Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Mechanisms of transcription modulation by low concentrations of rifampicin in Salmonella typhimurium Yim, Grace 2011

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

Item Metadata


24-ubc_2011_fall_yim_grace.pdf [ 4.09MB ]
JSON: 24-1.0103334.json
JSON-LD: 24-1.0103334-ld.json
RDF/XML (Pretty): 24-1.0103334-rdf.xml
RDF/JSON: 24-1.0103334-rdf.json
Turtle: 24-1.0103334-turtle.txt
N-Triples: 24-1.0103334-rdf-ntriples.txt
Original Record: 24-1.0103334-source.json
Full Text

Full Text

MECHANISMS OF TRANSCRIPTION MODULATION BY LOW CONCENTRATIONS OF RIFAMPICIN IN SALMONELLA TYPHIMURIUM by Grace Yim A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in The Faculty of Graduate Studies (Microbiology and Immunology) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) May 2011  © Grace Yim, 2011  Abstract. Screening of a S. typhimurium promoter-luxCDABE fusion library revealed that the antibiotic rifampicin modulated transcription of a subset of the fusions. This thesis examines promoter sequences of three down-modulated (fliA, flgK, invF) and three up-modulated (spvA, traS, STM3595) promoters in an effort to elucidate the mechanisms by which sub-minimal inhibitory concentrations of rifampicin modulate transcription.  Reverse transcriptase polymerase chain reaction was used as a second method to examine expression in response to rifampicin and the results were consistent with reporter data. Transcription modulation was abolished in a strain carrying a rifampicin resistance mutation in the gene for the beta subunit of RNA polymerase suggesting that rifampicin effects are mediated by its binding to RNA polymerase. Transcription modulation of the selected genes was observed after cells were treated with several antimicrobials; the effects on transcription differed from those of rifampicin.  DNA deletion studies of two promoters for genes involved in flagellar biosynthesis, PfliA and PflgK, showed that ~50 bp fragments containing the 28-dependent promoters were sufficient to observe rifampicin mediated down-modulation in culture. For PinvF, a promoter specific activator binding site was required for down-modulation. In vitro transcription from PinvF and PflgK was hypersensitive to rifampicin when compared to a control. The results indicate that the selectivity of rifampicin mediated down-modulation reflects variation in the direct interaction of rifampicin with different promoter:RNA polymerase complexes.  ii  In contrast to the down-modulated promoters, factors others than rifampicin and RNA polymerase holoenzyme may be involved in up-modulation. Transcription up-modulation from the promoters for STM3595 and traS could not be shown in vitro. In vivo, time course assays of rifampicin induced transcription up-modulation showed a lag period between rifampicin addition and induction. Use of spent media containing rifampicin was not able to shorten the lag phase, suggesting that an intracellular rather than a secreted factor was involved. DNA deletion studies suggested that nucleotides downstream of the transcription start site were involved with upmodulation. A potential intrinsic terminator was found in the untranslated region of the STM3595 transcript, suggesting that transcription attenuation may be involved in rifampicin mediated up-modulation.  iii  Preface. Chapter 3.1 includes work published in Yim, G., de la Cruz, F., Spiegelman, G. B. & Davies, J. (2006). Transcription modulation of Salmonella enterica serovar Typhimurium promoters by sub-MIC levels of rifampicin. Journal of Bacteriology 188:7988-7991. I conducted all research and wrote the corresponding manuscript.  iv  Table of contents. Abstract ........................................................................................................................................... ii Preface............................................................................................................................................ iv Table of contents ..............................................................................................................................v List of tables ................................................................................................................................. viii List of figures ................................................................................................................................. ix List of abbreviations ...................................................................................................................... xi Acknowledgements ...................................................................................................................... xiii Dedication .................................................................................................................................... xiv Chapter 1. Introduction ...................................................................................................................1 1.1  Antibiotics ............................................................................................................................1  1.2  Modulation of bacterial transcription by antimicrobials ......................................................6  1.3  Mode of action determination by reporter panels ................................................................8  1.4  Bacterial RNA polymerase ..................................................................................................9 1.4.1  Stages of transcription.........................................................................................12  1.4.2  Transcription regulation at initiation ..................................................................13  1.4.3  Regulation of transcription elongation, termination, antitermination and translational attenuation ...............................................................................14  1.5  1.6  Rifampicin:RNA polymerase interactions .........................................................................16 1.5.1  Mode of action ....................................................................................................16  1.5.2  Rifampicin resistance ..........................................................................................17  Pathogenicity and virulence of S. typhimurium ................................................................18 1.6.1  The virulence regulons ........................................................................................19  1.7  Flagellar regulon ................................................................................................................20  1.8  Thesis objective .................................................................................................................21  Chapter 2. Methods and materials ................................................................................................24 2.1  Bacterial strains, culture methods and reagents .................................................................24  2.2  Isolation of rifampicin resistant mutants............................................................................27 v  2.3  Luminescence time courses at various times of rifampicin addition .................................27  2.4  Pretreated cells and spent media experiments ..................................................................28  2.5  Liquid media lux reporter assays with sub-cloned reporters ............................................28  2.6  Reporter construction .........................................................................................................29  2.7  Disk diffusion assays .........................................................................................................31  2.8  RNA isolation and RT-PCR...............................................................................................33  2.9  Protein expression ..............................................................................................................33  2.10  Protein purification ............................................................................................................34  2.11  In vitro transcription assays ...............................................................................................34  2.12  5’ RACE.............................................................................................................................36  2.13  Primer extension ................................................................................................................36  2.14  Tn10d mutagenesis and Fluorescence Activated Cell Sorting ...........................................37  2.15  Inverse PCR methods.........................................................................................................38  Chapter 3. Results .........................................................................................................................40 3.1  In vivo characterization of transcription modulation by rifampicin...................................40 3.1.1  RT-PCR...............................................................................................................40  3.1.2  Measurement of rifampicin mediated transcription modulation in rifampicin resistant strains ....................................................................................................42  3.1.3  Disk diffusion assays ..........................................................................................43  3.1.4  Time courses .......................................................................................................47  3.1.5  Pretreated cells and spent media experiments ....................................................49  3.2  Mapping DNA regions necessary for transcription modulation by rifampicin .................51  3.3  In vitro transcription of rifampicin modulated promoters .................................................59 3.3.1  Rifampicin mediated down-modulation requires 28 and HilA ..........................60  3.3.2  Optimization of flgK and invF in vitro transcription .........................................62  3.3.3  Rifampicin hypersensitivity of flgK and invF transcription ..............................66  3.3.4  Mg2+and NTP concentration effects ...................................................................68  3.4  In vitro characterization of rifampicin up-modulated promoters .......................................71  3.5  Intracellular factors involved in rifampicin mediated up-modulation ...............................75 vi  3.6  Nucleotide sequence analysis of rifampicin modulated promoters ..................................79 3.6.1  Motif searches .....................................................................................................79  3.6.2  Preliminary exploration of nucleotide sequence motifs .....................................83  Chapter 4. Discussion ...................................................................................................................86 4.1  Mechanisms of rifampicin mediated down-modulation at RNA polymerase ...................87 4.1.1  The role of the +1 region in transcription initiation....................................91  4.1.2  A working model for rifampicin mediated down-modulation ....................92  4.2  Other possible mechanisms of rifampicin mediated down-modulation ............................93  4.3  Possible mechanisms of rifampicin mediated up-modulation ...........................................96 4.3.1  Known regulatory factors which bind RNA polymerase ............................98  4.3.2  Elongation and termination factors .............................................................99  4.3.3  Transcription attenuation and intrinsic terminators ..................................102  4.4  Rifampicin resistance .......................................................................................................105  4.5  Differential responses of reporter strains in liquid versus solid media ............................106  4.6  The response of rifampicin modulated promoters to various antibiotics .........................107  4.7  Biological relevance of transcription modulation by rifampicin .....................................109  4.8  Sub-MIC and its implications for virulence and motility ................................................112  4.9  Concluding remarks .........................................................................................................115  References ....................................................................................................................................117  vii  List of tables. Table 1.1  Characteristics of rifampicin responsive promoters in S. typhimurium 14028. .....23  Table 2.1  List of strains and phages .......................................................................................24  Table 2.2  List of plasmids ......................................................................................................25  Table 2.3  List of oligonucleotides..........................................................................................30  Table 3.1  Nucleotide alterations of common nucleotides in rifampicin down-modulated promoters and the resulting report plasmids ..........................................................84  viii  List of figures. Figure 1.1  The rif cluster and flanking genes rplL and rpoB of Amycolatopsis mediterranei and the structure of rifampicin ..................................................................................4  Figure 1.2  Structural overview of bacterial RNA polymerase ...................................................9  Figure 1.3  A model for the multi-step mechanism of transcription initiation ..........................13  Figure 2.1  Promoterless-lux reporter plasmid pCS26.. .............................................................32  Figure 2.2  Schematic of Tn10dtet and inverse PCR relevant primers and restriction sites .....39  Figure 3.1  Fold decreases of fliA and invF and fold increases of STM3595 and traS transcripts in response to rifampicin .......................................................................41  Figure 3.2  Comparison of luminescence produced by reporters carried by wildtype and rifampicin resistant strains.. ..............................................................................43  Figure 3.3  Disk diffusion assays with fliA, flgK, invF and STM2901 lux reporter strains .......44  Figure 3.4  Disk diffusion assays with traS, spvA and STM3595 lux reporter strains ..............46  Figure 3.5  Time courses of rifampicin activation and inhibition.. ...........................................49  Figure 3.6  A schematic representation of spent media experiments.. ......................................50  Figure 3.7  Spent media and pretreatment effects of rifampicin on STM3595 expression .......51  Figure 3.8  Mapping of rifampicin repressed promoters.. .........................................................53  Figure 3.9  Mapping of rifampicin activated promoters ............................................................56  Figure 3.10  Transcription start site of STM3595 as determined by 5’RACE ............................58  Figure 3.11  Induction and isolation of His-HilA and His-Sigma28 ............................................61  Figure 3.12  Optimization of flgK and invF transcription ...........................................................63  Figure 3.13  A schematic representation of incubation times used for in vitro transcription.. ....64  Figure 3.14  Identification of transcripts .....................................................................................65  Figure 3.15  Rifampicin inhibition of flgK, fliC, invF and abrB transcripts ...............................67  Figure 3.16  Graphs of transcription inhibition by rifampicin .....................................................67  Figure 3.17  Mg2+dependence of flgK and fliC transcription ......................................................69  Figure 3.18  Effect of NTP concentration on flgK and fliC transcription and rifampicin inhibition .................................................................................................................70 ix  Figure 3.19  Primer extensions of traS and STM3595 ................................................................72  Figure 3.20  A comparison of S. typhimurium and E. coli traS transcription start sites..............73  Figure 3.21  In vitro transcripts of traS, STM3595 and abrB in response to rifampicin.............74  Figure 3.22  Fluorescence of 14028 Tn10dTet mutants...............................................................75  Figure 3.23  Schematic representation of yciK and pcnB insertions ...........................................76  Figure 3.24  Luminescence of rifampicin up-modulated promoter reporter fusions in yciK::Tn10d and wildtype backgrounds. .................................................................77  Figure 3.25  Luminescence of rifampicin down-modulated promoter reporter fusions in wildtype and mutant backgrounds ...........................................................................78  Figure 3.26  Growth curves of the wildtype and pcnB mutant strains in the presence or absence of rifampicin .............................................................................................79  Figure 3.27  Motifs found in down- and up- modulated promoter sequences .............................80  Figure 3.28  Nucleotide sequence alignment of flgK, fliA and invF promoters ..........................81  Figure 3.29  Nucleotide sequences of STM3595, traS and spvA ................................................82  Figure 3.30  Luminescence values from mutated fusion reporters grown in the presence of varying amounts of rifampicin ............................................................................85  Figure 4.1  Sequence logo of the +1 region of E. coli promoters ..............................................92  Figure 4.2  A disk diffusion assay of the putative S. typhimurium 14028 hha lux reporter strain B9(2) ................................................................................................96  Figure 4.3  The ops sites of various E.coli and S. typhimurium genes ...................................102  Figure 4.4  Predicted RNA secondary structures of STM3595 and traS transcripts ...............104  Figure 4.5  Predicted secondary structure of the nucleotides from +1 to +65 nucleotides .....105  Figure 4.6  A schematic of putative virulence and motility pathways affected by rifampicin ..............................................................................................................114  x  List of abbreviations. ATP  adenosine 5’-triphosphate  bp  base pair  BSA  bovine serum albumin  CTD  carboxy terminal domain  CTP  cytidine 5’-triphosphate  DNA  deoxyribonucleic acid  EDTA  ethylenediaminetetra-acetic acid  EGTA  ethylene glycol bis(2-aminoethyl ether)-N,N,N'N'-tetra-acetic acid  FACS  Fluorescence Activated Cell Sorting  GTP  guanosine 5’-triphosphate  iNTP  initiating NTP  IPTG  isopropyl -D-1-thiogalactopyranoside  LB  Luria-Bertani  LBA  Luria-Bertani agar  MIC  minimal inhibitory concentration  mRNA  messenger ribonucleic acid  NTD  amino terminal domain  NTP  nucleoside triphosphate  OD  optical density  PAGE  polyacrylamide gel electrophoresis  PCR  polymerase chain reaction  ppGpp  guanosine tetraphosphate xi  RACE  rapid amplification of cDNA ends  RNA  ribonucleic acid  RNAP  RNA polymerase  RT-PCR  reverse transcriptase polymerase chain reaction  SDS  sodium dodecyl sulfate  SPI  Salmonella pathogenicity island  TBE  tris-borate EDTA  TE  tris-EDTA  UTP  uridine 5’-triphosphate  xii  Acknowledgements. I thank the Natural Sciences and Engineering Research Council of Canada (NSERC) for its funding to Julian E. Davies and George B. Spiegelman, the TALLY fund for its research support to Julian E. Davies and the Li Tze Fong Memorial Fellowship (UBC) and the NSERC PGS D for their support of Grace Yim. The suggestions and support of my committee members Drs. Charles Thompson and Erin Gaynor were greatly appreciated.  I would also like to acknowledge my many coffee drinking companions throughout the years: Leticia, Sachi, Geoff, Silke, Ivan, Raphael, Cedric and Tony. They made my research experience much more fun and enjoyable. The caffeine and conversation served as stimulants for many experiments. I would also like to thank Tony for drawing the structure of rifampicin with his excellent ChemDraw skills and for turning off the VictorII after my numerous weekend data collection runs.  xiii  Dedication. I would like to dedicate this dissertation to my family, my husband Alan and both the past and current members of the Davies and Spiegelman labs.  xiv  Chapter 1. Introduction. 1.1.  Antibiotics.  Antibiotics are molecules that kill or inhibit the growth of microbes. The word itself means “against life.” In the past 50 years, antibiotics have been critical to the control of many infectious diseases. Generally, antibiotics work by blocking crucial processes in microbial cells. Hence, antibiotics have been classified according to their structure and mode of action. Several examples of such classes are the rifamycin, beta-lactam, macrolide and aminoglycoside antibiotics. Rifamycins bind and inhibit bacterial DNA-dependent RNA polymerase (RNAP) (Campbell et al., 2001). Beta-lactams weaken the cell wall by inhibiting the transpeptidation of peptide strands within the bacterial peptidoglycan layer (Wickus & Strominger, 1972). Macrolide and aminoglycoside antibiotics inhibit protein synthesis. Macrolides block the elongation of the nascent peptide chain by binding to the 50S subunit of the ribosome (Tenson et al., 2003) while aminoglycosides bind to the A site of the 30S subunit (Moazed & Noller, 1987; Purohit & Stern, 1994).  Kohanski et al. has recently suggested a common mechanism of bacterial cell death in response to bactericidal antibiotics (Kohanski et al., 2007). Although the concept of antibiotic mode of action is well understood, the property by which an antibiotic is either lethal or inhibitory is unknown. Also unclear are the events following binding of the antimicrobial to its primary target, such as the consequences of protein synthesis inhibition, which lead to cell death. This is akin to disconnecting a tube or a wire in a car and the car stops running; one suspects key components are damaged but the exact process by which the car stopped is unknown. In E. coli, bactericidal but not bacteriostatic antibiotics were shown to induce hydroxyl radical formation.  1  In turn, hydroxyl radicals increased the rate of bacterial cell death. Deleterious effects of hydroxyl radicals could be ameliorated by the addition of thiourea and 2,2’-dipyridyl, agents which prevent the formation of radicals in bacteria (Kohanski et al., 2007).  Although the first antibiotic isolated for therapeutic use, penicillin G, is produced by a fungus, the majority of antibiotics have been isolated from bacteria of the genus Streptomyces. The first well characterized antibiotic isolated from a streptomycete was streptomycin. Selman Waksman, the discoverer of streptomycin, defined “true” antibiotics as natural products made by the metabolism of microorganisms that possess antimicrobial activity at low concentrations (Bryskier, 2005). Quinolones block DNA synthesis by targeting DNA gyrase and topoisomerase, enzymes responsible for DNA supercoiling and deconcatenation of interlinked daughter chromosomes following DNA replication (DiGate & Marians, 1988; Kato et al., 1990), respectively. Since quinolones are synthetic and not derived from microbes, by Waksman’s definition, the quinolones are not antibiotics but antimicrobials; however, they perform antibiotic functions in therapy.  Antimicrobial is a general term for natural, synthetic or semi-synthetic inhibitory small molecules. Natural products are isolated from cultures of microorganisms. Synthetic compounds are chemically derived and semi-synthetic products are modified (chemically or enzymatically) derivatives of natural products. Antimicrobials will either kill (bactericidal) or inhibit growth (bacteriostatic) of microbes. Most antibiotics are identified by screening growth medium of microorganisms for antibacterial activity. Beta-lactams and aminoglycosides are generally considered bactericidal while macrolides and tetracyclines are considered bacteriostatic  2  (Kohanski et al., 2007). For some antibiotics, such as rifampicin, bactericidal or bacteriostatic action on the cell is concentration dependent (Canadian Pharmaceutical Association, 2008).  Bacteria produce a plethora of diverse small molecules with many functions in nature. Some of these small molecules have been isolated for clinical use as antibiotics. Small molecule biosynthesis is a complex process. Many small molecule biosynthetic gene clusters are often in excess of 100 kb, encoding large multi-subunit enzyme complexes: non-ribosomal peptide synthases, polyketide synthases or hybrids thereof. Peptides synthesized by non-ribosomal peptide synthases have been found to contain over 100 unusual amino acids such as aminobutyrate, -alanine and -aminoadipate (Walsh, 2003). Genome analysis of a representative streptomycete, Streptomyces coelicolor, indicated the presence of 21 large biosynthetic clusters, each encoding proteins required for the synthesis of a specific molecule or a group of chemically related small molecules (Bentley et al., 2002). Three S. coelicolor clusters, including the well-studied polyketide antibiotic actinorhodin, had previously been identified to produce molecules having antibacterial properties (Bentley et al., 2002). Other predicted or known products of the encoded clusters include siderophores, pigments and lipids, as well as uncharacterized molecules (Bentley et al., 2002).  The role of small molecules in bacterial cell-cell signaling has been extensively studied as mediators of quorum sensing. Quorum sensing systems allow gene regulation in response to cell density. Examples of the “quorum” or cell density sensing by homoserine lactones in Gramnegative bacteria and peptides in Gram-positive bacteria have been widely documented. These signals can be passed within (Latifi et al., 1995) or between species (Surette et al., 1999).  3  Extensive work in this field has invigorated interest in bacterial cell-cell signaling (Novick & Geisinger, 2008; Parsek & Greenberg, 2000; Waters & Bassler, 2005; Williams & Camara, 2009).  Figure 1.1. The rif cluster and flanking genes rplL and rpoB of Amycolatopsis mediterranei and the structure of rifampicin. A. Genes encode the following functions: transport (red), post-polyketide synthase modification (orange), unknown (black), polyketide synthase (green), precursor synthesis (dark blue), regulation (pink) and sugar synthesis (light blue). Figure adapted from (Floss & Yu, 2005). B. The structure of rifampicin.  Once antibiotics are isolated for commercial use, a few naming conventions are loosely followed. The suffix “mycin” (as opposed to “micin”) is usually reserved for compounds isolated from Streptomyces. Derivatives of a natural product usually have the same ending as the original compound name (i.e. penicillin, ampicillin, oxacillin, carbenicillin, etc.) (Bryskier, 4  2005). Compounds have been named after the genus or species of the producer organisms, chemical characteristics of the antibiotic, places or even movie stars! For example, kanamycin was named after its producer S. kanamyceticus (Bentley & Bennett, 2009), bicyclomycin has a bicyclic structure (Bentley & Bennett, 2009), the producer organism of kasugamycin was first isolated from soil at the Kasuga shrine in Nara, Japan (Umezawa, 1967)) and melinamycin was named after the Greek actress and politician Melina Mercouri (J. Davies, personal communication).  Rifampicin is a clinically useful semi-synthetic compound derived from rifamycin B. Rifamycin B is produced by fermentation of Amycolatopsis mediterranei. Rifamycin B is subsequently modified to synthesize the final product rifampicin. The rifamycin B biosynthetic gene cluster is ~100 kb (Figure 1.1) and encodes a hybrid polyketide-non-ribosomal peptide synthase (Floss & Yu, 2005). Rifampicin inhibits growth of many Gram-positive bacteria, particularly Mycobacterium. It is used in the treatment of M. tuberculosis, M. leprae and other AIDSassociated mycobacterial infections. It is estimated that one third of the human population, primarily in Africa and South-East Asia, are infected with tuberculosis (World Health Organization, 2010). Interestingly, in May 2004, the Food & Drug Administration approved rifaximin, a derivative of rifampicin that is not absorbed by the body, for the treatment of traveller’s diarrhea caused by non-invasive strains of Escherichia coli (Adachi & DuPont, 2006).  What are the roles of antibiotics in nature? Antibiotics have been found to have different effects at high concentrations versus the effects at concentrations lower than the drug’s minimal inhibitory concentration (MIC, this term in discussed later in section 4.8). Clinically, antibiotics  5  are used in a concentrated and purified form to treat bacterial infections of humans. In the past decade, recognition of a second functionality of antibiotics has emerged, the modulation of bacterial gene transcription at sub-lethal dosages (Cheung et al., 2003; Evers et al., 2001; Goh et al., 2002; VanBogelen & Neidhardt, 1990). Antibiotic modulation of bacterial transcription raises the question of whether the natural purpose of antibiotic production by microbes is to inhibit growth of neighboring microbes for inter-microbial competition. It is likely that the concentrations of antibiotics found in the soil or in the environment are significantly lower than those encountered in clinical settings and would not aid in intermicrobial competition. Antibiotics are a type of naturally derived small molecule which have been isolated for inhibition of microbial growth. Other small molecules have been shown to have a diverse array of biological activities such as antiviral, antifungal, antitumor and immunosuppressive (Chadwick & Whelan, 1992; Demain & Fang, 2000).  1.2.  Modulation of bacterial transcription by antimicrobials.  Antibiotics and other small molecules have been observed to change bacterial transcription and protein expression patterns. Studies on the global effects of sub-lethal concentrations of various antimicrobials on different bacteria have enjoyed increasing popularity in the recent years (for a comprehensive review see (Davies et al., 2006)). For example, E. coli proteome studies have shown that various protein synthesis inhibitors induce expression of genes in a manner similar to a mild or strong heat shock (puromycin, kanamycin and streptomycin) or cold shock (chloramphenicol, erythromycin, fusidic acid, spiramycin, tetracycline) (VanBogelen & Neidhardt, 1990). In Pseudomonas aeruginosa, phenazines help maintain redox homeostasis by acting as electron acceptors (Price-Whelan et al., 2006). Quinolones stimulate expression of  6  integrase genes and integron recombination which may lead to antimicrobial induced acquisition of antibiotic resistance cassette or a reshuffling of cassettes (Guerin et al., 2009).  In addition to genes involved in metabolism and stress, antibiotics have been found to regulate the expression of virulence genes. Studies with enterohaemorrhagic E. coli have shown that expression of prophage genes, prophage encoded toxin genes and toxin release are increased by quinolones and trimethoprim (Herold et al., 2005; Kimmitt et al., 2000; Matsushiro et al., 1999; Zhang et al., 2000). Aminoglycosides induce biofilm formation via the bacterial second messenger cyclic di-GMP (Hoffman et al., 2005). In the Gram-positive Staphylococcus aureus and Streptococcus pyogenes, sublethal concentrations of antibiotics have been shown to suppress virulence factor synthesis by acting at the transcription and translation levels (Bernardo et al., 2004; Gemmell & Ford, 2002; Herbert et al., 2001). In Salmonella enterica serovar Typhimurium, S. typhimurium, cationic microbial peptides have been shown to affect virulence gene expression (Bader et al., 2003). A well studied precedent for the use of antibiotics to repress virulence function rather than growth inhibition is the use of macrolide antibiotics (such as erythromycin and azithromycin) in diffuse panbronchiolitis and cystic fibrosis infections. Macrolides do not reduce the bacterial load but inhibit the expression of virulence determinants and also have immunomodulatory effects (for a comprehensive review, see (Tateda et al., 2007)). The examples of transcription and protein expression changes elicited by sub-MIC antibiotics on different bacteria continually grow in number.  7  1.3.  Mode of action determination by reporter strain panels.  Gene or protein expression patterns of bacteria treated by antibiotics have been used as a means to group inhibitory small molecules of unknown structure and function with antimicrobials with known modes of action. Compounds with similar expression profiles have similar modes of action. A compound with a mode of action that differs from the reference compounds would putatively have a different expression profile. Early studies showed that different macrolidelincosamide-streptogramin antibiotics, with slightly different binding sites on the ribosome, could also be differentiated by their transcription profiles (Tsui et al., 2004). On a larger scale, reference libraries of transcriptome or proteome expression profiles from bacteria treated with different classes of antibiotics have been collected (Freiberg & Brotz-Oesterhelt, 2005). The modes of action of several small molecules have been identified or reclassified using expression profiles (Freiberg & Brotz-Oesterhelt, 2005; Hutter et al., 2004). Urban et al. (2007) identified a small number of promoters, chosen for their responsiveness to reference compounds, to create a panel of reporters that can be used in a similar fashion to the genome-scale expression patterns. These promoters can be either directly related to the compound mode of action, i.e the fabHB promoter to detect fatty acid biosynthesis inhibitors or can be promoters unrelated to the target of growth inhibition (i.e. involved in motility, virulence, metabolism) (Mesak et al., 2010) but are nonetheless responsive to the various reference compounds. These studies have used either lethal dosages of antibiotic for short exposure times (minutes) or sub-MICs for longer times (hours). These mode of action panels were employed to identify modes of action of unknown compounds present in small quantities in the supernatant of potential antibiotic producer organisms (Mesak et al., 2010); these compounds would likely be overlooked by traditional screening methods using growth inhibition of tester organisms to identify potential compounds.  8  1.4.  B Bacterial RNA A polymerasse.  w of bacteriaal RNA polyymerase. A schematic representatio r on of Figure 1.2 Structurral overview e coomplex. The main channnel secures the t RNA-DN NA hybrid inn the the T. theermophilus elongating hybrid-biinding site (ccenter). β doomains clam mp the downsstream DNA A in the DNA A duplex-binnding site (left)). The β zipp per, Zn fingeer, and lid allong with β-fl flap form thee RNA exit channel c (righht). The catallytic site (staar) is locatedd at the bottoom of the seccondary channnel (top lefft). Figure adapted from f Nudlerr (2009).  The enzyyme responsiible for RNA A synthesis in i the cell is RNAP. In the t best studdied organism m, E. coli, thhe RNAP co ore enzyme is i competentt for transcripption elongaation and is composed c off five subunits,, ’2 (Brrowning & Busby, B 2004) (Figure 1.22). The 3.3 Å crystal strructure of Thermus aquaticus RNAP R revealled a crab claw-like structure (Zhangg et al., 19999). The activve site whicch binds the DNA:RNA D c complex is formed f by thhe  and ’ subunits s (Koorzheva et all., 2000). A Mg2+ion iss bound at thhe beginning of the seconndary channeel through which w NTP diffuses into i the activ ve site (Zhanng et al., 19999). The 5’ end of the RNA R transcriipt exits the 9  active site via the main DNA/RNA channel (Korzheva et al., 2000). Each  subunit has two independently folded domains, the CTD (carboxy terminal domain) and the NTD (amino terminal domain) (Blatter et al., 1994). Dimerized NTD directs ’ assembly while CTD is a DNA binding module important for interactions with promoters and transcription factors (Gourse et al., 2000).  To begin transcription at a specific promoter sequence, the core enzyme requires a  factor to form the holoenzyme, ’2. In E. coli and many other species including S. typhimurium, 70 is the main  factor allowing RNAP to recognize the majority of promoters in the genome (Paget & Helmann, 2003). In Bacillus subtilis and many other organisms, 70 is referred to as A. E. coli has seven known  factors: 70, 28, 32, S, E, FecI and 54. Similarly, S. typhimurium has five known alternative sigma factors, 28, 32, S, E and 54 (Bang et al., 2005). S. coelicolor has 63 known sigma factors (Paget & Helmann, 2003). In bacteria, there are two families of  factors, 54 and 70. The members of the 70 family have four conserved domains as determined by sequence conservation in the 70 family (Paget & Helmann, 2003). In E. coli, 28, 32, S, E and FecI are members of the 70 family. The 70 family recognizes –10 and –35 hexamers while the 54 family recognizes –12 and –22 elements (Buck et al., 2000). Within bacteria, only regions two and four are well conserved in the 70 family (Paget & Helmann, 2003). The Salmonella 28 is a pared-down version of its 70 protein, containing only three of the four conserved domains (Schaubach & Dombroski, 1999).  10  Five DNA elements contribute to promoter recognition by RNAP:70. They are the -10 hexamer, -35 hexamer, extended -10, UP element and discriminator. The relative contribution of each element to promoter strength varies with the promoter (Browning & Busby, 2004). The consensus sequences are as follows: -10 (TATAAT), -35 (TTGACA), extended -10 (TGTG), UP elements (AAAWWTWTTTTNNNAAANNN where W = A or T and N is any base) and discriminator (GGG) (Haugen et al., 2008). The -10 hexamer is centered at -10 where +1 is the corresponding transcription start site. The -35 region is centered around -35, the extended -10 is located immediately upstream of the -10 hexamer, UP elements typically appear from -57 to -38 and the discriminator is directly downstream of the -10 from -6 to -4. Specific regions of 70 bind to the -35, extended -10, -10 and discriminator, while the CTD binds UP elements (Haugen et al., 2008). Consensus sequences for other sigma factors are not as well defined and can be quite different from the 70 consensus. For example, E.coli and S. typhimurium 28 (flagellar) consensus sequences are TAAAGTTT (-35) and GCCGATAA (-10) (Ide et al., 1999). An extended -10, UP element and discriminator have not been described for 28.  While natural promoters in which all elements exactly match all the consensus sequences do not exist (Browning & Busby, 2004), natural promoters have been altered to have a higher percent identity to consensus. Mutations causing higher nucleotide similarity to consensus and increased promoter strength are often referred to as “up” mutations while those changing sequences away from consensus and decreasing promoter strength are referred to as “down” mutations. Examples of classical “strong” promoters with “up” mutations are as follows: the lacUV5 promoter (-35, TTTACA, -10 TATAAT ) has two up mutations (bold) compared to the wildtype lac promoter (-35 TTTACA, -10 TATGTT) (Fuller, 1982). The tac promoter has consensus -10 11  and -35 elements and was derived by combining the -10 element of the trp promoter and the -35 element of the lacUV5 promoter, creating a promoter which was stronger than either of its progenitors (de Boer et al., 1983).  1.4.1. Stages of transcription. Transcription has been functionally separated into three parts: initiation, elongation and termination. The kinetics of initiation has been studied in detail but many questions still remain. During the process of initiation, RNAP and DNA will transition through a series of isomerizations that reflect structural changes in both RNAP and DNA. Figure 1.3 shows a simple schematic of transcription initiation. Some promoters have additional closed and open complexes that have been referred to as RPC1, RPC2, RPO1 and RPO2 (Record et al., 1996). Closed complex is characterized by specific binding of RNAP to promoter DNA without separation of the DNA strands. The open complex is formed by melting of the double stranded DNA template in the -10 region of the promoter. The transition from open to initiated complex occurs as the RNAP polymerizes NTPs. Once a 7 – 12 nucleotide transcript forms, RNAP forms a processive elongation complex. At some promoters, short RNAs (abortive transcripts) are released and the RNAP returns to an open complex to begin again (Record et al., 1996). The conversion from initiated complex to elongating complex involves promoter escape and clearance. The rate of chain elongation is approximately 50-100 nucleotides/s (Roberts et al., 2008). Termination occurs when the nascent RNA transcript is released from RNAP and the DNA template. It can be mediated by the protein Rho, the protein Mfd or by intrinsic terminator sequences (regions are characterized by a hairpin followed by a run of U residues) (Roberts et al., 2008). In some transcription units, terminator sequences can be found in the untranslated  12  region, between the core promoter elements and the start codon for translation. These sequences can cause attenuation, which is discussed below.  Figure 1.3. A model for the multi-step mechanism of transcription initiation. Figure modified from (Record et al., 1996)).  1.4.2  Transcription regulation at initiation.  Transcription factors down or up-regulate transcription by influencing RNAP. Regulatory factors will often affect transcription by stabilizing or destabilizing certain forms of the RNAP:DNA complex, such as open complex, or help recruit RNAP to specific promoters. Many transcription factors bind DNA specifically at or upstream of the core promoter elements (Browning & Busby, 2004). Some notable exceptions to this include the transcription factor DksA which potentiates transcription of rRNA promoters by binding in the RNAP secondary channel (Paul et al., 2004). Some global regulators have dual roles as activators and repressors, their action as either repressors or activators depends on their binding sites. For example, CRP and FNR have repressor binding sites located around the +1 and have upstream activator binding sites (Gralla & Collado-Vides, 1996).  13  1.4.3  Regulation of transcription elongation, termination, antitermination and translational attenuation.  Although most mechanisms of transcription regulation studied to date involve regulation of transcription initiation, there are a number of well documented mechanisms that involve regulation of transcription elongation and termination. The RNAP transitions into an elongating complex when RNAP has released  and has cleared the promoter region of the DNA; at this time the nascent RNA is approximately 7-12 nucleotides long (Record et al., 1996). Obstacles to elongation include DNA lesions and DNA binding proteins such as 70 which can bind to -10like sequences (Ring et al., 1996). These obstacles can cause stalling and a subsequent buildup of elongating complexes (Ring et al., 1996). Transcription elongation factors such as NusA, NusG, GreA and GreB are known to modulate the rate of elongation by stimulating or inhibiting paused elongating complexes (Borukhov et al., 2005; Burns et al., 1998; Lewis et al., 2008). Gre proteins can also enhance elongation by inhibiting abortive initiation complexes (Vassylyeva et al., 2007). When transcribing rRNA, elongating complexes form antitermination complexes which are highly sensitive to intrinsic and extrinsic pause and termination signals (Lewis et al., 2008).  Attenuation has been broadly defined as modulation of gene expression that influences aspects of transcription or translation but does not influence transcription initiation (Kolter & Yanofsky, 1982); others have described it as repression involving neither repressors nor activators (Turnbough & Switzer, 2008). Transcription attenuation is hallmarked by intrinsic terminators and a leader region. The first well-studied attenuation controlled transcript was the E.coli trp operon. The leader region is defined as the RNA from the transcription start site to the start of protein coding region and in the trp operon, this region contains a short peptide encoding ORF 14  with its own ribosome binding site. In general, attenuation involves the following mechanism (Turnbough & Switzer, 2008). When transcription occurs without pausing or slowing, the transcript will form an intrinsic terminator (hairpin structure) in the leader region of the nascent RNA causing transcription to terminate. When the elongating complex has slowed or paused, the ribosome which is translating the leader peptide follows quickly behind the elongating complex (transcription translation coupling), blocking the formation of the hairpin in the transcript and allowing transcription to continue without termination.  Examples of transcripts regulated by attenutation include pyrBI (Turnbough & Switzer, 2008), cat (Alexieva et al., 1988) and ermC (Mayford & Weisblum, 1989). The pyrBI operon encodes aspartate carbamylase, an enzyme which catalyzes the first step in the de novo synthesis of pyrimidine nucleotides. Attenuation control of pyrBI transcription coordinates the level of aspartate carbamylase synthesis to the intracellular levels of UTP. In the presence of high levels of UTP, ~98% of transcription terminates (Turnbough & Switzer, 2008). E.coli pyrBI has a 158 bp leader region which includes a 44 codon ORF, multiple clusters of U residues (UTP sensitive) and an intrinsic terminator 23 bp before the pyrB structural gene. When UTP levels are low, RNAP will stall at U-rich pause sites, allowing transcription translation coupling, prevention of terminator formation and transcription through the pyrBI operon. If UTP levels are high, RNAP will not pause and no coupling occurs. The terminator will form in the mRNA and terminate transcription, preventing transcription of the pyrBI genes (Turnbough & Switzer, 2008).  Attenuation can also control the level of translation. Translation of antibiotic resistance genes such as cat and erm, which confer chloramphenicol and erythromycin resistance to cells,  15  respectively, is regulated by attenuation. The leader region of the catA86 gene of B. subtilis requires a minimum 84 bp (Alexieva et al., 1988). The leader region has a nine codon ORF and a stemloop in the mRNA that when formed, sequesters the ribosome binding site required for translation of the cat-86 structural gene (Alexieva et al., 1988). If chloramphenicol binds to and stalls the ribosome, stalling at discrete sites in the leader peptides causes destabilization of the mRNA secondary structure and allows translation of the downstream resistance determinant (Alexieva et al., 1988).  1.5.  Rifampicin:RNA polymerase interactions.  1.5.1. Mode of action. The biochemical target for rifampicin in the bacterial cell is the  subunit of bacterial RNAP and is encoded by rpoB. The most highly cited in vitro study on rifampicin mode of action was published by McClure & Cech in 1978. They suggested that rifampicin inhibited transcription by sterically blocking elongation of nascent RNA. The crystal structure of rifampicin complexed with T. aquaticus RNAP was solved and supported the classical model for rifampicin mode of action (Campbell et al., 2001). A rarely cited study suggested that rifampicin may inhibit transcription by destabilizing the binding of intermediate oligonucleotides to the active RNAP:DNA complex (Schulz & Zillig, 1981). In 2005, a study using several rifamycin antibiotics challenged the classical model for the mode of action of rifampicin suggesting that rifamycins do not work solely by steric hindrance but also by inducing allosteric changes in RNAP that inhibit its catalytic activity (Artsimovitch et al., 2005). Subsequently, this newly proposed mode of action of rifampicin was challenged by Feklistov et al. (2008). At present, the  16  only known mode of action for rifampicin remains the one suggested by McClure and Cech (1978).  1.5.2. Rifampicin resistance. Rifampicin resistance can be conferred by mutations of the drug target, by drug inactivation or by efflux of the antibiotic. Resistance mutations usually occur in rpoB (Musser, 1995). This predominant site for resistance conferring mutations was early genetic evidence for RNAP as the target of rifampicin. Some environmentally and clinically isolated bacteria possess rifampicin modifying enzymes, such as those encoded by arr which inactivate rifampicin by ADPribosylation; these proteins seem to more clinically important in non-mycobacterial strains (Floss & Yu, 2005). Bacterial rifampicin resistance arises quickly during therapy. In an effort to slow the emergence of resistance in the treatment of M. tuberculosis, rifampicin is used in combination along with isoniazid, pyrazinamide and ethambutol (first-line drugs) (Musser, 1995). Widespread rifampicin resistance is of great clinical concern, as there are many cases of both multidrug resistance (resistant to isoniazid and rifampicin) and extensively drug resistant (resistant to rifampicin, isoniazid and second line drugs) M. tuberculosis (World Health Organization, 2010). This global problem has raised interest in the study of the mode of action of rifampicin, its derivatives (rifapentin, rifabutin, etc.), as well as further derivatization to create more drugs (Artsimovitch et al., 2005).  Since RNAP is a component of the basic cellular machinery, studies have examined the phenotypes of different rpoB mutants. In E.coli, particular phenotypes appear to be associated with different clusters of mutations in the rpoB gene. Phenotypes include altered transcription  17  termination, temperature sensitivity and uracil susceptibility (Jin et al., 1988; Jin & Gross, 1989). These studies elucidated the pleiotropic effects resulting from alterations to cellular machinery and functional specialization of certain areas of the RNAP  subunit.  1.6. Pathogenicity and virulence of S. typhimurium. Salmonella enterica serovar Tyhpimurium (S. typhimurium) is a facultative intracellular pathogen which lives a variety of pathogenic lifestyles. Salmonellosis is usually caused by the ingestion of contaminated food or water. S. typhimurium infection is usually limited to gastroenteritis (unable to invade intestinal epithelium) in humans but causes a systemic (typhoid like) infection in several strains of mice (Holden, 2002). S. enterica serovar Typhi causes typhoid fever, a systemic infection in humans but not in nonprimate mammals (Holden, 2002). Thus, mice infected with S. typhimurium are often used as an animal model for human typhoid fever (Holden, 2002). During systemic infection (S. typhimurium infection of mice or S. typhi infection of humans), bacteria travel from the stomach and colonize the intestinal epithelial cells. At the intestinal wall, injection of proteins using the type III secretion system induces epithelial cells to take up Salmonella (Galan, 1999). Salmonella disseminates to local mesenteric lymph nodes where it invades and survives in phagocytic cells. Salmonella then travels to extraintestinal sites, i.e. spleen and liver, resulting in a systemic infection. This thesis focuses on two S. typhimurium virulence genes invF and spvA. Genes such as invF, related to enteropathogenic infection, allow Salmonella to invade host cells. Genes such as ssrAB and spvABCD are related to systemic infection and allow S. typhimurium to survive and replicate within epithelial and macrophage cells inside a unique membrane-bound vacuole (Holden, 2002).  18  The majority of virulence factors present in Salmonella species are encoded within Salmonella Pathogenicity Islands (SPIs) or on the virulence plasmid (Hensel, 2004). Thus far, 21 SPIs distributed throughout the various serovars have been identified (Blondel et al., 2009; Hensel, 2004). The SPIs were likely acquired through horizontal gene transfer (Groisman & Ochman, 1996). The Salmonella virulence plasmids vary in size between the serovars and can restore virulence to plasmid cured Salmonella strains (Gulig et al., 1993). Examples of genes on the Salmonella virulence plasmids are the Salmonella Plasmid Virulence, spv, genes (Chu et al., 1999).  1.6.1. The virulence regulons. Virulence genes in S. typhimurium can be grouped into 1) those involved with the enteropathogenic phase of infection (invasion of intestinal cells) or 2) those involved in the systemic phase of infection. The SPI-1 genes and their associated effector proteins are central to the enteropathogenic phase of infection. The primary regulator for SPI-1 expression is HilA, a SPI-1 encoded protein in the OmpR/ToxR family (Bajaj et al., 1996). HilA expression is influenced by a multitude of proteins: CsrAB, Fis, Fad, FliZ, OmpR/EnvZ, HilC, HilD, PhoPQ, PhoBR, Lon protease, Ams, HupB and Hha (Teplitski et al., 2003). HilA has been shown to activate two multi-gene operons encoded on SPI-1. The first genes of these operons are prgH and invF, respectively (Lostroh et al., 2000). InvF in turn positively regulates SPI-1 encoded effectors which are secreted by the SPI-1 encoded type III secretion system (Eichelberg & Galan, 1999). The siiA operon, located in SPI-4, is required for the enteropathogenic phase of infection and is also HilA dependent (Thijs et al., 2007). There are at least seven SPI-1 effectors encoded by genes located outside SPI-1: SopA, SopD, SopE, SopE2, SlrP, SspH1 and SopB (SigD)  19  (Raffatellu et al., 2005). Their transcription is thought to be directly activated by InvF and therefore indirectly by HilA (Eichelberg & Galan, 1999).  Known regulators and genes that are involved in the systemic phase of infection include spvABCD and the genes encoded in SPI-2. SPI-2 genes are essential for systemic infection in mice. They are expressed upon entry into host cells and are required for survival in macrophages (Hensel et al., 1998). The regulator for SPI-2 is likely ssrAB which encodes a two-component regulatory system (Feng et al., 2004). The spv genes are also involved in systemic infection and survival in macrophages (Libby et al., 2000; Matsui et al., 2001), but it is unclear if their regulation is related to SsrAB.  1.7.  Flagellar regulon.  The genes responsible for flagella formation and motility in E. coli and S. typhimurium are transcriptionally organized into classes that form a hierarchy (Soutourina & Bertin, 2003). Products of the class I genes are required for transcription of class II genes and products of class II genes are required for transcription of class III genes. This thesis focuses on two flagellar genes, fliA (the flagellar sigma factor, class II) and flgK (hook associated protein, class III). The flagellar regulon consists of >50 genes and >17 operons. Class I genes encode flhD and flhC. FlhD2C2 activates the 70 dependent class II genes (Liu & Matsumura, 1994). Class II genes encode proteins that compose the flagellar export system, the basal body proteins, 28 (fliA) and FlgM, the anti-28 factor (Kutsukake & Iino, 1994; Kutsukake, 1997; Ohnishi et al., 1992). Transcription of class III genes is 28 dependent and is inhibited by FlgM. FlgM sequesters 28 until the basal body of the flagella is complete, at which time FlgM is exported out of the cell via 20  the flagellar export system (Hughes et al., 1993). Subsequently, the transcription of the class III genes, encoding the flagellar filament, hook-associated, motor function and chemotaxis proteins, proceeds and the flagella are completed.  1.8.  Thesis objective.  Davies et al. demonstrated transcription modulation in response to several antibiotic classes in S. typhimurium 14028 using a 6528-clone library constructed using a luxCDABE promoter trap vector (Goh et al., 2002; Tsui et al., 2004; Yim et al., 2011). Using this library of promoterreporter clones, it was shown that sub-MIC levels of the antibiotic rifampicin, dramatically upand down-modulated transcription from many reporter strains (Table 1.1). S. typhimurium genes involved with host cell invasion and motility were putatively found to be down-modulated by rifampicin and those involved with carbon metabolism or virulence were putatively upmodulated by rifampicin (Table 1.1). The following in vivo and in vitro experiments were conducted to elucidate the mechanism by which transcription modulation by rifampicin occurs.  i) To characterize the rifampicin response, the expression of promoter-reporter fusions were studied under various liquid and solid media growth conditions. To determine whether the target of rifampicin binding leading to transcription modulation by rifampicin was RNAP, the wild type strain and a rifampicin resistant strain were examined for their responses to rifampicin. To examine the specificity of the rifampicin response, strains harboring reporters for six rifampicin responsive promoters were exposed to rifampicin, various other antibiotics and oxidative stress. To characterize the timing of the rifampicin response, expression of the promoter fusions was measured after rifampicin was added at different times during culture growth. To determine  21  whether a compound accumulated intracellularly or was released into the supernatant to mediate the rifampicin response, experiments employing spent media and rifampicin pretreated cells were conducted.  ii) Two types of experiments (reporter constructs and RT-PCR) were used to confirm the identity of rifampicin responsive genes. Since the original lux reporter library used to screen for rifampicin responsive genes was constructed by cloning partial Sau3A digested DNA, deletion analysis of presumptive promoters was conducted. Expression of fusions carrying various truncations of the original DNA fragments grown in the presence and absence of rifampicin was examined. Smaller deletions were used to map regions of DNA necessary for rifampicin mediated transcription effects. RT-PCR was used a secondary method to check expression levels of rifampicin modulated genes.  iii) In vitro transcription was used to determine whether it was possible to replicate transcription modulation by rifampicin with isolated components in vitro.  iv) Since several lines of evidence suggested that an unknown cellular factor was involved in rifampicin mediated up-modulation, a screen for mutants no longer displaying up-modulation by rifampicin was conducted to identify genes or RNAs involved in transcription up-modulation by rifampicin. A strain carrying a plasmid with gfp fused to a promoter which was strongly upmodulated by rifampicin was randomly mutagenized with Tn10dtet. Mutants showing loss of transcription modulation by rifampicin were isolated by FACS and further studied using the Victor II plate reader.  22  v) To identify possible nucleotide motifs involved in transcription modulation by rifampicin, nucleotide sequences for the six rifampicin responsive promoters identified in ii) were analyzed for motifs. Luminescence of lux reporters carrying promoter DNA with altered motifs were examined for their responses to rifampicin.  Table 1.1. Characteristics of rifampicin responsive promoters in S. typhimurium 14028. Fold Repression (↓) or Activation (↑) by Promoter Identification Rifampicin Putative Function STM2899/invF ↓ 200 virulence; invasion STM1091/sopB ↓ 163 virulence; invasion STM2066/sopA ↓ 115 virulence; invasion STM4255-8/unknown, ssb, siiAB ↓ 13 virulence; invasion F ↓ 32 motility STM1956/fliA ( ) STM1914/flhBA ↓ 26 motility STM1183/flgK ↓ 18 motility STM2199/cirA ↓4 iron metabolism STM1328 ↓ 25 unknown STM1248 ↓ 19 unknown pSLT102-3/traST ↑5 virulence STM1154-55/yceE,htrB ↑4 virulence; systemic pSLT041-39/spvRAB ↑4 virulence; systemic STM4118/yijP ↑3 virulence STM4454/treB ↑ 13 carbon metabolism STM2445/ucpA ↑4 carbon metabolism STM1597/ydcW ↑3 carbon metabolism STM2473/talA ↑3 carbon metabolism ↑8 RNA modification STM0425/thiI ↑ 24 unknown STM3595 ↑4 unknown STM0389/yaiA ↑3 unknown STM2287 ↓ 1.6 unknown STM2901 (control) *modified from (Yim et al., 2006)  23  Chapter 2. Methods and materials. 2.1.  Bacterial strains, culture methods and reagents.  Luria-Bertani (LB) broth or agar (LBA) plates were used for growth and maintenance of strains. All media components were obtained from EMD Chemicals Inc. (Gibbstown, NJ, USA). Liquid cultures were grown at 37 C in shaking tubes while cultures in 96-well plates were grown without shaking. Media were supplemented with tetracycline (12.5 μg ml-1), chloramphenicol (30 μg ml-1), ampicillin (100 μg ml-1), kanamycin (25 μg ml-1) and rifampicin (5 μg ml-1) as appropriate and unless otherwise indicated. All antibiotics were obtained from Sigma-Aldrich (St. Louis, MO, USA) or from the laboratory collection. NTPs were obtained from Amersham Biosciences (Piscataway, NJ, USA); [-32P]GTP (111 TBq mmol-1), [-32P]ATP (111 TBq mmol-1) and [-32P]ATP (222 TBq mmol-1) were supplied by PerkinElmer Life Sciences (Waltham, MA, USA).  Table 2.1. List of strains and phages. Description  Source and/or reference  Strains 14028 DH10B BL21 star BL21(DE3) pLysS TOP10 GY10 GY11 R306 TT10423  S. typhimurium wildtype strain E. coli strain used for cloning E. coli strain used for protein expression E. coli strain used for protein expression E. coli strain used for protein expression Derivative of 14028, pcnB::Tn10d Derivative of 14028, yciK::Tn10d Derivative of 14028, rifampicin resistant proAB47 / F' pro(+) lac(+) zzf-1831::Tn10dtet  (Goh et al., 2002) Invitrogen Invitrogen Novagen Invitrogen This study This study (Yim et al., 2006) M.G. Surette  Phages P22 HT105/1-int P22 H5  Salmonella phage Salmonella phage  M.G. Surette S. Ramey 24  Table 2.2 List of plasmids. Plasmid Description pCS26 Promoterless luxCDABE reporter, kanr pCH112  hilA cloned into pBAD/Myc-His, ampr  pKH445  fliA cloned into pET15b, ampr  pAW44  template for abrB PCR  pGY7  Derivative of pCS26; fliA lux library reporter  pGY8 pGY9 pGY10 pGY11 pGY12  Sau3A fliA fragment of pGY7 inserted into pCS26 Derivative of pGY7, pCSFor/fliAP2 rev used for construction Derivative of pGY7, fliAP1 for/fliAP1 rev used for construction Derivative of pGY7, fliAP2 for/fliAP2 rev used for construction Derivative of pCS26; flgK lux library reporter  pGY13  Derivative of pGY12; flgKent2 for/flgKent rev used for construction Derivative of pGY12; flgKent for/flgKent rev used for construction Derivative of pGY12; flgKmin for/flgKmin rev used for construction Derivative of pCS26; invF lux library reporter  pGY14 pGY15 pGY16 pGY17  pGY21  Derivative of pGY16; invFent for/invFent rev used for construction Derivative of pGY16; invFent for/invFmin rev used for construction Derivative of pGY16; invFmin for/invFent rev used for construction Derivative of pGY16; invFmin for/invFmin rev used for construction Derivative of pCS26; STM2901 lux library reporter  pGY22  Derivative of pCS26; traS lux library reporter  pGY23  Derivative of pGY22; slyAent for/ slyAent rev used for construction Derivative of pGY22; traTent for/ traTent rev used for construction Derivative of pGY22; traSent for/ traSent rev used for construction Derivative of pGY22; traSent for/ traSmin rev used for construction  pGY18 pGY19 pGY20  pGY24 pGY25 pGY26  Source (Goh et al., 2002) (Lostroh et al., 2000) (Chadsey et al., 1998) M. Strauch, U. of Maryland (Goh et al., 2002) This study This study This study This study (Goh et al., 2002) This study This study This study (Goh et al., 2002) This study This study This study This study (Goh et al., 2002) (Goh et al., 2002) This study This study This study This study  25  Plasmid pGY27 pGY28 pGY29 pGY30 pGY31 pGY32 pGY33 pGY34 pGY35 pGY36 pGY37 pGY38 pGY39 pGY40 pGY41 pFPV25 pGY42 pGY43 pGY44 pGY45 pGY46 pGY47 pGY48 pGY49  Description Derivative of pGY22; traSmin for/ traSent rev used for construction Derivative of pGY22; traSmin for/ traSmin rev used for construction Derivative of pCS26; spvA lux library reporter Derivative of pGY29; spvAent2 for/ spvAent rev used for construction Derivative of pGY29; spvAent for/ spvAent rev used for construction Derivative of pGY29; spvAent for/ spvAmin rev used for construction Derivative of pGY29; spvAmin for/ spvAent rev used for construction Derivative of pGY29; spvAmin for/ spvAmin rev used for construction Derivative of pCS26; STM3595 lux library reporter Derivative of pGY35; STM3595ent2 for/ STM3595ent rev used for construction Derivative of pGY35; STM3595ent for/ STM3595ent rev used for construction Derivative of pGY35; STM3595min for/ STM3595ent rev used for construction Derivative of pGY35; STM3595ent for/ STM3595min rev used for construction Derivative of pGY35; STM3595min for/ STM3595min rev used for construction Derivative of pGY35; STM3595PE2 for/ STM3595ent rev used for construction Promoterless gfp reporter with a FACS optimized gfp Derivative of pFPV25; STM3595::gfpmut3a Derivative of pCS26; abrB for and abrB rev used for construction Derivative of pGY11, fliAP2 C for/fliAP2 rev used for construction Derivative of pGY11, fliAP2 for/fliAP2 GCC rev used for construction Derivative of pGY11, fliAP2 C for/fliAP2 GCC rev used for construction Derivative of pGY15; flgKmin C for/flgKmin rev used for construction Derivative of pGY15; flgKmin for/flgKmin GCC rev used for construction Derivative of pGY15; flgKmin C for/flgKmin GCC rev used for  Source This study This study (Goh et al., 2002) This study This study This study This study This study (Goh et al., 2002) This study This study This study This study This study This study (Valdivia & Falkow, 1997) This study This study This study This study This study This study This study This study 26  Plasmid  Description construction pGY50 Derivative of pGY18; invFent for/invFmin GCC rev used for construction pGY51 Derivative of pGY19; invFmin C for/invFent rev used for construction pNK2880 contains the transposase gene pLysS Used for protein expression  2.2.  Source This study This study M.G.Surette Novagen  Isolation of rifampicin resistant mutants.  Overnight cultures of S. typhimurium 14028 were spun down and spread on LBA supplemented with 300 μg ml-1 of rifampicin and incubated at 37 C for two days. Colonies were purified twice for single colonies on LBA supplemented with 300 μg ml-1 of rifampicin. To identify the mutation conferring rifampicin resistance, the following primer pairs were used to amplify regions of the rpoB gene: ST rpoBI sense/antisense and ST rpoBII sense/antisense. PCR products were nucleotide sequenced using the primers used for amplification (NAPS Unit, UBC, Canada). (The MIC for rifampicin when S. typhimurium 14028 is grown on LBA, as determined by Etest (AB Biodisk, Solna, Sweden) or by the microbroth dilution method, is ~12 g/ml.)  2.3.  Luminescence time courses at various times of rifampicin addition.  Overnight cultures of reporter strains were diluted 1:50 into 60 ml of LB supplemented with kanamycin and 15 ml aliquoted into four 50 ml flasks and were incubated with shaking at 37 °C. Rifampicin was added to a different flask every hour for the first two hours. 150 μl aliquots were removed from each flask every hour and luminescence (cps) and OD (595 nm) were measured in a Victor II Multi-label Counter (PerkinElmer, Waltham, MA, USA).  27  2.4.  Pretreated cells and spent medium experiments.  Spent medium and pretreated cells were collected as depicted in Figure 3.5. An overnight culture of 14028 carrying the STM3595 reporter pGY37 was diluted 1:100 into two flasks containing 35 ml of LB supplemented with kanamycin and incubated with shaking at 37 °C for three hours. After one and three hours, OD600 nm was measured in a Genesys 20 spectrophotometer (Thermo Electron, Milford, MA, USA). Rifampicin was added to one flask after one hour. Both flasks were incubated shaking at 37 °C for an additional two hours. Three hours after the initial dilution, cells from both flasks were spun down and 15 ml or 30 ml of supernatant/spent medium was removed from the rifampicin flask or LB flask, respectively. Excess medium was removed from centrifuge tubes and discarded. Pretreated cells were resuspended in fresh medium to an OD600 nm of 0.02. Resuspended cells were aliquoted into two 15 ml cultures and rifampicin was added to one flask. 30 ml of LB spent medium was inoculated at 1:50 with the overnight culture and split into two 15 ml cultures and rifampicin was added to one flask. 15 ml of rifampicin supplemented spent medium was inoculated 1:50 with the overnight culture. Flasks with resuspended cells and spent medium were incubated shaking an additional two to three hours at 37 °C. 150 μl of culture was removed from flasks every 30 minutes and luminescence and OD595 nm were measured in a Victor II Multi-label Counter.  2.5.  Liquid media lux reporter assays with sub-cloned reporters.  Luminescence measurements were taken in 96-well sterile, white, clear-bottomed microtitre plates (Nalge Nunc, Rochester, NY). 150 μl of overnight cultures of reporter strains were aliquoted into 96 well plates and a 96-pin replicator (V&P Scientific, San Diego, CA) was used to inoculate microtitre plates containing LB kanamycin supplemented with or without rifampicin  28  (5.0 μg ml-1 or as otherwise indicated). For luminescence assays described in sections 3.1.2, 3.2, 3.3.3 and 3.5, plates were sealed with Mylar Plate Sealers (Thermo Electron) or a Breathable Sealing Membrane (#163340, Nalge Nunc, discontinued), incubated at 37 C in a Victor II Multi-label Counter without shaking and luminescence measured every hour for 16-21 hours. Peak luminescence readings for replicates were used. For short assays described in section 3.6.2, inoculated plates covered with lids were incubated without shaking for three hours at 37 C at which point lids were removed, luminescence and OD595 nm read in the Victor II Multi-label Counter every 30 minutes for four additional hours. Using three different seed cultures, a minimum total of six sets of bacterial luminescence measurements were taken.  2.6.  Reporter construction.  All plasmid constructs (Table 2.2) were cloned in E. coli DH10B (Invitrogen, Carlsbad, CA). With the exception of pGY8, derivatives of pCS26 were made by insertion of BamHI-XhoI digested PCR products into the BamHI-XhoI fragment of pCS26. Primers for DNA amplification are listed in Table 2.3. S. typhimurium genomic DNA was used as a template for all PCR reactions with two exceptions: for pGY9 construction pGY8 plasmid DNA was used as a PCR template and for pGY43 construction pAW44 was used as a template. pGY8 was constructed by cloning the Sau3A fragment of pGY7 containing the fliA gene into pCS26 digested with BamHI. pGY42 was constructed by cloning the BamHI digested product of PCR amplification of genomic DNA using STM3595ent for and STM3595ent rev into the BamHISmaI (blunt end) fragment of pFPV25. Plasmid constructs were validated by nucleotide sequencing (Macrogen, Korea). Plasmids isolated from E. coli were electroporated into S. typhimurium 14028 by standard methods.  29  Table 2.3. List of oligonucleotides. Oligonucleotide Nucleotide Sequence 5’ to 3’ AAP GGCCACGCGTCGACTAGTACGGGIIGGGIIGGGIIG abrB for ACTCGAGAAACAAAATGATTGACG abrB rev TGGATCCTTACCTTCATAGCATAAC AUAP GGCCACGCGTCGACTAGTAC flgKent for ACTCGAGATTCCGCGTCTATAGCTC flgKent rev AGGATCCAAGCTGGACATGATGGTTC flgKent2 for TCTCGAGTGGTGCGCAAAGCCATAC flgKmin for ACTCGAGTTTTAAATTGCTCAAGTC flgKmin C for ACTCGAGTTTTAAATTGCTCAAGTCCAAGT flgKmin rev AGGATCCTCAATACTCGTTGTTATC flgKmin GCC rev AGGATCCTCAGCCCTCGTTGTTATC fliAP1min for TCTCGAGTTATTCCTTCGATAGAACCCTC fliAP1min rev TGGATCCGTTATCGGCATGATTATC fliAP2min for TCTCGAGGATAGAACCCTCTGTAGAAACG fliAP2min C for TCTCGAGGATAGAACCCTCTGTAGAAAAGG fliAP2min rev TGGATCCGCGTTAAATGAGTTATCG fliAP2min GCC rev TGGATCCGCGGCCAATGAGTTATCG fliC for GCTGTTGATACGCAGACCG fliC rev GCTATTTCGCCGCCTAAG GSP1 CTGCGTCTTGAGTATTCTT GSP2 TCCATCTTTGCCCTACCGTATA invFent for ACTCGAGGAATGCGTTTCAGGAATG invFent rev AGGATCCGTCTGTATAAACCATGCTTC invFmin for TCTCGAGTGATTGCATCAGGATTTT invFmin C for TCTCGAGTGATTGCATCAGGATTTTGACA invFmin rev TGGATCCAATATGTAAACAATACCG invFmin GCC rev TGGATCCAGCCTGTAAACAATACCG P2only5'A GCGAATTCAAACAAAATGATTGACG pCSFor  TGGCAATTCCGACGTCTAAG  pCSRev  CACTAAATCATCACTTTCGG  PRI1  ACATGAAGGTCATCGATAGCAGGA  PRI3  AACAGTAATGGGCCAATAACACCG  PRI5 PRI6 ST rpoBI sense ST rpoBI antisense ST rpoBII sense  CCAAAATCATTAGGGGATTCATCA CCAACGCTTTTCCCGAGATC GACAGATGGGTCGACTTGTCAGCG AGGTGGTCGATATCATCCACTT TCGAAGGTTCCGGTATCCTGAGC  Source Invitrogen This study This study Invitrogen This study This study This study This study This study This study This study This study This study This study This study This study This study This study This study This study This study This study This study This study This study This study This study M. Strauch, University of Maryland (Tsui et al., 2004) (Tsui et al., 2004) (Nichols et al., 1998) (Nichols et al., 1998) This study This study This study This study This study  30  Oligonucleotide ST rpoBII anti sense RT-16S-F RT-16S-R RT-fliA-F RT-fliA-R RT-invF-F1  Nucleotide Sequence 5’ to 3’ GGGTACATCTCGTCTTCGTTAAC  Source This study  GAACTGTGAGACAGGTGCTG GCAACAAAGGATAAGGGTTG TGTGGCAGCGTTATGTACC GCTTGTAGCAGATCGTCCAG GCAGGATTAGTGGACACGAC  RT-invF-R1  TTTACGATCTTGCCAAATAGCG  RT-STM2901 F2 RT-STM2901 R2 RT-STM3595 F RT-STM3595 R S187 slyAent for slyAent rev spvAent for spvAent rev spvAent2 for spvAmin for  CCGCAGTCATAGCACGATA TTTGCCTGAGTTCTCCTTGA TCATCCTTATACCGCAGGTG GATCATTTCCGCCATCAAC TGCAAGGCGATTAAG TCTCGAGGATCTTGTAAGGGCAATC TGGATCCTCAGAACCTAGTGGCGAT TCTCGAGCTGAAGAAACTGAATAACGG TGGATCCTTCGACCTGTGAAAGTGC TCTCGAGCCTTGTATTACCCGACCA CCTCGAGCGATTCCGCACAGCAGAAAAATAGC AC CGGATCCGGCTTATATTGAGTTTATTT GCTCGAGTGTCCAGATAGAGAAACGG GGGATCCAGAGATGACGCCAGTACAG TCTCGAGCAACAGTGGGAAGTCGCT CCTCGAGCGTTGTGAAAGTATTGCA TGGATCCGTTGACGGTTTGATGATAC CCTCGAGTCAAACCGTCAACTGCGC ACTCGAGCAGGTTGATAATATGGTC AGGATCCCAATAAACTGTACCACTG ACTCGAGATAAAACAGAAGCACTATC TGGATCCGCCTTTTCTTCATTATAT ACTCGAGGAGTGGGTGAAATGTCAG AGGATCCGCCAGAGTGGAACTGACC  This study This study This study This study (Choi et al., 2007) (Choi et al., 2007) This study This study This study This study M. Strauch This study This study This study This study This study This study  spvAmin rev STM3595ent for STM3595ent rev STM3595ent2 for STM3595min for STM3595min rev STM3595PE2 for traSent for traSent rev traSmin for traSmin rev traTent for traTent rev  2.7.  This study This study This study This study This study This study This study This study This study This study This study This study This study  Disk diffusion assays.  Overnight cultures were diluted 1:200 into 5 ml of 0.7% (w/v) agar overlays and poured over LBA plates. Antibiotic disks were placed on top of overlays. Antibiotic disks were premade (BD  31  Biosciences, Rockville, MD) or antibiotic solutions were spotted onto 6 mm filter paper discs (Advantec, Japan). Plates were incubated overnight at 37C. Images of luminescence produced by reporters were taken with a LB980 camera (Berthold Technologies, Oakridge, TN) or a MFChembis (Berthold Technologies).  Figure 2.1. Promoterless-lux reporter plasmid pCS26. The LuxCDE proteins catalyze the conversion of fatty acids into a long chain aldehyde, RCHO, and the LuxAB proteins are the two subunits of the luciferase enzyme. The reaction is as follows (R represents a long carbon chain where 13 is the most efficient): RCHO + FMNH2 + 02 → FMN + RCOOH + H2O + light490 nm (Meighen, 1993). Due to the pSC101 origin of replication, it is likely that the plasmid is maintained at five copies per chromosome (Wadood et al., 1997). No exogenous substrate is needed to produce light.  32  2.8.  RNA isolation and RT-PCR.  Cultures were grown in the presence or absence of rifampicin (5.0 μg ml-1) and treated with RNAprotect solution (Qiagen, Missassauga, ON). RNA was extracted with RNeasy columns (Qiagen), digested with DNase I and purified with RNeasy columns (Qiagen). Extracted RNA was amplified using a one-step SYBR Green Quantitative RT-PCR kit (Sigma, St. Louis, MO) in a MX3000P cycler (Stratagene, La Jolla, CA). The following primers were used for amplification of cDNA: RT-16S-F, RT-16S-R, RT-fliA-F, RT-fliA-R, RT-invF-F1, RT-invF-R1, RT-STM2901 F2 and RT-STM2901 R2. A minimum of three replicate cultures were used to prepare RNA samples and at least two separate amplifications were performed from each RNA sample. Relative quantities were calculated using the relative standard curve method. RNA amounts were normalized to 16S RNA. The effect of rifampicin was calculated using normalized values (RNAno rif/RNArif).  2.9.  Protein expression.  pKH445 was transformed into E. coli BL21(DE3) pLysS (Novagen, Gibbstown, NJ). Overnight cultures of transformants were diluted 1:10, incubated at 37 C, induced with 1 mM IPTG when the culture reached an OD600 of 0.8 and grown for an additional 2 hours before harvesting. pCH112 was transformed into E. coli TOP10 (Invitrogen). Overnight cultures of transformants were diluted 1:100, incubated at 37 C, induced with 0.02% (w/v) arabinose at an OD600 of 0.5 and grown for an additional 16 hrs at 22 C before harvesting.  33  2.10.  Protein purification.  RNAP was isolated from E. coli W3100 (28 g pellet) grown to mid-exponential phase by the University of Alabama at Birmingham Fermentation Facility. RNAP was isolated as described previously (Seredick & Spiegelman, 2007) except active fractions from the DNA-cellulose column were not concentrated. Active fractions were adjusted to a final concentration of 50% glycerol (v/v) and then stored at -80 C for direct use. HilA was purified from TOP10/pCH112 after induction (as described in section 2.9) using a Ni-NTA resin (Qiagen) according to manufacturer’s protocol for isolation of native proteins with an additional wash (Qiagen lysis buffer with 20 mM imidazole). HilA extracts were dialyzed into 1X transcription buffer (10 mM HEPES pH 8, 80 mM potassium acetate, 10 mM magnesium acetate, 0.1 mM DTT, 0.1 mM acetylated BSA) with 20% (v/v) glycerol. After dialysis, glycerol was added to a final concentration of 50% (v/v) and the protein concentration determined using the Bio-Rad protein assay using BSA as a standard (Mississauga, ON). After induction (as described in section 2.9), 28 was purified from BL21(DE3)/pLysS/pKH445 under denaturing conditions (8 M urea) using a Ni-NTA resin (Qiagen) according to the manufacturer’s protocol and reconstituted as described previously (Schaubach & Dombroski, 1999).  2.11.  In vitro transcription assays.  Linear templates for in vitro transcription were made using PCR with the following primer/template combinations: invFent for, pCSrev / pGY17 (invF); P2only5’A, S187 / pAW44 (abrB); flgKent2 for, pCSrev / pGY13 (flgK); and fliC for, fliC rev / S. typhimurium 14028 genomic DNA (fliC). For transcription assays with flgK and fliC, template and 1.5 μl of reconstituted RNAP-28 were incubated in transcription buffer for 2 min at 37 C. Transcription  34  was started by the addition of a mixture containing NTP, rifampicin (varying concentrations) and heparin to limit the transcription to a single round. For transcription assays with invF and abrB, template and HilA (or HilA dialysis buffer, respectively) were incubated for 2 min at 37 C in transcription buffer, RNAP added and incubated for another 2 min at 37 C. Transcriptions were then challenged with a mixture of NTP, rifampicin (varying concentrations) and heparin. See Figure 3.13 for a schematic of incubation times used. All reactions had a final volume of 10 μl and final concentrations were as follows unless otherwise indicated: 1X transcription buffer (10 mM HEPES pH 8, 10 mM magnesium acetate unless otherwise indicated, 80 mM potassium acetate, 0.1 mM DTT, 0.1 mg acetylated BSA ml-1), 16 nM for all templates except abrB (6 nM), 75 nM RNAP, 3 μM 28 (as appropriate), 0.3 μM HilA (as appropriate), 50 μg heparin ml-1, 400 μM CTP, 400 μM UTP, 400 μM ATP, 5 μM GTP and 111 kBq [-32P] GTP or 400 μM GTP, 5 μM ATP and 111 kBq [-32P]ATP, as indicated. Transcripts were elongated for 5 min at 37 C and then terminated by the addition of 5 μl of loading buffer (1.5X transcription buffer with 0.1% (w/v) bromophenol blue, 0.1% (w/v) xylene cyanol and 7 M urea). Prior to use in transcription assays, RNAP-28 was reconstituted by incubating in dilution buffer (10 mM HEPES pH 8, 10 mM magnesium acetate, 10% glycerol (v/v), 0.1 mM DTT, 0.1 mg ml-1 acetylated BSA) on ice for 10 min. In transcription assays with PinvF or PabrB, RNAP was diluted 1:1 with dilution buffer before use. Transcripts were electrophoresed through 8% denaturing acrylamide gels and imaged using a Storage Phospor screen (Amersham Biosciences, Piscataway, NJ), scanned by a Typhoon 9400 (Amersham Biosciences) and quantified using ImageQuant 5.2 software (Amersham Biosciences).  35  2.12.  5’ RACE.  The transcription start of the STM3595 transcript was determined with the 5' RACE System (Invitrogen). RNA isolated from S. typhimurium 14028 carrying pGY37 grown in the presence of rifampicin was used as a template for cDNA synthesis with the gene specific primer, GSP1. The AAP/GSP2 primer pair was used in first round PCR. The AUAP/pCS rev primer pair was used in second round PCR. Two PCR products were visible after second round PCR. DNA was isolated with a Qiaquick Gel Extraction kit (Qiagen) and nucleotide sequenced using pCS rev (Macrogen). Only one amplicon had a sequence related to STM3595; the other amplicon was likely a product of non-specific priming during PCR.  2.13.  Primer extension.  Oligonucleotides STM3595ent rev and traSminF rev were labeled using 22.2 MBq [-32P]ATP and T4 polynucleotide kinase, 3’ phosphatase free (Roche, Indianapolis, IN) according to manufacturer’s instructions for direct phosphorylation. Sequencing ladders for primer extension gels were created using 5’ labelled oligonucleotide in a Sequitherm Excel II kit (Epicentre, Madison, WI) according to manufacturer’s instruction. Templates for the sequencing reactions were created by PCR. PCR products using 5’ labeled oligonucleotides were generated using S. typhimurium 14028 genomic DNA as template with either of the following primer combinations: STM3595ent for/STM3595ent rev or traSmin for/traSent rev. M-MLV reverse transcriptase (Promega, Madison, WI) and 5’ labeled oligonucleotides were used to make radiolabelled cDNA using either RNA isolated from S. typhimurium 14028 grown in the presence of rifampicin or RNA synthesized in vitro. RNA was isolated from culture as done in section 2.8 and RNA synthesized as in section 2.11, but on a larger scale. Products were  36  separated by electrophoresis through 8% denaturing acrylamide gel, dried and visualized by radiography using Bioflex EconoFilm (Interscience, Markham, ON).  2.14  Tn10d mutagenesis and Fluorescence Activated Cell Sorting.  Phage infections and lysates were produced as described in various P22 protocols ( EGTA chelates Ca2+ and prevents reinfection by phage. Infected cells were grown in LB broth or agar supplemented with ampicillin and EGTA (pH 8, 10 mM) unless otherwise indicated.  S. typhimurium 14028 pNK2880 was infected with a P22HT int lysate made using S. typhimurium TT10423. Infected cells were plated on LBA supplemented with EGTA and tetracycline to create ~100,000 colonies/mutants. Colonies from plates were pooled and infected with P22 HT105/1-int creating a transducing phage library lysate. The lysate was used to infect S. typhimurium 14028 containing the STM3595::gfpmut3a reporter, pGY42. Infected cells were plated on LBA supplemented with EGTA, tetracycline and ampicillin to make approximately 40,000 14028/pGY42 Tn10dTet mutants. Colonies were pooled, resuspended (LB broth supplemented with ampicillin, EGTA and 20% glycerol (v/v)) and stored without growth at -80C.  Frozen stocks were diluted 1:20 into fresh media and grown overnight at 37 C. Overnight cultures were diluted 1:30 and supplemented with or without rifampicin (2.5 μg ml-1) to give a final optical density of 0.1 (600 nm, Genesys 20 Spectrophotometer, Thermo Scientific). Cells were grown for three hours and fluorescence at 530 nm (excitation at 488 nm) measured using an  37  Influx cell sorter (BD Biosciences). 107 events (cells and debris) representing a population of approximately ~ 40,000 Tn10d mutants were sorted. Particles with fluorescence intensities in the bottom 1% of the population were collected into 3 ml of diluted media (50% PBS and 50% media). 1 ml of the sorted particles was resorted. Particles with fluorescence intensities in the bottom 5% of the 3 x 104 particles sorted were collected in 300 μl of media and spread onto agar plates. After overnight growth, 150 colonies representing potential non-fluorescing mutants were visible. Colonies were restreaked twice on agar plates, single colonies were used to inoculate 150 μl broth in 96-well plates. A 96-well pin replicator (V&P Scientific) was used to stamp colonies into assay plates containing LB broth supplemented with ampillicin and rifampicin. These colonies were further analyzed for fluorescence using the Victor II plate reader (excitation at 485 nm, emission at 535 nm). Three of the 150 mutants analyzed by the Victor II plate reader had the desired phenotype in the presence of rifampicin, low expression of the STM3595::gfp reporter. Two different mutants were identified by nucleotide sequencing of the inverse PCR products (Figure 2.2). The two mutants were located in pcnB and yciK, respectively (Figure 3.24).  2.15  Inverse PCR methods.  Inverse PCR products were obtained from the Tn10d strains as described previously (Nichols et al., 1998; Ochman et al., 1988) with a few modifications: approximately 5 μg of genomic DNA were digested overnight at 37 C with 25 U of HpaII (Roche, Laval, Quebec) or RsaI (New England Biolabs, Pickering, Ontario) according to manufacturer’s instructions and inactivated by heating for 20 minutes at 70 C. For circularization, 500 ng of the digested DNA was ligated using 2.5 U of T4 DNA Ligase (Invitrogen) in a total volume of 25 μl for 16 hrs at 12 C as per  38  manufacturer’s instruction. Ligation mix (2.5 μl) was used as template in a 25 μl PCR reaction with the primers pri1 and pri6. PCR product from the pri1/pri2 PCR reaction (2.5 μl) was used as template in a 25 μl PCR reaction with primers pri3 and pri5. Figure 2.2 shows the restriction sites and primer binding sites within the Tn10d sequence. The nucleotide sequence of the final PCR product was determined using pri3 or pri5.  Figure 2.2. Schematic of Tn10dtet and inverse PCR relevant primers and restriction sites. Numbers in parentheses indicate relative nucleotide positions of primer binding sites (pri1, pri3, pri5 and pri6) and restriction endonuclease (HpaII and RsaI) recognition sequences. After digestion, DNA from restriction at a site 3’ to the transposon, located in the genomic DNA, is ligated to DNA resulting from restriction at RsaI(1882) or HpaII(2083). The resulting piece of circular DNA is used as a template in PCR.  39  Chapter 3. Results. 3.1.  In vivo characterization of transcription modulation by rifampicin.  Growth inhibition by rifampicin is caused by an inability of bacterial RNAP to produce full length RNA transcripts (Floss & Yu, 2005). The lowest concentration at which a given antibiotic inhibits growth of a given strain under the given conditions is the MIC. Using promoter::lux fusions, at concentrations below the MIC (sub-MIC) rifampicin both up-modulates and down-modulates transcription of various promoters (Table 1.1). Six promoters, three downmodulated promoters (fliA, flgK and invF) and three up-modulated promoters (spvA, traS and STM3595) are studied in this thesis.  3.1.1. RT-PCR. In addition to promoter::lux fusion assays, RT-PCR carried out on RNA harvested from rifampicin treated S. typhimurium 14028 was used as secondary method to test gene expression in response to rifampicin. Relative amounts of transcript in the presence and absence of rifampicin was determined for two rifampicin down-modulated promoters, fliA and invF, and two rifampicin up-modulated promoters, STM3595 and traS (Fig 3.1). Primers for 16S and STM2901 annealed to the 3’ end of the respective genes. Primers for invF and fliA annealed to the 5’ end of the coding region. Primers for STM3595 were in the center of the protein coding region. 16S rRNA was used to normalize for input RNA as it was presumably unaffected by sub-MIC rifampicin. Since expression of STM2901 in 14028/pGY21 grown in liquid culture was found to be unaffected by rifampicin (Table 1.1, Fig 3.8), STM2901 transcripts were used as a negative control.  40  Figure 3.1. Fold decreases of fliA and invF and fold increases of STM3595 and traS transcripts in response to sub-MIC rifampicin. Relative RT-PCR levels of transcripts identified as rifampicin responsive in reporter assays. RNA levels were measured in the presence and absence of rifampicin. After normalizing levels using 16S RNA levels, ratios were obtained to determine fold change. Error bars indicate standard deviation of at least three experiments.  In terms of directionality and magnitude, gene expression changes measured by lux reporters and RT-PCR on RNA from rifampicin treated cultures showed similar responses to rifampicin. Using RT-PCR, fliA was decreased approximately 27-fold in the rifampicin treated sample (Fig 3.1) compared to a 32-fold decrease in luminescence using the lux reporter assay (Table 1.1). The other rifampicin down-modulated transcript invF was 130-fold less abundant in the rifampicin treated sample (Fig 3.1) compared to a 200-fold decrease found using the lux reporter assay (Table 1.1). The traS transcript was 5 times more abundant in the rifampicin treated sample (Fig 3.1) compared to a 5-fold increase in luminescence (Table 1.1). The STM3595 transcript had the most disparate results when comparing the luminescence assay to RT-PCR but was up-modulated by rifampicin using both methods. This transcript was found to be 4-fold 41  more abundant in the rifampicin treated sample (Fig 3.1) compared to a 24-fold up-modulation (Table 1.1). RT-PCR analyses of transcripts from cells treated with and without rifampicin were comparable to the magnitude of luminescence changes obtained from analogous reporter strains grown in the presence or absence of rifampicin.  3.1.2. Measurement of rifampicin mediated transcription modulation in rifampicin resistant strains. It was theoretically possible that transcription modulation by rifampicin was due to interaction of rifampicin with one or more targets other than RNAP. To test whether transcription modulation by rifampicin and rifampicin mediated growth inhibition were both mediated by interaction with RNAP, a rifampicin resistant mutant was examined for transcription modulation by rifampicin at sub-MIC rifampicin. A spontaneous resistant mutant of S. typhimurium 14028, strain R306, was isolated by plating cells on media containing 300 μg ml-1 of rifampicin. By nucleotide sequencing, resistance in R306 was found to be conferred by a H526Y mutation (E. coli numbering) in the  subunit of RNAP. This mutation is in rifampicin resistance cluster I of the rpoB gene as named by Jin et. al. (1988).  Two rifampicin down-modulated promoters (invF and fliA), two rifampicin up-modulated promoters (STM3595 and traS) and a promoter unaffected by rifampicin when grown in liquid media (STM2901) were examined in a rifampicin resistant mutant. The basal level of expression from the rifampicin down-modulated promoters in R306 was within the error limits of repressed expression from the wild type exposed to rifampicin (not shown). The basal level of expression of rifampicin up-modulated promoters in the mutant was similar to the untreated levels of expression from the wild type (not shown). Transcription modulation of the four promoters 42  tested (STM3595, traS, invF and fliA) in response to rifampicin was abolished or greatly reduced in R306 (Fig 3.2). These data suggested that transcription modulation by rifampicin was mediated by rifampicin acting directly on RNAP rather than on secondary target.  Figure 3.2. Comparison of luminescence produced by reporters carried by wildtype and rifampicin resistant strains. Strains were grown in 2.5 μg ml-1 (grey) or 5.0 μg ml-1 (white) of rifampicin. Fold activation was calculated by dividing the level of luminescence of rifampicin treated cells by the luminescence of cells grown in the absence of rifampicin and vice versa for fold inhibition. A. From left to right, reporter plasmids are pGY21, pGY35 and pGY22. B. From left to right, reporter plasmids are pGY21, pGY16 and pGY7.  3.1.3. Disk diffusion assays. Sub-MIC antibiotics are known to have a plethora of effects on both bacterial transcription and translation. These effects might act through a common mechanism or through mechanisms unique to the antibiotic class. To examine whether other antibiotics had similar effects of transcription modulation and thus perhaps a similar mechanism of transcription modulation as 43  rifampicin, the response of strains carrying seven lux reporter fusions to various antibiotics and oxidative stressors was observed. Three rifampicin down-modulated promoters: fliA, flgK and invF (Fig 3.3), three rifampicin up-modulated promoters: traS, spvA and STM3595 (Fig 3.4) and one control promoter which is unaffected by rifampicin in liquid media (STM2901) were examined. The rifampicin resistant strain, R306, was also used as a host strain for these experiments.  Figure 3.3. Disk diffusion assays with fliA, flgK, invF and STM2901 lux reporter strains. In liquid media fliA, flgK and invF are down-modulated by rifampicin. Luminescence images of fliA (pGY8, i), flgK (pGY13, ii), invF (pGY17, iii) and STM2901 (pGY21, iv) promoter fusions in S. typhimurium 14028 (top row) or its rifampicin resistant derivative R306 (bottom row). Paper discs containing rifampicin (R, 30 μg), cumene peroxide (C, 5 μl of 1%), hydrogen peroxide (H, 5 μl of 3%), novobiocin (N, 30 μg), erythromycin (E, 300 μg) or tetracycline (T, 30 μg) are placed on top on the overlay. Luminescence intensity is converted to the color scale on the right, yellow being strong luminescence and dark blue being low luminescence.  44  The assay used to examine the expression in reporter strains was a combination disk diffusion and luminescence assay. Filter paper disks containing the compound of interest were placed on a bacterial agar overlay and cells were grown overnight. Compounds in the disks diffuse through the agar and as a result are concentrated near the disk and become more dilute farther away from the disk. Growth sensitivities to various compounds and antibiotics can be observed as zones of clearing where there is no growth (dark) and as bacterial lawns (light). The MIC of that particular compound occurs at the border of growth and no growth. For each antimicrobial shown, the diameter of the zone of clearing (zone of inhibition) was very similar between reporter strains; one representative plate is shown (Fig 3.4.iv). Luminescence images of the agar plates were recorded to monitor light production (Fig 3.3, Fig 3.4).  In liquid media, strains with fliA, flgK and invF reporters show rifampicin mediated downmodulation. In plate assays, these reporters showed no response to rifampicin (Fig. 3.3). However, it is difficult to detect down-modulation on solid media unless there is high basal luminescence. Thus, it is possible that rifampicin mediated down-modulation also occurs on solid media. Similar to the responses seen in liquid media, rifampicin induced a luminescence response from the traS, spvA and STM3595 (Figure 3.4). Interestingly, although unresponsive to rifampicin in liquid media, STM2901 was induced by rifampicin on solid media (Figure 3.4).  Promoters had varying response ranges to different antimicrobials. Outside the zone of growth inhibition, strains with fliA, flgK, invF, traS and spvA lux reporter fusions showed strong induction by tetracycline (Fig. 3.3, 3.4), a protein synthesis inhibitor that binds to the 30S ribosomal subunit. The fliA reporter strain was also weakly induced by erythromycin, a  45  macrolide which binds to the peptide exit tunnel of the ribosome, and by hydrogen peroxide but not to cumene peroxide (an organic peroxide) or novobiocin, a gyrase inhibitor (Fig. 3.3). The invF and STM3595 promoters were only induced by one compound tested, tetracycline and rifampicin, respectively, while the fliA promoter was induced by several compounds.  As expected, rifampicin produced a zone of inhibition in the overlay containing the wild type strain (Fig 3.4.iv, upper) but no zone of inhibition when R306 was used, as R306 is rifampicin resistant (Fig 3.4.iv, lower). For all other antibiotics, zones of inhibition were similar in 14028 and R306 regardless of the promoter-lux fusion used (Fig 3.4.iv).  Figure 3.4. Disk diffusion assays with traS, spvA and STM3595 lux reporter strains. Luminescence images of traS (pGY25, i), spvA (pGY30, ii), STM3595 (pGY37, iii) and a photograph of STM2901 (pGY21, iv) promoter fusions in S. typhimurium 14028 (top row) or its rifampicin resistant derivative R306 (bottom row). Compounds are labeled as in Fig 3.3.  46  R306 carrying the same reporters as the wild type strain was examined for responses to various compounds and stress induction (Fig.3.4). If a common mechanism was the basis of transcription modulation by tetracycline, erythromycin, hydrogen peroxide and rifampicin, rifampicin resistance should change the responses to the other compounds tested. Expression in the wildtype strain carrying the traS, spvA and STM3595 reporters showed strong up-modulation in response to rifampicin at ~9 mm away from the centre of the disk (Fig 3.4, upper row) but not close to the disk. At the same concentration range (~9 mm away from the centre of the disk), the response was abolished in the rifampicin resistant strains (Fig 3.4, lower row); although fliA and flgK reporter strains responded to rifampicin, these responses occurred at relatively high concentrations of rifampicin (at the center of the disk). Luminescence responses to other antibiotics and chemicals were largely unaffected by rifampicin resistance in the host (Fig 3.4). Rifampicin resistance only affected the response of rifampicin, supporting the notion that rifampicin has a different mechanism of transcription modulation compared to other active compounds tested.  3.1.4. Time courses. To observe when transcription modulation by rifampicin occurs in the growth cycle of a batch culture, growth and luminescence of reporter cultures were monitored after stationary phase cultures were diluted into fresh media. For these experiments, rifampicin was added at different times post-inoculation (Fig 3.5). One promoter fusion, STM2901, that is affected less than twofold by rifampicin, is shown as a control (Fig 3.5F). All reporter strains grew similarly, and the growth of one reporter is shown (Fig 3.5G). For the rifampicin down-modulated reporter, fliA,  47  the basal luminescence increased in the early exponential phase of growth, began to plateau during mid-exponential and declined in late exponential (Fig. 3.5A). For the rifampicin downmodulated reporter, flgK, the basal luminescence began to increase during mid-exponential and continued to increase for the duration of the assay (Fig. 3.5B). For the rifampicin downmodulated reporter, invF, the basal luminescence at the time of inoculation was relatively high, declined in the first hour of early exponential, began to increase in mid-exponential and continued to increase for the duration of the assay (Fig 3.5C). When rifampicin was added to cultures, the luminescence of down-modulated promoters showed similar patterns. When rifampicin was added before basal expression increased, rifampicin prevented the increase in expression levels (Fig 3.5A-C). If rifampicin was added after gene expression had already begun to increase, 2 hrs post inoculation for the fliA culture (Fig 3.5A), expression decreased. This suggested that for down-modulated promoters, rifampicin prevented expression around the time of addition.  In the case of the two rifampicin up-modulated promoters, traS and STM3595, the level of induced expression was higher when rifampicin was added earlier (Fig 3.5D, E). Furthermore, for the rifampicin up-modulated promoters, differential expression between treated and untreated cultures could not be seen until two to three hours after rifampicin addition. This type of lag was not observed in assays with the rifampicin down-modulated promoters (Fig. 3.5A-C). This lag suggested that induction of these promoters by rifampicin might require the accumulation of an unknown small molecule, protein or modification. This is addressed in the next section.  48  Figure 3.5. Time courses of rifampicin activation and inhibition. Luminescence (A-F) or growth (G) of strains carrying fliA (pGY7, A), flgK (pGY13, B), invF (pGY17, C,G), traS (pGY25, D), STM3595 (pGY37, E) or STM2901 (pGY21, F) lux reporters grown in the absence of rifampicin (diamond), or rifampicin added zero (square), one (triangle) or two hours (x) postinoculation. Error bars indicate standard deviation of at least three experiments.  3.1.5. Pretreated cells and spent media experiments. To determine whether accumulation of an unknown factor, either intracellularly or extracellularly, was involved with rifampicin mediated up-modulation, a series of rifampicin pretreatment and spent medium experiments were conducted. Luminescence and growth of the STM3595 reporter strain was examined as depicted in Figure 3.6. Untreated cells or rifampicin pretreated cells (Fig 3.6A, 3.7A) were grown in spent media or fresh media (Fig 3.6B-C, 3.7B-C). If a secreted, extracellular factor were involved in rifampicin mediated up-modulation,  49  one might expect to see stronger induction or induction occurring with a short lag period when untreated cells are grown in spent medium compared to fresh medium.  Figure 3.6. A schematic representation of spent medium experiments.  Luminescence of rifampicin pretreated cells grown in fresh medium (Fig 3.6B, 3.7B) had a higher starting luminescence but had a similar level of induction compared to control pretreated cells (Fig 3.7B). This higher starting luminescence was stable and may represent the accumulation of a factor or a modification of cellular components. Fresh cells grown in rifampicin spent media (rifampicin not removed) and fresh cells grown in control spent media supplemented with rifampicin were up-modulated similarly (Fig. 3.6C, 3.7C). Since growing 50  cells in spent media did not shorten the time of induction or amplify the fold increase, the data suggested that rifampicin mediated up-modulation does not involve a stable secreted factor. However, the data does not rule out accumulation of an intracellular factor.  Figure 3.7. Spent media and pretreatment effects of rifampicin on STM3595 expression. A. Luminescence of 14028/pGY37 grown in the absence (diamond) or the presence of rifampicin (square). B. Luminescence of cells grown in the absence (diamond, square) or presence (triangle, x) of rifampicin resuspended and grown in fresh media without (diamond, triangle) or with rifampicin (square, x). Time of resuspension is shown as zero hours. C. Luminescence of fresh cells grown in spent media from cultures grown in the absence of rifampicin (diamond), presence of rifampicin (triangle), or spent media from cultures grown in the absence of rifampicin but supplemented with rifampicin when fresh cells were added (square). Time of reinoculation is shown as zero hours. D. Growth of cultures shown in B (open symbols) and in C (closed symbols). See Fig. 3.6 for a schematic with corresponding lettering.  3.2.  Mapping DNA regions necessary for transcription modulation by rifampicin.  To more precisely identify the DNA fragments necessary for transcription modulation by rifampicin, six rifampicin responsive lux reporters were mapped by sub-cloning. DNA from 51  library plasmids was sub-cloned into pCS26 and luminescence of 14028 carrying these new constructs was measured. The three rifampicin repressed constructs chosen for examination contained promoters for the following genes: fliA, flgK and invF. The three rifampicin activated constructs chosen for examination contained the promoters for the following genes: STM3595, spvRAB and traS. These constructs were chosen as they displayed some of the strongest responses to rifampicin and were possibly important from a pathogenic perspective (Table 1.1). Several constructs had two fragments from different locations in the chromosome (Fig 3.8A, 3.9A) or had several genes with putative promoters.  In order to identify which fragment was involved in transcription modulation by rifampicin, subclones with approximately 300 bp (from 250 bp upstream to 50 bp downstream of the putative start codon) were constructed. Once the putative promoter was identified, smaller (~50 bp) fragments were cloned with the logic that if rifampicin acted directly on RNAP-promoter interactions alone, a fragment of this size should have a promoter sensitive to rifampicin. These fragments spanned nucleotides from approximately –45 to +5 relative to the published or determined +1 transcription start and would likely not contain complete binding sites for transcription factors. They should contain the -35 elements, -10 elements and +1 sequences and sufficient sequence to make specific contacts with RNAP. For each promoter, four to six reporter constructs which were derivatives of the 300 bp and 50 bp fragments were constructed. Luminescence of the strains carrying the variant constructs was monitored in the presence or absence of rifampicin. The data for these experiments is summarized in Figures 3.8 and 3.9.  52  Figure 3.8. Mapping of rifampicin repressed promoters. Schematics of fliA (A), flgK (B) and invF (C) promoter region truncations and the luminescence of the strains carrying the respective promoter-lux reporter fusion. Open circles indicate 70 dependent +1 transcription start sites while closed circles indicate 28 dependent +1 transcription start sites (Ikebe et al., 1999; Kutsukake & Ide, 1995; Lostroh et al., 2000). pGY11 contains the 70 dependent +1 but lacks the corresponding -35 element. Double lines indicate DNA from another region of the genome as compared to single lines in the same construct. The striped box indicates that all bases of the HilA binding site are present (Lostroh et al., 2000).  The original reporter construct from the library with the fliA promoter (pGY7) had two pieces of DNA from different regions of the chromosome. To identify the region of DNA required for rifampicin mediated down-modulation, a 377 bp fragment carrying the presumed fliA promoter region was cloned into pCS26 to create plasmid pGY8. pGY9, a sub-clone of pGY8 was created by deleting 143 bp 3’ from the 70 and 28 dependent promoters identified for the fliA gene 53  (Ikebe et al., 1999). Strains carrying pGY8 and pGY9 had similar levels of lux expression and rifampicin hypersensitivity as the strain with the original library construct, pGY7 (Fig. 3.8A). The other 336 bp fragment had no detectable promoter activity. A strain carrying plasmid pGY10 that contained a 50 bp fragment with only the 70dependent fliA transcription start site and 46 bp upstream of this site showed no lux activity (Fig. 3.8A). In contrast, lux expression in a strain carrying a 51 bp fragment containing the 28dependent transcription start site, pGY11, showed hypersensitivity to rifampicin, albeit with lower basal levels of activity than the larger fragments (Fig 3.8A). This suggested that in the case of the fliA reporter, rifampicin mediated down-modulation is associated with transcription by the RNAP holoenzyme containing 28.  A 632 bp fragment with the flgK 28dependent promoter (Kutsukake & Ide, 1995) was sub-cloned into pCS26 creating plasmid pGY13. Lux expression in the strain carrying pGY13 showed higher fold induction to rifampicin than a strain with the original construct (pGY12) but also showed much lower basal levels (Fig 3.8B). Removal of 310 bp of DNA 5’ to the transcription start site in pGY13, leaving 287 bp 5’ to the start in pGY14 did not change basal expression but decreased hypersensitivity to rifampicin (Fig. 3.8B). Additional 3’ and 5’ truncations of the flgK promoter yielded a 51 bp fragment containing the 28dependent transcription start site and 45 bp 5’ to that site (pGY15). Rifampicin repressed lux expression in the strain carrying was to a similar extent (fold change) as seen in the strain carrying the larger promoter fragment (pGY13, Fig. 3.8B).  A 352 bp fragment with the 70dependent promoter of invF was sub-cloned from the library construct (pGY16) to create plasmid pGY17. pGY17 showed similar basal expression levels and 54  fold hypersensitivity to rifampicin as the original library construct (Fig. 3.8C). 106 bp were deleted from pGY17, leaving 3 bp downstream of the invF +1 and creating plasmid pGY18. Transcription in strains carrying pGY18 retained rifampicin hypersensitivity but showed lower basal levels (Fig. 3.8C). A 198 bp truncation 5’ to the invF transcription start site fragment was made which removed half of the HilA binding site (Lostroh et al., 2000) (pGY19). The strain carrying pGY19 showed low basal lux expression and low rifampicin hypersensitivity (Fig. 3.8C) suggesting that the HilA binding site was either required for rifampicin mediated downmodulation of invF transcription or just required to increase basal luminescence so that downmodulation is detectable in this assay. Removal of 106 bp 3’ to the transcription start leaving DNA from -45 to +3 relative to the HilA dependent transcription start site fused to the lux genes (pGY20) had no basal activity (Fig. 3.8C).  In summary, for both flagellar genes fliA and flgK, transcription from ~ 50 bp fragments containing the transcription start sites of the 28dependent promoters exhibited rifampicin mediated down-modulation. The smallest invF promoter fragment that retained hypersensitivity to rifampicin was 246 bp and contained the binding site for the activator HilA (Lostroh et al., 2000).  55  Figure 3.9. Mapping of rifampicin activated promoters. Schematics of traS (A), spvA (B) and STM3595 (C) promoter region truncations and the luminescence of the strains carrying the respective promoter-lux reporter fusion. Open circles indicate 70 dependent +1 transcription start sites for traS (Ham et al., 1989) and STM3595 (primer extension and 5’ RACE using RNA harvested from cells), closed circles indicate S dependent +1 transcription start sites and boxes indicate SpvR binding sites (Sheehan & Dorman, 1998).  The original library reporter construct pGY22 contained both traGST and slyA. Six sub-clones of pGY22 were constructed and assayed in S. typhimurium 14028 for their luminescence responses to rifampicin to identify DNA regions required for rifampicin mediated up-modulation (Fig 3.9A). A 266 bp fragment carrying the slyA region, a 265 bp fragment region carrying the traT region and a 316 bp fragment carrying the traS region were each cloned into pCS26 to  56  create plasmids pGY23, pGY24 and pGY25, respectively. It is unlikely that the promoters associated with slyA and traT are up-modulated by rifampicin since the strains carrying the respective reporters had low basal activity and were not up-modulated by rifampicin. The traS reporter (pGY25) had moderate basal activity and displayed rifampicin mediated up-modulation. A 3’ truncation of this fragment deleting 155 bp and leaving 8 bp downstream of the transcription start site (pGY26) decreased basal activity but still displayed rifampicin mediated up-modulation. Removal of an additional 112 bp from the 5’ end leaving 41 bp upstream of the start site (pGY28), decreased basal activity and abolished rifampicin mediated up-modulation. Interestingly, the reporter created by a removal of 112 bp from this fragment in pGY25, pGY27, had reduced basal activity but was able to mediate up-modulation by rifampicin. This suggested that in the case of the traS promoter, up-modulation can be mediated by interaction with either DNA downstream or upstream of the transcription start site.  To analyze the regions upstream of the +1 of spvA that were required for rifampicin mediated upmodulation (Fig 3.9B), reporter strains carrying 642 bp (pGY30) and 350 bp (pGY31) were tested. These clones that have 506 bp and 208 bp 5’ to the +1 site, respectively, had similar basal activity and their level of rifampicin mediated up-modulation was similar to the original library reporter strain (pGY29). A strain carrying a construct made by truncating the 3’ end of the promoter in pGY31 to the+5 nucleotide (pGY32) had lower basal activity but a similar fold increase in rifampicin mediated up-modulation when compared to reporters carrying longer DNA fragments. Deletion of 163 bp 5’ truncation from the promoter in pGY31, leaving 45 bp 5’ to the +1 site (pGY33) had no basal activity and did not display rifampicin mediated upmodulation. Similarly, a 50 bp fragment containing -45 to +5 of the spvA promoter (pGY34) had  57  no basal activity and did not display rifampicin mediated up-modulation. The reporter with the shortest fragment displaying rifampicin mediated up-modulation, pGY32, had sequences from -208 to +5 and encodes the binding site of a known spvA activator, SpvR (Sheehan & Dorman, 1998).  STM3595  cgttgTgaAagtattgcaagacgtgttcgTATcATcaaaccGTcaac -35 -10 +1  E.coli 70 consensus  TTGACA  TATAAT  Figure 3.10. Transcription start site of STM3595 as determined by 5’RACE. The transcription start site is marked by GT and a +1 underneath. Sequences with similarity to the accepted E.coli consensus (Browning & Busby, 2004) are underlined.  Before studying the sequences required for rifampicin mediated up-modulation of the gene STM3595, the transcription start site was determined by 5’ RACE (Fig 3.10). The transcription start sites for all other promoters shown in Fig 3.8 and 3.9 had been previously determined in S. typhimurium except traS which had been determined in E. coli (Ham et al., 1989) and was assumed to be conserved in S. typhimurium. The transcription start site of STM3595 determined by 5’RACE using RNA isolated from S. typhimurium 14028 grown in the presence of rifampicin is indicated in Fig. 3.10.  Deletion analysis was used to narrow down sequences required for the rifampicin up-modulation in the STM3595 promoter (Fig 3.9C). Reporter strains carrying 626 bp (-512 to +114, pGY36) 58  and 366 bp (-252 to +114, pGY37) fragments had lower basal activities than the strain carrying the original library reporter (pGY35) and were up-modulated to similar absolute levels as pGY35. Thus the strains with pGY36 and pGY37 displayed higher levels of rifampicin mediated up-modulation than 14028/pGY35. Truncation of 211 bp 5’ to the +1 site of pGY37 (pGY38) had even lower basal activity and higher fold levels of rifampicin mediated upmodulation when compared to reporters carrying longer DNA fragments. A 108 bp 3’ truncation of the 366 bp fragment produced a reporter carrying -252 to +6 (pGY39) had low levels of basal activity and did not display rifampicin mediated up-modulation. A 52 bp fragment carrying -41 to +6 (pGY40) had no basal activity and did not display rifampicin mediated up-modulation. Since the smallest fragment of DNA mediating rifampicin mediated up-modulation spanned nucleotide sequences from -41 to +114 with respect to the +1 transcription start site, DNA downstream of +1 seemed to be necessary for rifampicin mediated up-modulation from the STM3595 promoter. Since examples of both upstream and downstream regions were seen to be required for rifampicin mediated up-modulation of different promoters, rifampicin responsive promoters may be indirectly affected by rifampicin and may work by multiple mechanisms.  3.3.  In vitro transcription of rifampicin modulated promoters.  In vivo data from RT-PCR and lux reporters showed that transcription of spvA, traS and STM3595 were up-modulated by sub-MIC rifampicin and that fliA, flgK and invF were downmodulated by sub-MIC rifampicin. Disk diffusion assays suggested that rifampicin elicited a pattern of expression different than that of other compounds tested. Deletion studies using truncated promoters fused to the lux reporter suggested that down-modulation of fliA and flgK was associated with 28, while down-modulation of invF may require the activator HilA.  59  Furthermore, since RNAP rpoB mutants no longer displayed transcription modulation by rifampicin (section 3.1.2), it was possible that RNAP holoenzyme, DNA and rifampicin alone were enough to mediate transcription modulation by rifampicin. That is no other regulatory elements such as transcription factors or small molecules were required to mediate this effect at some promoters. To investigate this, in vitro transcription experiments were performed with varying amounts of rifampicin, DNA and E. coli RNAP. Since E. coli and S. typhimurium RNAP are 99% identical in amino acid sequence, I presumed that the in vitro responses were equivalent. Large quantities of E. coli cells from which a mixture of RNAP holoenzymes could be isolated were also readily available. Other in vitro experiments studying S. typhimurium transcription have used E. coli RNAP (Schaubach & Dombroski, 1999). Linear DNA templates were prepared using PCR. Two rifampicin down-modulated S. typhimurium genes, invF and flgK, and two rifampicin up-modulated genes, STM3595 and traS, were examined.  3.3.1. Rifampicin mediated down-modulation requires 28 and HilA. For transcription at flgK and invF, preliminary experiments examining transcription by RNAP-70 using different buffer conditions and incubation times failed to produce strong transcripts (not shown). One 28 dependent transcription start site for S. typhimurium flgK transcription had been previously determined (Kutsukake & Ide, 1995). Deletion analysis (section 3.2) suggested that for flgK, rifampicin mediated down-modulation was associated with the previously determined RNAP- 28 transcription start site. In the case of invF, deletion analysis suggested that RNAP-70 in conjunction with the transcription activator HilA was required to observe transcription modulation by rifampicin from invF because removal of the HilA binding site eliminated rifampicin mediated down-modulation. As a consequence, 60  recombinant His-tagged versions of 28 and HilA were purified and added to in vitro transcription reactions for these promoters as appropriate.  Figure 3.11. Induction and isolation of His-HilA and His-Sigma28. A. SDS-PAGE of insoluble (i) and soluble (s) fractions of His-HilA from cultures grown in various induction conditions: 0.2% or 0.02% arabinose, growth at various temperatures or different induction times. B. SDS-PAGE of whole cell lysates from uninduced (u) or induced (i) cultures of BL21/pLys or BL21 star carrying pKH445 induced for 16 hrs at 22°C or 2 hrs at 37°C. C. SDSPAGE of purified extracts of RNAP, His-Sigma28 or His-HilA. BioRad prestained SDS-PAGE standard protein ladder was run on each gel (l).  61  Since two different conditions for isolation of soluble His-HilA had been reported (Lostroh et al., 2000; Rodriguez et al., 2002), several parameters were optimized to obtain the best yield of soluble protein (Fig 3.11). Two different arabinose concentrations were compared for their ability to induce His-HilA from the arabinose inducible promoter PBAD (Fig 3.11A). Four different post-induction incubation temperatures (15 °C, 22 °C, 30 °C and 37 °C) and three postinduction incubation times (2, 6 and 16 hrs) were also compared (Fig 3.11A). No conditions were found under which His-HilA was mostly soluble. Induction with lower arabinose, seemed to yield more protein. For isolation, expression was induced with 0.02 % arabinose and the culture incubated for 6 hrs at 30 °C or 16 hours at 22 °C (Fig. 3.11C).  As previously published, no conditions resulting in appreciable amounts of soluble His-28 could be identified. 28 could not be induced in BL21(DE3), as previously published (Chadsey et al., 1998), or in BL21 star (Invitrogen), but was induced in BL21(DE3) pLysS (Fig. 3.11B). If induced, more growth generally resulted in more total protein induction and more insoluble protein. His-28 induced for 2 hours at 37 °C was isolated under denaturing conditions and reconstituted as described elsewhere (Wilson & Dombroski, 1997). The resulting protein preparations appeared pure (Fig. 3.11C).  3.3.2. Optimization of flgK and invF in vitro transcription. Several in vitro transcription reaction conditions were optimized: protein concentration, template concentration, activation time, extension time and order of addition. High levels of transcription from the flgK template required the presence of 28 (Fig 3.12A). High levels of transcriptions were detected when the ratio of 28:RNAP was equal to 39:1 and stayed more or less the same 62  with higher ratios (Fig 3.12A). When the ratio of 28:RNAP was reduced, transcription weakened (Fig 3.12B, first 3 lanes), suggesting renaturation of 28 after isolation under denaturing conditions was approximately 2.5% effective. A ratio of 39:1 was used for subsequent experiments. Varying the level of flgK template showed the highest transcript production at 16 nM (Fig 3.12B); 16 nM template was used for subsequent experiments. Transcription from the invF template was stimulated by HilA and appeared optimal when HilA was 0.3 to 0.2 nM (Fig 3.12C). When invF template concentration was varied, HilA dependent invF transcript was optimal at 16 nM of template (Fig 3.12C).  Figure 3.12. Optimization of flgK and invF transcription. A,B. Phosphorimage of denaturing polyacrylamide gels used to separate transcripts from reactions with various amounts of HisSigma28 (M) or flgK template (nM). C. Phosphorimage of transcripts from reactions with various amounts of His-HilA (M) or invF template (nM). Arrows indicate specific transcripts.  Even after optimizing template and HilA amounts, transcript levels from invF were still low. Various incubation times and orders of addition were investigated to see if higher level of  63  transcription could be obtained. Activation of transcription by HilA was found to be stronger when HilA was added to DNA before RNAP (not shown). DNA-HilA preincubation times of 2 and 16 minute were examined; a 2 minute preincubation yielded higher levels of transcription (not shown). Extension times of 2, 5 and 22 min were examined and 5 min extension times were found to give the highest level of transcription with the least amount of non-specific transcription (not shown). Since transcription from the PflgK was efficient, transcription conditions were assumed to be sufficiently optimized. Incubation times used in subsequent experiments are outlined in Figure 3.13 unless otherwise indicated.  Figure 3.13. A schematic representation of incubation times used for in vitro transcription.  Heparin was added to limit transcription to a single round. Heparin is a polyanion which inactivates free RNAP and destabilizes some early binary complexes of RNAP and DNA. At most promoters, E. coli RNAP: 70 which has formed an initiation complex in open complex is resistant to heparin, one exception being complexes formed at the T7 A1 promoter (Pfeffer et al.,  64  1977). In assays shown in this thesis, heparin and NTP were added simultaneously after (E. coli) RNAP binding, so it is unlikely that the heparin affected open complex formation or stability.  Figure 3.14. Identification of transcripts. Phosphorimage of transcripts from flgK (F) and invF (I) templates with different 3’ end points separated by denaturing polyacrylamide gel electrophoresis. Templates with extended 3’ ends are indicated (ext). The abrB transcript (A) is shown as a reference. Specific transcripts are indicated by arrows. (Transcripts from the short invF template are shown in duplicate.)  In gels showing multiple transcripts, transcripts of interest were identified by making DNA templates of varying lengths by PCR, since DNA templates with a truncated 3’ end would lead to shorter transcripts. A shortened invF template was constructed using invFent for/invFent rev primers on pGY17 plasmid DNA and a shortened flgK template was synthesized using flgKent2 for/flgKent rev on genomic DNA. These templates were 101 bp shorter than the original templates. Transcripts from the shortened templates shifted in the expected manner and were the appropriate relative size when compared to transcripts of known length, such as the PabrB  65  transcript from the p2only5’A/S187 on pAW44 template which gives a transcript of 206 nucleotides (Fig 3.14).  3.3.3. Rifampicin hypersensitivity of flgK and invF transcription. Using optimized conditions, in vitro transcription reactions containing PflgK or PinvF , purified His-tagged 28 or HilA, respectively, were performed varying the concentration of rifampicin. Rifampicin sensitivity of transcription was compared to transcription from a control promoter. A promoter from B. subtilis that has -10 and -35 elements similar to the E. coli 70 consensus sequence, PabrB, was used as a 70 control promoter (Fig. 3.9, 3.29). PfliC was used as a control promoter transcribed by RNAP-28. Transcription from PflgK and PinvF was 3-fold and 10-fold more sensitive to inhibition by rifampicin than their respective control promoters (Fig 3.15, 3.16). These data suggest that the differences in the promoter sequences lead to differences in the inactivation of RNAP by rifampicin and suggested a mechanism by which promoters are down-modulated by rifampicin.  Since PabrB is a B. subtilis promoter and its response to rifampicin in S. typhimurium was unknown, the effect of rifampicin on PabrB was examined in Salmonella. An abrB::luxCDABE reporter fusion was constructed (pGY43) and the luminescence in S. typhimurium 14028 was determined. When luminescence of the invF and abrB reporter fusions in the presence of increasing amounts of rifampicin was compared (Fig 3.16C), the sensitivity of invF transcription to rifampicin was approximately three-fold greater when compared to abrB, comparable to results found in vitro (Fig 3.16B).  66  Figure 3.15. Rifampicin inhibition of flgK, fliC, invF and abrB transcripts. Phosphorimage of the products of transcription from flgK, fliC, invF and abrB promoters in the presence of varying amounts of rifampicin (50 to 0 g ml-1 as shown in Fig 3.16) separated by denaturing polyacrylamide gel electrophoresis.  Figure 3.16. Graphs of transcription inhibition by rifampicin. A,B. In vitro transcript amounts in the presence or absence of rifampicin represented in Fig. 3.15 determined by spot densitometry using ImageQuant 5.2 software plotted as log10 values of the percentage of the amount of transcript in the absence of rifampicin. Templates were fliC (diamond), flgK (square), abrB (x) and invF (triangle). C. Luminescence of abrB reporter (x, pGY43) and invF (triangle, pGY17) reporter plasmids carried in S. typhimurium 14028 grown in the presence of various concentrations of rifampicin. Luminescence values are plotted as log10 values. Error bars represent the standard deviation of at least three experiments. The slope of the resulting trend lines were used to calculate fold differences in rifampicin sensitivity.  67  3.3.4. Mg2+ and NTP concentration effects. In the past several years, several studies have reexamined the relationship between rifampicin, RNAP and the influence of Mg2+ on rifampicin sensitivity (Artsimovitch et al., 2005; Feklistov et al., 2008). Since some previously published data suggested that rifampicin inhibits transcription by inhibiting the catalytic activity of RNAP (Artsimovitch et al., 2005), a possible mechanism of differential rifampicin sensitivity of transcription from PflgK and PinvF, compared to the control promoters, PfliC and PabrB, may involve Mg2+. To test this hypothesis, transcription inhibition assays were carried out at different Mg2+ concentrations.  In vitro transcription assays were carried out as those shown in Fig. 3.15 except final Mg2+ concentrations were varied. The amounts of run-off transcript produced at each concentration of Mg2+ in the presence or absence of rifampicin are shown in Fig. 3.17. Sensitivity of flgK transcripts to rifampicin showed Mg2+ dependence over the range of concentrations tested, 1 to 16 mM, (Fig. 3.17). The effect of Mg2+ observed with flgK transcription was similar to rifampicin susceptibility of T7 A1 transcription by E coli RNAP (Artsimovitch et al., 2005) where higher levels of Mg2+ decreased the inhibition by rifampicin. In contrast, the effect of rifampicin on fliC transcription was not Mg2+ dependent over the concentration range tested. Sensitivity of transcription from invF and abrB did not differ over a similar Mg2+ range (not shown). Since only one of the two promoters showed Mg2+ dependent rifampicin sensitivity, this suggested that the level of Mg2+ was not a primary determinant in the polymerase promoter interactions leading to rifampicin responses.  68  Figure 3.17. Mg2+dependence of flgK and fliC transcription. A. Phosphorimages of in vitro transcription products from flgK and fliC promoters in the presence or absence of rifampicin (12.5 µg ml-1) with decreasing amounts of Mg2+. Transcripts were separated by denaturing polyacrylamide gel electrophoresis. B. In vitro transcription products were measured using ImageQuant 5.2 and are plotted as log10 of the percentage of the level seen in the no rifampicin transcript. Templates were fliC (diamond) and flgK (square). Error bars indicate standard deviation of transcript spot densities from at least three experiments.  Concentrations of initiating NTP (iNTP) have been shown to regulate transcription of rRNA in B. subtilis (Krasny & Gourse, 2004) and E. coli (Gaal et al., 1997). Both NTP and rifampicin are present in the RNAP active site. The possibility that rifampicin mediated down-modulation might be affected by the concentration of NTPs was examined in vitro. Standard transcription assays contain 0.4 mM of all NTPs with the exception of the 32P labelled NTP which is used at a lower concentration. To examine the effect of NTP, dilutions of a mixture of unlabelled nucleotide (ATP, CTP and UTP) were tested in in vitro transcription assays with or without rifampicin (Fig 3.18). GTP was not varied and remained at 5.1 M (total concentration).  Decreasing NTP (ATP, CTP and UTP) concentrations in the transcription assay did not alter rifampicin sensitivity of flgK or fliC transcription. However, the overall transcription level from PflgK declined sharply when the concentration of NTP was below 0.1 mM. The rifampicin 69  sensitivities of transcription from PinvF and PabrB were also tested and were affected equally by lowering the NTP concentration and in both cases, overall transcription declined when the NTP concentration was below 0.025 mM (not shown). The varied minimal requirement for NTP may reflect the degree of stabilization given to the open complex by iNTPs. In E.coli rRNA promoters, high iNTP levels are required to stabilize open complexes. Higher iNTP levels result in greater proportions of stabilized open complexes and result in a corresponding increase in transcription (Gaal et al., 1997). However, since rifampicin sensitive promoters were not preferentially affected by NTP levels, it suggested that NTP concentration was not a significant factor in rifampicin mediated down-modulation.  Figure 3.18. Effect of NTP concentration on flgK and fliC transcription and rifampicin inhibition. A phosphorimage of in vitro transcription products from flgK and fliC promoters in the presence or absence of rifampicin (12.5 µg ml-1) and various NTP concentrations. The undiluted NTP concentration was 0.4 mM. Dilutions of 1:4 (4) and 1:16 (16) were tested. Transcripts were separated using denaturing polyacrylamide gel electrophoresis.  70  3.4  In vitro characterization of rifampicin up-modulated promoters.  Since rifampicin mediated down-modulation of invF and flgK could be observed in vitro, several rifampicin up-modulated promoters were examined in vitro. As was the rationale with the rifampicin down-modulated promoters, in vitro transcription experiments were performed with varying amounts of rifampicin to test whether RNAP alone was sufficient to mediate rifampicin mediated up-modulation or if other factors might be required. Two promoters were examined in vitro, traS and STM3595.  Before in vitro transcription assays could be done, the transcription start sites for STM3595 and traS were determined. To map the transcription start site of RNA synthesized in vitro, primer extensions were conducted on the respective RNAs (Fig 3.19). Primer extension using the STM3595ent rev primer was also conducted on RNA harvested from cells. Analysis of PSTM3595 start transcription start sites suggested that transcription began at different nucleotides when initiated in vitro versus in vivo. As an additional method, the STM3595 in vivo transcription start site determined by 5’ RACE. The start site was similar to the one determined by primer extension using RNA harvested from cells (Fig 3.10).  Since it appeared that the STM3595 transcription start used in vitro and in vivo differed, the DNA region containing the in vitro transcription start site was tested to determine whether it was functional in vivo. A 121 bp fragment containing the -10 and -35 elements corresponding to the in vitro transcription start site but not the in vivo start site was cloned into the lux reporter. The strain carrying this fusion, pGY41, showed little promoter activity and was not stimulated by  71  rifampicin (Fig. 3.9C). This suggested that the start site observed in vitro was not functional in vivo.  Figure 3.19. Primer extensions of traS and STM3595. Autoradiography of primer extension products of RNA obtained from in vitro transcription of traS (A), in vitro transcription of STM3595 (B) and STM3595 RNA harvested from cells (C). B,C. For reference, the upper arrows indicate the position of the start of the in vitro transcript and the lower the in vivo transcript. Sample lanes “s” were loaded with varying amounts of sample. Nucleotide sequences are shown in Fig 3.29.  72  Since the traS transcription start had been previously determined in E.coli, the S. typhimurium +1 site of the in vitro PtraS transcript was determined (Fig 3.19). The S. typhimurium PtraS transcription start is 2 bp shifted from the transcription start found for the E.coli transcript (Ham et al., 1989) (Fig 3.20). Although the +1 to -10 spacing is smaller than the usual 6 to 9 bp, there are other known promoters with the same spacing of 4 bp (Walker & Osuna, 2002). Since recognizable-10 and -35 elements are located appropriately with respect to the +1 found by primer extension, I assumed that this is the biologically relevant start site in S. typhimurium.  S. typhimurium  ataaaacagaagcacTatCtcagaataataaaTATAATgaaGaaaagg  E. coli  ataaaacagaagcaTTaACtcagaataataaaTATAATgaagaAaagc -35  -10  +1  Figure 3.20. A comparison of S. typhimurium and E. coli traS transcription start sites. The S. typhimurium start site is labeled as determined in Fig. 3.19. An E.coli start site is shown for comparison (Ham et al., 1989). -35 and -10 are underlined and are capitalized where nucleotides match the 70 consensus. The transcription start site, +1, is capitalized.  In vitro transcriptions with templates containing PtraS and PSTM3595 did not show rifampicin mediated up-modulation (Fig. 3.21). In contrast to invF and flgK, in the absence of additional protein factors, traS transcription was quite strong (Fig 3.21). Although deletion analysis suggested that rifampicin mediated up-modulation from traS required DNA sequences either upstream or downstream of the transcription start site (Fig 3.9A), a transcription factor required for activation of traS that binds upstream of the +1 or at any other location has not been identified.  73  Thus, there were no known candidates for protein factors that could be examined for rifampicin stimulation of traS transcription in vitro.  Transcription of STM3595, one of the most promoters most highly up-modulated by rifampicin in reporter assays, showed two in vitro products, neither of which increased in the presence of rifampicin (Fig 3.19, 3.21). Deletion studies had suggested that DNA downstream of the +1 of STM3595 RNA was required for rifampicin mediated up-modulation (Fig 3.9C), but this region of STM3595 had not been previously examined for transcription or elongation factor binding. There are no known candidates for proteins involved in rifampicin mediated up-modulation at this promoter. Since rifampicin mediated up-modulation could not be shown in vitro, these data suggested that RNAP alone was not sufficient for rifampicin mediated up-modulation.  Figure 3.21. In vitro transcripts of traS, STM3595 and abrB in response to rifampicin. Phosphorimage of transcription products from traS, STM3595 and abrB promoters in the presence of varying amounts of rifampicin (12.5, 6.3, 3.1, 1.6 and 0 g ml-1). Transcripts are separated by denaturing polyacrylamide gel electrophoresis and arrows indicate transcript of interest. 74  3.5  Intracellular factors involved in rifampicin mediated up-modulation.  Figure 3.23. Fluorescence of 14028 Tn10dTet mutants. A. Fluorescence of S. typhimurium 14028 cells (black), 14028/pGY42 Tn10dTet mutants grown in the absence (blue) or presence (green) of rifampicin. B. Fluorescence of particles with fluorescence in the bottom 1% (black) or top 1% (blue) of the parent population grown in rifampicin (green). C. The trace of the bottom 1% shown in (B) with a red box representing the particles collected for a second sort.  Three lines of evidence suggested that an intracellular factor may be involved in rifampicin mediated up-modulation. First, the time course assays in section 3.1.4 showed a 2 to 3 hour lag between rifampicin addition and induction of expression. This lag may represent accumulation of a factor. Secondly, experiments with pretreated cells and spent media in section 3.1.5 showed that pretreated cells had higher promoter activity than untreated cells and spent media did not shorten the lag or increase the level of rifampicin induction. This suggested that a factor had accumulated in the cells, but not in the supernatant. Thirdly, rifampicin mediated up-modulation of transcription from STM3595 and traS templates could not be shown in vitro using only RNAP and the DNA template (section 3.4). As one approach to identify possible intracellular factor(s) involved, a screen was conducted to search for mutants that no longer displayed rifampicin mediated up-modulation of STM3595. Strain 14028 containing a STM3595::gfpmut3a reporter 75  plasmid (pGY42) was randomly mutagenized using Tn10dTet. Cells were screened for mutants that were not up-modulated by rifampicin using fluorescence activated cell sorting (FACS) followed by screening with the Victor II plate reader.  Two different mutants were identified in the Tn10dTet FACS screen (Fig. 3.22). Nucleotide sequencing of inverse PCR products using primer binding sites in Tn10dTet (Fig. 2.2) showed that in the two mutants, the transposon had inserted into the yciK or pcnB genes, respectively (Fig. 3.23). The yciK gene encodes a putative oxidoreductase with no reported function. However, it may exist in the same transcriptional unit as cobA (Fig. 3.23A) and the transposon insertion could prevent transcription of cobA. CobA is known to be involved in de novo synthesis of cobalamin (vitamin B12). The gene for pcnB encodes poly(A) polymerase I. Little is known about this protein in S. typhimurium. In E.coli, poly (A) polymerase affects the rate of transcript degradation by causing 3’ adenylation of transcripts. This facilitates the 3’ to 5’ exonucleolytic degradation of RNA with a structured 3’ end (Urban & Vogel, 2008).  Figure 3.24. Schematic representation of yciK and pcnB insertions.  76  Since the yciK insertion reduced rifampicin mediated up-modulation of STM3595 (Fig 3.24A), the effect of this mutation was examined for the five other genes that showed either rifampicin mediated up-modulation or down-modulation. The insertion moderately affected transcription modulation by rifampicin in one other strain containing the flgK reporter fusion (pGY13) (Fig 3.25C). The other promoters, including the promoter for the control gene STM2901 (pGY21), had similar levels of rifampicin mediated up-modulation or down-modulation in the yciK mutant and the wild type backgrounds (Fig. 3.24, 3.25). For the most part, the yciK insertion only affected rifampicin mediated up-modulation of the STM3595.  Figure 3.24. Luminescence of rifampicin up-modulated promoter reporter fusions in yciK::Tn10d and wildtype backgrounds. Luminescence of yciK mutants (black) or wildtype (grey) strains carrying STM3595 (pGY37, A), traS (pGY25, B) and spvA (pGY31, C) reporters. Cells were grown in LB supplemented with ampicillin and the indicated amount of rifampicin. Error bars indicate standard deviation of at least three experiments.  Since the yciK insertion may affect the levels of vitamin B12 in the cells, the role of vitamin B12 on rifampicin mediated up-modulation of STM3595 expression was examined in the yciK 77  mutant. Media for the yciK mutant and the wild type strains harboring the STM3595 lux reporter (pGY37) was supplemented with vitamin B12. Addition of vitamin B12 did not return rifampicin mediated up-modulation to levels observed in the wild type backgrounds (not shown). Therefore, it is unlikely that vitamin B12 is involved.  Figure 3.25. Luminescence of rifampicin down-modulated promoter reporter fusions in wildtype and mutant backgrounds. Luminescence of pcnB mutants (white), yciK mutants (black) or wildtype (grey) strains carrying invF (pGY17, A), fliA (pGY8, B), flgK (pGY13, C) and STM2901 (pGY21, D) reporters. Cells were grown in LB supplemented with ampicillin and the indicated amount of rifampicin. Error bars indicate standard deviation of at least three experiments.  The pcnB mutant showed increased sensitivity to growth inhibition by rifampicin compared to the wild type. Using a concentration of rifampicin slightly higher than the concentration used in reporter assays, growth of the mutant was noticeably more susceptible to growth inhibition by rifampicin compared to the wild type strain (Fig 3.26). When the effect of the pcnB mutation on the rifampicin response was examined in the other genes showing transcription modulation by rifampicin, the pcnB insertion mutants carrying reporter fusions to the invF and fliA promoters 78  showed more rifampicin mediated down-modulation than the wild type strains with the same reporter fusions (Fig. 3.25A, B).  Figure 3.26. Growth curves of the wildtype and pcnB mutant strains in the presence or absence of rifampicin. Growth of 14028 (diamond, square) and the pcnB mutant (x, triangle) in the absence (diamond, x) or presence (square, triangle) of rifampicin (6 µg ml-1). Error bars indicate standard deviation of at least three experiments. Cells were grown in LB supplemented with ampicillin and rifampicin as indicated.  3.6.  Nucleotide sequence analysis of rifampicin modulated promoters.  3.6.1. Motif searches. To search for common elements in the sequences of the rifampicin responsive promoters, in silico motif searches were conducted. Nucleotide sequences of down and up modulated promoters were examined for the presence of motifs that may be associated with transcription modulation by rifampicin. Three rifampicin down-modulated promoters, three rifampicin up-  79  modulated promoters and a set containing all six sequences were analyzed by the following algorithms with several settings for width, frequency, strand and palindromes: MEME (Bailey & Elkan, 1994), GLAM2 (Frith et al., 2008) and Gibbs motif sampler (McCue et al., 2001).  Figure 3.27. Motifs found in down- and up-modulated promoter nucleotide sequences. Motifs found in rifampicin down-modulated (A) and rifampicin up-modulated (B) promoters. See (Crooks et al., 2004) for a description of sequence logos.  When examining the motifs produced by the three software programs, most motifs differed in relative position when input sequences were aligned by their transcription start site and were disregarded. However, when examining the rifampicin down-modulated promoter sequences, one motif (Fig. 3.27A) was identified by several of the analysis programs. This motif was essentially the 28 -10 element with one extra nucleotide on the 5’ and 3’ end. The consensus sequence for Salmonella 28 promoters is TAAA-N15-GCCGATAA (Kutsukake et al., 1990). The 28 -10 like element was in a similar position, centered at -32, in the fliA and flgK promoters (Fig 3.28) but 22 nucleotides downstream of the HilA-dependent +1 in invF (upstream of the 80  invF coding region). This raised the possibility that rifampicin mediated down-modulation observed was due to an effect on the 28 holoenzyme. As a test of whether there was a 28 dependent promoter on the invF template, in vitro transcription assays using the invF promoter as template were carried out adding increasing amounts of 28 to the holoenzyme. The addition of several different concentrations of 28 (3.0, 1.5 and 0.75 M) to RNAP (75 nM), including amounts used for transcription of flgK transcription, yielded no transcripts from the invF template (data not shown). Therefore, it seemed unlikely that rifampicin mediated downmodulation from 28 dependent promoters was a common feature of rifampicin down-modulated promoters.  Figure 3.28. Nucleotide sequence alignment of flgK, fliA and invF promoters. Nucleotide sequences are numbered relative to the transcription start site (+1). -35 elements (broken lines), -10 elements (solid line), a partial HilA binding site (arrow) and similar sequences (shaded) are indicated.  In addition to the in silico analysis of rifampicin responsive promoters, a visual inspection for common nucleotide elements in sequences around the +1 was conducted. This region is relevant to rifampicin mode of action since rifampicin inhibits formation of RNA chains longer than 2 to 81  3 nucleotides long (Campbell et al., 2001; McClure & Cech, 1978) and is proposed to inhibit formation of the phosphodiester bond between the third and fourth nucleotide (Artsimovitch et al., 2005). Inspection showed four similar nucleotides -26C and TAW (where W represents T or A nucleotides) centered at +1 (Fig. 3.28, shaded); their significance is explored in section 3.6.2.  Figure 3.29. Nucleotide sequences of STM3595, traS and spvA. Alignments are numbered relative to the transcription start site (+1). Transcription start sites for traS (primer extension using in vitro RNA) , STM3595 (primer extension and 5’ RACE using RNA harvested from  82  cells) and spvA (Sheehan & Dorman, 1998) are indicated. Conserved residues are shown in yellow. A palindromic motif is shown in grey. In silico analysis of the rifampicin up-modulated promoters revealed no sequence motifs in similar positions in all three sequences. However, one palindrome (Fig. 3.27B) was found in a similar position in STM3595 and traS and in a different position in spvA (Fig. 3.29). This could be a biologically relevant motif considering deletion studies and the E.coli traS transcription start site. Deletion suggested that sequence downstream to the +1 was important for rifampicin mediated up-modulation of STM3595 and traS (Fig 3.8). Secondly, the in vitro transcription start site determined for traS may not reflect the in vivo transcription start site. Since high quality RNA is more readily isolated in vitro in comparison to RNA isolated from cells, the +1 site of traS transcription was determined for RNA synthesized in vitro (Fig 3.19). However, the E. coli start site identified from RNA harvested from cultured cells (Ham et al., 1989) is shifted 2 bp downstream from the in vitro S. typhimurium start site (Fig 3.20). It seems possible that the S. typhimurium and E.coli start sites are identical when RNA is harvested from cells. As such, the palindrome in STM3595 and traS would be in an identical position in both sequences, four bps downstream of the +1.  3.6.2. Preliminary exploration of nucleotide sequence motifs. To examine the biological significance of the four similar nucleotides found in the rifampicin down-modulated promoters (Fig. 3.28), they were mutated and the respective reporter fusions assayed for luminescence responses to rifampicin. The following base changes were -26C to A, -1+2TAW to GCC or both sets of mutations were made in flgK, fliA and invF promoter fusions  83  (Table 3.1). Due to cloning constraints resulting from employing a 9.4 kb plasmid, a mutant reporter with both sets of mutations in the invF sequence was not constructed.  Table 3.1. Nucleotide alterations of common nucleotides in rifampicin down-modulated promoters and the resulting reporter plasmids. Promoter  flgK  fliA P2  invF  Plasmid Original nucleotides  pGY47  pGY48  pGY49  pGY44  pGY45  pGY46  pGY50  pGY51  C  TAT  C TAT  C  TAA  C TAA  C  TAT  Base position -26 -26 -26 -1,+1,+2 -26 -1,+1,+2 -26 -1,+1,+2 of alteration(s) -1, +1,+2 -1,+1,+2 Replacement A A A GCC A GCC A GCC nucleotide(s) GCC GCC pGY19 Parent plasmid pGY15 pGY15 pGY15 pGY11 pGY11 pGY11 pGY18 Fold decrease in basal 1.1 4.4 1.01 7.4 8.0 0.30 3.7 3.2 expression  0.2 0.4 0.08 0.6 0.8 0.04 0.3 compared to wild type* *corresponds to luminescence of wild type divided by luminescence of mutant grown in 0 g/ml of rifampicin as seen in Figure 3.30  The TAW nucleotides were important for promoter activity from all three promoters, whereas -26C was only important for one. For all three promoters, flgK, fliA P2 and invF, alterations from TAW to GCC significantly reduced their basal activity (Table 3.1, Fig. 3.30, no rifampicin) suggesting that the -1,+1 and +2 nucleotides are significant for promoter strength. When examining the flgK and fliA P2 reporters, -26 C to A alterations did not affect basal activity or rifampicin hypersensitivity (Table 3.1, Fig. 3.30) indicating that -26C plays no role in rifampicin mediated down-modulation. In contrast, changing -26 C to A in the invF promoter increased basal activity (Table 3.1, Fig 3.30C,D). In cases where both sets of mutations were made (fliA 84  and flgK reporters), activity was similar to the mutants with only the TAW base changes (Fig. 3.30A, B diamond and x) having reduced basal activity when compared to the wild type. Again this suggested that positions -1 to +2 but not position -26 are important for rifampicin mediated down-modulation. Since mutations at position -1 to +2 reduced basal activity and rifampicin at 5 μg ml-1 reduced transcription of the reporters strains to the lower limit of detection in this assay, the sensitivity of transcription to rifampicin is reduced in the mutants when compared to the wild type reporter (Fig 3.30). The -1+1+2 positions of the flgK, fliA P2 and invF promoters are important for promoter activity and may be involved in rifampicin hypersensitivity in these cases.  Figure 3.30. Luminescence values from mutated fusion reporters grown in the presence of varying amounts of rifampicin. Luminescence at 6 hrs of 14028 carrying A) pGY15 (diamond), B) pGY11 (diamond), or derivatives thereof with -26C to A alterations (x), with -1+2 TAW alterations to GCC (triangle) or both alterations (square). Luminescence at six hours of 14028 carrying C) pGY19 (diamond), or a derivative thereof with a -26C to A alteration (x) or D) pGY18 (diamond), or a derivative thereof with a -1+2 TAT alteration to GCC (triangle). Each symbol represents one culture condition with growth in the indicated amount of rifampicin; error bars indicate standard deviation of least three experiments. 85  Chapter 4. Discussion. In S. typhimurium strain 14028, sub-MIC rifampicin modulated expression of at least 22 genes from 3 to 200-fold (Table1.1). Among the down-regulated genes were ones related to motility and host cell invasion. Some of the up-regulated genes encoded proteins of unknown function, while others encoded proteins related to carbon metabolism and virulence. This thesis examines the mechanisms by which sub-MIC rifampicin modulates bacterial transcription from six of these promoters (PfliA, PflgK, PinvF, PspvA, PSTM3595 and PtraS).  Since the mode of action of rifampicin has been studied in detail (Campbell et al., 2001; Floss & Yu, 2005), the simple hypothesis for the effect of sub-MIC rifampicin would be a general limitation of transcription capacity. However, the effect of sub-MIC rifampicin on transcription presented a paradox: rifampicin not only specifically down-modulated certain promoters but also up-modulated transcription for other promoters. Three promoters PflgK, PfliA and PinvF that were down-modulated by rifampicin were studied in this thesis. DNA deletion studies of PfliA and PflgK showed that ~50 bp fragments containing the 28-dependent promoters were sufficient to observe rifampicin mediated down-modulation in culture. For PinvF, a promoter specific activator binding site was required for down-modulation. In each case, transcription from the promoters showed increased rifampicin sensitivity in vitro relative to the controls which may reflect the down-modulation of these promoters by rifampicin when grown in culture. Since the magnitude of rifampicin mediated down-modulation observed in vitro was not equivalent to the effect seen in culture, an additional mechanism may contribute to down-modulation. The other three promoters examined are up-modulated by rifampicin in culture: PSTM3595, PspvA and PtraS. Evidence presented suggested that up-modulation may be mediated by an intracellular, but not an  86  extracellular factor and, in two of the three promoters, nucleotide sequences downstream of the transcription start sites were required.  Below, possible mechanisms of up-modulation are discussed. Since nucleotides downstream of the transcription start site are involved with up-modulation, particular attention is paid to mechanisms relating to this region such as transcription attenuation. Lastly, the biological relevance and applications for drug discovery and therapeutics are presented.  4.1.  Mechanisms of rifampicin mediated down-modulation at RNA polymerase.  As mentioned in the Introduction, in E. coli five DNA elements contribute to promoter recognition by RNAP:70: the -10 hexamer, -35 hexamer, extended -10, UP element and discriminator. For the most part, detailed mechanistic and kinetic studies of transcription initiation have been only conducted in E. coli as a model Gram-negative organism and B. subtilis as a model Gram-positive organism. Since E. coli and S. typhimurium RNAP are 99% identical in amino acid sequence, promoter recognition presumably works in a similar manner. Limited studies of transcription initiation has been conducted using alternate sigma factors such as S, the stationary phase sigma factor and 32, the heat shock sigma factor. There are a few investigations of transcription mechanism for RNAP containing 28 (flagellar). -10 and -35 elements have only been identified for the E. coli and S. typhimurium 28, TAAAGTTT (-35 element) and GCCGATAA (-10 element) (Ide et al., 1999).  Transcription has been functionally separated into three parts: initiation, elongation and termination. Most regulation of transcription occurs at the level of initiation. Transcription 87  factors will often activate or repress transcription by affecting the rate by which RNAP:DNA performs certain isomerizations during transcription initiation. These isomerizations (closed, open, initiated or elongating complex, Figure 1.3) involve rearrangements of RNAP and DNA. A closed complex is characterized by specific binding of RNAP to double stranded promoter DNA. The open complex is formed by melting of the double stranded DNA. Melting of DNA will nucleate in the -10 region of the promoter and extend from –11 to +3 while the RNAP footprint extends to +20 (Haugen et al., 2008). Open complex transitions to the initiated complex as the NTPs are polymerized. Once a 7 – 12 nucleotide transcript forms,  is released and RNAP forms a processive elongating complex. If an elongating complex is not formed, the abortive transcript is released and RNAP returns to an open complex (Record et al., 1996).  Some possibilities for the mechanism of down-modulation were that sub-MIC rifampicin slows growth, causes a stress response or that minor changes in growth rate may have a differential effect on some promoters. Alternatively, rifampicin may have specific effects at certain promoter-RNAP complexes. Using a bacterial lux-reporter system, RT-PCR and in vitro techniques, the work described in this thesis attempted to test various aspects of transcription and to characterize the mechanism by which sub-MIC rifampicin modulates transcription of certain S. typhimurium promoters.  Sub-cloning, testing of promoter fragments and in vitro transcription experiments described in this thesis suggested that rifampicin mediated down-modulation occurs directly at the level of RNAP-promoter interaction. Sub-cloning studies showed that ~50 bp fragments containing only the minimum promoter elements (-35, -10 and +1) of 28-dependent PflgK and PfliA were  88  sufficient for down-modulation by rifampicin, suggesting that no upstream trans-acting factors were necessary for down-modulation at these promoters (Figure 3.7). Furthermore, HilA was required for PinvF down-modulation by rifampicin, but this may just reflect the technical problem of having sufficient transcripts to see inhibition. Thus, it seemed unlikely that HilA was a general element of rifampicin mediated down-modulation at PinvF, PflgK and PfliA.  It is possible that different holoenzymes might be differentially sensitive to rifampicin. It has previously been reported for one pair of promoters that RNAP-70 (PL promoter) is less sensitive to rifampicin than RNAP-32 (PgroE promoter) (Wegrzyn et al., 1998). However, in vivo and in vitro transcription experiments in this thesis showed that differential rifampicin sensitivity occurred for both RNAP-70 transcription from PinvF and RNAP-28 transcription from PflgK (section 3.3.3). Transcription by the two forms of polymerase holoenzymes from the control promoters (PabrB and PfliC) were similarly sensitive to rifampicin, further indicating that, under the conditions used, rifampicin hypersensitivity was not due to the sigma factor associated with the holoenzyme. Thus, it seems that hypersensitivity to rifampicin inhibition results from differences in specific interactions between RNAP and rifampicin down-modulated promoters.  Based on recent literature on mechanisms of transcription initiation, I explored two reaction conditions for in vitro assays, varied nucleotide and Mg2+ concentrations. These two factors are linked as there is an active site Mg2+ in RNAP and NTPs are the substrates of RNAP. Furthermore, during the course of this work a study appeared suggesting that rifampicin inhibition was affected by Mg2+ (Artsimovitch et al., 2005). The rifampicin sensitivity of transcription from the flagellar promoter, PflgK, displayed Mg2+ dependence (Fig. 3.17) but 89  transcription from PinvF did not. While this is counter to the notion of a single mechanism for rifampicin hypersensitivity, it leaves open the possibility that the effect of Mg2+ on rifampicin action might be promoter dependent. Altering NTP concentrations had no differential effect on rifampicin inhibition in vitro when compared to control promoters. While no role for Mg2+ in rifampicin mediated down-modulation was uncovered, it was clear that rifampicin hypersensitivity of transcription in vitro was similar to the rifampicin mediated down-modulation observed in culture. This implies that DNA:RNAP interactions modulate rifampicin binding and in turn that small molecules such as antibiotics can directly modulate transcription.  Transcription from PflgK followed the previously suggested hypothesis that rifampicin inhibition of transcription acts by affecting binding of Mg2+ to the RNAP active site (Artsimovitch et al., 2005). A more recent study by Feklistov et al. (2008) challenged the relationship between rifampicin and Mg2+ found by Artsimovitch et al., tested several assertions made in 2005 and found them false. However, Feklistov et al. tested only one promoter. In the work presented here, it appeared that one of the four promoters tested, the flgK promoter but not the abrB, invF or fliC promoters, showed Mg2+ dependent rifampicin sensitivity (Fig 3.17). Since rifampicin binds to the RNAP and the same enzyme was used in all tests, one might expect that all of the promoters would display the same rifampicin sensitivity, or at least responses correlating to the sigma factor used. These findings suggest there may be unknown effects involved that could explain rifampicin mediated down-modulation. As with all in vitro experiments, buffer conditions or order of additions may contribute to the magnitude of Mg2+ dependent rifampicin sensitivity and may partially explain differences observed here and in other studies. Studies using alternative techniques to measure transcription initiation (Mekler et al., 2011) may need to  90  be employed to elucidate the role of Mg2+ in rifampicin inhibition of transcription initiation and certainly a wider range of promoters needs to be investigated.  4.1.1. The role of +1 region in transcription initiation. In some promoters, such as rRNA promoters, the +1 region and the iNTP, has been shown to be important for regulating transcription initiation. The iNTP is the first nucleotide of the transcript and base pairs at the +1 base of the template strand at the promoter. At the rrn promoters, the open complex is highly unstable, decreasing the possibility of transcription initiation. Therefore, efficient transcription will only occur when there are high concentrations of iNTP. When NTP pools are high, such as when the cell actively growing, the iNTP will base pair to the +1 nucleotide on the template strand of DNA, stabilize the open complex and increase the likelihood of transcription initiation (Gaal et al., 1997; Krasny & Gourse, 2004).  Data for consensus nucleotides in the +1 region of E. coli promoters shows a very weak nucleotide bias. Recently, a modified high throughput 5’ RACE protocol was used to determine approximately 1700 transcription start sites in E.coli (Mendoza-Vargas et al., 2009). The bases AAT seem to be weakly preferred at positions +1 to +3 (Fig 4.1). The information content of the motif is less than 0.2 bits at any position (Fig 4.1), 2 being the highest bits per position possible. Higher binding site conservation corresponds to higher information content (measured in bits). Typical information content of highly conserved positions in transcription factor binding sites, such as the CAP binding site, is usually between 1 and 2 bits (Crooks et al., 2004; Schneider et al., 1986). When the +1 to +3 motif of E.coli promoters compared to the TAW (Fig 3.28) consensus sequence at -1 to +2 of the rifampicin down-modulated promoters studied here, there  91  is one difference. The -1 of the rifampicin down-modulated promoters has a T bias where no bias exists in E. coli promoters. This base difference may reflect some characteristic of rifampicin down-modulated promoters that renders them hypersensitive to rifampicin inhibition.  Figure 4.1. Sequence logo of the +1 region of E. coli promoters. Data was obtained by a high throughput pyrosequencing strategy and compiled by Mendoza-Vargas et al.(2009). Their data was used to create a sequence logo using WebLogo software (Crooks et al., 2004).  4.1.2. A working model of rifampicin mediated down-modulation. A working model for rifampicin mediated down-modulation might involve the +1 region of the rifampicin down-modulated promoters. The accepted model for inhibition of transcription by rifampicin is that its binding to RNAP sterically inhibits extension of the RNA chain when it is 2-3 nucleotides long (Campbell et al., 2001; Feklistov et al., 2008; McClure & Cech, 1978). Elongating RNAP:DNA complexes are resistant to rifampicin (Carpousis & Gralla, 1985). When the elongating complex has formed, the nascent RNA chain presumably blocks binding of 92  rifampicin and subsequent inhibition. Certain DNA sequences of the rifampicin downmodulated promoters such as the TAW consensus may render rifampicin down-modulated promoter:RNAP complexes more sensitive to inhibition by rifampicin. This may involve rifampicin down-modulated promoter:RNAP complexes which form initiated complexes more slowly or have altered structures compared to typical promoters. This would allow more time for rifampicin to affect transcription initiation causing a higher proportion of the DNA:RNAP complexes at those particular promoters to be inhibited by rifampicin, would decrease the rate of transcription initiation and cause down-modulation of transcription.  4.2.  Other possible mechanisms of rifampicin mediated down-modulation.  The levels of rifampicin mediated down-modulation observed in vitro were lower than the responses seen in culture. It seems possible that a secondary mechanism exists which potentiates rifampicin mediated down-modulation in culture in addition to the rifampicin hypersensitivity of rifampicin down-modulated promoters. Several mechanisms which may act in addition to rifampicin hypersensitivity (discussed in section 4.1) are presented below. This includes the role of poly(A) polymerase (pcnB insertion mutant identified by FACS) and several Salmonella regulators (known regulators of invasion and motility) in rifampicin mediated down-modulation, the activator HilA being the most promising candidate.  FACS screening identified a mutant with an insertion in pcnB that had increased rifampicin mediated down-modulation relative to the wildtype strain (Fig. 3.25). In E.coli, pcnB encodes the enzyme poly(A) polymerase which adenylates specific transcripts at their 3’ end, destabilizing them; consequently, pcnB mutants usually have increased levels of the respective  93  transcript when compared to wildtype (Urban & Vogel, 2008). A direct effect of poly(A) polymerase seems unlikely to explain rifampicin mediated down-modulation since expression of the reporter construct was decreased in pcnB mutants. However, transcripts that are decreased in pcnB mutants are indirectly regulated by poly(A) polymerase. In these cases, poly(A) polymerase destabilizes the transcript of a repressor for these transcripts which increases the amount of gene of interest transcript (Urban & Vogel, 2008). A similar phenomenon may be occurring at rifampicin down-modulated promoters.  The Salmonella pcnB mutant defective in rifampicin mediated up-modulation of STM3595 that was isolated by FACS displayed increased susceptibility to growth inhibition by rifampicin (Fig 3.26). The increased susceptibility to rifampicin may reflect a general role of poly(A) polymerase in Salmonella physiology. Since rifampicin resistant mutants no longer display down-modulation, perhaps the opposite, mutants which are more susceptible to rifampicin mediated growth inhibition displaying stronger transcription down-modulation, is not surprising.  Other Salmonella regulators such as PhoP and SirA were tested for their involvement in rifampicin mediated down-modulation by rifampicin. The BarA/SirA two component system has been shown to negatively influence motility and invasion in Salmonella (Teplitski et al., 2003). Rifampicin also down-modulates expression of invasion and motility genes. Reporter fusions genes that showed rifampicin mediated down-modulation (invF and sopB, Table 1.1) known to be positively regulated by SirA, had a lower basal level expression in the sirA mutant but were still down-modulated by rifampicin (not shown). The PhoPQ system negatively regulates invasion genes (Teplitski et al., 2003). In Salmonella, PhoPQ is a two component  94  system regulated by magnesium ion concentrations, acidic pH and cationic peptides. PhoPQ is important for Salmonella survival in host cells, resistance to antimicrobial peptides and acid pH (Song et al., 2008). Promoter-reporter fusions carried in a phoP::cam mutant showed similar patterns of down-modulation as the same reporter fusions carried by the isogenic parent (not shown). SirA and PhoP are regulators of individual rifampicin down-modulated promoters, but are unlikely to be primary determinants of rifampicin mediated down-modulation.  The effect of rifampicin on some rifampicin down-modulated genes may be indirect, for example, involving HilA. Of the promoters examined in this thesis, this would only include PinvF. Deletion studies using invF promoter-reporter fusions suggested that rifampicin mediated down-modulation required the HilA binding site (Figure 3.8). HilA protein was required to observe rifampicin mediated down-modulation from in vitro transcripts (section 3.3) but this could have just reflected the technical problem of having sufficient transcripts to detect inhibition. Several genes (not examined in this thesis) such as sopA, sopB, and siiA, that are down-modulated by rifampicin (Table 1.1) are known to be HilA regulated (Thijs et al., 2007). Repression of the HilA regulon is mediated by Hha (Teplitski et al., 2003). Lux expression in B9(2), a S. typhimurium reporter strain containing the lux genes fused to the hha promoter region, was up-modulated by rifampicin (Figure 4.2). Thus, it is possible that rifampicin mediated up-modulation of hha may lead to down-modulation of HilA and consequent repression of invF, sopA, sopB and siiA. Strain B9(2) was initially isolated in a preliminary screen of a partial S. typhimurium lux reporter library that was discarded (Goh, unpublished). The original 6528-clone library used in the second large screen for rifampicin responsive reporters, from which the fusions used in this thesis were obtained, had approximately 2.7 fold  95  coverage (Bjarnason et al., 2003). This is likely not saturating. This may explain why a hha reporter was not re-isolated and thus not included in Table 1.1.  Figure 4.2. A disk diffusion assay of the putative S. typhimurium 14028 hha lux reporter strain B9(2). Luminescence (A) and growth inhibition (B) in the presence of various antibiotics are shown. Paper discs containing rifampicin (R, 30 μg), chloramphenicol (Ch, 30 μg), minocycline (M, 10 μg), erythromycin (E, 15 μg) or tetracycline (T, 10 μg) are placed on top on the overlay. C) The line labeled 2.1 kb indicates the Sau3A fragment cloned into the lux reporter plasmid carried by strain B9(2).  4.3.  Possible mechanisms of rifampicin mediated up-modulation.  Transcription of the rifampicin up-modulated promoters was not stimulated by rifampicin in vitro (section 3.4). Growth studies, testing of truncated promoter fragments, and in vitro transcription experiments suggested that rifampicin mediated up-modulation may involve factors other than RNAP. Time course (section 3.1.4) and spent media (section 3.1.5) experiments suggested that these factors, if they exist, accumulated intracellularly because a 2-3 hour lag existed between rifampicin addition and up-modulation. Since spent media from rifampicin treated cells did not increase the magnitude or speed of rifampicin mediated up-modulation, this unidentified factor must not be a stable secreted factor. DNA mapping studies (section 3.2) 96  suggested that rifampicin mediated up-modulation of the three promoters studied may involve at least two discrete mechanisms since DNA upstream and downstream of the core promoter elements (-35, -10, +1) were required for rifampicin mediated up-modulation in different cases. In the case of STM3595 and traS, DNA downstream of the +1 was needed for up-modulation. Nucleotide sequences upstream of the -35 element at the spvA and traS promoters affected rifampicin mediated up-modulation. These sequences are unlikely to interact directly with RNAP at initiation, suggesting that other intracellular factors interact with these upstream or downstream DNA sequences. Motif searches using DNA sequences from the three upmodulated promoters yielded only one consensus sequence, CTGCGCAG (Fig 3.27B). This sequence is a palindrome and an exact match is present in the downstream region of STM3595, with one mismatch in the downstream region of traS (sites are shifted two bp) and with one mismatch in the upstream region of spvA (Fig. 3.29). However, it difficult to hypothesize why this motif would be present in two different locations. In vitro transcription studies were consistent with the need for additional factors inasmuch as RNAP and DNA alone were not sufficient to demonstrate rifampicin mediated up-modulation in vitro (section 3.4).  A Tn10d mutant that was defective in rifampicin mediated up-modulation for all promoters tested was not found, this may be because that such a mutant(s) is lethal, a protein factor is not involved or a more extensive search is required. Transcription factors that are essential in E. coli and are presumably essential in Salmonella include elongation factors such as NusA, NusG or  (Cardinale et al., 2008). GreA and GreB are not essential in E. coli but are essential for organisms such as Mycoplasma pneumonia (Stepanova et al., 2007).  97  4.3.1. Known regulatory factors which bind RNA polymerase. A strong consensus DNA binding site may not exist in the sequence of the rifampicin upmodulated promoters if the protein or small molecule factor which mediates this effect binds RNAP and not DNA. Two known regulatory factors which bind RNAP are DksA and guanosine tetraphosphate (ppGpp). DskA is a transcription factor, like GreA and GreB, which binds the RNAP secondary channel (Paul et al., 2004). ppGpp is a small molecule important for stringent response, a bacterial program responsible for decreasing rRNA synthesis and consequently ribosome production when the cell undergoes stress conditions including amino acid starvation (Srivatsan & Wang, 2008). Recent studies have elucidated the mechanism of action of ppGpp at promoters involved in cell physiology as well as promoters related to virulence. In conjunction with DskA, ppGpp directly inhibits rRNA promoters by reducing the half life of open complexes. ppGpp also potentiates the expression of genes involved in amino acid biosynthesis by enhancing an isomerization step leading to open complex formation (Paul et al., 2005). Upon nutrient starvation or entry into stationary phase, ppGpp promotes expression of two regulators of enterohaemorrhagic E. coli virulence, Ler and Pch (Nakanishi et al., 2006). In addition, DksA and Hfq (a RNA chaperone and post-transcriptional regulator) play a role in intercellular spread of Shigella flexneri in epithelial monolayers (Sharma & Payne, 2006).  In E.coli, two enzymes synthesize ppGpp, RelA and SpoT. A microarray analysis of a relAspoT mutant of Salmonella and the isogenic parent strain found ppGpp regulates both the invasion and intracellular virulence programs (Thompson et al., 2006). Rifampicin appears to down-modulate genes involved in invasion and up-modulate other virulence genes, including the spvRAB operon (Table 1.1). Transcription modulation in response to rifampicin differs from the  98  response to ppGpp, which up-regulates both spvRAB and invF (Pizarro-Cerda & Tedin, 2004). Although rifampicin and ppGpp seem to regulate the same groups of Salmonella genes, the gene expression patterns are different, suggesting that ppGpp is not involved in transcription modulation by rifampicin.  4.3.2. Elongation and termination factors. The fact that regions downstream of the +1 were required for rifampicin mediated up-modulation of traS and STM3595 suggested that, at these promoters, rifampicin may act after transcription initiation. Rifampicin mediated up-modulation may involve regulation of transcription elongation, attenuation, termination or anti-termination. Elongation factors in E. coli include GreA, GreB, NusA, NusG; termination factors include Mfd and while RfaH is an antitermination factor related to the regulation of virulence genes (Roberts et al., 2008). Some studies have examined expression patterns in elongation factor deficient or over-expressing mutants. In many cases, these factors bind directly to a subunit of RNAP and have no known preference for DNA sequence. The evidence for each of these factors as mediators of rifampicin mediated up-modulation is discussed below.  An E. coli microarray study employing greA, greB, greAgreB and greA overexpressing strains analyzed the genes affected by these mutations (Stepanova et al., 2007). I compared the S. typhimurium genes up-modulated by rifampicin to the transcription patterns in elongation factor mutants. The only GreA regulated gene found by Stepanova et al. (2007) that was also affected by rifampicin was talA. Transcription of talA is positively regulated by rifampicin and negatively regulated by GreA (Stepanova et al., 2007). Overall, the lack of correspondence  99  between genes subject to transcription modulation by rifampicin and loss of GreA and GreB is inconclusive. Complicating the analysis is the observation that many genes modulated by rifampicin in Salmonella are related to virulence and are not found in E. coli.  The functions of E. coli NusA, NusB, NusE and NusG proteins are quite complex and are seemingly contradictory as they have been shown to be involved in either or both antitermination and termination. Antitermination complexes are multi-component RNAP elongating complexes, which are resistant to pausing and termination (Mooney et al., 2009) and contain NusA, NusB, NusE and NusG proteins (Prasch et al., 2009). Antitermination complexes at ribosomal (rrn) operons are comprised of NusA, NusB, NusE (ribosomal protein S10) and NusG, while those formed at  bacteriophage genes also include the N protein (Prasch et al., 2009). When at  anti-termination complexes, NusA binds to nut sites (Prasch et al., 2009). These nut sites are comprised of highly conserved boxA, boxB and boxC sites (Prasch et al., 2009). The  boxA consensus is thought to be 5’ CGCUCUUA 3’ (Prasch et al., 2009). However, when NusA forms an anti-termination complex at rrn promoters, NusA specificity is determined by boxA, and sometimes a boxC sequence; the rrn boxA consensus sequence is 5’ UGCUCUUUA 3’ and as such differs from the  boxA (Prasch et al., 2009).  anti-termination complexes are resistant to both dependent and independent termination while rrn complexes are only resistant to dependent termination (Prasch et al., 2009).  GreA and GreB bind to RNAP in the secondary channel of RNAP (Vassylyeva et al., 2007) and NusG has been proposed to bind the ’ subunit of RNAP (Mooney et al., 2009). The specificity of DNA binding sites for the loading of these factors onto RNAP is not known. NusA binds to 100  the  subunit of RNAP, its known binding sites are discussed below (Yang et al., 2009). Mfd will terminate arrested elongating complexes (Roberts et al., 2008). Mfd may also bind to the  subunit of RNAP (Roberts et al., 2008), while  does not bind to a specific consensus sequence but binds to naked, untranslated RNA (without ribosomes) with a preference for nucleotide sequences that are C-rich and lacking secondary structure (Cardinale et al., 2008). E.coli and Salmonella RfaH bind to ops sites, 5' GGCGGTAG 3'; this sequence often occurs with a direct repeat and a conserved thymine residue two nucleotides down from the ops (operon polarity suppressor) site (Fig 4.3).  Sequences of rifampicin up-modulated promoters were examined for rrn boxA and ops sites, as these were the only known consensus sequences for possible elongation and termination factors. Moderate matches to the rrn boxA consensus sequence were found in traS (matching in 6/9 positions) and spvA (one matching in 7/9, two matching in 6/9 positions) transcripts. A moderate match to the ops sequence was present in the sequence of STM3595 which starts at +35 relative to the transcription start site (Fig 4.3). RfaH, the protein which binds to ops sites, is involved in production of the bacterial cell envelope, hemolysin toxin, F pilus, as well as, siiA expression (SPI-4, a rifampicin modulated gene, Table 1.1) (Bailey et al., 1997; Main-Hester et al., 2008). Previous studies have shown that a 6 bp alteration in the consensus from 5’ GGCGGTAGCGC 3’ to 5’ CCGGGATGC 3’ only had a 2-fold affect on proximal gene expression but a 55-fold change in more distal gene expression as measured by RT-PCR (Main-Hester et al., 2008). Given that such a large nucleotide alteration had such a small effect on proximal gene expression, it is reasonable to hypothesize that a RfaH homologue may be involved in rifampicin mediated up-modulation of STM3595 and also traS. However, this cannot be the only  101  mechanism since siiA is down-modulated by rifampicin (Table 1.1), while STM3595 is upmodulated by RfaH but has a weak match to consensus.  EC EC EC EC EC EC EC EC ST ST ST  rfaQ rfbB wza kpsM pHly152 hlyC LE2001 hlyC F traR J96 hlyC SL1344 siiA LT2 rfbB 14028 STM3595  CAGTATTCAGGTAGCTGTTGAGCCTGGGGCGGTAGCGTG CAGTGCTCTGGTAGCTGTTAAGCCAGGGGCGGTAGCGTG CAGTGTATTGGTAGCTAAAAAGCCAGGGGCGGTAGCGTG CAGTGTATTGGTAGCTGTTAAGCCAAGGGCGGTAGCGTA TCCCGGTTTACGGGTAGTTTCCGGAAGGGCGGTAGCATG CCGGTTGATACGGGTAATTTCCGGAAGGGCGGTAGCATG CGAGTGCCCTGTGCGTGAAAAGGGATGGGCGGTAGCGTG GCTGGTTGATGACTGTTAATTCCAGAAGGCGGTAGTCTG AAGCGTATTGGTAGCAGGAAGCCAAGGGCGGTAGCGTT CAGTGCACTGGTAGCTGATGAGCCAGGGGCGGTAGCGTG CTGTGGGCAAGGTAGTAGTTTTGACCTGGCGTTATTTTT  Figure 4.3. The ops sites of various E.coli and S. typhimurium genes. The ops site is in bold and the direct repeat is underlined. This figure is modified from other sources (Bailey et al., 1997; Main-Hester et al., 2008; Wang et al., 1998).  4.3.3. Transcription attenuation and intrinsic terminators. Rather than protein elongation or termination factors being involved, an intrinsic terminator or transcription attenuation may be mediating up-modulation of PSTM3595 and PtraS by rifampicin. Intrinsic terminator sequences are regions are characterized by a hairpin followed by a run of U residues (Roberts et al., 2008). A brief description of attenuation is described in section 1.4.3. To investigate this possibility, secondary structures of these two rifampicin up-modulated transcripts were analyzed using RNA folding software. Several RNA secondary structures were predicted when the first 163 nucleotides were analyzed. Secondary structures with similar features in both traS and STM3595 transcripts are shown (Fig. 4.4), but no candidate intrinsic terminators were identified. Interestingly, the consensus sequence (Fig 3.27B) observed in 102  rifampicin up-modulated promoters appeared to be base paired in a similar location in both STM3595 and traS secondary structures (Fig. 4.4B and C, respectively). Finer deletions than those shown in Figure 3.9 or nucleotide alterations would need to be constructed to determine the importance of the consensus nucleotides in forming secondary structure involved with rifampicin mediated up-modulation.  Since intrinsic terminators could not be identified from RNA secondary structure predictions using the first 163 nucleotides of the traS and STM3595 transcripts (Fig 4.4), mRNA sequences were specifically examined for transcription attenuator features such as inverted repeats. The REPuter program (Kurtz et al., 2001) was used to search for the inverted repeats within the transcript sequences that would be present in the stem loop of a intrinsic terminator. Sequences downstream of the inverted repeats were examined for runs of U nucleotides. Since only one inverted repeat with a run of U’s was found, this allowed for a shorter input sequence into the RNA folding software. The software predicted one possible intrinsic terminator which encoded a peptide from the shorter STM3595 transcript sequence. The small 13 codon ORF in the leader sequence of STM3595 is displayed (Fig 4.5). Although leader peptides are usually found upstream of intrinsic terminators, a ribosome binding site within the terminator sequence would work by a functionally similar mechanism. In the presence of rifampicin, RNAP would pause allowing the ribosome to catch up to RNAP, preventing secondary structure formation and thereby preventing termination. If rifampicin is not present, RNAP would not pause, secondary structure would form and terminate transcription. This occurs in attenutation of the pyrC gene, in which the stem of a leader region hairpin includes part of the pyrC ribosome binding site  103  Figure 4.4. Predicted RNA secondary structures of STM3595 and traS transcripts. Selected secondary structures of traS (A,C) and STM3595 (B,D) transcripts from +1 to +163 as predicted by Mfold (Zuker, 2003). The consensus sequence shown in Fig. 3.27 is boxed.  104  (Turnbough & Switzer, 2008). Nucleotide sequences could be substituted within the putative stem loop structure or within the UTP tract to test this hypothesis.  Figure 4.5. Predicted secondary structure of the nucleotides from +1 to +65 nucleotides. Numbering is relative to the +1 transcription start. Capital letters follow the IUPAC single letter code for amino acids and correspond to the amino acid predicted by accepted codon usage tables. The structure is as predicted by Mfold (Zuker, 2003).  4.4  Rifampicin resistance.  Four rifampicin responsive reporter fusions tested in a S. typhimurium rifampicin resistant background (mutant in the  subunit of RNAP) no longer displayed transcription modulation by rifampicin (Fig 3.2). This suggested that transcription modulation by rifampicin was mediated at the level of RNAP and not by rifampicin binding to another target. This is consistent with a previous study showing that rifampicin resistance abolished rifampicin mediated up-modulation of quorum sensing promoters from various bacterial genera (Goh et al., 2002). Although 105  different rifampicin resistant mutants have varying phenotypes related to transcription termination, temperature sensitivity, etc. (Jin & Gross, 1989), only one rifampicin resistant mutant was employed in this study, S. typhimurium R306. R306, according to E. coli numbering, has a H526Y mutation in the RNAP beta subunit which corresponds to the E. coli allele rpoB2. Interestingly, in E. coli, rpoB2 confers a 20-fold decrease in F' plasmid stability, temperature sensitivity, increased 5-fluorouridine sensitivity (5-fluorouridine resistant mutants have altered Km’s for ATP and UTP) and altered termination at the trp transcriptional terminator (Jin &  Gross, 1989; Yanofsky & Horn, 1981). Jin and Gross (1989) proposed that rifampicin resistant phenotypes may reflect structure and function relationships within different areas of the  subunit. Similarly, different rifampicin resistant mutants may have different degrees of transcription modulation by rifampicin reflecting the involvement in that region in transcription modulation by  rifampicin and the associated phenotypes, such as defects in transcription termination, may give hints to the mechanism of transcription modulation by rifampicin at that promoter.  4.5.  Differential responses of reporter strains in liquid versus sold media.  Differences in transcription modulation by rifampicin from strains grown on solid or in liquid media were observed. The control strain (STM2901 fusion, pGY21) was unresponsive to rifampicin in liquid culture (Fig. 3.5) but showed increased luminescence in response to rifampicin on solid media (Fig. 3.3). Differential expression in response to sub-MIC antibiotics between strains grown on solid or liquid media was shown for rifampicin mediated upmodulation of quorum sensing promoters (Goh et al., 2002). Transcriptome analysis comparing expression between cultures grown on solid or liquid cultures showed that up to 32 % of the S. typhimurium functional genome can be expressed differentially under the two conditions  106  (Wang et al., 2004). Differences in STM2901 expression in liquid and solid media are an example of culture methods having an impact on expression.  Disk diffusion plate assays have many advantages over liquid media assays. Most studies examine the effects of antibiotics in broth culture. Sub-MIC transcription effects can occur over a narrow concentration range. In disk diffusion assays, this is illustrated by a narrow light induction zone (Fig 3.4ii). The use of sub-optimal antibiotic concentrations in liquid media may explain why these seemingly ubiquitous effects on bacterial transcription have not been described previously in literature. Compared to assays in liquid medium that can only employ one concentration at a time, disk diffusion assays test a large concentration range.  4.6.  The response of rifampicin modulated promoters to various antibiotics.  A wide response range of promoters to antibiotics has been previously been demonstrated with five other reporter fusions and four different classes of antibiotics (Goh et al., 2002). The responsiveness of promoter reporter fusions to a diverse range of antibiotics seen in disk diffusion assays (section 3.1.3) may hint at the sensitivity of these reporters to perturbations in bacterial metabolism. The ability of antibiotics to elicit so many effects may also emphasize that antibiotics are not randomly chosen compounds but compounds preselected for bioactivity. Antibiotics are not like off the shelf chemicals or compound libraries; they have undergone extensive selection for bioactivity: by evolution in bacteria and in the laboratory when screened for growth inhibitory properties at high concentrations.  107  The observation that many antibiotics induce transcription responses in many promoters led to the suggestion that detection and classification of pharmaceutically active compounds may be possible by monitoring transcription of selected promoter clones (Goh et al., 2002). As mentioned in section 1.3, it has proposed that mode of action identification or even preliminary compound identification is possible by comparing the expression profiles induced by unknown compound to profiles induced by known compounds. The pattern of luminescence produced by the rifampicin responsive reporter fusions to various compounds shown in section 3.1.3 could be considered a less comprehensive version of the mode of action identification panels. These seven promoter-lux reporter fusions may be limited in their ability to identify different modes of action, as their response pattern may only be able to determine whether compounds have a similar mode of action to rifampicin. Any compound that elicits a pattern of expression similar to rifampicin (inducing up-modulation of traS, spvA, STM3595 and STM2901 in the wildtype background that are abolished in the R306 background and shows negative responses with fliA, flgK and invF reporters) could be predicted to have a similar mode of action as rifampicin. These reporters may not be as useful for identifying other kinds of antibiotics since promoter reporter fusions were not specifically chosen for their unique and specific responses to other classes of antibiotics. Mode of action panels often include at least one-promoter fusion that has been chosen for its characteristic response to each class of antibiotic. For example, some macrolide responders, some aminoglycoside responders, some beta-lactam responders, etc.  The response of rifampicin modulated promoter-lux reporters to other compounds was examined (section 3.1.3). This was carried out by testing the rifampicin responsive reporter fusions against various antimicrobials. Many promoters responded to other antimicrobials but did not elicit a  108  pattern similar to rifampicin; as such, the promoter-lux reporters showed a rifampicin specific response pattern. Three of the seven reporter fusions tested were weakly responsive to hydrogen peroxide and erythromycin. Six of the seven tested reporter fusions tested in section 3.1.3, including promoters for invasion and flagellar genes, were strongly induced by tetracycline. Studies with S. typhimurium DT104 also showed tetracycline induced transcription of invasion and flagellar genes and in addition showed increased invasion of HeLa cells (Weir et al., 2008). Therefore, if the appropriate reporter strains are used (i.e. reporters for promoters related to virulence), antibiotic induced responses on disk diffusion assays could be useful predictors for screening compounds which elicit phenotypic changes. Since all the antimicrobials elicited different patterns of light induction, it is likely that these compounds induce transcription modulation by distinct mechanisms.  4.7.  Biological relevance of transcription modulation by rifampicin.  Screening of a S. typhimurium promoter-reporter library revealed that rifampicin differentially affected the transcription of many S. typhimurium virulence genes depending on their involvement in two virulence programs (Yim et al., 2006). Virulence genes associated with intracellular growth in macrophages spvAB (Table 1.1) and SPI-2 genes (not shown) were upmodulated by rifampicin. Other virulence genes, not known to be associated with SPIs, such as traS and yijP were also up-modulated by rifampicin (Table 1.1). Genes involved in intestinal invasion associated with the type III secretion system encoded on SPI-1 and its secreted effectors (Teplitski et al., 2003) showed rifampicin mediated down-modulation. These genes included invF, sopA and sopB. Such differential regulation in distinct environments and between types/routes of infections is often observed and discussed relative to SPI-2 (Knodler et al., 2002;  109  Watson et al., 1999). Rifampicin may be mimicking natural cues that cause switching of virulence modes from penetration of epithelial cells to growth and survival in macrophages occurring when the salmonellae move from the intestine into macrophages during systemic infection.  The genes found to show rifampicin mediated down-modulation are similar to genes repressed by cationic microbial peptides (Bader et al., 2003) and bile (Prouty & Gunn, 2000; Prouty et al., 2004b). Rifampicin, bile and cationic microbial peptides cause transcriptional repression of the SPI-1 genes, its effectors and motility genes in S. typhimurium (Bader et al., 2003; Prouty & Gunn, 2000; Prouty et al., 2004b; Yim et al., 2006). Both rifampicin and cationic antimicrobial peptides (e.g. polymyxin B) are broadly used antibiotics while bile has been shown to have antibacterial activity (Gunn, 2000). The active component of bile for the induction of bile and antimicrobial resistance genes has been identified as deoxycholate (Prouty et al., 2004a). Other than their antibacterial properties, it is unclear what relationship there is between these three molecules that could account for their similar transcription profiles. Structurally diverse natural products cause potassium leakage and biofilm formation (Lopez et al., 2009). Analogously, but highly unlikely, is the hypothesis that polymyxin B and deoxycholate may be RNAP inhibitors. Alternatively, host produced antimicrobial peptides and bile would be naturally present in the intestine, but it is unlikely that rifampicin, which is produced by a soil microorganism, would be present in the intestine except during therapy. Rifampicin may be mimicking a signal in the gut or macrophage that causes Salmonella to switch from invasion to intracellular survival.  110  The function of the gene most highly up-modulated by rifampicin identified in my studies, STM3595, is unknown but several observations suggest it is important for S. typhimurium survival in macrophages. In the PubMed Conserved Domain Database the protein encoded by STM3595 is annotated as a putative acid phosphatase with a PAP2 domain (Marchler-Bauer et al., 2009). Proteins with PAP2 domains have been classified as non-specific acid phosphatases, enzymes which catalyze phosphomonoester hydrolysis with optimal activity in low pH conditions (Ishikawa et al., 2000). Other genes encoding S. typhimurium proteins with PAP2 domains include phoN, mig-13 (ybjG) and pgpB. Like STM3595, phoN and ybjG may be regulated by PhoP (Eguchi et al., 2004; Kasahara et al., 1991; Song et al., 2008). The PhoP/Q system is activated by low pH (and low Mg2+) and PhoP activated genes are expressed maximally in acidified phagosomes (Alpuche Aranda et al., 1992). PtraS and PspvA-reporter fusions carried in a phoP::cam mutant showed similar up-modulation as the isogenic parent strain while the strain carrying the STM3595 reporter fusion (pGY35) was no longer activated by rifampicin suggested that STM3595 is positively regulated by PhoP (not shown). Others have also found that STM3595 is positively regulated by PhoP (Song et al., 2008). Interestingly, several PAP2-containing proteins have been identified in a FACS screen for Salmonella genes expressed within macrophages; one such gene was mig-13 (Valdivia & Falkow, 1997). A gene located beside the rifampicin up-modulated gene traS, traT, was identified in the same screen (Valdivia & Falkow, 1997). In E. coli, TraS and TraT are important for surface exclusion during conjugation and TraT has been shown to be important for complement resistance (Sukupolvi & O'Connor, 1990). Although STM3595 is known to be positively regulated by the transcription factor PhoP, little is known about the direct regulation and function of this gene.  111  4.8.  Sub-MIC and its implications for virulence and motility.  In laboratory studies, MIC values are often used to determine whether a given bacterial isolate is resistant or sensitive to a given drug. The MIC value is dependent on many factors; it can vary greatly from organism to organism, as well as strain to strain. It is also highly dependent on media, aeration, inocula, incubation time, temperature or even the type of agar (Lorian, 1991). This is why defined growth conditions are always used in medical laboratories. The accepted variability in an MIC when compared between multiple medical laboratories is four-fold (average MIC two-fold) (Espinel-Ingroff et al., 2005; Huys et al., 2010). This becomes a problem when the breakpoint that determines whether an organism is to be classified as sensitive or resistant to a given antibiotic straddles these values.  When referring to antibiotic concentrations in the body, antibiotic concentration will greatly vary depending on many factors. Given the plethora of organisms and their respective diverse MICs, non-uniform distribution of drug and patient noncompliance to proper dosing regimens, a given antibiotic will certainly exist at sub-MICs for most antibiotic treatments. Although the MIC value determined in culture does not reflect the amount of drug required in a given tissue to inhibit bacterial growth, the term Peak/MIC is often seen in clinical literature. The peak value is the concentration of given drug reached in a given body compartment (Mouton et al., 2005). The peak for a given drug will vary from adult to adult depending on age, health and other genetic and environmental factors (this is one of the aspects that personalized medicine wishes to address). The Peak/MIC value has no specific meaning since MICs are determined in culture and cannot be determined in the body but may be useful for estimations in the body such as following. On average, for a 600 mg oral dose of rifampicin, the serum peak is 7-10g/ml and  112  the serum half life is approximately three hours (Sanofi-Aventis Canada Inc., 2010). The rifampicin MIC for 20 M. tuberculosis strains in 7H-9 broth with Tween 80 varied between 0.005 to 0.02 g/ml (Lorian, 1991). The concentration of rifampicin in the serum is many times over the M. tuberculosis MIC with Peak/MIC ratios ranging from 350 to 2000. Relative to the rifampicin MIC of S. typhimurium 14028 in LB, ~12 g/ml, this is in the sub-MIC range.  Since promoters for virulence and motility are selectively repressed by sub-MIC rifampicin, this may prove to be an advantage in treating salmonellosis. Although flagellae are not absolutely required for the virulence of S. typhimurium, their presence increases the ability to invade mammalian cells (Jones et al., 1992; Schmitt et al., 2001). In addition to the genes invF, fliA and flgK which were studied in this thesis, screening of the S. typhimurium DNA-lux fusion library suggested that rifampicin down-modulated other genes related to host cell invasion and motility such as sopA, sopB, genes of SPI-4 and the class II flagellar operon flhBA (Table 1.1). Multiple mechanisms exist for co-regulation of motility and invasion (Ellermeier & Slauch, 2003; Iyoda et al., 2001; Teplitski et al., 2003) and down-modulation of PfliA and PinvF by rifampicin would be expected to interfere with some of these pathways (Fig. 4.6). For example, reduced transcription of fliAZY would reduce both motility and virulence. As fliA encodes 28, reduced fliA transcription would decrease expression of the class III flagellar genes and fliA (it is autoregulated) (Kutsukake, 1997). In addition, FliZ positively regulates hilA and positively affects secretion of invasion proteins (Iyoda et al., 2001). HilA and InvF are activators of transcription of virulence effectors secreted by SPI-1 (Darwin & Miller, 2001) and HilA directly down regulates sopA, sopB, and siiA (first gene of SPI-4, STM4257) (Thijs et al., 2007). All of these genes belong to the group of promoters selectively down-regulated by rifampicin (Table 1.1).  113  Thus we predict that S. typhimurium invasion and motility would be repressed by sub-MIC rifampicin and may contribute to the clinical functionality of rifampicin.  Figure 4.6. A schematic of putative virulence and motility pathways affected by rifampicin.  A well-studied precedent for the use of antibiotics to repress virulence function rather than growth inhibition is the use of macrolide antibiotics (such as erythromycin and azithromycin) in diffuse panbronchiolitis and cystic fibrosis infections. Macrolides do not reduce the bacterial load but inhibit the expression of virulence determinants and also have immunomodulatory effects (for a comprehensive review, see (Tateda et al., 2007)). Briefly, in P. aeruginosa, production of toxins, pigments, alginate, pili and flagella are inhibited by macrolides (Tateda et al., 2007). As a consequence, cell adherence and biofilm production is also inhibited (Tateda et al., 2007). These effects are proposed to be mediated, at least in part, by inhibition of the quorum sensing system (Tateda et al., 2007). Antibiotics may be used to kill or inhibit bacterial growth and to decrease bacterial pathogenicity. In addition, such studies may provide guidance on antibiotics-pathogen combinations not to be used if pathogenicity is induced. An example of 114  a poor combination would be using tetracycline to treat Salmonella since tetracycline induces invasion and motility in S. typhimurium strain DT104 (Weir et al., 2008). Clearly, this analysis would contribute to a more rational and effective use of antibiotics.  4.9.  Concluding remarks.  Sub-MIC rifampicin down and up-modulates transcription of many genes in S. typhimurium. Genes involved in host cell intestinal invasion and motility are down-modulated, while those involved in carbon metabolism and other virulence functions appeared up-modulated. The mode of action of rifampicin at PfliA, PflgK, PinvF, PspvA, PSTM3595 and PtraS is much more complicated than initially anticipated. The transcription of some of the down-modulated promoters is hypersensitive to rifampicin, causing transcription to be down-modulated by rifampicin. Other mechanisms may cause further down-modulation by rifampicin. Rifampicin mediated upmodulation may involve intracellular factors and transcription attenuation. It is unclear whether common factors mediated both up-modulation and down-modulation by rifampicin but it is clear several mechanisms are involved.  When any given promoter or set of promoters is induced or repressed by a given stimulus, one mechanism, such as a specific transcription factor, is often studied. From this, one may have the impression that gene regulation is simple. However, it is clear that multiple levels of regulation exist for expression of a given gene. A gene often has several promoters; a second promoter may be internal to an operon and correspond to a shorter transcript. Each promoter can be transcribed by a different set of sigma factors and transcription factors. This allows the integration of many different environmental stimuli and flexibility in the sets of genes transcribed in response to a  115  given stimulus. Regulation often exists at the level of transcription initiation, but can also occur during transcription elongation, termination or post-transcriptionally. Like any other group of promoters, modulation of this group of rifampicin responsive promoters has multiple levels.  Most antimicrobials are microbially produced, biologically active small molecules. They represent a fraction of the small molecules produced by microbes. Decades of work and billions of dollars worth of discovery, chemistry and testing were employed to find useful compounds to inhibit growth of other microbes. Given that antibiotics are not randomly selected molecules but molecules selected by evolution and selected again by pharmaceutical companies, perhaps it is not surprising that so many antibiotics have a wide variety of effects on transcription and consequently phenotype. They are used clinically at concentrations much higher than those likely to occur in the environment. It has long been thought that antibiotics are agents of intermicrobial warfare. One microbe will use these molecules to kill or inhibit the growth of competing microbes living in the same environmental niche. Since both functionalities of antibiotics have been shown in the laboratory, surely both are possible in nature. Studying the sub-MIC properties of antibiotics and other small molecules will help to understand how they affect gene expression and will allow a more detailed exploration of their potential as therapeutic agents, antimicrobial or otherwise. These studies may also elucidate the “natural” function of antibiotics: modulatory signaling molecules, weapons of inter-microbial competition or both!  116  References Adachi, J. A. & DuPont, H. L. (2006). Rifaximin: a novel nonabsorbed rifamycin for gastrointestinal disorders. Clin Infect Dis 42, 541-547. Alexieva, Z., Duvall, E. J., Ambulos, N. P., Jr., Kim, U. J. & Lovett, P. S. (1988). Chloramphenicol induction of cat-86 requires ribosome stalling at a specific site in the leader. Proc Natl Acad Sci U S A 85, 3057-3061. Alpuche Aranda, C. M., Swanson, J. A., Loomis, W. P. & Miller, S. I. (1992). Salmonella typhimurium activates virulence gene transcription within acidified macrophage phagosomes. Proc Natl Acad Sci U S A 89, 10079-10083. Artsimovitch, I., Vassylyeva, M. N., Svetlov, D. & other authors (2005). Allosteric modulation of the RNA polymerase catalytic reaction is an essential component of transcription control by rifamycins. Cell 122, 351-363. Bader, M. W., Navarre, W. W., Shiau, W., Nikaido, H., Frye, J. G., McClelland, M., Fang, F. C. & Miller, S. I. (2003). Regulation of Salmonella typhimurium virulence gene expression by cationic antimicrobial peptides. Mol Microbiol 50, 219-230. Bailey, M. J., Hughes, C. & Koronakis, V. (1997). RfaH and the ops element, components of a novel system controlling bacterial transcription elongation. Mol Microbiol 26, 845-851. Bailey, T. L. & Elkan, C. (1994). Fitting a mixture model by expectation maximization to discover motifs in biopolymers. Proc Int Conf Intell Syst Mol Biol 2, 28-36. Bajaj, V., Lucas, R. L., Hwang, C. & Lee, C. A. (1996). Co-ordinate regulation of Salmonella typhimurium invasion genes by environmental and regulatory factors is mediated by control of hilA expression. Mol Microbiol 22, 703-714. Bang, I. S., Frye, J. G., McClelland, M., Velayudhan, J. & Fang, F. C. (2005). Alternative sigma factor interactions in Salmonella: sigma and sigma promote antioxidant defences by enhancing sigma levels. Mol Microbiol 56, 811-823. Bentley, R. & Bennett, J. W. (2009). Name that antibiotic. In SIM News, pp. 4-11: Society for Industrial Microbiology. Bentley, S. D., Chater, K. F., Cerdeno-Tarraga, A. M. & other authors (2002). Complete genome sequence of the model actinomycete Streptomyces coelicolor A3(2). Nature 417, 141147. Bernardo, K., Pakulat, N., Fleer, S., Schnaith, A., Utermohlen, O., Krut, O., Muller, S. & Kronke, M. (2004). Subinhibitory concentrations of linezolid reduce Staphylococcus aureus virulence factor expression. Antimicrob Agents Chemother 48, 546-444.  117  Bjarnason, J., Southward, C. M. & Surette, M. G. (2003). Genomic profiling of ironresponsive genes in Salmonella enterica serovar typhimurium by high-throughput screening of a random promoter library. J Bacteriol 185, 4973-4982. Blatter, E. E., Ross, W., Tang, H., Gourse, R. L. & Ebright, R. H. (1994). Domain organization of RNA polymerase alpha subunit: C-terminal 85 amino acids constitute a domain capable of dimerization and DNA binding. Cell 78, 889-896. Blondel, C. J., Jimenez, J. C., Contreras, I. & Santiviago, C. A. (2009). Comparative genomic analysis uncovers 3 novel loci encoding type six secretion systems differentially distributed in Salmonella serotypes. BMC Genomics 10, 354. Borukhov, S., Lee, J. & Laptenko, O. (2005). Bacterial transcription elongation factors: new insights into molecular mechanism of action. Molecular microbiology 55, 1315-1324. Browning, D. F. & Busby, S. J. (2004). The regulation of bacterial transcription initiation. Nat Rev Microbiol 2, 57-65. Bryskier, A. (2005). Antimicrobial agents : antibacterials and antifungals. Washington, D.C.: ASM Press. Buck, M., Gallegos, M. T., Studholme, D. J., Guo, Y. & Gralla, J. D. (2000). The bacterial enhancer-dependent sigma(54) (sigma(N)) transcription factor. J Bacteriol 182, 4129-4136. Burns, C. M., Richardson, L. V. & Richardson, J. P. (1998). Combinatorial effects of NusA and NusG on transcription elongation and Rho-dependent termination in Escherichia coli. J Mol Biol 278, 307-316. Campbell, E. A., Korzheva, N., Mustaev, A., Murakami, K., Nair, S., Goldfarb, A. & Darst, S. A. (2001). Structural mechanism for rifampicin inhibition of bacterial rna polymerase. Cell 104, 901-912. Canadian Pharmaceutical Association (2008). Compendium of pharmaceuticals and specialties, 9th edn. Ottawa [etc.]: Canadian Pharmaceutical Association. Cardinale, C. J., Washburn, R. S., Tadigotla, V. R., Brown, L. M., Gottesman, M. E. & Nudler, E. (2008). Termination factor Rho and its cofactors NusA and NusG silence foreign DNA in E. coli. Science 320, 935-938. Carpousis, A. J. & Gralla, J. D. (1985). Interaction of RNA polymerase with lacUV5 promoter DNA during mRNA initiation and elongation. Footprinting, methylation, and rifampicinsensitivity changes accompanying transcription initiation. J Mol Biol 183, 165-177. Chadsey, M. S., Karlinsey, J. E. & Hughes, K. T. (1998). The flagellar anti-sigma factor FlgM actively dissociates Salmonella typhimurium sigma28 RNA polymerase holoenzyme. Genes & development 12, 3123-3136. 118  Chadwick, D. & Whelan, J. (1992). Secondary Metabolites:Their Function and Evolution. In CIBA Foundation Symposium. Chichester, UK: Wiley. Cheung, K. J., Badarinarayana, V., Selinger, D. W., Janse, D. & Church, G. M. (2003). A Microarray-Based Antibiotic Screen Identifies a Regulatory Role for Supercoiling in the Osmotic Stress Response of Escherichia coli. Genome Res 13, 206-215. Choi, J., Shin, D. & Ryu, S. (2007). Implication of quorum sensing in Salmonella enterica serovar typhimurium virulence: the luxS gene is necessary for expression of genes in pathogenicity island 1. Infect Immun 75, 4885-4890. Chu, C., Hong, S. F., Tsai, C., Lin, W. S., Liu, T. P. & Ou, J. T. (1999). Comparative physical and genetic maps of the virulence plasmids of Salmonella enterica serovars typhimurium, enteritidis, choleraesuis, and dublin. Infect Immun 67, 2611-2614. Crooks, G. E., Hon, G., Chandonia, J. M. & Brenner, S. E. (2004). WebLogo: a sequence logo generator. Genome Res 14, 1188-1190. Darwin, K. H. & Miller, V. L. (2001). Type III secretion chaperone-dependent regulation: activation of virulence genes by SicA and InvF in Salmonella typhimurium. Embo J 20, 18501862. Davies, J., Spiegelman, G. B. & Yim, G. (2006). The world of subinhibitory antibiotic concentrations. Curr Opin Microbiol 9, 445-453. de Boer, H. A., Comstock, L. J. & Vasser, M. (1983). The tac promoter: a functional hybrid derived from the trp and lac promoters. Proc Natl Acad Sci U S A 80, 21-25. Demain, A. L. & Fang, A. (2000). In History of Modern Biotechnology, pp. 2-39. Edited by A. Fiechter. Berlin: Springer. DiGate, R. J. & Marians, K. J. (1988). Identification of a potent decatenating enzyme from Escherichia coli. J Biol Chem 263, 13366-13373. Eguchi, Y., Okada, T., Minagawa, S., Oshima, T., Mori, H., Yamamoto, K., Ishihama, A. & Utsumi, R. (2004). Signal transduction cascade between EvgA/EvgS and PhoP/PhoQ twocomponent systems of Escherichia coli. J Bacteriol 186, 3006-3014. Eichelberg, K. & Galan, J. E. (1999). Differential regulation of Salmonella typhimurium type III secreted proteins by pathogenicity island 1 (SPI-1)-encoded transcriptional activators InvF and hilA. Infect Immun 67, 4099-4105. Ellermeier, C. D. & Slauch, J. M. (2003). RtsA and RtsB coordinately regulate expression of the invasion and flagellar genes in Salmonella enterica serovar Typhimurium. J Bacteriol 185, 5096-5108. 119  Espinel-Ingroff, A., Barchiesi, F., Cuenca-Estrella, M., Pfaller, M. A., Rinaldi, M., Rodriguez-Tudela, J. L. & Verweij, P. E. (2005). International and multicenter comparison of EUCAST and CLSI M27-A2 broth microdilution methods for testing susceptibilities of Candida spp. to fluconazole, itraconazole, posaconazole, and voriconazole. J Clin Microbiol 43, 38843889. Evers, S., Di Padova, K., Meyer, M., Langen, H., Fountoulakis, M., Keck, W. & Gray, C. P. (2001). Mechanism-related changes in the gene transcription and protein synthesis patterns of Haemophilus influenzae after treatment with transcriptional and translational inhibitors. Proteomics 1, 522-544. Feklistov, A., Mekler, V., Jiang, Q., Westblade, L. F., Irschik, H., Jansen, R., Mustaev, A., Darst, S. A. & Ebright, R. H. (2008). Rifamycins do not function by allosteric modulation of binding of Mg2+ to the RNA polymerase active center. Proc Natl Acad Sci U S A 105, 1482014825. Feng, X., Walthers, D., Oropeza, R. & Kenney, L. J. (2004). The response regulator SsrB activates transcription and binds to a region overlapping OmpR binding sites at Salmonella pathogenicity island 2. Mol Microbiol 54, 823-835. Floss, H. G. & Yu, T. W. (2005). Rifamycin-mode of action, resistance, and biosynthesis. Chem Rev 105, 621-632. Freiberg, C. & Brotz-Oesterhelt, H. (2005). Functional genomics in antibacterial drug discovery. Drug Discov Today 10, 927-935. Frith, M. C., Saunders, N. F., Kobe, B. & Bailey, T. L. (2008). Discovering sequence motifs with arbitrary insertions and deletions. PLoS Comput Biol 4, e1000071. Fuller, F. (1982). A family of cloning vectors containing the lacUV5 promoter. Gene 19, 43-54. Gaal, T., Bartlett, M. S., Ross, W., Turnbough, C. L., Jr. & Gourse, R. L. (1997). Transcription regulation by initiating NTP concentration: rRNA synthesis in bacteria. Science 278, 2092-2097. Galan, J. E. (1999). Interaction of Salmonella with host cells through the centisome 63 type III secretion system. Curr Opin Microbiol 2, 46-50. Gemmell, C. G. & Ford, C. W. (2002). Virulence factor expression by Gram-positive cocci exposed to subinhibitory concentrations of linezolid. J Antimicrob Chemother 50, 665-672. Goh, E. B., Yim, G., Tsui, W., McClure, J., Surette, M. G. & Davies, J. (2002). Transcriptional modulation of bacterial gene expression by subinhibitory concentrations of antibiotics. Proc Natl Acad Sci U S A 99, 17025-17030.  120  Gourse, R. L., Ross, W. & Gaal, T. (2000). UPs and downs in bacterial transcription initiation: the role of the alpha subunit of RNA polymerase in promoter recognition. Mol Microbiol 37, 687-695. Gralla, J. D. & Collado-Vides, J. (1996). Organization and Function of Transcription Regulatory Elements. In Eschericia coli and Salmonella: Cellular and Molecular Biology. Edited by F. C. Neidhardt & R. Curtiss. Washington, D.C.: ASM Press. Groisman, E. A. & Ochman, H. (1996). Pathogenicity islands: bacterial evolution in quantum leaps. Cell 87, 791-794. Guerin, E., Cambray, G., Sanchez-Alberola, N. & other authors (2009). The SOS response controls integron recombination. Science 324, 1034. Gulig, P. A., Danbara, H., Guiney, D. G., Lax, A. J., Norel, F. & Rhen, M. (1993). Molecular analysis of spv virulence genes of the Salmonella virulence plasmids. Mol Microbiol 7, 825-830. Gunn, J. S. (2000). Mechanisms of bacterial resistance and response to bile. Microbes Infect 2, 907-913. Ham, L. M., Cram, D. & Skurray, R. (1989). Transcriptional analysis of the F plasmid surface exclusion region: mapping of traS, traT, and traD transcripts. Plasmid 21, 1-8. Haugen, S. P., Ross, W. & Gourse, R. L. (2008). Advances in bacterial promoter recognition and its control by factors that do not bind DNA. Nat Rev Microbiol 6, 507-519. Hensel, M., Shea, J. E., Waterman, S. R. & other authors (1998). Genes encoding putative effector proteins of the type III secretion system of Salmonella pathogenicity island 2 are required for bacterial virulence and proliferation in macrophages. Mol Microbiol 30, 163-174. Hensel, M. (2004). Evolution of pathogenicity islands of Salmonella enterica. Int J Med Microbiol 294, 95-102. Herbert, S., Barry, P. & Novick, R. P. (2001). Subinhibitory clindamycin differentially inhibits transcription of exoprotein genes in Staphylococcus aureus. Infect Immun 69, 2996-3003. Herold, S., Siebert, J., Huber, A. & Schmidt, H. (2005). Global expression of prophage genes in Escherichia coli O157:H7 strain EDL933 in response to norfloxacin. Antimicrob Agents Chemother 49, 931-944. Hoffman, L. R., D'Argenio, D. A., MacCoss, M. J., Zhang, Z., Jones, R. A. & Miller, S. I. (2005). Aminoglycoside antibiotics induce bacterial biofilm formation. Nature 436, 1171-1175. Holden, D. W. (2002). Trafficking of the Salmonella vacuole in macrophages. Traffic 3, 161169.  121  Hughes, K. T., Gillen, K. L., Semon, M. J. & Karlinsey, J. E. (1993). Sensing structural intermediates in bacterial flagellar assembly by export of a negative regulator. Science 262, 1277-1280. Hutter, B., Schaab, C., Albrecht, S. & other authors (2004). Prediction of mechanisms of action of antibacterial compounds by gene expression profiling. Antimicrob Agents Chemother 48, 2838-2844. Huys, G., D'Haene, K., Cnockaert, M. & other authors (2010). Intra- and interlaboratory performances of two commercial antimicrobial susceptibility testing methods for bifidobacteria and nonenterococcal lactic acid bacteria. Antimicrob Agents Chemother 54, 2567-2574. Ide, N., Ikebe, T. & Kutsukake, K. (1999). Reevaluation of the promoter structure of the class 3 flagellar operons of Escherichia coli and Salmonella. Genes Genet Syst 74, 113-116. Ikebe, T., Iyoda, S. & Kutsukake, K. (1999). Structure and expression of the fliA operon of Salmonella typhimurium. Microbiology 145 ( Pt 6), 1389-1396. Ishikawa, K., Mihara, Y., Gondoh, K., Suzuki, E. & Asano, Y. (2000). X-ray structures of a novel acid phosphatase from Escherichia blattae and its complex with the transition-state analog molybdate. EMBO J 19, 2412-2423. Iyoda, S., Kamidoi, T., Hirose, K., Kutsukake, K. & Watanabe, H. (2001). A flagellar gene fliZ regulates the expression of invasion genes and virulence phenotype in Salmonella enterica serovar Typhimurium. Microb Pathog 30, 81-90. Jin, D. J. & Gross, C. A. (1988). Mapping and sequencing of mutations in the Escherichia coli rpoB gene that lead to rifampicin resistance. J Mol Biol 202, 45-58. Jin, D. J., Walter, W. A. & Gross, C. A. (1988). Characterization of the termination phenotypes of rifampicin-resistant mutants. J Mol Biol 202, 245-253. Jin, D. J. & Gross, C. A. (1989). Characterization of the pleiotropic phenotypes of rifampinresistant rpoB mutants of Escherichia coli. J Bacteriol 171, 5229-5231. Jones, B. D., Lee, C. A. & Falkow, S. (1992). Invasion by Salmonella typhimurium is affected by the direction of flagellar rotation. Infect Immun 60, 2475-2480. Kasahara, M., Nakata, A. & Shinagawa, H. (1991). Molecular analysis of the Salmonella typhimurium phoN gene, which encodes nonspecific acid phosphatase. Journal of bacteriology 173, 6760-6765. Kato, J., Nishimura, Y., Imamura, R., Niki, H., Hiraga, S. & Suzuki, H. (1990). New topoisomerase essential for chromosome segregation in E. coli. Cell 63, 393-404.  122  Kimmitt, P. T., Harwood, C. R. & Barer, M. R. (2000). Toxin gene expression by shiga toxinproducing Escherichia coli: the role of antibiotics and the bacterial SOS response. Emerg Infect Dis 6, 458-465. Knodler, L. A., Celli, J., Hardt, W. D., Vallance, B. A., Yip, C. & Finlay, B. B. (2002). Salmonella effectors within a single pathogenicity island are differentially expressed and translocated by separate type III secretion systems. Mol Microbiol 43, 1089-1103. Kohanski, M. A., Dwyer, D. J., Hayete, B., Lawrence, C. A. & Collins, J. J. (2007). A common mechanism of cellular death induced by bactericidal antibiotics. Cell 130, 797-810. Kolter, R. & Yanofsky, C. (1982). Attenuation in amino acid biosynthetic operons. Annu Rev Genet 16, 113-134. Korzheva, N., Mustaev, A., Kozlov, M., Malhotra, A., Nikiforov, V., Goldfarb, A. & Darst, S. A. (2000). A structural model of transcription elongation. Science 289, 619-625. Krasny, L. & Gourse, R. L. (2004). An alternative strategy for bacterial ribosome synthesis: Bacillus subtilis rRNA transcription regulation. Embo J 23, 4473-4483. Kurtz, S., Choudhuri, J. V., Ohlebusch, E., Schleiermacher, C., Stoye, J. & Giegerich, R. (2001). REPuter: the manifold applications of repeat analysis on a genomic scale. Nucleic Acids Res 29, 4633-4642. Kutsukake, K., Ohya, Y. & Iino, T. (1990). Transcriptional analysis of the flagellar regulon of Salmonella typhimurium. J Bacteriol 172, 741-747. Kutsukake, K. & Iino, T. (1994). Role of the FliA-FlgM regulatory system on the transcriptional control of the flagellar regulon and flagellar formation in Salmonella typhimurium. J Bacteriol 176, 3598-3605. Kutsukake, K. & Ide, N. (1995). Transcriptional analysis of the flgK and fliD operons of Salmonella typhimurium which encode flagellar hook-associated proteins. Mol Gen Genet 247, 275-281. Kutsukake, K. (1997). Autogenous and global control of the flagellar master operon, flhD, in Salmonella typhimurium. Mol Gen Genet 254, 440-448. Latifi, A., Winson, M. K., Foglino, M., Bycroft, B. W., Stewart, G. S., Lazdunski, A. & Williams, P. (1995). Multiple homologues of LuxR and LuxI control expression of virulence determinants and secondary metabolites through quorum sensing in Pseudomonas aeruginosa PAO1. Mol Microbiol 17, 333-343. Lewis, P. J., Doherty, G. P. & Clarke, J. (2008). Transcription factor dynamics. Microbiology 154, 1837-1844.  123  Libby, S. J., Lesnick, M., Hasegawa, P., Weidenhammer, E. & Guiney, D. G. (2000). The Salmonella virulence plasmid spv genes are required for cytopathology in human monocytederived macrophages. Cell Microbiol 2, 49-58. Liu, X. & Matsumura, P. (1994). The FlhD/FlhC complex, a transcriptional activator of the Escherichia coli flagellar class II operons. J Bacteriol 176, 7345-7351. Lopez, D., Fischbach, M. A., Chu, F., Losick, R. & Kolter, R. (2009). Structurally diverse natural products that cause potassium leakage trigger multicellularity in Bacillus subtilis. Proc Natl Acad Sci U S A 106, 280-285. Lorian, V. (1991). Antibiotics in laboratory medicine, 3rd edn. Baltimore: Williams & Wilkins. Lostroh, C. P., Bajaj, V. & Lee, C. A. (2000). The cis requirements for transcriptional activation by HilA, a virulence determinant encoded on SPI-1. Mol Microbiol 37, 300-315. Main-Hester, K. L., Colpitts, K. M., Thomas, G. A., Fang, F. C. & Libby, S. J. (2008). Coordinate regulation of Salmonella pathogenicity island 1 (SPI1) and SPI4 in Salmonella enterica serovar Typhimurium. Infect Immun 76, 1024-1035. Marchler-Bauer, A., Anderson, J. B., Chitsaz, F. & other authors (2009). CDD: specific functional annotation with the Conserved Domain Database. Nucleic Acids Res 37, D205-210. Matsui, H., Bacot, C. M., Garlington, W. A., Doyle, T. J., Roberts, S. & Gulig, P. A. (2001). Virulence plasmid-borne spvB and spvC genes can replace the 90-kilobase plasmid in conferring virulence to Salmonella enterica serovar Typhimurium in subcutaneously inoculated mice. J Bacteriol 183, 4652-4658. Matsushiro, A., Sato, K., Miyamoto, H., Yamamura, T. & Honda, T. (1999). Induction of prophages of enterohemorrhagic Escherichia coli O157:H7 with norfloxacin. J Bacteriol 181, 2257-2260. Mayford, M. & Weisblum, B. (1989). ermC leader peptide. Amino acid sequence critical for induction by translational attenuation. J Mol Biol 206, 69-79. McClure, W. R. & Cech, C. L. (1978). On the mechanism of rifampicin inhibition of RNA synthesis. J Biol Chem 253, 8949-8956. McCue, L., Thompson, W., Carmack, C., Ryan, M. P., Liu, J. S., Derbyshire, V. & Lawrence, C. E. (2001). Phylogenetic footprinting of transcription factor binding sites in proteobacterial genomes. Nucleic Acids Res 29, 774-782. Meighen, E. A. (1993). Bacterial bioluminescence: organization, regulation, and application of the lux genes. Faseb J 7, 1016-1022.  124  Mekler, V., Pavlova, O. & Severinov, K. (2011). Interaction of Escherichia coli RNA Polymerase sigma70 subunit with promoter elements in the context of free sigma70, RNA polymerase holoenzyme, and the beta'-sigma70 complex. J Biol Chem 286, 270-279. Mendoza-Vargas, A., Olvera, L., Olvera, M. & other authors (2009). Genome-wide identification of transcription start sites, promoters and transcription factor binding sites in E. coli. PLoS One 4, e7526. Mesak, L. R., Qi, S., Villanueva, I., Miao, V. & Davies, J. (2010). Staphylococcus aureus promoter-lux reporters for drug discovery. J Antibiot (Tokyo) 63, 492-498. Moazed, D. & Noller, H. F. (1987). Interaction of antibiotics with functional sites in 16S ribosomal RNA. Nature 327, 389-394. Mooney, R. A., Schweimer, K., Rosch, P., Gottesman, M. & Landick, R. (2009). Two structurally independent domains of E. coli NusG create regulatory plasticity via distinct interactions with RNA polymerase and regulators. J Mol Biol 391, 341-358. Musser, J. M. (1995). Antimicrobial agent resistance in mycobacteria: molecular genetic insights. Clinical microbiology reviews 8, 496-514. Nakanishi, N., Abe, H., Ogura, Y., Hayashi, T., Tashiro, K., Kuhara, S., Sugimoto, N. & Tobe, T. (2006). ppGpp with DksA controls gene expression in the locus of enterocyte effacement (LEE) pathogenicity island of enterohaemorrhagic Escherichia coli through activation of two virulence regulatory genes. Mol Microbiol 61, 194-205. Nichols, B. P., Shafiq, O. & Meiners, V. (1998). Sequence analysis of Tn10 insertion sites in a collection of Escherichia coli strains used for genetic mapping and strain construction. J Bacteriol 180, 6408-6411. Novick, R. P. & Geisinger, E. (2008). Quorum sensing in staphylococci. Annu Rev Genet 42, 541-564. Nudler, E. (2009). RNA polymerase active center: the molecular engine of transcription. Annu Rev Biochem 78, 335-361. Ochman, H., Gerber, A. S. & Hartl, D. L. (1988). Genetic applications of an inverse polymerase chain reaction. Genetics 120, 621-623. Ohnishi, K., Kutsukake, K., Suzuki, H. & Lino, T. (1992). A novel transcriptional regulation mechanism in the flagellar regulon of Salmonella typhimurium: an antisigma factor inhibits the activity of the flagellum-specific sigma factor, sigma F. Mol Microbiol 6, 3149-3157. Paget, M. S. & Helmann, J. D. (2003). The sigma70 family of sigma factors. Genome Biol 4, 203.  125  Parsek, M. R. & Greenberg, E. P. (2000). Acyl-homoserine lactone quorum sensing in gramnegative bacteria: a signaling mechanism involved in associations with higher organisms. Proc Natl Acad Sci U S A 97, 8789-8793. Paul, B. J., Barker, M. M., Ross, W., Schneider, D. A., Webb, C., Foster, J. W. & Gourse, R. L. (2004). DksA: a critical component of the transcription initiation machinery that potentiates the regulation of rRNA promoters by ppGpp and the initiating NTP. Cell 118, 311322. Paul, B. J., Berkmen, M. B. & Gourse, R. L. (2005). DksA potentiates direct activation of amino acid promoters by ppGpp. Proc Natl Acad Sci U S A 102, 7823-7828. Pfeffer, S. R., Stahl, S. J. & Chamberlin, M. J. (1977). Binding of Escherichia coli RNA polymerase to T7 DNA. Displacement of holoenzyme from promoter complexes by heparin. J Biol Chem 252, 5403-5407. Pizarro-Cerda, J. & Tedin, K. (2004). The bacterial signal molecule, ppGpp, regulates Salmonella virulence gene expression. Mol Microbiol 52, 1827-1844. Prasch, S., Jurk, M., Washburn, R. S., Gottesman, M. E., Wohrl, B. M. & Rosch, P. (2009). RNA-binding specificity of E. coli NusA. Nucleic Acids Res 37, 4736-4742. Price-Whelan, A., Dietrich, L. E. & Newman, D. K. (2006). Rethinking 'secondary' metabolism: physiological roles for phenazine antibiotics. Nat Chem Biol 2, 71-78. Prouty, A. M. & Gunn, J. S. (2000). Salmonella enterica serovar typhimurium invasion is repressed in the presence of bile. Infect Immun 68, 6763-6769. Prouty, A. M., Brodsky, I. E., Falkow, S. & Gunn, J. S. (2004a). Bile-salt-mediated induction of antimicrobial and bile resistance in Salmonella typhimurium. Microbiology 150, 775-783. Prouty, A. M., Brodsky, I. E., Manos, J., Belas, R., Falkow, S. & Gunn, J. S. (2004b). Transcriptional regulation of Salmonella enterica serovar Typhimurium genes by bile. FEMS Immunol Med Microbiol 41, 177-185. Purohit, P. & Stern, S. (1994). Interactions of a small RNA with antibiotic and RNA ligands of the 30S subunit. Nature 370, 659-662. Raffatellu, M., Wilson, R. P., Chessa, D., Andrews-Polymenis, H., Tran, Q. T., Lawhon, S., Khare, S., Adams, L. G. & Baumler, A. J. (2005). SipA, SopA, SopB, SopD, and SopE2 contribute to Salmonella enterica serotype typhimurium invasion of epithelial cells. Infect Immun 73, 146-154. Record, M. T., Jr., Reznikoff, W. S., Craig, M. L., McQuade, K. L. & Schlax, P. J. (1996). Escherichia coli RNA polymerase (Esigma70), promoters, and the kinetics of the steps of  126  transcription initiation. In Eschericia coli and Salmonella: Cellular and Molecular Biology. Edited by F. C. Neidhardt & R. Curtiss. Washington, D.C.: ASM Press. Ring, B. Z., Yarnell, W. S. & Roberts, J. W. (1996). Function of E. coli RNA polymerase sigma factor sigma 70 in promoter-proximal pausing. Cell 86, 485-493. Roberts, J. W., Shankar, S. & Filter, J. J. (2008). RNA polymerase elongation factors. Annu Rev Microbiol 62, 211-233. Rodriguez, C. R., Schechter, L. M. & Lee, C. A. (2002). Detection and characterization of the S. typhimurium HilA protein. BMC Microbiol 2, 31. Sanofi-Aventis Canada Inc. (2010). Product Monograph: Rifadin. Laval, Quebec. Schaubach, O. L. & Dombroski, A. J. (1999). Transcription initiation at the flagellin promoter by RNA polymerase carrying sigma28 from Salmonella typhimurium. J Biol Chem 274, 87578763. Schmitt, C. K., Ikeda, J. S., Darnell, S. C., Watson, P. R., Bispham, J., Wallis, T. S., Weinstein, D. L., Metcalf, E. S. & O'Brien, A. D. (2001). Absence of all components of the flagellar export and synthesis machinery differentially alters virulence of Salmonella enterica serovar Typhimurium in models of typhoid fever, survival in macrophages, tissue culture invasiveness, and calf enterocolitis. Infect Immun 69, 5619-5625. Schneider, T. D., Stormo, G. D., Gold, L. & Ehrenfeucht, A. (1986). Information content of binding sites on nucleotide sequences. J Mol Biol 188, 415-431. Schulz, W. & Zillig, W. (1981). Rifampicin inhibition of RNA synthesis by destabilisation of DNA-RNA polymerase-oligonucleotide-complexes. Nucleic Acids Res 9, 6889-6906. Seredick, S. D. & Spiegelman, G. B. (2007). Bacillus subtilis RNA polymerase recruits the transcription factor Spo0A approximately P to stabilize a closed complex during transcription initiation. Journal of molecular biology 366, 19-35. Sharma, A. K. & Payne, S. M. (2006). Induction of expression of hfq by DksA is essential for Shigella flexneri virulence. Mol Microbiol 62, 469-479. Sheehan, B. J. & Dorman, C. J. (1998). In vivo analysis of the interactions of the LysR-like regulator SpvR with the operator sequences of the spvA and spvR virulence genes of Salmonella typhimurium. Mol Microbiol 30, 91-105. Song, H., Kong, W., Weatherspoon, N., Qin, G., Tyler, W., Turk, J., Curtiss, R., 3rd & Shi, Y. (2008). Modulation of the regulatory activity of bacterial two-component systems by SlyA. J Biol Chem 283, 28158-28168.  127  Soutourina, O. A. & Bertin, P. N. (2003). Regulation cascade of flagellar expression in Gramnegative bacteria. FEMS Microbiol Rev 27, 505-523. Srivatsan, A. & Wang, J. D. (2008). Control of bacterial transcription, translation and replication by (p)ppGpp. Curr Opin Microbiol 11, 100-105. Stepanova, E., Lee, J., Ozerova, M., Semenova, E., Datsenko, K., Wanner, B. L., Severinov, K. & Borukhov, S. (2007). Analysis of promoter targets for Escherichia coli transcription elongation factor GreA in vivo and in vitro. J Bacteriol 189, 8772-8785. Sukupolvi, S. & O'Connor, C. D. (1990). TraT lipoprotein, a plasmid-specified mediator of interactions between gram-negative bacteria and their environment. Microbiol Rev 54, 331-341. Surette, M. G., Miller, M. B. & Bassler, B. L. (1999). Quorum sensing in Escherichia coli, Salmonella typhimurium, and Vibrio harveyi: a new family of genes responsible for autoinducer production. Proc Natl Acad Sci U S A 96, 1639-1644. Tateda, K., Ishii, Y., Kimura, S., Horikawa, M., Miyairi, S. & Yamaguchi, K. (2007). Suppression of Pseudomonas aeruginosa quorum-sensing systems by macrolides: a promising strategy or an oriental mystery? J Infect Chemother 13, 357-367. Tenson, T., Lovmar, M. & Ehrenberg, M. (2003). The mechanism of action of macrolides, lincosamides and streptogramin B reveals the nascent peptide exit path in the ribosome. J Mol Biol 330, 1005-1014. Teplitski, M., Goodier, R. I. & Ahmer, B. M. (2003). Pathways leading from BarA/SirA to motility and virulence gene expression in Salmonella. J Bacteriol 185, 7257-7265. Thijs, I. M., De Keersmaecker, S. C., Fadda, A., Engelen, K., Zhao, H., McClelland, M., Marchal, K. & Vanderleyden, J. (2007). Delineation of the Salmonella enterica serovar Typhimurium HilA regulon through genome-wide location and transcript analysis. J Bacteriol 189, 4587-4596. Thompson, A., Rolfe, M. D., Lucchini, S., Schwerk, P., Hinton, J. C. & Tedin, K. (2006). The bacterial signal molecule, ppGpp, mediates the environmental regulation of both the invasion and intracellular virulence gene programs of Salmonella. J Biol Chem 281, 3011230121. Tsui, W. H., Yim, G., Wang, H. H., McClure, J. E., Surette, M. G. & Davies, J. (2004). Dual effects of MLS antibiotics: transcriptional modulation and interactions on the ribosome. Chem Biol 11, 1307-1316. Turnbough, C. L., Jr. & Switzer, R. L. (2008). Regulation of pyrimidine biosynthetic gene expression in bacteria: repression without repressors. Microbiol Mol Biol Rev 72, 266-300, table of contents.  128  Umezawa, H. (1967). Antibiotic kasugamycin. Patent no. Patent number. United States Patent and Trademark Office. USA. Urban, A., Eckermann, S., Fast, B., Metzger, S., Gehling, M., Ziegelbauer, K., RubsamenWaigmann, H. & Freiberg, C. (2007). Novel whole-cell antibiotic biosensors for compound discovery. In Appl Environ Microbiol, pp. 6436-6443. Urban, J. H. & Vogel, J. (2008). Two seemingly homologous noncoding RNAs act hierarchically to activate glmS mRNA translation. PLoS Biol 6, e64. Valdivia, R. H. & Falkow, S. (1997). Fluorescence-based isolation of bacterial genes expressed within host cells. Science 277, 2007-2011. VanBogelen, R. A. & Neidhardt, F. C. (1990). Ribosomes as sensors of heat and cold shock in Escherichia coli. Proc Natl Acad Sci U S A 87, 5589-5593. Vassylyeva, M. N., Svetlov, V., Dearborn, A. D., Klyuyev, S., Artsimovitch, I. & Vassylyev, D. G. (2007). The carboxy-terminal coiled-coil of the RNA polymerase beta'-subunit is the main binding site for Gre factors. EMBO Rep 8, 1038-1043. Wadood, A., Dohmoto, M., Sugiura, S. & Yamaguchi, K. (1997). Characterization of copy number mutants of plasmid pSC101. J Gen Appl Microbiol 43, 309-316. Walker, K. A. & Osuna, R. (2002). Factors affecting start site selection at the Escherichia coli fis promoter. J Bacteriol 184, 4783-4791. Walsh, C. (2003). Antibiotics: actions, origins, resistance: ASM Press. Wang, L., Jensen, S., Hallman, R. & Reeves, P. R. (1998). Expression of the O antigen gene cluster is regulated by RfaH through the JUMPstart sequence. FEMS Microbiol Lett 165, 201206. Wang, Q., Frye, J. G., McClelland, M. & Harshey, R. M. (2004). Gene expression patterns during swarming in Salmonella typhimurium: genes specific to surface growth and putative new motility and pathogenicity genes. Mol Microbiol 52, 169-187. Waters, C. M. & Bassler, B. L. (2005). Quorum sensing: cell-to-cell communication in bacteria. Annu Rev Cell Dev Biol 21, 319-346. Watson, P. R., Paulin, S. M., Bland, A. P., Libby, S. J., Jones, P. W. & Wallis, T. S. (1999). Differential regulation of enteric and systemic salmonellosis by slyA. Infect Immun 67, 49504954. Wegrzyn, A., Szalewska-Palasz, A., Blaszczak, A., Liberek, K. & Wegrzyn, G. (1998). Differential inhibition of transcription from sigma70- and sigma32-dependent promoters by rifampicin. FEBS Lett 440, 172-174. 129  Weir, E. K., Martin, L. C., Poppe, C., Coombes, B. K. & Boerlin, P. (2008). Subinhibitory concentrations of tetracycline affect virulence gene expression in a multi-resistant Salmonella enterica subsp. enterica serovar Typhimurium DT104. Microbes Infect 10, 901-907. Wickus, G. G. & Strominger, J. L. (1972). Penicillin-sensitive transpeptidation during peptidoglycan biosynthesis in cell-free preparations from Bacillus megaterium. II. Effect of penicillins and cephalosporins on bacterial growth and in vitro transpeptidation. J Biol Chem 247, 5307-5311. Williams, P. & Camara, M. (2009). Quorum sensing and environmental adaptation in Pseudomonas aeruginosa: a tale of regulatory networks and multifunctional signal molecules. Curr Opin Microbiol 12, 182-191. Wilson, C. & Dombroski, A. J. (1997). Region 1 of sigma70 is required for efficient isomerization and initiation of transcription by Escherichia coli RNA polymerase. J Mol Biol 267, 60-74. World Health Organization (2010). Fact sheet No. 104. Yang, X., Molimau, S., Doherty, G. P., Johnston, E. B., Marles-Wright, J., Rothnagel, R., Hankamer, B., Lewis, R. J. & Lewis, P. J. (2009). The structure of bacterial RNA polymerase in complex with the essential transcription elongation factor NusA. EMBO Rep 10, 997-1002. Yanofsky, C. & Horn, V. (1981). Rifampin resistance mutations that alter the efficiency of transcription termination at the tryptophan operon attenuator. J Bacteriol 145, 1334-1341. Yim, G., de la Cruz, F., Spiegelman, G. B. & Davies, J. (2006). Transcription modulation of Salmonella enterica serovar Typhimurium promoters by sub-MIC levels of rifampin. J Bacteriol 188, 7988-7991. Yim, G., McClure, J., Surette, M. G. & Davies, J. E. (2011). Modulation of Salmonella gene expression by subinhibitory concentrations of quinolones. J Antibiot (Tokyo) 64, 73-78. Zhang, G., Campbell, E. A., Minakhin, L., Richter, C., Severinov, K. & Darst, S. A. (1999). Crystal structure of Thermus aquaticus core RNA polymerase at 3.3 A resolution. Cell 98, 811824. Zhang, X., McDaniel, A. D., Wolf, L. E., Keusch, G. T., Waldor, M. K. & Acheson, D. W. (2000). Quinolone antibiotics induce Shiga toxin-encoding bacteriophages, toxin production, and death in mice. J Infect Dis 181, 664-670. Zuker, M. (2003). Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Res 31, 3406-3415.  130  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items