plasmid pVIK112 (Kalogeraki and Winans, 1997) and the plasmid pJZ402 that carries a spectinomycin (Sp) resistance marker was electroporated into this strain EC4. The pJZ402 plasmid also carries a gene encoding dsRed fluorescence (gift from S.C. Winans). The Rif and Tc resistant derivative of wild type KLH11 strain described above and the strain EC4 labelled with Km and Sp were used to colonize the whole sponges. Both of these two strains were grown in MB2216 at 28?C with shaking at 200 rpm up to stationary phase. We used three different treatments to colonize the whole sponges. The Rif and Tc resistant derivative of wild type KLH11 was added to one tank containing three sponges; EC4 was added to another tank with three sponges; and wild type KLH11 and EC4 strains were added at 1:1 ratio to a third tank with three 200 sponges. Each of the two strains was added at a final concentration rougly 1.0 X 106 CFU ml-1. After 14 hours, each of the three sets of sponges was rinsed with ASW and transferred to a new tank filled with ASW. All sponges were processed for bacterial counts 48 h after this transfer. In brief, 1.5 g of sponge tissue was cut from each sponge, ground in 10 ml sterile ASW and serial dilutions of the sponge tissue were plated on MA2216 supplemented with Rif (200 !g ml-1) + Tc (20 !g ml-1) and /or supplemented with Km (100 !g ml-1) + Sp (100 !g ml-1). The plates were incubated at 30?C for 4 days and the number of colonies on each plate was counted. 201 5.4. Results 5.4.1 Screening of AHL-producing bacteria from marine sponges In total, 420 strains isolated from M. laxissima and I. strobilina collected in three different seasons over three years from Key Largo, FL, USA were screened for AHL production. Results showed that 109 out of 225 strains (48 ? 12%) from M. laxissima and 75 out of 195 strains (38 ? 8%) from I. strobilina could activate the AHL biological reporter A. tumefaciens KYC55 strain. Furthermore, 4 out of 86 strains (5%) isolated from the surrounding seawater could also activate the reporter. Results of 16S rRNA sequences analyses showed that 86 out of 109 AHL producing strains from M. laxissima and 55 out of 75 AHL producing strains from I. strobilina belong to the Silicibacter-Ruegeria (SR) subclade of the Roseobacter clade in Alphaproteobacteria. The phylogenetic analysis based on the 16S rRNA sequences of all the SR AHL producers is shown in Figure 5.1. All the SR isolates share >98% identity on the near full length (ca. 1300 bp) 16S rRNA gene sequence level. These isolates can be largely divided into nine different groups, each supported by a bootstrap value >50%. Group 1 contains isolates only from M. laxissima, including our model bacterium Ruegeria sp. KLH1, while groups 5 and 6 contain isolates from both M. laxissima and I. strobilina. All the other groups (2, 3, 4, 7, 8 and 9) contain isolates only from I. strobilina. Furthermore, group 1 contains isolates from four different collection seasons, including those collected in 2004 by Mohamed et al. (2008c), and is dominated by those collected in June 2011. Groups 2, 3,7, 8 and 9 contain 202 isolates from only one collection season. Group 5 contains isolates from I. strobilina collected over three different seasons in this study while it is dominated by those from M. laxisssima collected in March 2010. In contrast, group 6 is dominated by isolates from I. strobilina collected in June 2011. 203 204 205 Figure 5.1. Phylogenetic tree using neighbor-joining method of 16S rRNA genes (ca. 1300 bp) from SR AHL-producing bacteria. Sequences obtained from M. laxissima (JZ08ML, JZ10ML, JZ11ML and colored in red) and I. strobilina (JZ08IS, JZ10IS, JZ11IS and colored in blue) are included. Sequences obtained from Mohamed et al. (2008c) are labeled with N04ML, N05ML or N05IS. KLH11 was isolated from M. laxissima in the same study. The accession numbers are listed in parenthesis. Bootstrap values > 50% are shown at nodes. The bar indicates the number of nucleotide substitutions per site. The other non-SR AHL producers can be categorized into two groups: Alphaproteobacteria and Gammaproteobacteria. The three alphaproteobacterial genera Paracoccus, Pseudovibrio, and Stappia also belong to the Roseobacter clade (Figure 5.2). Among the gammaproteobacterial AHL producers, 25 strains belong to the Vibrio genus. These Vibrios were mainly isolated from the sponges collected in 2008. Strains that belong to the genera of Alteromonas, Pseudomonas and Shewanella were also obtained although at low numbers. AHL producers from surrounding seawater belong to the genera Vibrio and Alteromonas and not to the SR group. 206 207 Figure 5.2. Phylogenetic tree using neighbor-joining method of 16S rRNA genes (ca. 700 bp) from all the non-SR AHL-producing bacteria. Sequences of bacteria isolated from M. laxissima (JZ08ML, JZ10ML, JZ11ML, in red), I. strobilina (JZ08IS, JZ10IS, JZ11IS, in blue) and seawater (JZ08SW) in this study are included. Sequences obtained from Mohamed et al. (2008c) are labeled with the genus name followed by N05IS or N05ML. The accession numbers are listed in parentheses. Bootstrap values > 50% are shown at nodes. The scale indicates the number of nucleotide substitutions per site. 208 5.4.2 luxS genes from Vibrios in sponges Degenerate primers VluxsF and VluxR based on Vibrio luxS sequences were able to amplify partial luxS gene sequences from all the Vibrio strains isolated from M. laxissima and I. strobilina collected in October 2008. A phylogenetic tree of the luxS gene sequences from these vibrios is shown in Fig. 5.3. The majority of luxS genes from these sponge-derived Vibrio isolates are closely related to that of V. harveyi, which is consistent with the close relationship between these isolates and V. harveyi shown by 16S rRNA phylogeny (Fig. 5.2). A novel cluster (named Cluster MI) of luxS gene sequences, comprising those from isolates JZ08ML43, JZ08ML44, JZ08ML48, JZ08ML49, JZ08ML63, JZ08IS73, shared only 88%-90% identity to their closest relative in GenBank (boxed in Fig. 5.3). All the isolates were evaluated for AI-2 activity using reporter strain V. harveyi TL- 26. All isolates were able to induce light production (Fig. 5.4), showing that all the Vibrio isolates, all of which have luxS genes, are able to synthesize AI-2-type molecules. 209 .Figure 5.3 Phylogenetic tree using neighbor-joining method based on the predicted 96 aa residues encoded by luxS genes from Vibrio isolates. Sequences isolated from M. laxissima (JZ08ML), I. strobilina (JZ08IS) and seawater (JZ08SW) in this study are in bold and the accession number of each of these luxS sequences is listed after the isolate name. The novel luxS cluster is named Cluster MI. Bootstrap values > 50% are shown at nodes. The scale indicates the number of amino acid substitutions per site. 210 Figure 5.4. AI-2 activities of all Vibrio isolated from sponges detected by reporter strain V. harveyi TL-26. V. mimicus ATCC 33653 was used as positive control. Bars are average of three biological replicates. Error bars represented the standard deviations. 5.4.3 Sponge tissues contain AHLs and detectable levels of ssaI transcripts. Sponge tissues were extracted with a modified Bligh-Dyer procedure (Bligh and Dyer, 1959) and these samples were fractionated by reverse phase TLC. TLC plates were overlaid with the highly sensitive AHL biological reporter A. tumefaciens KYC55 (Zhu et al., 2003) and this revealed the presence of AHL- type compounds in the extracts (Fig. 5.5A), one with migration similar to octanoyl-HSL (C8-HSL). In addition, two of three extracts also had an activity (somewhat obscured due to co-extracted pigments) that barely migrated from the 211 point of application (Fig. 5.5A, Lanes 4 and 5), similarly to a synthetic 3-oxo- C16:1 "11-HSL standard. M. laxissima tissues were used to extract RNA from which cDNA was synthesized, and RT-PCR revealed that ssaI was actively expressed in these tissues (Fig. 5.5B). No amplification was detected in the control RNA sample of M. laxissima without the RT reaction step, suggesting that the RNA sample is free of DNA contamination. The PCR amplicons were sequenced and of 13 clones sequenced all were greater than 98.5% identical to the KLH11 ssaI gene on the nucleotide level. 212 Figure 5.5. Detection of ssaI gene expression and AHLs in sponge tissue. A) Reverse phase C18 thin layer chromatography plates overlaid with A. tumefaciens AHL reporter for detecting AHLs from sponge tissues. Lane 1: 3- oxo-C16:1 "11-HSL, lane 2: C8-HSL, lanes 3-5: M. laxissima individuals 1-3. B) RT-PCR detection of expression of ssaI gene in sponge tissue. The PCR amplicon is about 400 bp. The first lane used cDNA as template, the second lane used RNA as template to test for DNA contamination (RNA control) and the last lane was a no template negative control. 213 5.4.4 Colonization of sponge cell aggregates The results of sponge cell aggregate colonization experiments (Fig. 5.6) show that the combination of antibiotic markers used in this experiment selected very well against bacteria present in sponges. However, there was no consistent trend of higher or lower rates of colonization of M. laxissima cell aggregates by the ssaI mutant at the time points we examined (0 h, 6 h, and 24 h). 214 Figure 5.6. Bacterial colonization of M. laxissima sponge cells at different timepoints after inoculation (0 h, 6 h, and 24 h). The first two columns are the results from the negative control, which only had sponge cell aggregates. The ?mutant? column is the result from the cell aggregate with ssaI mutant inoculation only and the ?wild type? column is with wt KLH11 only. The competition/wild type and competition/mutant represented the numbers of wt KLH11 and ssaI mutant from sponge cells aggregates inoculated with both types of strains at a 1:1 ratio. Rif=rifampicin, Tc=tetracycline, Km=kanamycin, Sp=spectinomycin. 215 5.4.5 Colonization of whole sponges We used the whole sponge individuals maintained in tanks with circulating water to study the bacterial colonization process. The antibiotic resistance markers that were proven to select very well against bacteria present in sponges in the sponge cell aggregate colonization experiments were used to label the wild type and ssaR- mutant. The results showed that the combination of Rif and Tc still worked well while the combination of Km+Sp did not select well against bacteria present in the sponges as several different bacterial morphotypes grew on the plates. However, we were able to count the ssaR- mutant as its colony displays a yellow-green color because of the dsRed fluorescence carried on the plasmid pJZ402. The results presented in Fig. 5.7 indicate that the ssaR- mutant colonized the sponges better than the KLH11 wild-type derivative and the presence of wild type strains reduced the colonization efficiency compared to the addition of the ssaR mutant only (P<0.05). 216 Figure 5.7. Bacterial colonization of M. laxissima whole sponges. The ?WT? column is the result from whole sponge with wt KLH11 only and the ssaR- column is with ssaR- mutant inoculation only. The competition/ WT and competition/ ssaR- represented the numbers of wt KLH11 and ssaR- mutants from whole sponges inoculated with both types of strains at a 1:1 ratio. 217 5.5. Discussion Marine sponges harbor highly diverse and dense microbial communities and provide an ideal environment for bacterial quorum sensing (Mohamed et al., 2008c; Zan et al., 2011b). Our finding that a high proportion (40%-50%) of the 420 strains isolated from M. laxissima and I. strobilina can produce AHL molecules provides strong evidence to support this. However, it is unclear what proportion of bacteria that associate with marine sponges can produce AHLs as the majority of bacteria in sponges are resistant to cultivation (Webster and Taylor, 2012). All the AHL producers identified in this study belong to the phylum Proteobacteria, in which more than 100 different species have been shown to be capable of AHL production (Algren et al., 2011). The result that there is a higher proportion of AHL producers among isolated strains from sponges than from the surrounding seawater is consistent with what was previously found in these two sponges with roughly 40 strains screened (Mohamed et al., 2008c) and therefore confirms and extends this prior study. More importantly, these AHL producers are dominated by bacterial strains that belong to the SR group, which is also consistent with the findings of Mohamed et al. (2008c). Wagner-D?bler et al. (2005) showed that the majority of AHL producers among the marine isolates collected from various marine habitats including seawater are dominated by Roseobacter although no strains belonging to the SR group were identified. Taylor et al. (2004) were also able to isolate one AHL+ SR strain closely related to Ruegeria atlantica (99% identity on the 16S rRNA gene level) from an Australian sponge Cymbastela concentrica. Taken together, this suggests that 218 Roseobacter is an important group of AHL producers in the marine environment generally and in association with marine sponges. The consistent isolation of members in group 1 from four different seasons and only from M. laxissima suggest that they are specific to this sponge and also true sponge symbionts based on the definition used in Taylor et al. (2007). Thus Ruegeria sp. KLH11 represents an excellent model for studying quorum sensing in sponge symbionts as it falls into this group. Likewise, some isolates could only be isolated from I. strobilina. Although all our SR AHL producers from the two sponges share >98% identity on the near full length 16S rRNA gene sequence level, it is clear that they are not identical SR bacteria. This is similar to findings of Montalvo and Hill (2011), in which sequences of bacteria isolated from each of two closely related but geographically distant sponges Xestosongia muta and X. testudinaria cluster tightly according to which sponges they are derived. Both M. laxissima and I. strobilina were collected from the same geographic locations but they are only distantly related sponges and presumably provide different physiological environments. Thus it is possible that the last common ancestor of these two different sponges hosted tightly associated SR bacteria and as the sponges evolved, the SR bacteria diverged in the different sponges. Another possibility is that these SR bacteria are horizontally acquired from a pool of SR bacteria present at very low concentrations in the surrounding seawater and the different environmental conditions and/or active selection by the sponges results in the sponge specific SR grouping that we observed. Our study provides evidence that there might be sponge-specific bacteria as previously proposed 219 (Henstchel et al., 2002; Taylor et al., 2007), for example, six groups identified in this study can only be isolated from sponges. Mohamed et al. (2008c) showed that these closely related SR AHL producers (> 99% identity on the 16S rRNA level) produced diverse AHL profiles. It is possible that the same trend remains for the SR bacteria obtained in this tudy. Generally, little is known about the function of sponge symbionts in their hosts (Webster and Blackall, 2009). However, some functions can be speculated. For instance, some rosoebacterial strains have been shown to produce antibiotics and show potential for secondary metabolite production (Buchan et al., 2005; Martens et al., 2007). Furthermore, particle-associated Roseobacters were 10 times more likely to produce antibacterial compounds than their planktonic relatives (Long and Azam, 2001; Slightom and Buchan, 2009). It is possible that SR bacteria obtained in this study produce secondary metabolites and thus provide the host sponge with defensive tools against pathogenic bacteria and help to structure the bacterial community associated with the sponges. It would be interesting to explore this possibility and also to test whether QS mediated pathways play a role in controlling secondary metabolite production, including antibiotic production. Furthermore, these SR bacteria might play a role in sulfur metabolism, such as in the degradation of dimethylsulfoniopropionate (DMSP). Our model bacterium Ruegeria sp. KLH11 encodes one key enzyme involved in the enzymatic cleavage of DMSP (Curson et al., 2011; Zan et al., 2011a)(See Appendix 1). 220 We also observed some slight differences compared to the study of Mohamed et al. (2008c). For example, we did not obtain any AHL producers that belong to the genera Thalassomonas and Spongiobacter. Furthermore, the AHL producers isolated in this study that belong to the genera: Alteromonas, Pseudomonas, Paracocuss and Stappia were not reported in Mohamed et al. (2008c). This might reflect the variability of the bacteria associated with sponges that can be cultured or subtle shifts in the bacterial communities associated with the sponges. AHL is considered to be an intraspecies signaling molecule while AI-2 is an interspecies signaling molecule (Ng and Bassler, 2009). We took advantage of the fact that a relatively large amount of Vibrio strains were obtained from the collection of 2008 and designed degenerate primers to amplify the luxS gene, the synthase of AI-2, from these Vibrio strains. Not surprisingly, all the Vibrios contain luxS genes and were also able to trigger the AI-2 biological reporter, suggesting that AI-2 molecules are synthesized in all these strain. Our result also revealed a novel group of luxS sequence, thus extending the diversity of available luxS genes. However, it is unclear what the ecological significance of AI-2 mediating pathways in these Vibrios and this remains an interesting area for further investigation. In this study we not only showed the isolation of AHL-producing bacteria but also demonstrated the production of AHLs from tissues of the soft-bodied, shallow water sponge M. laxissima, similar to those produced by abundant cultivatable, AHL+ sponge-associated bacteria (Mohamed et al., 2008c). The 221 chemical identities of the molecules present in the M. laxissima need to be further confirmed. Likewise we also show expression of the ssaI AHL synthase gene directly in total RNA from the sponge, effectively connecting our laboratory findings on QS in these symbiotic bacteria with their native host environment. Similarly, Gard?res et al. (2012) identified the main AHL molecules present in the tissues of Suberites domuncula as 3-oxo-C12-HSL, the same compound that is produced by the cultivated isolates from this sponge. Taylor et al. (2004) showed that organic extracts of 24 out of 31 marine sponge species could activate the AHL reporter Chromobacterium violaceum (CV026), suggesting the presence of AHL molecules in the tissues. Taken together, these findings suggest that AHL- mediated signaling processes are important within sponge hosts. We invested efforts for the first time to study the role that QS plays in the colonization process. Ruegeria sp. KLH11 was used as a model as the ssaI gene in this bacterium is actively expressed in situ within sponge tissue. Wild type KLH11 strain and QS mutant strain were labeled with different antibiotic resistance markers to colonize the sponge cell aggregates. However, we did not observe a consistent trend in the role that QS plays in the colonization process in sponge aggregates. The approach we used has proven to be effective in several other systems (Visick et al., 2000; Gurich and Gonz?lez, 2009). We reasoned that several factors can contribute to the inconsistent results that we obtained: 1) there might be variation in the three sponge individuals, such as the age, weight and health status. 2) Sponge cell aggregates might not be a suitable experimental system in which to study the interaction between host and microbes 222 occurring naturally in intact and healthy sponges. 3) The ssaR receptor mutant would be a better strain to use since the sponge tissue contains bacteria that produce similar molecules to those encoded by ssaI, which could compensate for loss of production of AHLs in an ssaI mutant. 4) It is possible that within the time frame (24 h) we tested, a stable relationship is not yet established. We refined the design of colonization assay based on the lessons learned from the sponge cell aggregate colonization experiment. The wild type KLH11 and ssaR mutant strains were used to colonize whole sponges under laboratory conditions. We observed an interesting pattern in which the ssaR mutant strain colonized the sponges better than the labeled wild type KLH11 derivative and the presence of the wild type KLH11 strain seemed to reduce the colonization efficiency. KLH11 is motile but lacks chemotaxis machinery (Zan et al., 2011a) (See Appendix 1) and thus motility might function as a mechanism of random dispersion as previously proposed (Badger et al., 2006). This might explain why we failed to detect the wild type strain in sponges since the sponges used are relatively small and have a small surface area. The SsaR mutant cannot swim and also forms aggregates in culture and these characteristics might have facilitated attachment to the sponges. This does not explain why the colonization efficiency of ssaR mutant decreased when wild type strain was present. Another possibility is that QS affects the survival rate as reported in several other species (Joelsson et al., 2007; Goo et al., 2012) but this still does not explain why in the competition assay, the ssaR mutant colonized less efficiently in the presence of the wildtype. 223 Thus far, no clear conclusion can be made about how QS affects colonization process. To better understand this, a few additional aspects need to be explored: 1) How does QS affect the survival of Ruegeria sp. KLH11 in seawater? 2) How is Ruegeria sp. KLH11 transmitted? 3) Do adult sponges with established microbial communities actively acquire symbionts from surrounding environments besides through the process of filtering water for food? 4) It would be ideal to conduct colonization experiments in the field, such as enclosing the sponges with bags, rather than on sponges held in captivity. 224 Chapter 6. Conclusions and future directions 225 My research focused on the bacterial signaling in the two marine sponges, Mycale laxissima and Ircinia strobilina collected off Key Largo, Florida. I began my research project by isolating and screening approximately 420 bacterial strains from these two marine sponges (M. laxissima and I. strobilina) for acylhomoserine lactone (AHL) production. Previous studies in our laboratory and in collaboration with Dr. Clay Fuqua?s group from Indiana University had shown that bacteria belonging to the Silicibacter-Ruegeria (SR) subclade of the ecologically abundant and important Roseobacter clade are the main AHL producers in these two sponges (Mohamed et al., 2008c), although a relatively small number of isolates were screened. My work significantly expanded the number of isolates screened and showed that roughly 40% of the cultured isolates from these two sponges are able to produce AHL molecules and isolates from the SR group are the dominant and consistent AHL-producers, which strongly supports the findings by Mohamed et al. (2008c). Furthermore, the repeated isolation of the same AHL-producing SR bacteria regardless of the collection times and the failure to isolate them from the surrounding seawater suggests that these SR bacteria may be true sponge symbionts. We follow the definition of symbiosis as used by Taylor et al. (2007) and in the original sense of deBary (1879), in which symbiosis refers simply to two or more different organisms that live together over a long period of time without any judgment made regarding benefits to either partner. It is unclear what functions these SR bacteria can play to benefit their sponge hosts. However, it can be speculated that SR bacteria may participate in the defense mechanisms of the sponges since 226 many roseobacters have been shown to produce antibiotics. For instance, some rosoebacterial strains have been shown to produce antibiotics and show potential for secondary metabolite production (Buchan et al., 2005; Martens et al., 2007). Furthermore, particle-associated Roseobacters were found to be 10 times more likely to produce antibacterial compounds than their planktonic relatives (Long and Azam, 2001; Slightom and Buchan, 2009). It is possible that SR bacteria obtained in this study produce secondary metabolites and thus provide the host sponge defensive tools against pathogenic bacteria and help to structure the bacterial community associated with the sponges. It would be interesting to explore this possibility and also to test whether QS mediated pathways play a role in controlling secondary metabolite production, including antibiotic production. We recently sequenced 17 SR genomes, which provide a great opportunity to explore whether they have the potential to produce secondary metabolites and this analysis is in progress. Another interesting question to study is how broadly these AHL-producers are distributed. Many studies have supported the idea that sponges seem to have uniform sponge-specific microbial communities (Hentschel et al., 2002; Taylor et al., 2007). Do sponges harbor highly diverse AHL producers or are these SR bacteria the dominant AHL producers in sponges? Generally, sponges can be categorized as high microbial abundance or low microbial abundance sponges (Taylor et al., 2007). Is it possible that only these high microbial abundance sponges harbor AHL producers since quorum sensing is a cell-density related process? Bacteria isolated from different marine sponges, collected from different 227 geographic locations and belonging to different species, need to be screened for AHL production to answer these questions. What is the ecological significance of these QS pathways in the SR bacteria in the context of their native sponge host environment? This is the fundamental question that drove my dissertation research. In collaboration with Dr. Clay Fuqua from Indiana University and Dr. Mair Churchill from the University of Colorado Denver, we chose Ruegeria sp. KLH11 that was originally isolated from the sponge M. laxissima (Mohamed et al., 2008c) as our model. Genetic screening combined with genome sequencing identified two sets of luxR-luxI type QS pathways: ssaRI and ssbRI, and one luxI solo sscI. The results of mass spectrometry conducted in the laboratory of our collaborator Dr. Mair Churchill showed that all the three LuxI enzymes are able to synthesize long-chain AHLs, with SsbI and SscI showing similar profiles of AHLs and much stronger activity than SsaI. Furthermore, we now know the complex interconnected arrangement among all the three systems through a series of genetic studies (Chapters 2 and 3). However, it is still unclear what the exact mechanisms are for ssaI to indirectly affect ssbI and likewise how ssbI affects ssaI. It is possible that the proteins encoded by these genes compete for the substrate pool and thereby creates the regulatory network affecting each other?s activity. Furthermore experimental evidence is required to investigate this possibility, such as the expression of all three genes in a heterologous system and examination of the profile of the AHLs synthesized to provide support for the hypothesis that they compete for the substrate pool. Nevertheless, the architecture of the QS system in KLH11 is 228 indeed consistent with many other well-studied bacterial species, in which multiple QS pathways exist (Fuqua and Greenberg, 2002). Phenotypic studies showed that SsaRI activates flagellar motility and inhibits biofilm formation but SsbRI does not affect flagellar motility or biofilm formation. Furthermore, the SscI has only a very minor effect on motility. It is perplexing that KLH11 produces mainly the AHLs synthesized by SsbI and SscI in culture but these genes do not seem to play a significant role in regulating the two important phenotypes of motility and biofilm formation whereas SsaI that is key in these phenotypes produces much lower amounts of AHLs. Metatranscriptomic or microarray analyses where the gene expression profile of wild type KLH11 is compared with that of ssbR, ssbI and sscI mutants would be likely to provide clear insights into the genes or pathways that are affected by the ssb system and sscI. This would also provide clues about what is controlled by the systems similar to the ssb system in other Roseobacters. It has been shown by genomic analysis that the ssb-like system is more prevalent than the ssa system in Roseobacters (Cicirelli et al., 2008). Such future studies may provide new insights into why a single bacterium needs to encode multiple quorum sensing systems. Motility has been suggested to play a critical role in mediating interaction between roseobacters and the environment (Slightom and Buchan, 2009). We show that KLH11 exhibits flagellar motility and also established methods to study the interaction between sponges and KLH11. We do not yet know how general this system is among all the AHL-producing SR strains that we isolated. Do all of them exhibit motility? Does quorum sensing control the motility in all of these SR 229 bacteria? What is the distribution of the ssa and ssb systems and sscI that we have identified in Ruegeria sp. KLH11? The work accomplished in this study has established the basis to test all of these questions in the SR strains that we obtained in this study. The transcription data show that ssaI is expressed at a very high level but SsaI seems to have quite weak catalytic activity for AHL synthesis (Chapter 2). It is possible that SsaI is only translated at a very low level and the quantification at the protein level is required to test this hypothesis. It is also possible that the SsaI protein is unstable and subject to rapid proteolysis. A third possibility is that the protein has intrinsically low activity, which might be caused by the extra long C- terminus. It would be useful to perform a series of mutagenesis studies on the protein and then test its activity in synthesis of AHLs. Furthermore, we showed that SsaR complexed with AHLs could stimulate the expression of ssaI in a heterologous system. However, it is unclear if this is also true in the native KLH11 background. Two approaches could be used to test this: 1) since a ssaR deletion strain has now been constructed, the expression of PssaI-lacZ could be compared in the wild-type strain and in the "ssaR strain. The expression could also be examined in the absence and in the presence of cognate AHL molecules to test whether SsaR in the native KLH11 background requires AHL to stimulate ssaI expression. 2) Campbell insertion could also be used. One additional strain ("ssaR ssaI-lacZ) would need to be constructed. The ssaI-lacZ expression could then be compared to "ssaR ssaI-lacZ expression. The advantage of this approach over the first approach is that the ssaI promoter is in its native location. 230 Our data also show that SsaR very likely binds to the ssaI promoter region and the ssa box we identified is critical for this potential binding. However, experiments such as gel mobility shift assays are required to provide direct evidence to confirm this. Also, a more detailed understanding of the ssa box, such as what defines the minimal nucleotide set for SsaR binding, is needed. For instance, six critical nucleotides in the lux box define the LuxR-binding site (Egland and Greenberg, 2009). Understanding of the ssa box would be useful to predict what genes or pathways can be affected by the SsaR and also provide a baseline for defining this type of box in other closely related roseobacters. Our results show that the expression of the ssaI gene is greatly induced by the long chain AHL, 3-oxo-C16:1 "11-HSL. Currently, several biological reporters have been developed to detect the long chain AHLs (C12-C18), such as Sinorhizobium meliloti 1021 that can respond to AHLs with chain lengths of C16- C20 (Llamas et al., 2004). However, it elicits a weak response (Mohamed et al., 2008c). Thus it would be useful to develop a sensitive KLH11-derived reporter strain based on the fact that ssaI-lacZ fusion shows a very sensitive response to long chain AHLs, presumably via SsaR. The coordination of biofilm formation and motility control by the SsaRI system led us to hypothesize that quorum sensing is a key factor in the successful symbiosis between sponges and their SR alphaproteobacterial symbionts. One possibility is that quorum sensing is a determining factor whereby sponges select their symbionts from the huge bacterial load to which sponges are exposed through filtration of large volumes of water. We were able to show the presence 231 of AHLs and the expression of ssaI transcripts in the sponge tissues, which effectively connects what we found under laboratory conditions with its native host and thus makes KLH11 an excellent model system for understanding the complex symbiotic interactions with its host. The pattern observed from the whole sponge colonization experiment was very interesting but also very confusing. We do not yet have a good understanding of the role of quorum sensing in colonization of sponges by bacteria. Further efforts need to be directed towards the following to better study this process: 1) Ideally, experiments investigating the role of QS in colonization should be done under in situ conditions where the sponge individuals are covered with some kind of enclosure and wild type KLH11 and quorum sensing mutants labeled with the appropriate antibiotic resistance markers are released in these closed environments. The outcome of colonization of these strains could be determined at different timepoints. 2) Fluorescently tagged KLH11 and derivatives can be used in sponge colonization experiments. Advanced microscopy could be employed to monitor how these strains interact with the sponge tissues. 3) Several basic questions as listed in Chapter 5 need to be answered: i) How does QS affect the survival of bacteria in the seawater? ii) How is KLH11 transmitted? iii) Do adult sponges with established microbial community actively acquire symbionts from surrounding environments besides filtering for food? Another interesting aspect is to understand whether sponges can respond to AHL molecules i.e. whether inter-domain signaling occurs between the SR bacteria and sponge cells. Zoospores of the macroalga Ulva can ?probe? the 232 nature of surfaces for potential attachment by sensing the presence of AHLs in the biolfilm on the surface (Joint et al., 2002; Tait et al., 2009). It is possible that sponge cells can respond to AHLs to monitor the presence of bacteria. Proteomic approaches could be utilized to provide useful insights about the potential inter- domain signaling. The observation that QS inhibits biofilm formation indeed resembles the situation found in V. cholerae. At low cell density, V. cholerae forms biofilms while at high cell density, QS inhibits the biofilm formation, which can promote the dispersion of bacterial cells away from the biofilm and release into the environment (Hammer and Bassler, 2003; Zhu and Mekalanos, 2003; Higgins et al., 2007). It is possible that the same dispersal mechanism is at play in KLH11 to promote a uniform distribution of this symbiont in the sponge tissues or even to facilitate release back into a planktonic form in the surrounding seawater for colonizing other sponges. When KLH11 grows to high cell density, it would be beneficial for the motile cells to swim rather than attach to surfaces and in this situation, biofilm formation is inhibited. It is however unclear how this coordination works. One possibility is that QS regulates the concentration of a second messenger molecule called cyclic di-GMP (c-di-GMP). C-di-GMP is synthesized by diguanylate cylase enzymes that contain a GGDEF domain and degraded by phosphodiesterase enzymes that contain either an EAL or HD-GYP domain (Waters et al., 2008). It has been shown in several bacterial species that high concentrations of c-di-GMP promote biofilm formation and repress motility, facilitating transition from motile phase to sessile phase (Massie et al., 2012; 233 Purcell et al., 2012). In V. cholerae, QS has been shown to affect the expression of the genes that encode these related enzymes with the overall effect of reducing the concentration of c-di-GMP (Waters et al., 2008). It is worthwhile examining the relationship between QS and c-di-GMP in KLH11 to have a better understanding of the mechanisms to regulate the transition between sessile and motile phases. My research on the cckA-chpT-ctrA phosphorelay system makes Ruegeria sp. KLH11 an excellent model for understanding this pathway outside of the group in which this pathway is essential (Greene et al., 2012). The findings of the regulatory link between a QS pathway and a two-component system is not surprising per se since previous studies have shown that the GacS/GacA two- component system positively controls the quorum sensing pathways in Pseudomonas aeruginosa (Kay et al., 2006) and the quorum sensing SmaR protein in Serratia sp. ATCC 39006 can repress the response regulator PigQ (Williamson et al., 2008). The novelty is that we identified the SsaRI system as the regulator for the cckA-chpT-ctrA system outside of the model bacterium Caulobacter crescentus, in which this system is essential. The ctrA gene has been extensively studied in C. crescentus in the control of the cell cycle, polar development, and flagella biogenesis. Furthermore, in several species of the Alphaproteobacteria, the involvement of this phosphorelay system in flagellar motility has been well characterized, especially for the ctrA gene (Mercer et al., 2010; Greene et al., 2012). However, little is known about the regulation of the ctrA gene in these bacterial species. My study for the very first time identified QS 234 as the regulator for CtrA in Ruegeria sp. KLH11 and opens a new avenue for the regulation of the ctrA gene in the bacteria in which this gene is not essential. Issues remaining to be addressed include: 1) what is the nature of the phosphorelay between the cckA-chpT-ctrA ? 2) what component(s) exists in between the ssaRI and the ctrA genes to explain my results that show indirect transcriptional regulation of SsaRI on the ctrA gene? I optimized the protocol of using the mariner transposon pFD1 (Rubin et al., 1999) to create a large random pool of mutants in KLH11. The "ssaI mutant that has the PctrA-lacZ fusion shows very light blue color on the plates with X-Gal since the basal expression level of this fusion is very low. Therefore it would be possible to identify mutants that show very strong blue color on the plates. The mutants that might be identified by this approach could then be characterized and the target components might be revealed. 3) The difficulty in restoring the expression of the cckA-chpT-ctrA genes in the !ssaI mutant by exogenous addition of AHL seems to be related to the poor entry of the long-chain AHL into the cell. However, further evidence is required to support this explanation. One way to test this is to label the AHL molecules with fluorescence markers and then monitor the movement of these molecules and also localize their location. It would be technically challenging to label the small AHL molecule but this approach would provide very strong evidence. 4) Does quorum-sensing coordinate the expression and phosphorylation of the histidine kinase CckA? We showed that SsaRI transcriptionally regulates the expression of cckA while little is known about the phosphorylation of CckA. One way to test this is to quantify the level of 235 phosphorylated CckA in the wild type strain and also in the quorum sensing mutant strain at different stages of growth. If CckA can only be phosphorylated in late stage growth of the wild type KLH11, this would suggest that the signal(s) that CckA can recognize and respond to are associated with cell-density. The paradigm of the structure of AHLs as a fatty acyl chain attached to a homoserine lactone via an amide bond changed when the p-coumaroyl-HSL was discovered, in which the fatty acid chain has a ring structure. More interestingly, this molecule is synthesized by a LuxI homologue RpaI, which uses the nutrient coumaric acid as a precursor (Schaefer et al., 2008). It seems that KLH11 can also synthesize similar molecules detected by a biological reporter and this production is independent of the three LuxI homologues although the identity of the molecule is unknown (Chapter 3). It is reasonable to speculate that a novel type of enzyme exists in KLH11, although it is possible that an unknown luxI homologue that has eluded genome sequencing is responsible for the synthesis of p-coumaroyl-HSL. Genetic screening of the genomic library that was used to identify ssaI and ssbI genes could be applied to search for this potential enzyme. Identification of this enzyme in Ruegeria sp. KLH11 has broad significance as several bacterial species have been shown to respond to the addition of p- coumarate, the substrate for the synthesis of pC-HSL. For example, R. pomeroyi DSS-3, a close relative of KLH11, can produce a molecule biologically and chemically similar to pC-HSL (Schaefer et al., 2008). In Silicibacter sp. TM1040, a strain that does not encode LuxI or LuxM homologues, the compound named RMI can be induced by addition of p-coumarate (Sule and Belas, 2012). 236 Phaeobacter gallaeciensis BS107 (also known as DSM 17395), a member of the Roseobacter clade, responds to the presence of p-coumarate produced by the microalga Emiliania huxleyi to control the production of algaecides and thus convert itself into an opportunistic pathogen of the host microalga (Seyedsayamdost et al., 2011). Taken together, dissection of this pathway in these roseobacters represents an exciting area for further exploration. Marine sponges harbor diverse and highly abundant bacteria in their tissues. Do these bacteria living in the same environment interact with each other? Do they coordinate as a group or compete with each other for nutrients and resources? For example, it has been shown that antagonism among coral bacterial symbionts is common and can potentially structure the microbial communities associated with corals and might also be a protective mechanism against pathogens (Rypien et al., 2010). Quorum sensing might provide an entry point to study the interactions. Coordination of group behavior via chemical languages is advantageous. Thus the ability to disrupt this conversation, which is generally termed as quorum quenching (Dong and Zhang, 2005), might be an effective approach for other bacteria that compete with some of the AHL- producing sponge symbionts. Marine sponges are well known for being rich resources of bioactive natural compounds (Hill, 2004). Do they also produce quorum-quenching compounds? Our model organism KLH11 provides an excellent model for exploring these questions. We can use the motility phenotype as the readout to screen for compounds that can inhibit motility in the wild type, potentially through quorum quenching. We can also use the ssaI nonmotile 237 mutant to screen for interspecies signaling. These aspects will provide further insights into the interactions between bacteria living in the same host. Overall, my dissertation work combined with our collaborators? work established that KLH11 is an excellent model organism to study the complex quorum sensing regulatory circuits, motility control and biofilm formation in sponge symbionts and also other related roseobacterial relatives. Furthermore, approaches and tools have been established to further examine the interactions between the microbes and hosts. Our findings thus lay the groundwork for studies of this process and provide an important foundation for elucidating these pervasive and influential symbiotic relationships. 238 Appendix 1: Genome sequence of Ruegeria sp. Strain KLH11, an N- acylhomoserine lactone-producing bacterium isolated from the marine sponge Mycale laxissima. 239 Genome Sequence of Ruegeria sp. Strain KLH11, an N-Acylhomoserine Lactone-Producing Bacterium Isolated from the Marine Sponge Mycale laxissima! Jindong Zan,1 W. Florian Fricke,2 Clay Fuqua,3 Jacques Ravel,2 and Russell T. Hill1 Institute of Marine and Environmental Technology, University of Maryland Center for Environmental Science, Baltimore, Maryland 212021; Institute for Genome Sciences, Department of Microbiology & Immunology, University of Maryland School of Medicine, Baltimore, Maryland 212012; and Department of Biology, Indiana University, Bloomington, Indiana 474053 Ruegeria sp. strain KLH11, isolated from the marine sponge Mycale laxissima, produces a complex profile of N-acylhomoserine lactone quorum-sensing (QS) molecules. The genome sequence provides insights into the genetic potential of KLH11 to maintain complex QS systems, and this is the first genome report of a cultivated symbiont from a marine sponge. Sponges can harbor diverse assemblages of microbes, which in some cases can comprise up to 60% of the total biomass of the holobiont (6). This dense microbial community associated with sponges provides conditions in which quorum sensing (QS) may be important (3, 8). QS is a process by which bacteria coordinate group activities such as bioluminescence, antibiotic 240 production, and virulence at high cell density via signal molecules, typified by N- acylhomoserine lactone (AHL) in Proteobacteria (2). We isolated a Ruegeria sp. strain, KLH11 (99% 16S rRNA gene se- quence similarity with Ruegeria lacuscaerulensis [GenBank accession no. HQ908678]), which can produce a complex profile of long-chain AHLs (3). The Silicibacter-Ruegeria subgroup belongs to the Roseobacter clade, which is widespread and numerous in surface waters of the oceans and includes organisms that are involved in global sulfur cycling (1). Many roseobacters have been fully sequenced and annotated (5). Ruegeria sp. strain KLH11 is the first readily culturable sponge symbiont to have its genome sequenced. This microorganism has been isolated from several samples of the sponge Mycale laxissima and is termed a symbiont using the term as defined by Taylor et al. (6). The genome was sequenced by a combination of Roche/454 and Sanger whole-genome shotgun sequencing. Sequences were assembled with a Celera assembler (4), and genes were predicted and annotated using the automated CloVR-Microbe pipeline (7). The draft genome has a total assembly size of ca. 4.5 Mb, with an average G+C content of 57.3 mol%, and is predicted to have 4,493 coding sequences (CDS), of which 3,243 (ca. 72.0%) were functionally annotated. This assembly is comprised of 13 scaffolds, of which the largest is ca. 3.1 Mb, having 3,193 CDS. One scaffold of 81 kbp shows similarity to the megaplasmid identified in Ruegeria pomeroyi DSS-3 (accession number CP000032) and could indicate the presence of extrachromosomal elements in the genome. Approximately 82.0% of all the 241 predicted CDS have ATG sequences at the presumptive start position, and 17.8% may use GTG or TTG. One complete rRNA operon, one partial 16S rRNA gene, two 5S rRNA genes, and 52 tRNAs were identified. About 34 genes were found to be related to the synthesis of bacterial flagella and regulation of bacterial motility. No candidates for chemotaxis functions were observed in the genome. Two sets of quorum-sensing circuits were identified, implying the potential for a complex quorum-sensing mechanism. Roughly 44 genes were predicted to function in the biosynthesis and degradation of polysaccharides. Comparison of KLH11 genome sequences to closely related free-living bacteria, such as R. pomeroyi DSS-3, provides an opportunity to define the unique set of genes important for host-symbiont interactions and the ecological roles and bio- geochemical functions. Complex quorum-sensing circuits in KLH11 make it a perfect model for studying the ecological role of QS in the complex symbiosis between sponges and their associated bacteria. Nucleotide sequence accession number. The data of the genome sequence of KLH11 were deposited in GenBank under the accession number ACCW00000000. Sequencing was funded by the Gordon and Betty Moore Foundation and the Institute for Genome Sciences at the University of Maryland School of Medicine. This work was supported by the NSF Microbial Observatories Program (MCB-0238515), the Microbial Interactions and Processes Program (MCB- 0703467), and the BIO/IOS Program (IOS-0919728). The UNC National Underwater Research Center provided sponge sampling support. 242 This is contribution no. 4541 from the University of Maryland Center for Environmental Science. REFERENCES 1. Buchan, A., J. M. Gonz?lez, and M. A. Moran. 2005. Overview of the marine Roseobacter lineage. Appl. Environ. Microbiol. 71:5665?5677. 2. Fuqua, C., and E. P. Greenberg. 2002. Listening in on bacteria: acyl-homo- serine lactone signalling. Nat. Rev. Mol. Cell Biol. 3:685?695. 3. Mohamed, N. M., et al. 2008. Diversity and quorum-sensing signal production of Proteobacteria associated with marine sponges. Environ. Microbiol. 10:75?86. 4. Myers, E. W., et al. 2000. A whole-genome assembly of Drosophila. Science1287:2196?2204. 5. Newton, R. J., et al. 2010. Genome characteristics of a generalist marine bacterial lineage. ISME J. 4:784?798. 6. Taylor, M. W., R. Radax, D. Steger, and M. Wagner. 2007. Sponge- associated microorganisms: evolution, ecology, and biotechnological potential. Microbiol. Mol. Biol. Rev. 71:295?347. 7. White, J., et al. 2011. CloVR-Microbe: assembly, gene finding and functional annotation of raw sequence data from single microbial genome projects? standard operating procedure, version 1.0. Nature Proc. doi:10.1038/npre. 2011.5887.1. 8. Zan, J., C. Fuqua, and R. T. Hill. 2011. Diversity and functional analysis of luxS genes in vibrios from marine sponges Mycale laxissima and Ircinia strobilina. ISME J. 5: 1505-1516. 243 Literature cited Ahlgren, N.A., Harwood, C.S., Schaefer, A.L., Giraud, E., and Greenberg, E.P. (2011) Aryl-homoserine lactone quorum sensing in stem-nodulating photosynthetic bradyrhizobia. Proc Natl Acad Sci USA 108: 7183-7188. Akakura, R., and Winans, S.C. (2002) Mutations in the occQ operator that decrease OccR-induced DNA bending do not cause constitutive promoter activity. J Biol Chem 277: 15773-15780. Antunes, L.C.M., Ferreira, R.B.R., Lostroh, C.P., and Greenberg, E.P. (2008) A mutational analysis defines Vibrio fischeri LuxR binding sites. J Bacteriol 190: 4392-4397. Arora, S.K., Ritchings, B.W., Almira, E.C., Lory, S., and Ramphal, R. (1997) A transcriptional activator, FleQ, regulates mucin adhesion and flagellar gene expression in Pseudomonas aeruginosa in a cascade manner. J Bacteriol 179: 5574-5581. Atkinson, S., Chang, C.Y., Patrick, H.L., Buckley, C.M.F., Wang, Y., Sockett, R.E. et al. (2008) Functional interplay between the Yersinia pseudotuberculosis YpsRI and YtbRI quorum sensing systems modulates swimming motility by controlling expression of flhDC and fliA. Mol Microbiol 69: 137-151. Austin, B., and Zhang, X.H. (2006) Vibrio harveyi: a significant pathogen of marine vertebrates and invertebrates. Lett Appl Microbiol 43: 119-124. 244 Badger, J.H., Hoover, T.R., Brun, Y.V., Weiner, R.M., Laub, M.T., Alexandre, G. et al. (2006) Comparative genomic evidence for a close relationship between the dimorphic prosthecate bacteria Hyphomonas neptunium and Caulobacter crescentus. J Bacteriol 188: 6841-6850. Barnett, M.J., Hung, D.Y., Reisenauer, A., Shapiro, L., and Long, S.R. (2001) A homolog of the CtrA cell cycle regulator is present and essential in Sinorhizobium meliloti. J Bacteriol 183: 3204-3210. Bassler, B.L., Wright, M., Showalter, R.E., and Silverman, M.R. (1993) Intercellular signalling in Vibrio harveyi: sequence and function of genes regulating expression of luminescence. Mol Microbiol 9: 773-786. Bassler, B.L., Wright, M., and Silverman, M.R. (1994) Multiple signalling systems controlling expression of luminescence in Vibrio harveyi: sequence and function of genes encoding a second sensory pathway. Mol Microbiol 13: 273-286. Belas, R., Horikawa, E., Aizawa, S.-I., and Suvanasuthi, R. (2009) Genetic determinants of Silicibacter sp. TM1040 motility. J Bacteriol 191: 4502-4512. Berger, M., Neumann, A., Schulz, S., Simon, M., and Brinkhoff, T. (2011) Tropodithietic acid production in Phaeobacter gallaeciensis is regulated by N-acyl homoserine lactone-mediated quorum sensing. J Bacteriol 193: 6576-6585. Bergquist, P.R. (1978) Sponges. Lodon: Hutchinson CO. 245 Biondi, E.G., Reisinger, S.J., Skerker, J.M., Arif, M., Perchuk, B.S., Ryan, K.R., and Laub, M.T. (2006) Regulation of the bacterial cell cycle by an integrated genetic circuit. Nature 444: 899-904. Bird, T.H., and MacKrell, A. (2011) A CtrA homolog affects swarming motility and encystment in Rhodospirillum centenum. Arch Microbiol 193: 451-459. Bligh, E.G., and Dyer, W.J. (1959) A rapid method of total lipid extraction and purification. Can J Biochem Physiol 37: 911-917. Brilli, M., Fondi, M., Fani, R., Mengoni, A., Ferri, L., Bazzicalupo, M., and Biondi, E.G. (2010) The diversity and evolution of cell cycle regulation in alpha- proteobacteria: a comparative genomic analysis. BMC Syst Biol 4: 52-67. Buchan, A., Gonz?lez, J.M., and Moran, M.A. (2005) Overview of the marine Roseobacter lineage. Appl Environ Microbiol 71: 5665-5677. Capra, E.J., and Laub, M.T. (2012) Evolution of two-component signal transduction systems. Annu Rev Microbiol 66: 325-347. Carroll, R.K., Liao, X., Morgan, L.K., Cicirelli, E.M., Li, Y., Sheng, W. et al. (2009) Structural and functional analysis of the C-terminal DNA binding domain of the Salmonella typhimurium SPI-2 response regulator SsrB. J Biol Chem 284: 12008-12019. 246 Cheng, Z., Miura, K., Popov, V.L., Kumagai, Y., and Rikihisa, Y. (2011) Insights into the CtrA regulon in development of stress resistance in obligatory intracellular pathogen Ehrlichia chaffeensis. Mol Microbiol 82: 1217-1234. Choi, S.H., and Greenberg, E.P. (1992) Genetic dissection of DNA binding and luminescence gene activation by the Vibrio fischeri LuxR protein. J Bacteriol 174: 4064-4069. Churchill, M.E.A., and Chen, L. (2011) Structural basis of acyl-homoserine lactone-dependent signaling. Chem Rev 111: 68-85. Cicirelli, E.M., Williamson, H., Tait, K., and Fuqua, C. (2008) Acylated homoserine lactone signaling in marine bacterial systems. In Chemical communication among bacteria. Winans, S.C., and Bassler, B.L. (eds). Washington, DC: ASM press, pp. 251-272. Clay, K.L., and Murphy, R.C. (1979) New procedure for isolation of amino acids based on selective hydrolysis of trimethylsilyl derivatives. J Chromatogr 164: 417-426. Curson, A.R., Todd, J.D., Sullivan, M.J., and Johnston, A.W. (2011) Catabolism of dimethylsulphoniopropionate: microorganisms, enzymes and genes. Nat Rev Microbiol 9: 849-859. Daniels, R., Vanderleyden, J., and Michiels, J. (2004) Quorum sensing and swarming migration in bacteria. FEMS Microbiol Rev 28: 261-289. 247 Davey, M.E., Caiazza, N.C., and O?Toole, G.A. (2003) Rhamnolipid surfactant production affects biofilm architecture in Pseudomonas aeruginosa PA01. J Bacteriol 185: 1027-1036. Davies, D.G., Parsek, M.R., Pearson, J.P., Iglewski, B.H., Costerton, J.W., and Greenberg, E.P. (1998) The involvement of cell-to-cell signals in the development of a bacterial biofilm. Science 280: 295-298. de Caralt, S., Uriz, M.J., Ereskovsky, A.V., and Wijffels, R.H. (2007) Embryo development of Corticium candelabrum (Demospongiae: Homosclerophorida). Invertebr Biol 126: 211-219. Devine, J.H., Shadel, G.S., and Baldwin, T.O. (1989) Identification of the operator of the lux regulon from the Vibrio fischeri strain ATCC7744. Proc Natl Acad Sci USA 86: 5688-5692. Dekimpe, V., and D?ziel, E. (2009) Revisiting the quorum-sensing hierarchy in Pseudomonas aeruginosa: the transcriptional regulator RhlR regulates LasR- specific factors. Microbiology 155: 712-723. Domian, I.J., Reisenauer, A., and Shapiro, L. (1999) Feedback control of a master bacterial cell-cycle regulator. Proc Natl Acad Sci USA 96: 6648-6653. Dong, Y.H., and Zhang, L.H. (2005) Quorum sensing and quorum-quenching enzymes. J Microbiol 43: 101-109. 248 Eberhard, A., Burlingame, A.L., Eberhard, C., Kenyon, G.L., Nealson, K.H., and Oppenheimer, N.J. (1981) Structural identification of autoinducer of Photobacterium fischeri luciferase. Biochemistry 20: 2444-2449. Eberl, L., Winson, M.K., Sternberg, C., Stewart, G.S.A.B., Christiansen, G., Chhabra, S.R. et al. (1996) Involvement of N-acyl-l-homoserine lactone autoinducers in controlling the multicellular behaviour of Serratia liquefaciens. Mol Microbiol 20: 127-136. Egland , K.A., and Greenberg, E.P. (1999) Quorum sensing in Vibrio fischeri: elements of the luxI promoter. Mol Microbiol 31: 1197-1204. Engebrecht, J., Nealson, K.H., and Silverman, M. (1983) Bacterial bioluminescence: isolation and genetic analysis of functions from Vibrio fischeri. Cell 32: 773-781. Enticknap, J.J., Kelly, M., Peraud, O., and Hill, R.T. (2006) Characterization of a culturable alphaproteobacterial symbiont common to many marine sponges and evidence for vertical transmission via sponge larvae. Appl Environ Microbiol 72: 3724-3732. Fenchel, T. (2001) Eppur si muove: many water column bacteria are motile. Aquat Microb Ecol 24: 197-201. Fieseler, L., Horn, M., Wagner, M., and Hentschel, U. (2004) Discovery of the novel candidate phylum 'Poribacteria' in marine sponges. Appl Environ Microbiol 70: 3724-3732. 249 Freeman, J.A., Lilley, B.N., and Bassler, B.L. (2000) A genetic analysis of the functions of LuxN: a two-component hybrid sensor kinase that regulates quorum sensing in Vibrio harveyi. Mol Microbiol 35: 139-149. Fuqua, C. (2006) The QscR quorum-sensing regulon of Pseudomonas aeruginosa: an orphan claims its identity. J Bacteriol 188: 3169-3171. Fuqua, C., and Greenberg, E.P. (2002) Listening in on bacteria: acyl-homoserine lactone signalling. Nat Rev Mol Cell Biol 3: 685-695. Fuqua, C., Parsek, M.R., and Greenberg, E.P. (2001) Regulation of gene expression by cell-to-cell communication: acyl-homoserine lactone quorum sensing. Annu Rev Genet 35: 439-468. Fuqua, C., and Winans, S.C. (1996) Conserved cis-acting promoter elements are required for density-dependent transcription of Agrobacterium tumefaciens conjugal transfer genes. J Bacteriol 178: 435-440. Fuqua, W.C., Winans, S.C., and Greenberg, E.P. (1994) Quorum sensing in bacteria: the LuxR-LuxI family of cell density-responsive transcriptional regulators. J Bacteriol 176: 269-275. Gard?res, J., Taupin, L., Sa?din, J.B., Dufour, A., and Pennec, G.L. (2012) N-acyl homoserine lactone production by bacteria within the sponge Suberites domuncula (Olivi, 1792) (Porifera, Demospongiae). Mar Biol 159: 1685-1692. 250 Geng, H., and Belas, R. (2010) Expression of tropodithietic acid biosynthesis is controlled by a novel autoinducer. J Bacteriol 192: 4377-4387. Geng, H., and Belas, R. (2011) TdaA regulates tropodithietic acid synthesis by binding to the tdaC promoter region. J Bacteriol 193: 4002-4005. Geng, H., Bruhn, J.B., Nielsen, K.F., Gram, L., and Belas, R. (2008) Genetic dissection of tropodithietic acid biosynthesis by marine roseobacters. Appl Environ Microbiol 74: 1535-1545. Gilson, L., Kuo, A., and Dunlap, P.V. (1995) AinS and a new family of autoinducer synthesis proteins. J Bacteriol 177: 6946-6951. Gonz?lez, J.M., and Moran, M.A. (1997) Numerical dominance of a group of marine bacteria in the alpha-subclass of the class Proteobacteria in coastal seawater. Appl Environ Microbiol 63: 4237-4242. Goo, E., Majerczyk, C.D., An, J.H., Chandler, J.R., Seo, Y.S., Ham, H. et al. (2012) Bacterial quorum sensing, cooperativity, and anticipation of stationary- phase stress. Proc Natl Acad Sci USA 109: 19775-19780. Gora, K.G., Tsokos, C.G., Chen, Y.E., Srinivasan, B.S., Perchuk, B.S., and Laub, M.T. (2010) A cell-type-specific protein-protein interaction modulates transcriptional activity of a master regulator in Caulobacter crescentus. Mol Cell 39: 455-467. 251 Gould, T.A., Herman, J., Krank, J., Murphy, R.C., and Churchill, M.E. (2006) Specificity of acyl-homoserine lactone synthases examined by mass spectrometry. J Bacteriol 188: 773-783. Gould, T.A., Schweizer, H.P., and Churchill, M.E. A. (2004) Structure of the Pseudomonas aeruginosa acyl-homoserinelactone synthase LasI. Mol Microbiol 53: 1135-1146. Graf, J., Dunlap, P.V., and Ruby, E.G. (1994) Effect of transposon-induced motility mutations on colonization of the host light organ by Vibrio fischeri. J Bacteriol 176: 6986-6991. Gram, L., Grossart, H.P., Schlingloff, A., and Kiorboe, T. (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. Greenberg, E.P., Hastings, J.W., and Ulitzur, S. (1979) Induction of luciferase synthesis in Beneckea harveyi by other marine bacteria. Arch Microbiol 120: 87- 91. Greene, S.E., Brilli, M., Biondi, E.G., and Komeili, A. (2012) Analysis of the CtrA pathway in Magnetospirillum reveals an ancestral role in motility in alphaproteobacteria. J Bacteriol 194: 2973-2986. Gurich, N., and Gonz?lez, J.E. (2009) Role of quorum sensing in Sinorhizobium meliloti-Alfalfa symbiosis. J Bacteriol 191: 4372-4382. 252 Hammer, B.K., and Bassler, B.L. (2003) Quorum sensing controls biofilm formation in Vibrio cholerae. Mol Microbiol 50: 101-114. Hanzelka, B.L., Parsek, M.R., Val, D.L., Dunlap, P.V., Cronan, J.E., and Greenberg, E.P. (1999) Acylhomoserine lactone synthase activity of the Vibrio fischeri AinS protein. J Bacteriol 181: 5766-5770. Hengge, R. (2009) Principles of c-di-GMP signalling in bacteria. Nat Rev Microbiol 7: 263-273. Henke, J.M., and Bassler, B.L. (2004) Three parallel quorum-sensing systems regulate gene expression in Vibrio harveyi. J Bacteriol 186: 6902-6914. Hentschel, U., Schmid, M., Wagner, M., Fieseler, L., Gernert, C., and Hacker, J. (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. Hentschel, U., Hopke, J., Horn, M., Friedrich, A.B., Wagner, M., Hacker, J., and Moore, B.S. (2002) Molecular evidence for a uniform microbial community in sponges from different oceans. Appl Environ Microbiol 68: 4431-4440. Hentschel, U., Usher, K.M., and Taylor, M.W. (2006) Marine sponges as microbial fermenters. FEMS Microbiol Ecol 55: 167-177. Hentschel, U., Piel, J., Degnan, S.M., and Taylor, M.W. (2012) Genomic insights into the marine sponge microbiome. Nat Rev Microbiol 10: 641-654. 253 Hibbing, M.E., and Fuqua, C. (2011) Antiparallel and Interlinked Control of Cellular Iron Levels by the Irr and RirA Regulators of Agrobacterium tumefaciens. J Bacteriol 193: 3461-3472. Hickerson, M.J., Meyer, C.P., and Moritz, C. (2006) DNA Barcoding will often fail to discover new animal species over broad parameter space. Syst Biol 55: 729- 739. Higgins, D.A., Pomianek, M.E., Kraml, C.M., Taylor, R.K., Semmelhack, M.F., and Bassler, B.L. (2007) The major Vibrio cholerae autoinducer and its role in virulence factor production. Nature 450: 883-886. Hill, R. T. (2004) Microbes from marine sponges: a treasure trove of biodiversity for natural products discovery. In Microbial diversity and bioprospecting. A. T. Bull (ed.). ASM Press, Washington, D.C. p177-190. Hirakawa, H., Oda, Y., Phattarasukol, S., Armour, C.D., Castle, J.C., Raymond, C.K. et al. (2011) Activity of the Rhodopseudomonas palustris p-Coumaroyl- homoserine lactone-responsive transcription factor RpaR. J Bacteriol 193: 2598- 2607. Holtzendorff, J., Hung, D., Brende, P., Reisenauer, A., Viollier, P.H., McAdams, H.H., and Shapiro, L. (2004) Oscillating global regulators control the genetic circuit driving a bacterial cell cycle. Science 304: 983-987. 254 Hooper, J.N.A., and van Soest, R.W.M. (2002) In Systema Porifera: a guide to the classification of Sponges. Hooper, J.N.A., and van Soest, R.W.M. (eds). New York: Kluwer Academic/Plenum Publishers, p. 1718. Huang, Y.L., Ki, J.-S., Case, R.J., and Qian, P.-Y. (2008) Diversity and acyl- homoserine lactone production among subtidal biofilm-forming bacteria. Aquat Microb Ecol 52: 185-193. Hwang, I., Cook, D.M., and Farrand, S.K. (1995) A new regulatory element modulates homoserine lactone-mediated autoinduction of Ti plasmid conjugal transfer. J Bacteriol 177: 449-458. Iniesta, A.A., McGrath, P.T., Reisenauer, A., McAdams, H.H., and Shapiro, L. (2006) A phospho-signaling pathway controls the localization and activity of a protease complex critical for bacterial cell cycle progression. Proc Natl Acad Sci USA 103: 10935-10940. Jacobs, C., Ausmees, N., Cordwell, S.J., Shapiro, L., and Laub, M.T. (2003) Functions of the CckA histidine kinase in Caulobacter cell cycle control. Mol Microbiol 47: 1279-1290. Jacobs, C., Domian, I.J., Maddock, J.R., and Shapiro, L. (1999) Cell cycle? dependent polar localization of an essential bacterial histidine kinase that controls DNA replication and cell division. Cell 97: 111-120. Joelsson, A., Kan, B., and Zhu, J. (2007) Quorum sensing enhances the stress response in Vibrio cholerae. Appl Environ Microbiol 73: 3742-3746. 255 Joint, I., Callow, M.E., Callow, J.A., and Clarke, K.R. (2000) The attachment of Enteromorpha zoospores to a bacterial biofilm assemblage. Biofouling 16: 151- 158. Joint, I., Tait, K., Callow, M.E., Callow, J.A., Milton, D., Williams, P., and C?mara, M. (2002) Cell-to-cell communication across the prokaryote-eukaryote boundary. Science 298: 1207. Kalogeraki, V.S., and Winans, S.C. (1997) Suicide plasmids containing promoterless reporter genes can simultaneously disrupt and create fusions to target genes of diverse bacteria. Gene 188: 69-75. Kanbe, M., Yagasaki, J., Zehner, S., G?ttfert, M., and Aizawa, S.-I. (2007) Characterization of two sets of subpolar flagella in Bradyrhizobium japonicum. J Bacteriol 189: 1083-1089. Kay, E., Humair, B., D?nervaud, V., Riedel, K., Spahr, S., Eberl, L. et al. (2006) Two GacA-dependent small RNAs modulate the quorum-sensing response in Pseudomonas aeruginosa. J Bacteriol 188: 6026-6033. Khan, S.R., Gaines, J., Roop, R.M., 2nd, and Farrand, S.K. (2008) Broad-host- range expression vectors with tightly regulated promoters and their use to examine the influence of TraR and TraM expression on Ti plasmid quorum sensing. Appl Environ Microbiol 74: 5053-5062. 256 Kim, J., Kang, Y., Choi, O., Jeong, Y., Jeong, J.E., Lim, J.Y. et al. (2007) Regulation of polar flagellum genes is mediated by quorum sensing and FlhDC in Burkholderia glumae. Mol Microbiol 61: 165-179. Kiratisin, P., Tucker, K.D., and Passador, L. (2002) LasR, a transcriptional activator of Pseudomonas aeruginosa virulence genes, functions as a multimer. J Bacteriol 184: 4912-4919. Klose, K.E., and Mekalanos, J.J. (1998) Distinct roles of an alternative sigma factor during both free-swimming and colonizing phase of the Vibrio cholerae pathogenic cycle. Mol Microbiol 28: 501-520. Kovach, M.E., Elzer, P.H., Hill, D.S., Robertson, G.T., Farris, M.A., Roop II, R.M., and Peterson, K.M. (1995) Four new derivatives of the broad-host-range cloning vector pBBR1MCS, carrying different antibiotic-resistance cassettes. Gene 166: 175-176. Krasteva, P.V., Fong, J.C., Shikuma, N.J., Beyhan, S., Navarro, M.V., Yildiz, F.H., and Sondermann, H. (2010) Vibrio cholerae VpsT regulates matrix production and motility by directly sensing cyclic di-GMP. Science 327: 866-868. Kuo, A., Blough, N.V., and Dunlap, P.V. (1994) Multiple N-acyl-L-homoserine lactone autoinducers of luminescence in the marine symbiotic bacterium Vibrio fischeri. J Bacteriol 176: 7558-7565. Kutsukake, K., Ohya, Y., and Iino, T. (1990) Transcriptional analysis of the flagellar regulon of Salmonella typhimurium. J Bacteriol 172: 741-747. 257 Lang, A.S., and Beatty, J.T. (2000) Genetic analysis of a bacterial genetic exchange element: The gene transfer agent of Rhodobacter capsulatus. Proc Natl Acad Sci USA 97: 859-864. Latifi, A., Foglino, M., Tanaka, K., Williams, P., and Lazdunski, A. (1996) A hierarchical quorum-sensing cascade in Pseudomonas aeruginosa links the transcriptional activators LasR and RhIR (VsmR) to expression of the stationary- phase sigma factor RpoS. Mol Microbiol 21: 1137-1146. Laub, M.T., Chen, S.L., Shapiro, L., and McAdams, H.H. (2002) Genes directly controlled by CtrA, a master regulator of the Caulobacter cell cycle. Proc Natl Acad Sci USA 99: 4632-4637. Laub, M.T., and Goulian, M. (2007) Specificity in two-component signal transduction pathways. Annu Rev Genet 41: 121-145. Laub, M.T., McAdams, H.H., Feldblyum, T., Fraser, C.M., and Shapiro, L. (2000) Global analysis of the genetic network controlling a bacterial cell cycle. Science 290: 2144-2148. Lee, O.O., Chui, P.Y., Wong, Y.H., Pawlik, J.R., and Qian, P.-Y. (2009) Evidence for vertical transmission of bacterial symbionts from adult to embryo in the Caribbean sponge Svenzea zeai. Appl Environ Microbiol 75: 6147-6156. Lenz, D.H., Mok, K.C., Lilley, B.N., Kulkarni, R.V., Wingreen, N.S., and Bassler, B.L. (2004) The small RNA chaperone Hfq and multiple small RNAs control quorum sensing in Vibrio harveyi and Vibrio cholerae. Cell 118: 69-82. 258 Lequette, Y., Lee, J.-H., Ledgham, F., Lazdunski, A., and Greenberg, E.P. (2006) A distinct QscR regulon in the Pseudomonas aeruginosa quorum-sensing circuit. J Bacteriol 188: 3365-3370. Leung M.M., Brimacombe C.A., and Beatty J.T. (2013) Transcriptional regulation of the Rhodobacter capsulatus response regulator CtrA. Microbiology 159: 96- 106. Levi, C. (1957) Ontogeny and systematics in sponges. Syst Zool 6: 174-183. Lindemann, A., Pessi, G., Schaefer, A.L., Mattmann, M.E., Christensen, Q.H., Kessler, A. et al. (2011) Isovaleryl-homoserine lactone, an unusual branched- chain quorum-sensing signal from the soybean symbiont Bradyrhizobium japonicum. Proc Natl Acad Sci USA 108: 16765-16770. Limmer, S., Haller, S., Drenkard, E., Lee, J., Yu, S., Kocks, C. et al. (2011) Pseudomonas aeruginosa RhlR is required to neutralize the cellular immune response in a Drosophila melanogaster oral infection model. Proc Natl Acad Sci USA 108: 17378-17383. Liu, X., and Matsumura, P. (1994) The FlhD/FlhC complex , a transcriptional activator of the Escherichia coli flagellar class II operons. J Bacteriol 176: 7345- 7351. Llamas, I., Keshavan, N., and Gonz?lez, J.E. (2004) Use of Sinorhizobium meliloti as an indicator for specific detection of long-chain N-acyl homoserine lactones. Appl Environ Microbiol 70: 3715-3723. 259 Long, R.A., and Azam, F. (2001) Antagonistic interactions among marine pelagic bacteria. Appl Environ Microbiol 67: 4975-4983. Lyczak, J.B., Cannon, C.L., and Pier, G.B. (2000) Establishment of Pseudomonas aeruginosa infection: lessons from a versatile opportunist. Microb Infect 2: 1051-1060. Ma, S., Selvaraj, U., Ohman, D.E., Quarless, R., Hassett, D.J., and Wozniak, D.J. (1998) Phosphorylation-independent activity of the response regulators AlgB and AlgR in promoting alginate biosynthesis in mucoid Pseudomonas aeruginosa. J Bacteriol 180: 956-968. Maclouf, J., Grassi, J., and Pradelles, P. (1987) Development of enzyme- immunoassay techniques for the measurement of eicosanoids. In Prostaglandin and lipid metabolism in radiation injury. Walden, T.L., Jr, and Hughes, H.N. (eds). New York: Plenum press, pp. 355-364. Maidak, B.L., Cole, J.R., Parker, C.T., Jr., Garrity, G.M., Larsen, N., Li, B. et al. (1999) A new version of the RDP (Ribosomal Database Project). Nucleic Acids Res 27: 171-173. Mandel, M.J., Wollenberg, M.S., Stabb, E.V., Visick, K.L., and Ruby, E.G. (2009) A single regulatory gene is sufficient to alter bacterial host range. Nature 458: 215-220. 260 Martens, T., Gram, L., Grossart, H.P., Kessler, D., Muller, R., Simon, M. et al. (2007) Bacteria of the Roseobacter clade show potential for secondary metabolite production. Microb Ecol 54: 31-42. Mashburn, L.M., and Whiteley, M. (2005) Membrane vesicles traffic signals and facilitate group activities in a prokaryote. Nature 437: 422-425. Massie, J.P., Reynolds, E.L., Koestler, B.J., Cong, J.-P., Agostoni, M., and Waters, C.M. (2012) Quantification of high-specificity cyclic diguanylate signaling. Proc Natl Acad Sci USA 109: 12746-12751. Mayfield, C.I., and Inniss, W.E. (1977) A rapid, simple method for staining bacterial flagella. Can J Microbiol 23: 1311-1313. McFall-Ngai, M. J., and Ruby, E. G. (1991) Symbiont recognition and subsequent morphogenesis as early events in an animal-bacterial mutualism. Science 254: 1491-1494. Meeks, J.C., Wolk, C.P., Lockau, W., Schilling, N., Shaffer, P.W., and Chien, W.S. (1978) Pathways of assimilation of [13N]N2 and 13NH4+ by cyanobacteria with and without heterocysts. J Bacteriol 134: 125-130. Mercer, R.G., Callister, S.J., Lipton, M.S., Pasa-Tolic, L., Strnad, H., Paces, V. et al. (2010) Loss of the response regulator CtrA causes pleiotropic effects on gene expression but does not affect growth phase regulation in Rhodobacter capsulatus. J Bacteriol 192: 2701-2710. 261 Mercer R.G., Quinlan M., Rose A.R., Noll S., Beatty J.T., and Lang, A.S. (2012) Regulatory systems controlling motility and gene transfer agent production and release in Rhodobacter capsulatus. FEMS Microbiol Lett 331: 53-62. Merritt, P.M., Danhorn, T., and Fuqua, C. (2007) Motility and chemotaxis in Agrobacterium tumefaciens surface attachment and biofilm formation. J Bacteriol 189: 8005-8014. Mersereau, M., Pazour, G.J., and Das, A. (1990) Efficient transformation of Agrobacterium tumefaciens by electroporation. Gene 90: 149-151. Meyer, C.P., and Paulay, G. (2005) DNA Barcoding: error rates based on comprehensive sampling. PLoS Biol 3: e422. doi:410.1371/journal.pbio.0030422. Miller, J.H. (1972) Experiments in molecular genetics. Cold Spring Harbor, New York: Cold Spring Laboratory Press. Miller, M.B., and Bassler, B.L. (2001) Quorum sensing in bacteria. Annu Rev Microbiol 55: 165-199. Miller, M.B., Skorupski, K., Lenz, D.H., Taylor, R.K., and Bassler, B.L. (2002) Parallel quorum sensing systems converge to regulate virulence in Vibrio cholerae. Cell 110: 303-314. Miller, T.R., and Belas, R. (2006) Motility is involved in Silicibacter sp. TM1040 interaction with dinoflagellates. Environ Microbiol 8: 1648-1659. 262 Milton, D.L. (2006) Quorum sensing in vibrios: complexity for diversification. Int J Med Microbiol 296: 61-71. Milton, D.L., Chalker, V.J., Kirke, D., Hardman, A., C?mara, M., and Williams, P. (2001) The LuxM homologue VanM from Vibrio anguillarum directs the synthesis of N-(3-hydroxyhexanoyl) homoserine lactone and N-hexanoylhomoserine lactone. J Bacteriol 183: 3537-3547. Minogue, T.D., Carlier, A.L., Koutsoudis, M.D., and von Bodman, S.B. (2005) The cell density-dependent expression of stewartan exopolysaccharide in Pantoea stewartii ssp. stewartii is a function of EsaR-mediated repression of the rcsA gene. Mol Microbiol 56: 189-203. Mohamed, N.M., Colman, A.S., Tal, Y., and Hill, R.T. (2008a) Diversity and expression of nitrogen fixation genes in bacterial symbionts of marine sponges. Environ Microbiol 10: 2910-2921. Mohamed, N.M., Enticknap, J.J., Lohr, J.E., McIntosh, S.M., and Hill, R.T. (2008b) Changes in bacterial communities of the marine sponge Mycale laxissima on transfer into aquaculture. Appl Environ Microbiol 74: 1209-1222. Mohamed, N.M., Cicirelli, E.M., Kan, J., Chen, F., Fuqua, C., and Hill, R.T. (2008c) Diversity and quorum-sensing signal production of Proteobacteria associated with marine sponges. Environ Microbiol 10: 75-86. 263 Montalvo, N.F., and Hill, R.T. (2011) Sponge-associated bacteria are strictly maintained in two closely related but geographically distant sponge hosts. Appl Environ Microbiol 77: 7207-7216. Moran, M.A., Belas, R., Schell, M.A., Gonz?lez, J.M., Sun, F., Sun, S. et al. (2007) Ecological genomics of marine Roseobacters. Appl Environ Microbiol 73: 4559-4569. Moran, M.A., Buchan, A., Gonzalez, J.M., Heidelberg, J.F., Whitman, W.B., Kiene, R.P. et al. (2004) Genome sequence of Silicibacter pomeroyi reveals adaptations to the marine environment. Nature 432: 910-913. Moran, M.A., Reisch, C.R., Kiene, R.P., and Whitman, W.B. (2012) Genomic insights into bacterial DMSP transformations. Annu Rev Mar Sci 4: 523-542. Mor?, M.I., Finger, L.D., Stryker, J.L., Fuqua, C., Eberhard, A., and Winans, S.C. (1996) Enzymatic synthesis of a quorum-sensing autoinducer through use of defined substrates. Science 272: 1655-1658. Nealson, K.H., Platt, T., and Hastings, J.W. (1970) Cellular control of the synthesis and activity of the bacterial luminescent system. J Bacteriol 104: 313- 322. Ng, W.L., and Bassler, B.L. (2009) Bacterial quorum-sensing network architectures. Annu Rev Genet 43: 197-222. 264 Nyholm, S.V., and McFall-Ngai, M.J. (2004) The winnowing : establishing the squid- Vibrio symbiosis. Nat Rev Microbiol 2: 632-642. O?Toole, G.A., and Kolter, R. (1998) Flagellar and twitching motility are necessary for Pseudomonas aeruginosa biofilm development. Mol Microbiol 30: 295-304. Paggi, R.A., Martone, C.B., Fuqua, C., and De Castro, R.E. (2003) Detection of quorum sensing signals in the haloalkaliphilic archaeon Natronococcus occultus. FEMS Microbiol Lett 221: 49-52. Pearson, J.P., Gray, K.M., Passador, L., Tucker, K.D., Eberhard, A., Iglewski, B.H., and Greenberg, E.P. (1994) Structure of the autoinduer required for expression of Pseudomonas aeruginosa virulence genes. Proc Natl Acad Sci USA 91: 197-201. Pearson, J.P., Passador, L., Iglewski, B.H., and Greenberg, E.P. (1995) A second N-acylhomoserine lactone signal produced by Pseudomonas aeruginosa. Proc Natl Acad Sci USA 92: 1490-1494. Pearson, J.P., Pesci, E.C., and Iglewski, B.H. (1997) Roles of Pseudomonas aeruginosa las and rhI quorum-sensing systems in control of elastase and rhamnolipid biosynthesis genes. J Bacteriol 179: 5756-5767. Pellock, B.J., Teplitski, M., Boinay, R.P., Bauer, W.D., and Walker, G.C. (2002) A LuxR homolog controls production of symbiotically active extracellular polysaccharide II by Sinorhizobium meliloti. J Bacteriol 184: 5067-5076. 265 Pesci, E.C., Pearson, J.P., Seed, P.C., and Iglewski, B.H. (1997) Regulation of las and rhl quorum sensing in Pseudomonas aeruginosa. J Bacteriol 179: 3127- 3132. Piekarski, T., Buchholz, I., Drepper, T., Schobert, M., Wagner-D?bler, I., Tielen, P., and Jahn, D. (2009) Genetic tools for the investigation of Roseobacter clade bacteria. BMC Microbiol 9: 265-276. Pompeani, A.J., Irgon, J.J., Berger, M.F., Bulyk, M.L., Wingreen, N.S., and Bassler, B.L. (2008) The Vibrio harveyi master quorum-sensing regulator, LuxR, a TetR-type protein is both an activator and a repressor: DNA recognition and binding specificity at target promoters. Mol Microbiol 70: 76-88. Purcell, E.B., McKee, R.W., McBride, S.M., Waters, C.M., and Tamayo, R. (2012) Cyclic diguanylate inversely regulates motility and aggregation in Clostridium difficile. J Bacteriol 194: 3307-3316. Puskas, A., Greenberg, E.P., Kaplan, S., and Schaefer, A.L. (1997) A quorum- sensing system in the free-living photosynthetic bacterium Rhodobacter sphaeroides. J Bacteriol 179: 7530-7537. Quon, K.C., Marczynski, G.T., and Shapiro, L. (1996) Cell cycle control by an essential bacterial two-component signal transduction protein. Cell 84: 83-93. Rezzonico, F., and Duffy, B. (2008) Lack of genomic evidence of AI-2 receptors suggests a non-quorum sensing role for luxS in most bacteria. BMC Microbiol 8: 154-173. 266 Riebesell, U., Zondervan, I., Rost, B., Tortell, P.D., Zeebe, R.E., and Morel, F.M.M. (2000) Reduced calcification of marine plankton in response to increased atmospheric CO2. Nature 407: 364-367. Robertson, G.T., Reisenauer, A., Wright, R., Jensen, R.B., Jensen, A., Shapiro, L., and Ropp, R.M., 2nd (2000) The Brucella abortus CcrM DNA methyltransferase is essential for viability, and its overexpression attenuates intracellular replication in murine macrophage. J Bacteriol 182: 3482-3489. Romero, M., Avendano-Herrera, R., Magarinos, B., Camara, M., and Otero, A. (2010) Acylhomoserine lactone production and degradation by the fish pathogen Tenacibaculum maritimum, a member of the Cytophaga-Flavobacterium- Bacteroides (CFB) group. FEMS Microbiol Lett 304: 131-139. Rose, T.M., Henikoff, J.G., and Henikoff, S. (2003) CODEHOP (COnsensus- DEgenerate Hybrid Oligonucleotide Primer) PCR primer design. Nucleic Acids Res 31: 3763-3766. Rubin, E.J., Akerley, B.J., Novik, V.N., Lampe, D.J., Husson, R.N., and Mekalanos, J.J. (1999) In vivo transposition of mariner-based elements in enteric bacteria and mycobacteria. Proc Natl Acad Sci USA 96: 1645-1650. Ryan, R.P., and Dow, J.M. (2008) Diffusible signals and interspecies communication in bacteria. Microbiology 154: 1845-1858. Rypien, K.L., Ward, J.R., and Azam, F. (2010) Antagonistic interactions among coral-associated bacteria. Environ Microbiol 12: 28-39. 267 Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989) Molecular cloning: a laboratory manual. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press. Schaefer, A.L., Greenberg, E.P., Oliver, C.M., Oda, Y., Huang, J.J., Bittan-Banin, G. et al. (2008) A new class of homoserine lactone quorum-sensing signals. Nature 454: 595-600. Schaefer, A.L., Taylor, T.A., Beatty, J.T., and Greenberg, E.P. (2002) Long-chain acyl-homoserine lactone quorum-sensing regulation of Rhodobacter capsulatus gene transfer agent production. J Bacteriol 184: 6515-6521. Schaefer, A.L., Val, D.L., Hanzelka, B.L., Cronan, J., J.E., and Greenberg, E.P. (1996) Generation of cell-to-cell signals in quorum sensing: acyl homoserine lactone synthase activity of a purified Vibrio fischeri LuxI protein. Proc Natl Acad Sci USA 93: 9505-9509. Sch?r, J., Sickmann, A., and Beier, D. (2005) Phosphorylation-independent activity of atypical response regulators of Helicobacter pylori. J Bacteriol 187: 3100-3109. Schmitt, S., Angermeier, H., Schiller, R., Lindquist, N., and Hentschel, U. (2008) Molecular microbial diversity survey of sponge reproductive stages and mechanistic insights into vertical transmission of microbial symbionts. Appl Environ Microbiol 74: 7694-7708. 268 Schuster, M., and Greenberg, E.P. (2008) LuxR-type protein in Pseudomonas aeruginosa quorum sensing: distinct mechanisms with global implication. In Chemical communication among bacteria. Winans, S.C., and Bassler, B.L. (eds). Washington, D.C.: ASm press, pp. 133-144. Schuster, M., Urbanowski, M.L., and Greenberg, E.P. (2004) Promoter specificity in Pseudomonas aeruginosa quorum sensing revealed by DNA binding of purified LasR. Proc Natl Acad Sci USA 101: 15833-15839. Schuster, M., Lostroh, C.P., Ogi, T., and Greenberg, E.P. (2003) Identification, timing, and signal specificity of Pseudomonas aeruginosa quorum-controlled genes: a transcriptome analysis. J Bacteriol 185: 2066-2079. Seyedsayamdost, M.R., Case, R.J., Kolter, R., and Clardy, J. (2011) The Jekyll- and-Hyde chemistry of Phaeobacter gallaeciensis. Nat Chem 3: 331-335. Sharif, D.I., Gallon, J., Smith, C.J., and Dudley, E. (2008) Quorum sensing in Cyanobacteria: N-octanoyl-homoserine lactone release and response, by the epilithic colonial cyanobacterium Gloeothece PCC6909. ISME J 2: 1171-1182. Sharp, K.H., Eam, B., Faulkner, D.J., and Haygood, M.G. (2007) Vertical transmission of diverse microbes in the tropical sponge Corticium sp. Appl Environ Microbiol 73: 622-629. Shrout, J.D., Chopp, D.L., Just, C.L., Hentzer, M., Givskov, M., and Parsek, M.R. (2006) The impact of quorum sensing and swarming motility on Pseudomonas 269 aeruginosa biofilm formation is nutritionally conditional. Mol Microbiol 62: 1264- 1277. Slightom, R.N., and Buchan, A. (2009) Surface colonization by marine roseobacters: integrating genotype and phenotype. Appl Environ Microbiol 75: 6027-6037. Smith, T.G., and Hoover, T.R. (2009) Deciphering bacterial flagellar gene regulatory networks in the genomic era. Adv Appl Microbiol 67: 257-295. Soutourina, O.A., and Bertin, P.N. (2003) Regulation cascade of flagellar expression in Gram-negative bacteria FEMS Microbiol Rev 27: 505-523. Spencer, W., Siam, R., Ouimet, M.-C., Bastedo, D.P., and Marczynski, G.T. (2009) CtrA, a global response regulator, uses a distinct second category of weak DNA binding sites for cell cycle transcription control in Caulobacter crescentus. J Bacteriol 191: 5458-5470. Srivastava, M., Simakov, O., Chapman, J., Fahey, B., Gauthier, M.E., Mitros, T. et al. (2010) The Amphimedon queenslandica genome and the evolution of animal complexity. Nature 466: 720-726. Stabb, E.V., Schaefer, A.L., Bose, J.L., and Ruby, E.G. (2008) Quorum signaling and symbiosis in the marine luminous bacterium Vibrio fischeri. In Chemical communication among bacteria. Winans, S.C., and Bassler, B.L. (eds). Washington, D.C.: ASM press, pp. 233-250. 270 Stevens, A.M., and Greenberg, E.P. (1999) Transcriptional activation by LuxR. In Cell-cell signaling in Bacteria. Dunny, G.M., and Winans, S.C. (eds). Washington, DC: ASM press, p231. Subramoni, S., and Venturi, V. (2009) LuxR-family 'solo': bachelor sensors/regulators of signalling molecules. Microbiology 155: 1377-1385. Sule, P., and Belas, R. (2013) A novel inducer of Roseobacter motility is also a disruptor of algal symbiosis. J Bacteriol 195: 637-646. Surette, M.G., and Davies, J. (2008) A new look at secondary metabolites. In Chemical communication among bacteria. Winans, S.C., and Bassler, B.L. (eds). Washington, D.C.: ASM press, pp. 307-322. Tait, K., Joint, I., Daykin, M., Milton, D.L., Williams, P., and C?mara, M. (2005) Disruption of quorum sensing in seawater abolishes attraction of zoospores of the green alga Ulva to bacterial biofilms. Environ Microbiol 7: 229-240. Tait, K., Williamson, H., Atkinson, S., Williams, P., C?mara, M., and Joint, I. (2009) Turnover of quorum sensing signal molecules modulates cross-kingdom signalling. Environ Microbiol 11: 1792-1802. Taylor, M.W., Radax, R., Steger, D., and Wagner, M. (2007) Sponge-associated microorganisms: evolution, ecology, and biotechnological potential. Microbiol Mol Biol Rev 71: 295-347. 271 Taylor, M.W., Schupp, P.J., Baillie, H.J., Charlton, T.S., de Nys, R., Kjelleberg, S., and Steinberg, P.D. (2004) Evidence for acyl homoserine lactone signal production in bacteria associated with marine sponges. Appl Environ Microbiol 70: 4387-4389. Taylor, M.W., Tsai, P., Simister, R.L., Deines, P., Botte, E., Ericson, G. et al. (2013) 'Sponge-specific' bacteria are widespread (but rare) in diverse marine environments. ISME J 7: 438-443. Temp?, J., Petit, A., Holsters, M., van Montagu, M., and Schell, J. (1977) Thermosensitive step associated with transfer of the Ti plasmid during conjugation: possible relation to transformation in crown gall. Proc Natl Acad Sci USA 74: 2848-2849. Tomlinson, A.D., Ramey-Hartung, B., Day, T.W., Merritt, P.M., and Fuqua, C. (2010) Agrobacterium tumefaciens ExoR represses succinoglycan biosynthesis and is required for biofilm formation and motility. Microbiology 156: 2670-2681. Thakur, N.L., Anil, A.C., and M?ller, W.E.G. (2004) Culturable epibacteria of the amrine sponge Ircinia fusca: temporal variations and their possible role in the epibacterial defense of the host. Aquat Microb Ecol 37: 295-304. Tsai, C.-S., and Winans, S.C. (2010) LuxR-type quorum-sensing regulators that are detached from common scents. Mol Microbiol 77: 1072-1082. 272 Tu, K.C., and Bassler, B.L. (2007) Multiple small RNAs act additively to integrate sensory information and control quorum sensing in Vibrio harveyi. Genes Dev 21: 221-233. Tu, K.C., Waters, C.M., Svenningsen, S.L., and Bassler, B.L. (2008) A small- RNA-mediated negative feedback loop controls quorum-sensing dynamics in Vibrio harveyi. Mol Microbiol 70: 896-907. Vacelet, J. (1975) Etude en microscopie eletronique de l'association entre bacteries et spongiaires du genre Verongia (Dictyoceratida). J Microsc Biol Cell 23: 271-288. Vacelet, J., and Donadey, C. (1977) Electron microscope study of association between some sponges and bacteria. J Exp Mar Biol Ecol 30: 301-314. van Soest, R.W.M., Boury-Esnault, N., Vacelet, J., Dohrmann, M., Erpenbeck, D., De Voogd, N.J. et al. (2012) Global diversity of sponges (Porifera). PLoS One 7: e35105. Visick, K.L., Foster, J., Doino, J., McFall-Ngai, M., and Ruby, E.G. (2000) Vibrio fischeri lux genes play an important role in colonization and development of the host light organ. J Bacteriol 182: 4578-4586. von Bodman, S.B., and Farrand, S.K. (1995) Capsular polysaccharide biosynthesis and pathogenicity in Erwinia stewartii require induction by an N- acylhomoserine lactone autoinducer. J Bacteriol 177: 5000-5008. 273 von Bodman, S.B., Majerczak, D.R., and Coplin, D.L. (1998) A negative regulator mediates quorum-sensing control of exopolysaccharide production in Pantoea stewartii subsp. stewartii. Proc Natl Acad Sci USA 95: 7687-7692. Wagner-D?bler, I., Ballhausen, B., Berger, M., Brinkhoff, T., Buchholz, I., Bunk, B. et al. (2010) The complete genome sequence of the algal symbiont Dinoroseobacter shibae: a hitchhiker's guide to life in the sea. ISME J 4: 61-77. Wagner-D?bler, I., Thiel, V., Eberl, L., Allgaier, M., Bodor, A., Meyer, S. et al. (2005) Discovery of complex mixtures of novel long-chain quorum sensing signals in free-living and host-associated marine Alphaproteobacteria. ChemBioChem 6: 2195-2206. Warrens, A.N., Jones, M.D., and Lechler, R.I. (1997) Splicing by overlap extension by PCR using asymmetric amplification: an improved technique for the generaton of hybrid proteins of immunological interest. Gene 186: 29-35. Waters, C.M., Lu, W., Rabinowitz, J.D., and Bassler, B.L. (2008) Quorum sensing controls biofilm formation in Vibrio cholerae through modulation of cyclic di-GMP levels and repression of vpsT. J Bacteriol 190: 2527-2536. Watson, W.T., Minogue, T.D., Val, D.L., von Bodman, S.B., and Churchill, M.E.A. (2002) Structural basis and specificity of acyl-homoserine lactone signal production in bacterial quorum sensing. Mol Cell 9: 685-694. Webster, N.S., and Blackall, L.L. (2009) What do we really know about sponge- microbial symbioses? ISME J 3: 1-3. 274 Webster, N.S., and Hill, R.T. (2001) The culturable microbial community of the Great Barrier Reef sponge Rhopaloeides odorabile is dominated by an $- Proteobacterium. Mar Biol 138: 843-851. Webster, N.S., Taylor, M.W., Behnam, F., Lucker, S., Rattei, T., Whalan, S. et al. (2010) Deep sequencing reveals exceptional diversity and modes of transmission for bacterial sponge symbionts. Environ Microbiol 12: 2070-2082. Webster, N.S., and Taylor, M.W. (2012) Marine sponges and their microbial symbionts: love and other relationships. Environ Microbiol 14: 335-346. Wei, Y., Ng, W.L., Cong, J., and Bassler, B.L. (2012) Ligand and antagonist driven regulation of the Vibrio cholerae quorum-sensing receptor CqsS. Mol Microbiol 83: 1095-1108. Wheeler, G.L., Tait, K., Taylor, A., Brownlee, C., and Joint, I.A.N. (2006) Acyl- homoserine lactones modulate the settlement rate of zoospores of the marine alga Ulva intestinalis via a novel chemokinetic mechanism. Plant, Cell and Environment 29: 608-618. Wilkinson, C.R. (1983) Phylogeny of bacterial and cyanobacterial symbionts in marine sponges. Endocytobiology 2: 993?1002. Wilkinson, C.R., and Fay, P. (1979) Nitrogen fixation in coral reef sponges with symbiotic cyanobacteria. Nature 279: 527-529. 275 Williamson, N.R., Fineran, P.C., Ogawa, W., Woodley, L.R., and Salmond, G.P.C. (2008) Integrated regulation involving quorum sensing, a two-component system, a GGDEF/EAL domain protein and a post-transcriptional regulator controls swarming and RhlA-dependent surfactant biosynthesis in Serratia. Environ Microbiol 10: 1202-1217. Yates, E.A., Philipp, B., Buckley, C., Atkinso, S., Chhabra, S.R., Sockett, R.E. et al. (2002) N-acylhomoserine lactones undergo lactonolysis in a pH-, temperature-, and acyl chain length-dependent manner during growth of Yersinia pseudotuberculosis and Pseudomonas aeruginosa. Infect Immun 70: 5635-5646. Zan, J., Cicirelli, E.M., Mohamed, N.M., Sibhatu, H., Kroll, S., Choi, O. et al. (2012) A complex LuxR-LuxI type quorum sensing network in a roseobacterial marine sponge symbiont activates flagellar motility and inhibits biofilm formation. Mol Microbiol 85: 916-933. Zan, J., Fricke, W.F., Fuqua, C., Ravel, J., and Hill, R.T. (2011a) Genome sequence of Ruegeria sp. Strain KLH11, an N-acylhomoserine lactone-producing bacterium isolated from the marine sponge Mycale laxissima. J Bacteriol 193: 5011-5012. Zan, J., Fuqua, C., and Hill, R.T. (2011b) Diversity and functional analysis of luxS genes in Vibrios from marine sponges Mycale laxissima and Ircinia strobilina. ISME J 5: 1505-1516. 276 Zhang, G., Zhang, F., Ding, G., Li, J., Guo, X., Zhu, J. et al. (2012) Acyl homoserine lactone-based quorum sensing in a methanogenic archaeon. ISME J 6: 1336-1344. Zhang, R.G., Pappas, K.M., Brace, J.L., Miller, P.C., Oulmassov, T., Molyneaux, J.M. et al. (2002) Structure of a bacterial quorum-sensing transciption factor complexed with pheromone and DNA. Nature 417: 971-974. Zhu, J., Beaber, J.W., More, M.I., Fuqua, C., Eberhard, A., and Winans, S.C. (1998) Analogs of the autoinducer 3-oxooctanoyl-homoserine lactone strongly inhibit activity of the TraR protein of Agrobacterium tumefaciens. J Bacteriol 180: 5398-5405. Zhu, J., Chai, Y., Zhong, Z., Li, S., and Winans, S.C. (2003) Agrobacterium bioassay strain for ultrasensitive detection of N-acylhomoserine lactone-type quorum-sensing molecules: detection of autoinducers in Mesorhizobium huakuii. Appl Environ Microbiol 69: 6949-6953. Zhu, J., and Mekalanos, J.J. (2003) Quorum sensing-dependent biofilms enhance colonization in Vibrio cholerae. Dev Cell 5: 647-656. Zhu, J., and Winans, S.C. (2001) The quorum-sensing transcriptional regulator TraR requires its cognate signaling ligand for protein folding, protease resistance, and dimerization. Proc Natl Acad Sci USA 98: 1507-1512.