UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Evolution and regulation of the gene transfer agent (GTA) of Rhodobacter capsulatus Taylor, Terumi Anne 2004

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2004-0655.pdf [ 5.65MB ]
Metadata
JSON: 831-1.0099793.json
JSON-LD: 831-1.0099793-ld.json
RDF/XML (Pretty): 831-1.0099793-rdf.xml
RDF/JSON: 831-1.0099793-rdf.json
Turtle: 831-1.0099793-turtle.txt
N-Triples: 831-1.0099793-rdf-ntriples.txt
Original Record: 831-1.0099793-source.json
Full Text
831-1.0099793-fulltext.txt
Citation
831-1.0099793.ris

Full Text

Evolution and Regulation of the Gene Transfer Agent (GTA) of Rhodobacter capsulatus by TERUMI A N N E T A Y L O R B.Sc, Simon Fraser University, 2000  A THESIS SUBMITTED IN P A R T I A L F U L F I L L M E N T OF THE REQUIREMENTS FOR THE D E G R E E OF M A S T E R OF SCIENCE in THE F A C U L T Y OF G R A D U A T E STUDIES (Department of Microbiology and Immunology) We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH C O L U M B I A June 2004 © Terumi Anne Taylor, 2004  Library Authorization  In presenting this thesis in partial fulfillment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.  Name of Author (please print)  Title of Thesis:  Degree:  AK  Departmentof  & / Q  Datef (dd/mm/yyyy)  (uj~)G^ (hnck fUiC^ jq ~f}py\  Year:  £ ^ AM  '  (Tob)  G  The University of British Columbia Vancouver, BC Canada  D.QO  ^  11  ABSTRACT  The gene transfer agent (GTA) of Rhodobacter capsulatus is an unusual entity that resembles a bacteriophage (phage), but does not appear to cause cell lysis and does not form plaques. G T A packages random 4.5 kb fragments of double-stranded bacterial genomic D N A and is capable of transducing recipient R. capsulatus cells. G T A has mainly been used as a mechanism to create genetic mutants, however little about the evolutionary history and regulation has been investigated, considering its relevance as a model of unusual viral-bacterial interactions. This work describes a phylogenetic analysis of G T A genes that indicates predominantly vertical descendence and limited horizontal transfer between related viruses. Investigation into the regulation of G T A revealed a link to quorum sensing through the discovery of the gtal gene. The Gtal protein is a Luxl homologue that acts as an acyl-homoserine lactone synthase. When gtal is knocked out, G T A activity is reduced significantly (5-7 times less activity). Also described is the creation of a tool to measure G T A activity that is independent of the packaging, transfer and incorporation of genetic markers that is typically used in the investigation of G T A activity, also known as the bioassay. Instead, an immunoassay that requires less time to perform (two days instead of four) was created using a tagged G T A capsid protein as an antigen to obtain antisera. This G T A immunoassay was tested and a standard protocol is described for optimal use in R. capsulatus strains. The immunoassay was employed for preliminary investigation of environmental factors affecting G T A production. The results show that under phosphate limitation in a minimal medium, G T A production increases to levels similar to the rich medium YPS. It was also found that G T A production increased in response to carbon limitation in a minimal medium, but this appeared to affect only intracellular  amounts  of G T A , with  little release  into  the  culture  medium.  iii T A B L E OF CONTENTS LIST OF TABLES  vi  LIST OF FIGURES  vii  ABBREVIATIONS  viii  ACKNOWLEDGEMENTS  ix  1. INTRODUCTION  1  1.1  D N A exchange  1  1.2  Transduction and G T A ofR. capsulatus  1  1.3  Several G T A gene homologues and relationships to R. capsulatus  3  1.4  Regulation of G T A  3  1.5  Quorum sensing and G T A production  6  1.6  G T A and R. capsulatus  6  2. PHYLOGENETIC ANALYSIS OF GTA  8  2.1  Introduction  8  2.2  Materials and Methods  8  2.3  2.4  2.2.1 Sequence analyses  8  2.2.2 Multiple sequence alignments  9  2.2.3 Tree construction and presentation  9  Results  10  2.3.1  GTA-related genes  10  2.3.2  G T A structural proteins  10  2.3.3  G T A regulatory proteins  11  2.3.4  16S rRNA  17  Discussion  3. QUORUM SENSING AND GTA  19 20  3.1  Introduction  20  3.2  Materials and Methods  20  3.2.1  Detection of R. capsulatus acyl-HSLs  20  3.2.2  Identification of acyl-HSLs  21  3.2.3  Chemical synthesis of hexadecanoyl H S L  21  iv 3.2.4  Identification and genetic analysis of the R. capsulatus acyl-HSL synthase  3.2.5  Assessment of gene transfer agent transcription and production  3.3  3.4  21  Results  21  22  3.3.1  R. capsulatus synthesizes a C16-HSL  22  3.3.2  The R. capsulatus acyl-HSL synthase gene  22  3.3.3  Induction of R. capsulatus G T A by C16-HSL  23  Discussion  25  4. GTA IMMUNOASSAY AND EPIFLUORESCENCE MICROSCOPY ASSAY  26  4.1  Introduction  26  4.2  Materials and Methods  28  4.3  4.2.1  Bacterial strains, growth conditions and plasmids  28  4.2.2  Plasmid D N A preparation and manipulation  28  4.2.3  Construction of protein over expression plasmids  28  4.2.4  D N A sequencing  32  4.2.5  Protein over expression  32  4.2.6  SDS polyacrylamide gel electrophoresis (SDS PAGE)  34  4.2.7  Protein purification using N i - N T A columns  34  4.2.8  Protein purification using A K T A - F P L C  34  4.2.9  Polyclonal antibody production  35  4.2.10 Cross-absorption of antibodies  35  4.2.11 Western blotting  35  4.2.12 Protein sequencing and mass spectrometry  36  4.2.13 Epifluorescence microscopy using SYBR-green  36  4.2.14 G T A bioassays (transfer ofpuhA gene)  37  4.2.15 Nutrient limitation  37  Results  38  4.3.1  Optimization of overproduction of tagged capsid proteins  38  4.3.2  Purification of tagged capsid proteins  40  4.3.3  Confirmation of identity of purified capsid proteins  40  4.3.4  Anti-capsid polyclonal antibodies  43  4.3.5  G T A immunoassay using anti-capsid antibodies  46  4.3.6  Environmental effects on G T A production  4.3.7 4.4  using the G T A immunoassay  46  Epifluorescence microscopy and particle counts  50  Discussion 4.4.1  50  G T A immunoassay as a tool to investigate G T A production is practical and functional with some limitations  50  4.4.2  Resolving the extra bands  53  4.4.3  Implications of preliminary and environmental studies on G T A production  55  4.4.4  Epifluorescence microscopy and viral-like particle counts  57  4.4.5  Future uses of the G T A immunoassay  58  5. CONCLUSION  60  REFERENCES  61  APPENDIX 1  67  Publications arising from this thesis research  vi LIST OF TABLES Table 1:  Percent identities in alignments between R. capsulatus G T A structural proteins and homologues  Table 2:  12  Percent identities in alignments between and protein lengths of R. capsulatus CtrA and CckA proteins and homologues in other a-proteobacteria  Table 3:  13  Effects of acyl-HSL on G T A transcription and transducing particle production  24  Table 4:  Plasmids and strains  29  Table 5:  Primers designed and used  33  vii LIST OF FIGURES Figure 1:  Regulatory components of G T A in R. capsulatus  Figure 2:  Map of G T A structural genes of R. capsulatus and homologues  Figure 3:  14  Phylogenetic analyses of G T A structural proteins and homologues  Figure 4:  4  15  Phylogenetic analyses of G T A regulatory proteins and homologues  16  Figure 5:  Phylogenetic analyses of 16S rRNA sequences  18  Figure 6:  Amino acid sequence of ORFg5, G T A capsid protein  27  Figure 7:  Cloning scheme to create tagged ORFg5 protein  30  Figure 8:  SDS P A G E of initial over production of tagged proteins  39  Figure 9:  SDS P A G E of primary purification (Ni-NTA column) of tagged capsid proteins  Figure 10:  41  SDS P A G E of secondary purification of mid-tagged capsid protein using a Resource Q column in FPLC (used for antibody production)  Figure 11:  Mass spectrometry and N-terminal sequence of the purified mid-tagged capsid protein  Figure 12:  48  Western blots of environmental effects on G T A production  Figure 15:  45  Western blot of time course of G T A production in a Y262 culture grown in YPS medium  Figure 14:  44  Western blots of final antibodies and "clean" antibodies from last bleed  Figure 13:  42  49  Epifluorescence microscopy particle counts compared to G T A transduction (bioassay)  51  Figure 16:  Potential proteolysis sites of ORFg5 expressed inE. coli.  54  Figure 17:  Growth curve of nutrient limited samples  56  ABBREVIATIONS  aa  amino acids  Ap(Ap )  ampicillin (ampicillin resistance)  BSA  bovine serum albumin  DMSO  dimethyl sulphoxide  EPM  epifluorescence microscopy  FPLC  fast performance liquid chromatography  GTA  gene transfer agent  HSL  homoserine lactone  IPTG  isopropyl-/3-D-thiogalactopyranoside  kb  kilobase pair  kDa  kilodalton  R  Km(Km )  kanamycin (kanamycin resistance)  KU  Klett unit  MOPS  3-(N-morpholino)propanesulfonate  MS  mass spectrometry  Ni-NTA  nickel-nitrilotriacetic acid  OD  optical density  ORF  open reading frame  phage  bacteriophage  RCV  Rhodobacter capsulatus minimal growth media  RifCRif*)  rifampicin (rifampicin resistance)  SDS P A G E  sodium dodecyl sulfate polyacrylamide gel electroph  Sp (Sp )  spectinomycin (spectinomycin resistance)  Tc (Tc )  tetracycline (tetracycline resistance)  YPS  yeast-peptone salt (rich media)  R  R  R  ix ACKNOWLEDGEMENTS First, I thank my supervisor; Dr. J.T. Beatty. I thank all past and present members of the Beatty Lab, especially Jeanette Beatty and Andrew Lang. Members of the Department of Microbiology and Immunology have also been helpful, thank you.  1 INTRODUCTION  1.1 DNA exchange.  The exchange of D N A among bacteria is important in bacterial evolution. Recent estimates indicate that approximately 18 percent of the Escherichia coli open reading frames (ORFs), many of which are known genes, were introduced by lateral transfer events (Lawrence and Ochman 1998).  D N A is exchanged by transformation,  conjugation, and transduction. Transformation is the uptake of naked D N A from the environment.  Conjugation is cell-to-cell transfer of plasmids or transposons, and  sometimes, chromosomal D N A i f mobilization elements are present in the chromosome. Lastly, transduction entails D N A exchange from one cell to another via a bacteriophage (phage) particle (Levy and Miller 1989; Mazodier and Davies 1991). Most bacterial cells are susceptible to phage infection. Phage consist of proteincoated nucleic acids and have two methods of replication. The virulent phage life cycle is usually characterized by host cell lysis after phage nucleic acid replication and packaging.  The other life cycle is lysogeny, in which phage D N A can be stably  maintained in the bacterial host genome in the form of a prophage. The prophage genes are replicated along with the host chromosome, either integrated (X phage) or as a separate molecule (PI phage). At some point, a prophage may be activated and enter into the lytic life cycle. Transduction occurs when part of the host bacterium's chromosome is packaged in a phage particle as a rare event during phage replication. Subsequently, injection of this D N A into a recipient cell may introduce new genes or allelic variants of existing genes (Kokjohn 1989).  1.2 Transduction and the gene transfer agent (GTA) oiR. capsulatus.  A novel D N A exchange system called the gene transfer agent was discovered in the purple non-sulphur bacterium Rhodobacter capsulatus (Marrs 1974). R. capsulatus G T A resembles a small tailed phage with a head diameter of approximately 30 nm. G T A  2 packages genomic D N A in an apparently random manner as linear double-stranded pieces of approximately 4.5 kb (Solioz and Marrs 1977). G T A has no detectable cell lysing ability and thus does not produce plaques (Solioz et al. 1975; Yen et al. 1979). A structural gene cluster of G T A has been found and characterized (Lang and Beatty 2000). The sole activity of G T A appears to be to transfer genes. However, some strains of R. capsulatus are only able to produce G T A or to take up D N A from G T A , and some strains neither produce G T A nor take up GTA-borne D N A (Marrs 1974; Wall et al. 1975). The first studies performed with G T A were limited by the low levels of G T A production (approximately 10 gene transfer units/ml) (Solioz 1975). The isolation of a 5  G T A overproducer mutant that produces G T A at approximately 1000 times the level of the wild type strain greatly advanced G T A studies (Yen et al. 1979). The purification of particles from the overproducer strain allowed visualization using electron microscopy of the phage-like particle involved in the gene transfer activity (Yen et al. 1979). G T A activity is also enhanced under photosynthetic conditions as opposed to aerobic growth. Furthermore G T A production is greater in rich media than in minimal media and peaks as cultures reach stationary phase (Solioz 1975; Yen et al. 1979). GTA-like entities are significantly different from phage. There have been four other GTA-like entities discovered in other prokaryotes. A l l package double-stranded near random bacterial D N A , have a tailed-phage structure, and seem to package less D N A than would be expected to code for a complete phage genome. The first is found in Desulfovibrio desulfuricans, named D d l , and packages and transduces random 13.6 kb fragments (Rapp and Wall 1987). The next one found in Brachyspira (Serpulina) hyodysenteriae is named VSH-1, and appears to package and transduce random 7.5 kb chromosomal pieces (Humphrey et al. 1997). A third GTA-like entity named V T A was discovered in Methanococcus voltae and appears to transduce random 4.4 kb genomic D N A fragments (Bertani 1999; Eiserling et al. 1999). Fourth, Bartonella species bacilliformis and henselae produce GTA-like particles (BLPs) that package and transduce 14 kb chromosomal D N A fragments (Barbian and Minnick 2000).  3  1.3 Several GTA gene homologues and their evolutionary relationships to R. capsulatus.  G T A structural gene-like clusters have been found in: Rhodopseudomonas palustris, Rhodobacter sphaeroides, Caulobacter crescentus, Agrobacterium tumefaciens and Brucella melitensis (Lang et al. 2002). In addition, ctrA and cckA homologues (see section 1.4) were found in all of these species as well as other a-proteobacteria: Mesorhizobium loti and Sinorhizobium meliloti. Our phylogenetic analyses suggest a predominantly vertical descendence of G T A genes in the o>proteobacteria and limited horizontal transfer between related viruses (Lang et al. 2002).  1.4 Regulation of GTA.  Two proteins were first implicated in regulation of G T A transcription: CtrA and CckA (Lang and Beatty 2000; Lang and Beatty 2001). Although the details of how they work are not known, both of these proteins are thought to function as part of a cellular sensor kinase/response regulator system that controls the transcription of G T A structural genes (Lang and Beatty 2000; Lang and Beatty 2001) (Figure 1). The response regulator CtrA is encoded by the gene ctrA that was first discovered in Caulobacter crescentus, where it is an essential protein required for transcriptional control of D N A replication, D N A methylation, cell division, flagellar biosynthesis, and pilus biosynthesis genes (Domian et al. 1997; Quon et al. 1996; Skerker and Shapiro 2000). The R. capsulatus ctrA homologue was discovered as an ORF disrupted by transposon insertion, and the predicted protein is 71% identical to C. crescentus CtrA (Lang and Beatty 2000). R. capsulatus ctrA is present in only one copy, is not essential for cell division, and when knocked out cannot be rescued by the C. crescentus ctrA. The phenotype of the R. capsulatus ctrA' strain is loss of G T A production and loss of motility (Lang and Beatty 2002). It appears that CtrA is required to activate transcription of the G T A genes, motility genes and perhaps other genes as well (Lang and Beatty 2001; Lang and Beatty 2002). CtrA positively regulates G T A structural and flagellar genes  4  CckA  — n t h (putative histidine kinase, membrane bound dimer)  Gene Repression  CtrA  (putative transcription regulator! soluble) ...-'  Gene Induction  t  -  -Flagellar genes  Type III restriction subunit Hypothetical proteins (next to putative phage related genes)  -GTA genes  Gtal  Figure 1: Regulatory components of G T A in R. capsulatus. CckA is a putative autophosphorylating membrane bound histidine kinase that is thought to phosphorylate CtrA, a putative soluble signal regulator that is known to regulate G T A genes in R. capsulatus. Other regulators include the autoinducer synthase of a LuxIR-type of quorum sensing system, produced by Gtal (Schaefer et al. 2002). Additional possible regulators are a Type III restriction-modification system restriction subunit and a hypothetical protein found next to phage related genes (S. Florizone, personal communication). The question mark indicates the possible presence of another regulator that is capable of activating/phosphorylating CtrA.  5 (Lang and Beatty 2000; Lang and Beatty 2001; Lang and Beatty 2002). It appears that CtrA may not be a direct effector of G T A transcription, as a putative CtrA DNA-binding sequence has not been found upstream of the G T A gene cluster. CtrA may, in principle, either positively or negatively regulate the transcription of other genes. One possible missing link could be a CtrA-induced R N A polymerase sigma factor that promotes transcription of G T A genes. The gene for the putative histidine kinase cckA also has an effect on motility and G T A transcription. In a cckA  strain, the negative effect on motility and G T A  transcription observed is approximately one-tenth of the effect seen in the ctrA strain (Lang and Beatty 2002). Thus, assuming that the function of CckA is to activate CtrA, there may be an alternative pathway for CtrA activation that functions at approximately 1/10 the level of the CckA pathway. Other genes required for G T A gene expression are currently being investigated. A transposon library constructed by A . Lang has been used to search for genes required for G T A gene expression. This is the same library that was used initially to implicate CckA in G T A gene expression (Lang and Beatty 2000). So far, the screen of this transposon library has yielded 15 candidates that have a transposon insertion in a gene potentially required for G T A gene expression (S. Florizone, personal communication). The initial cloning and sequencing of five candidates has been successful. One of the disrupted genes is predicted to encode a Type III restriction-modification system restriction enzyme subunit, which is located between genes annotated as encoding a Type III restrictionmodification system restriction subunit and a recombinase. The second disrupted gene encodes a hypothetical protein that is amongst genes encoding phage-homologous and hypothetical phage proteins. A third transposon insertion that reduces G T A production was found to be located in ORF 142 that is a weak homologue of cckA genes, and located near ORF 139 that is weakly similar to an autoinducer synthase (cerl) and an ORF that is clearly an acyltransferase homologue. It is possible that ORF 142 (cckA-Yiko) and perhaps the adjacent ORF 139 (cer/-like) encode proteins involved in either the quorum sensing system described below, or in a parallel regulatory pathway. The other candidates are annotated  to code for an omega-amino  acid-pyruvate aminotransferase,  and a  6  hypothetical  cytosolic protein with  glutamyl-tRNA  aminotransferase  subunit  B  immediately downstream.  1.5 Quorum sensing and GTA production.  Quorum sensing is the control of gene expression in response to cell density, and is used by many bacterial species to regulate a variety of cellular functions (Fuqua et al. 1994). Quorum sensing involves the production and detection of extracellular signalling molecules called autoinducers. In Gram-negative bacteria, communication is commonly based on homologues of the LuxI/LuxR signal-response system. This system includes the synthesis of an acyl-homoserine lactone (acyl-HSL) autoinducer by a Luxl homologous enzyme, as well as a luxR homologue encoding a transcriptional activator that induces genes which encode proteins responsible for cell responses to acyl-HSL (Fuqua et al. 1994). In R. capsulatus the luxI/luxR homologues were found by a B L A S T search (Altschul et al. 1997) and a knockout of the luxl gene was made (Schaefer et al. 2002). This knockout abolished the production of a novel, long-chain acyl-HSL, but no other phenotype was detected (Schaefer et al. 2002). Because G T A production peaks in the stationary phase (Solioz et al. 1975), when quorum sensing responses are exhibited (Fuqua et al. 1994), we thought that G T A production might be controlled by the R. capsulatus acyl-HSL. Indeed, in collaborative experiments with A . Schaefer and E.P. Greenberg, we discovered that the quorum sensing knockout (gtal') strain had decreased production of G T A , and decreased expression of /3-galactosidase  encoded by a  GTA::/acZ gene fusion [pYP; (Lang and Beatty 2000)]. G T A production and GTA: :lacZ expression in the gtal' strain supplemented with acyl-HSL were returned to wild type levels (Schaefer et al. 2002).  1.6 GTA and R. capsulatus.  R. capsulatus is a purple non-sulphur a-proteobacterium that is capable of growth under a variety of conditions including aerobically "by respiration and anaerobically by  7 photosynthesis, and is able to fix nitrogen and carbon dioxide (Madigan 1995; Tabita 1995). These properties of R. capsulatus make it ideal for photosynthesis research, as mutations that abolish photosynthesis are not lethal. G T A was and still is significant in applications to R. capsulatus genetics research, as G T A transduction allows manipulations that would otherwise require transformation that is not possible in R. capsulatus. G T A is exploited to perform gene replacement by homologous recombination, thus allowing specific mutations and knockouts without the difficulties associated with conjugative suicide plasmids (Scolnik and Haselkorn 1984). The evolutionary origin of G T A has been postulated to be either as precursor phage that has yet to attain independence, or as a defective phage that has been trapped and taken over by cellular regulatory mechanisms (Lang and Beatty 2001). This thesis investigates the phylogeny of G T A and a quorum sensing regulatory component that controls G T A production. In addition, a new tool to look at G T A production was created to enable new ways to study G T A under a variety of conditions in a reproducible and sensitive manner.  8 2. PHYLOGENETIC ANALYSIS OF GTA  2.1 Introduction.  The gene transfer agent (GTA) of the a-proteobacterium Rhodobacter capsulatus is a cell-controlled genetic exchange vector. Genes that encode the G T A structure are clustered in a 15 kb region of the R. capsulatus chromosome, and some of these genes show sequence similarity to known bacteriophage head and tail genes (Lang and Beatty 2000). However, the production of G T A is controlled at the level of transcription by a cellular two-component signal transduction system (Lang and Beatty 2000). This section describes homologues of both the G T A structural gene cluster and the G T A regulatory genes in the a-proteobacteria Rhodopseudomonas palustris, Rhodobacter sphaeroides, Caulobacter crescentus, Agrobacterium tumefaciens and Brucella melitensis. These sequences were used in a phylogenetic tree approach to examine the evolutionary relationships of selected G T A proteins to these homologues and phage proteins, which was compared to a 16S rRNA tree. The data indicate that a GTA-like element was present in a single progenitor of the extant species that contain both G T A structural cluster and regulatory gene homologues. The evolutionary relationships of G T A structural proteins to phage proteins indicated by the phylogenetic tree patterns suggest a predominantly vertical descent of GTA-like sequences in the Q!-proteobacteria and little past gene exchange with phage. This is in contrast to phylogenetic studies of doublestranded D N A phage, which indicated extensive lateral exchange of genes (Casjens et al. 1992; Juhalaetal. 2000).  2.2 Materials and Methods.  2.2.1 Sequence analyses. Genome  sequence  data  for  R.  http://wit.IntegratedGenomics.com/IGwit/.  capsulatus  were  obtained  from  Preliminary sequence data for R. palustris  and R. sphaeroides were obtained from The D O E Joint Genome Institute (JGI) at http://spider.jgi-psf.org/JGI_microbial/html/, and preliminary annotation was obtained  9 from http://compbio.ornl.gov/channel/. Genome sequence data for Methanothermobacter marburgensis [formerly Methanobacterium thermoautotrophicum Marburg (Wasserfallen and Nolling J 2000)], Escherichia coli, Streptomyces coelicolor, B. melitensis, A. tumefaciens  and  M.  loti  were  http://www.ncbi.nlm.nih.gov/PMGifs/Genomes/micr.html.  obtained  from  The values of  percent identity in Tables 1 and 2 were calculated from alignments made with the program Geneworks (Oxford Molecular, Campbell, CA), such that the % identities are relative to R. capsulatus sequences.  2.2.2 Multiple sequence alignments. Each protein sequence group was aligned using C L U S T A L X vl.81 (Thompson and Gibson TJ 1997).  The Gonnet weight matrix (Gonnet et al. 1994) was used in the  sequence alignments. The 16S rRNA sequence data were aligned using C L U S T A L X vl.81 (Thompson and Gibson TJ 1997). The complete sequence alignments are available from the authors (Lang et al. 2002) upon request.  2.2.3 Tree construction and presentation. The protein sequence alignments were transformed into maximum likelihood distances using T R E E - P U Z Z L E v5.0 (Strimmer and A v 1996) and 10000 "puzzling" steps. The multiple substitution matrix used was BLOSUM62 (Henikoff and JG 1992). Trees were plotted using NJplotWIN95 (Perriere and M 1996). For comparison, unrooted neighborjoining bootstrapped trees were constructed using C L U S T A L X vl.81 (Thompson and Gibson TJ 1997) and plotted using NJplotWIN95 (Perriere and M 1996). The 16S rRNA alignments were also plotted as above using the neighbor-joining algorithm. Use of the Phylip Interface at the Ribosomal Database (Maidak and Cole JR 2001) gave the same topology.  10 2.3 Results.  2.3.1  GTA-related genes.  B L A S T searches (Altschul et al. 1997) of sequence databases were used to identify G T A gene homologues and GTA-like structural gene clusters. GTA-like gene clusters were found in the a-proteobacteria Caulobacter crescentus, Rhodobacter sphaeroides, Brucella melitensis, and Agrobacterium tumefaciens. G T A homologues were previously found in Rhodopseudomonas palustris (Lang and Beatty 2001)(Figure 2). In addition, G T A regulatory homologues of ctrA and cckA were found in R. palustris, C. crescentus, R. sphaeroides, B. melitensis, and A. tumefaciens. Other aproteobacteria, such as Mesorhizobium loti and Sinorhizobium meliloti, have cckA/ctrA homologues, but do not have identifiable clusters of GTA-like structural gene homologues. Only R. sphaeroides and A. tumefaciens have all of the homologues of known R. capsulatus G T A genes. The most significant similarity to genuine phage genes is to phage head and tail structural genes. Lengths and percentage identities of R. capsulatus G T A structural protein sequences and the homologues are presented in Tables 1 and 2.  2.3.2  GTA structural proteins.  Evolutionary relationships between R. capsulatus G T A proteins and the homologous amino acid sequences encoded by GTA-like gene clusters and (pro)phage were evaluated using phylogenetic trees. The R. capsulatus G T A putative large terminase, portal and major capsid protein sequences form distinct clusters with the other Q!-proteobacterial G T A homologues, and always branch closest to the homologues from R. sphaeroides (Figure 3). Interestingly, the homologues from R. capsulatus putative prophage 0 R c P l and 0 R c M l group more closely to genuine phage proteins than to the GTA-like sequences in other species (Figure 3). This implies an evolutionary difference between G T A and the prophage 0 R c P l and </>RcMl, although the G T A sequences and the two prophage are carried on the R. capsulatus genome.  11 2.3.3  GTA regulatory proteins.  The relationships indicated by the phylogenetic analyses of the G T A regulatory proteins, CckA and CtrA are very similar to the phylogenetic analyses of the G T A structural proteins (compare Figures 4 and 3).  12 Table 1: Percent identities in alignments between R. capsulatus G T A structural proteins and homologues. R. capsulatus G T A protein  R. sphaeroides  Orfgl Orfg2 Orfg3 Orfg4 Orfg5 Orfg6 Orfg7 Orfg8 Orfg9 Orfgl 0 OrfglO-10.1 Orfgl 1 Orfgl 2 Orfgl 3 Orfgl4 Orfgl 5  57 75 68 61 73 39 58 56 74 37 48 66 53 48 62 58  c  Percent identity with homologue from: R. palustris C. crescentus A. tumefaciens a  a  54 50 40 42 21 25 32 56 44 38 37 42 38 43 39  51 51 43 42 a a a  53 31 32 27 47 26 44 43/29 b  d  21 46 45 43 42 20 25 26 50 38 32 41 47 42 51 28  B. melitensis a  53 48 a a a a a  54 39 31 36 34 31 46 40/25  Dash indicates that sequence was not found. Because this sequence has no start codon, a sequence of the same length as the G T A Orfgl4 was used for the alignment. The available sequences from the two ends of this ORF were combined and aligned with the analogous regions from the G T A Orfgl5 sequence. Because there is a frameshift in the C. crescentus and B. melitensis orfgl 5 sequences, the two values reported are from alignments of the two reading frames with the corresponding regions of the R. capsulatus G T A Orfgl 5 sequence. a  b  c  d  d  Table 2: Percent identities in alignments between and protein lengths of/?, capsulatus CtrA and CckA proteins and homologues in other a-proteobacteria. Species R. capsulatus R. sphaeroides R. palustris C. crescentus M. loti B. melitensis A. tumefaciens S. meliloti  length (aa) length (aa) % identity length (aa) % identity length (aa) % identity length (aa) % identity length (aa) % identity length (aa) % identity length (aa) % identity  CtrA  CckA  237 237 92 233 74 231 71 231 75 232 75 234 76 233 76  774 (376, 391) 766 (382, 384) 59 (46, 70) 882 (485, 397) 27 (14, 48) 691 (299, 392) 26(15, 42) 898 (505, 393) 29 (10, 50) 767 (376, 391) 28 (11, 51) 861(460, 401) 31 (14, 47) 869 (466, 403) 30(18, 47)  a  a  b  a  b  a  b  a  b  a  b  a  b  a  b  The lengths of just the N-terminal and C-terminal regions, respectively, are given in brackets. The values of percent identity for alignments of just the N-terminal and C-terminal regions, respectively, are given in brackets. a  14  A. R. capsulatus (Rhodobacter sphaeroides is identical) temuiiase  portal  1 kb  10.1  capsid  12  10  B.  15  14  Rhodopseuclonumaspalustris portal  5  1  10 10.1  capsid"  F  13  C.  13  1  1  2  «7 8 5  11  13 |l4  1%  n 1  Caulobacter crescentus tcnniiust  1  15  1 kb  portal  10  cipsid _4~  5  •  9  10.1  Li  nostartcodon  11 •  12  13  14  fiameshift 15  15  D. Agrobacterimn tumefaciens i  E.  terminue  portal  2  3  1  Brucella melitensis ttrminas* 2  6 1  5  8  9 10  10.1  port .a] 3  4  B  11  12  13  14  15  ftameshift  I 9 10  1 kb  10.1  c&psid  1 kb 11  12  13  14  15  15  Figure 2: Maps of G T A structural genes of R. capsulatus and homologues. Boxes above the line are ORFs transcribed from left to right, whereas boxes below the line are ORFs transcribed from right to left. (A) Map of R. capsulatus G T A structural gene cluster (identical to R. sphaeroides); ctrA is approximately 15 kb 5' (transcriptionally upstream); cckA is approximately 55 kb 3' (transcriptionally downstream). (B) Map of the R. palustris G T A structural gene homologues. (C) Map of the C. crescentus G T A structural gene homologues. (D) Map of the A. tumefaciens G T A structural gene homologues. (E) Map of the B. melitensis G T A structural gene homologues.  kb  15  A. Terminase  100  .0.2  B. Portal  r- FM2 ( M marburgensis) <  4>ML1 (M. loti) 4>C31 (S. coelicolor) HK97 (E. coli) <(>RcPl {R. capsulatus) <()RcMl (R. capsulatus) I (|>ML2(Af. loti) | r ^ N E (M. marburgensis) i w l ^ M l O O (Af. wolfeii) i—C. crescentus A . tumefaciens B. melitensis R. palustris [_r-iR. sphaeroides iooL G X A capsulatus)  WmIOO ( M  wo//«0  dpRcMl (R. capsulatus) <j)RcPl (/?. capsulatus) -<J»ML1 (M.tofi) -<(>C31 (5. coelicolor) -HK97 (E. coii)  0.2  A . tumefaciens B. melitensis C. crescentus R. palustris \_r-R. sphaeroides G T A ( « . capsulatus)  C. Capsid  -HK97 (£ co/0 -<>RcPl (K. capsulatus) -<|>RcMl (R. capsulatus) -<>C31 (S. coelicotor) • $MLl(Af./orf) • C. crescentus -A. tumefaciens • R. palustris -R. sphaeroides i « U G T A {R. capsulatus)  Figure 3: Phylogenetic analyses of G T A structural proteins and homologues. GTA-like sequences are in bold face. A . Maximum likelihood tree for large terminase proteins. B. Maximum likelihood tree for portal proteins. C. Maximum likelihood tree for major capsid proteins. Support values are shown as percentages based on 10000 puzzling steps. The scale bar represents the expected number of changes per residue position. See (Lang et al. 2002) for sequence sources and methodology.  16  B. CtrA  A. CckA M. marburgensis S. coelicolor — E. coli j—M. loti V—B. melitensis  69  9?  I r l tumefaciens 84 " L- 5. metiloti ' if. palustris C. crescentus —K.R. .sphaeroides —R. capsulatus  M. marburgensis -S. coelicolor — E. coli M. loti 95. hB. melitensis A. tumefaciens  |ioo  3  95  621 0.2  J  S. palustris meliloti m *-R. — C . crescentus 1001 I t *R. sphaeroides ifeL^ capsulatus 1  0.1  Figure 4: Phylogenetic analyses of G T A regulatory proteins and homologues.  A.  Maximum likelihood tree for CckA proteins. B. Maximum likelihood tree for CtrA proteins. Support values are shown as percentages based on 10000 puzzling steps. The scale bar represents the expected number of changes per residue position. Sequence sources and methodology described in (Lang et al. 2002).  17 2.2.4 16S rRNA. Neighbour-joining trees of the 16S rRNA sequences, a generally accepted method to infer cellular evolutionary relationships (Woese 1987) was used as a control for the previously described phylogenetic analyses. The branching of the 16S rRNA tree is similar to the CckA and CtrA trees (Figure 5). This similarity supports the idea that cckA, ctrA and 16S rRNA all give similar representations of cellular evolution.  18  •hi marlmrg^nsis coelicolor Ecbli dcrescentus 'R, palustBis B. melitensis A: tumefaciens -M.loti S. mjzltloti & sphaeroides Kcapsulatus  Figure 5: Phylogenetic analyses of 16S rRNA sequences. Tree construction based in neighbour-joining with bootstrap values as percentages based on 1000 iterations. The scale bar represents the expected number of changes per base position. Accession numbers of sequences used are available in (Lang et al. 2002).  19 2.4 Discussion.  These phylogenetic analyses indicate that an ancestral form of the R. capsulatus G T A structural gene cluster, and homologues in R. palustris, C. crescentus, R. sphaeroides, B. melitensis and A. tumefaciens, was present in an evolutionary ancestor of these species. This suggests that these GTA-like gene clusters could be non-functional remnants of a genetic exchange element that has retained its functionality in R. capsulatus. Alternatively, a non-functional "proto-GTA" may have existed in an ancestor, with development of gene transfer capabilities in R. capsulatus. The phylogenetic analyses of the CckA and CtrA regulatory proteins that contain GTA-like structural gene clusters are similar to the 16S rRNA. As noted above, this indicates that the cckA and ctrA regulatory genes are descended from a single evolutionary ancestor. This indicates that the association between the bacterial regulatory components of R. capsulatus and G T A has had a long evolutionary history in which to become integrated. This supports my view that there was a long time that allowed evolution of a complex cascade of regulatory factors involved in the regulation of G T A in R. capsulatus. In conclusion, it can be said that G T A resembles a defective prophage. The phylogenetic analyses indicate the vertical transmission of a GTA-like entity from an aproteobacterial ancestor, that has resulted in the homologous sequences that are retained by R. capsulatus, R. palustris, C. crescentus, R. sphaeroides, B. melitensis and A. tumefaciens. If G T A genes were transferred laterally, it would have been only prior to the divergence of the a-proteobacteria, possibly when an ancestral prophage was active as a virus.  20 3. QUORUM SENSING AND GTA  3.1 Introduction.  Rhodobacter capsulatus produces a novel genetic exchange element called GTA. which is produced in greatest numbers when cells are in the stationary phase (i.e., high cell density) (Yen et al. 1979). Many proteobacteria use acyl-homoserine lactones (acylHSLs) as quorum-sensing signals (Whitehead et al. 2001). However, long-chain acylHSLs are not always readily detected by standard bioassays. This section describes a nontraditional, more sensitive approach that did not require a bioassay to detect production of long-acyl-chain H S L production by R. capsulatus. The most abundant acyl-HSL in R. capsulatus was N-hexadecanoyl-HSL (C16-HSL). The long-chain acyl-HSLs were concentrated in cells but were also found in the culture fluid. A n R. capsulatus gene responsible for long-chain acyl-HSL synthesis was identified. A mutation in this gene, which we named gtal, resulted in decreased production of the R. capsulatus GTA. G T A production was increased by addition of exogenous C16-HSL. Thus, long-chain acylHSLs serve as quorum-sensing signals to enhance genetic exchange in R. capsulatus.  3.2 Materials and Methods.  3.2.1  Detection of if. capsulatus acyl-HSLs.  Prior discovery of acyl-HSLs has relied on bioassays, which are limited by signal specificity constraints (Puskas et al. 1997). Novel acyl-HSLs might not be detected by any of the available bioassays. A n assay was adapted to detect the incorporation of 14Clabel into acyl-HSLs (Piper et al. 1993) to screen R. capsulatus for acyl-HSL production regardless of whether the molecules can be readily detected in any bioassay. The method of (Parsek and Greenberg 2000)was followed. Four ml of scintillation cocktail 3a70b (Research Products Inc.) were added to each of the seventy 1-ml fractions collected and radioactivity was determined by scintillation counting.  3.2.2  Identification of acyl-HSLs.  21 To identify acyl-HSLs we extracted late logarithmic-phase culture fluid with acidified ethyl acetate. The acyl-HSLs were purified from extracts by using C18-reverse phase H P L C as described above. The purified material was analyzed by chemical ionization mass spectrometry (CI-MS) with a V G Trio-1 quadropole mass spectrometer using methane as the reagent gas.  3.2.3  Chemical synthesis of hexadecanoyl HSL.  C16-HSL was synthesized in a manner similar to that described elsewhere (BlosserMiddleton and Gray 2001).  3.2.4  Identification and genetic analysis of the R. capsulatus acyl-HSL synthase.  We  performed  a  BLAST  search  (http://ergo.integratedgenomics.com/ERGO;  of  the  Integrated  R.  capsulatus  Genomics)  for  genome translation  products showing similarity to the Rhodobacter sphaeroides Cerl protein (accession number AAC46022). One putative R. capsulatus gene, ORF RRC03805, coded for a polypeptide that showed significant similarity to Cerl and was adjacent to a gene coding for a probable LuxR-type regulatory protein, ORF RRC03806. Construction of the R. capsulatus acyl-HSL knockout, ALS1 by A . Schaefer is described in (Schaefer et al. 2000).  3.2.5  Assessment of gene transfer agent transcription and production.  Plasmid pYP, which contains the G T A promoter, orfg2 and an in-frame translational fusion of the G T A structural gene orfg2 to lacZ (Lang and Beatty 2002), was introduced into R. capsulatus strains SB 1003 (Yen and Marrs 1976) and ALS1 by conjugation. Exconjugants were grown phototrophically in YPS (Wall et al. 1975) at 30°C and /3galactosidase activity was measured in stationary phase cells (-22 h) (Solioz et al. 1975). As a control, similar experiments were performed using pYNP, which contains a promoterless orfg2::lacZ fusion (Lang and Beatty 2002). With pYNP /3-galactosidase activity was less than 0.1 unit.  22 Transducing particle production was measured as GTA-mediated transfer of a Rif* marker from the G T A donor strains (SB 1003 or ALS1) to the Rif GTA-recipient 6  strain [BIO, (Marrs 1974)] as described previously (Lang and Beatty 2002).  R.  capsulatus SB 1003 and ALS1 were grown as described above and synthetic C16-HSL (2 mM) was added to cultures where indicated.  3.3 Results.  3.3.1  R. capsulatus synthesizes a C16-HSL.  Acyl-HSL discovery has relied on detection with bioassays that use heterologous reporter constructs (Schaefer et al. 2000). A major limitation of this bioassay is that it is biased towards detection of acyl-HSLs close enough to the natural signal (i.e., of similar molecular structure) to be recognized by previously known LuxR homologues. A radiotracer technique that overcomes the limitations of the bioassay was developed to detect acyl-HSLs of novel chain length or configuration (Schaefer et al. 2000). This new radiotracer assay identified a C16-HSL that was produced by R. capsulatus (Schaefer et al. 2002).  3.3.2  The R. capsulatus acyl-HSL synthase gene.  Bioassays for quorum sensing signalling in R. capsulatus yielded negative results, despite the closely related Rhodobacter sphaeroides having a Luxl homologue Cerl, that is known to be an acyl-HSL synthase involved in quorum sensing, producing an autoinducer that was detected using the bioassay [(Puskas et al. 1997); (Lang and Beatty, unpublished)]. Based on the sequence for the R. sphaeroides Luxl homologue, Cerl, the R. capsulatus genome was searched for a putative homologue. A n ORF with 26% identity and 43% similarity, containing all key Luxl homologous amino acids was found in R. capsulatus (Schaefer et al. 2002). A n insertional mutation was made in the R. capsulatus luxl homologue, subsequently named gtal, to create the R. capsulatus mutant strain ALS1 (Schaefer et al. 2002). Although this mutation abolished production of the C16-HSL, there was no  23 obvious phenotype (Schaefer and Greenberg, personal communication). As a result of a fortuitous conversation between A . Lang and P. Greenberg, we decided to investigate whether ALS1 was impaired in the expression of G T A genes.  3.3.3  Induction of R. capsulatus GTA by C16-HSL.  R. capsulatus G T A production is maximal in stationary phase, when cell densities are the highest (Solioz et al. 1975). It is also known that the Agrobacterium conjugative plasmid gene transfer system is controlled by quorum sensing (Fuqua and Winans 1993). Therefore, R. capsulatus G T A was investigated for regulation by the C16-HSL. Two methods were used to examine the role of the acyl-HSL synthase in G T A production. First, a G T A bioassay was used to assess the relative level of G T A particles produced by the wild type and the ALS1 mutant. G T A transduction was 5 to 6-fold less in the acyl-HSL deficient strain (Table 3). However, addition of exogenous C16-HSL restored G T A transduction of mutant ALS1 to that of wild type levels (Table 3). Secondly, a plasmid that carries the G T A gene orfg2 fused to lacZ and driven by the G T A promoter [pYP (Lang and Beatty 2000)], was conjugatively transferred into both the wild type and the acyl-HSL mutant (ALS1). Measurements of /3-galactosidase activity showed that orfgl expression was reduced approximately 7-fold compared to the wild type, and /3-galactosidase activity in the mutant was restored to the wild type level with the addition of C16-HSL (Table 3). These data indicate that G T A production is controlled at the level of G T A structural gene transcription by a regulatory system that is dependent on C16-HSL for maximal induction.  24 Table 3: Effects of acyl-HSL on G T A transcription and transducing particle production.  Wild type (SB 1003)  j8- galactosidase activity from p Y P 69.2  Number of GTA transductants 28.0  ALS1 (gtal')  9.9  5.3  ALS1 + 2 jLtM C16-HSL added exogenously  61.2  37.3  Strain  8  a  Average of two samples  b  Average of three samples  b  25 3.4 Discussion.  A radiotracer assay, rather than the traditional and limited bioassays was employed to show that R. capsulatus synthesizes a long-chain C16-HSL. The ORF that is responsible for the acyl-HSL synthase was identified and named gtal, based on sequence similarity to Luxl family members. Analysis of a mutant strain (ALS1) with an insertion in gtal, indicated that this gene product does indeed direct the synthesis of an acyl-HSL with a 16-member carbon chain. On the R. capsulatus genome there is a LuxR homologue adjacent to the gtal acyl-HSL synthase gene. There are also other apparent LuxR homologues elsewhere on the R. capsulatus genome. This is obviously an avenue for further research. It would be interesting to discover which LuxR homologues are the cognates for Gtal, and what is the function of each of these homologues. The R. capsulatus C16-HSL induced G T A gene expression 5- to 7-fold. The other known regulators, CckA and CtrA have been found to reduce the expression of G T A and flagellar genes by 10- to 100-fold when mutations are made in either the cckA or ctrA genes. The connections between the cckA/ctrA regulatory system and that of the acylH S L dependent regulation will be interesting material for future investigations. The regulation of G T A could be by independent, but overlapping signals, which act in cooperation. As new regulators of G T A are discovered, their places in a hypothesised G T A signalling cascade (Figure 1) will provide further insight into this novel phage-like particle, and its relationships with the host bacterium. Further research should provide exciting insights into the regulation of G T A in R. capsulatus.  26 4. GTA IMMUNOASSAY AND EPIFLUORESCENCE MICROSCOPY ASSAY  4.1 Introduction.  The current assay for G T A is a bioassay of gene transduction between a donor and a recipient strain containing different alleles of a gene, and takes about four days to obtain results. The measurements from the bioassay are quite variable (e.g., I have found +/50% variation even in duplicate samples). Clearly, to design and implement a specific and rapid immunoassay would assist in obtaining accurate measurements of G T A production under a variety of conditions. The other advantage of an immunoassay is that it could allow measurement  of G T A protein production independently of D N A  packaging, release of particles from cells, binding of particles to a recipient, and injection and homologous recombination of D N A in a recipient (as are needed to do the bioassay). Anti-GTA antibodies would provide a useful tool to investigate the mechanism of regulation, by accurate and rapid measurement of G T A protein levels in response to environmental changes and gene mutations that might affect G T A expression. In order to purify large quantities of a G T A protein to produce polyclonal antibodies for the G T A immunoassay, I tagged a component of G T A . The gene encoding the putative capsid protein (ORFg5; Figure 6) was selected and tagged with six sequential histidine codons either at the 3' (carboxyl) terminus or in the middle of the gene [at a site corresponding to amino acids that may be exposed on the surface of the capsid; (Wilkoff et al. 2000); R. Hendrix, personal communication]. The hexahistidyl-tagged G T A capsid proteins were over expressed in E. coli and purified via nickel-agarose and FPLC. The carboxyl-terminus tagged protein was not as well expressed as the mid-tagged capsid protein. Polyclonal antibodies were then obtained using the purified protein injected into rabbits, and a protocol was established for their use as a G T A immunoassay in R. capsulatus cultures. I also did experiments to evaluate the use of a fluorescent dye to count D N A containing virus-like particles in cultures of R. capsulatus.  27  MKTETKARAGTGMPEGADPVAEVKTALAGFLKEVKGFQDDVKTRLQQQEER VTMLQTKTYAGRHALAAAATEEAPHQKAFAAYLRTGDDDGLRGLSLEGKAL NSAVAAEGGYLVDPQTSETIRGVLRSTASLRQIASWNVEATSFDVLVDKT DMGSGWASETAALSETATPQIDRITIPLHELAAMPKASQRLLDDSAFDIET ML WLANRIADKFARAEAAAFISGDGVDKPTGFLTKTKVANGAWAWGSLGYVAT GAAGD FAAVNAS D A W D L V Y A L G A E YRANAS FVMNS KTAGAVRKMKDADGR F LWAD S L A A G E PARLMGY P V L I A E D M P D I A A N A Y A I A F G D F G N G Y T I A E R P DLRVLRD P F S A K P H V L F YAS KRVGGD VS D F A A I K L L K F A A J ^ KA  [399 aa]  - proteolytic cleavage site - location of 6X histidines  Figure 6: Amino acid sequence of 0RFg5, GTA capsid protein. This protein appears to be made as a pro-protein that is cleaved to produce the protein found in GTA particles (Lang and Beatty 2000). The uncleaved protein with 6 histidines is approximately 43.1 kDa, and once N-terminally cleaved the approximate size is 32.4 kDa. The 2 independent locations where the 6 histidines were inserted are marked by the grey triangles to create a carboxyl-terminus tagged capsid protein and a mid-tagged protein. The location of the mid-tag histidines was selected in the hope that in a completed capsid head the histidines would be exposed exteriorly.  28 4.2  Materials and Methods.  4.2.1  Bacterial strains, growth conditions and plasmids.  The bacterial strains and plasmids used in this study are listed in Table 4.  E. coli strains used for cloning and subcloning were DH5ct (Invitrogen) and DH10B (Invitrogen). The E. coli strain used for protein over expression was M l 5 (pREP4) (Qiagen). E. coli strains were grown at 37°C in Luria-Bertani medium (Sambrook et al. 1989) supplemented with antibiotics as necessary at the following concentrations (u.g/ml): ampicillin, 100; tetracycline-HCl, 10; kanamycin sulfate, 25; gentamycin sulfate, 10; and spectinomycin-HCl, 100.  R. capsulatus strains B10 (Marrs 1974), Y262 (Yen et al. 1979), SB 1003 (Yen and Marrs 1976), Y C K F  (Lang and Beatty 2000) and Y G T 9 (Lang 2000) were used to monitor  G T A production through the immunoassay and the bioassay. R. capsulatus strains were grown aerobically or photosynthetically in R C V medium (Beatty and Gest 1981) or YPS (Wall et al. 1975) medium at 30-35°C, supplemented with antibiotics as necessary at the following concentrations (u.g/ml): tetracycline-HCl, 0.5  and gentamycin sulfate, 3.  Culture turbidity was measured with a Klett-Summerson photometer, in which 100 Klett units equals approximately 4 x 10  8  colony forming units (CFU)/ml or using a  spectrophotometer to measure light-scattering at 660 nm. 4.2.2  Plasmid DNA preparation and manipulation.  Standard methods of D N A purification, restriction enzyme digestion, and other modification techniques were used (Sambrook et al. 1989).  4.2.3  Construction of protein over expression plasmids.  Cosmid p9H2 (Lang and Beatty 2000) that contains a large fragment of G T A structural genes, including ORFg5, was digested with Pstl and the fragment containing ORFg5 was inserted into pUC16 Pstl sites to create pUCG5. The cloning is described below and shown in Figure 7.  29 Table 4: Plasmids and strains  Strain R. capsulatus BIO Y262 SB 1003 YCKF YGT9 DW5 E. coli DH5cc DH10B M l 5 (pREP4)  Source  Phenotype  (Marrs 1974) (Yenetal. 1979) (Yen and Marrs 1976) (Lang and Beatty 2000) (Lang 2000) (Wongetal. 1996)  Wild type G T A overproducer Rif*, Cured of phage CtrA", Y262 background ORFg2", Y262 background PuhA", no photosynthesis  Invitrogen Invitrogen Qiagen  Cloning strain Cloning strain Protein over expression strain  Plasmid p9H2  Source (Lang 2000)  Markers* Km ' Tc  pQE-60 pQECAPC  Qiagen This work  Ap Ap  R  pQECAPM  This work  A  R  pUCG5  This work  Ap  R  pUCG5mid  This work  Ap  R  pUCseqC  This work  A  R P  pUCseqM  This work  A  P  pUC19  (Norrander et al. 1983)  A  P  K  K  R  P  R  R  Remarks Cosmid containing part of the G T A gene cluster Over expression plasmid Over expression plasmid with Cterminal histidine tag on ORFg5 Over expression plasmid with histidine tag in middle of ORFg5 pUC19 containing Pstl ORFg5 fragment from p9H2 PCR mutagenized pUCG5 containing 6 histidines in middle of ORFg5 Subcloned C-tagged ORFg5 from pQECAPC for sequencing Subcloned mid-tagged ORFg5 from p Q E C A P M for sequencing Cloning vector  * Resistance markers; A p : ampicillin, K m : kanamycin, Rif*: rifampicin, T c : tetracycline. R  R  R  Figure 7: Cloning scheme to create tagged ORFg5 protein  31  Pst I Pst I  ORFg5  * 6X  His *  ORFg5 QuikChange mutagenesis to add 6 histidines to the middle of ORFg5 (Primers 3 and 4). Tagging was confirmed by sequencing with the pUC19 universal forward primer  Pst I PCR amplification of ORFg5 to add a BspHI site (primer 1) and a BamHl site (primer 2).  BspHI  BamHl * PCR product digested with BspHI and BamHl and inserted into pQE-60 Ncol and BamHl. <Ncol/Bspl>  PCR amplification of ORFg5 mid-tagged to add a BspHI site (primer 1) and a Hindlll site (primer 5).  BspHI  6X His  Hindlll  PCR product digested with BspHI and Hindlll and inserted into pQE-60 Ncol and Hindlll. <Ncol/BspHI> 6X His  ORFgS  Hindlll  Prior to transformation of to M l 5 cells, plasmids were digested with Xhol and Hindlll, subcloned to pUC19 SaWHindlll sites and sequenced using pUC19 universal forward and reverse primers to confirm PCR fidelity and presence of tags, pUCseqC and pUCseqM.  32 In order to tag ORFg5 with 6 histidines in the middle of the gene, QuikChange mutagenesis (Stratagene) was used (Primers 3 and 4, see Table 5) as per the manufacturer's suggested protocol with modifications. Due to the high G C content of R. capsulatus D N A , D M S O was added to 5% to the QuikChange reaction. This created pUCG5Mid that was subsequently P C R amplified using primers 1 and 5 (Table 5) to incorporate a BspHl site and a terminal Hindlll site to allow subcloning to pQE-60 Ncol and Hindlll sites to create the over expression plasmid p Q E C A P M . For the carboxyl-terminus tagged capsid gene, pUCG5 was P C R amplified using primers 1 and 2 (Table 5) to incorporate a BspHl site and a BamHl to allow subcloning to pQE-60 Ncol and Bamlll sites to create the over expression plasmid pQECAPC. To confirm the P C R fidelity and presence of appropriate tags, plasmids pQECAPC and p Q E C A P M were digested with Xhol and Hindlll  and subcloned in  parallel to pUC19 Sail /Hindlll to create pUCseqC and pUCseqM. The over expression strains were then created by heatshock transformation (Sambrook et al. 1989) of either pQECAPC or p Q E C A P M to M l 5 cells containing the repressor plasmid pREP4 (Qiagen). 4.2.4  DNA sequencing.  Sequencing was completed in Dr. I. Sadowsky's laboratory (Biochemistry and Molecular Biology Department, University of British Columbia) or the University of British Columbia Nucleic Acid and Protein Service Unit (NAPS) using universal primers (New England BioLabs) from plasmid templates purified using QIAprep Miniprep Kit (Qiagen).  4.2.5  Protein over expression.  Protein over expression was performed following the QIAexpress kit manufacturer's suggested protocol (Qiagen). Optimal conditions for protein over expression for the midtagged protein M l 5 (pREP4, pQECAPM) were L B medium, 37°C and 4 hours post induction (addition of 1 m M IPTG) to harvest protein. Optimal conditions for the carboxyl-terminus tagged protein M l 5 (pREP4, pQECAPC) were L B medium, 30°C and 6 hours post induction to harvest protein.  33 Table 5: Primers designed and used * Primer sequence 1 5'- C A G G A G T G T G G A A A T C A T G A A G A C C -3' 2 5'- C C G C C C G T T A G G A T C C G G C A A A C T T C - 3 ' 3 5'- C G A G A C G G C G G C G C A T C A C C A T C A C C A T C A C C T G A G C G A G A C - 3 ' 4 5'- G G T C T C G C T A G G T G A T G G T G A T G G T G A T G C G C C G C C G T C T C G - 3 ' 5 5'- C C C C T T C A C C G A A G C T T A C G A G G C - 3 ' * Number as referred to in Figure 7 and related text.  Name BSPHI BAMHI HSENS HANTI HIND  34 4.2.6  SDS polyacrylamide gel electrophoresis (SDS PAGE).  The Laemmli buffer system (Laemmli 1970) was used to separate proteins on 12% acrylamide gels with 4% stacking gels (acrylamide/bis 37.5:1). Protein samples were boiled for 10 minutes in sample loading buffer [50 m M Tris-Cl (pH 6.8), 2% SDS, 10% glycerol, 0.1% bromophenol blue, 1% 2-mercaptoethanol] prior to loading the gel. Gels were run with the M i n i - P R O T E A N II systems (Bio-Rad) according to manufacturer's protocols. SDS P A G E gels were stained using Coomassie blue (40% methanol, 10% acetic acid, 0.025% Coomassie dye).  4.2.7  Protein purification using Ni-NTA columns.  Protein was purified following the QIAexpress kit manufacturer's suggested protocol to yield native protein (Qiagen) with some modifications. 100 ml cultures were grown in optimal conditions (section 4.2.5), harvested by centrifugation and resuspended in a 1 ml volume. The cells were then treated with lysozyme, sonicated, and protein was bound to 500 /xl of Ni-nitrilotriacetic acid (NTA) agarose resin in a mini-column made by packing glass wool into a 1 ml plastic pipette tip. The column was equilibrated with (50 m M N a H P 0 , 300 m M NaCl, pH 8.0). Four washes of 1.5 ml each were performed [1:(50 4  m M N a H P 0 , 300 m M NaCl, pH 8.0), 2:(50 m M N a H P 0 , 300 m M NaCl, pH 6.5), 3:(50 4  4  m M N a H P 0 , 300 m M NaCl, 20 m M Imidazole, pH 8.0), 4:(50 m M N a H P 0 , 20 m M 4  4  Imidazole, pH 8.0)]. The protein was then eluted with 1.5 ml of (50 m M N a H P 0 , 250 4  m M Imidazole, pH 8.0). The eluate was buffer exchanged for (50 m M Tris-Cl, pH 8.0) on a pre-rinsed (50 m M Tris-Cl, pH 8.0) 10K Macrosep centrifugal concentrator according to manufacturer's suggested protocol ( P A L L Life Sciences).  4.2.8  Protein purification using AKTA-FPLC.  The eluate in (50 m M Tris-Cl, pH 8.0) was applied to a Resource Q anion exchange column (6 ml volume: Amersham) attached to a A K T A - F P L C following manufacturer's suggested protocols (Amersham) using binding buffer (50 m M Tris-Cl, pH 8.0) and eluted with a continuous gradient of NaCl. The purified mid-tagged capsid protein elutes at approximately 200 m M NaCl, and the C-tagged capsid protein elutes at approximately  35 400 m M NaCl. The purified fractions were collected and concentrated on a pre-rinsed (50 m M Tris-Cl, pH 8.0) 10K Macrosep centrifugal concentrator according to manufacturer's suggested protocol ( P A L L Life Sciences).  4.2.9  Polyclonal antibody production.  The purified protein was injected into rabbits following Current Protocols in Molecular biology (Cooper and Paterson 1997) and advice from the University of British Columbia Animal Care Facility. A primary inoculation of purified protein homogenized in complete Freund's adjuvant (0.5 mg mid-tagged or 0.2 mg C-tagged ORFg5 protein preparation per rabbit) and 3 subsequent boosts of purified protein homogenized in incomplete Freund's adjuvant (0.1 mg mid-tagged or 0.07 mg C-tagged protein per rabbit) were given. Homogenization of purified protein with adjuvant was accomplished using 20second sonication blasts, 3 times. Two rabbits were used for the mid-tagged capsid protein antibody production and one rabbit was used for the carboxyl-terminus tagged capsid protein.  4.2.10  Cross-adsorption of antibodies.  In order to decrease the background of the Western blots, the antibodies were adsorbed to a lysate of M l 5 (pREP4) cells. A 3 ml overnight culture of M l 5 (pREP4) cells was centrifuged and the cell pellet resuspended in 500 y\ of phosphate buffer (50 m M Na2HP04, pH 7.5). The cell suspension was sonicated on ice on medium frequency for 2 pulses of 20 seconds each. 100 [i\ of a 50% mixture of the polyclonal antibodies was added and incubated on ice for 2 hours with occasional mixing. The mixture was then centrifuged for 4 minutes and the supernatant was decanted to a fresh tube for use as the primary probe in Western blots (M. Toporowski, personal communication).  4.2.11  Western blotting.  Photosynthetically grown R. capsulatus cells were harvested at the desired point in the growth phase, based on culture turbidity. Cell samples were pelleted by centrifugation, the culture supernatant was transferred to a fresh tube, and then dried down using a vacuum centrifugator (Savant Speed Vac) on high heat for 60 to 90 minutes. Samples  36 were then dissolved based on the turbidity of the culture at the time the sample was taken, so that the equivalent of 100 ul of OD660 0.071 of the original culture was loaded in each lane. The samples were run on 12% SDS P A G E gels and blotted onto a nitrocellulose membrane (Bio-Rad). The blotting was performed in Trans-Blot or Mini Trans-Blot apparatus (Bio-Rad) according to the manufacturer's protocols in Towbin buffer (Towbin etal. 1979). The primary antibodies ( M l or M 2 ; rabbit anti-capsid) were detected using horseradish  peroxidase-linked anti-rabbit  Ig secondary  antibody (from  donkey;  Amersham) as part of the enhanced chemiluminescence (ECL) kit according to manufacturer's protocols (Amersham).  4.2.12  Protein sequencing and mass spectrometry.  For sequencing and mass spectrometry, capsid protein purified using N i - N T A columns and F P L C as described above was analysed at the University of British Columbia Nucleic Acid and Protein Service Unit (NAPS). The approximate concentration of the protein determined by SDS P A G E was 230 pmoles/jul. The submitted sample yielded a protein mass estimate of 42.980 kDa with a computer called N-terminal sequence of A K T E T N and A K T E T G or M R A G D K called by the technician with ambiguity at amino acids 1-4 and 6, this matches the tagged G T A capsid protein M K T E T K (Figure 6).  4.2.13  Epifluorescence microscopy using SYBR-green.  Epifluorescence microscopy (EPM) was performed following the protocol of (Hennes and Suttle 1995), with some modifications. Filters containing samples ranging from 100800 ixl were visualised using SYBR-green (Molecular probes) [70 p\ 0.02 /mi filtered Milli-Q water, 2 LII SYBR-green (10% solution, resuspended)]. Filters were incubated with SYBR-green for approximately 15 minutes in the dark. Filters were mounted using a mounting solution containing 50% glycerol, 50% phosphate buffer solution (50 m M N a H P 0 , 0.85% NaCl, pH 7.5) and 0.1% p-phenylenediamine to prevent fading. Slides 2  4  were viewed using a 100X lens with a wide blue filter set. Uniform bright pinpricks were counted as "particles".  37 4.2.14  GTA bioassays (transduction of puhA gene).  Bioassays for G T A activity were performed as described in (Solioz et al. 1975), with some modifications. A 100 /il volume of 0.2 urn filtered culture liquid to be assayed was mixed with 100 ju.1 of the indicator strain cells and 400 /xl of G-buffer [10 m M Tris-Cl (pH 7.8), 1 m M M g C l , 1 m M CaCl , 1 m M NaCl, 500 /ig/ml BSA] (Solioz et al. 1975). 2  2  The indicator strain was a puhA deletion mutant strain DW5, that is incapable of photosynthetic growth (Wong et al. 1996), prepared by centrifuging an overnight culture and resuspending cells in an equal volume of G-buffer. The entire mixture was incubated for 1 hour at 30-35°C with gentle agitation and then 900 /il of R C V medium was added and incubation under the same conditions was continued for 3-4 hours. The cells were spread on R C V plates and incubated under photosynthetic conditions in anaerobic GasPak jars for 2-3 days. Plates were removed and the numbers of puhA  +  colonies counted;  because DW5 is incapable of photosynthetic growth, all colonies were transductants.  4.2.15  Nutrient limitation.  R. capsulatus cultures were grown photosynthetically in R C V medium to log phase and harvested by centrifugation. Cell pellets were resuspended and centrifuged twice, once with fresh R C V and a second time with phosphate-free R C V [RCV without ( N H ) S 0 ] . 4  2  4  The twice-washed cells were resuspended in phosphate-free R C V and inoculated into R C V media variants in 16.5 ml screw top glass tubes to equal approximately 20 K U . The samples were incubated photosynthetically and 50 /il samples were taken after 24 hours of growth. Each sample was centrifuged and the supernatant was separated from the cell pellet. The supernatant was dried down using a vacuum centrifugator (Savant Speed Vac) on high heat for 60 to 90 minutes. Media variants of R C V investigated included; low phosphate [RCV with 15 (JM  phosphate, supplemented with MOPS pH 6.8 to maintain  buffering (9.6 m M phosphate is standard)], low nitrogen [RCV with 0.025% ( N H ) S 0 4  2  4  (1/4 the standard amount)], low carbon [RCV with 0.13% malate (1/3 the standard amount)] and R C V phosphate-buffered to pH 7.8 (above the usual starting pH of 6.8). It was determined that these conditions were nutrient limited in that the final culture density was approximately one-half the density using R C V media.  38 4.3  Results. 4.3.1  Optimization of overproduction of tagged capsid proteins.  G T A ORFg5, which codes for the major capsid protein was successfully subcloned and tagged with 6 consecutive histidines either at the carboxyl-terminus or in the middle (see Materials and Methods). Once these overproducer strains were created in an E. coli background host [ M l 5 (pREP4, pQECAPC), M l 5 (pREP4, pQECAPM)], it was possible to optimize the overproduction of the tagged proteins in order to obtain the largest quantity of protein per cell. Ideal conditions for the overproduction were determined by loading whole cells on SDS P A G E gels (see Materials and Methods). The best conditions for the carboxylterminus tagged protein were found to be growth in L B medium, 30°C allowing 6 hours of growth post induction prior to harvesting. Optimal conditions for the mid-tagged protein were growth in L B medium, 37°C with 4 hours of growth post induction prior to harvesting. The mid-tagged capsid protein was always expressed in larger quantities than the carboxyl-terminus tagged protein. Even by adjusting temperature, media or incubation times, cells containing the carboxyl-terminus tagged protein always yielded lower amounts (Figure 8). Maximizing protein yield at this early step is critical to obtain sufficient material for subsequent purifications and the end use: antibody production. Due to the relatively good production of the desired protein (greater than 1% of the total cell protein content) (Figure 8) subsequent purifications were possible, without having to scale up culture volumes.  39  Figure 8: SDS P A G E of initial overproduction of tagged proteins. The overproducer strains M l 5 (pREP4, pQECAPC)] and M l 5 (pREP4, pQECAPM) were grown in L B medium at different temperatures (30 and 37°C) to optimize protein production. Odd numbered lanes (1, 3, 5, 7) show whole cells loaded prior to addition of EPTG (0 hours); even numbered lanes (2, 4, 6, 8) show whole cells 4 hours after addition of IPTG.  40 4.3.2  Purification of tagged capsid proteins.  In order to produce the most specific polyclonal antibodies possible, it is necessary to use the purest antigen, which would ideally be a single protein of known concentration. In my case, two purification steps were employed. The first purification step was to use N i - N T A columns that specifically bind oligo-histidines, such as the hexahistidine tags of the over expressed capsid proteins (see Materials and Methods). The protein obtained was relatively pure, with few contaminants visible in the SDS P A G E gel (Figure 9). Once again, less of the carboxyl-terminus tagged protein was obtained compared to the mid-tagged capsid protein, compare lanes 2 and 1 in Figure 9. The second step of purification was F P L C with a Resource Q anion exchange column (see Materials and Methods). This technique should yield extremely pure protein, as most proteins elute at a specific and reproducible salt concentration in a limited set of fractions. The mid-tagged capsid protein eluted at approximately 200 m M NaCl as a single sharp peak (data not shown), and appeared to be relatively free from contaminating proteins when concentrated (see Materials and Methods) and run on an SDS P A G E gel (Figure 10). The carboxyl-terminus tagged capsid protein eluted at approximately 400 m M NaCl as a single sharp peak (data not shown); however when the peak fraction was concentrated and run on an SDS P A G E gel, there was no band visible using Coomassie staining (data not shown).  4.3.3  Confirmation of identity of purified capsid protein.  In order to confirm the identity of the protein purified by N i - N T A column and F P L C that was used for antibody production, mass spectrometry and N-terminal sequencing were performed on the mid-tagged capsid protein (see Material and Methods).  41  1 2 (Mid-tagged) (C-tagged)  47.5 kDa  32.5  Figure 9: SDS P A G E of primary purification (Ni-NTA column) of tagged capsid proteins. Lane 1 represents primary purification of the mid-tagged capsid protein, lane 2 is the carboxyl-terminus tagged capsid protein. Lower amounts of the carboxyl-terminus tagged protein were routinely obtained in comparison to the mid-tagged protein.  Figure 10: SDS P A G E of secondary purification of mid-tagged capsid protein using a Resource Q column in FPLC (used for antibody production). Lanes 1 and 2 contain 1 / i l and 10 /il, respectively, of a concentrated 5 ml fraction. Band a represents the full-length uncleaved tagged capsid protein and band b appears to be a potential cleavage product(s), perhaps due to E. coli proteases.  43 Mass spectrometry yielded a single dominant peak as expected at approximately 43 kDa, consistent with the mass of the mid-tagged capsid protein that has a theoretical value of 43.1 kDa (Figure 11). A secondary peak was also found at approximately 36.3 kDa (Figure 11), which could represent a cleavage product due to the presence of proteases in the E. coli strain used to express the tagged capsid protein (see section 4.4.2). N-terminal sequencing of the mid-tagged capsid protein yielded a computer-called N terminal sequence of AKTETN, whereas AKTETG or MRAGDK were called by the technician with ambiguity at amino acids 1-4 and 6. The consensus of these calls matches the tagged GTA capsid protein MKTETK. The N-terminal sequence data was deemed "noisy" by the technician and this could be due to more than one protein being copurified (Figure 10) and the potential cleavage product being present. These data indicate that the predominant protein purified and used to create the anti-capsid antibodies is indeed the correct protein; the tagged GTA capsid protein. 4.3.4  Anti-capsid polyclonal antibodies.  Anti-capsid polyclonal antibodies were made in rabbits using both the mid-tagged capsid protein and the carboxyl-terminus tagged capsid protein (see Materials and Methods). The sera were assayed for specificity and titre using Western blotting techniques. The polyclonal antibody preparation was then "cleaned" using sonicated E. coli M l 5 (pREP4) cells to adsorb non-specific antibodies (see Materials and Methods). Using the anti-capsid antibodies (either unprocessed or "cleaned") as a primary probe of R. capsulatus cells known to produce GTA yielded a band at approximately 32 kDa (Lanes 2 and 3 in Figure 12A-D). This band correlates to the capsid protein that is known to be N-terminally cleaved at a site that would yield a protein of 32.4 kDa (Lang and Beatty 2000). The wild type strain BIO appears to produce low amounts of GTA (Figure 12B lanes 4 and 5), this is consistent with previous BIO GTA bioassays (Solioz et al. 1975). Thus the polyclonal anti-capsid antibodies appear to be useful in identifying GTA production, based on the capsid protein.  44  42980.10  43137.11 8066.75  Figure 11: Mass spectrometry of the purified mid-tagged capsid protein. The major peak is at approximately 43 kDa, this represents the tagged capsid protein as expected. A secondary peak located at 36.4 kDa could represent a C-terminal cleavage product due to the presence of proteases in the E. coli strain used to express the tagged capsid protein.  45  Final serum  "Cleaned" antibodies  Figure 12: Western blots of final antibodies and "clean" antibodies from last bleed. M l and M 2 are the anti-capsid antibodies obtained from two different rabbits using the same mid-tagged purified capsid protein (see Materials and Methods). The bands of A represent the M l anti-capsid antibody final serum, B represents the M l anti-capsid antibody "cleaned" by adsorption to E. coli cells. C and D represent the M2 anti-capsid antibodies, final serum and "cleaned", respectively. Lane 1 is the purified mid-tagged protein. Lanes 2 and 3 show samples of the overproducer strain Y262 grown to stationary phase in YPS, whole cells or supernatant, respectively. Lanes 4 and 5 show samples of the wild type strain BIO grown to stationary phase in YPS, whole cells or supernatant, respectively.  46 4.3.5  GTA immunoassay using anti-capsid antibodies.  A G T A immunoassay with "clean" anti-capsid antibodies was used to evaluate a more rapid and reproducible method to look at G T A production under a variety of conditions. In order to develop the G T A immunoassay a clear protocol was prepared for their use in R. capsulatus strains (see Materials and Methods), and this was tested using the G T A overproducer strain Y262 in Y P S rich medium. Photosynthetically grown Y262 cultures were harvested at early, mid and late stationary phase, and the supernatant was processed by means of Western blots (see Materials and Methods). Analysis of G T A production based on a Western blot of the supernatant in stationary phase showed a 32 kDa band that is attributed to the G T A capsid protein (Figure 13). In addition, even in very old cultures (135 hours), this band as measured by the immunoassay appeared to be high. The G T A immunoassay can be also used to estimate the relative amounts of G T A capsid protein within and outside of cells as they enter the stationary phase of maximal production of G T A transducing particles, by comparing the cell-associated 32 kDa band (cell pellets) to the band in cell-free G T A (supernatant) samples. This can be seen in lanes 2 versus 3 in Figure 12D, where there appears to be equivalent amounts of the 32 kDa band in the cells and outside the cells. This implies that there is some inefficiency in the escape of G T A particles, consistent with the lack of cell lysis, and relates to the unknown mechanism of G T A particle release. The band at approximately 36 kDa that was associated with the cell could represent the capsid precursor (uncleaved), but the gel migration was to a slightly higher position than the histidine-tagged control protein.  4.3.6  Environmental effects on GTA production using the GTA immunoassay.  The anti-capsid antibodies were used to investigate environmental effects on G T A production. A series of nutrient-limiting media types were made based on the defined medium, R C V (see Materials and Methods). They were low carbon, low phosphate, low nitrogen and starting pH 7.8. Photosynthetic cultures of Y262 were grown to stationary phase and cell pellets and supernatants were collected and processed using the G T A immunoassay (Western blots with the anti-capsid antibodies as the primary probe).  47 The Western blots shown in Figure 14 indicate differences in G T A production under differing nutrient limiting conditions. These preliminary results indicate that under phosphate limitation the 32 kDa band was present in cell samples and in the supernatant, in an amount that appears to be similar to the amount obtained using the rich medium, YPS (compare Figure 14C lanes P to YPS and 14D lanes P to YPS). It is also interesting to notice that under carbon limitation, there appears to be significant amounts of G T A production in the cells and little or no G T A in the supernatant (see Figure 14C lane C and 14D lane C).  48  Figure 13: Western blot of time course of G T A production in a Y262 culture grown in YPS medium. Supernatant samples grown in YPS, harvested at time points indicated and loaded in equal amounts, calculated per cell based on OD readings. G T A production in the supernatant appears to peak in late stationary phase, and is present even in old cultures  (135  hours).  49  N  P  C pH  g  £  A.  > Experiment 1 B.  N  P  C pH  g  p 32.5 kDa S. Experiment 2  D.  32.5 kDa  Figure 14: Western blots of environmental effects on G T A production, probed with "cleaned M2 anti-capsid antibodies and using samples from two independently grown cultures. Samples from stationary phase Y262 cultures grown photosynthetically (see Materials and Methods). A . Western blot showing cellular amounts of G T A in experiment I. B. Western blot showing supernatant amounts of G T A in experiment 1. C . Western blot showing cellular amounts of G T A in experiment 2. Replicate independent experiment. D. Western blot showing supernatant amounts of G T A in experiment 2. Lanes: N ; low nitrogen R C V , P; low phosphate R C V , C ; low carbon R C V , pH; pH 7.8 RCV.  50 4.3.7  Epifluorescence microscopy and particle counts.  SYBR-green was used to label viral-like particles in R. capsulatus strains Y262, Y C K F and Y G T 9 on the basis of the bioassay (see Materials and Methods) for counting by epifluorescence microscopy (EPM). It was observed that there were significantly more particles in the overproducer strain, Y262 than the CtrA regulatory mutant, Y C K F or the G T A structural gene ORFg2 mutant, Y G T 9 (Figure 15A). It was also found that there were significantly more particles in the rich medium YPS, as opposed to the minimal medium R C V (Figure 15A). Higher G T A production in rich media has been reported previously (Solioz 1975). However the number of particles observed by means of E P M does not reflect the number of transductants from the G T A bioassay (Figure 15B). Specifically, SYBR-green labels all D N A present in the sample, and so G T A cannot be distinguished from other particles, and are all counted as long as they appeared as small, uniform pinpricks. The number of particles observed in the G T A structural mutant, YGT9 were approximately 1 X 10 particles/ml in YPS medium (Figure 15A), however 8  in the G T A bioassay there were no transductants obtained from the Y G T 9 strain (Figure 15B). These data indicate that the amounts of G T A particles detected by E P M are overwhelmed by non-GTA virus-like particles.  4.4  Discussion.  4.4.1  GTA immunoassay as a tool to investigate GTA production is practical and functional with some limitations.  As previously mentioned, the current assay for G T A is a gene transduction bioassay that takes about four days to obtain results. In contrast, the immunoassay takes two days to complete. The measurements from the bioassay are quite variable, even with duplicate samples (e.g., I have found +/- 50% variation in duplicate samples). Therefore, this immunoassay should assist in obtaining rapid and accurate measurements of G T A production under a variety of conditions. The other advantage of an immunoassay over the gene transduction bioassay is that the immunoassay allows measurement of G T A protein production independently of D N A packaging, release of particles from cells, binding of particles to a recipient, and injection  51 1.0E+10  1.0E+09 _i  E  "35 •g  1.0E+08  '€ 1.0E+07  1.0E+06 Y262  YCKF  YGT9  Y262  YCKF  YGT9  1000  B.  c re  o  3 "D  </) C  re  Figure 15: Epifluorescence microscopy particle counts compared to G T A transduction (bioassay). A . Particles per ml stained with SYBR-green. Strains grown in YPS media exhibit greater numbers of particles compared to those grown in R C V . Y262 is the G T A overproducer, Y C K F is the CtrA regulatory mutant, and Y G T 9 is the G T A structural mutant. Standard deviation is minimal and is shown on plot. Viewed using 100X lens with "wide blue" lenses: see Materials and Methods. B. Number of successful G T A transductants/ml, using the same samples as in the YPS particle counts; average of three replicates, G T A bioassay used to obtain data. Standard deviations as follows; Y262: 170, Y C K F : 11.5, YGT9: 0, transductants/ ml.  52 and homologous recombination of D N A in a recipient. The anti-capsid antibodies provide a useful tool to investigate the mechanism of regulation by measurement of G T A capsid protein levels in response to environmental changes and gene mutations (see below). The protocol described for the use of the anti-capsid antibodies in the G T A immunoassay is relatively straightforward. The G T A immunoassay involves acquiring cell or supernatant samples, running an SDS P A G E gel, immunodetection with the "cleaned" anti-capsid antibodies as the primary probe. There are no special apparatus requirements for performing the immunoassay, as opposed to the G T A bioassay that requires anaerobic Gas-Pak jars and photosynthetic incubation equipment. There are some limitations to the G T A immunoassay. Due to the nature of the polyclonal antibodies, there are extra bands that appear in the Western blots that cannot be explained. Although there were variations in the bands obtained in the Western blots, these appear to be due to transfer of protein to the membrane. Loading 2-3 replicate lanes, and taking the average could overcome these variations. It is also difficult to visualise the appropriate band when strains that produce low levels of G T A are used, such as the wild type strain BIO or when minimal media, such as R C V are used. These problems could be overcome by loading more sample, longer exposure (with the disadvantage of higher background), longer incubations with the anti-capsid antibody, or use of richer growth media, such as YPS. Overall, the G T A immunoassay is an excellent addition to the limited set of tools currently available to investigate G T A . The G T A immunoassay is rapid and more reproducible than the G T A bioassay because the immunoassay does not rely on the release, uptake and homologous recombination of packaged D N A between donor and recipient cells. The G T A immunoassay will allow a larger array of possible applications and avenues of study (see section 4.4.5). The disadvantage is that the immunoassay measures only one G T A protein, and so does not necessarily reflect the amount of G T A particles. This is especially a problem for evaluation of intracellular G T A particles because; this protein could be made but not assembled as part of a particle. However a combination of dye-labels, SYBR-green for D N A and Texas Red labelled anti-capsid antibodies could over come this problem (see section 4.4.5 below).  53 4.4.2  Resolving the extra bands.  As with all work done with polyclonal antibodies, there is the possibility of crossreactivity and non-specific binding of antibodies to irrelevant protein products. This is due in part to the purity of the antigen used to make the antibodies in the first place, differences in the innate immune system of the animals used to obtain the antibodies, and the nature of polyclonal antibodies being a pool of many different antibodies (Cooper and Paterson 1997). To decrease the background reactivity of the anti-capsid antibodies, they were adsorbed to lysates of E. coli cells that did not produce the target protein, to take irrelevant antibodies out of the polyclonal antibody pool. This was successful in diminishing some of the extra bands that appeared in the Western blots. In order to explain the secondary band (approximately 36 kDa according to SDS P A G E gel analysis, Western data and M S results) that was always found along with the purified capsid protein (43.1 kDa), potential sites of protease activity were investigated. Sites specific to the E. coli strain that was used to over express the tagged capsid protein were considered. A single cleavage site for chymotrypsin of 99% probability was found on the carboxyl end of the capsid protein (Figure 16). Chymotrypsin-like proteases are found in most strains of E. coli such as M l 5 , which was used to express the capsid protein (Maurizi et al. 1990). Cleavage at this site would yield a protein that is approximately 37.7 kDa. If there is proteolysis at or near this site, this could explain the band(s) found in Figure 10 (Lane 2, band 'b'). In addition, cleavage at the carboxyl end of the tagged capsid protein would explain the difficulty in the purification of significant amounts of the carboxyl terminus tagged capsid protein, as any proteolysis at this end of the protein would cleave off the histidine tag necessary for the primary purification step using the N i - N T A agarose columns. This also indicates an advantage of over expressing the carboxyl terminus tagged capsid protein at lower temperatures (30°C as opposed to 37°C) in order to be below the temperature of optimal protease activity (Wingfield 1995).  54  MKTETKARAGTGMPEGADPVAEVKTALAGFLKEVKGFQDDVKTRLQQQEERVTMLQTKTY  6i  AGRHALAAAATEEAPHQKAFAAYLRTGDDDGLRGLSLEGKHHHHHHALNSAVAAEGGYLV + +  +  +  +  120  +  121  DPQTSETIRGVLRSTASLRQIASWNVEATSFDVLVDKTDMGSGWASETAALSETATPQI + + + + + +  • 180  181  DRITIPLHELAAMPKASQRLLDDSAFDIETWLANRIADKFARAEAAAFISGDGVDKPTGF + + + + + +  240  241  LTKTKVANGAWAWGSLGYVATGAAGDFAAVNASDAWDLVYALGAEYRANASFVMNSKTA + +  +  +  +  Chym  +  3  0  0  (99.5%)  I  301  GAVRKMKDADGRFLWADSLAAGEPARLMGYPVLIAEDMPDIAANAYAIAFGDFGNGYTIA + + + + +  361  ERPDLRVLRDPFSAKPHVLFYASKRVGGDVSDFAAIKLLKFAASZ + + + +  +  3 o 6  405  Figure 16: Potential proteolysis site of ORFg5 expressed in E. coli. If this site were cleaved in E. coli the size of the protein would be approximately 37.7 kDa, which could represent the second band(s) that co-purifies at approximately 36 kDa and the secondary peak in the M S data at approximately 36.3 kDa (Figures 10 and 11). Figure output obtained at http://ca.expasy.org/tools/peptidecutter/ using the protease chymotrypsin at 99% probability of cleavage.  55  4.4.3  Implications of preliminary environmental studies on GTA production.  Using variants of the defined medium R C V , with specific nutrients limited or starting growth at pH 7.8 instead of the R C V pH 6.8, preliminary results indicated that when nutrients such as carbon or phosphate are limited the G T A production profile is altered (Figure 14). It appears that under carbon limitation, amounts of G T A in the cells is far greater than that found in the supernatant (compare Figure 14C lane C to 14D lane C). Also observed was the production of equivalent amounts of G T A in cells and supernatant under minimal medium phosphate limitation and the rich medium Y P S (compare Figure 14C lanes P to Y P S or 14D lanes P to YPS). It is interesting that the limitation of certain nutrients in the R C V minimal medium (where G T A production is very low) can trigger G T A production that is similar to that of the rich medium YPS. We know that as cultures enter the stationary phase there is an increase in G T A production (Yen and Marrs 1976), however during the onset of stationary phase there is a complex set of phenomena that occur at the level of individual cells and between cells of the group (quorum sensing) undergoing stationary phase. We know that quorum sensing (cell groups) is involved in G T A production (Schaefer et al. 2002), but the other triggers have not been thoroughly investigated. How G T A is released from R. capsulatus cells is another area where little is known. It is intriguing that limiting a nutrient such as phosphate could lead to increased G T A production and release from the cells, whereas the limitation of another component, carbon, leads to an increase in G T A production but only intracellularly with little or no release into the culture medium. The growth curve for the media types investigated provides no real insight into the differences observed (Figure 17). The maximum density for the cells grown under carbon limitation or phosphate limitation is essentially identical, although there appears to be a difference in the growth rates. It is therefore conceivable that the growth rate prior to stationary phase may affect overall G T A production within cells, and the subsequent release into the supernatant. However this does not explain the similarity in G T A production between low phosphate grown cells and cells grown in YPS rich medium. Overall it is an interesting observation that under nutrient limiting conditions G T A production can be significantly altered, to the point where G T A production in a  56  0  5  10  15  20  25  30  Time (hours)  Figure 17: Growth curve of nutrient limited samples. Y262 samples grown under a variety of nutrient limited conditions in the defined media, R C V . N , low nitrogen. C, low carbon. P, low phosphate. pH, pH 7.8 (see Materials and Methods).  57 nutrient limited R C V minimal medium can appear like that of G T A production in Y P S rich medium. As well, these data could be used to further investigate the mechanism of G T A release, as the limitation of carbon appears to increase G T A production but block G T A release from the cells. These results highlight the usefulness of the G T A immunoassay as a tool to investigate a variety of questions relating to the regulation of G T A production under differing conditions. These experiments, although preliminary appear to have the potential to lead to a better understanding of environmental signals that are sensed by cells to result in the production of GTA.  4.4.4  Epifluorescence microscopy and viral-like particle counts.  Epifluorescence microscopy (EPM) using SYBR-green is a method to detect D N A , specifically DNA-containing particles such as in D N A viruses (Hennes and Suttle 1995). Using SYBR-green stained cell-free samples of the R. capsulatus overproducer strain Y262, it was observed that there are significantly more particles than obtained from the CtrA regulatory mutant, Y C K F and the G T A structural mutant, Y G T 9 (Figure 15). The rich medium (YPS) was also found to enhance particle production in comparison to the minimal medium R C V . A l l of these observations are consistent with previous publications and confirmed with parallel experimental results using the G T A bioassay (Figure 15B). However the number of particles observed by means of E P M does not directly correlate with the number of transductants from the G T A bioassay (compare Figure 15A and 15B). This may be explained by the fact that most R. capsulatus strains are lysogenic for a number of prophage (Wall et al. 1975). As SYBR-green stains all D N A , and under a microscope one cannot distinguish G T A particles from any other type of virus-like particles, it is plausible that R. capsulatus produces a variety, (or at least one) phage, in addition to GTA. The overall trend of the E P M data shows fewer particles present in the mutant strains, which supports the typical G T A bioassay results. It could be possible to perform a baseline subtraction of particle numbers, assuming that the G T A structural gene mutant Y G T 9 (ORFg2") should have zero G T A particles and all other particles present are due to the lysogenic phage present in R. capsulatus. Although it should be noted that it is assumed that the G T A structural gene mutant YGT9, has no G T A particle  58  production, it is conceivable that G T A particles are still being produced, but are not capable of transduction due a structural modification. To overcome, or at least to investigate this issue, samples from an R. capsulatus strain that has been cured of lysogenic phage could be used in E P M . A strain that was reported to be cured of 2 known lysogenic phage is SB 1003 (Yen and Marrs 1976), and so it would be of interest to count virus-like particles in cultures of SB 1003. It is possible that there are genes present on the R. capsulatus chromosome in addition to the G T A structural gene cluster, which are needed for production of G T A particles. For example, one or more of the prophage-like gene clusters (0RcMl, 0RcPl) or other genes may contribute to the intracellular assembly of G T A particles, and/or release of G T A from the cell. 4.4.5  Future uses of the GTA immunoassay.  The anti-capsid antibodies that are the basis of the G T A immunoassay described in this thesis will provide a useful tool to investigate the mechanisms of G T A regulation by measurement of G T A capsid protein levels in response to environmental changes and gene mutations. The antisera can be used to differentiate between intracellular and cell-free G T A , to investigate questions relating to possible differences between intracellular synthesis of G T A proteins and extracellular release of G T A particles. We know that G T A structural gene transcription is maximal in the stationary phase of laboratory cultures, and that release of mature (protein-encapsidated D N A ) G T A transducing particles peaks at the same time (Yen and Marrs 1976). The mechanism of G T A release from cells is unknown, and it is possible that some of the G T A production mutants now being obtained (S. Florizone, personal communication) will synthesize G T A proteins but will not release G T A from cells. The G T A immunoassay can be used to estimate the amounts of the G T A capsid protein within and outside of cells as they enter the stationary phase of maximal production of G T A transducing particles. If there were great differences between the amounts of intracellular and extracellular G T A capsid protein (detected by Western blots of intact cells and cell-free culture filtrates; normalized by loading equivalent culture volumes on SDS PAGE), this would provide information about the amounts of G T A made within cells relative to the amounts of G T A released from cells. Generally,  59 equivalent amounts of the capsid protein are found in intracellular and cell-free fractions (Lane 2 and 3, Figure 12D), implying that there is some inefficiency in the assembly or the escape of G T A particles, as already indicated by the lack of cell lysis. Data from a variety of mutants could be related to the particular mutation being screened, allowing for further inferences into G T A regulation at the level of particle release. The G T A immunoassay could also be used to bind my rabbit antibodies to G T A particles followed by a second (anti-rabbit) antibody that was dye-labelled (such as with Texas Red). Evaluation of the colours and numbers of particles would allow one to distinguish among the numbers of: i) G T A particles that contain D N A (combined Texas Red and SYBR-green signals should be yellow; ii) G T A particles that lack D N A (Texas Red signal red, 615 nm); iii) non-GTA particles that contain D N A (SYBR-green signal green, 510 nm). [Most commonly used R. capsulatus strains are lysogenic for a variety of phage (Haselkorn et al. 2001; Wall et al. 1975).] It might be interesting to harvest intact cells of R. capsulatus in the stationary phase, disrupt the cells to release their contents, and evaluate their contents using mass-fractionated (e.g., sucrose gradients) samples in Westerns of SDS P A G E gels in order to measure the amounts of intracellular particles versus unassembled protein relative to the extracellular particles. Additionally, the G T A immunoassay could be used to survey other bacteria known to have GTA-like structural components (Figure 2), and the G T A non-producing strains of R. capsulatus (Wall et al. 1975), to screen for G T A protein synthesis in the absence of any G T A activity as assayed using the traditional bioassay method. It would also be possible to perform immuno-gold labelling, to observe the exact location of G T A particles in relation to individual donor and recipient cells. In summary, the G T A immunoassay will allow differentiation between: G T A protein synthesis; G T A particle assembly +/- D N A packaging; G T A assembly +/- release from cells. This differentiation can be applied to mutant strains of R. capsulatus (random or site-directed) in order to identify other genes involved in G T A regulation, or other species that are known to contain GTA-like gene clusters, such as R. sphaeroides.  60 5. CONCLUSION The results presented in this thesis develop a clearer picture of the evolutionary history of G T A and its homologues in a-proteobacteria as having vertical descent amongst bacteria. In addition, the regulation of G T A by quorum sensing was discovered. A Luxl homologue named Gtal is involved not only in the synthesis of a long chain (16 carbon) acyl-HSL, but also G T A activity. These results, along with previous experiments on CtrA and CckA, indicate a complex regulatory cascade to control G T A production and hence gene transfer activity. This is interesting because CtrA and CckA are typical bacterial regulators yet control the virus-like GTA. How R. capsulatus and G T A co-evolved to obtain such complex control mechanisms that appear to be helpful only to the bacterium is intriguing. The work presented in this thesis also includes the development of an additional tool to look at the regulation of GTA, the G T A immunoassay. The G T A immunoassay relies on anti-capsid polyclonal antibodies to allow differentiation between G T A in cells and cell-free G T A , as well as differences in G T A production over time and in different environmental conditions. This should be a powerful tool in future studies. R. capsulatus G T A is an unusual biological entity, which could provide insight into the evolution of viruses and their intimate relationship with bacteria. This work enabled the discovery of a new regulatory factor Gtal, involved in G T A production, as well as evolutionary insights and a new tool to look at G T A production, the G T A immunoassay. These results can now be used in concert to enable a better understanding of G T A biogenesis and function.  61 REFERENCES  Altschul SF, Madden TL, Schaffer A A , Zhang J, Zhang Z, Miller W, Lipman D J (1997) Gapped B L A S T and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25:3389-3402 Barbian K D , Minnick M F (2000) A bacteriophage-like particle from Bartonella bacilliformis. Microbiol. 146:599-609 Beatty JT, Gest H (1981) Generation of succinyl-coenzyme A in photosynthetic bacteria. Arch. Microbiol. 129:335-340 Bertani G (1999) Transduction-like gene transfer in the methanogen Methanococcus voltae. J. Bacteriol. 181:2992-3002 Blosser-Middleton R, Gray K (2001) Multiple JV-acyl homoserine lactone signals of Rhizobium leguminosarum are synthesized in a distinct temporal pattern. J. Bacteriol. 183:6771-6777 Casjens S, Hatfull G, Hendrix R (1992) Evolution of dsDNA tailed-bacteriophage genomes. Sem. Virol. 3:383-397 Cooper H M , Paterson Y (1997) Preparation of polyclonal antibodies. In: Ausbel F M , Brent R, Kingston RE, Moore DD, Smith JA, Seidman JG, Struhl K (eds) Current protocols in molecular biology. John Wiley and Sons, Inc., New York, p 11.12.111.12.9 Domian IJ, Quon K C , Shapiro L (1997) Cell type-specific phosphorylation and proteolysis of a transcriptional regulator controls the Gl-to-S transition in a bacterial cell cycle. Cell 90:415-424 Eiserling F, Pushkin A , Gingery M , Bertani G (1999) Bacteriophage-like particles associated with the gene transfer agent of Methanococcus voltae PS. J. Gen. Virol. 80:3305-3308 Fuqua W C , Winans SC (1993) A LuxR-LuxI type regulatory system activates Agrobacterium T i plasmid conjugal transfer in the presence of a plant tumor metabolite. J. Bacteriol. 176:2796-2806  62 Fuqua W C , Winans SC, Greenberg EP (1994) Quorum sensing in bacteria: the LuxRLuxl family of cell density-responsive transcriptional regulators. J. Bacteriol. 176:269-275 Gonnet G, M A C, SA. B (1994) Analysis of amino acid substitution during divergent evolution: the 400 by 400 dipeptide substitution matrix. Biochem Biophys Res Commun. 199:489-496 Haselkorn R, Lapidus A , Kogan Y , Vlcek C, Paces J, Paces V , Ulbrich P, Pecenkova T, Rebrekov D, Milgram A , Mazur M , Cox R, Kyrpides N , Ivanova N , Kapatral V , Los T, Lykidis A , Mikhailova N , Reznik G, Vasieva O, Fonstein M (2001) The Rhodobacter capsulatus genome. Photosyn. Res. 70:43-52 Henikoff S, JG H (1992) Amino acid substitution matrices from protein blocks. Proc Natl Acad Sci U S A 89 Hennes K P , Suttle C A (1995) Direct counts of viruses in natural waters and laboratory cultures by epifluorescence microscopy. Limnol. Oceanogr. 40:1050-1055 Humphrey SB, Stanton T B , Jensen NS, Zuerner R L (1997) Purification  and  characterization of VSH-1, a generalized transducing bacteriophage of Serpulina hyodysenteriae. J. Bacteriol. 179:323-329 Juhala RJ, Ford M E , Duda R L , Youlton A , Hatfull GF, Hendrix R W (2000) Genomic sequences of bacteriophages HK97 and HK022: pervasive genetic mosaicism in the lambdoid bacteriophages. J. Mol. Biol. 299:27-51 Kokjohn T A (1989) Transduction: mechanism and potential for gene transfer in the environment. In: Levy SB, Miller R V (eds) Gene transfer in the environment. McGraw-Hill, New York, p 73-97 Laemmli U K (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685 Lang AS (2000) The gene transfer agent (GTA) of Rhodobacter capsulatus. Ph.D. thesis, University of British Columbia, Vancouver, British Columbia Lang A S , Beatty JT (2000) Genetic analysis of a bacterial genetic exchange element: The gene transfer agent of Rhodobacter capsulatus. Proc. Natl. Acad. Sci. U S A 97:859-864  63 Lang A S , Beatty JT (2001) The gene transfer agent of Rhodobacter capsulatus and "constitutive transduction" in prokaryotes. Arch. Microbiol. 175:241-249 Lang A S , Beatty JT (2002) A bacterial signal transduction system controls genetic exchange and motility. J. Bacteriol. 184:913-918 Lang A S , Taylor T A , Beatty JT (2002) Evolutionary implications of phylogenetic analyses of the gene transfer agent (GTA) of Rhodobacter capsulatus. J. M o l . Evol. 55:534-543 Lawrence JG, Ochman H (1998) Molecular archaeology of the Escherichia coli genome. Proc. Natl. Acad. Sci. U S A 95:9413-9417 Levy SB, Miller R V (1989) Gene transfer in the environment. In: Atlas R M , Demain A L , Greenberg EP, Olson B H , Sayler G (eds). McGraw-Hill, New York Madigan M T (1995) Microbiology of nitrogen fixation by anoxygenic photosynthetic bacteria. In: Blankenship R E , Madigan M T , Bauer C E (eds) Anoxygenic photosynthetic bacteria. Kluwer Academic Publishers, Boston, p 915-928 Maidak B , Cole JR LT, Parker CT Jr, Saxman PR, Farris RJ, Garrity G M , Olsen GJ, Schmidt T M , Tiedje J M (2001) The RDP-II (Ribosomal Database Project). Nucleic Acids Res. 29:173-174 Marrs B L (1974) Genetic recombination in Rhodopseudomonas capsulata. Proc. Natl. Acad. Sci. U S A 71:971-973 Maurizi M R , Clark WP, Katayama Y , Rudikoff S, Pumphrey J, Bowers B , Gottesman S (1990) Sequence and structure of ClpP, the proteolytic component of the ATPdependent Clp protease of Escherichia coli. J. Biol. Chem 265:12536-12545 Mazodier P, Davies J (1991) Gene transfer between distantly related bacteria. Annu. Rev. Genet. 25:147-171 Norrander J, Kempe T, Messing J (1983) Construction of improved M13 vectors using oligodeoxynucleotide-directed mutagenesis. Gene 26:101-106 Parsek M , Greenberg E (2000) Acyl-homoserine lactone quorum sensing in gramnegative bacteria: a signaling mechanism involved in associations with higher organisms. Proc Natl Acad Sci U S A 97:8789-8793 Perriere G, M G (1996) WWW-Query: A n on-line retrieval system for biological sequence banks. Biochimie 78:364-369  64 Piper K , Beck S, Bodman V , Farrand S (1993) Conjugation factor of Agrobacterium tumefaciens regulates T i plasmid transfer by autoinduction. Nature (London) 362:21670-21676 Puskas A , Greenberg EP, Kaplan S, 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 GT, Shapiro L (1996) Cell cycle control by an essential bacterial two-component signal transduction protein. Cell 84:83-93 Rapp B J , Wall JD (1987) Genetic transfer in Desulfovibrio desulfuricans. Proc. Natl. Acad. Sci. U S A 84:9128-9130 Sambrook J, Fritsch EF, Maniatis T (1989) Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory Press, Plainview, N . Y . Schaefer A , Taylor T A , Beatty JT, Greenberg EP (2002) Long-chain acyl-homoserine lactone quorum-sensing regulation of Rhodobacter capsulatus gene transfer agent production. J Bacteriol. 184:6515-21 Schaefer A L , Hanzelka B L , Parsek M R , Greenberg EP (2000) Detection, purification and structural elucidation of the acylhomoserine lactone inducer of Vibrio fischeri luminescence and other related molecules. Methods Enzymol. 305:228-301 Scolnik P A , Haselkorn R (1984) Activation of extra copies of genes coding for nitrogenase in Rhodopseudomonas capsulata. Nature 307:289-292 Skerker J M , Shapiro L (2000) Identification and cell cycle control of a novel pilus system in Caulobacter crescentus. E M B O J. 19:3223-3234 Solioz M (1975) The gene transfer agent of Rhodopseudomonas capsulata. Ph.D. thesis, Saint Louis University, Saint Louis, M O Solioz M , Marrs B (1977) The gene transfer agent of Rhodopseudomonas capsulata: Purification and characterization of its nucleic acid. Arch. Biochem. Biophys. 181:300-307 Solioz M , Yen H-C, Marrs B (1975) Release and uptake of gene transfer agent by Rhodopseudomonas capsulata. J. Bacteriol. 123:651-657 Strimmer K , A v H (1996) Quartet puzzling: A quartet maximum likelihood method for reconstructing tree topologies. Mol. Biol. Evol. 13:964-969  65 Tabita FR (1995) The biochemistry and metabloic regulation of carbon metabolism and CO2 fixation in purple bacteria. In: Blankenship R E , Madigan M T , Bauer C E (eds) Anoxygenic photosynthetic bacteria. Kluwer Academic Publishers, Boston, p 885-914 Thompson J, Gibson TJ PF, Jeanmougin F, Higgins D G (1997) The ClustalX Windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucl. Acids Res 24:4876-4882 Towbin H , Staehelin T, Gordon J (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and applications. Proc. Natl. Acad. Sci. U S A 76:4350-4354 Wall JD, Weaver PF, Gest H (1975) Gene transfer agents, bacteriophages,  and  bacteriocins of Rhodopseudomonas capsulata. Arch. Microbiol. 105:217-224 Wasserfallen A , Nolling J PP, Reeve J, Conway de Macario E. (2000) Phylogenetic analysis of 18 thermophilic Methanobacterium isolates supports the proposals to create a new genus, Methanothermobacter gen. nov., and to reclassify several isolates in three species, Methanothermobacter thermautotrophicus comb, nov., Methanothermobacter  wolfed  comb,  nov.,  and  Methanothermobacter  marburgensis sp. nov. Int J Syst Evol Microbiol. 50:43-53 Whitehead M K , Barnard A M L , Slater H , Simpson NJL, Salmond G P C (2001) Quorum sensing in gram-negative bacteria. F E M S Microbiol. Rev. 25:217-224 Wilkoff WR, Liljas L, Duda R L , Tsuruta H , Hendrix R W , Johnson JE (2000) Topologically linked protein rings in the bacteriophage HK97 capsid. Science 289:2129-2133 Wingfield PT (1995) Overview of the purification of recombinant proteins produced in Escherichia coli. In: Coligan JE, al. e (eds) Current protocols in protein science. John Wiley and Sons, Inc., New York, p 6.1.1-6.1.22 Woese C R (1987) Bacterial evolution. Microbiol. Rev. 51:221-271 Wong D K - H , Collins WJ, Harmer A , Lilburn TG, Beatty JT (1996) Directed mutagenesis of  the  Rhodobacter  capsulatus puhA  gene and  pleiotropic effects  on  photosynthetic reaction center and light-harvesting I complexes. J. Bacteriol. 178:2334-2342  66 Yen H-C, Hu NT, Marrs B L (1979) Characterization of the gene transfer agent made by an overproducer mutant of Rhodopseudomonas capsulata. J. Mol. Biol. 131:157168 Yen H - C , Marrs B (1976) Map of genes for carotenoid and bacteriochlorophyll biosynthesis in Rhodopseudomonas capsulata. J. Bacteriol. 126:619-629  67 APPENDIX 1:  1.  Publications arising from this research  Lang A S , Taylor T A , and Beatty JT (2002) Evolutionary Implications of Phylogenetic Analyses of the Gene Transfer Agent (GTA) of Rhodobacter capsulatus. J. Mol. Evol. 55:534-543.  2.  Schaefer A L , Taylor TA, Beatty JT, and Greenberg EP (2002) Long-chain acylhomoserine lactone quorum-sensing regulation of Rhodobacter capsulatus gene transfer agent production. J. Bacteriol. 184: 6515-6521.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0099793/manifest

Comment

Related Items