UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The gene transfer agent (GTA) of Rhodobacter capsulatus Lang, Andrew Stephen 2000

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

Item Metadata

Download

Media
831-ubc_2000-565734.pdf [ 7.48MB ]
Metadata
JSON: 831-1.0089687.json
JSON-LD: 831-1.0089687-ld.json
RDF/XML (Pretty): 831-1.0089687-rdf.xml
RDF/JSON: 831-1.0089687-rdf.json
Turtle: 831-1.0089687-turtle.txt
N-Triples: 831-1.0089687-rdf-ntriples.txt
Original Record: 831-1.0089687-source.json
Full Text
831-1.0089687-fulltext.txt
Citation
831-1.0089687.ris

Full Text

THE GENE TRANSFER AGENT (GTA) OF RHODOBACTER CAPSULATUS by ANDREW STEPHEN LANG B.Sc.(Hons.), Brock University, 1994 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Microbiology and Immunology) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA March 2000 ©Andrew Stephen Lang, 2000 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, 1 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. Department of /h&friftfjG^O-f f-jfrtoWNdLOG*/ The University of British Columbia Vancouver, Canada Date ffy/lS/tf) DE-6 (2/88) ABSTRACT An unusual system of genetic exchange exists in the purple non-sulfur bacterium Rhodobacter capsulatus. DNA transmission is mediated by a small bacteriophage-like particle called the gene transfer agent (GTA) that transfers random 4.5 kb segments of the producing cell's genome to recipient cells, where allelic replacement occurs. This thesis presents the results of gene mutagenesis, cloning, and analysis experiments which show that GTA resembles a defective prophage related to bacteriophages from diverse genera of bacteria, that has been adopted by R. capsulatus for genetic exchange. A pair of cellular proteins, CckA and CtrA, appear to constitute part of a sensor kinase/response regulator signaling pathway that is involved in expression of GTA structural genes. This signaling pathway controls growth phase-dependent regulation of GTA gene messages, yielding maximal gene expression in the stationary phase. I suggest that GTA is an ancient prophage remnant that has evolved in concert with the bacterial genome, resulting in a genetic exchange process Controlled by the bacterial cell. i i i TABLE OF CONTENTS ABSTRACT ii TABLE OF CONTENTS i i i LIST OF TABLE vi LIST OF FIGURES vii ACKNOWLEDGEMENTS x 1. INTRODUCTION 1 1.1. Transduction in nature 1 1.2. Transduction, the gene transfer agent of Rhodobacter capsulatus, and other unusual transducing agents 3 1.3. Sequence relationships among bacteriophages 8 1.4. GTA and R. capsulatus 9 2. MATERIALS AND METHODS 11 2.1. Bacterial strains, growth conditions, and plasmids 11 2.2. Motility tests 17 2.3. GTA bioassays 17 2.4. Purification of GTA 17 2.5. SDS-polyacrylamide gel electrophoresis (SDS-PAGE) 18 2.6. Protein concentration. 18 2.7. Protein sequencing 18 2.8. Creation of and screening a Y262::Tn5 library for mutants with reduced production of GTA. 19 2.9. DNA manipulations 20 2.10. DNA sequencing. 20 2.11. Calculation of codon adaptation index (CAI) values 22 2.12. Alignments and database homology searches 23 2.13. Isolation of GTA gene expression mutants 23 2.14. Construction of mutants 24 2.15. Construction of ctrA expression plasmids 24 2.16. (3-Galactosidase assays 25 2.17. Southern blotting 25 2.18. Colony lifts 26 2.19. RNA analysis 27 2.20. Western blotting 27 RESULTS 29 3.1. Identification of mutants deficient in GTA production and isolation of wild type versions of the mutated genes 29 3.2. DNA sequence analysis of mutants and cloned chromosomal fragments 33 3.3. Identification of a sensor kinase involved in GTA gene expression 42 3.4. GTA purification and protein analysis 44 3.5. Sequence analysis of proteins from a GTA preparation 47 3.6. Directed mutations of GTA genes and effects on GTA production 49 3.7. Analysis of the ctrA gene 49 3.8. Transcription of the GTA structural gene cluster 63 3.9. Analysis of GTA-producing and non-producing R. capsulatus strains for the presence of GTA genes 71 3.10. Analysis of R. capsulatus strains for GTA binding capability... 71 4. DISCUSSION v 75 4.1. The GTA structural gene cluster 75 4.2. Gene expression of the GTA structural gene cluster 77 4.3. GTA release by R. capsulatus. 79 4.4. Relationship of GTA to other phages and prophages 80 4.5. Growth phase-dependence and two-component system control of GTA production 88 4.6. Genes near the ctrA gene in R. capsulatus 94 4.7. Presence of GTA genes in R. capsulatus strains in nature 95 4.8. Concluding remarks 95 5. REFERENCES 97 vi LIST OF TABLES Table 1. R. capsulatus strains 13 Table 2. Plasmids 15 Table 3. Characterization of GTA- mutants and cosmids 9H1 and 9H2...33 Table 4. Sizes, relative locations and codon adaptation indices (CAI) for predicted open reading frames 37 Table 5. Summary of database search results for predicted open reading frame products 39 vii LIST OF FIGURES Figure 1. Comparison of prototypical lysogenic phage and GTA "life cycles" 6 Figure 2. Representation of a side view through an overlay plate used for screening the Y262::Tn5 library for GTA production mutants 21 Figure 3. Map of transposon Tn5 30 Figure 4. Digitized image of an autoradiogram of a Southern blot used for characterization of Tn5 insertion sites in GTA production mutants 31 Figure 5. Map of a region of the R. capsulatus chromosome containing genes necessary for GTA production 35 Figure 6. Alignment of CckA protein sequences 43 Figure 7. Sucrose gradient centrifugation of GTA. 45 Figure 8. Equilibrium centrifugation of GTA in RbCl 46 Figure 9. Digitized image of a Coomassie blue-stained SDS-PAGE of a GTA preparation sample 48 Figure 10. Capsid protein processing by GTA and phage HK97 50 Figure 11. Comparisons of CtrA proteins and DNA binding sites 51 Figure 12. Digitized image of a result from a Southern blot used to determine the ctrA copy number. 53 Figure 13. Digitized image of an autoradiogram used for RNA blot analysis Vlll of ctrA transcripts 54 Figure 14. Digitized image of an autoradiogram used for RNA blot analysis of orfg2 transcripts 55 Figure 15. Digitized image of an autoradiogram used for RNA blot analysis of orfg4 transcripts 56 Figure 16. Evaluation of anti-CtrA antibodies and expression of the CtrA protein from C. crescentus in R. capsulatus by western blotting 58 Figure 17. Evaluating growth phase variation in CtrA levels by western blotting 59 Figure 18. Growth phase- and CtrA-dependence of p-galactosidase activities of orfg2'::7acZplasmid-borne gene fusions 61 Figure 19. CtrA-dependent motility. 64 Figure 20. Effect of ctrA mutation on photosynthetic growth rate 65 Figure 21. RNA blot analysis for comparison of GTA gene expression levels between strains Y262 and BIO 68 Figure 22. Strain differences in p-galactosidase activities of orfg2'::'iacZ plasmid-borne gene fusions 70 Figure 23. Digitized image of an autoradiogram used to analyze R. capsulatus strains for the presence of GTA genes 72 Figure 24. GTA binding capability of R. capsulatus strains 74 Figure 25. Map of the R. capsulatus chromosomal DNA present on ix cosmids 9H1 and 9H2 and the GTA structural gene cluster 78 Figure 26. Sequence relationships between GTA and <])RcMl proteins 83 Figure 27. Modular structure of the putative GTA terminase protein 87 ACKNOWLEDGEMENTS I am indebted to many people who have helped me over the years of my work on this thesis: I thank all past and present members of the Beatty lab for their support and friendship. I would especially like to thank my supervisor, Tom Beatty, who, in addition to many other things, always invested significant effort in trying to make me a better scientist. I thank M. Aklujkar for allowing me to use the computer program he wrote for calculation of codon adaptation indices. I would like to thank my advisory committee (C. Suttle, T. Warren and G. Weeks) for their helpful comments and advice; I also thank D. Rochon who served on my committee for several years before moving to the Okanagan. I am also indebted to G. Spiegelman and R. Redfield for their interest and helpful discussions. I am grateful to Robert Haselkorn (University of Chicago) for allowing me access to unpublished sequence data, and to Honza Paces for getting this data to me. I am also grateful to Lucy Shapiro (Stanford University) for provision of the anti-CtrA antibodies and the clone of the C. crescentus ctrA gene (pSALFl). I thank John Smit (UBC) for the Tn5L32 sequencing primer. Over the years I begged and borrowed many things from many labs in the Department of Microbiology and Immunology. Those that stand out most as frequent victims are the Gold, Smit and Tufaro labs so I thank the members of those labs (and especially John Nomellini from the Smit lab). Finally I thank my partner, Yolanda Morbey, for her support over the years. 1. I N T R O D U C T I O N 1.1. Transduction in nature. Bacteria exchange D N A by three mechanisms: natural transformation, conjugation, and transduction (Levy and Miller 1989; Mazodier and Davies 1991). Transformation occurs when naked D N A is taken up from the environment. Conjugation results in transfer of plasmid or transposon sequences, and possibly genomic sequences if a mobile genetic element has integrated into the chromosome. Transduction occurs when bacterial D N A is packaged in a bacteriophage (phage) particle and subsequently injected into another cell. In all three mechanisms, incoming D N A may either be degraded and the nucleotides recycled, replicate autonomously of the chromosome, or be incorporated into the chromosome. Exchange of D N A by bacteria, or horizontal gene transfer, plays an important role in the evolution of bacterial genomes. A recent analysis of the complete genome sequence of Escherichia coli was very revealing in this respect (Lawrence and Ochman 1998). The authors identified orfs that originated from horizontal gene transfer, and subsequently assessed the length of time those orfs had been present in the genome. These orfs were identified on the basis of their unusual (for E. coli) sequence characteristics (GC content). They estimated that 17.6% of orfs in the genome (approximately 550 kb) arrived by horizontal transfer in the past 100 million years. These new genes were often found nearby tRNA genes, which commonly are phage integration sites (Cheetham and Katz 1995). This implicates many past transduction events, although there were usually no corresponding phage genes in these sites. Observations in aquatic environments show that there are staggering numbers of virus particles (up to lO^/ml) in nature (Bergh et al. 1989; Hennes and Suttle 1995). Transducing bacteriophages have been found for a wide variety of bacterial species and it seems possible that transduction occurs in each of these species. Transduction is therefore likely to be important in the exchange of D N A in nature (Kokjohn 1989), although it is limited by several factors. The first of these is a specific interaction between the bacteriophage and a cell-surface receptor which limits potential hosts to those that have an appropriate receptor. This is reflected by the observation that most phages infect only one species. Phage D N A (of most phages) is double-stranded and is therefore susceptible to recipient cell restriction systems if it lacks the required modification. Also, cell-free phage are susceptible to damage from proteases or ultra violet radiation that may be present in the environment. Phage mediated transfer has been shown to be an important mechanism for spread of virulence genes (Cheetham and Katz 1995) and a good illustration of this is the vap region from Dichelobacter nodosus. The vap region consists of many orfs and has been found to be involved in virulence of D. nodosus (Cheetham and Katz 1995). The region is bounded on each side by a repeated sequence that resembles a phage attachment site, and an orf at one end of the region is highly similar to integrase genes from several different phages (Cheetham et al. 1995). This region is located next to a tRNA gene in the D. nodosus chromosome, a common integration site for phages (Cheetham and Katz 1995). This seems to be a compelling example of transmission of virulence gene by a bacteriophage vector, and there are other similar examples (Cheetham and Katz 1995). Another, more direct example of phage-dependent virulence that warrants mention by virtue of its novelty is in the bacterium Vibrio cholerae. In V. cholerae there are two requirements for virulence: a pathogenicity island (VPI) (Karaolis et al. 1998) and the phage C T X O that encodes the cholera toxin (Waldor and Mekalanos 1996). What makes this situation novel is that the pathogenicity island itself turns out to be a filamentous phage, V P I O , and it is the coat protein of VPIO that acts as the receptor for C T X O (Karaolis et al. 1999). Transduction is also one mechanism by which antibiotic resistance genes are transmitted between bacteria (Davies 1994) which is currently a serious medical concern. 1.2. Transduction, the gene transfer agent of Rhodobacter capsulatus, and other unusual transducing agents. Transduction was discovered during attempts to demonstrate conjugation in Salmonella typhimurium (Zinder and Lederberg 1952). Instead, generalized transduction mediated by the phage now known as P22 was found. This was an important discovery since transduction has become an important tool in the study of bacteria. Two important uses have been strain construction and gene mapping. In 1974, a novel system of genetic exchange, termed the gene transfer agent (GTA) , was discovered in the purple nonsulfur bacterium Rhodobacter capsulatus (Marrs 1974). G T A resembles a small tailed bacteriophage, with a density of 1.33 g/ml (Solioz and Marrs 1977), and a head diameter of approximately 30 nm (Yen et al. 1979). The nucleic acid within the particle is linear double-stranded D N A of approximately 4.5 kb in length (Solioz and Marrs 1977). G T A packages genomic D N A indiscriminately; restriction digestion patterns and Cot analysis indicated that there are no detectable preferred D N A sequences in particles (Yen et al. 1979). G T A also displays no plaque forming activity (Marrs 1974). Most strains of R. capsulatus can both produce G T A and take up GTA-borne D N A , although some strains only do one or the other, and some strains do neither (Marrs 1974; Wall et al. 191 A). G T A mediates generalized transduction because it has been found that all genetic markers tested can be transferred from donor to recipient cells via G T A , whether the trait is present in donor cells on the chromosome or on extra-chromosomal elements (Solioz and Marrs 1977; Yen et al. 1979; Scolnik and Haselkorn 1984). When G T A is in excess, the maximum frequency of transfer to recipient cells for a single marker is approximately 4 x 10~4 per recipient (Solioz et al. 1975). Early studies of G T A were complicated by the fact that it was produced in very low levels, although its activity could be detected with a transduction bioassay (Solioz et al. 1975). The study of G T A became somewhat easier after the isolation of an overproducer mutant that produces G T A at levels approximately three orders of magnitude higher than wild-type strains (Yen et al. 1979). It had been shown that G T A activity was DNase resistant (Marrs 1974) and pronase sensitive (Solioz 1975), but the purification of particles from the overproducer strain allowed direct demonstration that gene transfer activity occurred via a phage-like particle, as visualized by electron microscopy (Yen et al. 1979). It was concluded that the overproducer strain Y262 contains two mutations separated by more than 4.5 kb, because two successive treatments with G T A , followed each time by screening for increases in G T A production, were required to transfer the overproducing trait to a different strain (Yen et al. 1979). G T A is produced in the greatest quantities when growth occurs in a rich as opposed to a minimal medium (Solioz 1975), similar to bacteriophages in Bacillus subtilis (Webb et al. 1982), Pseudomonas aeruginosa (Kokjohn et al. 1991) and Escherichia coli (Hada et al. 1997), in which it has been shown that the amount of progeny virus released is dependent upon the metabolic activity of the host. G T A yield is also much greater from photosynthetic as opposed to aerobic cultures (Solioz 1975), but it is not clear why. For the virulent R. capsulatus phage, RC1, there is no difference in burst size between the two growth conditions when the virus is added at low multiplicity of infection (moi), but more virus is produced under photosynthetic conditions at higher moi (Schmidt et al. 1974). The levels of G T A in the medium reach a maximum after the culture enters stationary phase (Solioz et al. 1975). The G T A "life cycle" is unlike that of a typical phage (Figure 1). Unlike a phage, the path to production of G T A particles does not begin with injection of D N A into a cell, but rather with expression of genes somewhere within the existing genome. As mentioned above, G T A particles contain random segments of the genome which can then be transduced to other cells. Although phages capable of transduction are common in nature, the frequency of a transducing particle in a population of phage particles produced after replication is usually rare (Hayes 1968), which contrasts directly with G T A . However, the end product which is released from cells is the most striking difference between G T A and a typical phage. Most particles produced by a phage infection will be infectious particles capable of infecting, and eventually lysing, another target cell whereas G T A only donates host D N A to its target cell. Other phage-like agents of genomic gene transfer that functionally resemble G T A have subsequently been discovered in several other bacteria. In Desulfovibrio desulfuricans, a tailed phage-like particle named D d l was discovered that packages and transduces random 13.6 kb fragments (Rapp and Wall 1987). A similar agent named VSH-1 was found in the bacterium Serpulina hyodysenteriae (Humphrey et al. 1997). VSH-1 has a tailed phage structure and appears to package and transduce 7.5 kb random chromosomal fragments. A transducing phage named V T A was found in Methanococcus voltae which apparently packages random 4.4 kb replication ^ induct ion ^ lysogeny divis ion B . transduction lysis (?) Figure 1. Comparison of prototypical temperate phage and G T A "life cycles". A . Simplified outline of a lysogenic phage life cycle. A lytic phage would exclusively follow the lysis pathway (dashed arrow). Phage-specific D N A is indicated by a hatched pattern, and this is the material most often packaged in phage particles after replication. B. G T A life cycle. G T A particles contain fragments from the producing cell's genome, and the D N A responsible for production of the particles (hatched) is not packaged more frequently than other sequences. Released particles can transduce another R. capsulatus cell with the packaged genomic D N A . chromosomal D N A fragments (Bertani 1999). V T A also has a tailed-phage structure with dimensions very similar to those reported for G T A (Eiserling et al. 1999). Therefore, at present there are 4 known constitutive generalized transducing phages. A l l four package double-stranded D N A , have a tailed-phage structure, and seem to package less D N A than would be expected to represent a complete tailed phage genome (with the possible exception of Ddl) . 1.3. Sequence relationships among bacteriophages. A recent paper examined the sequence relationships between double-stranded D N A phages from diverse genera of bacteria (Hendrix et al. 1999). The model that emerged from this work is that most, if not all, double-stranded D N A phages appear to share a common ancestry and are active in horizontal genetic exchange. Phages from distantly related bacterial genera are unlikely to directly exchange genes, and the exchange is thought to proceed through a long set of intermediary phages and prophages. This has produced phage genomes wherein the genes show a mosaic pattern of sequence relationships with other phage genes. Some support for this model includes the sequence relationships found between the phages HK97 (from E. coli), L5 (from Mycobacterium spp.) and (|)C31 (from Streptomyces spp.). There are protein sequence similarities between L5 and (J)C31, between HK97 and 0C31, but not between L5 and HK97. Therefore, some HK97 and L5 protein sequences have recognizable similarity to (bC 31 protein sequences, but HK97 and L5 do not show any direct protein sequence relationships. In addition, all three phages have a similar genomic organization. This evidence supports the idea that there was a common ancestor of these phages. The sequence relationships found for the genes of a cryptic prophage found in Haemophilus influenzae, (j)flu, provide a good example in support of the horizontal exchange part of the model. Different genes from this prophage show similarity to phage genes from a broad spectrum of bacterial genera (Bacillus, Escherichia, Haemophilus, Lactococcus, Mycobacterium, Salmonella, and Streptococcus), although adjacent <))flu genes are not similar to genes from the same phage. For example, a cpflu gene that is similar to a Mycobacterium phage gene is next to a gene that is similar to an Escherichia phage gene which is next to a gene that is similar to a Lactococcus phage gene. Overall, these and other examples support the model of common ancestry for double-stranded D N A phages and the subsequent occurrence of horizontal genetic exchange by the evolutionary precursors of the modern day phages and prophages (Hendrix et al. 1999). 1.4. GTA and R. capsulatus. R. capsulatus, a purple nonsulfur bacterium, is able to use several different means of energy generation including anaerobic photosynthesis and aerobic respiration, and is capable of fixing nitrogen and carbon dioxide (Madigan 1995; Tabita 1995). This makes it a useful organism for the study of photosynthesis, since mutations that abolish photosynthetic growth are not lethal, and R. capsulatus has also been used to study a variety of other interesting metabolic processes (Scolnik and Marrs 1987). In addition, the expression of photosynthesis genes, and therefore the subsequent assembly of the photosynthetic apparatus, can be modulated easily by controlling oxygen tension. Because the bacteria continue to grow under these conditions, temporal processes during induction and repression can be studied in genetic analyses (Scolnik and Marrs 1987). The discovery of G T A was and still is of significance due to its use in genetic manipulations of R. capsulatus, particularly since there is no method for transforming R. capsulatus. G T A is commonly used for mediating allelic exchange as long as there is sufficient similarity in 10 flanking sequences to allow gene replacement by homologous recombination. This gene replacement is dependent upon a cellular system that includes reck (Genthner and Wall 1984). This allows the replacement of chromosomal genes with disrupted versions of cloned genes ("interposon" mutagenesis) to be performed without the complexities associated with the use of conjugative integrative (or suicide) plasmids (Scolnik and Haselkorn 1984). G T A has also been used for gene mapping by measuring cotransfer frequencies to establish linkages for the genes involved in carotenoid and bacteriochlorophyll biosynthesis (Yen and Marrs 1976). The unusual properties of G T A led to the suggestion that it might represent an evolutionary precursor of a phage or, alternatively, a defective phage (Yen et al. 1979). This thesis was undertaken to attempt to distinguish between these two possibilities. 2. M A T E R I A L S A N D M E T H O D S 2.1. Bacterial strains, growth conditions, and plasmids. The E. coli strains used for cloning and subcloning were DH5a (GIBCO-BRL) , DH10B (GIBCO-BRL) , and the dam mutant RB404 (Brent and Ptashne 1980). Strains S17-1 (Simon et al. 1983), C600(pDPT51) (Taylor et al. 1983), and HB101(pRK2013) (Ditta et al. 1985) were used to conjugate plasmids into R. capsulatus. The E. coli strains were grown in L B medium (Sambrook et al. 1989) supplemented with the appropriate antibiotics at the following concentrations: ampicillin, 200 n g / m l ; tetracycline-HCl, 10 pg/ml; kanamycin sulfate 50 pg/ml; gentamycin sulfate, 8 u,g/ml; spectinomycin-2HC1, 30 n g / m l . Caulobacter crescentus strain CB15A ( A T C C 19089) was grown in P Y E medium (Ely 1991). The R. capsulatus strains used are described in Table 1. Strains Y262 (Yen et al. 1979) and B10 (Marrs 1974) were used as the starting strains for subsequent strain construction (see below) as well as the sources of D N A for sequencing. Other D N A sequences used for analysis were from the closely related strain SB 1003 (Yen and Marrs 1976), which is the source of D N A used by the group sequencing the R. capsulatus genome (http://rhodoL.uchicago.edu/capsulapedia/capsulapedia/capsulapedia.sht ml). R. capsulatus strains were grown aerobically in R C V medium (Beatty and Gest 1981) supplemented with the appropriate antibiotics at the following concentrations: tetracycline-HCl, 0.5 n g / m l ; kanamycin sulfate, 10 u,g/ml; gentamycin sulfate, 3 n g / m l ; spectinomycin-2HCl, 10 p-g/ml; or photosynthetically in YPS medium (Wall et al. 1974). The turbidity of cultures was monitored by measuring light scattering with a Klett-Summerson photometer (filter #66; red); 100 Klett units represents approximately 4 x 10$ colony forming units per ml. Table 1. R. capsulatus strains. 1 3 Strain Genotype Phenotype Reference BIO wild type SB 1003 rif-10 Y262 unknown mutations BY1653 bchA165, crtB4 DW5 YGT2 YGT7 YGT9 YGT24 YGT27 YGT31 YGT23 Y T L 3 puhA orfgl8::Tn5 orfgl8::Tn5 orfg2::Tn5 ctrA::Tn5 ctrA ::Tn5 orfgl2::Tn5 orfUl::Tn5 cckA: :Tn5 Y T L 4 intergenic Tn5 insertion YCKF ctrA: :KIXX BCKF crM::KIXX YKKR2 cckA::KlXX B6 natural isolate YW1 natural isolate wild type wild type G T A overproducer no photosynthesis, pale colour no photosynthesis no G T A production no G T A production no G T A production no G T A production no G T A production no G T A production no G T A production reduced G T A production reduced G T A production no G T A production no G T A production reduced G T A production no G T A production no G T A production, no G T A uptake (Marrs 1974) (Yen and Marrs 1976) (Yen et al. 1979) (Marrs 1981) (Wong et al. 1996) this thesis this thesis this thesis this thesis this thesis this thesis this thesis this thesis this thesis this thesis this thesis this thesis (Wall et al. 1974) (Wall et al. 1974) Y W 2 natural isolate no G T A production (Wall et al. 1974) SP36 natural isolate „ no G T A production, (Wall et al. 197'4) no G T A uptake Table 2. Plasmids. Plasmid Description Source or reference pUC13 pUC18 pUC19 pBluescript-I-KS (pBS) pDPT51 pRK2013 pSUP2021 pXCA601 pUC4KIXX p L A F R l 9H1, 9H2, C H I pRR5C pCTRA cloning vector, lacZa, A m p r cloning vector, lacZa, A m p r cloning vector, lacZa, A m p r cloning vector, lacZa, A m p r mobilizing vector, A m p r , T p r mobilizing vector, K a n r Tn5 donor plasmid, A m p r , C m r , K a n r promoter probe vector, T c r source of K I X X cartridge, Amp r , K a n r cosmid vector, T c r (Vieira and Messing 1982) (Norrander et al. 1983) (Norrander et al. 1983) Stratagene (Taylor et al. 1983) (Ditta et al. 1985) (Simon et al. 1983) (Adams et al. 1989) (Barany 1985) (Friedman et al. 1982) J.T. Beatty (Young et al. 1998) R. capsulatus genomic D N A cloned in p L A F R l expression vector for R. capsulatus, Gm1" R. capsulatus ctrA gene in pRR5C as this thesis a EcoR I to Sma I fragment 1 6 pCCC pRRR p9KC-2 p E / S 4 p X / S 5 pYP pYNP p9HSTU pSTU12 p9H52 pKHF7 C. crescentus ctrA gene in pRR5C as a this thesis Apa I to Sma I fragment R. capsulatus ctrA gene in pRR5C as this thesis a Apa I to Sma I fragment Kpn I to Cla I fragment from strain this thesis Y G T 9 containing Tn5 and sequences surrounding the insertion site blunted EcoR I to Sal I fragment this thesis from p9KC-2 in Sma I site of pBS blunted Xho I to Sal I fragment from this thesis p9KC-2 in Sma I site of pBS Pst I to BamK I fragment from this thesis pX/S5 in pXCA601 Pst I to BamH I fragment from this thesis pE/S4 in pXCA601 Kpn I to Hind III fragment from 9H1 this thesis in pUC19 Pst I to blunted Hind III fragment this thesis from p9HSTU in pXCA601 3.7 kb EcoR I fragment from 9H1 in this thesis pUC13 pUC18 with blunted EcoR I fragment this thesis from p9H52 in Hinc II site 17 2.2. Motility tests. Motility was tested as described (Krieg and Gerhardt 1981) by stabbing YPS agar (0.4%) and incubating the tubes under photosynthetic conditions for 4 days. 2.3. GTA bioassays. Bioassays for G T A activity were done as described (Solioz et al. 1975), with minor modifications. A small volume (not more than 100 pi) of liquid to be assayed was mixed with 100 pi of the indicator strain cells and 400 pi of G-buffer [10 m M Tris-Cl (pH 7.8), 1 m M MgCl2, 1 m M CaCl2, 1 m M NaCl, 500 pg/ml B S A Fr. V] (Solioz et al. 1975). The indicator strain (the pub,A deletion mutant strain DW5) was prepared by centrifuging an overnight culture and resuspending the cells in an equal volume of G-buffer. This mixture was incubated with gentle agitation for 1 hour at 3 0 - 3 5 ° C after which time 900 pi of R C V medium was added and the mixture incubated under the same conditions for a further 3-4 hours. The cells were plated on R C V plates which were incubated under photosynthetic conditions in anaerobic Gas-Pak jars for 2-3 days. Plates were removed and the numbers of puhA+ colonies counted. 2.4. Purification of GTA. G T A was purified essentially as described (Yen et al. 1979). A culture of strain Y262 was grown under photosynthetic conditions in YPS medium in a 15 litre fermentor. After 3 days (at least 24 hours after entry into the stationary phase), the cells were removed by centrifugation followed by filtration (0.45 pm) of the supernatant liquid, and the filtrate was concentrated 10 fold on a 100 kDa molecular weight cut-off membrane (Millipore). Bulk macromolecules were precipitated from this concentrate by addition of NaCl to a final concentration of 0.5 M and polyethylene glycol 6000 (PEG) to 10%. This mixture was incubated on ice overnight and the resulting precipitate recovered by centrifugation at 17 000 x g in a Beckman JA-10 rotor for 30 minutes. The precipitate was resuspended in 10 ml of G-buffer and 0.5 ml aliquots were layered over 30 ml of 10-30% (w/w) linear sucrose gradients centrifuged at 40 000 x g in a Beckman SW28 rotor for 15 hours at 4 ° C . Sucrose gradient solutions were made in G-buffer without B S A to prevent contamination of S D S - P A G E samples by excessive amounts of this protein. Fractions of 75 drops were collected and those found to contain peak G T A activity were pooled. This was then mixed with 0.5 volumes of saturated (at 4 ° C ) rubidium chloride (RbCl) and this mixture used for equilibrium centrifugation (at 140 000 x g in a Beckman 50Ti rotor for 50 hours at 2 0 ° C ) . In some cases, the sucrose gradient fractions were concentrated 10-fold on 10 kDa molecular weight cut-off spin membranes (Millipore) before equilibrium centrifugation. Fractions were collected and assayed for G T A activity and refractive index (RI). The RI values were converted to density using the equation p=8.7719nD-10.7054, where p is the density and nD is the refractive index at 25 °C (Solioz 1975). 2.5. SDS-polyacrylamide gel electrophoresis (SDS-PAGE). The Laemmli buffer system (Laemmli 1970) was used to separate proteins on 10% or 12% acrylamide separating gels with 4% stacking gels (acrylamide/bis 29:1). Protein samples were boiled for 5 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 on the gel. Gels were run with the P R O T E A N II or M i n i - P R O T E A N II systems (Bio-Rad) according to the manufacturer's specifications. 2.6. Protein concentration. Proteins were sometimes precipitated with 4 volumes of - 2 0 ° C acetone (Tufaro et al. 1987) and the resultant pellet dried, dissolved in sample loading buffer and boiled for 5 minutes before loading onto S D S - P A G E gels. 2.7. Protein sequencing. For sequencing, proteins were separated by S D S - P A G E and blotted onto a P V D F membrane (Schleicher & Schuell). The S D S - P A G E were performed in the same manner as described above with minor modifications. Protein samples were boiled in sample buffer for 2 minutes prior to loading on the gel, and all electrophoresis and blotting solutions were made in Mi l i -Q water. The blotting was done in a Trans-Blot apparatus (Bio-Rad) according to manufacturer's recommendations in Towbin buffer (Towbin et al. 1979). The N-terminal sequences of four proteins of approximately 30, 33, 40 and 43 kDa were obtained by the University of British Columbia Nucleic Acid and Protein Service Unit (NAPS). 2.8. Creation of and screening a Y262::Tn5 library for mutants with reduced production of GTA. A Tn5 mutant library of approximately 2000 members was constructed from the R. capsulatus strain Y262 by use of E. coli S17-l(pSUP2021) (Simon et al. 1983). This involved performing a conjugation between Y262 and S17-l(pSUP2021) and subsequently selecting for kanamycin resistance and growth on R C V medium ( which auxotrophic E. coli donors cannot use). The plasmid pSUP2021 carries Tn5 and cannot replicate in R. capsulatus, and so the resulting kanamycin resistant Y262 cells should have Tn5 inserted somewhere in the genome due to a transposition event. Approximately 2000 colonies that resulted from this process were scraped from plates and pooled to form the Y262::Tn5 library. The screening method used to obtain G T A - mutants was essentially the same as described (Yen et al. 1979) but the opposite result was sought. Three ml of soft-agar (0.6%) G-buffer that contained 50-100 Y262::Tn5 cells from the library was layered over 15 ml of YPS agar and these plates were incubated for 2 days under photosynthetic conditions. These plates were then overlayed with 9 ml of soft-agar G-buffer, followed by 3 ml of the same containing 2-4 x 10^ cells of the pale coloured non-20 photosynthetic mutant strain BY1653. These plates were then incubated under aerobic conditions for 1 day to allow for G T A uptake by BY1653 cells before being placed under photosynthetic conditions. This selected for BY1653 transductants in the overlay that received a functional copy of the bchA gene via G T A , and hence were capable of photosynthetic growth. Colonies in the Y262::Tn5 layer that had a cloud of BY1653 transductants above them in the top layer were still capable of producing G T A (and therefore capable of transferring the wild type version of the bacteriochlorophyll biosynthesis gene which is mutated in strain BY1653). Colonies in the Y262::Tn5 layer that did not have a cloud of BY1653 transductants above them in the top layer were isolated and quantitatively evaluated for G T A production in more accurate bioassays. This screening method is represented in Figure 2. 2.9. D N A manipulations. Standard methods of D N A purification, restriction enzyme digestion and other modification techniques were used (Sambrook et al. 1989). 2.10. D N A sequencing. Sequencing was performed by either the U B C NAPS facility or in I. Sadowsky's lab (Biochemistry and Molecular Biology Department, U B C ) using plasmid templates that had been purified using the QIAprep Miniprep Kit (Qiagen). Tn5 insertion sites were sequenced by cloning the region of interest from the chromosome of mutants into pUC13 (Vieira and Messing 1982), pUC19 (Norrander et al. 1983) or pBluescript-I-KS (pBS) (Stratagene). Mutant chromosomal D N A was digested with EcoR I, BamH I, or BamH I and EcoR I together, and the appropriately sized restriction fragments (determined by Southern blotting) were purified from agarose gels. These D N A fragments were ligated with the vector and clones selected by resistance to kanamycin conferred by the neo-containing portion of Tn5. 2 1 '-'ii.ii.'CZ BY1653 buffer-agar Y262::Tn5 YPS agar Figure 2. Representation of a side view through an overlay plate used for screening the Y262::Tn5 library for G T A production mutants. Y262::Tn5 colonies lacking a visible "halo" of photosynthesis-competent BY1653 transductants in the top layer were chosen as potential G T A production mutants. 22 Inserts were then sequenced with the appropriate vector-based primer(s), and a primer complementary to the end of Tn5 (Tn5L32: 5'-A A A C G G G A A A G G T T C - 3 ' ; a generous gift of J. Smit) if appropriate. Restriction fragments (EcoR I) from cosmids 9H1 and 9H2 (see Results section) were subcloned in pUC13 and partially sequenced with the ml3R and -21ml3 primers (NAPS). One 3762 bp EcoR I fragment, from plasmid p9H52 and which contains the orfg2-orfg5' region (see Results section), was made blunt with Klenow enzyme and ligated into the Hinc II site of pUC18 (Norrander et al. 1983). The resultant plasmid, pKHF7, was then used to construct a set of plasmids by performing unidirectional deletions with the Exo-Size kit (NEB). This was done by digesting with Sst I and Xba I and subsequently performing a timed reaction with exonuclease III, followed by treatment with mung bean nuclease, and then ligase. Resultant clones were screened on the basis of size and 15 were sequenced with the M13/pUC reverse sequencing primer (-48) (NEB). This gave an overlapping set of sequences that comprised most of the 3762 bp fragment. Three gaps in the sequence data were filled by synthesis of specific primers adjacent to these gaps. The ctrA gene sequence and the partial cckA gene sequence were obtained using a sub-cloning approach and the -2 lml3 , ml3R, T3 and T7 primers (NAPS). Other sequences used for analysis were obtained from the research group sequencing the genome of R. capsulatus strain SB 1003 (http://rhodoL.uchicago.edu/capsulapedia/capsulapedia/capsulapedia.sht ml). 2.11. Calculation of codon adaptation index (CAI) values. The preferred R. capsulatus codon usage was calculated based on the highly expressed genes of the photosynthesis gene cluster (accession number 23 Z l 1165) and this information used as described (Sharp and L i 1987) in a computer program written by M . Aklujkar to calculate C A I values for open reading frames (orfs). These C A I values, which can range from 0 to 1, are a measure of the degree to which individual orfs contain the preferred codons used in highly expressed genes. A high C A I value indicates a potential for high-level expression whereas a low C A I value indicates a gene that is not well adapted for high-level expression. 2.12. Alignments and database homology searches. D N A and protein sequence alignments were done with the program GeneWorks (Oxford Molecular). Database searches were performed with the predicted amino acid sequences of orfs using the B L A S T P program (Altschul et al. 1997). The protein sequence alignments that resulted from the B L A S T P searches were also used for calculation of % identities of amino acid sequence alignments. 2.13. Isolation of GTA gene expression mutants. The plasmid pSTU12 was constructed by inserting an orfg2 Pst I to Hind III fragment (from plasmid p9HSTU, and originally from strain BIO) that had been made blunt at the Hind III end with mung bean nuclease into the plasmid pXCA601 (Adams et al. 1989), which had been digested with Pst I and BamR 1 (with the BamH I end made blunt with the Klenow enzyme). This resulted in a translationally in-frame fusion between orfg2' and 'lacZ under the control of the G T A gene promoter. This plasmid (pSTU12) was then conjugated into the Y262::Tn5 library. This library was screened for colonies displaying weak (3-galactosidase activity after growth under photosynthetic conditions on YPS plates containing X-gal. Colonies with little or no blue colour were then tested individually for G T A production in a G T A bioassay. 24 2.14. Construction of mutants. Most gene disruptions were made by ligating the neo gene-containing Sma I fragment (KIXX cartridge) from the plasmid p U C 4 K I X X (Barany 1985) into the gene to be disrupted. These mutations were made on plasmids and subsequently transferred to the R. capsulatus chromosome by G T A transduction as described (Scolnik and Haselkorn 1984). Strains B C K F and Y C K F were constructed from strain BIO and Y262, respectively, by deleting the ctrA coding region between nucleotides 124 and 672 by digestion with Hind III and Bel I, making the ends blunt with Klenow enzyme, and inserting the K I X X cartridge in this site. Strain Y K K R 2 was constructed from strain Y262 by inserting the K I X X cartridge in the cckA gene at the blunted (Klenow) Bgl II site, 606 nucleotides from the 3' end of the coding sequence. Strain Y P K F was made by inserting the K I X X cartridge in the G T A gene orfg4. It was inserted in an EcoR V site which disrupts orfg4 366 nucleotides 3' of the predicted start codon. Strain Y K 8 was made by inserting the K I X X cartridge in the blunted (Klenow) Cla I site of orfg2, which disrupts the gene 269 nucleotides from the start codon. 2.15. Construction of ctrA expression plasmids. The plasmid p C T R A was constructed by inserting the strain BIO ctrA gene 3' of the puf promoter in plasmid pRR5C (Young et al. 1998) as an EcoR I to Sma I fragment. The plasmid p C C C was constructed by inserting the C. crescentus ctrA gene (a generous gift from L . Shapiro) 3' of the puf promoter in plasmid pRR5C as an Apa I to Sma I fragment. Plasmid pRRR was constructed by insertion of the R. capsulatus ctrA gene 3' of the puf promoter in pRR5C as an Apa I to Sma I fragment. 25 2.16. (3-Galactosidase assays. A Xho I to Hind III fragment containing the G T A promoter region and some G T A gene coding sequences from the Y262 chromosome was obtained from p9KC-2, and used as the source of orfg2 5' sequences for construction of the orfg2'::7acZ fusions used in p-galactosidase assays. A Xho I to Sal I fragment, made blunt with the Klenow enzyme, was ligated into the Sma I site of pBS, and a Pst I to BamH I fragment removed from the resultant plasmid and inserted into the promoter probe vector pXCA601 (Adams et al. 1989) to generate pYP, which contains a translationally in-frame fusion between orfg2' and 'lacZ under the control of the G T A promoter. An analogous procedure was used to generate pYNP, which lacks the G T A promoter: a Klenow blunted EcoR I to Sal I fragment was ligated into the Sma I site of pBS and a Pst I to BamH I fragment from the resultant plasmid subcloned into pXCA601, again producing an orfg2'::7acZ translational fusion. Photosynthetically grown R. capsulatus cells were harvested at the desired point in the growth phase, as determined by culture turbidity, and P-galactosidase activities determined (by a colourometric assay with o-nitrophenol-P-D-galactoside) on duplicate samples as described (Miller 1992). Activities are reported as Miller units which are proportional to the increase in o-nitrophenol per minute per bacterium (it is approximated that 3 Miller units of P-galactosidase hydrolyze 1 nmole of o-nitrophenol-P-D-galactoside per minute per mg protein) (Miller 1992). 2.17. Southern blotting. D N A samples to be analyzed were separated on 1 % agarose gels and blotted onto nylon membranes (ICN) as described (Sambrook et al. 1989). D N A probes were 32p radio-labeled with the Rediprime kit (Amersham) or digoxigenin-dUTP (DIG)-labeled with the DIG D N A labeling and detection kit (Boehringer Mannheim). DIG-labeled probes were used for hybridization and detection according to kit 26 specifications. Radio-labeled probes were hybridized overnight at 6 5 ° C in hybridization buffer [5X SSPE (Sambrook et al. 1989), 1% SDS, 0.1 mg/ml denatured sheared salmon sperm D N A ] . Blots were then washed twice for 5 minutes at room temperature in 2X SSPE-0.1% SDS followed by two 15 minute washes at 6 8 ° C in 0.2X SSPE-0.1% SDS. Blots with radio-labeled probes were placed on X-ray film in the presence of an intensifying screen at - 8 0 ° C for varying lengths of time. Blots were stripped before reprobing according to the manufacturer's (ICN) recommendations. The D N A fragment used as a ctrA probe was a 1.1 kb EcoR I to Pvu I fragment that contains all of the ctrA gene as well as 170 bp of dnlJ coding sequences. The 3762 bp EcoR I fragment that was completely sequenced (containing from 50 bp 5' of the orfg2 start codon to 300 bp 3' of the orfg5 start codon) was used for panel B of Figure 23. The D N A fragment used to probe the original Tn5 mutants (Figure 4) was a 500 bp Hinc II to Pst I fragment from the "left end" of Tn5 (Figure 3). 2.18. Colony lifts. A R. capsulatus strain B10 genomic library (kindly provided by J.T. Beatty) in cosmid p L A F R l (Friedman et al. 1982) was screened for clones that contained the D N A sequences present next to the transposon in mutants Y G T 9 and Y T L 3 . Colonies to be screened were blotted onto 541 filter paper (Whatman). The filters were then treated with 0.5 N NaOH for 5 minutes, twice with 1 M Tris-Cl (pH 7.0) for 10 minutes, and 1.5 M NaCl-0.5 M Tris (pH 7.5) for 5 minutes. The filters were then briefly rinsed with ethanol (squirted from a bottle), dried, and baked at 8 0 ° C for 2 hours. Hybridizations were carried out with radio-labeled probes as described above. The D N A fragment from mutant Y G T 9 used as a probe was an approximately 3 kb Pst I fragment that contained 0.7 kb from the "left end" of Tn5 and 2.3 kb of R. capsulatus sequences next to the insertion site. 27 The D N A fragment used as a probe from mutant Y T L 3 was a 1.5 kb Hind III fragment that contained approximately 1.2 kb of Tn5 sequences and 300 bp of R. capsulatus genomic sequences next to the Tn5 insertion site. 2.19. R N A analysis. Photosynthetically grown R. capsulatus cells were harvested at the desired point in the growth phase, as determined by culture turbidity, and R N A isolated using the RNeasy kit (Qiagen). R N A electrophoresis and blotting were done as described (LeBlanc et al. 1999). D N A probes were 32p radio-labeled with the Rediprime kit (Amersham). Hybridizations were done overnight at 4 2 ° C in hybridization buffer [50% formamide, 10% dextran sulfate, 1.0 M NaCl, 50 m M Tris-Cl (pH 7.5), 0.2% BSA, 0.2% ficoll, 0.1% sodium pyrophosphate, 1% SDS, 0.2% polyvinylpyrrolidone, 0.1 mg/ml denatured sheared salmon sperm D N A ] . Hybridized blots were washed as follows: two 5 minute washes in 2X SSC (Sambrook et al. 1989) at room temperature, two 15 minute washes in 2X SSC-1 % SDS at 60 °C, and one 15 minute wash in 0.1X SSC at room temperature. Blots were then placed on X-ray film in the presence of an intensifying screen at - 8 0 ° C for varying lengths of time. Blots were stripped before reprobing according to manufacturer's (ICN) recommendations. The D N A fragment used as a ctrA probe was a Hind III to Pvu I fragment that contains from 121 bp 3' of the ctrA start codon through the 3' end of ctrA (approximately 600 bp) and 170 bp of dnlJ. The orfg2 probe was a EcoR I to Hind III fragment that contains approximately 500 bp of the coding region of orfg2. The orfg4 probe was a 500 bp Pst I fragment that contains almost the entire orfg4 coding sequence. 2.20. Western blotting. Photosynthetically grown R. capsulatus cells were harvested at the desired point in the growth phase, as determined by culture turbity. The cells were concentrated 10- or 15-fold by 28 centrifugation and sonicated. Cell debris and membranes were removed by centrifuging at 540 000 x g for 10 minutes in a Beckman TLA100.3 rotor and soluble proteins recovered by collecting the resulting supernatant fluid. Protein concentration was determined using a Lowry assay (Peterson 1983). The samples were run on 10% or 12% S D S - P A G E gels and blotted onto a nitrocellulose membrane (Schleicher and Schuell). The blotting was done in a Mini Trans-Blot apparatus (Bio-Rad) according to manufacturer's specifications in Towbin buffer (Towbin et al. 1979). The primary antibody was a mouse anti-CtrA antibody raised against the C. crescentus CtrA protein (a generous gift from L . Shapiro). Primary antibody binding was detected using a peroxidase-linked anti-rabbit Ig secondary antibody (from donkey; Amersham) as part of the enhanced chemiluminescence (ECL) kit according to the manufacturer's instructions (Amersham). 3. R E S U L T S 29 3.1. Identification of mutants deficient in G T A production and isolation of wild type versions of the mutated genes. Screening of the Y262::Tn5 library by the overlay method initially identified 37 potential mutants, as indicated by the absence of a visible halo of photosynthetically competent transductants in the overlaying agar layer (Figure 2). These potential mutants were then tested individually for G T A production by the standard transduction assay (see Materials and Methods). Seven isolates that no longer produced detectable levels of G T A and three isolates that produced reduced, but still detectable, levels of G T A were obtained. Genomic D N A from the seven mutants that did not produce detectable amounts of G T A was purified, restriction enzyme digested and used for Southern blots. A fragment of D N A containing the Tn5 neo gene (Figure 3) was used as a probe for hybridization to determine the sizes of fragments containing the transposon insertions. Based on the hybridization patterns it was determined that among the seven isolates there were five independent insertion sites. For example, Figure 4 shows that Y G T 2 and Y G T 7 have the same hybridization pattern with BamH I, EcoR I and a double digest with both enzymes, and are therefore presumably the same insertion. In contrast, mutants Y G T 9 and YGT23 have different hybridization patterns for all three digests and therefore represent independent insertions. A segment of Tn5 and flanking sequences were cloned from the chromosome of the five independent mutants by digestion with either BamH I alone, or BamH I and EcoR I together, and ligation with similarly digested pUC13 (see Materials and Methods). The D N A flanking the insertion site of one mutant, Y G T 9 , was used as a probe to hybridize with 30 neo Figure 3. Map of transposon Tn5. The approximate locations of the neo gene, the BamH I site used for sub-cloning, the primer used for sequencing (Tn5L32; arrowhead) and the inverted repeats (IR) are shown. There are no EcoR I sites in Tn5. Y G T 2 YGT7 Y G T 9 YGT23 B E D B E D B E D B E D 23 10 6 Figure 4. Digitized image of an autoradiogram of a Southern blot used for characterization of Tn5 insertion sites in G T A production mutants. The neo gene from transposon Tn5 was used to probe Southern blots of genomic D N A from the mutant strains shown above the lines digested with the restriction endonucleases BamH I (B), EcoR I (E), and BamH I and EcoR I together in a double digest (D). Approximate sizes of D N A standards are shown on the left in kb. 32 colony lifts of a R. capsulatus strain BIO chromosomal cosmid library. Eight cosmids were recovered by this procedure and these cosmids were found to have similarities in their restriction enzyme digestion patterns (data not shown). Two large overlapping cosmids, 9H1 and 9H2, each with approximately 20 kb of insert D N A , were used for complementation and hybridization experiments with the five original Tn5 mutants (summarized in Table 3). The complementation experiments showed that mutants Y G T 2 , Y G T 9 and YGT31 were complemented by cosmid 9H2, mutant YGT24 was complemented by cosmid 9H1 and mutant YGT23 was complemented by both cosmids. The hybridization experiments showed that the insertion site from Y G T 2 hybridized with cosmid 9H2, mutant Y G T 2 4 hybridized with cosmid 9H1, mutants Y G T 9 and YGT31 hybridized with both cosmids and mutant YGT23 did not hybridize with either cosmid. Thus I concluded that four of the five insertion sites mapped to sequences present on the two cosmids. These experiments also showed that although some insertion sites were located on both cosmids (mutants Y G T 9 and YGT31), only one of the two cosmids (9H2) complemented the loss of G T A production. This could indicate that a promoter required for expression of the mutated genes is present only on cosmid 9H2, or the Tn5 insertions have a polar effect on other G T A genes that are present only on cosmid 9H2. For mutant Y G T 2 3 , both cosmids complemented the loss of G T A production, but the insertion site was not present on the cosmids. This led me to conclude that the reason this strain does not produce G T A is not due to the Tn5 insertion (i.e., this mutant contains a spontaneous mutation unrelated to the Tn5 insertion), and so only the four mutants Y G T 2 , Y G T 9 , YGT24 and YGT31 were further analyzed. 33 T A B L E 3. Characterization of G T A " mutants and cosmids 9H1 and 9H2. Mutant Complementation Hybr id izat ion 5 strain 9H1 9H2 9H1 9H2 YGT2 + - + YGT9 - + + + YGT23 + + - -YGT24 + - + YGT31 - + + + a using D N A from next to each Tn5 insertion site as a probe 3.2. DNA sequence analysis of mutants and cloned chromosomal fragments. The insertion sites of the mutants were sequenced, using the cloned fragments described above as templates and a primer complementary to the end of Tn5. The sequences obtained for three of the four mutants displayed no homology to known genes in the GenBank database. A l l of the EcoR I fragments detected on cosmids 9H1 and 9H2 were sub-cloned and at least partially sequenced. B L A S T analyses of resultant sequences showed this region to include the glnA gene (coding for glutamine synthase), which allowed the localization of the insertion sites on the genome physical map since the location of this gene was known (Fonstein et al. 1995). The D N A sequence of the glnA region of the chromosome was obtained from a research group headed by R. Haselkorn that is sequencing the genome of R. capsulatus strain SB 1003. The exact locations of the Tn5 insertions were found and all mapped to a 30 kb region. The sequences were analyzed for the presence of open reading frames (orfs) and a genetic 34 map was constructed (Figure 5). The orfs were analyzed for codon usage (Table 4) and most were similar to the R. capsulatus photosynthesis gene codon usage; the C A I values for 15 highly expressed photosynthesis genes tested ranged from 0.61 to 0.77, and the lacZ gene from E. coli scored 0. Two of the orfs, orfgl7 and orfgl9, showed CAI scores well out of this range, at 0.28 and 0.27 respectively, and may be unexpressed orfs. The predicted amino acid sequences of each orf were used to search the GenBank sequence database using the program B L A S T P (Altschul et al. 1997), and the results of these searches are summarized in Table 5. The three insertion sites that demonstrated no homology to known genes mapped to a 15 kb cluster of 19 orfs with the same transcriptional polarity. However, several orfs within this cluster exhibit significant amino acid sequence similarity to known or putative phage and prophage proteins. This cluster is flanked on each side by orfs with high homology to conserved bacterial cellular proteins (see below). Amino acid sequence alignments showed that the G T A orfg4 is most similar to the protease protein from E. coli phage HK97. The predicted products of orfg3 and orfg5 are homologous to the portal and capsid proteins, respectively, from Streptomyces spp. phage (J>C31. O r f g l l is similar in sequence to a gene from Salmonella typhimurium that is homologous to a tail protein of the E. coli phage A,. Two different regions of the orfg2 sequence display similarities to two different phage proteins. A n N-terminal region is 27% identical over 114 amino acids to the D N A packaging protein GP17 from the E. coli phage T4, and a C-terminal region is 28% identical to the putative large terminase protein (Accession AF065411) from the Methanobacterium thermoautotrophicum phage *FM2 (Pfister et al. 1998). The G T A gene cluster is bounded on both sides by orfs with very high similarity to known cellular "housekeeping" genes. It £8 dip dm >1 flop j 36 Figure 5. Map of a region of the R. capsulatus chromosome containing genes necessary for G T A production. Locations of Tn5 and K I X X cartridge insertion found to abolish G T A production are shown. A l l orfs are predicted to be transcribed left to right except for trmU. Orfs with weak or no similarity to known genes are shown in red or pink; orfs homologous to known phage genes are shown in light blue; orfs highly homologous to known cellular genes are shown in yellow. A l l predicted orfs match R. capsulatus codon usage well, as defined by the highly expressed photosynthesis genes, except for those shown in pink (orfgl7 and orfgl9). More detailed information about sequence similarities is given in Table 5. Table 4. Sizes, relative locations and codon adaptation indices (CAI) for predicted open reading frames orf predicted size 3' distance to next C A I value 0 (amino acids) orf (bp) a trmU 354 198 0.75 II. I 92 97 0.55 ctrA 237 62 0.75 dnlJ 707 -4 0.69 recG 680 0.70 fabF 420 -4 0.74 chp 390 192 0.69 or fg l 107 114 0.53 orfg2 393 184 0.73 orfg3 396 246 0.69 orfg4 184 3 1 0.71 orfg5 398 171 0.72 orfg6 179 48 0.53 orfg7 112 -4 0.73 orfg8 135 4 1 0.70 orfg9 137 8 0.72 orfg lO 108 211 0.72 o r f g l 1 219 1 1 0.64 o r f g l 2 247 47 0.57 or fg l3 243 -4 0.65 or fg l4 150 1 0.69 or fg l 5 346 82 0.71 or fg l 6 100 103 0.56 orfg l7 1 13 87 0.28 or fg l8 651 93 0.74 o r f g l 9 69 60 0.27 cysE 279 0.72 anegative values indicate translational overlap; D C A I is defined in Materials and Methods (Section 2.11) Table 5. Summary of database search results for predicted open reading frame products orf database match (Accession number) E value a trmU probable tRNA methyltransferase (P73755) 8x10" 56 II. 1 orf in flagellar hook gene operon (AF072135) 5x10- 17 ctrA CtrA/response regulator (AF051939) 3x10" 87 dnlJ D N A ligase (Z11910) 1x10" 166 recG RecG/recombination (X59550) 1x10" 123 fabF K A S II/fatty acid synthesis (AJ235273) 1x10" 122 chp conserved hypothetical protein (U32728) 1x10- 43 or fg l none orfg2 T4 D N A packaging protein (X52394) 5x10" 03 orfg3 0C31 portal protein (AJ006589) 2x10" 15 orfg4 HK97 prohead protease protein (U18319) 3x10- 08 orfg5 0C31 capsid protein (AJ006589) 2x10" 12 orfg6 none orfg7 none orfg8 none orfg9 antigen A (Y09161) 2x10" 03 orfg lO none o r f g l l X tail component homologue (AF007380) 9x10" 07 o r f g l 2 none or fg l3 none or fg l4 putative lipoprotein (D86610) 8x10- 05 or fg l5 none or fg l6 none or fg l7 none orfg l8 similar to rhamnosyl transferase (AL008967) 4.5xl0~02 or fg l9 none cysE CysE/cysteine biosynthesis (X59594) 4 x l 0 ' 6 9 anumber of equal scoring matches expected by random chance; results with a value less than 10"! are included 4 1 is flanked on the 3' side by a cysE (cysteine biosynthesis) homologue and on the 5' side by an orf similar to a widespread hypothetical bacterial protein (chp) and afabF (fatty acid biosynthesis) homologue. A search of the genomic sequencing database showed that there are no other phage gene homologues within 90 kb of the G T A structural gene cluster. Three other orfs within the G T A gene cluster (orfg9, orfgl4 and orfgl8; Table 5) had low scoring matches to non-phage genes and the potential significance of these matches is discussed below (see Discussion section). The fourth initial Tn5-generated mutation that abolished detectable G T A production mapped to the cellular gene ctrA, located approximately 15 kb from the G T A gene cluster (Figure 5). This gene is predicted to encode a response regulator protein similar to the CtrA proteins from Brucella abortus (Table 5) and C. crescentus (Accession AF133718). The R. capsulatus ctrA gene is located 5' of dnlJ (DNA ligase) and recG (DNA recombination) homologues, and 3' of an orf whose predicted product is similar to the II. 1 protein from C. crescentus that is located in a flagellar hook gene cluster (Ohta et al. 1984). A 3762 bp EcoR I fragment from the cosmid 9H1 (which contains genomic D N A from strain BIO) was entirely sequenced, as was the region containing the ctrA gene. Comparison of these sequences with those from strain SB 1003 showed one difference, the presence of an extra "G" in the BIO sequence. The absence of this "G" causes a frameshift that has functional significance, since it alters the predicted orf composition of the G T A structural gene cluster. The BIO sequence predicts one large orf (orfg4) with continuous similarity to a known phage gene while the SB 1003 sequence predicts two smaller orfs similar to different parts of the same phage gene. Thus, I trust in the biological precedent of the homologous phage genes and propose that the sequence determined for 42 the SB 1003 gene is incorrect, whereas the sequence I have determined for the BIO gene is correct. 3.3. Identification of a sensor kinase involved in GTA gene expression. The plasmid pSTU12, which carries a translationally in-frame fusion between orfg2 (from strain BIO) and lacZ, was used to identify genes necessary for orfg2 expression (see Materials and Methods section). This procedure produced five colonies that were identified as weakly blue on plates that contained X-Gal . G T A transduction assays showed that two of these isolates, Y T L 3 and Y T L 4 , produced extremely low levels of G T A (less than 10 transductants as compared with thousands produced by the parental strain Y262), and so the insertion sites from these two mutants were cloned. The D N A sequence of the insertion site in Y T L 4 revealed that it was located between divergent orfs; it was located approximately 230 bp upstream of a conserved bacterial protease homologue which is followed closely by formate acetyl transferase and butanol dehydrogenase homologues, and in the other direction it was approximately 400 bp upstream of an elastin homologue that is followed by an acetate kinase homologue. This mutant was not further characterized. The insertion in Y T L 3 was in the R. capsulatus homologue of the C. crescentus cckA gene. The wild type gene was recovered from the R. capsulatus strain B10 genomic cosmid library on cosmid C H I , and 1257 bp of the gene were sequenced. The translation of the D N A sequence indicated a protein that is 44% identical to the C. crescentus C c k A protein over the 419 amino acids of the R. capsulatus gene that were sequenced (Figure 6). CckA is a hybrid kinase, which means that it contains an extra response regulator receiver domain at the C-terminus of the protein in addition to the usual histidine kinase transmitter domain. Examination of the partial R. capsulatus CckA sequence revealed that it possesses the R. capsulatus LQVTJllARVTE C. crescentus RIAFflJYLYRA EGRPdll E G R r quAKM arJtySEdKpaE pJCjLaCHQKMQ AIGQLAGGVAl dF^radQKMQ AIGQLAGGVAJ 50 3 2 1 R. capsulatus C. crescentus HDFNNLLTAI HDFNNLLTAI BGHdEBJ: bLRllEIH: DynJR H I]LI|R H IR H DHGDPfc Y TD iH PMGDEEYEG | L V 3 C L L A L V 32 L L A 1 0 0 3 7 1 R. capsulatus C. crescentus FSRKQltLKPR it DllRDTTlSp LTHLlWRliTG F S R K Q W Q R E VUDICELasp F E \ I L I M R I J L R D T L T H D PI^|LKPI|RADK 1JITDYG RqqPQ\|RADK 1 5 0 4 2 1 J?. capsulatus flQLF^VTjMNL C. crescentus S. V^NARDAMPG G-G EIRIETE NLll QLEJTA\(MNL pfVNARDA vTRA A¥C 3 G W R I R TAI I E E L K R D R A - A V P F G N T 1 9 8 ' F E E A I Q L G F P A A E G D 4 7 1 R. capsulatus C. crescentus TAFI EEG ECG WGIPHEKIGK IF E fcblPHEWGK IFC PFiTTKK HGEGTGLGL 3 T P. YGXVKQ rG 2 4 8 PFETTKP UGEGTGLGL A TvYGIVKQSD 5 2 1 R. capsulatus C. crescentus I F C E IETvT SEIE .GAVA IJAUP ITME LPSIEENSAk . \ A E E A K P R A A R D L S G A 3 2 9 6 5 7 1 R. capsulatus C. crescentus R. R]JLJE jVEDFJAP IIMVEDEJDA IRGYt^iJEAb Nq^EAlMljlJE^DQLQ^ 3 4 6 ^[RGY|EJVJL|EA^ EXEEAI]I°E ENAGTIELLI 6 2 1 R. capsulatus C. crescentus 3 TlDVIMPGMDG DVIMPGfllDG P T ATVAI A L K T RPDT4 P T L L K J A R G Y L G T A : ^^ VWFWS IV^FJIS GYAE D\ GYAE AE RPPTPNS\FL LEGETGVT FL 3 9 6 6 7 1 R. capsulatus C. crescentus PKE F S L S PKEIDIK' ^CJIARR AFjA ICQL QPh 4 1 9 6 9 1 Figure 6. Alignment of C c k A protein sequences. Residues 274-691 of the C. crescentus protein (Accession AF133718) are aligned with the available 419 amino acids of the R. capsulatus sequence (Accession AF181079). The proteins are 44% identical over the sequences shown. 44 conserved histidine, asparagine, and glycine-rich sensor kinase domains (Stock et al. 1989). In C. crescentus the CckA protein is involved in the activation (by phosphorylation) of CtrA (Jacobs et al. 1999); thus I hypothesize that the R. capsulatus CckA protein is involved in CtrA activation for G T A gene expression. When the Y262 cckA gene was disrupted with the K I X X cartridge, in strain Y K K R 2 , there was an approximately 10-fold reduction in G T A production in comparison to the parental strain, Y262. It is interesting that there was a low level of G T A production in this mutant, in contrast to the undetectable levels of G T A production in the ctrA mutants. This may indicate that there are CckA-independent mechanisms in R. capsulatus by which CtrA becomes activated (by phosphorylation) and proceeds to induce transcription of the G T A genes, although my results indicate that CckA is the major pathway. 3.4. GTA purification and protein analysis. The GTA-containing concentrate from P E G precipitation of the cell-free fraction of a 17 litre fermentor-grown Y262 culture was subjected to velocity centrifugation on linear sucrose gradients (see Materials and Methods). Fractions were collected from the gradient and assayed for G T A activity and absorbance at 280 nm. The results for a typical gradient are shown (Figure 7). Peak GTA-containing fractions (on the basis of transduction assays) were pooled [and sometimes concentrated (see Materials and Methods)] and subjected to equilibrium density centrifugation (isopycnic banding) in a RbCl solution. Fractions from equilibrium density gradients were collected and assayed for G T A activity and density, and the results of one gradient are shown (Figure 8). This gradient shows the peak G T A activity to band at an approximate density of 1.29 g/ml, as compared to the previously calculated value of 1.32 g/ml (Solioz 1975). Peak GTA-containing fractions Fraction Figure 7. Sucrose gradient centrifugation of G T A . The results of a typical gradient are shown. Fractions (75 drops) were collected and assayed for G T A activity (closed symbols) and absorbance at 280 nm (open symbols). G T A activity indicates the number of transductants resulting from a transduction assay with 20 pi from each fraction. 46 Figure 8. Equilibrium centrifugation of G T A in RbCl . Each 15 drop fraction of the RbCl gradient was assayed for G T A activiy (open symbols) and the density determined (closed symbols) by measuring the refractive index (see Materials and Methods). G T A activity indicates the number of transductants resulting from a transduction assay with 50 p i from each fraction. 47 were pooled and dialyzed against G-buffer at 4 ° C . Proteins were precipitated with acetone and the resulting precipitate dissolved in loading buffer and run on a S D S - P A G E gel (Figure 9). A larger sample was then used to run another gel which was blotted onto a P V D F membrane for N -terminal amino acid sequence analysis. 3.5. Sequence analysis of proteins from a G T A preparation. The four most prominent protein bands of a G T A preparation, approximately 30, 38, 40 and 43 kDa, were chosen for N-terminal sequence analysis. The 38 kDa protein sequence was determined to be N -S S I N T N T S A M V A L Q T L K G I N S N L A K T Q S - C . The database search results showed that this protein fragment is similar to the flagellin Fl jK protein from C. crescentus (Accession AF089835). The 43 kDa protein band was a mixture of several proteins which made analysis difficult. Three potential sequences were provided as results, each of less than 10 amino acid residues, and searches of the database and the G T A orfs for these sequences were negative. The 40 kDa protein band also contained more than one protein and was further complicated by a failure of the H P L C system during analysis. A 13 residue sequence ( N - Q F F N T N T A A V A I D - C ) was obtained but searches of both the database and the G T A orfs with this sequence were negative. The sequence of the 30 kDa protein was determined to be N - A L N S A V A A E G G Y L V D P Q T S E T I R G V L R - C . A GenBank database search with this sequence gave no matches, but I searched the predicted G T A protein sequences and discovered that this sequence perfectly matches amino acid residues 101-127 of the Orfg5 sequence. On the basis of the database search results, Orfg5 is predicted to be the G T A major capsid protein (Table 5). Thus, it appears that the full-length G T A capsid protein precursor is cleaved on the C-terminal side of a lysine residue to produce the mature capsid protein during maturation of the 48 Figure 9. Digitized image of a Coomassie blue-stained S D S - P A G E of a G T A preparation sample. Molecular weight standards (lane 1) are shown with sizes on the left (kDa). A sample from a G T A preparation is in lane 2. Three of the four protein bands used for N-terminal sequencing are visible (the 40 kDa band sequenced is not visible on this gel; see Results). The identities of the proteins determined by N-terminal sequencing are shown on the right. 49 head structure (Figure 10), similar to phage HK97 (Duda et al. 1995). The fact that this G T A protein migrates at this size indicates that, similar to the phage (|)C31 (Suarez et al. 1984; Smith et al. 1999), there is not cross-linking between G T A capsid protein subunits. For bacteriophage HK97 the capsid protein subunits are cross-linked between lysine and asparagine residues (K-169 and N-356) on neighbouring subunits to form closed rings of either 5 or 6 subunits (Duda 1998). Examination of the G T A capsid protein sequence revealed that these residues are not conserved; the nearest candidates are a lysine at position 152 and an asparagine at position 349, but these apparently do not function in cross-linking. The presence of the flagellin protein in the purified G T A preparation indicates that the G T A was contaminated with other proteins that must have co-purified with the G T A particles through the procedure. Therefore, it is difficult to evaluate if the unsuccessfully sequenced protein bands seen are G T A proteins and which genes they might represent. 3.6. Directed mutations of GTA genes and effects on GTA production. Site-specific insertions of the K I X X cartridge were created and used to confirm the specific requirement of some orfs for G T A production. Transcriptionally congruent insertions, which rarely produce a polar effect on 3' genes (Bollivar et al. 1994), as well as oppositely oriented insertions in orfg2, orfg4 and ctrA, reduced G T A production to undetectable levels in comparison with parental strains. Complementation of the ctrA mutant with the plasmid p C T R A (which contains the ctrA gene transcribed from the puf promoter) restored G T A production (see below). 3.7. Analysis of the ctrA gene. Alignment of the deduced CtrA protein sequences from R. capsulatus and C. crescentus showed that the proteins are 71% identical (Figure 11 A). The C. crescentus CtrA protein is A. HK97 MSELTALIQKMEESQQ^^ 40 SDIjyiKVQEr^TlCSC^^ 80 AAEt^IKSWDGKQG^^ 120 processing site B. GTA MKTKKARAG^ 40 VKIKLQQQEEWTiy^^ 80 A A Y m T G D D r a 120 • processing site Figure 10. Capsid protein processing by GTA and phage HK97. The first 120 amino acid residues of the HK97 (A) and GTA (B) capsid proteins and their proteolytic processing sites are shown. A . 5 1 R. capsulatus C. crescentus I E L M L r r H A N m v T f l MFffLlJaEDqS ATAQTllELML KSEGHNVYMT DLGEEGUDLp' T DLGEEGplDLR. KLYDYDLILL KIYDYDLILL 50 50 R. capsulatus C. crescentus DJ^PDldtJGJL FJVLRdLRJLJAJR VlJTp3[|lThJG DLNLPDMflG|l qVLRHL^I^K INTPIMILHG MDDTHS KIK 3 3 S E i m K \ , K T F 31GADDYMT F \C GADDYMT 100 100 R. capsulatus C. crescentus KPFERI KPFF. KI , \ JA RIHAfenJRRSK GHyQsfafSEtTG JE&VNLDAKT VLwdGJKltVHL LT|A R I H A | V \ R R S K GH P QS i I E TG [RVNLDAKT V E V S G O T V H L 150 150 R. capsulatus C. crescentus TGKEYQMLEL LSLRKGTTLT KEMFLNHLYG GMDEPELKII' DVTICKLRKK TGKEYQMLEL LSLRKGTTLT KEMFLNHLYG GMDEPELKIi ITvFICKLRKK 200 200 R. capsulatus C. crescentus LiAEVTd LA&.SAH 3GENY : G K H H IETVWGRGYV LRDEJDOGDLD RRl^ vk/dA IETVWGRGYV LRDPSTE —CVU2A 237 231 B . C. crescentus consensus CtrA binding site: TTAA.. .N7. . .TTAAC R. capsulatus site 1: R. capsulatus site 2: GTA A ...N7...TTAAC TGAT.. .N7. . .TTAAT Figure 11. Comparisons of CtrA proteins and DNA binding sites. A. Alignment of the CtrA amino acid sequences from R. capsulatus and C. crescentus. Identical amino acids (71%) are boxed and the putative sequence-specific DNA recognition domains of the proteins are shaded. B. The consensus CtrA binding site of C. crescentus compared to potential CtrA binding sites located 5' of the R. capsulatus ctrA gene. Sites 1 and 2 are approximately 70 bp and 3 0 bp, respectively, 5' of the ctrA start codon. 52 phosphorylated on an aspartate residue at position 51, and this residue is conserved in the R. capsulatus protein. The sequence-specific D N A binding domain of the C. crescentus CtrA protein was identified by alignment with the OmpR protein of E. coli (Martinez-Hackert and Stock 1997), and the R. capsulatus and C. crescentus CtrA proteins were found to be identical in this region (Figure 11 A). This indicates that these two CtrA proteins likely recognize the same D N A sequence. D N A binding sites for the C. crescentus protein are known (Quon et al. 1996) and two sequences similar to a consensus site are present upstream of the R. capsulatus ctrA gene (Figure 11B). The two sites are approximately 70 bp and 30 bp 5' of the start codon for the ctrA gene. If these are CtrA binding sites, it would suggest autoregulation of expression by CtrA in R. capsulatus. The ctrA gene is essential for survival in C. crescentus (Quon et al. 1996) but my ability to mutate the ctrA gene suggested it is not essential in R. capsulatus. To evaluate if there is only one copy of the ctrA gene in R. capsulatus I analyzed the ctrA::Tn5 strain Y GT24 and its parental strain Y262 in a Southern blot. D N A from these two strains was digested with the restriction enzyme EcoR I, which does not cut within Tn5 (see Figure 3), and the blot was probed with the ctrA probe. The results indicate that there is only one copy of the ctrA gene in R. capsulatus, and therefore it is not an essential gene (Figure 12). R N A blot analysis showed ctrA gene expression to be growth phase-dependent (Figure 13). Disruption of the ctrA gene with the K I X X cartridge abolished G T A production, and no detectable transcripts corresponding to the G T A orfg2 or orfg4 genes were visible (Figures 14 and 15). The orfg2 and orfg4 transcripts were restored when ctrA was supplied in trans on a plasmid. However the orfg2 and orfg4 transcripts showed the same growth phase-dependent pattern in the complemented strain as in strain 1 2 Figure 12. Digitized image of a result from a Southern blot used to determine the ctrA copy number. A fragment of D N A containing the ctrA gene was used to probe chromosomal D N A digested with the restriction endonuclease EcoR I from strains Y262 and YGT24 (ctrA::Tn5) (lanes 1 and 2, respectively). YCKF Y262 YCKF (pCTRA) Figure 13. Digitized image of an autoradiogram used for R N A blot analysis of ctrA transcripts. R N A was isolated from the strains indicated above the lines at three time points over the growth cycle: a, b and c represent mid-log, late-log and early stationary phases, respectively. Equal amounts (10 pg) of R N A were loaded in each lane, separated by electrophoresis, and transferred to a nylon membrane before probing with a 3 2 P radio-labeled D N A fragment containing the ctrA gene (see Materials and Methods). Approximate locations of molecular weight R N A standards are shown on the left (in kb). Figure 14. Digitized image of an autoradiogram used for R N A blot analysis of orfg2 transcripts. R N A was isolated from the strains indicated above the lines at three time points over the growth cycle: a, b and c represent mid-log, late-log and early stationary phases, respectively. Equal amounts (10 pg) of R N A were loaded in each lane, separated by electrophoresis, and transferred to a nylon membrane before probing with a 3 2 P radio-labeled D N A fragment containing orfg2 (see Materials and Methods). Approximate locations of molecular weight R N A standards are shown on the left (in kb). YCKF Y262 YCKF (pCTRA) a b c a b c a b c Figure 15. Digitized image of an autoradiogram used for R N A blot analysis of orfg4 transcripts. R N A was isolated from the strains indicated above the lines at three time points over the growth cycle: a, b and c represent mid-log, late-log and early stationary phases, respectively. Equal amounts (10 pg) of R N A were loaded in each lane, separated by electrophoresis, and transferred to a nylon membrane before probing with a 3 2 P radio-labeled D N A fragment containing orfg4 (see Materials and Methods). Approximate locations of molecular weight R N A standards are shown on the left (in kb). 57 Y262, despite constitutive ctrA expression from the puf promoter. The ctrA gene from C. crescentus, on plasmid p C C C , did not complement the loss of G T A production caused by mutation of the native ctrA gene. There was an initial question of whether the plasmid p C C C would allow ctrA expression since the Apa I site used had never been tested before. However, construction of pRRR, which had the R. capsulatus ctrA gene inserted at the same site, complemented the mutant, which suggests that this site is suitable for insertion of fragments for expression by the puf promoter. This result, as well as the visualization of a C. crescentus CtrA protein band on a western blot (Figure 16), lead me to conclude that the C . crescentus protein does not substitute for the R. capsulatus protein in G T A production. I attempted to follow the level of the R. capsulatus CtrA protein over the growth cycle using western blots probed with antibodies against the C . crescentus CtrA protein. The antibodies seemed to bind to the R. capsulatus CtrA protein, but also bound to other proteins (Figure 16). Protein samples were collected over the growth cycle and analyzed in this way, but there was another protein band, present in the mid-log and late-log samples but absent from the stationary phase sample, at a slightly higher molecular weight than the presumed CtrA band [Figure 17; compare lanes with Y C K F (ctrA~) samples]. This band shows a pattern of expression opposite to what would be predicted for CtrA, as the levels of this other band are highest in the mid-log phase, lower in the late-log phase, and it seems to be absent from the stationary phase. This band may represent a logarithmic growth response regulator protein that has enough similarity to CtrA to bind the antibodies used. The western blot with the samples collected over the growth cycle provided some information about CtrA stability and regulation. In the 1 2 3 4 CtrA • (27 kDa) Figure 16. Evaluation of anti-CtrA antibodies and expression of the CtrA protein from C. crescentus in R. capsulatus by western blotting. Cell samples were collected from C. crescentus CB15A (lane 1), R. capsulatus Y262 (lane 2), R. capsulatus Y C K F (lane 3), and R. capsulatus YCKF(pCCC) (lane 4). Lanes 1-3 contain soluble proteins (sonicated cell samples) from approximately 0.1 ml of culture and lane 4 contains whole cells from 0.1 ml of culture; R. capsulatus samples are from stationary phase cultures. YCKF Y262 YCKF (pCTRA) a b c a b c a b c ^j^^^g:gj^^n ^^ ^^ S «Z3 C t r A — " — (27 kDa) Figure 17. Evaluating growth phase variation in CtrA levels by western blotting. Cell samples were taken from the strains indicated above the lines at three points over the growth cycle: a, mid-log phase; b, late-log phase; c, early stationary phase. Each lane contains an equal amount of protein (10 pg) from a sonicated cell sample. 60 mid-log and late-log phase samples of Y C K F ( p C T R A ) there is a low molecular weight band that presumably corresponds to a CtrA degradation product. This strain presumably overproduces CtrA since there is clearly a lot of ctrA R N A made (Figure 13). There is a much smaller amount of this protein band in the stationary phase sample, and this could indicate that the protein is stabilized in the stationary phase (or destabilized in the other phases). The R N A blots show that this gene is maximally expressed in the stationary phase of strain Y262 cultures (Figure 13), and so protein stability may be another aspect of regulation of CtrA activity. In C . crescentus, the CtrA protein is subject to proteolysis as part of a complex regulatory regime, which was suggested to eliminate this protein from certain times in the cell cycle (Domian et al. 1997). If the C. crescentus interpretations are correct, my data suggest that there may be a similar system of control in R. capsulatus. R. capsulatus strains containing plasmid-borne G T A gene fusions to lacZ (orfg2'y.'lacZ) showed growth phase-dependent p-galactosidase activity (Figure 18), similar to the pattern seen for R N A levels (Figure 14). The high level of activity seen in the stationary phase samples was dependent on ctrA [compare Y262(pYP), YCKF(pYP)] and the loss of p-galactosidase activity in strain Y C K F was partially restored by supplying ctrA on a second, compatible plasmid [compare YCKF(pYP) and Y C K F ( p Y P , pCTRA)]. Therefore, similar to the R N A blot results for orfg2, the P-galactosidase activities of the orfg2'::7acZ fusions show a growth phase- and CtrA-dependent pattern. The C. crescentus protein regulates expression of flagellar genes (Quon et al. 1996; Domian et al. 1997), and so I attempted to determine if there was any effect on flagellum production in R. capsulatus. I attempted both flagellum staining and electron microscopy to determine if strains A. B pYP pYNP 'chp Y262c Y262(pYNP)c Y262(pYP)a Y262(pYP)b Y262(pYP)c Y262(pYP)d YCKF(pYP)d YCKF(pCTRA, pYP)a YCKF(pCTRA, pYP)b YCKF(pCTRA, pYP)c YCKF(pCTRA, pYP)d g l H g2' g2' HacZ 'lacZ 100 activity (Miller units) 200 62 Figure 18. Growth phase- and CtrA-dependence of P-galactosidase activities of orfg2'::7acZ plasmid-borne gene fusions. A . Representations of the orfg2'::7acZ fusions used to investigate G T A gene expression. B. P-galactosidase specific activities. The cell samples indicated to the left of the graph designate: Y262, plasmid-free Y262; Y262(pYNP), Y262 containing plasmid pYNP; Y262(pYP), Y262 containing plasmid pYP; YCKF(pYP), Y C K F containing plasmid pYP; Y C K F ( p C T R A , pYP), Y C K F containing plasmids p C T R A and pYP. Cells were harvested over the growth phase at a, mid-log phase; b, late-log phase; c, early stationary phase; d, late stationary phase. Miller units are defined in Materials and Methods (Section 2.16). 63 Y262 (ctrA+) and Y C K F (ctrA') produced flagella but the results were inconclusive. I never saw a Y C K F cell with a flagellum by these methods, although I could rarely find Y262 cells with a flagellum (data not shown). However, motility tests using soft agar stabs showed that ctrA mutants were non-motile and that this phenotype was complemented by supplying the ctrA gene on a plasmid (Figure 19). This suggests that the R. capsulatus CtrA protein is required for motility, presumably for expression of the flagellar genes as in C. crescentus. Disruption of the ctrA gene in R. capsulatus strains BIO and Y262 had no significant effect on photosynthetic growth rate in YPS medium (Figure 20). Similarly, disruption of the ctrA gene in strain BIO had no significant effect on aerobic growth rate in YPS and R C V media, photosynthetic growth rate in R C V medium, or on the levels of the photosynthetic apparatus (data not shown). It is of note that the growth rates for the complemented ctrA mutants, Y C K F ( p C T R A ) and B C K F ( p C T R A ) , are slightly slower than either the wild type or the mutant strains. This effect is visible in Figure 20 (more obvious for the Y262 strains) and I also observed it on many other occasions. This is presumably due to overproduction of CtrA in these strains (Figure 17). This slow growth may be due to adverse effects of excess CtrA during non-stationary phases of growth, or perhaps degradation of this "unwanted" protein leads to the slower growth rate. 3.8. Transcription of the GTA structural gene cluster. R N A blot analysis showed that orfg2 and orfg4 transcription are growth phase-dependent. There was very little or no detectable R N A corresponding to these genes in mid- or late-log phase, but there was a strong signal for these genes in stationary phase (Figures 14 and 15). The transcripts for both genes show size heterogeneity, but over different size ranges. The expression of orfg2 and orfg4 was dependent on ctrA: when ctrA was Figure 19. CtrA-dependent motility. Strains BIO (ctrA+), B C K F (ctrA-) and B C K F ( p C T R A ) were stabbed into YPS agar (0.4%) and the tubes incubated under photosynthetic conditions for four days. 1000 time (hours) Figure 20. Effect of ctrA mutation on photosynthetic growth rate. A . Growth curves for strains BIO, open circles; B C K F , closed circles; and B C K F ( p C T R A ) , open squares. B. Growth curves for strains Y262, open circles; Y C K F , closed circles; and Y C K F ( p C T R A ) , open squares. Klett units are defined in Materials and Methods (Section 2.1). 67 disrupted by insertion of the K I X X cartridge there were no detectable transcripts for either orfg2 or orfg4, but expression of both orfs was restored when ctrA was supplied in trans on a plasmid. M y results show that the ctrA gene is required for maximal expression of the G T A gene cluster, and orfs in this cluster display growth phase-dependent expression (see above). The differences in p-galactosidase activities obtained with the orfg2'v.'lacZ fusions in Y262(pYP) and Y262(pYNP) show that a promoter required to transcribe orfg2 is located between the middle of chp and approximately 50 bp 5' of the start codon of orfg2 (a region of approximately 900 bp), since Y262(pYNP) cells had the same low activity as plasmid-free cells (Figure 18). Examination of the D N A sequence in the region delimited by pYP and pYNP revealed a sequence ( 5 ' - A A A A C C C C G G C T T C G T C C G G G G T T T T T T C T T T T - 3 ' ) that could encode a Rho-independent transcription terminator (Chen et al. 1988; d'Auberton Carafa et al. 1990) beginning 3 bp 3' of the stop codon for chp, and 27 bp 3' of this motif is a sequence (5 ' -TTGACT-N17-TAGAGAT-3 ' ) that is similar to known sigma factor binding sites. The 5' half of the latter sequence is similar to o^O sites from R. capsulatus (Nickens and Bauer 1998) and E. coli (Hawley and McClure 1983), and the 3' portion is similar to the analogous region of a rjD (a^8) site from B. subtilis (Gilman et al. 1981). There are no sequences in this 900 bp region that resemble potential CtrA binding sites. Strain Y262 is indirectly a derivative of strain B10, although there are likely many mutations that separate the two strains since Y262 was created by performing nitrosoguanidine mutagenesis (Yen et al. 1979). By probing a Northern blot with the orfg4 probe I found that one reason why strain Y262 produces more G T A than strain B10 is because of elevated G T A R N A levels (Figure 2IB). The orfg4 message was not visible in the strain A . B . H H U U Cu Cu o S w s s * - U U <N u U PQ PQ PQ >H >H >H RA) /—\ < RA) H U U a 3 P - i E ld o s 2 1—1 U u • u C Q P Q >-4.4 1.4 « 0.24 Figure 21. R N A blot analysis for comparison of G T A gene expression levels between strains Y262 and BIO. R N A was isolated from the strains indicated above the lanes from early stationary phase cultures. Equal amounts (10 pg) of R N A were loaded in each lane, separated by electrophoresis, and transferred to a nylon membrane before probing with a 3 2 P radio-labeled D N A fragments (see Materials and Methods). A . The blot probed with the ctrA gene probe. B. The blot probed with orfg4. Approximate locations of molecular weight R N A standards are shown in the middle (in kb). BIO sample whereas there were no apparent differences in the R N A levels between Y262 and BIO when the same blot was probed with ctrA (Figure 21A). This difference in G T A gene expression levels was also observed with orfg2'::7flcZ fusions. The fusions made with the BIO promoter region produced very low (unmeasurable) P-galactosidase activities whereas fusions with the Y262 promoter region were clearly measurable (Figure 18). This suggests there are sequence differences between the Y262 and BIO promoter regions in these fusions. This difference in expression is also manifested in E. coli; Y262 fusion-containing cells were clearly blue on X -gal plates but, under the same conditions, a blue colour was barely visible in cells containing the BIO fusion. The lacZ fusion experiments showed that the promoter required for expression of orfg2, and presumably the other G T A genes, is located in a 900 bp region 5' of orfg2. I sequenced a 470 bp region of this 900 bp region from strain Y262 and comparison of this sequence to the SB 1003 sequence revealed no differences. This was surprising since this was the region I predicted to contain the promoter; the region from strain Y262 that I sequenced starts upstream of the proposed chp stop codon, proceeds through the potential transcriptional terminator and sigma factor recognition sequences discussed above, and ends more than 100 bp into the proposed orfgl gene. I conclude that the mutation(s) that leads to higher expression levels in strain Y262 must be in a region that I did not sequence. Interestingly, when the fusion to the Y262 promoter region was assayed in strain B10, the resulting activity was lower than that measured in strain Y262 although it showed the same growth phase dependence (Figure 22). This suggests that, in addition to the promoter mutation(s), there is at least one more mutation in a separate locus in Y262 responsible for part of the observed increase in expression. B10(pYP)a B10(pYP)b Y262(pYP)c activity (Miller units) Figure 22. Strain differences in P-galactosidase activities of orfg2'::7acZ plasmid-borne gene fusions. The p-galactosidase activities in the wild type strain BIO and the G T A overproducer strain Y262, from the same Y262 orfg2'::7acZ fusion, are compared. Cell samples were harvested over the growth phase at a, mid-log phase; b, late-log phase; c, early stationary phase. The gene fusion on plasmid pYP is described in Materials and Methods and represented in Figure 18. Miller units are defined in Materials and Methods (Section 2.16). 7 1 3.9. Analysis of GTA-producing and non-producing R. capsulatus strains for the presence of GTA genes. D N A fragments that contain the ctrA gene and G T A structural genes (orfg2, orfg3, orfg4 and part of orfg5) were used separately as probes of Southern blots to test for the presence of these genes in several R. capsulatus strains. Both fragments hybridized with all strains tested (Figure 23). The ctrA fragment was present in all strains on the same size of EcoR I fragment, but the G T A structural genes were present on different sized fragments. Strains Y262, SB 1003 and BIO all produce G T A and contain the structural genes on the same sized fragment (3.8 kb) as strains YW1 and YW2, which do not produce detectable levels of G T A . Strains B6 and SP36, which do not produce G T A , contain the genes on different sized fragments (5 and 6.5 kb respectively). Thus, I conclude that there must be a reason other than the complete lack of the G T A genes to explain why these G T A non-producing strains do not produce G T A (i.e., they are missing some other G T A genes not tested for, or they contain mutations that lead to loss of G T A gene expression, etc.). Another explanation could be that these strains do produce G T A particles but that these particles are not able to bind to recipient cells and therefore are undetectable. There is greater polymorphism among the structural G T A gene £ c o R I sites than the sites around the ctrA gene. 3.10. Analysis of R. capsulatus strains for GTA binding capability. This experiment involved mixing samples of a G T A -containing culture filtrate with various strains of R. capsulatus and, after a timed incubation, removing the cells and measuring the amount of G T A remaining by a transduction bioassay (see Materials and Methods). If the cells in the initial incubation step were able to bind the G T A , there should have been a loss of detectable G T A in the second filtrate compared to an A. B . CO CO o o CO ^ g O ^ r o > > ! o - H V O ^ f> ? H r ) ^ ? H aj c s ffl CO ^ ffl M >H PQ t>0 >H >-< p Q >H : 9.4 6.6 4.4 2 . 3 Figure 23. Digitized image of an autoradiogram used to analyze R. capsulatus strains for the presence of GTA genes. Chromosomal D N A from the strains shown above each lane was digested with the restriction endonuclease EcoR I, electrophoresed, blotted onto a nylon membrane and hybridized with probes corresponding to the ctrA gene (A) and orfg2-orfg5' (B). Approximate sizes are shown in the middle, in kb. initial incubation that contained no cells. Neither strain SP36 nor YW1 are supposed to be transduced by G T A (Wall et al. 1974) in contrast to strain DW5 (Wong et al. 1996). Strain SP36 produced no detectable loss of G T A from the filtrate (Figure 24), whereas strain YW1 seemed to bind G T A comparable to strain DW5 (a capable transductant). This indicates that strain YW1 bound G T A whereas strain SP36 did not. Therefore, it seems the reason strain SP36 is not transduced by G T A is because it lacks a G T A receptor. In contrast to previously published results (Wall et al. 1974), I found that strain YW1 is transduced at a low frequency (at least-10 fold lower than strain DW5). The results of my binding experiments suggest that strains YW1 and DW5 are capable of binding G T A particles equally well, and so there is another reason for the difference in transduction frequency observed. One possibility is that there are differences in restriction/modification systems between strain Y W 1 and the G T A donor strains used. It might seem that one way to test this would be to attempt an intra-strain transfer of a plasmid-borne marker via G T A , but strain YW1 does not produce detectable levels of G T A , and so this experiment cannot be done. 200 no cells SP36 YW1 D W 5 Figure 24. G T A binding capability of R. capsulatus strains. G T A binding capability is reflected by low values of residual G T A activity. The recipient strains used are shown below the bars. 4. D I S C U S S I O N This thesis was undertaken to attempt to distinguish between two possible evolutionary origins of G T A : it was suggested that G T A might represent an evolutionary precursor of a phage or, alternatively, a defective phage (Yen et al. 1979). While I believe that my work has demonstrated an evolutionarily long relationship between G T A and R. capsulatus, I am still unable to distinguish between the two possible origins of G T A . Despite this, my work has significantly advanced our understanding of G T A . I found that many genes are involved in the production of G T A particles. Some of these genes map to a cluster wherein the only significant sequence similarities to known genes are to known phage genes, and phage genes from diverse bacterial genera are represented. I also found that the expression of the G T A genes, and thus the production of G T A particles, is regulated by cellular proteins that make up at least part of a (growth-phase-dependent) two-component signal transduction pathway. This two-component system is also involved in regulating motility in R. capsulatus. These and other results are discussed in more detail below. 4.1. The G T A structural gene cluster. I suggest that the orfgl to orfgl9 region (Figure 5) is a cluster of G T A structural genes. This 15 kb G T A gene cluster is too small to encode a genuine phage, which by analogy with the phages HK97 (Accession AF069529) and <pC31 (Smith et al. 1999) that contain G T A gene homologues, would require a genome of approximately 40 kb. However the head-tail region of phage 0C31 is approximately 17 kb (Smith et al. 1999), which is similar in size to the cluster shown in Figure 5. There could be additional G T A structural genes 76 in a separate locus that I did not find or, alternatively, the gene cluster shown in Figure 5 is sufficient to encode G T A particles. Since G T A randomly packages 4.5 kb linear genomic D N A fragments (Solioz and Marrs 1977; Yen et al. 1979; Scolnik and Haselkorn 1984), phage DNA-specific replication and excision functions are not required to produce D N A destined for packaging. Similarly, phage DNA-specific integration functions should not be required for gene transmission by G T A since GTA-dependent allelic replacement seems to be mediated by a cellular (recA-dependent) recombination system (Genthner and Wall 1984). On the basis of the database search results the G T A structural gene cluster does not appear to contain a lysin/holin system, as seems to exist in the related phages HK97 (Accession AF069529) and <j)C31 (Smith et al. 1999). This is consistent with the report that GTA-induced cell lysis is undetectable (Marrs 1974). Control of G T A gene transcription is positively regulated by the cellular proteins CckA and CtrA and there is no evidence for phage regulatory proteins. Therefore, the gene cluster shown in Figure 5 could contain all the information needed for assembly of the G T A head and tail structures and D N A packaging. Gene organization of the head-tail region is a highly conserved feature amongst double-stranded D N A phages (Casjens et al. 1992) and, as best as I can determine, the G T A genes seem to conform to this organization. Within the G T A gene cluster are three orfs that had low scoring matches of uncertain significance (Table 5). Despite the low scores, some of these matches may provide a clue to the function of these predicted G T A proteins. Orfg9 is 20% identical over 131 amino acids to a secreted antigen from pathogenic strains of Listeria monocytogenes, but the function of this protein is unknown (Schaferkordt and Chakraborty 1997) making it difficult to propose a potential function for Orfg9. The predicted product of orfgl4 is 29% identical over 136 amino acids to a putative lipoprotein, and so could interact with the inner and/or outer membranes of recipient cells. Orfgl8 is 26% identical over a 204 amino acid stretch to the hypothetical protein Rv2739c from Mycobacterium tuberculosis, which in turn is similar in sequence to a rhamnosyl transferase. Since rhamnose is a component of the R. capsulatus capsule (Omar et al. 1983), I speculate that the orfgl8 product may interact with the capsule of recipient cells. 4.2. Gene expression of the GTA structural gene cluster. The results of complementation experiments with the original Tn5 mutants and the cosmids 9H1 and 9H2 suggest that the G T A structural gene cluster constitutes an operon under the control of a single promoter. Mutant Y G T 9 is not complemented by cosmid 9H1 despite the fact that the promoter required for expression of orfg2, as demonstrated by the gene fusion experiments, and the complete coding sequences for orfgl to orfg l l (and part of orfgl2) are present on 9H1 (Figure 25). Only cosmid 9H2, which contains the entire G T A structural gene cluster including the proposed promoter region, complemented the G T A " phenotype of Y G T 9 . This suggests that the Tn5 insertion in orfg2 has a polar effect on downstream G T A genes and only when the entire operon is supplied in trans is G T A production restored. In the related phage 0C31, transcription of the head and tail genes is driven by a single promoter (Suarez et al. 1992; Howe and Smith 1996). I attempted to discover the sequence differences between strains Y262 and SB 1003 that lead to increased G T A gene expression in Y262, but the sequenced region showed no differences. It is possible that the changes are located further 3' than I sequenced. I sequenced only a central 470 bp region of the 900 bp region in which I found a promoter. It would be interesting to sequence the rest of this region to discover what ctrA 9H1 — 9H2 G T A structural gene cluster ^ ^ m ^ m ^ ^ m ^ ^ ^ ^ ^ ^ K m A A A Y G T 9 YGT31 Y G T 2 Figure 25. Map of the R. capsulatus chromosomal D N A present on cosmids 9H1 and 9H2 and the G T A structural gene cluster. The locations of the Tn5 insertions in mutants Y G T 9 , YGT31 and Y G T 2 are shown (the exact locations of these insertions in the G T A orfs is shown in Figure #). 79 changes might exist and be responsible for the cz's-active properties of this region. It would also be interesting to find out what other mutations in Y262 contribute to the increased expression levels. I do not think that this elevation is due to differences in the ctrA gene because the Y262 orfg2 and orfg4 R N A levels are restored to Y262 levels in the ctrA' strain Y C K F when the strain BIO ctrA gene is supplied on a plasmid (Figures 14 and 15). The mutated gene could be found by making a Y262 genomic library and putting it into strain BIO cells that contained the orfg2'::7acZ fusion. This library of cells could then be screened for colonies that showed increased P-galactosidase activity (as judged by the intensity of blue colour in colonies on plates containing X-gal). 4.3. G T A release by R. capsulatus. It is unclear how G T A particles are released from cells. Some possibilities are GTA-induced lysis, fortuitous release upon natural cell lysis, or some sort of non-lytic excretion process, although I am unaware of a biological precedent for a tailed phage to be released by a non-lytic event. As mentioned above, GTA-induced cell lysis is undetectable and there is no evidence for a G T A lysin/holin system within the G T A structural gene cluster. The production of G T A by R. capsulatus cultures is quite low, as shown by the magnitude of effort required to purify an amount of G T A sufficient for analysis of protein or D N A composition (Yen et al. 1979), and this correlates with the low level of p-galactosidase produced in the gene fusion experiments. This indicates that G T A gene expression is quite low as averaged over a population of cells in laboratory culture. One possible explanation for this is that all cells express the G T A genes at an equal but low level. Alternatively, there could be differential expression within a population such that most cells express the G T A genes at a low level while a sub-population of cells express these genes at a much higher level. This would 80 be analogous to the system of competence development in Bacillus subtilis where a subset of the population (1-10%) develops competence at the transition from the exponential to the stationary phase of growth (Grossman 1995). An investigation of this potential cell to cell variation in G T A gene expression levels would be a good experiment to follow up on my work. This could be done by observing cells containing G T A gene::lacZ fusions with fluorescence microscopy in the presence of the fluorogenic (3-galactosidase substrate fluorescein di-P-D-galactopyranoside, a published technique (Russo-Marie et al. 1993). Such an experiment could distinguish between the "all low" or "some high" gene expression hypotheses. Without a GTA-specific lytic pathway, even if a cell highly expresses the G T A genes the particles would not be released until the cell naturally lysed. Similarly, it may take time to accumulate enough of the proteins to lyse a cell when there is a low expression level of lytic genes. 4.4. Relationship of G T A to other phages and prophages. Sequence differences between the G T A genes and phage homologues evidently account for the differences in terms of quantity and quality of D N A that is packaged. However, the details of how 4.5 kb segments of genomic D N A are encapsidated by G T A are unclear. There are likely to be several mutations responsible for the aberration in D N A packaging by G T A . There appears to be no specificity in what D N A is packaged and this could be due to mutations in the terminase and portal proteins. The model for double-stranded D N A phages is that these proteins intimately interact and are involved in both the initiation and completion of packaging, as well as the translocation of D N A into the head (Black 1989). Assuming that G T A evolved from a bacteriophage, the particle at one time would have held more than 4.5 kb if my proposal that 15 kb is required to produce the G T A particles is correct. A "headful" mechanism more easily explains why all 8 1 G T A particles contain the same amount of D N A , as opposed to a mechanism relying on recognition of specific packaging sequences (pac sites) (Black 1989) that are evenly spaced, once every 4.5 kb, on the R. capsulatus chromosome. These points suggest that the major capsid protein must also have significant changes, and that these changes lead to the difference in the quantity of D N A contained within G T A particles in comparison with GTA's phage ancestor, if G T A really is a defective phage. The purification of G T A particles and subsequent N-terminal amino acid sequencing of proteins from the purification showed that the G T A capsid is made up of monomer units of the major capsid protein. This is in contrast to phage HK97, which shows head subunit cross-linking (Duda et al. 1995) and protein chainmail (Duda 1998). The consequences of cross-linking and chainmail organization are that rings of 5 or 6 capsid protein subunits are crosslinked together to form covalently closed circles, and that these circles are also looped through neighbouring rings within the capsid to form a chainmail structure. The same phenomenon is also believed to occur in the P. aeruginosa phage D3 (Gilakjan and Kropinksi 1999). Phage L5 from Mycobacterium spp. displays head subunit cross-linking (Hatfull and Sarkis 1993), but the existence of a chainmail organization has not been reported. These observations fit with the fact that the G T A capsid protein is more similar to the capsid protein from the Streptomyces spp. phage <|)C31 than to the HK97 protein, and phage (j)C31 lacks cross-linking (Smith et al. 1999). There are at least two putative prophages on the R. capsulatus genome that have sequence similarities to the same family of phages that G T A resembles. These were referred to as (j)RcPl and (|)RcMl (Smith et al. 1999) due to their presence on the sequencing contigs named PI and M l , respectively, and are located approximately 300 to 500 kb and 650 to 850 82 kb, respectively, from the G T A genes. The sequence relationships between these putative prophage genes, known phage homologues and the G T A genes is very interesting. Both of these putative prophages contain terminase and portal genes (and other phage gene homologues). Phage 0RcMl contains a capsid protein coding homologue, but it is fused with a protease gene. Overall, the gene organization of these prophages (most specifically the head genes) does not appear to conform to the conserved order seen for the G T A genes; there appear to be missing genes and possibly rearrangements. The genes mentioned above all show some sequence identity with previously mentioned GTA-related phages (D3 from P. aeruginosa, HK97 from E. coli and (|)C31 from Streptomyces spp.). I find it interesting that only two of the (|)RcPl or (|)RcMl genes exhibit the best sequence alignment with an orf from either the other prophage or G T A , despite the fact that these three all co-exist in the same genome. The (|)RcMl protease (the 185 amino acid protease fragment of the protease-capsid fusion) shows 36% overall identity (44% identity over 137 amino acids) to the G T A putative protease, whereas the <])RcMl and <|)RcPl terminases are 23% identical (31% identical over 504 amino acids). The surprising character of these sequence relationships is exemplified by comparisons of the <j)RcMl and G T A portal, terminase and protease protein sequence similarities (Figure 26). A B L A S T search with the 396 amino acid G T A portal protein (orfg3) sequence yields the 457 amino acid (|)C31 portal protein, with 27% identity over 354 amino acids. A B L A S T search with the 358 amino acid <j)RcMl portal sequence also yields the <))C31 portal as the highest match, with 31% identity over 346 amino acids; the G T A portal protein scores sixth, with 26% identity over 342 amino acids. Thus, the G T A and 0RcMl portal proteins each appear to be more closely related to a portal protein from the Streptomyces spp. phage A . Portal protein relationships 26% (342 aa) GTA « • (SRcMl 4>C31 B. Protease protein relationships 44% (137 aa) GTA « • (HRcMl HK97 84 C . Terminase protein relationships (j)RcMl < ^ • GTA D3: 27% (513 aa) <|>C31: 26% (503 aa) D3, <|)C31 T4: 27% (114 aa) ¥ M 2 : 28% (128 aa) T4, *¥M2 D. G T A gene organization and summary of sequence relationships — terminase (orfg2) portal (orfg3) protease (orfg4) similar to T4 and ¥ M 2 not similar to <|>RcMl and $C3l similar to <))RcMl similar to <))RcMl and <|>C31 and HK97 Figure 26. Sequence relationships between G T A and (|)RcMl proteins. The % identities shared by proteins are shown beside the double-headed arrows, with the length of the homologous region in parentheses. A double-headed arrow with an X through it signifies that there is no recognizable sequence similarity between the proteins. The relationships of the G T A and (|)Rc]Vll portal (A), protease (B) and terminase (C) protein sequences to each other and to those from other phages are shown. The genetic organization for the G T A proteins shown with a summary of the sequence relationships (D). 86 ())C31 than they are to each other (Figure 26A). As mentioned above, the G T A and <|)RcMl protease protein sequences are 36% identical; both are also similar to the protease protein from the E. coli phage HK97 (Figure 26B). Despite the close sequence similarity between the portal and protease proteins, the terminase proteins show a much different pattern. The G T A terminase is most similar to the proteins from phages T4 (E. coli) and ^VMl (M. thermoautotrophicum), and shows no recognizable similarity to the lambdoid group of phages. The <|)RcMl terminase is similar to proteins from the lambdoid group such as D3 and (j)C31, but not to any of G T A -related terminases (Figure 26C). This suggests that at some point in the past either the (|)RcMl or G T A ancestor underwent a horizontal exchange that replaced the terminase gene. This patchwork quality of sequence relationships (Figure 26D) seems to be further evidence in support of the model (Hendrix et al. 1999), that genetic exchanges between double stranded D N A phages (and prophages) occurred during evolutionary history. This patchwork or modular structure also exists within a single G T A gene: orfg2, the presumed G T A terminase. As mentioned previously, the predicted protein product of this gene shows sequence similarity to terminase proteins from two different phages; the E. coli phage T4, and the archaeophage from M. thermoautotrophicum, V FM2. A N-terminal region is similar to the T4 terminase whereas a C-terminal region is similar to the WM2 terminase (Figure 27). These similarities are also interesting because all of the other G T A genes that are homologous to phage genes are homologues of genes from the lambdoid group of phages, or phages related to the lambdoid group (Hendrix et al. 1999). Similarities between proteins from the lambdoid family and T4 have been seen before (Hendrix et al. similar to T4 similar to *FM2 B. GTA 31 AHRVALVGETFDQVRDVMI FGESGILACSPPDRRPEWEATKRRLVWA AH+ ++ E D+ + + F + GI+ EW K + T4 190 AHKGSMSAEVLDRTKQAIELLPDFLQPGIV EWN--KGSIELD GTA 78 NGATAQAYSAQEPEALRGPQFDAAWVDELAKWRRAEETWDMLQFALRLGKH NG++ AY A P+A+RG F ++DE A ++W +Q + G+ T4 230 NGSSIGAY-ASSPDAVRGNSFAMIYIDECAFIPNFHDSWLAIQPVISSGRR GTA 129 PQQVITTTPRNV + +ITTTP + T4 280 SKIIITTTPNGL c. GTA 256 FVLEDASVRGRPTDWARAAIAAMERWGAEKLVA EVNQGGEMVESVLRQ +VL+ R P + E G E ++A E G++V LR psiM2 325 YVLDVRRFRESPGKVKSKVLRTAEEDGREVIIAKEEEPGSSGKrVTDYLRS GTA 304 IDPLVPFKALPASRGKSARAEPVAALYEQGRVKHCRIX?RLGALEDQMCRMT + F+A R + K RA PV++ E GR+K R A D++ psiM2 376 LLQGYTFRADRVTGDKVTRALPVSSYAESGRIKVLRASWTRAFLDELEAFP GTA 354 VRGYAGKGSPDRVDALVWAMTELMIE + G D+VDA A L +E psiM2 422 MEGV HDDQVDAFSGAFNILSME Figure 27. Modular structure of the G T A putative terminase protein. The two regions that are similar to known phage terminase proteins are highlighted (A), and the B L A S T P alignments for these regions to the phage T4 (B) and phage V FM2 (C) terminases are shown. 88 1999), but the archaeophage WM2 exhibited similarities only with proteins from phages that infect Gram positive hosts (Pfister et al. 1998). 4.5. Growth phase-dependence and two-component system control of GTA production. M y discovery that disruption of the R. capsulatus cckA gene impaired G T A production, as did disruption of ctrA, strongly indicates that the proteins encoded by these genes are part of a sensor kinase/response regulator system that controls transcription of G T A genes. The results of the R N A analyses and P-galactosidase assays show that this system acts in response to a stationary phase signal. It seems that there may be at least one missing link in the signal transduction cascade that leads from activation of CckA to the increased expression of the G T A genes. One missing piece seems to be between activation of CtrA and the increase in transcription of the G T A genes, since there do not appear to be any CtrA binding sites within the region I found to contain the promoter for the G T A structural genes. Because I found sequences in the putative G T A promoter region with similarities to known sigma factor binding sites, it is plausible that activation of CtrA positively affects the expression of an alternative (stationary phase) sigma factor that directs increased G T A structural gene transcription. It is also possible that my prediction of a R. capsulatus CtrA binding site (DNA sequence) is incorrect, although I think that the logic I used was sound. The identity of the signal that activates the CckA/CtrA pathway and how this system became involved in the expression of G T A genes requires additional research. When the ctrA gene is mutated (strain Y C K F ) G T A production is undetectable. However, the orfg2'::7acZ fusion experiments showed that there is still G T A gene expression in strain Y C K F (Figure 18), because the Y C K F level of stationary phase expression was similar to the amount in the Y262 mid- and late-log samples. This may indicate a multi-component 89 G T A promoter/regulatory region, such that there is a low basal level of CtrA-independent expression and a CtrA-dependent stationary phase increase in expression. Many gram-positive bacteria, such as B. subtilis, differentiate into dormant spores in response to starvation conditions (Brock and Madigan 1991). Although gram-negative bacteria such as E. coli do not differentiate into spores in response to starvation, they undergo significant physiological changes upon reaching the stationary phase of the growth cycle [reviewed in (Kolter et al. 1993; Huisman et al. 1996)]. These include changes in the inner membrane, the cell envelope, motility, protein synthesis, and metabolism. These physiological changes are brought about by specific changes in gene expression and an important part of this is the production of a stationary phase-specific sigma factor (o^) which stimulates the expression of many other genes [reviewed in (Hengge-Aronis 1996)]. This sigma factor is not essential for survival of E. coli in the lab but is essential for long term survival under starvation conditions. It would be interesting to investigate the effect of mutating the ctrA gene on long term stationary phase survival in R. capsulatus. If the R. capsulatus CtrA protein activates the expression of additional stationary phase genes (either directly or indirectly), long term survival could be affected in a ctrA mutant. This could be addressed by monitoring the number of viable cells in a ctrA mutant culture over time and comparing these results to the values for the parental strain. The C. crescentus CtrA and CckA proteins are required for viability and are key proteins in the C. crescentus cell cycle. Specifically, they are involved in cell cycle transitions (Domian et al. 1997), flagellum development (Quon et al. 1996), D N A replication (Quon et al. 1996; Quon et al. 1998; Jacobs et al. 1999), D N A methylation (Quon et al. 1996) and cell 90 division (Kelly et al. 1998; Jacobs et al. 1999). M y data show that neither CtrA nor CckA is required for viability of R. capsulatus because disruptions of the genes encoding these proteins were not lethal. Although there might be an analogy between the C. crescentus cell cycle and the R. capsulatus growth phase, these two concepts are different. It is interesting that the C. crescentus CtrA protein did not complement the loss of the R. capsulatus CtrA protein despite the high level of sequence conservation between these proteins (71% identity). The differences that exist between the two CtrA proteins must be great enough to prevent interaction between the C. crescentus CtrA protein and critical elements in R. capsulatus. These elements could be a protein(s) responsible for activation of CtrA by phosphorylation, or perhaps the C. crescentus CtrA protein is properly activated but not able to interact with the requisite R. capsulatus target D N A sequences or transcriptional machinery. The high degree of sequence conservation in the proposed D N A binding regions (Figure 11 A) makes it unlikely that the absence of a target D N A sequence is the reason for the absence of complementation. The R. capsulatus CtrA is even more similar to the protein from the pathogen B. abortus (75%) than to the C. crescentus homologue (71%), but although it is supposedly essential in B. abortus (L. Shapiro 1998, personal communication), there is currently no published information about this protein. The question of how this CtrA system came to be present in these distant genera (although they are all oc-Proteobacteria) is a perplexing question. It will be interesting to see if other bacteria have a CtrA protein and what functions it performs in these species. The role of CtrA in motility is conserved between C. crescentus and R. capsulatus. Analogous to C. crescentus, the R. capsulatus CtrA seems to affect flagellar gene expression because R. capsulatus ctrA mutants were 9 1 impaired in motility (Figure 19). In C. crescentus, CtrA is a class I flagellar gene that is required for expression of class II flagellar genes, through direct interactions at promoter sequences (Quon et al. 1996; Domian et al. 1997). Class II genes include the inner ring structure, flagellar protein-export machinery of the flagellum and also a sigma factor required for expression of the class III genes, which include the basal body and hook [reviewed in (Roberts et al. 1996)]. Therefore, it would be useful to attempt to identify the R. capsulatus class II homologues and investigate the possibility of CtrA-dependent expression. It also would be interesting to determine if CtrA control is mediated directly or through the hypothetical sigma factor discussed above. Since ctrA expression is growth phase-dependent (Figure 13), I speculate that motility is also growth phase-dependent, analogous to G T A gene expression. There is a biological precedent for growth phase-dependent motility: flagellar gene expression is growth phase-dependent in B. subtilis (Haldenwang 1995). There is also another connection between my findings in R. capsulatus and B. subtilis motility. The potential promoter sequence I identified upstream of the G T A genes is similar to the B. subtilis rjD recognition site (see Results section 3.8), and o ^ is the sigma factor which allows expression of flagellar genes in B. subtilis (Helmann 1991). E. coli cells are motile throughout the growth phase, but there is a transient increase in motility, expression of flagellar genes, and production of flagellin that lasts from the late exponential phase to the early stationary phase of growth (Amsler et al. 1993). In C. crescentus, control of CtrA activity is very complex. There is temporal control at the level of transcription initiation (Quon et al. 1996) by feedback control (Domian et al. 1999). The identification of two potential CtrA binding sites 5' of the ctrA gene suggests that there may be 92 feedback control of ctrA transcription in R. capsulatus. A good follow up experiment would be to test if either of these sites are genuine by using purified CtrA protein for D N A footprinting. Another level of control is protein turnover by proteolysis. The C . crescentus CtrA protein is controlled by proteolysis dependent upon two C -terminal alanine residues, although replacement of these alanine residues with valine residues did not affect the proteolytic control (Domian et al. 1997). The R. capsulatus protein also has a hydrophobic C-terminus ( V V G A ) , and so a similar control mechanism may exist in R. capsulatus. Also, analogous to other response regulators, the activity of the C . crescentus CtrA protein is controlled by phosphorylation (Quon et al. 1996) at a conserved aspartate residue (D-51). This residue is conserved in the R. capsulatus protein (Figure 11 A) , which suggests that there is the same mechanism of control of activity by phosphorylation at this site of the R. capsulatus CtrA protein. The control of the phosphorylation state of CtrA in C. crescentus is complex, because there appear to be as many as three pathways whereby CtrA can be activated through phosphorylation. The histidine kinase protein CckA is one of the ways by which CtrA is activated, but it is unknown if the activation of CtrA by CckA is direct or mediated through a phosphorelay system. After D N A replication has begun, C c k A moves from a delocalized state in the cell membrane and localizes to the pole of the predivisional cell, and it is believed this polar localization allows more effective spatial control of CtrA activation (Jacobs et al. 1999). Another pathway to CtrA phosphorylation starts with the histidine kinase DivJ (and possibly also the histidine kinase PleC) and proceeds through the single-domain type response regulator protein DivK, and at least one other unknown protein (Hecht et al. 1995; Wu et al. 1998). A third pathway 93 involves direct phosphorylation of CtrA by DivL, a histidine kinase homologue (Wu et al. 1999). The CckA, DivL and DivK proteins are all essential for cell viability in C. crescentus. The redundancy of mechanisms that activate CtrA in C. crescentus may help explain one of my results in R. capsulatus. When I made a disruption of the cckA gene, the resultant mutant strain, Y K K R 2 , produced G T A at 10-fold reduced levels compared to the parental strain! When the ctrA gene was disrupted there was no detectable G T A activity. Therefore, it seems likely that there is more than one pathway that leads to activation of CtrA in R. capsulatus. I would have liked to have obtained the entire R. capsulatus cckA sequence, but it seemed a redundant effort to continue with sequencing this gene since the entire genome will soon be finished (http://rhodoL.uchicago.edu/capsulapedia/capsulapedia/capsulapedia.sht ml); it is unfortunate that this gene appears to be located in one of the regions not yet completed (at least I was unable to find it on the web site). Presumably, when the rest of the cckA sequence is available, the protein will be predicted to contain a putative membrane-spanning domain analogous to other sensor kinases, specifically the C. crescentus CckA protein (Jacobs et al. 1999). It would be interesting to investigate if the R. capsulatus CckA protein displays skewed localization in the same way as the C. crescentus protein but this seems unlikely given the life-cycle differences between the two species (Brock and Madigan 1991). On the basis of the G T A gene expression data, the R. capsulatus CckA protein is presumed to become activated in the stationary phase, but it is difficult to imagine why it would be advantageous for polar localization of this protein to occur in R. capsulatus because R. capsulatus does not have a dimorphic life cycle (Brock and Madigan 1991). 94 There are many missing details concerning the exact pathway and mechanism by which G T A gene expression is activated by CtrA. These include the possibility of another protein acting downstream of CtrA, to directly activate G T A gene transcription. Also, the mechanism by which CtrA presumably becomes activated by phosphorylation needs further study. M y experiments suggest that the putative sensor kinase CckA is involved in this process, and that there may be other proteins that activate CtrA (analogous to the C. crescentus system). Additional screening of the Y262::Tn5 library with the plasmid-borne orfg2'::7acZ fusion may identify other genes involved in this pathway. The initial screening was difficult since the BIO orfg2'::7acZ fusion I used is weakly expressed, and so the use of a Y262 orfg2'::7acZ fusion should facilitate the identification of additional mutants. 4.6. Genes near the ctrA gene in R. capsulatus. The amino acid sequence similarity between the R. capsulatus orf located 5' of the ctrA gene and the II. 1 protein from C. crescentus is curious (the protein sequences are 44% identical). No function has been reported for the C . crescentus II. 1 protein although it is located in a cluster of class III (hook) flagellar genes (Mullin and Newton 1989; Mullin and Newton 1993). The C . crescentus CtrA protein regulates the expression of different (class II) flagellar genes (Quon et al. 1996; Domian et al. 1997). Given this relationship in C. crescentus, it is puzzling that homologues of these two genes are located next to each other in R. capsulatus. The location of the ctrA gene with respect to the genes for D N A ligase (dnlJ) and RecG (Figure 5) is interesting because the proteins predicted to be encoded by these two genes are likely to be required for recombination of D N A from a G T A particle into the recipient cell's chromosome. Given the growth phase-dependence of ctrA expression, it is not surprising that dnlJ 95 appears to be transcriptionally independent from ctrA. This is because, by analogy with E. coli, the dnlJ gene should be essential due to its role in D N A replication (Gottesman et al. 1973). The Northern blot with the ctrA probe (Figure 13) showed a R N A molecule of higher molecular weight than the proposed ctrA message, which is present at constant levels in all lanes. This could be the message for dnlJ since there was a small amount of the dnlJ sequences on the D N A fragment used to probe this blot (see Materials and Methods). 4.7. Presence of GTA genes in R. capsulatus strains in nature. The fact that most present-day strains of R. capsulatus, isolated from geographically distant areas of the world, are capable of producing G T A (Marrs 1974; Wall et al. 1974) suggests that G T A is an evolutionarily old property of R. capsulatus. This idea was strengthened by the finding that the four G T A non-producing strains I tested contain sequences which strongly hybridize with known G T A genes (Figure 23). For two of the strains there was a polymorphism at one or more of the E c o R I sites tested. These or other changes could be a reason for the lack of G T A production. In the producing strains, one of the polymorphic EcoR I sites occurs in the major capsid protein. A l l four of these non-producing strains presumably have one or more mutations that abolish either the production of G T A transducing particles or the detectability of the particles' activity. 4.8. Concluding remarks. The results presented in this thesis indicate that there has been an evolutionarily long relationship between G T A and R. capsulatus. However, the question of whether G T A is a defective bacteriophage or a phage precursor remains. Fewer assumptions are required to propose a phage ancestor of G T A , as opposed to vice versa, and so the simplest interpretation is that G T A is a defective prophage. Another possibility that I believe is supported by my results is that G T A is a 96 combination of a defective bacteriophage and a cell-evolved genetic exchange vector. Perhaps future experiments will help clarify this question. Regardless of its genesis, G T A is used by R. capsulatus as a genetic exchange vector, such that the production of G T A is controlled by a set of cellular proteins that activate the process of gene transfer in the stationary phase of growth of laboratory cultures. I suggest that a response to growth-limiting environmental conditions causes a signal to be transduced through the CtrA pathway to activate G T A gene expression. This results in the production of G T A particles that transduce genomic D N A fragments to other R. capsulatus cells, enhancing the genetic diversity of a population of cells confronted by a hostile environment. 97 5. REFERENCES Adams, C. W., M . E . Forrest, S. N . Cohen and J. T. Beatty (1989). Structural and functional analysis of transcriptional control of the Rhodobacter capsulatus puf operon. / . Bacteriol. 171: 473-482. Altschul, S. F. , T. L . Madden, A . A . Schaffer, J. Zhang, Z. Zhang, W. Miller and D. J. Lipman (1997). Gapped B L A S T and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25: 3389-3402. Amsler, C. D. , M . Cho and P. Matsumura (1993). Multiple factors underlying the maximum motility of Escherichia coli as cultures enter post-exponential growth. J. Bacteriol. 175: 6238-6244. Barany, F. (1985). Single-stranded hexameric linkers: a system for in-phase insertion mutagenesis and protein engineering. Gene 37: 111-123. Beatty, J. T. and H . Gest (1981). Generation of succinyl-coenzyme A in photosynthetic bacteria. Arch. Microbiol. 129: 335-340. Bergh, 0., K. Y . B0rsheim, G. Bratbak and M . Heldal (1989). High abundance of viruses found in aquatic environments. Nature 340: 467-468. Bertani, G . (1999). Transduction-like gene transfer in the Methanogen Methanococcus voltae. J. Bacteriol. 181: 2992-3002. Black, L . W. (1989). D N A Packaging in dsDNA Bacteriophages. Ann. Rev. Microbiol. 43: 267-292. Bollivar, D. W., J. Y . Suzuki, J. T. Beatty, J. M . Dobrowski and C. E . Bauer (1994). Directed mutational analysis of bacteriochlorophyll a biosynthesis in Rhodobacter capsulatus. J. Mol. Biol. 237: 622-640. Brent, R. and M . Ptashne (1980). The lexA gene product represses its own promoter. Proc. Natl. Acad. Sci. USA 77: 1932-1936. Brock, T. D. and M . T. Madigan (1991). Biology of microorganisms. (6th ed.) Prentice Hall, Englewood Cliffs, New Jersey. Casjens, S., G. Hatfull and R. Hendrix (1992). Evolution of dsDNA tailed-bacteriophage genomes. Sent. Virol. 3: 383-397. Cheetham, B. F. and M . E . Katz (1995). A role for bacteriophages in the evolution and transfer of bacterial virulence determinants. Mol. Microbiol. 18: 201-208. Cheetham, B. F., D. B. Tattersall, G . A . Bloomfield, J. I. Rood and M . E . Katz (1995). Identification of a gene encoding a bacteriophage-related integrase in a vap region of the Dichelobacter nodosus genome. Gene 162: 53-58. Chen, C . -Y . A . , J. T. Beatty, S. N . Cohen and J. G. Belasco (1988). An intercistronic stem-loop structure functions as an m R N A decay terminator necessary but insufficient for puf m R N A stability. Cell 52: 609-619. d'Auberton Carafa, Y . , E . Brody and C. Thermes (1990). Prediction of Rho-independent Escherichia coli transcription terminators: a statistical analysis of their R N A stem-loop structures. J. Mol. Biol. 216: 835-858. Davies, J. (1994). Inactivation of antibiotics and the dissemination of resistance genes. Science 264: 375-382. Ditta, G. , T. Schmidhauser, E . Yakobsen, P. Lu, X . - W . Liang, D. R. Finlay, D. Guiney and D. R. Helinski (1985). Plasmids related to the broad host range vector, pRK290, useful for gene cloning and for monitoring gene expression. Plasmid 13: 149-153. Domian, I. J. , K. C. Quon and L . Shapiro (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. Domian, I. J. , A . Reisenauer and L . Shapiro (1999). Feedback control of a master bacterial cell-cycle regulator. Proc. Natl. Acad. Sci. USA 96: 6648-6653. Duda, R. L . (1998). Protein chainmail: catenated protein in viral capsids. Cell 94: 55-60. Duda, R. L . , J. Hempel, H . Michel, J. Shabanowitz, D. Hunt and R. W. Hendrix (1995). Structural transitions during bacteriophage HK97 head assembly. J. Mol. Biol. 247: 618-635. Eiserling, F. , A . Pushkin, M . Gingery and G. Bertani (1999). Bacteriophage-like particles associated with the gene transfer agent of Methanococcus voltae PS. J. Gen. Virol. 80: 3305-3308. Ely, B. (1991). Genetics of Caulobacter crescentus. Meth. Enzymol. 204: 372-384. Fonstein, M . , E . G. Koshy, T. Nikolskaya, P. Mourachov and R. Haselkorn (1995). Refinement of the high-resolution physical and genetic map of Rhodobacter capsulatus and genome surveys using blots of the cosmid encyclopedia. EMBO J. 14: 1827-1841. Friedman, A . M . , S. R. Long, S. E . Brown, W. J. Buikema and F. M . Ausubel (1982). Construction of a broad host range vector and its use in the genetic analysis of Rhizobium mutants. Gene 18: 289-296. Genthner, F. J. and J. D. Wall (1984). Isolation of a recombination-deficient mutant of Rhodopseudomonas capsulata. J. Bacteriol. 160: 971-975. Gilakjan, Z. A . and A . M . Kropinksi (1999). Cloning and analysis of the capsid morphogenesis genes of Pseudomonas aeruginosa bacteriophage D3: another example of protein chain mail? / . Bacteriol. 181: 7221-7227. 100 Gilman, M . Z . , J. L . Wiggs and M . J. Chamberlin (1981). Nucleotide sequences of two Bacillus subtilis promoters used by Bacillus subtilis sigma-28 R N A polymerase. Nucleic Acids Res. 9: 5991-6000. Gottesman, M . M . , M . L . Hicks and M . Gellert (1973). Genetics and function of D N A ligase in Escherichia coli. J. Mol. Biol. 77: 531-547. Grossman, A . D. (1995). Genetic networks controlling the initiation of sporulation and the development of competencein Bacillus subtilis. Annu. Rev. Genetics 29: 477-508. Hada, H . , M . Einav, I. Fishov and A . Zaritsky (1997). Bacteriophage T4 development depends on the physiology of its host Escherichia coli. Microbiol. 143: 179-185. Haldenwang, W. G. (1995). The sigma factors of Bacillus subtilis. Micrbiol. Rev. 59 : 1-30. Hatfull, G . F. and G. J. Sarkis (1993). D N A sequence, structure and gene expression of mycobacteriophage L5: a phage system for mycobacterial genetics. Mol. Microbiol. 7: 395-405. Hawley, D. K. and W. R. McClure (1983). Compilation and analysis of Escherichia coli promoter D N A sequences. Nucleic Acids Res. 11: 2237-2255. Hayes, W. (1968). The genetics of bacteria and their viruses. (2nd ed.) John Wiley & Sons Inc., New York. Hecht, G. B., T. Lane, N. Ohta, J. M . Sommer and A . Newton (1995). An essential single domain response regulator required for normal cell division and differentiation in Caulobacter crescentus. EMBO J. 14: 3915-3924. Helmann, J. D. (1991). Alternative sigma factors and the regulation of flagellar gene exression. Mol. Microbiol. 5: 2875-2882. 1 Hendrix, R. W., M . C. M . Smith, R. N. Burns, M . E . Ford and G. F. Hatfull (1999). Evolutionary relationships among diverse bacteriophages and prophages: A l l the world's a phage. Proc. Natl. Acad. Sci. USA 96: 2192-2197. Hengge-Aronis, R. (1996). Regulation of gene expression during entry into stationary phase, pp. 1497-1512. Escherichia coli and Salmonella: Cellular and molecular biology. (2nd ed.) F. C. Neidhardt, R. C. Ill, J. L . Ingrahamet al. A S M Press, Washington, D . C . Hennes, K. P. and C. A . Suttle (1995). Direct counts of viruses in natural waters and laboratory cultures by epifluorescence microscopy. Limnol. Oceanogr. 40: 1050-1055. Howe, C. W. and M . C. Smith (1996). Gene expression in the cos region of the Streptomyces temperate actinophage 0C31. Microbiol. 142: 1357-1367. Huisman, G. W., D. A . Siegele, M . M . Zambrano and R. Kolter (1996). Morphological and physiological changes during stationary phase, pp. 1672-1682. Escherichia coli and Salmonella: Cellular and molecular biology. (2nd ed.) F. C. Neidhardt, R. C. Ill, J. L . Ingrahamet al. A S M Press, Washington, D . C . Humphrey, S. B., T. B. Stanton, N. S. Jensen and R. L . Zuerner (1997). Purification and characterization of V S H - 1 , a generalized transducing bacteriophage of Serpulina hyodysenteriae. J. Bacteriol. 179: 323-329. Jacobs, C , I. J. Domian, J. R. Maddock and L . Shapiro (1999). Cell cycle-dependent polar localization of an essential bacterial histidine kinase that controls D N A replication and cell division. Cell 97: 111-120. Karaolis, D. K. R., J. A . Johnson, C. C. Bailey, E . C. Boedeker, J. B. Kaper and P. R. Reeves (1998). A Vibrio cholerae pathogenicity island associated with 102 epidemic and pandemic strains. Proc. Natl. Acad. Sci. USA 95: 3134-3139. Karaolis, D. K. R., S. Somara, D. R. Maneval, J. A . Johnson and J. B. Kaper (1999). A bacteriophage encoding a pathogenicity island, a type-IV pilus and a phage receptor in cholera bacteria. Nature 399: 375-379. Kelly, A . J. , M . J. Sackett, N. Din, E . Quardokus and Y . V . Brun (1998). Cell cycle-dependent transcriptional and proteolytic regulation of FtsZ in Caulobacter. Genes & Dev. 12: 880-893. Kokjohn, T. A . (1989). Transduction: mechanism and potential for gene transfer in the environment, pp. 73-97. Gene transfer in the environment. S. B. Levy and R. V . Miller. McGraw-Hill , New York. Kokjohn, T. A . , G . S. Sayler and R. V . Miller (1991). Attachment and replication of Pseudomonas aeruginosa bacteriophages under conditions simulating aquatic environments. J. Gen. Microbiol. 137: 661-666. Kolter, R., D. A . Siegele and A . Tormo (1993). The stationary phase of the bacterial life cycle. Annu. Rev. Microbiol. 47: 855-874. Krieg, N. R. and P. Gerhardt (1981). Solid culture, pp. 143-150. Manual of methods for general bacteriology. P. Gerhardt and R. N . Costilow. American Society for Microbiology, Washington, D C . Laemmli, U . K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680-685. Lawrence, J. G. and H . Ochman (1998). Molecular archaeology of the Escherichia coli genome. Proc. Natl. Acad. Sci. USA 95: 9413-9417. LeBlanc, H . , A . S. Lang and J. T. Beatty (1999). Transcript cleavage, attenuation, and an internal promoter in the Rhodobacter capsulatus puc operon. J. Bacteriol. 181: 4955-4960. Levy, S. B. and R. V . Miller, Eds. (1989). Gene transfer in the environment. McGraw-Hill , New York. 103 Madigan, M . T. (1995). Microbiology of nitrogen fixation by anoxygenic photosynthetic bacteria, pp. 915-928. Anoxygenic photosynthetic bacteria. R. E . Blankenship, M . T. Madigan and C. E . Bauer. Kluwer Academic Publishers, Boston. Marrs, B. (1981). Mobilization of the genes of the photosynthesis from Rhodopseudomonas capsulata by a promiscous plasmid. J. Bacteriol. 146: 1003-1012. Marrs, B. L . (1974). Genetic recombination in R. capsulata. Proc. Natl. Acad. Sci. USA 71: 971-973. Martinez-Hackert, E . and A . M . Stock (1997). Structural relationships in the OmpR family of winged-helix transcription factors. J. Mol. Biol. 269: 301-312. Mazodier, P. and J. Davies (1991). Gene transfer between distantly related bacteria. Annu. Rev. Genet. 25: 147-171. Miller, J. H . (1992). A short course in bacterial genetics: a laboratory manual and handbook for Escherichia coli and related bacteria. Cold Spring Harbor Laboratory Press, Plainview, N Y . Mullin, D. A . and A . Newton (1989). Ntr-like promoters and upstream regulatory sequence ftr are required for transcription of a developmentally regulated Caulobacter crescentus flagellar gene. / . Bacteriol. 111. 3218-3227. Mullin, D. A . and A . Newton (1993). A a^^ promoter and downstream sequence elements ftr2 and ftr3 are required for regulated expression of divergent transcription units flaN and flbG in Caulobacter crescentus. J. Bacteriol. 175: 2067-2076. Nickens, D. G. and C. E . Bauer (1998). Analysis of the puc operon promoter from Rhodobacter capsulatus. J. Bacteriol. 180: 4270-4277. 104 Norrander, J . , T. Kempe and J. Messing (1983). Construction of improved M13 vectors using oligodeoxynucleotide-directed mutagenesis. Gene 26: 101-106. Ohta, N . , E . Swanson, B. Ely and A . Newton (1984). Physical mapping and complementation analysis of Tn5 mutations in Caulobacter crescentus: organization of transcriptional units in the hook cluster. J. bacteriol. 158: 897-904. Omar, A . S., J. Weekesser and H . Mayer (1983). Different polysaccharides in the external layers (capsule and slime) of the cell envelope of Rhodopseudomonas capsulata S p l l . Arch. Microbiol. 136: 291-296. Peterson, G . L . (1983). Determination of total protein. Meth. Enzymol. 91: 95-111. Pfister, P., A . Wasserfallen, R. Stettler and T. Leisinger (1998). Molecular analysis of Methanobacterium phage V PM2. Mol. Microbiol. 30: 233-244. Quon, K. C , G. T. Marczynski and L . Shapiro (1996). Cell cycle control by an essential bacterial two-component signal transduction protein. Cell 84: 83-93. Quon, K. C , B. Yang, I. J. Domian, L . Shapiro and G. T. Marczynski (1998). Negative control of bacterial D N A replication by a cell cycle protein that binds at the chromosome origin. Proc. Natl. Acad. Sci. USA 95: 120-125. Rapp, B. J. and J. D. Wall (1987). Genetic transfer in Desulfovibrio desulfuricans. Proc. Natl. Acad. Sci. USA 84: 9128-9130. Roberts, R. C , C. D. Mohr and L . Shapiro (1996). Developmental programs in bacteria. Curr. Top. Dev. Biol. 34: 207-257. Russo-Marie, F. , M . Roederer, B. Sager, L . A . Herzenberg and D. Kaiser (1993). (3-Galactosidase activity in single differentiating bacterial cells. Proc. Natl. Acad. Sci. USA90: 8194-8198. 105 Sambrook, J., E . F. Fritsch and T. Maniatis (1989). Molecular cloning: a laboratory manual. (2nd ed.) Cold Spring Harbor Laboratory Press, Plainview, N .Y . Schaferkordt, S. and T. Chakraborty (1997). Identification, cloning, and characterization of the Ima operon, whose gene products are unique to Listeria monocytogenes. J. Bacteriol. 179: 2707-2716. Schmidt, L . S., H . - C . Yen and H . Gest (1974). Bioenergetic aspects of bacteriophage r eplication in the photosynthetic bacterium Rhodopseudomonas capsulata. Arch. Biochem. Biophys. 165: 229-239. Scolnik, P. A . and R. Haselkorn (1984). Activation of extra copies of genes coding for nitrogenase in Rhodopseudomonas capsulata. Nature 307: 289-292. Scolnik, P. A . and B. L . Marrs (1987). Genetic research with photosynthetic bacteria. Annual Review of Microbiology 41: 703-726. Sharp, P. M . and W . - H . L i (1987). The codon adaptation index - a measure of directional synonymous codon usage bias, and its potential applications. Nucleic Acids Res. 15: 1281-1295. Simon, R., U . Priefer and A . Ptihler (1983). A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in Gram-negative bacteria. Bio/Technology 1: 37-45. Smith, M . C. M . , R. N . Burns, S. E . Wilson and M . A . Gregory (1999). The complete genome sequnec of the Streptomyces temperate phage <J)C31: evolutionary relationships to other viruses. Nucleic Acids Res. 27: 2145-2155. Solioz, M . (1975). The gene transfer agent of Rhodopseudomonas capsulata. Ph.D. thesis, Saint Louis University, Saint Louis. 106 Solioz, M . and B. Marrs (1977). The gene transfer agent of Rhodopseudomonas capsulata: Purification and characterization of its nucleic acid. Arch. Biochem. Biophys. 181: 300-307. Solioz, M . , H . - C . Yen and B. Marrs (1975). Release and uptake of gene transfer agent by Rhodopseudomonas capsulata. J. Bacteriol. 123: 651-657. Stock, J. B. , A . J. Ninfa and A . M . Stock (1989). Protein phosphorylation and regulation of adaptive responses in bacteria. Microbiol. Rev. 53: 450-490. Suarez, J. E . , J. L . Caso, A . Rodriguez and C. Hardisson (1984). Structural characteristics of the Streptomyces bacteriophage 0C31. FEMS Microbiol. Lett. 22: 113-117. Suarez, J. E . , T. M . Clayton, A . Rodriguez, M . J. Bibb and K. F. Chater (1992). Global transcription pattern of 0C31 after induction of a Streptomyces coelicolor lysogen at different growth stages. J. Gen. Microbiol. 138: 2145-2157. Tabita, F. R. (1995). The biochemistry and metabloic regulation of carbon metabolism and CC»2 fixation in purple bacteria, pp. 885-914. Anoxygenic photosynthetic bacteria. R. E . Blankenship, M . T. Madigan and C. E . Bauer. Kluwer Academic Publishers, Boston. Taylor, D. P., S. N. Cohen, W. G. Clark and B. L . Marrs (1983). Alignment of the genetic and restriction maps of the photosynthetic region of the Rhodopseudomonas capsulata chromosome by a conjugation-mediated marker rescue technique. J. Bacteriol. 154: 580-590. Towbin, FL, T. Staehelin and J. Gordon (1979). Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and applications. Proc. Natl. Acad. Sci. USA 76: 4350-4354. 1 Tufaro, F. , M . D. Snider and S. L . McKnight (1987). Identification and characterization of a mouse cell mutant defective in the intracellular transport of glycoproteins. J. Cell Biol. 105: 647-657. Vieira, J. and J. Messing (1982). The pUC plasmids, and M13mp7-derived system for insertion mutagenesis and sequencing with synthetic universal primers. Gene 19: 259-268. Waldor, M . K. and J. J. Mekalanos (1996). Lysogenic conversion by filamentous phage encoding a cholera toxin. Science 272: 1910-1914. Wall, J. D. , P. F. Weaver and H . Gest (1974). Gene transfer agents, bacteriophages, and bacteriocins of Rhodopseudomonas capsulata. Arch. Microbiol. 105: 217-224. Webb, V . , E . Leduc and G. B. Spiegelman (1982). Burst size of bacteriophage SP82 as a function of growth rate of its host Bacillus subtillus. Can. J. Microbiol. 28: 1277-1280. Wong, D. K . - H . , W. J. Collins, A . Harmer, T. G. Lilburn and J. T. Beatty (1996). Directed mutagenesis of the Rhodobacter capsulatus puhA gene and pleiotropic effects on photosynthetic reaction center and light-harvesting I complexes. / . Bacteriol. 178: 2334-2342. Wu, J . , N . Ohta and A . Newton (1998). An essential, multicomponent signal transduction pathway required for cell cycle regulation in Caulobacter. Proc. Natl. Acad. Sci. USA95: 1443-1448. Wu, J. , N . Ohta, J . -L. Zhao and A . Newton (1999). A novel bacterial tyrosine kinase essential for cell division and differentiation. Proc. Natl. Acad. Sci. U.S.A. 96: 13068-13073. Yen, H . - C , N. T. Hu and B. L . Marrs (1979). Characterization of the gene transfer agent made by an overproducer mutant of Rhodopseudomonas capsulata. J. Mol. Biol. 131: 157-168. Yen, H . - C . and B. Marrs (1976). Map of genes for carotenoid and bacteriochlorophyll biosynthesis in Rhodopseudomonas capsulata. J. Bacteriol. 126: 619-629. Young, C. S., R. C . Reyes and J. T. Beatty (1998). Genetic complementation and kinetic analyses of Rhodobacter capsulatus ORF1696 mutants indicate that the ORF1696 protein enhances assembly of the light-harvesting I complex. J. Bacteriol. 180: 1759-1765. Zinder, N . D. and J. Lederberg (1952). Genetic exchange in Salmonella. J. Bacteriol. 64: 679-699. 

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.831.1-0089687/manifest

Comment

Related Items