UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An RNA secondary structure regulates sxy expression and competence development in haemophilus influenzae Bannister, Laura Anne 2000

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

Item Metadata

Download

Media
831-ubc_2000-486001.pdf [ 11.16MB ]
Metadata
JSON: 831-1.0089846.json
JSON-LD: 831-1.0089846-ld.json
RDF/XML (Pretty): 831-1.0089846-rdf.xml
RDF/JSON: 831-1.0089846-rdf.json
Turtle: 831-1.0089846-turtle.txt
N-Triples: 831-1.0089846-rdf-ntriples.txt
Original Record: 831-1.0089846-source.json
Full Text
831-1.0089846-fulltext.txt
Citation
831-1.0089846.ris

Full Text

A N RNA SECONDARY STRUCTURE REGULATES SXY EXPRESSION AND COMPETENCE DEVELOPMENT IN HAEMOPHILUS INFLUENZAE by LAURA ANNE BANNISTER A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES Department of Biochemistry and Molecular Biology We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA November 1999 © Laura Anne Bannister, 1999 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of X? ' Q tktr*' S-hry flod fV\g>| fciJtfr 6«.'«l»^y The University of British Columbia Vancouver, Canada Date N<>v7 «Ag* , 199 9 DE-6 (2/88) A B S T R A C T The process of natural transformation is the ability of cells to take up DNA from their environment and recombine this DNA into their genome to yield stable inheritance of an altered genotype. In Haemophilus influenzae, competence for transformation is tightly regulated, and is induced to maximal levels when cells undergo nutrient limitation. One signal for competence development is a rise in concentration of intracellular cAMP. A second proposed signal is an increase in the concentration of Sxy, a protein that appears to activate competence by regulating the expression of other competence genes. In this thesis, I examined the evolution and expression of sxy, focusing on whether the function of Sxy is specific to competence development, and whether the pattern of expression of sxy is consistent with competence being controlled by levels of Sxy within the cell. I found potential Sxy homologs in both transformable and nontransformable bacteria, indicating that Sxy may have a function that is unrelated to competence development. Neither Sxy nor any of its homologs contain amino acid motifs that share similarity with those of known transcription factors; therefore, my results do not confirm the hypothesis that Sxy directly activates transcription of competence genes. Sxyv.lacZ reporter gene analysis indicates that sxy expression is increased within cells well before maximum competence development in rich medium, suggesting that an increase in Sxy concentration is not the limiting step of competence development. Reporter gene analysis also indicates that maximal amounts of Sxy are not any higher in starvation medium than in rich medium, suggesting that an increase in Sxy concentration is not the second, cAMP-independent signal that increases the competence of cells in starvation medium relative to those in rich medium. Finally, I suggest that sxy may be regulated at the translational level, perhaps by the inhibition of translation by an mRNA secondary structure that had been previously proposed. I confirmed the existence of this secondary structure, using both in vivo and in vitro experiments. ii In this thesis, I also discuss attempts to find repressors of competence following H. influenzae mutagenesis with Tn926. Although I did not recover any such mutants, I did recover mutants in the fis operon of H. influenzae that have delayed spontaneous competence, and further experiments showed that fis null mutants have transformation frequencies in rich medium that are reduced - 500-fold relative to wild type cells. I suggest some ways in which Fis may be involved in natural transformation. iii TABLE OF CONTENTS Title Page I Abstract II Table of Contents IV List of Tables XI List of Figures XII List of Abbreviations XVII Acknowledgments XXI Chapter 1: General Introduction 1 1.1. Overview of the Distribution, Function and Evolution of Natural Transformation 1 1.2. The Mechanism of Natural Transformation in H. influenzae 7 1.2.1. H. influenzae preferentially binds and takes up DNA fragments containing a 9 bp core USS sequence 8 1.2.2. Processing and translocation of donor DNA 10 1.2.3. Homologous recombination of single-stranded donor DNA into the recipient chromosome 12 1.3. Regulation of competence development in H. influenzae occurs at the level of nutrient availability 13 1.3.1. Competence is induced by a rise in the concentrations of intracellular cAMP 14 1.3.2. A cAMP-independent signal is required for maximal competence induction 16 Chapter 2: Materials and Methods 18 2.1. General Methods 18 2.1.1. Strains and plasmids 18 2.1.2. Culture methods 18 iv 2.1.2.1. Haemophilus influenzae culture methods 18 2.1.2.2. Escherichia coli culture methods 24 2.1.3. Media 24 2.1.4. Natural transformation of H. influenzae and artificial transformation of E. coli 25 2.1.4.1. H. influenzae transformation 25 2.1.4.2. E. coli plasmid transformation 27 2.1.5. Transposon Mutagenesis 27 2.1.5.1. Mutagenesis with mmiTnlOkan 27 2.1.5.2. Tn926 mutagenesis 28 2.1.6. DNA methods 28 2.1.6.1. Isolation of plasmid and chromosomal DNAs 28 2.1.6.2. DNA sequencing 29 2.1.6.3. DNA agarose electrophoresis and cloning 29 2.1.6.4. Polymerase chain reaction 30 2.1.6.5. Southern analysis 30 2.2. Sxy methods 34 2.2.1. Sxy homology searches 34 2.2.2. Construction of sxy.dacZ operon and protein fusions 35 2.2.2.1. Construction of operon and protein fusions of lacZ to sxy codon 89 35 2.2.2.2. Construction of operon and protein fusions of lacZ to sxy codon 11 37 2.2.3. P-galactosidase assays 37 2.2.4. Cloning of sxy and sxy-1 into pGEM7 for RNA secondary structure mapping 39 2.2.5. Sxy and sxy-1 RNA secondary structure mapping 40 2.2.5.1. In vitro RNA transcription 40 v 2.2.5.2. RNase secondary structure mapping 42 2.2.6. Mfold analysis of sxy mRNA secondary structures 44 2.2.7. Site-directed mutagenesis 44 2.2.8. Transferring site-directed mutations to the KW20 chromosome 50 2.2.8.1 Transferring the sxy-5 mutation to the KW20 genome 50 2.2.8.2. Transferring the sxy-6 and sxy-7 mutations to the KW20 genome 50 2.2.9. Northern analysis 53 2.3 Fis methods 55 2.3.1. Cloning the Tn926 insertion in mutant B-l.. 55 2.3.2. Cloning the fis operon by PCR 57 2.3.3. Cassette mutagenesis of fis and orfl 58 Chapter 3 : Sxy function and regulation 62 3.1. Background: The role of Sxy in competence development 62 3.1.1. Discovery of the role of Sxy in competence development... 62 3.1.2. Sxy regulates competence development 63 3.1.3. Regulatory features of the sxy gene 67 3.1.3.1. The sxy promoter 67 3.1.3.2. Translation of the sxy message 68 3.1.3.3. The sxy hypercompetence mutations may disrupt base pairing of a regulatory RNA secondary structure 68 3.1.4. A working model for competence regulation by the control of sxy expression 71 3.1.5. Specific research objectives 72 3. 2. Results 73 vi 3.2.1. Database searches to find Sxy homologs and functional motifs 73 3.2.1.1. Proteins that are similar to Sxy are found in both transformable and nontransformable bacteria 73 3.2.1.2. Sxy and its putative homologs are soluble, leucine-rich proteins with no known homology to other proteins80 3.2.1.3. Sxy and its putative homologs have conserved amino acid motifs 81 3.2.2 Site-directed mutagenesis confirms that the sxy-5 mutation is sufficient to cause hypercompetence 84 3.2.3. Transcriptional regulation of sxy expression 85 3.2.3.1 The production of Sxy is regulated at the level of mRNA availability 87 3.2.3.2. Regulation of sxy transcription may be independent of cAMP 91 3.2.4. Sxy expression is posttranscriptionally regulated 95 3.2.5. Eliminating the 3' half of sxy RNA structure A results in an increase in sxy transcription and translation 100 3.2.6. Point mutations that strengthen or weaken base pairing in sxy RNA structure A affect sxy expression and competence development 106 3.2.6.1. Mfold predicts a sxy RNA secondary structure that is destabilized by the sxy hypercompetence mutations 106 3.2.6.2. The sxy-1 point mutation increases expression of (3-galactosidase from sxy operon and protein fusions 110 3.2.6.3. Site-directed mutations designed to destabilize or stabilize sxy RNA structure A affect sxy expression and competence development 114 3.2.6.3.1. Construction of the sxy-7 point mutant and transfer of this mutation to the KW20 genome 115 3.2.6.3.2. The sxy-7 mutation prevents competence development, and decreases lacZ expression from operon fusions 121 3.2.6.4. Preliminary Northern analysis confirms that sxy mutations increase or decrease sxy mRNA levels 124 3.2.7. In vitro RNase mapping supports the folding of sxy mRNA into the modified structure A. l or A.2 127 Chapter 4 : Fis mutants of H. influenzae have delayed and reduced spontaneous competence 145 4.1. Background 145 4.1.1. Rationale for Tn916 mutagenesis and isolation of competence mutants 145 4.1.2. The evolution and function of Fis in E. coli and other entericsl46 4.2. Results 149 4.2.1. Tn916 mutagenesis and isolation of mutants with delayed spontaneous competence development 149 4.2.2. Mapping of the transposon hops: The Tn926 mutation causing delayed spontaneous competence is in the conserved fis operon 155 4.2.3. Mutations in fis cause a 500-fold reduction in transformation frequencies during spontaneous competence development 159 4.2.4. Epistasis between fis and the competence regulatory genes cya and sxy 166 Chapter 5 : Discussion 170 5.1. Discussion of sxy results 170 5.1.1. Sxy function and evolution 170 5.1.2. The correlation of sxy expression with competence development 172 5.1.2.1. The correlation of sxy expression and competence development in sBHI 173 5.1.2.2. The correlation of sxy expression and competence development in MIV 177 5.1.3. Sxy expression is mediated by transcriptional and posttranscriptional regulation 179 5.1.4. The 5' region of sxy mRNA folds into a secondary structure that negatively regulates sxy translation 180 5.1.5. Regulation of gene expression by RNA secondary structure: a review 184 5.1.5.1. Sxy expression is probably not regulated by an attenuation mechanism 184 5.1.5.2. Sxy RNA secondary structure is probably not a cleavage determinant for endonucleolytic mRNA decay 187 5.1.5.3. Sxy RNA secondary structure may affect translational initiation from the sxy ribosome binding sitel90 5.1.5.4. What is the signal for the increase in sxy mRNA translation during stationary phase in sBHI and in MIV? : 193 5.1.6. A modified model for competence development in sBHI 196 5.2. Discussion of Fis results 203 5.2.1. Repressors of competence development were not identified by Tn916 mutagenesis 203 5.2.2. fis mutants are unlikely to have reduced transformation frequencies simply as a result of slow growth 204 5.2.3. Fis may modulate a mechanistic step of natural transformation 204 5.2.4. Fis may directly or indirectly affect competence by transcriptional regulation 205 5.2.5. Summary of Fis results 207 Bibliography 208 x LIST OF TABLES Table 1.1 Phylogenetic distribution of natural transformation within Eubacteria and Archaebacteria 4 Table 2.1 Bacterial strains used in this study 19 Table 2.2 Plasmids and phage used in this study 21 Table 3.1 Results of blast searches of H. influenzae Reel and Sxy amino acid sequences 75 Table 3.2 Extent of similiarity between Sxy and its homologs 78 Table 3.3 Mfold RNA secondary structure predictions for sxy wild type and sxy-1 to sxy-5 sequences 109 Table 5.1 Summary of the effects of sxy mutations on cellular transformation frequencies and on (3-galactosidase expression from sxy.dacZ operon and protein fusions 174 Table 5.2 Mechanisms of transcriptional attenuation 186 xi LIST OF FIGURES Figure 1.1 An overview of the phylogenetic distribution of natural transformation in Eubacteria and Archaebacteria • 3 Figure 1.2 An overview of the mechanistic steps of natural transformation in H. influenzae 9 Figure 2.1 Sequences and annealing sites for sxy primers 31 Figure 2.2 Location and sequences of primers used to amplify the KW20 fis operon by PCR 32 Figure 2.3 Creation of pLBSFl 36 Figure 2.4 Creation of pLBSF3 38 Figure 2.5 Cloning strategy for the creation of pGEMsxy 41 Figure 2.6 Cloning strategy for the creation of pAltersxy 46 Figure 2.7 Sequences and annealing sites for mutagenic oligos used to create the site-directed mutations sxy-5, sxy-6 and sxy-7 46 Figure 2.8 Site-directed mutagenesis steps carried out to isolate the sxy-5 point mutation 47 Figure 2.9 Cloning strategy for the creation of pLBS5 49 Figure 2.10 Directed integration of pLBS6K into the RR699 chromosome and regeneration of a single chromosomal copy of sxy carrying the sxy-6 allele 51 Figure 2.11 Cloning strategy for the creation of pLBTnl 56 Figure 2.12 Cloning strategy for the creation of pLBFl 59 Figure 2.13 Cloning strategy for the creation of pLBF2 60 Figure 2.14 Cloning strategy for the creation of pLBF3 61 Figure 3.1 Chromosomal location of sxy and location of sxy promoter and sxy hypercompetence mutations 64 xii Figure 3.2 Growth and transformation frequencies of KW20, RR648 and RR554 in sBHI 65 Figure 3.3 Transformation frequencies and growth of KW20, RR699, RR700, RR724 and RR723 in sBHI 69 Figure 3.4 Base pair and amino acid alterations caused by the sxy hypercompetence mutations, and predicted sxy RNA secondary structure 70 Figure 3.5 Chromosomal maps of sxy and putative sxy homologs and surrounding genes 79 Figure 3.6 Clustal W alignment of Sxy amino acid sequence with the sequences of putative Sxy homologs 82 Figure 3.7 Conserved motifs of Sxy and Sxy homologs 83 Figure 3.8 Results of colony transformation assays for KW20 transformed with pDJM90 and pLBS5 86 Figure 3.9 The operon fusion plasmid pLBSFl 88 Figure 3.10 fj-galactosidase expression and growth of RR844 in sBHI 89 Figure 3.11 (3-galactosidase expression and growth of RR844 in sBHI and MIV 92 Figure 3.12 (3-galactosidase expression and growth of RR844 and RR874 in sBHI and MIV 94 Figure 3.13 (3-galactosidase expression and growth of RR844 in sBHI with and without cAMP addition '. 96 Figure 3.14 The protein fusion plasmid pLBSF2 98 Figure 3.15 P-galactosidase expression and growth of RR844 and RR845 in sBHI and MIV 99 Figure 3.16 The operon fusion plasmid pLBSF3 102 Figure 3.17 The protein fusion plasmid pLBSF4 103 xiii Figure 3.18 P-galactosidase expression and growth of RR844 and RR846 in sBHI 104 Figure 3.19 P-galactosidase expression and growth of RR845 and RR847 in sBHI 105 Figure 3.20 Mfold RNA secondary structures obtained by folding sxy wild type and mutant RNAs 107 Figure 3.21 P-galactosidase expression and growth of RR844 and RR848 in sBHI and MIV 112 Figure 3.22 P-galactosidase expression and growth of RR845 and RR849 in sBHI and MIV 113 Figure 3.23 Sxy-6 mutation created by site-directed mutagenesis 116 Figure 3.24 Sxy-7 mutation created by site-directed mutagenesis 117 Figure 3.25 Transformation frequencies and growth of KW20 and RR854in sBHI 122 Figure 3.26 P-galactosidase expression and growth of RR844 and RR860 in sBHI 123 Figure 3.27 Northern analysis of KW20, RR699, RR724 and RR854 125 Figure 3.28 Plasmid pGEMsxy used to generate in vitro sxy transcript 128 Figure 3.29 Mapping of nuclease-sensitive sites in end-labeled sxy RNA : 130 Figure 3.30 Mapping of nuclease-sensitive sites in end-labeled sxy RNA 132 Figure 3.31 Mapping of nuclease-sensitive sites in end-labeled sxy-1 RNA 134 xiv Figure 3.32 Mapping of nuclease-sensitive sites in end-labeled sxy-1 RNA.... 136 Figure 3.33 Summary of RNase structure mapping of sxy RNA 138 Figure 3.34 Summary of RNase structure mapping of sxy-1 RNA 139 Figure 3.35 sxy RNA structures A. l and A.2 144 Figure 4.1 Restriction map of pAM120 150 Figure 4.2 Results of colony assays for Tn926 mutants 152 Figure 4.3 Transformation frequencies and growth of KW20 and mutant B-l in sBHI 154 Figure 4.4 Restriction map of Tn926 insertion in mutants with delayed spontaneous competence 156 Figure 4.5 Locations of Tn926 insertion in mutant B-l and spectinomycin cassettes in fis::spec and orfl::spec mutants 158 Figure 4.6 Transformation frequencies and growth of KW20 and RR859 in sBHI 160 Figure 4.7 Growth of KW20 and RR858 in sBHI 162 Figure 4.8 Transformation frequencies and growth of KW20 and RR858 in sBHI 163 Figure 4.9 Transformation frequencies and growth of KW20 andRR859 in MIV 164 Figure 4.10 Transformation frequencies and growth of KW20 and RR858 in MIV 165 Figure 4.11 Effect of ImM cAMP on transformation frequencies of RR858 in sBHI 168 Figure 4.12 Transformation frequencies and growth of RR873 and RR866 in sBHI 169 xv Figure 5.1 Modified model for competence development in sBHL version 1 198 Figure 5.2 Modified model for competence development in sBHI, version 2 200 xvi LIST OF ABBREVIATIONS Amp r ampicillin resistance ATP adenosine 5'-triphosphate BAP bacterial alkaline phosphatase bp base pair cAMP 3', 5' cyclic adenosine monophosphate cfu colony forming unit C m r chloramphenicol resistance CPSD (Disodium 3-(4-methoxyspiro{l,2-dioxetane - 3, 2' - (5'chloro) tricyclo [3.3.1.1 /^7] decan} - 4 - yl) phenyl phosphate CRE competence regulatory element CRP cAMP receptor protein CTP cytidine 5'-triphosphate DEPC diethylpyrocarbonate DIG digoxigenin DNA deoxyribonuleic acid i DNase deoxyribonuclease dNTP deoxyribonucleoside 5'-triphosphate E value expected value, xvii DTT dithiothreitol EDTA ethylenediaminetetraacetic acid Fis factor for inversion stimulation GTP guanosine 5'-triphosphate HI# Haemophilus influenzae protein number (T.I.G.R.) HTH helix turn helix Kan r kanarhycin resistance kbp kilobase pair kDa kiloDalton LacZ [3-galactosidase LAMA Local Alignment of Multiple Alignments LB Luria Bertani broth MAST Motif Alignment and Search Tool MIV Haemophilus influenzae defined non-growth media mRNA messenger RNA NAD nicotinamide adenine dinucleotide Nal r nalidixic acid resistance NAOAc sodium acetate N H 4 O A C ammonium acetate xviii N o v r novobiocin resistance OD optical density ONPG O-nitrophenyl-fJ-D-galactopyranoside PBS phosphate-buffered saline PEG polyethylene glycol pfu plaque forming unit pnk polynucleotide kinase PBS phosphate buffered saline PCR polymerase chain reaction PTS phosphotransferase system RBS ribosome binding site RNA ribonucleic acid RNase ribonuclease rNTP ribonucleoside 5'-triphosphate sBHI supplemented Brain Heart Infusion broth SAPS Statistical analysis of protein sequences SDS sodium dodecyl sulfate Sm r streptomycin resistance Sp r spectinomycin resistance SSC standard saline citrate TAE tris acetate EDTA Taq Thermus aquaticus TBE tris borate EDTA TBMM tryptone broth supplemented with magnesium and maltose TE tris EDTA Tet r tetracycline resistance T.I.G.R. The Institute For Genomic Research Tn transposon uCi microCurie UAS upstream activating site USS uptake signal sequence UTP uridine 5'-triphosphate UV Ultraviolet X-GAL 5-bromo-4-chloro-3-indolyl-(3-D-galactopyranoside xx ACKNOWLEDGMENTS I would like to thank my research supervisor, Dr. Rosemary Redfield, both for academic instruction as wells as financial support. I would also like to thank each of the members of the Redfield lab, past and present, whom I have enjoyed working with over the years. In addition, I would like to thank my supervisory committee, Dr. George Spiegelman and Dr. Caroline Astell, for their helpful advice. I would also like to thank Dr. Mackie, and the members of the Mackie lab, especially Kristian Baker, for providing laboratory research space for me and kindly instructing me in RNA techniques. I would also like to thank several individuals for providing me with moral support and friendship throughout the completion of my Ph. D, including my family members, and my friends Holger Ebhardt, Laura Barjaktarovic, Claire Sutherland, Heather Scott, Shaun Cordes, Lillian Ford, Stephen Latham, and Kevin Peters. xxi C h a p t e r 1; G e n e r a l I n t r o d u c t i o n To understand how natural transformation has arisen and been maintained in a diverse number of bacteria, we are studying the regulation of this process in Haemophilus influenzae. Identification of the signals that regulate competence for transformation should provide information about its function and evolution. One signal in H, influenzae is an increase in the concentration of intracellular cAMP. A second regulatory component may be increased concentrations of the Sxy protein. In this thesis, I have investigated the evolution and function of Sxy and regulation of sxy expression. In addition, I identified mutations in the gene encoding the H. influenzae Fis homolog that severely reduce competence development. In the following sections of the Introduction, I first discuss the phylogenetic distribution of natural transformation and the selective forces that may govern its evolution; I then discuss in detail the mechanism of natural transformation and the regulation of competence development in H. influenzae. 1.1. Overview of the Distribution, Function and Evolution of Natural Transformation Natural transformation is a process of horizontal gene transfer that allows bacteria to take up DNA from the environment and incorporate the acquired genetic information into their genome by homologous recombination. Several comprehensive reviews about this process have been published (62,63,97,119,168,170,173). As a process of genetic exchange, natural transformation is distinguished from bacterial transduction and conjugation in that it is the only form of DNA exchange that is encoded by chromosomal genes, and it may be the only process considered to have evolved for genetic exchange (156). Natural transformation can be divided into several steps. The first of these is the development of genetic competence, which is the ability of bacteria to bind and take up DNA. In most bacteria, induction of competence is regulated by a subset of genes which are coordinately expressed only under particular environmental 1 conditions. Following binding, DNA is taken up and translocated into the cytoplasm. Transformation is completed upon homologous recombination of the incoming DNA into the recipient chromosome. More than 40 species of naturally transformable bacteria have been described (118). Figure 1.1 and Table 1.1 describe the phylogenetic distribution of natural transformation within Eubacteria and Archaebacteria. The phenomenon of natural transformation is widespread but there is no clear evolutionary pattern which describes its distribution, since the ability to transform is distributed sporadically within evolutionary subdivisions and families. Furthermore, even within a given species there are both transformable and nontransformable isolates (118). This sporadic distribution makes it difficult to determine whether the ability to transform evolved independently in many bacterial lineages or evolved once and was frequently lost. Many bacterial species have been shown to be transformable under laboratory conditions, but do bacteria in natural environments exchange DNA by transformation? To address this issue, several factors need to be considered — are bacteria capable of transforming in situ, does DNA exists in sufficient quantities in the environment, and does the analysis of bacterial genomes reveal evidence of nonclonality? Many bacterial species have been shown to be capable of natural transformation in situ, and high molecular weight DNA is abundant in some environments (118). DNA concentrations are approximately 300 ug/mL within respiratory mucosa of healthy individuals, and have been measured at approximately lug/g within sediments (116,118). Extracellular DNA in the environment appears to be protected against nucleases by its association with soil, sediments, cell walls etc. (118). Most environmental free DNA probably originates from cellular lysis, but there is also evidence that some bacteria may be able to actively extrude genomic DNA into the environment [reviewed in (118)]. The most compelling evidence for the occurrence of genetic exchange in the environment both 2 CD o CO J* L U o CO . O o CO u_ o + CD o CO o \_ c CD CO +—> o > CO O JO f w u CD ' +-< (D £ O - O CO Z CD (3 CO to E a> CD & p o o \ 3 03 s o <*> 3 03 03 4-1 OS c 4-1 M-H O •r-1 > t-i > O 3 < to a> DH o 3 03 •s O CN OJ 3 a» i-i OJ "H-H a» 3 o •I-H f-i 4-» • F-I 4-1 OJ 3 OJ 0 0 3 £ -c OJ o at as (-H O OJ H a> ai -a a» 3 TH1 ai T3 3 3 OJ ta in OJ 'o a» DH in a» 03 S r-H 1/3 3 03 .2 -2 >-l a» 4-* u 03 OJ 03 u l-H < T3 3 03 OS • i-H l - l cu -I-I U 03 X 3 W 3 o u 4-1 03 X l 4-* <Si DH 3 O too u •43 ai 3 cu too o X OH DH 3 O rH too X CJ 03 a» 3 •F-H X ! oj 'u OJ D H in __0J x> 03 l « 3 03 H CO > •r-1 <*> 3 OJ xi a> U l DH S o u Table 1.1. Phylogenetic distribution of natural transformation within Eubacteria and Archaebacteria. Adapted from Lorenz and Wackernagel (118). The evolutionary classification of each microorganism was obtained from the NCBI taxonomy home page which can be accesssed at http://www.ncbi.gov/Taxonomy. DOMAIN SUBDIVISION GROUP/FAMILY GENUS AND SPECIES 1 Bacteria Proteobacteria; Methylbacterium Methylobacterium OC subdivision organophilum Rhizobiaceae group/ Rhizobium meliloti Rhizobiaceae Proteobacteria; Neisseriaceae Neisseria gonorrhoea P subdivision Neisseria meningitidis Bukholderia group? Ralstonla solanacearum Thiobacillus thiopatvs Thiobacillus sp. strain Y Proteobacteria; Campylobacter Group Campylobacter coli £ subdivision Campylobacter jejuni Helicobacter Group Helicobacter pylori Proteobacteria Alcaligenaceae Achromobacter spp. Y subdivision Azotobacteriaceae Azotobacter vinelandii Leglonellaceae Legionella pneumophila Pasteurellaceae Actinobacillus actinomycetemcomitans Haemophilus influenzae Haemophilus parainfluenzae Pseudomonas Group Acinetobacter spp. Pseudomonas stutzeri Vibrioaceae Vibrio sp. strain D19 Vibrio sp. strain WJT-1C Vibrio parahaemolyticus Firmicutes Bacillus/Clostridium Bacillus subtilis group Bacillus licheniformis Baclllacae Staphylococcus aureus Bacillus/Clostridium Lactobacillus lactis group Lactobacillaceae Bacillus/Clostridium Streptococcus aureus group Streptococcus pneumoniae Streptococcaceae Streptococcus sanguis Streptococcus mutans Streptococcus gordonii 4 Table 1.1. continued.. DOMAIN SUBDIVISION GROUP/FAMILY GENUS AND SPECIES Bacteria Firmicutes Actlnobacteria/ Streptomycetaceae Actinobacter/ Mycobacterlaceae Actinobacterla/ Themoactinomycetes Streptomyces spp. Mycobacterium smegmatis Thermoactinomyces vulgaris Cyanobacteria Chroococcales/ Synechococcus Anacystis nidulans Synechocystis sp. strain 6803 Synechocystis sp. strain OL50 Green Sulfur bacteria Nostocales/ Nostocaceae Nostoc muscorum Chloroblum limicola Thermus/ Delnococcus group Thermus thermpophilus Thermus tlavus Thermus caldophilus Thermus aquaticus Delnococcus radiodurans Archaea Archaebacleria Euryarchaeota Methanobacteriales/ Methanobacterlaceae Methanococcales/ Methanococcaceae Methanobacterium thermoautotrphicum Methanococcus voltae 5 within and between species comes from observations of genetic mosaicism within naturally transformable species [reviewed in (118)]. What is the underlying function of transformation, and what are the selective pressures under which natural transformation has arisen and been maintained in a subset of bacterial species? One potential benefit of transformation stems from recombination, similar to that suggested for sexual species (107). The majority of bacteria produce genetically identical progeny, and genetic changes arising from mutation are confined to the clonal progeny of those bacteria. Transformation can result in novel combinations of favorable alleles, or in the disruption of combinations of harmful mutations (170). For example, transformation may enable pathogenic bacterial species to evade host defenses via recombination-mediated variation in antigenic properties such as pili (reported for Neisseria and Campylobacter species) and penicillin binding proteins (reported for Streptococcus species) (118,170). A second suggested benefit is the use of transforming DNA as a template against which damaged recipient chromosomal DNA can be repaired (131). A third suggested benefit is the utilization of incoming nucleotides of transforming DNA as an energy source (156,173). Which, if any, of these benefits account for the prevalence of natural transformation? Redfield has suggested that by considering the signals which induce competence development in bacteria, we may obtain information about which of these processes -i.e. recombination, DNA repair, or nutrient acquisition - is the main selective force underlying the function of bacterial transformation (156). For example, if DNA repair was the underlying function, we would expect damaged DNA to induce competence development. Similarly, if nutrient acquisition was the main selective force underlying the evolution of transformation, we would expect competence to be induced under conditions of nutrient limitation. However, Redfield also suggested that the selective advantages engendered by each of the described benefits might not be mutually 6 exclusive, and may in fact act in a synergistic fashion. For example, transforming DNA might be used as a nutritional source as well as for homologous recombination (156). 1.2. The Mechanism of Natural Transformation in H. influenzae Natural transformation has been studied most extensively in the Gram positive bacteria Bacillus subtilis and Streptococcus pneumoniae, and in the Gram negative bacteria Haemophilus influenzae and Neisseria gonorrhoeae. Historically, the mechanism of natural transformation has been considered to differ significantly between Gram positive and Gram negative bacteria. However, it is unclear whether the differences in transformation mechanisms between these two types of bacteria actually represent different processes, or are simply a consequence of the differences in cell wall structure. In Gram negative bacteria, the presence of a negatively charged outer membrane is an additional barrier to the translocation of negatively charged DNA (60). In the following section of the Introduction, I will focus on the mechanism of natural transformation of H. influenzae. H. influenzae is a small Gram negative bacterium that is a member of the Pasteurellaceae family within the y subdivision of the Proteobacteria. This bacterium causes meningitis in children and also causes respiratory infections, especially in immunocompromised and elderly adults. H. influenzae strains are fastidious and require a source of NAD and hemin for growth. Our study organism, KW20, lacks a capsule and is therefore a nonpathogenic rough derivative (Rd) strain. The capsule of H. influenzae is a negatively charged, porous matrix consisting of a phosphodiester linked ribose-ribitol copolymer that is unlikely to represent a barrier to DNA translocation (97). Encapsulated clinical isolates transform as efficiently as non encapsulated strains (161); therefore, the lack of a capsule does not affect the natural transformation process of KW20. 7 1.2.1. H. influenzae preferentially binds and takes up DNA fragments containing a 9 bp core USS. An overview of the process of natural transformation in H. influenzae is shown in Figure 1.2. Competent H. influenzae cells bind double-stranded DNA at the cell surface. DNA that is irreversibly bound is resistant to DNasel and is not elutable by high salt washing (97). H. influenzae cells selectively bind only DNA from the same or closely related species, unlike Gram positive bacteria, which bind DNA in a non- sequence-specific manner (13,54). The specificity of binding is due to the recognition of an uptake signal sequence (USS), consisting of the 9 bp core sequence 5' AAGTGCGGT 3' and flanking DNA, by an unidentified receptor protein or complex (51, 54, 78,167). Heterologous DNA that lacks USSs transforms poorly and competes inefficiently against USS-containing DNA sequences in DNA uptake experiments. N. meningitidis also binds homospecific DNA and recognizes an unrelated USS of similar length (65,79). The specificity of USS recognition in H. influenzae and N. gonorrhoeae indicates that DNA binding and uptake in these microorganisms is mediated by a specific receptor protein or proteins and is not a diffuse property of the cell envelope. In H. influenzae, DNA uptake saturates at about 120-140 ng/10^ cells (15). Each competent cell has been estimated as having approximately 8-10 receptors capable of binding USS-containing double-stranded DNA fragments (168). The existence of a discrete number of bacterial DNA receptors that cannot be reutilized is supported by the fact that competent cells transformed with one marker become resistant to transformation with a second marker (54). The protein or proteins which bind theUSS have not been identified. Several candidate proteins have been identified. These proteins are present in the outer membrane of competent cells but are not present in noncompetent cells. Genes for the putative receptor protein(s) have not been identified (168). 8 Figure 1.2. An overview of the mechanistic steps of natural transformation in H. influenzae. Adapted from Redfield (unpublished). Step 1 shows irreversible binding of USS-containing double- stranded DNA by the bacterial cell. Step 2 shows uptake of double-stranded DNA into the competent cell into a 'transformasome' or the periplasmic space, where it is shielded from DNase and cellular restriction enzymes. Step 3 shows DNA translocation into the cytoplasm. The 3' end of the leading strand is partially degraded and the 5' end of the leading strand is completely degraded. Degraded nucleotides are reutilized as precursors for subsequent rounds of DNA replication. Step 4 shows the Reel-mediated recombination of a single strand of donor DNA into the homologous region of the recipient chromosome. 9 The H. influenzae USS is overrepresented in the genome. There are 1465 copies of a 29 bp sequence that includes the perfect core USS and conserved flanking sequences, whereas only 8 would be expected by chance (98,169). It is not known what selective forces maintain the high abundance of USSs. Several functional roles unrelated to transformation have been proposed for USSs. One suggestion is that USSs are transcriptional terminators, because 17% of the USSs occur in pairs that can form stem loops at the 3' ends of genes, an arrangement also observed for USSs in Neisseria (169). It has also been proposed that USSs may function as sites of membrane attachment or as sites contributing to genome packaging (98). 1.2.2. Processing and translocation of donor DNA During transformation of Gram positive bacteria, DNA bound irreversibly at the cell surface is nicked and one strand of the DNA is degraded by a surface endonuclease that creates a single-stranded DNA intermediate for uptake (62,119). In contrast, H. influenzae takes up DNA in a double-stranded form that becomes resistant both to extracellular DNase and cellular DNA restriction systems. The compartmentalization of DNA during this uptake process is not clearly understood. Membranous blebs, referred to as transformasomes, have been visualized on the surface of competent cells by electron microscopy (95, 96). These appear to be vesicles in which the inner and outer membranes are fused to form a pore through which DNA could be translocated into the cytoplasm of the cell. However, membranous blebs are seen for numerous bacteria under general conditions of stress, casting doubt on the idea that transformasomes are a specialized membranous structure formed in response to competence induction. Therefore it remains unclear whether transforming DNA becomes sequestered in specialized transformasomes or simply within the periplasmic space. 10 Following uptake, DNA is processed and transported into the cell for homologous recombination. DNA translocation occurs extremely rapidly - DNA uptake is completed within a 2 minute period, and there is no detectable single-stranded DNA intermediate or eclipse phase as is seen in Gram positive bacteria (115). Translocation of DNA into the cytoplasm is an energy-dependent process, but the precise source of the required energy - membrane potential (proton or voltage gradient) or ATP hydrolysis - has not been elucidated (63). DNA requires a free end to be directionally exported from the transformasome/periplasmic space - circular plasmid DNA molecules and linear DNA with hairpin ends or termini covalently bound by proteins are inefficiently translocated into the cytoplasm (25,96). DNA translocation is accompanied by a degradation of the 5' incoming strand and limited degradation of the 3' incoming strand, which becomes available for homologous recombination with the recipient genome (Figure l.A; step 4) (25,174). Measurements obtained with radiolabeled DNA indicate that 85-90% of incoming donor DNA is degraded to yield nucleotides that are randomly incorporated into the genome during subsequent rounds of DNA replication, while 10-15% of the DNA remains undegraded and is available for homologous recombination (97). Several H. influenzae proteins that are required for DNA processing and translocation have been identified. The Rec-2 protein is a putative integral membrane protein with homology to the ComEC protein of B. subtilis (28,48, 125). The Dpr protein is also required for DNA processing (99,100). The protein encoded by the comF gene (also called comlOlA) is homologous to the integral membrane ComFA protein of B. subtilis (112-114,182). It has been proposed that the ComFA protein uses the energy of ATP hydrolysis to transport single-stranded DNA into the cytoplasm during transformation (63). A H. influenzae periplasmic oxidoreductase (Por) protein is also required for transformation, but its precise role has not been determined (180). 11 A model for the translocation of DNA and other macromolecules across Gram positive and Gram negative bacterial membranes has been proposed (60, 63). The model stems from the observation that many of the com genes of B. subtilis and other transformable bacteria share similarity with proteins that function in other membrane translocation systems, such as bacterial conjugation, T DNA transfer, and the secretion and/or assembly of cell envelope proteins such as pullanase and the prepilin proteins of type IV pili. [Mutations in competence proteins resembling type IV pilin proteins have been identified in all naturally transformable bacteria, including H.influenzae (58)]. This similarity has led to the hypothesis that complex channels composed of several proteins spanning the cell envelope may be assembled for the translocation of macromolecules across the bacterial envelope. The energy for the translocation process is proposed to come from ATP hydrolysis by membrane-bound nucleotide binding proteins; for example, the ComFA protein in the case of B. subtilis transformation (62). 1.2.3. Homologous recombination of single-stranded donor DNA into the recipient chromosome Following translocation into the cytoplasm, DNA is incorporated by homologous recombination through strand displacement of the homologous region of the recipient chromosome (145,152). There have been reports that DNA synthesized during competence development contains single-stranded gaps and tails that could serve as recombinogenic sites for synapsis with donor cell DNA (115,126). An average of 1.5 kb of the 3' strand of the donor DNA is degraded during the search for homology (Figure 1.2; Steps 3 and 4) (25). Once recombination is initiated, it is rapidly completed and proceeds to the 5' end of the integrating DNA (25). Each competent cell is predicted to take up 4 to 8 fragments of transforming DNA, and either strand of the donor duplex can be taken up and integrated with equal probability (168). A significant proportion of the recipient chromosome can be replaced during a single round of transformation 12 (168). Under saturating DNA conditions, a cell can be transformed at a frequency of 0.5-5% for a single marker (96). Reel (the homolog of E. coli RecA), is the only known recombination protein which has an identified role in H. influenzae transformation. Although RecBCD is required for transformation of B. subtilis (171), RecBCD mutants of H. influenzae are not transformation-deficient [(190), (R.S. Myers unpublished data)]. Reel mutants are transformation defective; this protein is required for homologous recombination, but not for DNA binding, uptake or translocation (27, 28,166,175). Is Reel expression increased upon competence development? As discussed above, if the underlying function of natural transformation is adaptive recombination, we would expect conditions that induce competence development to increase the expression of the homologous recombination machinery. In fact, RecA activity is increased by competence induction in both B. subtilis and in S. pneumoniae (119,170). However, although phage recombination in H. influenzae increases 10 to 100-fold upon competence induction (37), the expression of Reel is not increased (199). It remains uncertain if any components of the homologous recombination pathway of H. influenzae are induced upon competence development. 1.3. Regulation of competence development in H. influenzae occurs at the level of nutrient availability. The majority of naturally transformable bacteria (with the notable exception of N. gonorrhoeae) become competent only under certain environmental conditions, possibly because the synthesis of competence proteins is energetically expensive, and possibly because the presence of the DNA uptake and processing machinery compromises the integrity of the cell envelope. In addition, it may not be advantageous for naturally transformable bacteria to be constitutively competent if the surrounding DNA arises primarily from the lysis of cells that may contain many mutations. 13 In Gram positive bacterial populations, transformation is induced by quorum sensing of peptide hormones (170). In B. subtilis, competence stimulating factor induces the expression of ComK, which in turn activates the expression of other com genes. In H. influenzae, competence is not induced by cell to cell signalling events; instead, it appears to be an autonomous response mediated by individual cells in response to nutrient limitation. In the following section of the Introduction, I will focus on the regulation of competence induction in H. influenzae. I will discuss evidence that shows that competence is activated by high concentrations of cAMP, and by increased expression of the regulatory protein known as Sxy. 1.3.1. Competence is induced by a rise in the concentrations of intracellular cAMP. When measuring transformation to antibiotic resistance using chromosomal markers, transformation frequencies of H. influenzae are low to unmeasurable (<10~8) for cells in the exponential phase of growth in rich medium (BHI supplemented with NAD and hemin) and are transiently induced by about 10,000 fold (to 5 X 10"^ ) when cells reach the stationary phase of growth. This rise in transformation frequency is proposed to be due to an increase in the concentration of intracellular cAMP, because the addition of ImM cAMP to exponentially growing cells can also raise transformation frequencies to this moderate level (193). Exponentially growing cells transferred from rich broth to a defined nongrowth medium known as MIV achieve transformation frequencies of about 10"2 with chromosomal DNA markers, or 10 to 20% with cloned DNA fragments, corresponding to a population in which most of the cells are competent (77,87). The transfer of cells to MIV, which contains low levels of precursors for protein synthesis, and lacks precursors for DNA synthesis, is also proposed to increase cellular concentrations of cAMP (120). Exponentially growing cells subjected to transient oxygen limitation also achieve transformation frequencies approaching 10~2 (77); however, it is not known whether oxygen limitation increases cAMP concentrations in H. influenzae. It should be noted that the hypothesis that cAMP concentrations are 14 increased in H. influenzae cells during stationary phase in rich medium and in nongrowth medium is based on indirect measurements of cAMP levels using cAMP-dependent lacZ reporter strains, and on empirical observation. Direct measurements of H. influenzae cAMP concentrations have not been achieved (57). The stimulation of competence by exogenous cAMP led to the hypothesis that catabolite repression might be important in regulating natural transformation in H. influenzae. This prediction was confirmed by the subsequent isolation and characterization of genes for proteins homologous to those that mediate catabolite repression in E. coli. The H. influenzae homologues of both the cyclic AMP receptor protein regulatory protein (Crp) (44) and Adenylate Cyclase, the enzyme that synthesizes cAMP, are essential for transformation, as null mutations in either of these genes prevent competence development (44, 57). In E. coli, cAMP levels are regulated by a variety of mechanisms, including the regulation of adenylate cyclase activity and the degradation or active excretion of cAMP from cells (18). The phosphotransferase (PTS) system of E. coli regulates intracellular cAMP levels by activating Adenylate Cyclase in the absence of available PTS sugar. In H. influenzae, a simple phosphotransferase system for the utilization of fructose has been revealed both by sequencing of the H. influenzae genome and by experimental studies (68,84,120,121). H. influenzae strains carrying null mutants of the ptsl (EI) and err (EIIAG l u) genes have reduced transformation frequencies that are increased to wild type levels upon the addition of exogenous cAMP, indicating that the PTS components of H. influenzae interact with Adenylate Cyclase to regulate cAMP levels and competence development. The H. influenzae ice gene product encodes a homologue of an E, coli 3'-5' cAMP phosphodiesterase that moderates intracellular cAMP concentrations by degrading excess cAMP (122). Strains carrying ice null mutations, expected to have elevated cAMP 15 concentrations, were observed to exhibit elevated transformation frequencies during logarithmic growth (122). 1.3.2. A cAMP-independent signal is required for maximal competence induction. The experimental results discussed above indicate that competence is induced when cAMP concentrations are high. However, the 100-fold higher transformation frequencies of cells in MIV compared to those in stationary phase in rich medium is not currently understood. It has been proposed that this difference exists because relief of catabolite repression is necessary but not sufficient for maximal competence development in MIV, and that a second regulatory signal is required for maximal competence devlopment (192). In addition to genes known to play a role in catabolite repression, competence development in H. influenzae also appears to be tightly linked to the expression of another gene, sxy, whose product is thought to encode an activator of competence. High concentrations of Sxy, induced upon starvation of cells, may constitute the second regulatory event required for maximal competence development. Several lines of evidence, discussed comprehensively in Chapter 3 (Sections 3.1.1 and 3.1.2), indicate that Sxy is an early competence gene which activates the expression of genes required for the mechanistic steps of transformation. Strains carrying sxy null mutations are transformation-deficient, showing that Sxy is essential for competence development. In addition, the overexpression of sxy in cells in rich medium results in high levels of transformation achieved in wild type cells only when they are in MIV. Finally, a number of point mutations within sxy cause an approximate 100-fold activation of transformation in cells in rich medium (155,193) (Redfield, unpublished). These mutations are proposed to exert their affects by decreasing the stability of a sxy 5' 16 mRNA secondary structure which negatively regulates sxy expression (Redfield, unpublished). In Chapter 3,1 discuss the function of Sxy and the regulation of sxy expression in sBHI and in MIV. I present evidence that both transformable and nontransformable bacterial strains contain Sxy homologs, indicating that the function of Sxy may not be specific to competence development. I also show that sxy expression is negatively regulated by an mRNA secondary structure, and that the sxy hypercompetence point mutations increase sxy expression through destabilization of this structure. My results also show that Sxy concentrations are probably not higher in starvation medium than in rich medium, suggesting that an increase in Sxy expression is probably not the second regulatory signal required for the induction of maximal competence in starvation medium. In Chapter 4,1 discuss mutations in the H. influenzae fis gene that reduce competence development in rich medium. After discovering that Tn916 mutations in the KW20 fis operon delayed spontaneous competence development, I created a fis null mutant by cassette mutagenesis, and determined that/z's mutants in rich medium are 500-fold less competent than wild type cells. I discuss the implications for the possible role of Fis competence development. 17 Chapter 2; Materials and Methods 2.1. General Methods 2.1.1. Strains and plasmids H. influenzae and E. coli strains, plasmids and phage are listed in Tables 2.1 and 2.2. All H. influenzae strains are derivatives of Alexander and Leidy's original Rd strain (14). Plasmids pLZK80 and pLZK83 were obtained from G. Barcak (26), plasmids pSU20 and pSU21 were obtained from B. Bartolome (29), pAM120 was obtained from D. Clewell (70), and pKRP13 was obtained from G. Phillips (157). pAlter-1 and pGEM-7Zf- were purchased from Promega. The phage XNK1316 was obtained from N. Kleckner (103). 2.1.2. Culture methods 2.1.2.1. Haemophilus influenzae culture methods H. influenzae cells were routinely cultured by shaking (in flasks) or rolling (in tubes) at 37°C in Brain Heart Infusion Broth (BHI; Difco) supplemented with Nicotinamide Adenine Dinucleotide (NAD) at 2 mg/ml and hemin at 10 ug/ml (sBHI or rich medium). For plating, cells were serially diluted in a solution containing 1% phosphate-buffered saline (PBS) and 10% BHI and inoculated on plates containing sBHI plus 1.2% Bacto agar (Difco). For scoring (3-galactosidase activity, cells were plated on sBHI agar containing 0.008% 5-Bromo-4-chloro-3-indolyl-(3-D-galactopyranoside (X-GAL). Antibiotics were used at the following concentrations: novobiocin, 2.5 ug/ml; kanamycin, 7ug/ml; chloramphenicol, lug/ml; tetracycline, 5 ug/ml; spectinomycin 15 ug/ml; streptomycin, 250 ug/ml. 18 Table 2.1. Bacterial strains used in this study Strains Relevant Genotype Source or reference H. influenzae Rd KW20 Wild type Alexander and Leidy (15) RR514 kans nals novs strr R. Redfield (155) RR518 kans nals novr strs R. Redfield (155) RR520 kanr nals novs strs R. Redfield (155) RR648 sxy: miniTnl Okan P. Williams (192) RR736 cyav.cat L. Macfadyen (120) MAP7 kanr nal r novr strr vio r J. Setlow (26) RR699 sxy-1 P. Williams (192) RR700 sxy-2 R. Redfield, unpublished RR724 sxy-3 R. Redfield, unpublished RR723 sxy-4 R. Redfield, unpublished RR850 sxy-5 R. Redfield, unpublished RR852 sxy-6 This study RR854 sxy-7 This study RR844 sxy$9::lacZkan operon fusion This study RR845 sxyg9::lacZkan protein fusion This study RR846 sxyu::lacZkan operon fusion This study RR847 sxyndacZkan protein fusion This study RR848 sxy-1 S9::lacZkan operon fusion This study RR849 sxy-lsg::lacZkan protein fusion This study RR856 sxy-6S9::lacZkan operon fusion This study RR857 sxy-6S9::lacZkan protein fusion This study 19 Table 2.1. continued.. Strains Relevant Genotype Source or reference H. influenzae Rd continued... RR860 sxy-7 S9:.lacZkan operon fusion This study RR861 sxy-7S9::lacZkan protein fusion This study RR874 cya sxyS9::lacZ kan operon fusion This study RR872 fis::Tn916 This study RR858 fisv.spec This study RR859 orfl:: spec This study RR873 sxy-1 fis This study E. coli DH5oc recAl D. Hanahan JM109 recAl Promega ES1301 mutS mutSlOl Promega LE392 supE N. Murray NM554 A(lac) N. Murray GM2163 dam New England Biolabs 20 Plasmid Table 2.2. Plasmids and phage used in this study Relevant genotype or phenotype Source or reference pAlter-1 Plasmid for site-directed mutageneisis (AmpsTetr) pGEM7-Zf- pBR322 derivative (Ampr) pSU20 pACYC derivative (Cmr) pSU21 pACYC derivative (Cmr) pSU2718 pACYC derivative (Cmr) pAlter sxy 1.8 kb EcoRI - BamHI fragment of pDJM90 cloned into pAlter-1 (AmpsTetr) pAltersxy3 1.8 kb EcoRI - BamHI fragment of pRRS3 cloned into pAlter-1 (AmpsTetr) pAltersxy5 pAlter-1 containing the sxy-5 site-directed mutation (AmprTets) pAltersxy6 pAlter-1 containing the sxy-6 site-directed mutation (AmpTet8) pAltersxy7 pAlter-1 containing the sxy-7 site-directed mutation (AmprTets) pAM120 Tn926 cloned onto the EcoRI site of pGLl (AmprTetr) pDJM90 1.8 kb fragment containing rec-1 and 5' half of sxy cloned into the EcoRI site of pGEM7 (Ampr) pGEMsxy sxy, including transcriptional start site, cloned into the Apal and EcoRI sites of pGEM7 (Ampr) pGEMsxy-1 sxy-1, including transcriptional start site, cloned into the Apal and EcoRI site of pGEM7 (Amp r) Promega Promega Bartolome (29) Bartolome (29) M. Chandler This study This study This study This study This study Gawron-Burke (70) Williams (192) This study This study pKRP13 Spectinomycin/streptomycin resistance cassette Phillips (158) 21 Table 2.2. continued... Plasmid Relevant genotype or phenotype Source or reference pLBFl Clal -BamHl fragment containing the KW20 fis This study operon cloned into pSU21 (Cmr) pLBF2 Spec1- cassette cloned into the SnaBl site of pLBFl This study (CmrSpr) pLBF3 Specr cassette cloned into the EcoRl site of pLBFl This study (CmrSpr) pLBS5 sxy-5 mutation subcloned into pDJM90 (Ampr) This study pLBS6 sxy-6 mutation subcloned into pDJM90 (Ampr) This study pLBS7 sxy-7 mutation subcloned into pDJM90 (Ampr) This study pLBS6K pLBS6 containing a miniTnlOkan inertion (Kanr) This study that disrupts the ampicillin resistance gene pLBS7K pLBS7 containing a mini TnlOkan inertion (Kanr) This study that disrupts the ampicillin resistance gene pLBSFl pLBSF2 pLBSF3 pLBSF4 pLBSF5 pLBSF6 LacZKan r operon fusion cassette cloned into the This study Bell site (at Sxy codon 89) of pDJM90 (Amp rKan r) LacZKan 1 protein fusion cassette cloned into the This study Bell site (at Sxy codon 89) of pDJM90 (AmprKanr) LacZKan r operon fusion cassette cloned into an This study engineered EcoRl site (at Sxy codon 11) within pDTM90 (AmprKanr) LacZKan r protein fusion cassette cloned into an This study engineered EcoRl site (at Sxy codon 11) within pDJM90 (AmpTCanO LacZKan1" operon fusion cassette cloned into the This study Bell site (at Sxy codon 89) of pRRS6 (AmpTCan1-) LacZKan1" protein fusion cassette cloned into the This study Bell site (at Sxy codon 89) of pRRS6 (AmpTCan1*) 22 Plasmid Table 2.2. continued... Relevant genotype or phenotype Source or reference pLBSF7 LacZKan r operon fusion cassette cloned into the Bell site (at Sxy codon 89) of pLBS6 (AmprKanr) pLBSF8 LacZKan r protein fusion cassette cloned into the Bell site (at Sxy codon 89) of pLBS6 (AmpTCan1-) pLBSF9 LacZKan r operon fusion cassette cloned into the Bell site (at Sxy codon 89) of pLBS7 (AmprKan1") pLBSFlO LacZKan r protein fusion cassette cloned into the Bell site (at Sxy codon 89) of pLBS7 (Amp^n 1") pLBTnl Mutant B-l 18 kb chromosomal fragment containing Tn926 and surrounding DNA cloned into the Kpnl site ofpSU20 (CmrTetr) pLZK80 LacZKan cassette for generating operon fusions (AmprKanr> pLZK81 LacZKan cassette for generating protein fusions (AmprKanr) pRRNovl Nov r gene of KW20 cloned into pSU2718 (Cm rNov r) pRRS3 sxy-3 mutation subcloned into pDJM90 (Ampr) pRRS6 sxy-1 mutation subcloned into pDJM90 (AmpO pRRS6K pRRS6 containing a miniTnl Okan inertion (Kanr) that disrupts the ampicillin resistance gene This study This study This study This study This study G. Barcak G. Barcak Redfield (155) Redfield, (unpublished) Redfield (unpublished) This study Phage Relevant genotype or phenotype Source or reference XNK1316 xmmTnlOkan N. Kleckner / 23 2.1.2.2. Escherichia coli cul ture methods E. coli cells were routinely cultured at 37°C in Luria-Bertani (LB) broth (162). For plating, cells were serially diluted (1/100 or 1/1000) with 1% PBS and inoculated on plates containing LB plus 1.2% Bacto agar. For scoring P-galactosidase activity, cells were plated on LB agar containing 0.008% X-Gal, or on MacConkey agar (Difco) plates containing 1% lactose. Antibiotics were used at the following concentrations: kanamycin, 50 ug/ml; chloramphenicol, 25 ug/ml; tetracycline, 10 ug/ml; spectinomycin, 100 ug/ml; streptomycin, 50 ug/ml; ampicillin, 100 ug/ml. 2.1.3. Media sBHI, MacConkey lactose and LB media were purchased from Difco. Non-commercial media were prepared as follows: i) Terrific broth (162): Components, per litre (pH 7.0): Bacto-yeast extract 12 g Bacto-tryptone 24 g Glycerol 4 ml K H 2 P 0 4 K2HPO4 2.31 g 12.54 g ii) MIV: Components, per litre (pH 7.0): Sodium chloride Asparate KH 2P04 4.7 g 1.74 g 24 MIV components per litre continued... Fumarate 1 g Glutamate 314 mg Tween-80 200 mg CaCl 2 I4 7 m g MgS0 4 1 2 4 m g Serine 65 mg Leucine 61 mg Proline 50 mg Alanine 48 mg Phenylalanine 46 mg Tyrosine 42 mg Lysine 35 mg Valine 35 mg Isoleucine 33 mg Arginine 21 mg Alanine 20 mg Methionine 18 mg Histidine 13 mg Citrulline 12 mg Cystine 6 mg Glycine 2.5 mg 2.1.4. Natural transformation of H. influenzae and artificial transformation of E. coli 2.1.4.1. H. influenzae transformation i) Transformation in sBHI. H. influenzae cells were diluted from cultures grown overnight in sBHI, and grown with shaking to permit several generations of exponential growth (OD6001L0.05) before the first time point samples were taken. Cultures were then sampled at regular time intervals. 1 ml of cells was withdrawn from the main culture, transferred to a 16 X 100 ml culture tube, and incubated for 15 minutes at 37°C with 1 ug of novobiocin-resistant DNA from strain MAP7 (26). Excess DNA was then degraded by treatment with 1 ug of DNase for 5 minutes. Cells were either plated immediately, or grown for an additional 30 minutes to one hour to allow cells to express antibiotic resistance. Cells were diluted and plated on sBHI agar to determine cfu/ml, and on sBHI agar containing the appropriate antibiotic to select for 25 transformants. Transformation frequency was scored as the fraction of total cells which had acquired antibiotic resistance. ii) Transformation in MIV. Cells were diluted from cultures grown overnight in sBHI and grown to permit several generations of exponential growth. When cells reached a density of approximately 10^  cfu/ml (ODgnn of 0.2-0.25), 15 ml of cells was collected by filtration using a Nalgene Analytical Test Filter Funnel (0.2 um pore size), and rinsed with an equal volume of MIV (191). Cells were then resuspended in an equal volume of MIV and shaken (100 RPM) at 37°C. For time courses of competence development, 1 ml samples of cells were removed at regular time intervals and assayed for competence as described above. iii) Colony assays. Colony assays were used to crudely assay the competence of a large number of strains simultaneously. Colonies were grown for approximately one day on sBHI agar. Individual colonies were inoculated into 5 ml of prewarmed sBHI containing 0.1 ug/ml of antibiotic-resistant DNA from MAP7. After 15 minutes at 37°C, cells were diluted and plated on sBHI agar (diluted cultures) or on sBHI agar containing novobiocin or kanamycin (undiluted cultures). Transformation frequency was scored as previously described. iv) Plasmid transformations. Circular plasmid DNA lacks free ends and thus transforms H. influenzae inefficiently. Plasmid transformations of KW20 and related strains were carried out using a previously described method of enhancing plasmid transformation by treating MlV-competent cells with 32% glycerol (176). 26 2.1.4.2. £. coli plasmid transformation E. coli cells were made artificially competent with CaCl2 (163), with the exception of strain ES1301rau£S, which was made competent with a modification of the RbCl technique (Promega Altered Sites II in vitro Mutagenesis System Technical Manual). 2.1.5. Transposon Mutagenesis 2.1.5.1. Mutagenesis with miniTnlOfcan Titering of phage, preparation of phage DNA, and transposon mutagenesis with miniTnl Okan were carried out as described in Kleckner and Gottesmann (103). Phage ?iNK1316 was titered on strain LE392 (supE). For transposon mutagenesis, ANK1316 was used to infect NM554 cells(swp°) carrying the appropriate plasmid. Cells were grown to late exponential phase (OD600 of approximately 0.75) in TBMM [Tryptone broth (Difco) containing 0.2% maltose and 0.01 M MgS04] and concentrated approximately 10-fold by centrifugation and resuspension in LB. Concentrated cells (0.1 ml) were then infected with 0.1 ml of undiluted (approximately 5 X 10^ pfu/mL) or diluted phage for 15 minutes at room temperature and 15 minutes at 37°C. Unadsorbed phage were removed by washing cell/phage mixtures with LB containing 0.05 M sodium citrate. Infected cells were diluted and plated on TBI agar (Tryptone broth plus 1.2% Bacto Agar) containing kanamycin (30 ug/ml) and the antibiotic encoded by the plasmid resistance gene. To isolate plasmids with transposon insertions, plasmid DNA was isolated from pools of colonies containing transposon insertions, and used to retransform NM554 to kanamycin resistance and plasmid-encoded antibiotic resistance. Transposon insertion sites were mapped by restriction enzyme analysis. 27 2.1.5.2. Tn916 mutagenesis Tn926 mutagenesis was carried out as described previously (101). pAM120 (at a final concentration of 200 ng/ml) was used to transform MlV-competent KW20 by glycerol-mediated plasmid transformation. Strains that had acquired transposon insertions were selected on sBHI agar containing tetracycline. 2.1.6. DNA methods 2.1.6.1. Isolation of plasmid and chromosomal DNAs i) Isolation of plasmid DNA. For the preparation of plasmid DNA, H. influenzae cells were grown in sBHI, and E. coli cells were grown in either LB or Terrific broth (163). Plasmid DNA was isolated using the alkaline lysis protocol (162). Large scale plasmid preparations, to isolate template DNA for sequence analysis or for in vitro transcription, were further purified by precipitation with LiCl and polyethylene glycol (PEG 8000; Sigma) (22). ii) Isolation of high molecular weight chromosomal DNA. High molecular weight H. influenzae chromosomal DNA was used for bacterial transformations, Southern analysis, and PCR (polymerase chain reaction). 10 ml of bacterial cells were grown to late exponential phase of growth in sBHI (OD600 approximately 1.0), and extraction of chromosomal DNA was carried out as previously described (26). Briefly, cells were pelleted and resuspended in 0.15 M NaCI, 0.1 M EDTA; pH 8.0, and subsequently lysed by treatment with 1% sodium dodecyl sulfate (SDS) for 10 minutes at 52°C. The lysed extract was treated with proteinase K at 50 ug/ml for 1 to 2 hours at 37°C, and subsequently extracted with an equal volume of phenol chloroform. DNA was precipitated by adding 2 volumes of 95% ethanol, and the precipitated DNA was 28 isolated by spooling it out of solution with a glass rod. Spooled DNA was dried at room temperature for 30 minutes to 1 hour, and then resuspended in 200 to 500 pi of TE8 [10 mM Tris-HCl, pH 8.0; 1 mM ethylenediamine-tetraacetic acid (EDTA)]. Resuspended DNA was treated with RNase A (0.2 mg/ml, Sigma) at 37°C for 30 minutes to 1 hour. RNase-treated DNA was further purified by subsequent extractions with equal volumes of phenol, phenol chloroform, and chloroform. Extracted DNA was re-precipitated with 2 volumes of 95% ethanol and 0.15 M NaCI, and resuspended in 200 to 500 pi of TE8. 2.1.6.2. DNA sequencing Automated DNA sequencing of plasmids and PCR products was carried out by the University of British Columbia Nucleic Acid-Protein Service (NAPS) unit, using AmpliTaq Dye Terminator Cycle Sequencing chemistry. 2.1.6.3. DNA agarose electrophoresis and cloning Restriction analysis of DNA, separation of DNA fragments by agarose electrophoresis, and DNA cloning were carried out according to Sambrook et al. (162). Partial restriction digests were carried out by varying the amount of restriction enzyme (22). DNA 'i fragments were separated on 0.6 to 0.8% Tris Acetate EDTA (TAE, pH 8) agarose gels (ultraPure agarose, Gibco BRL) containing 0.25 ug/ml ethidium bromide. Separated DNA fragments were photographed under UV light. A, Hindlll and 1 kb ladders (Gibco BRL) were used as size markers for agarose electrophoresis. DNA fragments were purified for cloning by excision from agarose gels and subsequent isolation using a Gene-Clean Kit (Bio 101 inc.). DNA ligations were carried out according to Sambrook et al. (162) using T4 DNA ligase (Boerhinger Mannheim). 15% PEG8000 was added to 29 blunt-end ligation reactions (162). Concentrations of DNAs, oligonucleotide primers and RNAs were determined using a Beckman Du-65 Spectrophotometer. 2.1.6.4. Polymerase chain reaction Primers used in PCR were designed using the program Amplify (version 1.2 (3; W. Engels, University of Wisconsin, Genetics), using sequence information obtained from the H. influenzae Rd database at The Institute For Genomic Research (T.I.G.R.; 2, 68). Primers used to amplify regions of the KW20 genome containing sxy are shown in Figure 2.1, and primers used to amplify the KW20/z's operon are shown in Figure 2.2. All primers were synthesized by the UBC NAPS Unit, using Applied Biosystems Oligonucleotide Synthesizers. Primers were resuspended in dH20 and primer concentrations were determined by spectroscopy at OD260- PCR was carried out in volumes of 100 uL in 0.5 ml microfuge tubes; a standard PCR contained IX PCR buffer (Gibco BRL), MgCl2 (1-10 mM), primers (500 nM), dNTPs (240 uM, Boehringer Mannheim) and Taq polymerase (approximately 5 units, Gibco BRL). Taq extender (Stratagene) was included in PCR reactions when the PCR product was expected to be > 2 kb in length. PCR was performed using a Perkin Elmer Cetus DNA Thermal Cycler 480 machine. Standard PCR amplification parameters were as follows, where Tm was the calculated annealing temperature for the primer set: [1 X (94°C for 2 minutes.; Tm for 1 minute; 72°C for 1 minute, 30 seconds)]; [30 X (92°C for 15 seconds; Tm for 1 minute; 72°C for 1 minute, 30 seconds)]; [1 X (92°C for 15 seconds; Tm for 1 minute; 72°C for 10 minutes)]. PCR products were photographed after separation on 2.0% TAE agarose gels, and purified either by precipitation with 2.5 M NH4CI and 1 volume of 95% ethanol (22), or with QIAquick PCR purification kits (Qiagen). 2.1.6.5. Southern analysis Southern blotting was performed using the nonradioactive digoxigenin (DIG) system of Boehringer Mannheim, according to the modified Southern protocol of 30 I Primers that anneal to the top strand f I Primer 6 (-288 to-267) EcoRI I 5'GAATTCTGTGATTATATCTGTATTGATG 3* | Primer 13 (-52 to-32) Aoa\ I 5 ' AAAGGGCCCCAGAAGTACTTCTACTGACTC 3 ' | I Primer 4 (+182 to+201) | 5 ' AACTAAATTAGGTTGTGAAC 3 ' I Primers that anneal to the bottom strand f | Primer 15 (+11 to+20) I 3'CCTACTCGTATATCTATCGCCTJAMfifiA 5 ' EcoRI I Primer 10 (+153 to+172) I 3 ' GCGCCACTCCCACAAGAGCG 5 ' I | Primer 12 (+659 to+678) I 3 ' CMGTMGTTTATTAAAAMGCTJMCXAJ | EcoRI Figure 2.1. Sequences and annealing sites for sxy primers. The +1 position denotes the A of the ATG translational start site for the sxy open reading frame. 31 F S l 5 ' CTTGGTGCGAAAAGTGCGAAAAGTGCGGTGGG 3 ' C/al \ /C/al 5 ' I V I 3- r orf1 1 kb EcoRl SnaBIBglH ^, 3 ' GGCTAAAACTGAAAACGGGTGCG 5 ' F S 2 Figure 2.2. Location and sequences of primers FSl and FS2 used to amplify the KW20 fis operon by PCR. 32 Engler-Blum (66). Bacterial chromosomal DNA and /or plasmid DNA (approximately 1 ug) was digested with the appropriate restriction enzyme(s), and digested DNA fragments were separated on 0.6 to 0.8% TAE agarose gels. Electrophoresed DNA was subsequently transferred to nitrocellulose membranes (Hybond N+, Amersham) using a rapid, downward alkaline Southern blotting technique (106). Membranes were rinsed in 2 X SSC (standard saline citrate; 3 M NaCl, 0.3 M sodium citrate, pH 7.0) and DNA was fixed by baking membranes at 80°C for 2 hours. DNA probes were generated by restriction digestion of plasmid DNA followed by random-primed labelling with DIG-dUTP (Boehringer Mannheim). Membranes were prehybridized at 68°C for several hours in prehybridization solution [25 M Na2HP04, pH 7.2; 1 mM EDTA, 7% SDS, 0.5% blocking reagent (Boeringher Mannheim)] and subsequently hybridized overnight at 68°C in hybridization solution (prehybridization solution with 10 pmol/ml of DIG-labeled probe). Membranes were then washed 3 times at 65°C with 50 ml of hybridization wash buffer (20 mM Na2HPC>4; 1 mM EDTA, 1% SDS) and subsequently incubated for 1 hour in block buffer (0.1 M maleic acid; 3 M NaCl; 0.3% Tween 20, pH 8 and 0.5% blocking reagent). Membranes were then treated for 30 minutes at room temperature with diluted (1: 7500) anti-DIG antibody (Boehringer Mannheim). Excess anti-Dig antibody was removed by washing membranes 4 times with 50 ml of washing buffer (0.1 M maleic acid; 3 M NaCl; 0.3% Tween 20, pH 8). Membranes were then equilibrated in substrate buffer (0.1 M Tris-HCI; 0.1 M NaCl; 50 mM MgCl2) and incubated in darkness for 5 minutes with diluted (1:100) CSPD (Disodium 3-(4-methoxyspiro{l,2-dioxetane -3, 2'-(5'chloro) tricyclo [3.3.1.1 /^7] decan}-4-yl) phenyl phosphate; Boehringer Mannheim). Membranes were then exposed on X-ray film for varying amounts of time for signal detection. 33 2.2. Sxy methods 2.2.1. Sxy homology searches The Sxy 217 amino acid protein sequence was searched against the unfinished microbial genomes database at the National Centre for Biotechnological Information (NCBI) using Blast [Blast version 2.0; (17) (11)]. Protein sequences of Sxy and putative Sxy homologs were used in searches against known protein motifs in the Prints and Blocks protein databases using the program eMotif Search (1,20). Individual protein sequences were searched for putative transmembrane segments and analyzed for charged clusters, amino acid content and repetitive amino acid clusters using the program SAPS (Statistical Analysis of Protein Sequence; 40) at the Baylor College of Medicine Protein Secondary Structure Prediction web site (4). Individual protein sequences were also analyzed for the presence of a helix turn helix (HTH) DNA binding motif, using an HTH Motif Prediction web program (56,7), and for the presence of leucine zippers using the leucine zipper identification program 2Zip (3,38). Sxy and putative Sxy homologue amino acid sequences were aligned using Clustal W 1.7 (178) at the Baylor College of Medicine Multiple Sequence Alignments web page (5). Conserved blocks of homology were identified from aligned sequences with the web program Block Maker (86) at the Baylor College of Medicine Multiple Sequence Alignments web page (5). Conserved motifs within the aligned protein sequences were used to search the Blocks and Prints databases using the web programs L A M A [Local Alignment of Multiple Alignments, (8,150)] and MAST [Motif Alignment and Search Tool, version 2.2; (10,23)]. 34 2.2.2. Construction of sxy.dacZ operon and protein fusions 2.2.2.1. Construction of operon and protein fusions of lacZ to sxy codon 89 pBR322-derived plasmids containing lacZkan cassettes for generating lacZ operon (pLZK83) and protein (pLZK80) fusions were obtained from Barcak (26). The cloning scheme for creation of sxyv.lacZ fusions is shown in Figure 2.3. pLZK80 and pLZK81 were digested with BamHl, and the lacZkan cassettes (~4.5 kb) were ligated to pDJM90 (sxy), pRRS6 (sxy-1), pLBS6 (sxy-6) and pLBS7 (sxy-7) digested with Bell (Bell cuts within codon 89 of sxy in each of these plasmids). Ligation mixtures were transformed into DH5a, and transformants were selected on LB agar containing ampicillin and kanamycin. Recombinant plasmids were screened by restriction analysis using Sacl + EcoRl and BamHl + EcoRV digests. Plasmids with operon fusions in the + orientation were named pLBSFl (sxyS9"lacZkan operon fusion); pLBSF5 (sxy-l89--lacZkan operon fusion); pLBSF7 (sxy-6s9'-'-lacZkan operon fusion) and pLBSF9 (sxy-7%9::lacZkan operon fusion). Plasmids with protein fusions in the + orientation were named pLBSF2 (sxyggv.lacZkan protein fusion); pLBSF5 (sxy-2 ggv.lacZkan protein fusion); pLBSF7 (sxy-6%9\:lacZkan protein fusion) and pLBSF9 (sxy-7%9::lacZkan protein fusion). Plasmid inserts were transferred to the KW20 or RR736 (cya) genome to generate single copy chromosomal fusions. This was accomplished by digesting plasmid DNA with Apal + BamHl to release the inserts, and transforming MlV-competent KW20 or RR736 cells to kanamycin resistance using linearized DNA. KW20 transformants were screened for P-galactosidase activity on sBHI agar containing X-GAL. 35 pLZK80 lacZwkan cassette FrnRT Figure 2.3. Creation of pLBSFl (sxy89::lacZ operon fusion plasmid) by cloning the LacZKan operon fusion cassette of pLZK80 into the unique Bell site of pDJM90. Cloning strategies to create pLBSF5, pLBSF7 and pLBSF9 were equivalent, except that the cassette was cloned into the Bell sites of pRRS6, pLBS6 and pLBS7, respectively. pLBSF2 (sxy89::lacZ protein fusion plasmid) was created by digesting pLZK81 (containing a lacZkan protein fusion cassette) with BamHI and cloning the cassette into the Bell site of pDJM90. Cloning strategies to ceate pLBSF6, pLBSF8 and pLBSFlO were equivalent, except that the cassette was cloned into the Bell sites of the plasmids pRRS6, pLBS6 and pLBSF7, respectively. 36 2.2.2.2. Construction of operon and protein fusions of lacZ to sxy codon 11 I constructed lacZ operon and protein fusions to sxy codon 11 by engineering an EcoRl restriction site within the wild type sxy sequence. The cloning strategy for construction of these fusions is outlined in Figure 2.4. Primer 6 and Primer 15 (containing an EcoRl site at its 5' end) were used to amplify sxy KW20 genomic DNA. PCR generated a -355 bp product that was subsequently cleaved with Seal + EcoRl to generate a~333 bp fragment. pLZK80 and pLZK81 were digested with EcoRl + Bell to generate -1400 bp DNA fragments containing the 5' half of lacZ gene. The digested PCR product and 1400 bp fragments containing the 5' half of the lacZ gene were ligated, and the ligation mixtures were subsequently ligated to pLBSFl that had been digested to completion with Bell and partially digested with Seal. The resultant plasmids were screened by restriction analysis with EcoRl, Apal + BamHl, and Apal + SnaBl digests. Strains carrying recombinant plasmids with the expected restriction enzyme digestion patterns were scored for (3-galactosidase activity on MacConkey lactose agar. Plasmids with the desired insertions were named pLBSF3 (sxynv.lacZ operon fusion) and pLBSF4 (sxy\\::lacZ protein fusion). Plasmid inserts were transferred to the KW20 genome to generate single copy chromosomal fusions. Plasmid DNA was digested with Apal + SnaBl to release the inserts, and linearized DNA was used to transform MlV-competent KW20 to kanamycin by additive replacement. KW20 transformants were screened for (3-galactosidase activity on sBHI agar containing X-GAL. 2.2.3. p-galactosidase assays P-galactosidase assays were carried out according to a modified protocol derived from Miller (133,160). H. influenzae cells were grown in sBHI or MIV and sampled 37 KW20 genome P6 <t rec-1 •*xy S e a l P 1 5 EcoRI 1. PCR with P6 and P15 2. Digest PCR product with Seal and EcoRI FrnRT RmHT p L Z K 8 0 lacZr.kan cassette EcoRV B e l l S p c l kan R«TtiT FrnRT 3. Digest pLZK80 with EcoRI and Bell and ligate to sxy PCR product digested with Seal and EcoRI * sxy' •sxv' RnnHT EcoRV B e l l I I W///////////A S e a l EcoRI A p a l 4. Clone ligation mixture into pLBSFl digested with Seal (partial digestion) and Bell Figure 2.4. Creation of pLBSF3 (sxy^::/flcZ operon fusion plasmid) by cloning the LacZkan operon cassette of pLZK80 into an engineered EcoRI site at codon 11 of sxy. The cloning strategy to create pLBS4 (sxy^::kcZ protein fusion plasmid) was equivalent to the procedure shown except that the lacZKan protein cassette of pLZK81 was used in step 3. 38 in duplicate at regular time intervals. For cells growing in sBHI in mid to late logarithmic growth (OD600 ^ 0.05), and for cells in MIV, 0.1 ml of cells was usually sampled; for cells growing in sBHI during early exponential growth, larger samples of cells were taken and concentrated by centrifugation and/or filtration. After sampling, cells were immediately pelleted by centrifugation at 13 kRPM in a microcentrifuge. Supernatants were removed using pasteur pipettes, and cell pellets were frozen in microfuge tubes at -80°C. Simultaneously, the main cell culture was assayed for OD600 and, in some cases, for cfu/ml. For (i-galactosidase assays, frozen pellets were thawed on ice and resuspended in 1.0 ml of Z buffer (60 mM Na2HP04/7H20; 40 mM NaH2P04-H20; 10 mM KC1; 1 mM MgS04-7H20; 50 mM f3-mercaptoethanol) containing o-nitrophenyl-B-D-galactopyranoside (ONPG; 4 mg/ml); sodium deoxycholate (0.1 mg/ml), and hexadecyltrimethylammonium bromide (0.2 mg/ml). Assays were allowed to proceed at 28°C until tube contents had turned yellow, and terminated by the addition of 0.5 ml 1 M Na2CC>3. Following termination of the assay, cell debris was pelleted by briefly centrifuging tubes at 13 kRPM in a microfuge. Tube contents were monitored at OD420 to quantify yellow colour (product) and also at OD550 to correct for cell debris. Measurements were carried out using a LKB Novaspec II spectrophotometer. Miller units of fj-galactosidase activity were calculated using the following formula: Miller units = 1000 X [OD420 - (1.75 X OD550) ] / (t X v X OD6oo) where t = time of reaction and v = volume of cells assayed (133). 2.2.4. Cloning of sxy and sxy-1 into pGEM7 for RNA secondary structure mapping For in vitro transcription with T7 RNA polymerase, sxy and sxy-1 were cloned into pGEM7 immediately downstream of the T7 promoter. The cloning scheme is shown in Figure 2.5. Primer 12 (containing an EcoRI site at its 5' end) and Primer 13 (containing an Apal site at its 5' end) were used to amplify KW20 (sxy) and RR699 (sxy-1) genomic 39 DNA. The PCR products (approximately 750 bp) were digested with Apal + EcoRl and ligated to pGEM7 digested with Apal + EcoRl. The ligation mixtures were transformed into DH5oc, and recombinant plasmids containing the desired inserts were identified by restriction analysis with EcoRl + Apal, Seal + EcoRl, and Clal digestions. Plasmid insert sequences were confirmed by sequencing with Ml3 forward and reverse primers. Plasmids with the desired inserts were named pGEMsxy and pGEMsxy-1. 2.2.5. Sxy and sxy-1 RNA secondary structure mapping 2.2.5.1. In vitro RNA transcription pGEMsxy and pGEMsxy-1 were linearized with SnaBl. This enzyme cuts within sxy and sxy-1, downstream of the mRNA sequences predicted to form the regulatory RNA secondary structure. Linearized template DNAs were separated from uncut plasmid DNA on TAE agarose gels, and purified by dialysis electroelution using SPECTRA/POR 1 membrane tubing (162). Unlabeled sxy and sxy-1 RNA were transcribed from the pGEM T7 promoter in reactions containing ~2 ug of SnaBI-linearized plasmid DNA, 100 units of T7 RNA polymerase (Pharamcia), rNTPs (500 uM, Pharmacia), DTT (10 mM), IX T7 transcription salts [40 mM Tris-HCI, pH 8; 8 mM MgCl 2; 2 mM spermidine-(HCl3); 25 mM NaCl], RNA guard (50 units, Pharmacia) and diethylpyrocarbonate (DEPC)-treated dH20. Reactions were allowed to proceed at 37°C for 45 minutes, at which time 50 additional units of T7 polymerase were added to the reaction. After 30 minutes of further reaction time, RNA samples were diluted with 5 mM EDTA, 2 M N H 4 O A C , 0.1 mg/ml E. coli tRNA and DEPC-treated dH20. Diluted RNA samples were then extracted with an equal volume of phenol chloroform, precipitated overnight at -20°C with 2.5 volumes of 95% ethanol, and centrifuged. Pellets were rinsed with 80% ethanol, dried at room temperature, and redissolved in 100 pi of DEPC-treated dH 2 0 . Full-length transcripts were purified away from abortive transcripts by passing RNA samples through Microspin-S-200 HR columns (Pharmacia). 40 KW20 genome Apal P13 rec-7 sxy v. P72 EcoRl 1. PCR withP13 and P12 2. Digest PCR product with Apal and EcoRl 3. Ligate into pGEM7 digested withApal and EcoRl Apal SnaBl EcoRl Figure 2.5. Cloning strategy for the creation of pGEMsxy. The cloning strategy for the creation of pGEMsxy-1 was equivalent except that PCR (step 1) was carried out using RR699 (sxy-1) genomic template DNA. 41 2.2.5.2. RNase secondary structure mapping RNA secondary structure mapping was carried out as previously described (123). Prior to end-labeling, RNA samples were dephosphorylated in reactions containing ~2 pmol unlabeled RNA, 150 units of BAP (Bacterial Alkaline Phosphatase, Gibco BRL), 1 X BAP buffer (10 mM Tris-HCl, pH 8) and DEPC-treated d H 2 0 . Dephosphorylation reactions were allowed to proceed at 37°C for 45 minutes, and then were diluted with 0.3 M NaOAc, 10 mM EDTA and DEPC-treated d H 2 0 . Diluted RNAs were extracted with an equal volume of phenol chloroform, precipitated overnight at -20° with 2.5 volumes of 95% ethanol, and centrifuged. Pellets were rinsed with 80% ethanol, dried at room temperature, and redissolved in 100 pi of DEPC-treated d H 2 0 . RNAs were end-labeled in reactions containing ~2 pmol dephosphorylated RNA, 15 units T4 polynucleotide kinase (pnk, NEB), 1 X T4 pnk buffer (70 mM Tris-HCl, pH 7.6; 10 mM M g C l 2 ; 5 mM DTT), RNA guard (50 units, Pharmacia), y 3 2 ATP (40 pCi at 50 mCi/mmol; Amersham) and DEPC-treated water. Reactions were allowed to proceed at 37°C for 45 minutes, and then were diluted with 2 M N H 4 O A C , 5 mM EDTA, tRNA 0.1 mg/ml and DEPC-treated d H 2 0 . Diluted RNAs were extracted with an equal volume of phenol chloroform, precipitated overnight at -20°C with 2.5 volumes of 95% ethanol, and centrifuged. Pellets were rinsed with 80% ethanol, dried at room temperature, and redissolved in 100 pi of DEPC-treated d H 2 0 . End-labeled RNAs were further purified by preparative gel electrophoresis on 6% denaturing polyacrylamide gels and subsequent electroelution using a Biorad Model 422 Electroeluter. For folding and subsequent partial RNase digestions, end-labeled RNAs (~0.5 to 1 pmol) were diluted with ribonuclease assay buffer [25 mM Tris-HCl; 5 mM MgCl 2; 60 mM K C 1 ; 100 mM N H 4 C I ; 0.1 mM DTT; 5 % (w/v) glycerol, pH 8] containing ~1 pmol of unabeled RNA. RNAs were then incubated at 50°C for 2 minutes and 37°C for 10 42 minutes, and cooled on ice. Partial RNase digestions were carried out with the enzymes RNaseTl (Pharmacia), RNase T2 (Sigma), RNase CL3 (Boehringer Mannheim), and Physarum M RNase (Pharmacia). Enzyme digestions, carried out at 30°C, were initiated by the addition of 1 ul of enzyme into 9 ul aliquots of folded RNAs in ribonuclease buffers. Reaction times and enzyme concentrations varied for each enzyme used. Typical enzymatic reaction conditions were as follows: RNase T l , 0.1 units for 2 minutes; RNase T2, 0.01 units for 5 minutes; RNase CL3, 0.002 units for 50 minutes; Physarum M RNase, 10 units for 30 minutes. Reactions were terminated by dilution in 10 volumes of RNase stop buffer (0.2 M NaOAc, pH 6; 50 ug/ml yeast RNA; 5 mM EDTA; 0.1% SDS). Diluted reactions were extracted with an equal volume of phenol chloroform and precipitated overnight at -20°C with 2.5 volumes of 95% ethanol. Partially digested RNAs were recovered by centrifugation, rinsed with 80% ethanol, bench dried and redissolved in electrophoresis dyes. RNAs were analysed on 8% and 12% denaturing polyacrylamide sequencing gels. Sequencing ladders were prepared as follows. Hydroxyl (OH) ladders were generated by incubating unfolded, labeled RNAs in carbonate buffer, pH 9, for 5 minutes at 100°C. T l ladders were generated by treating unfolded, labeled RNAs in urea citrate buffer, pH 5.5, with 0.5 units of RNase Tl for 6.5 minutes at 50°C. PhyM ladders were generated by treating unfolded, labeled RNAs in urea citrate buffer, pH 5.5, with 10 units of Physarum RNase for 20 minutes at 50°C. Sequencing films were exposed on intensifying screens at -80°C for one to several days prior to development. 43 2.2.6. Mfold analysis of sxy mRNA secondary structures Computer-based RNA folding predictions were generated for sxy RNAs using the program Mfold [Mfold version 3, University of Washington (9,124,198)]. The program loopDloop (version 1.2a64,1992, D.G. Gilbert, Indiana University) was used to generate sxy RNA secondary structures from Mfold data. 2.2.7. Site-directed mutagenesis Site-directed mutagenesis was carried out using the pAlter system of Promega (Altered Sites II in vitro Mutagenesis System Technical Manual). In preparation for mutagenesis, the 1.8 kb EcoRl-BamHI fragment of pDJM90 (sxy) was cloned into the EcoKL-BamHI site of pAlter-1 to create the plasmid pAltersxy (Figure 2.6). Similarly, the EcoRl-BamHI fragment of pRRS6 (sxy-1) was cloned into pAlter-1 to create the plasmid pAltersxyl. Recombinant plasmids containing the desired inserts were identified by restriction analysis. The mutations sxy-5, sxy-6 and sxy-7 were created by site-directed mutagenesis, as outlined in Figures 2.7 and 2.8. The mutagenic oligos RRS7 (containing the sxy-5 mutation) and RRS8 (containing the sxy-7 mutation) were annealed to pAltersxy template DNA. RRS3 (containing the sxy-3 mutation) was annealed to pAltersxy-1 template DNA. The Tetracycline Knockout oligo (5' GCCGGGCCTCTTGCGGGCGTCCATTCC 3') and Ampicillin Resistance oligo ( 5' GTTGCCATTGCTTGCAGGCATCGTGGT 3') were provided by Promega. pAltersxy or pAltersxyl plasmid DNA was denatured in 2M NaOH, 0.2mM EDTA to generate single-stranded DNA. The DNA was precipitated with 0.2 M N H 4 O A C and 2 volumes of 95% ethanol, and resuspended in TE8. Single-stranded DNA was annealed with the Tetracycline Knockout, Ampicillin Resistant and mutant oligos 44 Figure 2.6. Cloning strategy for the creation of pAltersxy. The cloning procedure to create pAltersxy-1 was equivalent except that the EcoRl-BamHI insert of pRRS6 (sxy-1) was cloned into pAlter-1. 45 Sequences and annealing positions of sxy mutagenic oligos 01 iQQ MUTATION RRS3 (-48 to -31) 5' AGTACTTCTATTGACTCT 3' C -> T at position -38 (sxy-3) RRS7 (-46 to -26) 5' TACTTCTACTTACTCTTTTAA 3' G -> Tat position -36 (sxy-5) RRS8 (- 43 to -18) 5' TTCTACTGACTGATTTAAAATAATTA 3' C T - > G A at positions -31 and -32 (sxy-7) 5 ' . RRS3 ^ R R S 7 ^ RRS8 I Sea Bel! SnaBl Cjal Seal 3 — r ~ -250 sxy open reading f rame; r +250 +500 5 ' Figure 2.7. Sequences and annealing sites for mutagenic oligos used to ceate the site-directed mutations sxy-5, sxy-6 and sxy-7. The +1 position denotes the A of the ATG translational start site for the sxy open reading frame. 46 1. Alkaline denaturation of plasmid DNA to generate single- stranded DNA 2. Annealing of mutagenic oligos (Ampicillin Repair Oligo, Tetracycline Knockout Oligo and Mutagenic Oligo) to single- stranded DNA RRS7 Amp Tet 8 3. Synthesis of the mutant strand of DNA using T4 DNA polymerase and T4 DNA ligase. Ampr 0 <> Tet 8 5. Sequence inserts of tetracycline resistant, ampicillin sensitive plasmids to screen for point mutation of interest. 4. Transform ligated double stranded DNA into MutS. Grow cells and transfer single stranded DNA into JM109. Plate JM109 cells and select for ampicillin resistance. Screen cells for tetracycline resistance. Figure 2.8. Site-directed mutagenesis steps carried out to isolate the sxy-5 point mutation. For site directed mutagensis to create sxy-6, the parent plasmid was pAlter sxy-1 and the mutagenic oligo was RRS3; for site-directed mutagenesis to create sxy-7, the parent plasmid was pAltersxy and the mutagenic oligo was RRS8. (Diagram adapted from Promega Altered Sites II in vitro Mutagenesis Technical Manual). 47 in 1 X annealing buffer (200 mM Tris-HCl, pH 7.5; 100 mM MgCl2; 500 mM NaCI) by heating annealing reactions to 75°C for 5 minutes and then cooling them slowly to room temperature. The mutant strand of plasmid DNA was generated by adding 10 units of T4 DNA polymerase, 2 units of T4 DNA ligase and Synthesis buffer (100 mM Tris-HCl, pH 7.5; 5 mM dNTPs; 10 mM ATP and 20 mM DTT; to IX) to annealing reactions and incubating the reactions for 90 minutes at 37°C. Synthesis reactions, along with R408 Helper Phage DNA, were transformed into ES1301 mutS cells made competent with the RbCl technique (Promega Altered Sites II in vitro Mutagenesis System Technical Manual). Transformed cells were incubated at 37°C for 3 hours to allow for production of infectious phagemid. Cells were then pelleted in a microfuge, and the supernatant (approximately 3 ml) was added to 100 pi of JM109 cells. Transfected JM109 cells were incubated for 30 minutes at 37°C and subsequently plated on LB agar containing ampicillin. Ampicillin resistant colonies were screened for tetracycline sensitivity, and plasmids isolated from ampicillin resistant, tetracycline sensitive colonies were screened by restriction analysis and by sequence analysis with primers 4 and primer 10, which included the DNA sequence from the Clal to the Seal site (see Figure 2.1). For those plasmids containing the desired mutations, the sequenced region of the plasmid between Scal-Clal was subcloned into pDJM90 in order to ensure that the plasmid inserts contained no additional, undesirable mutations. The scheme for subcloning is shown in Figure 2.9. pDJM90 contains two Seal sites, so it was necessary to digest the plasmid to completion with Clal and then to carry out a partial Seal digest. The Scal-Clal fragment (~400 bp) from the mutant pAlter plasmids was then ligated into pDJM90 backbone deleted for this fragment. Ligation reactions were transformed into DH5oc, and transformants were selected on LB agar containing ampicillin. Recombinant plasmids containing the appropriate inserts were confirmed by restriction analysis and sequence analysis with Primer 10. These 48 Figure 2.9. Cloning strategy for the creation of pLBS5. The cloning strategies for the creation of pLBS6 and pLBS7 were equivalent except that the starting plasmid was either pAltersxy-6 or pALtersxy-7, respectively. 49 plasmids were named pLBS5 (pGEM::sxy-5), pLBS6 (pGEM::sxy-6) and pLBS7 (pGEM::sxy-7). 2.2.8. Transferring site-directed mutations to the KW20 chromosome 2.2.8.1 Transferring the sxy-5 mutation to the KW20 genome The site-directed mutation sxy-5 was transferred to the KW20 chromosome as previously described for the sxy-1 mutation, using a marker rescue experiment (193). pLBS5 (pGEM::sxy-5) was digested with EcoRI + Clal and the linearized DNA was used to transform MlV-competent KW20. Since the sxy-5 mutation causes cells to be hypercompetent, transformants that had acquired the sxy-5 were selected for by selecting for strains that were hypercompetent. pLBS5 transformants were put through a second round of transformation by growing cells to an OD600 ^ 0-1 in sBHI and transforming them to nalidixic acid resistance with MAP7 DNA. Colonies that were resistant to nalidixic acid were then tested for hypercompetence by colony transformation assays (see Section 2.1.4.1). 2.2.8.2. Transferring the sxy-6 and sxy-7 mutations to the KW20 genome Unlike the situation for the sxy hypercompetence mutations, there was no direct method by which the presence of the sxy-6 and sxy-7 mutations could be selected for, since I didn't know a priori what effects these mutations would have on competence. Sxy-6 and sxy-7 mutations were transferred to the KW20 genome using a site-directed integration protocol that is outlined in Figure 2.10. The rationale for each of these steps is discussed in more detail in Section 3.2.6.3.1. Because the ampicillin resistance marker of pBR322 is not functional in H. influenzae, the ampicillin resistance markers of pLBS6 and pLBS7 were converted to kanamycin resistance. This was accomplished by miniTnlO mutagenesis with A.NK1316 (described in Section 2.1.5.1). Plasmids that had acquired transposon 50 51 03 H a» o cC >> X v> O) Xi 60 C • i-H *H CO u >> X V3 M l O a, o cc s O 1/3 o s o u a; 1/3 CC o c o a» T3 •43 o CN 3 60 to '0 6 co cc o a» u Pi 03 1/3 X> CO a> 03 13 > '3 cr <u cC 03 -»-» CO a; C a; 60 i/i CC co 03 wo a 60 Pi •p—I >-l I-H u O c > 03 C o g g ou o» l - H CO t 03 X\ +-> 60 G l - H l - l CO u 3 O 60 O o a o o u d) 1/3 cc 60 C • i-H O 03 -t-t cn o u CO u o CM O o 03 1/3 co o u l i b CC 60 o5 c pa i— i a, CC a, 03 V X 03 V 60 4-> 0) CO cC 52 insertions within the ampicillin resistance marker were isolated by selecting for kanamycin resistant cells, and screening those cells for ampicillin sensitivity. The locations of transposon insertions within plasmids were confirmed by restriction analysis. Recombinant plasmids with transposon insertions in the desired locations were named pLBS6K and pLBSZK. pLBS6K and pLBS7K were linearized with SnaBI to create a double-stranded breaks within these plasmids downstream of the locations of the sxy-6 and sxy-7 mutations (Figure 2.10.A.). Linearized DNA was used to transform MlV-competent RR699 to kanamycin resistance following a single crossover event. The competence phenotypes of several kanamycin resistant isolates were tested by colony assays, and the chromosomal rearrangements in these cells (Figure 2.10.A) were confirmed by Southern analysis. KW20 cells carrying single chromosomal copies of sxy-6 or sxy-7 were generated by growing cells for several generations in the absence of kanamycin selection and screening for cells that had become kanamycin sensitive due to a recombination event equivalent to that depicted in Figure 2.10.B. The chromosomal structure of kanamycin sensitive strains carrying the sxy-6 and sxy-7 mutations was confirmed by Southern analysis. The presence of the mutations was confirmed by cycle sequencing of PCR products obtained by amplifying genomic DNA with the Primers 6 and 10 (see Figure 2.1). 2.2.9. Northern analysis RNA was isolated using the hot phenol method, as previously described (55). Cells were grown to the mid-exponential phase of growth in sBHI (OD600 ~0.7), and RNA (approximately 100 pg) was isolated from 25 ml of cells. Sample RNAs were quantified by OD26O/ a n d the quality of RNA preparations was assessed by agarose electrophoresis of samples on 0.8% TAE agarose gels. 53 For Northern analysis, approximately 5 pg of each RNA sample was electrophoresed on 6% denaturing polyacrylamide gels. Following electrophoresis, RNAs were transferred to nitrocellulose membranes (Hybond NX, Amersham) in blotting buffer (0.5 X Tris Borate EDTA, TBE), using a Trans-Blot Cell (Bio-Rad). RNAs were transferred for 1.5 hours at 90 amps. Subsequently, transferred RNA was fixed onto membranes by UV crosslinking with a G-S Gene Linker UV Chamber (Bio-Rad). Uniformly radiolabeled RNA antisense sxy probe was generated by in vitro transcription as follows. pGEMsxy was linearized with Apal, and linearized DNA was isolated by preparative agarose electrophoresis and dialysis as described in Section 2.2.5.1. In vitro transcription reactions contained linearized plasmid DNA (approximately 2 pg), SP6 polymerase (40 units, Gibco BRL), 1 X SP6 transcription buffer (40 mM Tris-HCl, pH 7.9; 6 mM MgCl 2; 2 mM spermidine-(HCl)3), 5 mM DTT, 1 mM each of ATP, GTP and UTP, cx32-CTP (50 pCi at 50 mCi/mmol, Amersham) and DEPC-treated-dH20. Reactions were allowed to proceed at 37°C for 45 minutes, at which time 20 additional units of SP6 polymerase were added to the reaction. After 30 minutes of further reaction time, RNA samples were diluted with 4 M N H 4 O A C , and DEPC-treated dH 2 0. Diluted RNA samples were then extracted with an equal volume of phenol chloroform, precipitated overnight at -20°C with 2.5 volumes of 95% ethanol, and centrifuged. Pellets were and rinsed with 80% ethanol, dried at room temperature and redissolved in 100 pi of DEPC-treated dH 2 0. For further purification, the RNA precipitation step was repeated. Antisense E. coli 5S probe was provided by K.E. Baker. For Northern analysis, membranes were incubated for several hours at 55°C in 8 ml prehybridization buffer (50% deionized formamide; 5 X SSPE (3M NaCI; 200 mM NaH 2 P0 4 ; 25 mM EDTA; pH 7.4), 5 X Denhardt's Solution; 0.25 mg/ml yeast tRNA; 0.06 mg/ml sheared calf thymus DNA and 0.1% SDS). Membranes were hybridized overnight at 55°C in 8 ml prehybridization buffer containing approximately 10 to 20 X 106 cpm of antisense probe. Following hybridization, membranes were washed four 54 times at 55°C with 250 ml of wash buffer (2 X SSC; 0.1% SDS). Following washing, membranes were exposed overnight on a Molecular Dynamics Storage Phosphor Screen. Screens were subsequently scanned, and RNAs were quantitated, using a Phosphorlmager (Molecular Dynamics Storage Phospholmager SI). Prior to reprobing, membranes were stripped by boiling in 0.5% (w/v) SDS. 2.3 Fis methods 2.3.1. Cloning the Tn916 insertion in mutant B - l To identify the chromosomal location of the Tn926 insertion in mutant B-l, I cloned an approximately 20 kb Kpnl chromosomal fragment that included most of Tn916 and approximately 5 kb of H. influenzae genomic DNA flanking the right side of the insertion. The cloning strategy is outlined in Figure 2.11. Mutant B-l chromosomal DNA was digested with Kpnl. Digested chromosomal DNA was electrophoresed on a 0.6% TAE agarose gel, and DNA fragments of approximately 20 kb were excised from the gel and subsequently purified. Purified, size-selected DNA was then ligated to pSU20 that had been digested with Kpnl and dephosphorylated with Calf Intestinal Alkaline Phosphatase (Pharmacia). The ligation mixture was transformed into DH5oc, and plasmids with inserts were identified using blue/white-screening on LB agar containing chloramphenicol and X-GAL. Since the cloned Tn916 sequence did not contain the entire TetM gene, plasmids with the desired inserts were identified by colony hybridization using a subfragment of Tn916 as a probe. Colony hybridizations were carried out according to Boehringer Mannheim specifications. Approximately 200 white colonies were patched onto LB agar containing chloramphenicol (50 colonies each on four 100 mM X 55 MutB-1 chromosomal DNA Figure 2.11. Cloning strategy for the creation of pLBTnl 56 15 mM agar plates). DH5oc and DH5oc (pAM120) were also patched onto each plate as negative and positive controls, respectively. Colonies were grown overnight at 37°C and then chilled at 4°C for one hour. Each plate was replica plated prior to colony hybridization. Nitrocellulose membranes (Hybond, Amersham) were overlaid on the agar surfaces of each plate for 5 minutes. Bacterial cells on the membrane were then lysed by incubating the membranes for 15 minutes in gel-blotting paper saturated with denaturing solution (0.5 N NaOH, 1.5 M NaCl), and neutralized by incubating membranes for 15 minutes in neutralization solution (1.0 M Tris-HCI, pH 8.0; 1.5 M NaCl). Membranes were baked at 80°C for two hours to fix DNA. Prior to prehybridization, membranes were incubated at 55 °C for 30 minutes in a solution of 1 X SSC, 0.2 % SDS, and 50 ug/ml proteinase K, and then washed briefly in 1 X SSC. Subsequent prehybridization and hybridization steps were carried out as previously described (Section 2.1.6.5). A 1.5 kb DIG-labeled pAM120 Hindlll-Kpnl fragment was used as a probe. Three out of 200 colonies tested had the desired recombinant plasmid as determined by restriction digestion with Hindi and Kpnl. One of these plasmids was designated as pLBTnl and used for further studies. Primer TnRl (5' GATAAAGTGTGATAAG TCCAG 3') binds to Tn926 sequences 36 to 56 and is directed outwards toward the 5' end of Tn926 (Genbank accession # U09422). This primer was used to identify the precise position of the Tn926 insertion within mutant B-l. 2.3.2. Cloning the fis operon by PCR The strategy for cloning the KW20 fis operon is outlined in Figure 2.12. The KW20 genome was amplified using the primers FSl and FS2 (Figure 2.2). The PCR product (approximately 1.9 kb) was digested with CM and Bglll to yield a 1.8 kbp product that was cloned into the unique CM and Bglll sites of the multiple cloning site of pSU21. Ligation reactions were transformed into MlV-competent KW20, and transformants 57 were selected on sBHI agar containing chloramphenicol. Clones containing the desired insert were identified by restriction analysis with Xhol and HmdIII +SnaBl digests. One plasmid, pLBFl, was retained for further study. 2.3.3. Cassette mutagenesis of fis and orfl Null alleles of fis and orfl were created by cloning the spectinomycin/streptomycin resistance cassette of pKRP13 into the SnaBI and EcoRI sites within the/is operon, respectively. The/z's null allele was created by digesting pKRP13 with Smal and ligating the Sp/Sm r cassette into the SnaBI site of pLBFis (Figure 2.13). The orfl null allele was created by digesting pKRP13 with EcoRI and ligating the Sp/Sm r cassette into the EcoRI site of pLBFis (Figure 2.14). The ligation reactions were transformed into DH5cx, and transformants were selected on LB agar containing chloramphenicol and spectinomycin. The desired recombinant plasmids were identified by restriction digestion with the enzymes Kpnl and HmdIII. The respective plasmids were named pLBFl (pSU21 fis::spec) and pLBF2 (pSU21 orfl::spec). The fisr.spec and orflwspec null alleles were transferred to the KW20 chromosome by digesting pLBFl and pLBF2 with Kpnl + Clal (pLBFl) or with SnaBI + Clal (pLBF2) and transforming MlV-competent KW20 or RR699 (sxy-1) to spectinomycin resistance. 58 FS1 Gal KW20 genome ORF1 t5 fioZII 1. PCR with FSl andFS2 2. Digestion of PCR product with CM and Bgm 3. Ligation of digested PCR product with pSU21 digested with CM and BamHl FS1 EcoRl Xhol C l a l Figure 2.12. Cloning strategy for the creation of pLBFl (psu21:: fis). 59 Figure 2.13. Cloning strategy for the cr eation of pLBF2 [pSU21(fis::spec)]. 60 Kpnl Smal Figure 2.14. Cloning strategy for the creation of pLBF3 [ pSU21(orfl::spec)]. 61 C h a p t e r 3 : S x y f u n c t i o n a n d r e g u l a t i o n In this chapter, I address the evolution and function of Sxy through the identification of Sxy homologs in other bacteria. I examine the regulation of sxy expression in sBHI, and in MIV. I propose that sxy expression is regulated at both the transcriptional and posttranscriptional level, and present evidence that supports the folding of sxy mRNA into an RNA secondary structure that negatively regulates sxy expression during logarithmic growth in rich medium. 3.1. Background: The role of Sxy in competence development In the General Introduction (Section 1.3.2), I discussed the existing evidence that Sxy is an early competence protein that regulates the expression of late competence genes. In the Background of Chapter 3,1 discuss this concept in further detail. I also discuss the hypothesis that sxy expression is negatively regulated by an mRNA secondary structure that is destabilized in the sxy hypercompetent mutants. I outline a simple model for the regulation of competence by two signals (Redfield, unpublished). The first of these is a general signal mediated by a rise in cAMP levels, and the second, competence-specific, signal is an increase in the concentration of Sxy. Finally, I present an outline of my specific research objectives in the context of the model for competence development. 3.1.1. Discovery of the role of Sxy in competence development The sxy gene of H. influenzae encodes a small basic protein of unknown function (HI # 601, position 62,2876 to 62,3526 of the genome; Figure 3.1.A). It is located in the chromosome between the £. coli RecA homolog (reel)(HI # 600) and rrnA, the 62 ribosomal RNA A operon. The role of Sxy in competence development was discovered independently by two research groups. Following EMS mutagenesis of KW20, Redfield identified several point mutations that caused mutant cells to develop competence during early logarithmic growth in sBHI, a stage of growth during which transformation is not observed in cultures of wild type cells (155). Several of the mutations (sxy-1 to sxy-5, Figure 3.1.A and Figure 3.4) were mapped to a gene adjacent to the recA gene, which was named sxy (192; Redfield, unpublished). It had been previously reported that the multicopy plasmid pHKrec, which contains a cloned 3.1 kb EcoRI fragment carrying the reel gene and flanking DNA, caused cells to exhibit constitutive competence in rich medium (i.e. cells had transformation frequencies of ~10~2/ whereas the transformation frequencies of wild type cells in sBHI do not exceed ~ 5 x 10"4) (175). However, at that time it was not known whether cells were hypercompetent due to elevated levels of Reel, or due to the overexpression of another gene encoded by the cloned fragment. It was subsequently reported that the hypercompetence of strains carrying pHKrec was Reel-independent, and was due to the overexpression of sxy (Figure 3.2) (192, 200). 3.1.2. Sxy regulates competence development Because it was known that sxy overexpression increases competence, it was expected that cells lacking functional Sxy would have reduced transformation frequencies. It was not known if sxy was an essential gene, or if the sxy gene product had a function outside of competence. Sxy null mutants were created and found to be unable to transform in either rich medium or starvation medium (transformation frequencies are <10~8), indicating that Sxy is essential for competence development (Figure 3.2) (192, 200). The addition of exogenous cAMP to sxy null mutants doesn't raise their competence level, indicating that Sxy affects competence development at a step that is independent of, or downstream of, the regulation of competence 63 A. rec-1 CRP [ c r £ sxy rrnA i****«*J'.v.'.v,v,v,v.'!Tr| EcoRI 345 21 Ecofl/ ' 1 1 1 1 r 1.0 2.0 3.0 B . AATAAAAAAATATGGGGTATTATATATTTCAGTAAAGAGGATGGGGAAGGCTTAAAATAA 60 AAGTTTTAAATATAGTATATTTATATATAAACCAATCGTTTAGATAAAAATAAATTTTAA 120 CRP 1 ATTTAC(^KTiGCTCACAMTCTGACATTTATATTTAGGTGTAGTAAMATTTC 180 CRP 2 T Seal R B SI CTAAAXGAAATCAGMGTACTTCTACTGACTCTmAAAATAATTATTCATTGGAGGTTT 240 SxyL R B S s AAT ATG AAT ATA AAG GAT GAG CAT ATA GAT AGC GTT TGC TCC TTG TTA GAT 291 M N I K D E H I D S V C S L L D 17 CAG Q TTA L GTA v GGA G AAT N GTT V TCC TTT S F AAA AAT CTT K N L TTT F ACT T GGT G TAC GGT TTG Y G L 342 34 TTT F CAC K AAG K GAA E GAG E ACA T SxyS ATG TT1 M F r GCT ATT TGG CAA AAT AAA AAA CTT TAT A I W Q N K K L Y 393 51 TTA L CGC R GGT G GAG E GGT G GTT V CTC GCA L A ATT I CAA TTA Q L ACT T AAA K TTA L GGT TGT G C GAA E 444 68 CCT P TTT F ACA T ACG T Bch AAT N GAA E TTG AAT L N S M AAA Jl< CGG TTT R F , GTG V CTT L TCA S CAA TAT Q Y TAT Y 495 85 GCA A CTT L TCT S GAT D CAG Q ATT I TTA CGT L R AGT S AAT AGA N R TTG L TGT C AGA R AAA TTG K L ATT I 546 102 ATT I CTT TCT ATT L S I -AAG K CAG Q ATT CTT I L GAG E CAG AAG Q K CTA L GAA E TGT C ACG TTA T L AGA R 597 119 AAA K TTG L AAT N CGA R TTA L AAG K GAT TTA D L CCC P AAT TTA N L ACG ATT AAA T I K CAT GAA H E AGA R 648 136 GCT A TTA L ATA I AAA K GTT V GGT G ATT ACA I T AAT N GTT GCG V A ATG M CTA L AGA R GAG ATT E I GGC G 699 153 GCA A GAA E AAT N GCA A TTG L GTG V GAA TTA E L AAG K AAA AGT K S GGC AGT G S GGT G GCT ACG A T CTT L 750 170 GAT D TTT F TAT TGG Y W AAA K TTA L GTA TGT V C GCT TTG CAA A L Q AAT N AAA K AAT N AGT CAG S Q ATG M 801 187 TTA L AGT S CAA GCT Q A GAA AAA E K GAG CGT E R TTA L TTG AAG L K AAA K TTA L AAC N GAA GTT E V TTG L 852 204 AGA AAA AAT R K N GGC G TTA L AAA K GGC TAT AGA AAA TTA G Y R K L GAT D GAT D GAA TAA AATTTGT E STOP 904 Figure 3.1. (A). 3.1 kb EcoRI chromosomal fragment showing the locations of the rec-1, sxy and rrnA 16S genes, and the location of the sxy promoter and sxy-1 to sxy-5 point mutations. (B). The nucleotide and amino acid sequence of the sxy gene. Amino acid positions are numbered on the left and DNA sequence is numbered on the right of the diagram. The sxy transcriptional start site is indicated by an arrow. Putative CRP binding sites and ribosome binding sites for the major open reading frame (SxyL) and smaller open reading frame (Sxys) are indicated by dashed and solid lines, respectively. Both (A) and (B) were adapted from Zulty and Barcak (200). 64 A. B. i i i 0 100 200 300 Time in sBHI (min) Figure 3.2. Growth (A) and transformation frequencies (B) of KW20 (•), RR648 (Sxy-) (O) and RR554 [KW20 (pHKrec)] (A) in rich medium. Transformation frequencies were assayed as described in Section 2.1.4.1. Williams etal. (192). development by cAMP/CRP (192). There are no other obvious phenotypic defects of Sxy" cells - mutant strains grow with doubling times close to that of wild type cells (200; data not presented), are not UV sensitive, and appear to have wild type levels of intracellular cAMP, as assessed by sugar fermentation assays (192). There is no evidence that Sxy has a function outside of competence. Does Sxy function as a regulatory protein? Transformation-deficient strains lacking known regulatory proteins - for example, CRP and TopA (Topoisomerase I) - are defective in DNA binding, DNA uptake, and competence-induced phage recombination (44,45,181). These proteins probably act early in competence development, by affecting the expression of several genes required for the mechanistic steps of transformation. Sxy null mutants were also shown to be deficient in DNA binding and uptake (200). In addition, DNA uptake and phage recombination are increased in sxy-1 hypercompetent mutant strains (192). These results imply that Sxy acts at a regulatory step of competence development. How does Sxy activate competence? One suggestion is that Sxy is a transcription factor. Several competence genes, including dprA, comF, rec-2, pilA, and comA, contain a 26 bp palindromic sequence upstream of their core promoters. These sequences, referred to as Competence Regulatory Elements (CREs), may be sites for the binding of a competence specific transcriptional activator, possibly Sxy (99). The transcription of the genes comF, rec-2, and dprA, whose products are required for DNA translocation during transformation, is induced upon transfer of cells to starvation medium (99. 200). The induction of dprA and comF transcription is dependent upon Sxy (99, 200) so Sxy must directly or indirectly activate their transcription. However, there is currently no biochemical evidence that Sxy binds DNA. Furthermore, Macfadyen has suggested that CRE sites are CRP binding sites (120), and expression of a rec-2::lacZ operon fusion was decreased in crp cells (83). Therefore, the mechanism by which the transcription of the 66 rec-2, dprA, pilA and comF genes is coupled to sxy expression and cAMP levels remains to be determined. 3.1.3. Regulatory features of the sxy gene In this section, I will discuss the known or proposed features of sxy expression, including the effects of the sxy hypercompetence mutations. 3.1.3.1. The sxy promoter The proposed regulatory features governing sxy transcription and translation are shown in Figure 3.I.B. (200). The sxy transcriptional start site was mapped in MIV-competent cells by primer extension (200). A single transcriptional start site was identified, as shown in Figure 3.1.B. However, the possibility that a different start site is used in noncompetent cells has not been ruled out. Sxy and rec-1 are transcribed from divergent promoters (199, 200). sxy and the rrnA operon are transcribed in the same direction; although I have not identified the rrnA promoter, the possibility that sxy and rrnA are cotranscribed from the sxy promoter is remote, because ribosomal RNA promoters are highly conserved, and are usually positioned approximately ~150 nucleotides upstream of the 5' end of the 16S gene (102). The 3' end of the sxy message has never been mapped. However, sequence analysis reveals a putative rho- independent terminator at its 3' end. Therefore, sxy is probably transcribed as a monocistronic mRNA. Zulty and Barcak showed that sxy transcription is induced 3 to 4-fold immediately upon transfer of cells to MIV, and is maximal 50 minutes after transfer to MIV (200). What specific signals are responsible for this induction? Since high cAMP concentrations induce competence, the sxy promoter is an obvious target for CRP/cAMP transcriptional activation. Zulty and Barcak identified two regulatory sequences resembling CRP sites upstream of the sxy transcriptional start site (Figure 3.1.B), and 67 reported that expression of sxyr.lacZ operon fusions in sBHI was reduced 3 to 4-fold in CRP- cells (200). However, these putative CRP binding sites have only weak sequence similarity to the 22 bp E. coli consensus sequence, which is AAATGTGATCT*AGATCACATTT (underlined sequences are highly conserved; the star separates sequences of dyad symmetry) (33). 3.1.3.2. Translation of the sxy message The sxy coding sequence is predicted to encode a single protein of 217 amino acids. Zulty and Barcak analyzed polypeptide synthesis from sxy DNA cloned and expressed in E. coli (200). These authors observed two proteins, one of approximately 24.9 kDa, similar to the predicted size of the major sxy orf (24.3 kDa), and another of approximately 20.8 kDa, which could be produced by in-frame translational initiation at a downstream ATG (Figure 3.I.B). It remains to be determined if the appearance of two proteins in a heterologous host reflects two functional forms of Sxy in H. influenzae. This issue is not addressed in this thesis. 3.1.3.3. The sxy hypercompetence mutations may disrupt base pairing of a regulatory RNA secondary structure. The sxy-1 to sxy-5 hypercompetence mutations have equivalent effects on competence development; each mutation increases spontaneous competence by 100-fold (Figure 3.3). These mutations are unlikely to alter Sxy structure; sxys, sxy-4 and sxy-5 are in the 5' untranslated region of the message; sxy-2 is a silent substitution at codon 19; and sxy-1 is a conservative substitution of valine for isoleucine at amino acid 18 (Figure 3.4.B). Therefore, these mutations were proposed to act at the level of sxy regulation. One obvious possibility is that the mutations affect RNA secondary structure. Preliminary examination of the region of sxy RNA containing these mutations suggested a potential RNA secondary structure (hereafter referred to as sxy RNA structure A) that could be disrupted by each of the mutations (Figure 3.4.C) 68 I 1 1 1 ° o o o o © o - H m Time in sBHI (min) Figure 3.3. Transformation frequencies (A) and growth (B) of KW20 (•), RR699 (sxy-1; •), RR700 (sxy-2;A), RR724 (sxy-3;Y) and RR723 (sxy-4;B) in sBHI. Sxy mutations were created by site-directed mutagenesis and transferred to the KW20 chromosome as previously described for the sxy-1 mutation (192). Transformation frequencies were assayed as described in Section 2.1.4.1. Growth and transformation frequencies for KW20, RR724 and RR700 are the mean values of replicate cultures. Growth and transformation frequency values for RR699 and RR723 are from single cultures. (Redfield and Ma, unpublished). 69 C R P C R P J U A 345 21 B. Allele sxy-3 ^Mutation C - > T Posit ion_....^^ 38 5' untranslated sxy-4 T - > C -37 5' untranslated sxy-5 G -> T -36 5' untranslated sxy-2 G - > A +51 Gln-|7 silent sxy-1 G -> A +55 V a l 1 9 -> lie Figure 3.4. A. Chromosomal location of the sxy-1 to sxy-5 hypercompetence mutations. The arrow shows the sxy transcriptional start site. B. Base pair and amino acid alterations caused by the sxy hypercompetence mutations. Sequence positions are numbered relative to the +1 position, defined here as the A of the sxy ATG translational start site. C. Predicted RNA secondary stucture (RNA secondary structure A) that supports base pairing between sequences that are mutated in the sxy hypercompetent mutants. Adapted from Redfield (unpublished). 70 (Redfield, unpublished). (This secondary structure is shown in more detail in Figure 3.20). 3.1.4. A w o r k i n g m o d e l for competence regu la t ion by the cont ro l o f sxy exp ress ion As discussed, several lines of evidence indicate that Sxy is a positive regulator of competence: a) overexpression of Sxy on a multicopy plasmid results in constitutive competence; b) sxy is absolutely required for competence development in sBHI and MIV; c) point mutations in sxy lead to hypercompetence in sBHI; d) transcriptional levels of some genes involved in the mechanistic steps of transformation are drastically reduced in sxy cells. Our current model for the development of competence includes the premise that the induction of competence requires increased concentrations of both cAMP and Sxy (57, 155,192). Under this premise, an increase in the amounts of cellular Sxy is predicted to occur as cells proceed from logarithmic growth in rich medium to stationary phase; this increase may be mediated via cAMP/CRP transcriptional activation. A second, cAMP-independent, level of sxy regulation is predicted to occur via negative regulation of sxy expression by sxy structure A. Relief of this negative regulation occurs in wild type cells only upon starvation. Strains carrying the sxy hypercompetence mutations are predicted to be hypercompetent because the mutations destabilize sxy RNA structure A; however, these strains don't exhibit constitutive competence because sxy transcription remains dependent upon high concentrations of cAMP or the expression of another regulatory factor. 71 3.1.5. Specific research objectives 1. The precise function of Sxy in transformation has not been determined, nor is it known whether Sxy has a function outside of competence. To address these issues, I searched for Sxy homologs in transformable and nontransformable bacteria, and attempted to identify functional motifs within Sxy and identified Sxy homologs. Our model for sxy expression and competence development generated several testable predictions which were addressed experimentally: 2. Cellular transformation frequencies should be positively correlated with Sxy concentrations. Sxy protein concentrations should increase at the end of logarithmic growth in rich medium, and increase even further upon transfer of cells to MIV. 3. Transcription of wild type and mutant sxy promoters should be increased by high concentrations of cAMP, both in rich medium and in MIV, as described by Zulty and Barcak (200). 4. sxy RNA structure A should negatively regulate expression of sxy in rich medium only. Cells with sxy point mutations should be hypercompetent because sxy expression is maximal even in the absence of the starvation-specific signal, providing that intracellular levels of cAMP are sufficiently high. Conversely, the point mutations should not affect levels of sxy expression in MIV. 5. Compensatory mutations that restore base pairing of sxy RNA structure A should lower sxy expression and competence development to wild type levels. Point mutations which stabilize sxy RNA structure A through additional base pairing should decrease sxy expression and competence development. 72 The model makes no predictions about the mechanism by which sxy RNA structure A regulates sxy expression. This structure could affect transcriptional initiation, mRNA elongation rates, translational initiation or ribosomal translation rates. Several of these possibilities were addressed experimentally. 3. 2. Results 3.2.1. Database searches to find Sxy homologs and functional motifs 3.2.1.1. Proteins that are similar to Sxy are found in both transformable and nontransformable bacteria. Most studies on Sxy, including those described in this thesis, have focused on the regulation of the gene, rather than on the biochemical function of the protein. The precise role of Sxy in competence development is undefined. Although the phenotype of sxy cells leads us to think that this protein could function specifically in competence development, it may have an additional function. An example of this type of protein is Reel. Reel was initially isolated because mutations in its gene caused a transformation-deficient phenotype; however, Reel has a general role in DNA repair and homologous recombination. One way to determine whether Sxy functions specifically in competence development may be to examine the phylogenetic distribution and natural transformation of bacteria that contain Sxy homologs. The NCBI Microbial Genomes Blast database (11,17) includes sequences from naturally transforming bacteria, such as B. subtilis and S. pneumoniae, that are distant relatives of H. influenzae, and also includes sequences from bacteria that are members of the y subdivision of the Proteobacteria, including A. 73 actinomycetemcomitans which, along with H. influenzae, is a member of the Pasteurellaceae family. Sxy Blast searches were expected to yield two possible results: (See Figure 1.1 for a reminder of the phylogenetic distribution of natural transformation across bacterial taxa): A. Only transformable bacterial species contain Sxy homologs. If Sxy has a function that is specific to competence, and the role of Sxy in competence development is conserved amongst transformable bacteria, I would expect to identify Sxy homologs in transformable bacteria closely related to H. influenzae, such as A. actinomycetemcomitans, and also in more distantly related naturally transformable bacteria such as N. gonorrhoeae, B. subtilis, and S. pneumonia. I would not expect to identify Sxy homologs in closely related nontransformable bacteria such as P. multocida. B. Transformable and nontransformable bacteria contain Sxy homologs. If Sxy has a functional role not specific to the induction of competence, I would expect to identify Sxy homologs amongst those bacteria most closely related to H. influenzae, regardless of whether those bacteria were naturally transformable or not. I performed Blast searches with the Sxy protein sequence (217 amino acids) against the NCBI Microbial Genomes Database, which includes sequences from both completed and incomplete genome projects. As a control, I performed Blast searches using the Reel protein sequence. These results are shown in Table 3.1. For Reel, only the 6 most similar proteins scored in the search are included. For Sxy, similar proteins were included in Table 3.1 only if they had an E value of ^ _le~5. (Note: E values are a statistical score that represent the number of times a particular protein match would 74 Table 3.1. Results of blast searches of H. influenzae Reel and Sxy amino acid sequences against the NCBI unfinished and finished microbial genomes database. Organism H. influenzae Red Blast Results Evolutionary classification Score E value Haemophilus influenzae 622556-621495 of complete genome (spP43705) Actinobacillus actinomycetemcomitans 5343-4288 of contig 645 Escherichia coli 2821792-2820734 of complete genome (spP03017) Salmonella typhi 252260-253318 of contig 350 Yersinia pestis 149177-150244 of contig 656 Shewanella putrefaciens 5883-4848 of contig 4320 Proteobacteria y subdivision; Pasteurellaceae Proteobacteria y subdivision; Pasteurellaceae Proteobacteria y subdivision; Enterobacteriaceae Proteobacteria y subdivision; Enterobacteriaceae Proteobacteria y subdivision; Enterobacteriaceae Proteobacteria y subdivision; Alteromonadaceae 693 589 533 522 524 517 o.o e-168 e-151 e-148 e-148 o-146 Organism H.influenzae Sxy Blast Results Evolutionary classification Score E value Haemophilus influenzae 622876-623526 of complete genome (spP43779) Actinobacillus actinomycetemcomitans 7772-6410 of contig 703 Pasteurella multocida 121-282 of contig 252 Vibrio cholerae 1883929-1884480 of contig 1752 Escherichia coli 1020415-1020933 of complete genome (spP75864) Salmonella typhi 571-1089 of contig 333 Yersiniapestis 1236-1655of contig 673 Proteobacteria y subdivision; Pasteurellaceae Proteobacteria y subdivision; Pasteurellaceae Proteobacteria y subdivision; Pasteurellaceae Proteobacteria y subdivision; Vibrionaceae ; Proteobacteria y subdivision; Enterobacteriaceae Proteobacteria y subdivision; Enterobacteriaceae Proteobacteria y subdivision; Enterobacteriaceae 436 e-122 138 3 e 3 2 _85 77 58 52 48 2 e - l6 7e- 1 4 3 e 8 2e"6 2 e 5 75 be expected to arise by chance; an E value of le~5 means a similar match would arise by chance only one time in 105 searches). As expected, since RecA is a highly conserved protein, those organisms that contain proteins most similar to the H. influenzae Reel protein are members of the Proteobacterial g subdivision. Blast results obtained for Sxy were qualitatively similar to those obtained for Reel - those organisms with proteins most similar to Sxy are also members of the Proteobacterial y subdivision. However, Blast scores for the possible Sxy homologs were relatively low, indicating that Sxy is not a highly conserved bacterial protein. Of the bacteria that may contain Sxy homologs, only A. actinomycetemcomitans is known to be naturally transformable (183). If we assume that the proteins identified in nontransformable bacteria are true Sxy homologs (see below), we can conclude that Sxy is likely to have a function unrelated to competence. It is unclear from transformation studies what this function might be. Sxy is not an essential protein; sxy null mutants grow with a doubling time similar to that of wild type cells and have no obvious phenotype other than transformation deficiency. Are the proteins in Table 3.1 true Sxy homologs? The identification of homologs based on sequence similarity can give rise to false identification of homology (41). For example, proteins may share sequence similarity within a protein domain because of a similar biochemical function without being true homologs related by evolutionary descent. On the other hand, the identification of true homologs doesn't necessarily mean that the homologs have equivalent functions, due to sequence divergence over evolutionary time (41). Despite the comparatively low Blast scores obtained for some of the proteins in Table 3.1, they meet the criteria proposed by Huynen and Bork for proteins truly related by evolutionary descent (91). Proteins that are truly homologous should have the highest 76 level of pairwise identity when compared with the identities of either protein to all other proteins in the other's genome. 2). The pairwise identity should be significant (E < 0.01). 3). The similarity should extend to at least 60% of one of the proteins (i.e. accounting for gaps, deletions, insertions, etc.). According to these minimal criteria, each of the proteins in Table 3.1 should be considered as a true Sxy homolog (Table 3.2 and Figure 3.6). Two additional lines of evidence suggest homology. Each of the possible Sxy homologs listed in Table 3.1 was used to perform Blast searches against the NCBI Microbial Genomes Blast database. Each protein, as well as showing similarity to Sxy, also shows similarity to the other proteins in Table 3.1, at E values of < 0.01. In addition, the proteins are similar in overall size (ranging in size from 195 to 217 amino acids). (Protein lengths were determined by visually identifying start and stop codons in sequences of the appropriate genomic DNA or contig). We might expect the chromosomal position of homologous genes to be conserved between closely related bacteria (91). Therefore, it seemed possible that the chromosomal arrangement of sxy and surrounding genes might be conserved within families, but not between the Enterobacteriaceae, Pasteurellaceae, and Vibrionaceae families. In order to determine the identity of genes surrounding the putative sxy genes, I searched each contig or genomic sequence in a Basic Blast search of the nonredundant database (6), This allowed me to tentatively identify proteins of known function based on sequence similarity. The results are shown in Figure 3.5. The chromosomal location of the putative sxy gene and surrounding genes is conserved between the enterics E. coli, S. typhi and Y. pestis. This provides additional evidence that the putative Sxy homologs within these bacteria are true homologs of one another. Unfortunately, the chromosomal location of the Sxy homologs in the enterics didn't 77 Table 3.2. Extent of similarity between Sxy and its homologs. Similarity and alignment length values were obtained from Blast searches of Sxy against the NCBI Microbial Genomes Blast database (11). Organism containing Sxy homolog % Similarity a Al ignment length A. actinomycetemcomitans 59% 208/217 (96%); 6 gaps P. multocida 63% b84/217(44%) V. cholerae 52% 186/217(86%); 1 gap E. coli 47% 138/217(64%) S. typhi 47% 177/217 (82%); 2 gaps Y. pestis 46% 143/217 (66%); 4 gaps a Given as the length of the Sxy amino acid sequence across which the similarity extends b The contig did not contain the whole gene encoding the sxy homolog 78 Haemophilus influenzae - 3.0 kb of KW20 chromosomal DNA reel sxy ^ rrnA 16S. — i 1 r 1000 2000 3000 Pasteurella multocida Contig 252 - 1.2 kb comG comF sxy horn ^ horn horn 1 1 500 1000 Actinobacillus actinomycetemcomitans Contig 703 RNA -1-9 kb polymerase sxy horn ^ ^ sigma 32 1 1 1 500 1000 1500 Vibrio cholerae Contig 1752 - 2.6 kb hypothetical purR fc sxy horn ^ protein ^ 1 1 1 1 r 500 1000 1500 2000 Escherichia coli, Yersinia pestis and Salmonella typhi - 4 kb of chromosomal DNA sulA yccR/sxy ^ m yccS I I I I I i I 500 1500 2500 3500 Figure 3.5. Chromosomal maps of sxy or sxy putative homologs and surrounding genes in H. influenzae, P. multocida, A. actinomycetemcomitans, V. cholerae, E. coli, Y. pestis and S. typhi. Dashed lines indicate genes that are only partially encoded by the particular DNA fragment shown. provide any further information about the function of Sxy. The products of the flanking genes yccR(sxy) and yccS are of unknown function. SulA is an inhibitor of cells division (Genbank report); however, since sxy/yccR is not in an operon with the sulA gene, we cannot infer any functional relatedness between the sulA and sxy gene products. The chromosomal location of the putative Sxy homologs was not conserved within the members of the Pasteurellaceae family, or between the different Proteobacterial families. This cannot be construed as evidence that the Sxy proteins within these bacteria are not true homologs: Huynen and Bork suggest that there is little conservation of the order of genes in genomes if the amino acid identity between orthologous genes falls below 50% (91). Interestingly, the putative sxy homolog of P. multocida is located in the genome adjacent to genes encoding proteins that share sequence similarity with the H. influenzae comF and comG proteins. In H. influenzae, comF and comG are part of a larger competence operon encoding the ComA to ComG proteins (182). The physical proximity of the P. multocida sxy homolog to other com genes may indicate interaction between these gene products within this microorganism. 3.2.1.2. Sxy and its putative homologs are soluble, leucine-rich proteins with no known homology to other proteins. Does Sxy or its putative homologs contain amino acid motifs that help to identify the function of the protein(s)? I attempted to identify functional domains within Sxy and its putative homologs by several methods (as outlined in Section 2.2.1). Competence proteins involved in DNA translocation are often integral or peripheral membrane proteins. On the other hand, if Sxy and its putative homologs are transcription factors, we would expect them to be soluble proteins that contain DNA binding motifs. Each protein sequence was analysed for the presence of transmembrane segments or clusters of hydrophobic residues. None of the proteins contained transmembrane domains, indicating that they are not likely to be membrane proteins. Each protein 80 sequence was also analyzed for amino acid content, charge clustering, and periodic spacings of amino acids. Several of the proteins contained a high distribution (15-17%) of leucine residues. However, none of the protein sequences were found to contain leucine zippers, which are common dimerization domains for DNA binding proteins. I also searched each protein sequence for similarity to known functional motifs of other proteins. None of the proteins had motifs that significantly matched any motifs in the Prints and Blocks databases. Several of the proteins contained sequences that were weakly similar to helix turn helix domains of transcription factors; however, the similarity of the matches were not strong enough to imply a true functional similarity. 3.2.1.3. Sxy and its putative homologs have conserved amino acid motifs. Do Sxy and its homologs share conserved amino acid motifs? This was addressed by aligning Sxy with its homologs, and searching the aligned sequences for regions of high similarity. Figure 3.6. shows a modified Clustal W alignment of Sxy with its putative homologs. Although the protein sequences are not highly conserved overall, they share distinct regions of homology across all of the aligned protein sequences. Three conserved blocks of similarity (Blocks A, B and C) were identified amongst the set of 7 sequences, as shown in Figure 3.7. In addition, two tryptophan (YY) residues (at positions 83 and 84 in Sxy) are completely conserved. Blocks A, B and C were searched against conserved motifs of protein families in the Prints and Blocks databases. No similarity was found between any of the Sxy motifs and motifs in the databases. However, the identification of more than one block of homology provides additional evidence that these proteins are true homologs. The identification of conserved sequences, likely to encode biochemical or functional 81 •H <a & o * J o p, K ^ pi > W W 5M' -H fd g U 4J O PI BG < pi > W W JM' r> <H IT) LD r-\ rH H H Ol Ol Ol O H CM CM * H H M M 5 H CO U H H •H «J fi O .p O ft K < pi > TO W >i CD O CH CD 3 cr o» u CC o Pi g CC X co 6 • i-H "cc OH o "3 5 a N R <u S f 3 ^R a . o 5 A CO 3 u to <u u 3 60 cc «3 o R O (« 9 £ CC o ri £ ^ .5 to a» cc u •«-H Pi • l-H + - a . • S s C CC R O o g o X CO > 3 a , o •s -R en a> o Pi CD 3 cr V3 g cc -R a . R O co a> X a , a> Pi a> 3 cr a» cn o CC O Pi •f—< g cC Pi g P! ai »—< a , 6 o u S £ in ^3 •'- 13 Si cn 13 X g to a» xi cC CC 1c • r-i 4-1 1-1 CC * a; > o cc to CD Pi o Pi 60 -e a . R O s Ol x\ o Pi o g o Pi a; 60 CD X i Pi CD ^ -t-» O ° c 60 <u 1/3 CU VH Pi CD 3 cr CD l« CD cn 3 CC CJ a> CD Pi o T3 O u a , o CC CC • t—4 Pi a» u Pi CD 3 cr CD C« CC 5 ^ CC CD l-H CC on Pi O TJ O a , o o -R u Pi CD tn CD X I-H CD CD Pi O o g T 3 CD > l-H CD to Pi O U ai 4-> CC u -a .g on CD Pi 3 cr CD (/I TJ • t—< U CC o Pi ~ i T 3 j_, - 3CC O Pi 82 Block A: width = 14 Consensus Moti f H . i . A . a . V . c . S . t . E.c . Y .p . Block B: width = 12 Consensus Moti f H . i . A .a . P.m. V .c . S . t . E.c. Y.p. Block C: width = 29 Consensus Moti f : H . i . : A . a . : P.m.: V . c . : S . t . : E . c . : Y . p . : T-FA-V-NGELYLR 38 TVFAMVANGELYLR 36 IMFALHQNDSIFLR 35 TMFALVVNDTLHIR 38 TVFAMVANGELYLR 38 TVFAMVSDGELYLR 38 IMFAVVSEGELYLR RLKDLPNLS L 122 RLKDLPNLTIKH 120 RIKELANFSIKL * RIKELPNLSIKH 117 RLKDLPNLRLAT * RLKDLPXMTFHL 118 RLKDLPNMSFHL 119 RLKDLPNLSASL ERLL-KVGI--V— L--LGAK—Y-RL-130 ERALIKVGITNVAMLREIGAEBNALELKK 132 ERLLAKINIYTVSEFQKIGAIHSYVRLKK * ERLL-KVEIHDVATFQALGAKNAYIRLKK 129 ERMLKKAGIDTVESLQTLGSVEAYKAVQR * ETLLNESGIKDENMLXILGAKMCWLRLQS 130 EAILGEVGIKDVRALRILGAKMCWLRLRQ 131 ERLLWKVGIKNATELRLEGAKCCYLKLRA Figure 3.7. Conserved motifs of Sxy and Sxy homologs identified using Block Maker (4,86). Consensus sequences were compiled by visual comparison of the data set at each amino acid position. A particular amino acid was identified as a consensus amino acid sequence if it appeared at least 4 out of 7 times within the data set. Perfectly conserved amino acid sequences are underlined. * Amino acid positions not numbered because the entire protein sequence was unavailable. ^ activities of these proteins, may make it possible to find Sxy homologs in more distantly related ancestors. The sxy-1 hypercompetence mutation alters amino acid 19, which is not a highly conserved amino acid residue. This observation supports the hypothesis that the sxy hypercompetence mutations do not exert their effects by affecting Sxy protein structure. 3.2.2 Site-directed mutagenesis confirms that the sxy-5 mutation is sufficient to cause hypercompetence. Each of the sxy-1, sxy-2, sxy-3 and sxy-4 mutations, originally identified by Redfield following EMS mutagenesis, were recreated by site-directed mutagenesis and shown to be sufficient to cause hypercompetence in an otherwise unmutagenized background (Figure 3.3 and 3.4) (Ma and Redfield, unpublished). The sxy-5 mutation is a G-36 to U-36 transversion that alters the potential base pairing of sxy mRNA between G-36 and U+53 (Figure 3.4), to LL36 U4-53; a sequence that is not able to form a base pair (Note: bases are numbered relative to translational start site of sxy; bases with negative numbers are in the 5' untranslated region of the sxy message). Because G-U base pairs are energetically less favorable than the G-C base pairs disrupted by the other sxy hypercompetent mutations, it was surprising that strains carrying the sxy-5 allele are as competent as strains carrying the sxy-1, sxy-2. sxy-3 and sxy-4 mutations (data not shown). Therefore, it seemed important to confirm that the sxy-5 mutation causes this hypercompetence, by creating this mutation in an otherwise nonmutagenized background (i.e. KW20). I constructed the sxy-5 point mutation by site-directed mutagenesis, and subcloned this mutation into pDJM90 to create pLBS5 (as described in Section 2.2.7). The sxy-5 mutation was transferred to the KW20 chromosome using a marker rescue experiment, as described in Section 2.2.8.1. Briefly, pLBS5 was digested with EcoRI + CM, and approximately 1 pg of linearized DNA was used to transform MlV-competent KW20. As a negative control, linearized pDJM90 was simultaneously used to transform KW20. 84 To enrich for hypercompetent strains that had acquired the sxy-5 allele, transformants were grown to the early phase of logarithmic growth in sBHI (OD600 ^ 0.1), at a stage when transformants are not detectable for wild type cells, and transformed to nalidixic acid resistance with MAP7 DNA. (pDJM90 transformants were also put through this second round of transformation as a negative control). Nal r colonies were then tested for competence by colony transformation assays as described in section 2.1.4.1. The results of colony assays are shown in Figure 3.8. Panel A shows results obtained with pDJM90 transformants. As expected, all of the colonies from the negative control population had transformation frequencies of 5L1X10"5. Panel B shows results obtained with pLBS5 transformants. 52% of the colonies were scored as hypercompetent, with transformation frequencies between 1.0 X 10"4 to 5.0 X 10"3. In a duplicate experiment (data not shown), 72% of colonies were scored as hypercompetent. (Note: A time course of spontaneous competence was not carried out for any of the hypercompetent strains). These results indicate that the sxy-5 mutation is sufficient to cause hypercompetence. Therefore, disrupting the G-36-U4-53 base pair within sxy RNA structure A may be sufficient to destabilize this secondary structure. An alternative (but not mutually exclusive) hypothesis is that the sequence alteration, rather than an alteration of secondary structure, affects sxy expression by altering recognition by an RNA binding protein or another regulatory factor that binds to this region of the sxy mRNA. 3.2.3. Transcriptional regulation of sxy expression According to the model for competence development presented in Section 3.1.4, sxy expression is expected to increase by the end of logarithmic growth in sBHI, possibly as a result of a rise in the concentration of cAMP. A further, cAMP-independent 85 40 - f 10-5 10_4 Colony transformation frequency Figure 3.8. Results of colony transformation assays for KW20 transformed with EcoRI- and Clal- digested pDJM90 (sxy)(A) or EcoRI-and Clal- digested pLBS5 (sxy-5) (B). Colonies were picked into 5 ml sBHI containing 0.1 Mg/ml MAF7 DNA. After 15 minutes at 37°C, cultures were diluted and plated onto sBHI agar and sBHI agar containing novobiocin. The x axis gives the transformation frequency exhibited by the cells in each tested colony. The y axis indicates the number of colonies showing that transformation frequency. increase in sxy expression is expected for cells transferred to MIV. Therefore, I addressed the following questions: 1. Does the amount of available sxy transcript vary throughout growth in sBHI and in MIV? Do differences in the amount of sxy transcript correlate with differences in the magnitude of transformation frequencies? 2. Is sxy transcription regulated by cAMP? I addressed these questions by measuring p-galactosidase activities of wild type and cya strains carrying sxydacZ operon fusions. The construction of pLBSFl (sxy::/acZ operon fusion) is described in Section 2.2.1. In pLBSFl (Figure 3.9), lacZ is fused to sxy within codon 89 of the sxy open reading frame, downstream of where sxy RNA secondary structure A ends. This operon fusion is referred to as the sxyS9'-'lacZ operon fusion, to distinguish it from the corresponding protein fusion, and from operon and protein fusions having lacZ fused to another position within the sxy open reading frame (Section 3.2.5). Note that this fusion contains an in-frame TAA stop codon upstream of the lacZ RBS, which ensures that translation of lacZ does not initiate from the sxy RBS. pLBSFl was digested with Apal + BamHI and used to transform MlV-competent KW20 and RR736 (cya). Transformants containing single copy chromosomal gene fusions (Sxy-) were isolated by selecting for Kan r transformants following additive replacement of sxy DNA. 3.2.3.1 The production of Sxy is regulated at the level of mRNA availability. The (3-galactosidase activity of RR844 (sxy89"lacZ operon fusion) grown in sBHI is shown in Figure 3.10. During logarithmic growth, activity of the operon fusion increases approximately 22- fold, from 10 Miller units to approximately 220 Miller units (see Section 2.2.3 for a description of the calculation of Miller units). This peak of 87 ATC CCA CAG CCG CCA GTT CCG CTG GCG GCA TTT TAA CTT TCT TTA TCA Stop sxy t ranslat ion CAC AGG AAA CAG CT ATG A C C . . . 3 ' lacZ R B S lacZ cod ing sequence Figure 3.9. The operon fusion plasmid pLBSFl., showing the sxy promoter (small arrow), and sxy and lacZ ribosome binding sites and translational start sites (dark and light boxes, respectively) * sxy/lacZ junction sequence at sxy codon 89. 88 5 0 0 " > 4 5 0 " (0 J 0) 350 H (0 • § 3 0 0 " 10 5 i 1 1 1 1 1 ° 8 8 8 8 8 *—< m in Time in sBHI (min) Figure 3.10. P-galactosidase expression (A) and growth (B) of RR844 (sxygy.-.lacZ operon fusion) in sBHI. Two replicate cultures of RR839 cells were grown in sBHI. At the time points indicated, each culture was plated, to determine cfu/ml, and assayed (in duplicate) for p-galactosidase activity (as described in Section 2.2.3). cfu/ml values are the mean values of replicate cultures. For P-galactosidase activities, duplicate samples were averaged for each replicate culture, and the means of the replicate values were plotted. Error bars, which represent the range of the replicate activities, are shown only for data points that deviated from the mean replicate value by > 15 Miller units. 89 P-galactosidase activity reaches a plateau by the time cells have reached a density of ~ 5 X 107 cfu/ml, and remains at this level until cells reach a density of ~ 1 to 5 X 109 cfu/ml, at which time this peak of activity levels off and decreases slightly (by ~ 10%). As cells approach stationary phase (density of > 5 X 109 cfu/ml), at the time when competence development is maximal in wild type cells, there is a second increase in p-galactosidase activity to about 260 Miller units. This biphasic pattern of P-galactosidase expression was highly reproducible. The apparent increase in sxy expression throughout logarithmic growth in sBHI is most likely due to an increase in sxy mRNA availiability, since sxyv.lacZ protein fusions gave an identical pattern of P-galactosidase expression throughout this stage of growth (see Section 3.2.4). However, since P-galactosidase measurements provide only an indirect measurement of relative mRNA levels, it is unclear whether the increase in activity of the operon fusion during exponential growth is due to an increase in transcriptional initiation at the sxy promoter, or to another factor, such as a change in sxy message stability or mRNA elongation rate. The data in Figure 3.10 implies that sxy mRNA increases about 20-fold during logarithmic growth, and that mRNA levels are high well before the onset of stationary phase, many generations prior to maximal competence development. Therefore, although sxy expression was increased during growth of cells in sBHI, consistent with our model, this increase in expression occurred at an earlier growth stage than expected. The fact that sxy mRNA levels are high several generations prior to maximum competence development is surprising, and will be considered further in the Discussion. The biphasic pattern of p-galactosidase production could arise from one of several phenomena. One is transcriptional initiation at more than one sxy promoter i.e. transcription initiates from a different region of sxy DNA during logarithmic growth than it does during stationary phase in sBHI. Zulty and Barcak identified only one 90 major sxy promoter using primer extension; however, as previously discussed (Section 3.1.3.1), these authors did not assay cells prior to OD600 0.2 to 0.3 in rich media. Therefore, the possibility of more than one sxy promoter cannot be ruled out. The biphasic pattern of sxy operon fusion expression could also result because of growth-regulated differences in sxy mRNA stability, as has been reported for the ompA message (139). Currently, it is not known if the observed biphasic pattern of sxy operon fusion expression is functionally significant. f3-galactosidase expression from RR844 following transfer of cells to MIV is shown in Figure 3.11. Activity of the operon fusion increases rapidly upon transfer of cells to MIV, and levels off at approximately 400 Miller units after 90 minutes. Maximal (3-galactosidase activities are moderately higher (1.5 fold) in MIV than in sBHI, and B-galactosidase activities increase much more rapidly upon transfer to MIV than in sBHI, indicating that the kinetics of sxy expression may be a factor in the development of the high levels of competence seen in MIV. Zulty and Barcak, using primer extension, also showed that sxy transcription is increased upon transfer of cells to MIV (200). The signal responsible for increased sxy transcription in MIV is unknown. 3.2.3.2. Regulation of sxy transcription may be independent of cAMP. According to the simple model for competence development, transcriptional activation of sxy should be regulated by changes in the concentration of cAMP. Zulty and Barcak identified potential CRP sites within the sxy promoter region (see Figure 3.1), and reported that expression of a sxy.lacZ operon fusion was decreased 4-fold in a crp strain (200). Sxy operon fusion results reported above (Section 3.2.3.1) imply that sxy transcription is moderately high during logarithmic growth, and may be induced to even higher levels as cells are starved in MIV. Macfadyen's indirect measurements 91 u (0 a> CO (0 T 3 w In o c Co »_ 75 J) a 6 (A C <u 3 CO Q 3 & 0.01 Time (min) in sBHI Figure 3.11. P-galactosidase expression (A) and growth (B) of R R 8 4 4 (sxyggidacZ operon fusion) in sBHI (open symbols) and MIV (filled symbols). sBHI P -galactosidase data is replotted from figure 3.10. Two replicate cultures of R R 8 4 4 were grown in sBHI and transferred to MIV at the time indicated by the arrow. At each time point, cultures were assayed for O D ^ Q Q , and assayed (in duplicate) for P-galactosidase activity, as described in Section 2.2.3 (note that cell samples in sBHI prior to 180 minutes were too dilute for an accurate determination of O D ^ Q Q ) . O D 6 0 0 values are the mean values of replicate cultures. P-galactosidase activities were determined as described in the legend of Figure 3 .10. Error bars, which represent the range of the r eplicate P-galctosidase activities, are shown only for data points that deviated from the mean replicate value by >15 Miller units. 9 2 of cAMP concentrations using cAMP-dependent reporter fusions suggested that cAMP levels increase as cells enter stationary phase, and are increased even further immediately after cells are transferred to MIV (120). Therefore, it seemed reasonable that the changes in sxy operon fusion activities could reflect increasing cAMP levels in sBHI and in MIV. Cells with cya mutations cannot synthesize cAMP. In order to address whether cAMP activates sxy transcription, I measured (3-galactosidase production from cya cells carrying sxy89'-lacZ operon fusions (RR874). I also tested the effects of the addition of exogenous cAMP on wild type strains carrying sxysgdacZ operon fusions (RR844). (3-galactosidase activities of RR844 (sxy%9\:lacZ operon fusion) and RR874 (cya sxyS9'-'-lacZ operon fusion) in sBHI and in MIV are shown in Figure 3.12. Surprisingly, (3-galactosidase activities of RR874 in sBHI were not dramatically lower than those of RR844. For cells in MIV, f3-galactosidase levels of RR874 were only slightly lower (by approximately 20%) than those of RR844. The reason for the discrepancy between these results and those reported by Zulty and Barcak is unclear (200). (Note: these authors did not specify the sequence of sxy to which lacZ was fused to create their sxydacZ operon fusion; it is possible that the discrepancy between their results and those reported here may result because of the use of different gene fusion points). Results presented here suggest that the increased p-galactosidase expression from sxydacZ operon fusions in sBHI occurs independently of cAMP/CRP transcriptional activation. In MIV, cAMP may slightly (< 2-fold) activate sxy expression, but can't be the primary regulatory determinant of sxy transcription. These results are consistent with other observations, as follows. Neither of the proposed CRP sites identified by Zulty and Barcak are strong matches to the E. coli CRP consensus sequence (Section 3.1.3.1). In fact, one of the proposed CRP sites (CRP2; see Figure 3.10) overlaps the 93 u CO 0) CO CO •g "co 'to-co c CO !_ 75 a> • > CO c T 3 ^ d> O i= O 3 <o 0.01 Time (min) in sBHI Figure 3.12. P-galactosidase expression (A) and growth (B) in sBHI (open symbols) and MIV (filled symbols) of RR844 (sxygy.-.lacZ operon fusions; circles) and RR874 (cya sxy^.-.lacZ operon fusions; squares). Cells growing in sBHI were transferred to MIV at the time indicated by the arrow. At the indicated times, cell samples were removed and assayed for O D 6 0 0 and (in duplicate) for P-galactosidase activity, as described in Section 2.2.3. P ~ galactosidase activities were averaged from duplicate samples at each time point. sxy transcriptional start site; whereas functional CRP sites are usually centered about 40 base pairs upstream of the transcriptional start site. CRP sites overlapping the core promoter usually cause repression (39). In addition, Macfadyen's indirect measurements of cAMP concentrations indicate that cAMP levels within KW20 in sBHI begin to increase only when cells reach a density of ~109 cfu/ml, whereas results presented in Figure 3.10 show that sxy mRNA levels increase well before this stage (120). Since cAMP addition increases cellular transformation frequencies ten thousand- fold during exponential growth of cells in sBHI (194), I tested whether cAMP increases (3-galactosidase expression from sxyv.lacZ operon fusions. Figure 3.13 shows the P -galactosidase activities of RR844 in sBHI with or without the addition of 1 mM cAMP. Contrary to expectation, cAMP actually moderately decreased p-galactosidase activities of RR839 in both sBHI and MIV. The reason for this reduction in sxy expression is unclear, but may simply have been due to the slight reduction of growth observed here, and previously reported (193), for strains to which exogenous cAMP have been added. It is possible that the high concentrations of cAMP repressed transcription at the sxy promoter (and perhaps competence development) by binding to sxy CRP site 2. Overall, the results presented here, which are reproducible, do not support a role for CRP/cAMP in the transcriptional activation of sxy. 3.2.4. Sxy expression may be posttranscriptionally regulated. Results described above indicated that sxy expression is regulated at the transcriptional level. I was also interested in determining whether sxy expression was regulated at the posttranscriptional level, for example, at the level of either translational initiation or translational attenuation. The sxy hypercompetent mutations disrupt a potential RNA secondary structure that may negatively regulate sxy expression; these mutations are therefore unlikely to be exerting their effects at the level of transcriptional initiation. 95 A. B. Time (min) in MIV l 1 i 1— — (N CO ITi Time (min) in sBHI Figure 3.13. P-galactosidase expression (A) and growth (OD 6 0 0) (B) of RR844 (sxy89::lacZ operon fusions) with (triangles) and without (circles) the addition of 1 mM cAMR Open symbols; cells in sBHI; closed symbols, cells in MIV. Data for cells without cAMP addition is replotted from Figure 3.12. Cells growing in sBHI were transferred to MIV at the time indicated by the arrow. At the indicated times, cell samples were removed and assayed for O D 6 0 0 and (in duplicate) for P-galactosidase activity, as described in Section 2.2.3. P-galactosidase activities were averaged from duplicate samples at each time point. The RBS and AUG of the major sxy open reading frame may be sequestered within sxy RNA secondary structure A (see Figure 3. 20), so this structure could repress translational initiation of sxy mRNA. There are many examples of genes whose translational initiation is negatively regulated by the sequestering of a RBS in an RNA secondary structure (see Discussion, Chapter 5). The protein fusion plasmid pLBSF2 (Figure 3.14) was constructed as described in Section 2.2.1. The protein fusion in this plasmid, referred to as the sxy89'-lacZ protein fusion, has the same sxy/lacZ sequence junction as the sxy^.lacZ operon fusion (i.e. lacZ is fused to sxy at codon 89), permitting direct comparisons of P-galactosidase expression from the two types of fusion. In protein fusion constructs, the lacZ gene lacks its own RBS and AUG, so lacZ gene expression is governed by both the transcriptional and translational regulatory signals of sxy. pLBSF2 was digested with Apal + BamHI, and linearized DNA was used to transform MlV-competent KW20. Single copy chromosomal gene fusions (Sxy") were isolated by selection for Kan r transformants obtained by additive replacement of sxy DNA. A comparison of P-galactosidase production from RR844 (sxyg9'lacZ operon fusion) and RR845 (sxyS9'-lacZ protein fusion) is shown in Figure 3.15. During exponential growth • of cells in sBHI (open symbols), lacZ expression was the same for both fusions. However, as cells began to enter stationary phase, lacZ expression increased more rapidly from the protein fusion than from the operon fusion. At the last data point measured in Figure 3.15, lacZ expression from the protein fusion was about 2-fold higher than that from the operon fusion. The observed differences in lacZ expression from sxydacZ operon and protein fusions under specific culture conditions implies that translatability of sxy may be regulated. For cells in MIV (Figure 3.15, filled symbols), maximal lacZ expression from sxydacZ protein fusions was not higher than that from operon fusions. However, maximal 97 Sxy coding sequence 85 86 87 88 89 TAT GCA CTT TCT GAT CCG TCG ACC TGC AGC CAA GCT TGC GAT CCC BclI/BanHI LacZ codon 8 Figure 3.14. The protein fusion plasmid pLBSF2, showing the sxy promoter (small arrow), and sxy ribosome binding site and translational start site . sxy/lacZ junction sequence at sxy codon 89. 98 o (0 <D (A (0 •g o •*-> c 75 a) o> = c V •o ^ 0) o O 5 to •= Q 0.01 Time (min) in sBHI Figure 3.15. P-galactosidase expression (A) and growth (OD^QQ) (B) of R R 8 4 4 (sxygg-.-.lacZ operon fusion; circles) and R R 8 4 5 (sxyggy.lacZ protein fusion; squares) in sBHI (open symbols) and MIV (filled symbols). (3- galactosidase activities of R R 8 4 4 in sBHI and MIV are replotted from Figures 3 .10 and 3 .11 , respectively. Two replicate cultures each of R R 8 4 5 were grown in sBHI and transferred to MIV at the time indicated by the arrow. All samples were measured for O D ^ Q Q and P-galactosidase activities as described in the legend of Figure 3 .10. Error bars, which represent the range of the replicate p-galactosidase activities, are shown only for data points that deviated from the mean replicate value by > 1 5 Miller units. 99 lacZ expression from sxyv.lacZ protein fusions occurred more rapidly (after 45 minutes in MIV) than maximal expression from operon fusions (achieved after- 90 minutes in MIV). This result indicates that sxy translatability, in addition to being increased at the end of logarithmic growth in sBHI, may also be increased upon transfer of cells to starvation medium. Overall, these results suggest that the difference in lacZ expression from sxyv.lacZ operon and protein fusions may be due to regulation of sxy translatability. This hypothesis is strengthened by results in Section 3.2.5. which show that sxy translation is increased by the removal of sequences corresponding to the 3' half of stem IA of sxy RNA structure A. However, another interpretation of the data which should not be discounted is that the exogenous RBS of the E. coli lacZ (which governs lacZ translation from operon fusions) is not utilized as efficiently as the H.influenzae sxy endogenous RBS (which governs lacZ translation from protein fusions). In this situation, however, we would expect the differential P-galactosidase expression from operon and protein fusions to be independent of culture conditions, which does not seem to be the case. 3.2.5. Eliminating the 3' half of sxy RNA structure A results in an increase in sxy transcription and translation. Our simple model for competence development predicts that 5' sequences of sxy mRNA fold into an RNA secondary structure that negatively regulates sxy expression. The point mutations sxy-l-sxy-5 are presumed to cause hypercompetence because they destabilize this secondary structure, thereby eliminating or reducing negative regulation of sxy (evidence of this is presented in Section 3.2.6.1). Because sxy expression may be regulated at the level of translation (described above), a possible function of sxy RNA structure A is the regulation of sxy translatability. Therefore, I 100 investigated whether elimination of structure A would increase P-galactosidase expression from sxyv.lacZ operon or protein fusions. Operon and protein fusions plasmids in which the 3' half of sxy RNA structure A was deleted (pLBSF3 and pLBSF4, respectively) were constructed as described in Section 2.2.2.2. These fusions are referred to as sxy\\:lacZ operon and protein fusions because lacZ is fused to codon 11 of the sxy open reading frame (Figures 3.16 and 3.17). pLBSF3 and pSLBSF4 were digested with Apal + BamHI, and linearized DNA was transferred to the KW20 chromosome by transformation of MlV-competent KW20 to Kan r . (3-galactosidase production from RR844 and RR846 (sxyf&v.lacZ operon fusion and sxy\\::lacZ operon fusion, respectively) in sBHI is shown in Figure 3.18. P-galactosidase activities were about twofold higher (450 vs. 250 Miller units) in RR846 than in RR844 throughout growth in sBHI. These data suggest that disruption of sxy RNA structure A increases the amount of sxy mRNA by about 2 fold. A comparison of RR844 and RR846 in MIV was not done. P-galactosidase production from RR845 and RR847 (sxy89'-lacZ protein fusion and sxy\\\\\acZ protein fusion, respectively) in sBHI is shown in Figure 3.19. Two significant observations come from these data. First, maximum P-galactosidase activities of RR847 are 5 to 7-fold higher than those of RR845. Second, P-galactosidase production in RR847 is high even at low cell densities (i.e. during logarithmic growth). A comparison of P -galactosidase activities of RR845 and RR847 in MIV was not done. These data strongly suggest that sxy RNA structure A limits the accumulation and translatability of sxy mRNA and provide further evidence for the proposed regulation of sxy translation (Section 3.2.4) in post-logarithmic growth in sBHI. This level of regulation may be mediated by limiting access of ribosomes to the sxy RBS 101 B. A u G U A G A U U U Al A u l C G U A U rt UL sxy RNA structure A IA U A A U A . A A A U U u u c u c A G U C A U C U C U 5' Figure 3.16. A. The operon fusion plasmid pLBSF3. B. Sxy RNA showing the elimination of the 3' strand of stem IA of structure A by sxyu::lacZ operon and protein fusions ). The solid line outlines sxy protein coding sequence, which is interrupted within codon 11 by the lacZ fusions. The small arrow shows the sxy promoter, and the sxy and lacZ ribosome binding sites and translational start sites are indicated by dark and light boxes, respectively. 102 Sxy coding sequence.. 1 2 3 4 5 6 7 8 9 10 11 5' ATG AAT ATA AAG GAT GAG CAT ATA GAT AGC GGA ATT CCC GGG GAT CCG TCG AGC EcoRI CAA GCT TGC GAT CCC LacZ codon 8 Figure 3.17. A. The protein fusion plasmid pLBSF4. sxy IlacZ junction sequence at sxy codon 11. The small arrow shows the sxy promoter, and the location of the sxy ribosome binding site and translational start site is indicated by the box. 103 U.U1 1 1 1 1 1 O O Q O © U~l O u-l © r-l H <N Time (min) in s B HI Figure 3.18. p-galactosidase expression (A) and growth (B) of RR839 (sxy39::lacZ operon fusion; circles) and RR863 (sxyn::lacZ operon fusion; squares). One culture of RR839, and two replicate cultures of RR863, were grown in sBHI. For each time point, samples of cultures were removed and assayed for O D 6 0 Q and (in duplicate) for P-galactosidase activity, as described in Section 2.2.3. For RR863, O D 6 0 Q values are the mean values of replicate cultures. For RR839, P-galactosidase activities were averaged from duplicate samples at each time point. For RR863, P-galactosidase activities were determined as described in the legend of Figure 3.10. Error bars, which represent the range of the r eplicate P-galactosidase activities for RR863, are shown only for averaged data points that deviated from the mean replicate value by >15 Miller units. 104 1600 o.o H 1 1 1 1 O O o o o ir> O *r> O —c — i cs Time (min) in sBHI Figure 3.19. P-galactosidase expression (A) and growth (B) of RR845 (sxyggr.lacZ protein fusion; circles) and RR847 (sxy^::/acZ protein fusion; squares). Two replicate cultures of RR845, and three replicate cultures of RR847, were grown in sBHI. At the indicated times, samples of each culture were removed and assayed for O D 6 0 0 and (in duplicate) for p-galactosidase activity, as described in Section 2.2.3. O D 6 0 0 values are the mean values of replicate cultures. P-galactosidase activities were determined as descibed in the legend of Figure 3.10. Error bars, which represent the range of the r eplicate p-galactosidase activities, are shown only for averaged data points that deviated from the replicate mean by > 50 Miller units. 105 and AUG sites. The increased amount of sxy mRNA may be due to a direct affect of the secondary structure on sxy transcription or mRNA stability, or it may be due to an indirect effect of increased sxy translatability (see Discussion, Chapter 5). 3.2.6. Point mutations that strengthen or weaken base pairing in sxy RNA structure A affect sxy expression and competence development. The specific questions addressed in this section are: 1. Is wild type sxy mRNA predicted to fold into RNA secondary structure A? Do the sxy hypercompetence mutations alter predicted sxy RNA folding patterns? 2. Does the sxy-1 mutation cause hypercompetence by increasing sxy expression? 3. Do point mutations that strengthen base pairing of sxy RNA structure A decrease transformation frequencies and sxy expression? 3.2.6.1. Mfold predicts a sxy RNA secondary structure that is destabilized by the sxy hypercompetence mutations. The folding of the sxy wild type and hypercompetent mutant mRNAs into RNA secondary structure was predicted using the program Mfold (9,199). The results are shown in Table 3.2 and Figure 3.20. Mfold results obtained with the wild type sequence predicted predominantly sxy structure A, shown in detail in Figure 3.20. While structure A is not thermodynamically more favorable than the other predicted RNA secondary structures, it is believed to represent the biologically relevant structure for the following reasons. All of the hypercompetence mutations dirsrupt base pairs in stem IA of this structure, strongly suggesting that the mutations might act by altering this base pairing and thus the formation of structure A. Structure B also contains one 106 -10*G ,U A U , A „ - G , A U Struct ure A IB u IC - 2 0 U A A G G U A U u A A U » » S 1 „ UG c u C G U U U U C G **» C G •30 A A ' U A „ A U . G U — A ti -30U U UC A C * » cu',© • u • A -o u IA A - A •** -40 IJ II U •7 0 j A — U.60 A~U A vn A -u A ~u • ~ A c u A A " G U*9° u u u \ A*110 u V G A U \ A IV . u A U 5' A -SO A G A U , u*130 A " , / " " G ^ * G U ° G »,l A * A •^140 A A A St ructure B III IV © • u © . u •60 C C — G A , 3' VI S G A •V A VII A U « \ „ u • u v V » -50 G 5' 3' Figure 3.20. MFold RNA secondary structures obtained by folding sxy wild type and mutant RNAs. Bases are numbered relative to the translational initiation codon (+1). The sxy Shine Dalagarno sequence is underlined; bases mutated in the sxy hypercompetent strains are circled. 107 U A A -20 J U u -30" Structu re C A U U < A ^  C*30 U u * 2 0 A ~ u U * V c G ^ c G I A ^ „ U „ A A. G GG U • U c • 60 • 10 A A A G C U •1 u A •50.11 \ A U A C U , u o '\ c * u G '•90 IV C A . G . G 5' 3' IV III .70 . c \ J G U A C -•30 U A .+60 'u u . + 50A - u C — G C / ,A / U •20 A "A U V / A ' 1 0 A « » „ ' / A C u A A - M U u V 0 * c / A - U* 8 0 • 4 0 u © u U A © * U A c u G A A -40 u u c u I I I I 0 A •50 5' • 100 . C . A+110 U \ G A G u \ c U \ A° A •<« VI .130 3' Figure 3.20. continued... 108 Table 3.3. Mfold RNA secondary structure predictions for sxy wild type and sxy-1 to sxy-5 sequences. For each sequence analyzed, many alternative RNA secondary structures were predicted (from 21 to 26, depending on the sequence used). Each structure was classified according to whether the 5' region of the RNA folded into Structure A through B in Figure 3.20, or into an additional secondary structure whose structure is not shown. For each sequence analyzed, the number of RNA foldings that supported a particular RNA secondary structure are shown below. RNA secondary structures predicted by Mfold (Figure 3.20) Sxy sequence Structure A \ Structure B i Structure C 1 Struct ure D 1 Structure E Other structure sxy 19/21 * • 1/ 21 j 1/ 21 sxy -1 sxy -2 I 16/2 3 ! 5 > k > I 17/2 6 9/ 26 sxy-3 2/ 22 i } 18 /23 sxy -4 1/ 26 I W/2 6 | | sxy -5 | 19/2 4 109 stem that is formed by base pairing between bases that are mutated by the sxy-3, -4 and -5 mutations (Stem I of structure B), and another stem that is formed by base pairing between bases mutated by the sxy-1 and -2 mutations (Stem IV of structure B). However, sxy\\::lacZ operon and protein fusions (Section 3.2.5) would be expected to eliminate only Stem IV of structure B, and the bases disrupted by the sxy hypercompetence mutations in this stem are both weak G-U base pairs. The predicted structures obtained by folding mutant RNAs also support the hypothesis that these mutants exert their effectsby disrupting stem I of structure A (Table 3.2). Predicted structures of sxy-1 and sxy-2 RNAs didn't include structure A. Instead, structures B and E, respectively, were predominant. Predicted structures of sxy-3 included structure A in 2/22 foldings, but predominantly (18/23 foldings) supported structure B. Mfold predictions obtained with sxy-4 RNA included structure A in 1/26 foldings, but predominantly (13/26 foldings) supported structure C. Finally, Mfold predictions obtained with sxy-5 RNA didn't include structure A; the predominant (19/24 foldings) structure predicted for this mutant sequence was structure D. These results strongly suggest that the sxy hypercompetence mutations destabilize sxy RNA structure A. 3.2.6.2. The sxy-1 point mutation increases expression of P-galactosidase from sxy operon and protein fusions. I examined the effect of the sxy-1 mutation on lacZ expression from sxy&ydacZ operon and protein fusions. This mutation was chosen because its phenotype has been extensively characterized (155,192). Because all of the hypercompetent mutations cause equivalent mutant phenotypes (Figure 3.3), they are likely to affect sxy expression by the same mechanism. Sxylg9'-lacZ operon and protein fusions plasmids pLBSF5 and pLBSF6 were created as described in Section 2.2.2. pLBSF5 and pLBSF6 are identical to pLBSFl and pLBSF2, 110 respectively, except for the presence of the sxy-1 mutation. pLBSF5 and pLBSF6 were digested with Apal + BamHl, and the linearized DNA was used to transform KW20 to Kan r resistance to yield single copy chromosomal fusions. P-galactosidase production from RR844 (sxyS9'.-lacZ operon fusion, circles) and RR848 (sxyl89'-lacZ operon fusion, squares) in sBHI and MIV is shown in Figure 3.21. P -galactosidase activities of RR848 were approximately 1.5 to 2-fold higher than those of RR844 at all stages of growth in sBHI and in MIV, consistent with the hypothesis that the sxy-1 mutation increases sxy expression. P-galactosidase production from RR845 (sxy89"lacZ protein fusion) and RR849 (sxyl89"lacZ protein fusion) throughout growth of cells in BHI and MIV is shown in Figure 3.22. P-galactosidase activities of RR849 were only slightly higher (250 units compared to 175 units) than those of RR845 at the beginning of the experiment (OD600~0-05). However, the difference in lacZ expression from protein fusions in RR844 and RR849 increased to about 3-fold during post-logarithmic growth and also after cells were transferred to MIV. These observations have several implications. They confirm the prediction that the sxy-1 mutation increases sxy expression. The sxy-1 mutation increases the amount of available sxy mRNA by about 2-fold, by an unknown mechanism. In addition, the proposed negative regulation of sxy translation in wild type cells appears to be partially relieved by the sxy-1 mutation, by about 1.5-fold, because lacZ expression from sxy-1 protein fusions is approximately 3-fold higher than that from sxy protein fusions, whereas lacZ expression from sxy-1 operon fusions is 2-fold higher than that from sxy operon fusions. Therefore, the increased expression of lacZ from sxy-1 protein fusions can partially be explained by increased translatability of the sxy-1 RBS compared to translatability of the wild type RBS. The initiation of translation at the sxy-1 RBS appears to be regulated by culture conditions, unlike the situation observed for sxy\\ protein fusions, in which sxy translatability seemed to be high even during logarithmic 111 o CO 0) CO CO •g "CO o c O l 4 0 0 i 1600-1400 • 1200" 1000 H 800 "J 600-A. CO c T 3 3 § 0.01 Time (min) in sBHI Figure 3.21. fi-galactosidase expression (A) and growth (B) of RR39 (sxy^v.lacZ operon fusion; circles) and RR865 (sxy-1 ^ wlacZ operon fusion; squares) in sBHI (open symbols) and MIV (filled symbols). Replicate cultures of RR839 and RR865 were grown in sBHI and transferred to MIV at the time indicated by the arrow. O D 6 0 0 values are the mean values of replicate cultures. (5- galactosidase activities were determined as described in the legend of Figure 3.10. Error bars, which represent the range of the replicate (3-galactosidase activities, are shown only for averaged data points that deviated from the replicate mean by > 50 Miller units 112 o CO a> co co T 3 "Jo O <2 o c •S ^ co a5 o> = co c a> T 3 i o f t ? O S . 1600 1400 H 1200H 1000 H 0.01 Time (min) in sBHI Figure 3.22. p-galactosidase expression (A) and growth of RR862 (sxy^v.lacZ protein fusion; circles) and RR844 (sxyl^::lacZ protein fusion; squares) protein fusions in sBHI (open symbols) and MIV (filled symbols). Replicate cultures of RR862 and RR844 were grown in sBHI and transferred to MIV at the time indicated by the arrow. O D 6 0 0 values are the mean values of replicate cultures. P ~ galactosidase activities were determined as described in the legend of Figure 3.10. Error bars, which represent the range of the replicate P-galactosidase activities, are shown only for averaged data points that derviated from the replicate mean by > 50 Miller units. 113 growth of cells. This implies that the sxy-1 mutation reduces but doesn't completely eliminate the negative regulation of sxy expression mediated by sxy RNA structure A, which may explain why competence is not constitutive in cells carrying the sxy hypercompetence mutations. It was surprising that lacZ expression from sxy-1 fusions of cells in MIV was higher than that from sxy fusions. Our model predicts that transformation frequencies should be correlated with the amount of Sxy produced, and that maximal amounts of transformation are limited by amounts of Sxy available. Although strains carrying the sxy-1 mutation are no more competent than wild type cells in MIV, my results indicate that strains carrying the sxy-1 mutation probably produce more Sxy in MIV than do wild type cells. These data indicate that the level of MIV competence is not limited by the concentration of Sxy, but is limited instead by other factor(s), such as the cellular recombination machinery. 3.2.6.3. Site-directed mutations designed to stabilize sxy RNA structure A affect sxy expression and competence development. In order to provide additional in vivo evidence that sxy mRNA folds into RNA structure A, I created point mutations designed to stabilize this secondary structure, and examined the effects of these point mutations on competence and on lacZ expression from sxydacZ gene fusions. Because each of the known hypercompetence mutations occurs within Stem IA of sxy RNA structure A, I mutated residues within this stem. The sxy-6 mutation (Figure 3.23) was constructed as a potential compensatory mutation that combined the sxy-1 and sxy-3 mutations, thereby potentially restoring the base pairing disrupted by each single mutation. Strains carrying this mutation were no longer hypercompetent; as predicted. However, lacZ expression from sxy-6dacZ operon and protein fusions was approximately 1.5-fold higher than that from the 114 corresponding wild type fusions. Because of the discrepancy of these results, the corresponding data is not presented in the thesis. The sxy-7 mutation was designed to introduce two additional base pairs into stem IA of sxy RNA structure A (Figure 3.24). This mutation doesn't alter the Sxy coding sequence, but was expected to increase the stability of sxy RNA structure A, and to lower sxy expression and competence development below that of wild type cells. 3.2.6.3.1. Construction of the sxy-7 point mutation and transfer of this mutation to the KW20 genome. pLBS7 (pGEM::sxy-7) was created by site-directed mutagenesis, as described in Section 2.2.7. The sxy-7 mutation was transferred to the KW20 genome as described in Section 2.2.8.2 (Figure 2.10). There was no means of directly selecting for the presence of the sxy-7 mutation in the KW20 genome (the competence phenotype of cells carrying this mutation might not differ from that of wild type cells). Instead, this mutant was obtained in two steps. pLBS7 was first transferred to the KW20 genome by Campbell recombination involving a single crossover event (note: a Campbell crossover initiating within sxy generates one full length and one partial copy of sxy; Figure 2.10.A), and strains cured of the plasmid and retaining a single mutant copy of sxy were then isolated. Because pBR322 cannot replicate in H. influenzae, cells that acquire plasmid integrants will have acquired the plasmid-encoded antibiotic resistance. However, the pBR322 gene encoding antibiotic resistance is not functional in H. influenzae (personal observation). Therefore, the ampicillin resistance markers of pLBS7 and pRRS6 (pGEM::sxy-2; positive control) were converted to kanamycin resistance by transposon mutagenesis with 1NK1316 (Section 2.1.4.1). Plasmids with transposon insertions in their ampicillin resistance gene were isolated by selecting for kanamycin resistance and 115 5' A - U -10 G • U A I B J : g u u : A A - 2 0 ^ - A U I C A A+10 U A A G G U G A + 2 0 A - U U U U r U " A +30 C GA -30 e A , u u „ C n A C £ U C + 8 0 IA ^"G+50 A% U II , A G~U + > V " • U 8 : ^ A u u 'A U A - U ^ X , ,C U - A G u C-G „ U G U • G+60 U n „ -50 ^ G U A C U - A ^ G j / ^ O A G G U 3' Figure 3.23. Sxy-6 mutation (C. 3 8G + 54 ->U. 3 8A + 5 4) created by site-directed mutagenesis. A - U -10 G • U A n ^-A U+1 A - U §'-$ IB UG, +20 U , A A G G y " A \ G U U 'c 'cu' ? ' C A U A U A G A - U G + 4 f J U C G / , i . A A - U U U U r U U A + 3 0 C G A Si A C ^ U C + 8 0 H-G+50 A % U , . A - U +70 A ^ . U « A-U „C\ tC U-A UC_ G U C-G „„ U G U • G+60 U U «„ A G U A C U - A A A G ^ # + 9 0 * Q A G G J- l A G y 5 3' Figure 3.24. Sxy-7 mutation (C .3 2U. 3 1 ->G.32A.3i) created by site-directed mutagenesis. screening for ampicillin sensitivity. Approximately 10% of Kan r plasmids were Amp s . The location of transposon insertions in Kan r Amp s plasmids was confirmed by restriction analysis. Recombinant plasmids were referred to as pLBS7K (pGEM::sxy-7 kan) and pRRS6K (pGEM::sxy-2 kan). In order to generate a sxy merodiploid strain in which the full length copy of sxy contains the sxy-1 or sxy-7 mutations, Campbell recombination would have to involve a single crossover between the KW20 genome and plasmid sequences 3' to the mutations (Figure 2.10. A). However, pLBS7K and pRRS6K contain only about -400 bp of homologous DNA 3' to the mutations, and -1400 bp of homologous DNA 5' to the mutations. Therefore, in order to increase the probability of a single crossover event occurring 3' to the mutations, plasmid DNAs were linearized with SnaBl to create a double-stranded break on the 3' side of the mutations (Figure 2.10.A). It was hoped that a large percentage of single crossover events would be initiated within the region of the double-stranded break. As a positive control for this experiment, I transformed MlV-competent KW20 (sxy) with SnaBI-linearized pRRS6K (sxy-1; final concentration -200 ng/ml). In this situation, plasmid integration events arising from crossovers occurring within the region of the double-stranded break were expected to generate a full-length copy of sxy containing the sxy-1 mutation; and therefore at least some of the transformants were expected to be hypercompetent. 30 Kan r KW20 transformants of pRRS6K were tested by colony assays (Section 2.1.4.1). Sixty seven % of the colonies were found to be hypercompetent (Class II transformants, transformation frequencies of 5 X 10 -4 to 2 X 10~3), indicating that the desired recombination event had occurred. The remainder of the of Kan r colonies had transformation frequencies equivalent to those of wild type cells, indicating that the crossover event in these strains had occurred 5' to the sxy-1 mutation. 118 I then transformed MlV-competent RR699 (sxy-1) with SnaBI-linearized pLBS7K (sxy-7; final concentration ~ 200 ng/ml). Kan r transformants were recovered at a frequency of approximately 1 to 5 X 10~5. In this case, I chose RR699 as the recipient strain so that I could score for the desired crossover events, which were expected to generate a full length copy of sxy-7, and a partial copy of sxy-1; therefore reverting the hypercompetence phenotype of the recipient strain. For RR699 (sxy-1) transformed with linearized pLBS7K (sxy-7), plasmid integration events arising from crossovers occurring within the region of the double-stranded break were expected to yield a full-length copy of sxy containing the sxy-7 mutation. As discussed above, sxy-7 strains were predicted to have reduced competence relative to wild type cells. Twenty-five Kan r RR699 transformants of pLBSZK were tested by colony assays. Eighty-three % of the tested colonies were hypercompetent (Class I transformants, transformation frequencies of 5 X 10"4 to 2 X 10"3), indicating that the crossover in these strains occurred within plasmid sequences 5' to the sxy-7 mutation. The remainder of the tested colonies were no longer hypercompetent (Class II transformants, transformation frequencies <1 X 10"6 to 1 X 10~5), indicating that the crossover in these strains occurred within plasmid sequences 3' to the sxy-7 mutation. Southern analysis (as described in Section 2.1.6.5) was used to confirm single copy plasmid integration into the genomes of several Class II Kan r transformants of pLBS7K. Chromosomal DNAs were digested with EcoRI and Seal, and separated on 0.8% TAE agarose gels. Digested DNAs were then transferred to membranes, and the membranes were probed with either pGEM7 DNA (pGEM7 linearized with EcoRI) or with sxy DNA (pDJM90 1.8 kb EcoRI - BamHI insert). Results of the Southern analysis confirmed that each of the Class II Kan r transformants was a sxy merodiploid strain containing a single copy plasmid integration. The sxy merodiploid constructs in Kan r transformants were likely to be unstable because of the potential for homologous recombination between duplicated sequences 119 of rec-1 and sxy. To create strains containing single copies of sxy-7,1 grew Kan r transformants (2 each of pLBSTK Class II transformants) in the absence of kanamycin to allow for the elimination of duplicated sequences by homologous recombination (Figure 2.10.B). Crossovers 5' to the sxy-7 mutation were expected to generate full-length, single copies of sxy-7. Crossover events that occurred 3' of the mutation could be readily scored because they were expected to regenerate hypercompetent strains carrying a full-length copy of the sxy-1 gene. Class II Kan r isolates were grown for ~70 hours (approximately 160 generations) in the absence of kanamycin, and were diluted 1/5000 every 12 hours. Colonies were then screened for kanamycin sensitivity; Kan s strains were recovered at a frequency of between 1% and 10%. To identify Kan s isolates with the desired genotypes, 8 colonies from each original Class II Kan r isolate were screened using colony assays. Kan s colonies originating from pLBS7K Class II transformants colonies were either hypercompetent (3 X 10 -4 to 1 X 10" 3) or had transformation frequencies slightly lower than wild type (<2 X 10~6). Southern analysis was carried out on chromosomal DNA isolated from Kan s isolates. Chromosomal DNAs were digested with Seal or Apal + Bglll, and separated on 0.8% TAE agarose gels. Digested DNAs were transferred to membranes, and the membranes were probed with either pGEM7 DNA (pGEM7 linearized with EcoRl) or with sxy DNA (pDJM90 1.8 kb EcoRl - BamHl insert). Results of the Southern analysis confirmed that Kan s isolates contained single copies of the rec-1 and sxy genes, and no longer contained integrated plasmids. PCR with primers P6 and P12 (Figure 2.2.1) was carried out on Kan s isolates that were not hypercompetent. PCR products were sequenced with P10 to confirm the presence of the sxy-7 mutation. Strains carrying single copies of sxy-7 (RR854) were frozen for further studies. 120 3.2.6.3.2. The sxy-7 mutation prevents competence development and decreases lacZ expression from operon fusions. Transformation frequencies of KW20 and RR854 (sxy-7) in sBHI are compared in Figure 3.25 . The sxy-7 allele completely prevented cells from transforming in sBHI or MIV (MIV transformation frequencies <10~8; data not shown). P-galactosidase,production from RR844 (sxyggdacZ operon fusions) and RR860 (sxy-7gg::lacZ operon fusions) in sBHI is shown in Figure 3.26. P-galactosidase production from the sxy-7 fusion was approximately 2-fold lower than from the sxy fusion during logarithmic growth. At the end of logarithmic growth, p-galactosidase production from the sxy-7 operon fusion was approximately 4-fold lower than from the wild type fusion, since p-galactosidase expression was increased from sxy /usion at this stage of growth but was not increased from the sxy-7 fusion. (P-galactosidase production from the sxy-7&)::lacZ operon fusion in MIV, and from a sxy-7g9-:lacZ protein fusion, was not measured). Overall, these results suggest that the sxy-7 mutation, which was designed to increase the stability of sxy structure A, reduces the amount of available sxy transcript by 2 to 4-fold, depending on the stage of growth in sBHI. The effects of the sxy-7 mutation on the expression of lacZ from protein fusions were not assessed. The decrease in sxy operon fusion levels caused by the sxy-7 mutation may be caused by alterations in sxy transcriptional initiation or elongation, or by alterations in sxy mRNA stability (see Discussion). 121 O 10-7 ~j 4— co -zr c > 1 008 i , , 1 1 o o © o © ir> © u-> J5 Time in sBHI (min) Figure 3.25. Transformation frequencies (A) and growth (B) in sBHI of KW20 (circles) and RR853 (sxy-7; squares). Transformation frequencies were assayed as described in Section 2.1.4.1. 122 500 0.01 -T m e (mi n) i n s BH I Figure 3.26. P-galactosidase expression (A) and growth (B) of RR844 (sxy89::lacZ operon fusion; circles) and RR860 (sxy-789::/acZ operon fusion; squares) in sBHI. One culture of RR844, and two replicate cultures of RR860, were grown in sBHI. For each time point, samples of cultures were removed and assayed for O D 6 0 0 and for P-galactosidase activity, as described in Section 2.2.3. For RR860, O D 6 0 0 values are the mean values of replicate cultures. For RR844, P-galactosidase activities are the mean values of duplicate samples. For RR860, p-galactosidase activities were determined as described in the legend of Figure 3.10 (error bars ar e not shown because none of the p-galactosidase activities deviated from the mean replicate value by > 15 Miller units). 123 3.2.6.4. Preliminary Northern analysis confirms that sxy mutations increase or decrease sxy mRNA levels. Here, I discuss Northern analysis carried out to confirm that the sxy-1 mutation causes an increase in the amount of sxy mRNA and that the sxy-7 mutation causes a reduction in the amount of sxy mRNA. I expected measurements of steady state sxy mRNA amounts of KW20 and strains carrying sxy-1 and sxy-7 mutations to confirm indirect measurements of sxy mRNA from analysis of P-galactosidase from reporter strains. P-galactosidase assays indicated that the sxy-1 mutation caused a 2-fold increase in sxy mRNA (Figure 3.21). Although reporter strains carrying the sxy-3 mutation were not created, I also expected the presence of the sxy-3 mutation to increase sxy mRNA 2-fold. Finally, analysis of lacZ expression from sxy-7::lacZ operon fusions indicated that the sxy-7 mutation lowered sxy mRNA by 3 to 4-fold (Figure 3.26). KW20, RR699 (sxy-1), RR724 (sxy-3) and RR854 (sxy-7) were grown to an OD6oo ~0.7 in sBHI, and total RNA was isolated from cells as described in Section 2.2.9. Approximately 5 ug of each RNA was loaded and separated on a 6% denaturing polyacrylamide gel. RNAs were transferred to membranes, and the membranes were probed with uniformly radiolabeled antisense sxy RNA; membranes were then stripped and reprobed with antisense E. coli 5S RNA to confirm even loading of RNA samples. Membranes were exposed overnight on Phosphorlmager screens, and RNAs were quantified by segment analysis of the resulting images. The results are shown in Figure 3.27 (all samples except KW20 are shown in duplicate). For all samples except sxy-7, one major transcript was identified, as indicated by the arrow in Figure 3.27. Results for RR699 and RR854 agreed with p-galactosidase activity measurements. RR699 (sxy-1) and RR724 (sxy-3) each had approximately twice as 124 Sxy mRNA 5S rRNA^ Figure 3.27. Northern analysis of steady state sxy mRNA levels in KW20 (lane 1), RR699 (sxy-1, lanes 2 and 5); RR724 (sxy-3, lanes 3 and 6); RR854 (sxy-7, lanes 4 and 7). Each lane contains approximately 5 pg of RNA isolated from cells grown to an OD600 of 0.7 in sBHI. The membrane was probed with uniformly radiolabeled antisense sxy RNA (top panel). The membrane was then stripped and reprobed with antisense 5S rRNA (bottom panel). 126 much of the major transcript as did KW20 (Note: the results for RR699 (lanes 2 and 6) were inconsistent between samples in Figure 3.30; however, in an additional experiment (data not shown), the results confirmed those of lane 6). RR854 (sxy-7) contained no detectable transcript, confirming that this mutation affects sxy transcriptional initiation, mRNA elongation, or mRNA stability. Other minor bands of lower molecular weight were also present in the Northern; the significance of these bands is unknown - they may reflect degradation of sample RNAs. 3.2.7. In vitro RNase mapping supports the folding of sxy mRNA into the modified structure A . l or A.2. In vivo evidence suggested that the sxy-1 to sxy-5 point mutations destabilize stem IA of the sxy RNA structure A, thereby increasing Sxy production and competence development. The existence of this stem was also suggested by data obtained with sxy\\:lacZ operon and protein fusions (Section 3.2.5) and by analysis of the sxy-7 mutation (Section 3.2.6). However, in vivo experiments did not address whether stems other than Stem IA of sxy structure A exist within cells (i.e. stems IB and IC of structure A; Figure 3.20). Below, I present data on RNase structure mapping of in vitro transcribed sxy RNA. Sxy and sxy-1 DNA sequences were cloned into pGEM7 as described in Section 2.2.4. Figure 3.28. shows the sequence of the T7 promoter and 5' end of the in vitro sxy and sxy-1 transcripts. There are 15 extraneous nucleotides on 5' end of the in vitro sxy transcripts, however, MFold analysis showed the presence of these nucleotides should not alter the formation of structure A. Sxy RNAs were transcribed, end-labeled, folded and cleaved by partial RNase digestion as described in Sections 2.2.5.1 and 2.2.5.2. Partial cleavage products and sequencing ladders were run on either 12.5% (Figure 3.29; sxy and Figure 3.31; sxy-1) or 8% (Figure 3.30, sxy and Figure 3.32, sxy-1) acrylamide 127 SnaBI T7Transcription Start r sxy +1 I 5' TGTAA TACGA CTCAC TATAG GGCGA ATTGG GCCCC AGAAG TACTT CTACT 3' Apal Figure 3.28. Plasmid pGEMsxy used to generate in vitro sxy transcript from the T7 RNA polymerase promoter. The T7 promoter and sxy sequences (Sxy sequence is shown in bold). pGEMsxyl is identical to the construct shown above, except for the presence of the sxy-1 mutation. 128 sequencing gels. Undigested RNA samples were also run on the gels as controls; however, in some cases, undigested RNAs were partially digested, presumably due to contaminating RNAses. This analysis permitted structure mapping of residues from approximately -50 to +60 of in vitro transcripts. Results of structure mapping of sxy and sxy-1 structure mapping are summarized in Figure 3.33 and Figure 3.34, respectively. Overall, the RNase structure mapping of sxy provided evidence for the existence of two slightly modified versions of structure A (structures A. l and A.2, see below); however, structure mapping of the sxy-1 transcript did not confirm MFold analyses that suggested that the 5' sequences of sxy-1 mRNA fold into a structure that is different from the wild type structure. Below, I separately discuss cleavage patterns for RNase T l , RNAse CL3, and RNase Physarum M plus RNase T2 (which yielded equivalent results). RNase T l : RNase Tl cleaves unpaired G residues (64). Figures 3.20 and 3.33 show that unpaired G residues in structure A occur at residues -47, -50, -11, -10, -8, -7, +16, +25, +29, +31 and +42. Figure 3.33 shows a summary of partial RNase T l cleavage sites of sxy RNA. Many of the G residues predicted to be single- stranded in Structure A, including unpaired G residues in stems IA, IB and IC, were cleaved by RNase T l . There were some exceptions of residues that were expected to be single- stranded but were not cleaved by partial RNase Tl digestion. These included the -47 and -49 residues, whose cleavage signal may have just been too weak to detect, and the G residue at +16, which is contained in a small bulge that may be sterically hindered and not accessible to ribonuclease digestion (64). Most of the G residues predicted to be double-stranded in structure A were not cleaved by RNase T l . These included G residues at -36 (site of the sxy-3 mutation), +3, +12, +13, + 35, + 46, +51 (site of the sxy-2 mutation), +55 (site of the sxy-1 mutation), +58 and +59. 129 Ladder Structure Figure 3.29. Mapping of nuclease-sensitive sites in end-labeled sxy RNA. sxy RNA was transcribed from pGEMsxy linearized with SnaBI, and purified and end-labeled as described in Sections 2.2.5.1 and 2.2.5.2. RNA partial digestions were carried out as described in Section 2.2.5.2. Digested products were separated on a 12.5% sequencing gel. Lanes 1 to 4 contain sequencing ladders: lanes 1 and 2; RNase Tl sequencing ladder; lane 3, Physarum M RNAse sequencing ladder; lane 4; O H ladder. Lane 5 contains undigested RNA. Lanes 6-11 contain folded, partially digested RNA samples: lanes 6 and 7, RNase T l ; lane 8, RNase T2, lane 9, RNase Physarum M, lanes 10 and 11, RNase CL3. Partial digests were carried out as described in Section 2.2.5.2, except for lane 1, which contained RNA was treated with 0.02 units of RNase T l for 2 minutes, and lane 10, which contained RNA digested with 0.004 units of RNase CL3 for 50 minutes. 131 Ladder Structure 1 2 3 4 5 6 132 Figure 3.30. Mapping of nuclease-sensitive sites in end-labeled sxy RNA. sxy RNA was transcribed from pGEMsxy linearized with SnaBI and purified and end-labeled as described in Sections 2.2.5.1 and 2.2.5.2. RNA partial digestions were carried out as described in Section 2.2.5.2. Digested products were separated on a 8% sequencing gel. Lanes 1 and 2 contain sequencing ladders: lane 1, RNaseTl sequencing ladder; lane 2, O H ladder. Lanes 3-5 contain folded, partially digested RNA samples: lane 3, RNase T l ; lane 4, RNase T2; lane 5, RNase CL3. Lane 6 contains undigested RNA. 133 1 3 4 F igu re 3.31. Mapping of nuclease-sensitive sites in end-labeled sxy -1RNA. sxy -1 RNA was transcribed from pGEMsxy-1 linearized with SnaBl, and purified and end-labeled as described in Sections 2.2.5.1 and 2.2.5.2. RNA partial digestions were carried out as described in Section 2.2.5.2. Digested products were separated on a 12.5% sequencing gel. Lanes 1 and 2 contain sequencing ladders: lane 1, RNase Tl sequencing ladder; lane 2, O H ladder. Lanes 3 to 5 contain folded, partially digested RNA samples: lane 3, RNAse T l ; lane 4, RNase T2; lane 5, RNase CL3. Lane 6 contains undigested RNA. 135 Figure 3.32. Mapping of nuclease-sensitive sites in end-labeled sxy-1 RNA. sxy -1 RNA was transcribed from pGEMsxy-1 linearized with SnaBI and purified and end-labeled as described in Sections 2.2.5.1 and 2.2.5.2. RNA partial digestions were carried out as described in Section 2.2.5.2. Digested products were separated on a 8% sequencing gel. Lanes 1 and 2 contain sequencing ladders: lane 1, RNAseTl sequencing ladder; lane 2, O H ladder. Lanes 3 to 5 contain folded, partially digested RNA samples: lane 3, RNAse Tl ; lane 4, RNase T2; lane 5, RNase CL3. Lane 6 contains undigested RNA. 137 Enzyme RNase T1 RNase C L 3 RNase 121 Physarum M RNase Specificity Unpaired G's Unpaired C's Unpaired A's and U's Relative Cleavage strength Strong Moderate Weak • • IB • V • • A - V • 1 0 ^ G ' ° A • 6 G A • u u U - A +1 • A - u • • C - G U - A U - A A - U - ? P A u U - A L U • A U • A A+10 +20 IC • A - U - U U U G U « . U - A -ao^u-Ay • C C A ^ U - A +50 C - G A - u G • u +30 IA -50 A A * G A - U C - G •* U « G +60 Q U A C U - A A A U Figure 3.33. Summary of RNase structure mapping of end -labeled sxy RNA. Relative sizes of symbols indicate relative cleavage strengths. Note: Identical patterns were obtained for cleavage with RNase T 2 and Physarum M, so data from these enzymes are combined. Enzyme Specificity Relative Cleavage strength Strong Moderate Weak RNase T1 RNase CL3 RNase T2/Phsarum M RNase Unpaired G's Unpaired C's Unpaired A's and U's IB -10 • A U • A - U » G « U A * G A • U „ , U - A U + 1 • A - U C ~ G • U - A A - U , U U " A , V U A+10 A A G G +20 IC U " " O G J A f a • A - U « + 4 0 C D C G ^ T / . A • A - U U - A U . G -3tt U - A +50 3 0 u u > c c • G ~ U U - 7 A - C - 4 0 U - A * C - G « U ' G « + 6 0 -50 L F L ^ A C U - ^ 0 u u G u C G A . A. +30 IA 3' Figure 3.34. Summary of RNase structure mapping of end -labeled sxy-1 RNA. Relative sizes of symbols indicate relative cleavage strengths. The sxy-1 mutation; G+55 -A+55, is circled. It should be noted that, for the gels shown in Figures 3.30,3.31 and 3.32, undigested RNA samples were contaminated with an RNAse that gave a digestion pattern that in some cases looked very similar to that produced by RNase digestion with RNase T l . This may be because undigested samples were contaminated with RNase T l . In these cases, I have assessed RNase Tl cleavage sites based on band intensity compared to undigested samples; however, these results should be repeated using control samples not contaminated with nucleases. RNA structure mapping of sxy-1 RNA gave results that were very similar to those obtained with the sxy message (Figure 3.34); however, some of the cleavage sites were weaker or absent in the sxy-1 RNA. Cleavage at G residues -11, -10, -8 and -7 were weak cleavages, and cleavage at positions +25, +29, +31 and +42 were not detectable in sxy-1 RNA. However, although some cleavage sites in sxy-1 RNA were weaker or absent compared to wild type RNA, no novel positions of RNase T l cleavage were apparent. Therefore, although Mfold predicted that sxy-1 mRNA would fold into structure B of Figure 3.20, this structure was not supported by RNAse T l mapping. RNase CL3: RNase CL3 cleaves at unpaired C residues (64). Figure 3.20 and Figure 3.33 show that unpaired Cs in structure A occur at positions -44,-32, +30 and +49. Results for RNase CL3 cleavage of sxy RNA are summarized in Figure 3.33. Most of the C residues expected to be single-stranded were cleaved either moderately or strongly. Paired C residues in structure A occur at positions -41, -38, -34, -15, +19, +36, +38 and +39. Many of these bases were protected from cleavage; however, several were cleaved, either weakly, in the case of bases at positions -15, and + 36, or strongly, in the case of the base at position +19. Because the conditions used for RNAse CL3 cleavage resulted in weak cleavage at most C residues, it is difficult to detemine whether the weak cleavages at positions -15 and -36 are significant. However, the cleavage event at C +19 truly indicates an unpaired C residue at this position; the cleavage of this residue 140 does not support sxy RNA structure A, especially because the corresponding G at position +35 was not cleaved by RNAse T l . RNase CL3 structure mapping of sxy-1 RNA (Figure 3.34) gave results that were equivalent to those obtained with the wild type message, including a strong cleavage event at C +19. Physarum M RNase and RNase T2. RNase T2 cleaves at unpaired residues, with a preference for unpaired As (64). Physarum M RNase cleaves at unpaired A and U residues. These two enzymes gave almost identical cleavage patterns for sxy structure mapping, so they will be considered here as a single digestion pattern. Figure 3.20 and Figure 3.33 show the regions of sxy RNA structure A that are expected to contain unpaired A and U residues. Within Stem IA, most of the A and U residues predicted to be paired were not cleaved by ribonuclease treatment. However, some residues predicted to be paired were cleaved weakly. These included residues at the base of the stem, including + 54 and +57, as well as residues at the apex of this stem, including residues -31 through -25. These later results indicate that residues within the apex of Stem IA may be unpaired or weakly paired. Several unpaired A and U residues at positions -16 through -21 in sxy RNA were predicted to be unpaired. Many of these were cleaved weakly. Several residues within the unpaired region of Stem IB were cleaved, including those at positions -5, -6, -9 and -12. However, some residues predicted to be paired- i.e. those at positions 12, -13 and -14, were also cleaved, indicating that this region of sxy mRNA may be unpaired. Single-stranded regions in structure A at positions -1 through -4 were not cleaved by RNase T2 and Physarum M RNase, indicating that they may be paired within sxy RNA. RNAse T2 and Physarum M RNase also cleaved several unpaired residues within Stem IC. These included unpaired A's and U's at positions +24, +26, +28 and +32. Many of 141 the base pairs in Stem IC predicted to be paired were not cleaved by these enzymes. Unpaired U residues at positions +15, +21, and +33 were not cleaved by RNAse T2 and Physarum RNase. Again, this may be because these residues are sterically inaccessible to ribonuclease. RNase T2 and Physarum M mapping data obtained for sxy-1 RNA was similar to that obtained with sxy RNA. The majority of the cleavage sites were equivalent to those obtained with sxy RNA; some of these cleavage events appeared to be weaker, or absent, particulary for residues within Stem IC. Again, no novel cleavage sites for RNAse T2 or Physarum M were detected within sxy-1 RNA. Overall, RNase secondary structure mapping data provided evidence that confirmed base pairing of sxy structure A Stem IA. In addition, it provided evidence for other parts of sxy mRNA structure A that were not addressed by in vivo experiments. However, sxy structure A is probably not the true sxy RNA secondary structure, since some of the bases expected to be paired were cleaved by RNAse digestion and, conversely, some bases expected to be single-stranded were not cleaved. I utilized information obtained from structure mapping to re-analyse sxy mRNA structure using Mfold. This analysis generated two novel structures - sxy structure A. l and sxy structure A.2 - shown in Figure 3.35. Each of these structure fits the structure mapping data better than structure A; however, neither satisfies all of the structure mapping data. Sxy mRNA secondary structure should be further analyzed by the construction of additional point mutations, and also by further RNase and chemical mapping. .. The results presented here did not provide evidence that the secondary structure of sxy-2 mRNA is different from that of the wild type message. However, other evidence does suggest that the sxy-1 mutation acts by destabilizing structure A. Perhaps trans-142 acting factors or other in vivo components are necessary to destabilize sxy mRNA secondary structure in the hypercompetent mutants. 143 - 1 0 G A G G U IB u + 1 U G A a , U , > A A c u u A 2 " 0 u ' A ' " 2 ° A A+10 u ' A IC 5' U - A • U • G - 3 0 U - A El u U - A +50 ^ 9 A -40 U - A -50 C - G + 6 0 A G A A G U A C U U • G A A U . / / , G A A + 2 0 A - U U C ' / 3' G C G +30 sxy s t ructure A . 2 -10 ( G U U A A G U U A A U IB u U - A+1 A — 0 U • G • 2 ° U - A * IC A A A U U -30 U A+10 I I o u A G A . I u u +20z G A G C A A " A C u C G U U U G C G +30 -50 A G A A G U A C U ' E L u • u - A C c _ © + 5 ° A - U © * U | A © - A I A © - © A - u -40 U - A C - G n . ^+60 A A A U Figure 3.35. sxy RNA secondary structures A. l and A.2. These modified structures were obtained by incorporating RNase structure mapping data into Mfold analysis. The bases mutated in the sxy hypercompetence mutants are circled. The bases mutated by the sxy-7 mutation are boxed. 144 Chapter 4 : Fis mutants of H. influenzae have reduced spontaneous competence. 4.1. Background 4.1.1. Rationale for Tn916 mutagenesis and isolation of competence mutants Most genes involved in the process of natural transformation in H. influenzae have been isolated by mutagenesis of the KW20 genome followed by screening for transformation-deficient mutants. Several mutagenesis methods have been employed, including chemical mutagenesis (43), mutagenesis by transformation with heterologous DNA (31), transposon mutagenesis 0(83,180,181) and ligation-mediated cassette mutagenesis (58). Redfield described a unique approach for the selection of hypercompetent mutants, and used this approach to isolate gain-of-function mutations, rather than null alleles, of the sxy and murE genes (155,192) (Ma and Redfield, unpublished). H. influenzae loss-of-function mutations that result in hypercompetence have never been described. One of my research goals was to determine whether a repressor of competence development existed; I reasoned that null mutations in such a repressor would allow cells to become competent during early logarithmic growth and/or to become hypercompetent during late logarithmic growth. As a mutagenesis tool, I chose the conjugative transposon Tn926, originally identified on the chromosome of Enterococcus faecalis (69). Tn916 mutagenesis has been used for the isolation of mutants in H. influenzae and other bacterial species (72,82,88,148). Tn916 mutagenesis of KW20 failed to yield hypercompetence mutations. There are several possible reasons: because repressors of competence don't exist in H. influenzae, because repressors of competence are essential genes for which gene knockouts are lethal, or because of the insertion bias of Tn916 (see Discussion). However, as described 145 below, one class of mutant I isolated had a delayed spontaneous competence phenotype. The transposon insertion in this class of mutant was mapped within the H. influenzae fis operon, which is conserved between the Enterobacteriaceae and Pasteurellaceae (30). In E. coli, the Fis protein has been well characterized; in the following section, I discuss the known functions and regulation of Fis. 4.1.2. The evolution and function of Fis in £. coli and other enterics Fis (factor for inversion stimulation) was first recognized for its ability to promote site-specific recombination in the Hin, Gin and Cin recombination systems, and in the excision/integration system of phage X 0(85,92,94,179). Subsequently, several additional functions of Fis were identified. Along with the proteins IHF, H U and H-NS, Fis is one of the small abundant histone-like DNA binding proteins that contribute to the structure of the bacterial nucleoid (61,163). Fis also plays a role in the initiation of replication at OriC (62, 67, 73,194), and as a transcriptional activator or repressor of several genes and operons. Fis activates transcription of stable RNA operons and several metabolic and structural genes 0(12,21,47,76,141,159,195). Fis is also able to repress the transcription of a number of genes, including its own (24,47, 76,144,195). Fis binds as a homodimer to the degenerate sequence N N Y R N N T / A N N Y T N N G / C and induces DNA bending at enhancers of site-specific recombination, at OriC, and within promoters of genes and operons (90, 93,146, 179). The crystal structure of Fis shows that its 98 amino acids are organized into four a helices tightly intertwined to form a globular dimer with two protruding helix-turn-helix motifs (108,197). The N terminus of Fis, which is unresolved in the crystal structure, is essential for the activation of site-specific recombination (104,146,186). The mechanism of transcriptional activation by Fis, analogous to that of CRP, has been most extensively characterized for the stable RNA operons. Fis binds to conserved sites 146 in the upstream activating site (UASs) of several rRNA and tRNA operons and activates transcription 5-fold (35,36,80,81). Fis is thought to recruit RNA polymerase to the UASs of these operons through a combination of DNA bending and direct protein-protein interactions between Fis and the a subunit of RNA polymerase (36,135,136). In E. coli, the fis gene is in an operon with another gene of unknown function, orfl. This operon is conserved in the Enterobacteriaceae Salmonella typhi, Klebsiella pneumoniae, Serratia marcescens, Proteus vulgaris and Erwinia carotovora; and also in Haemophilus influenzae (30). The Fis amino acid sequence is highly conserved amongst the enteric bacteria (98-100% identity), and also between H. influenzae and the Enterobacteriaceae (80% identity). Beach and Osuna showed that the H. influenzae Fis protein could stimulate Hin-mediated DNA inversion and A, DNA excision (30). The conservation of Fis sequence between members of the Enterobacteriaceae and Pasteurellacaea, and the fact that H. influenzae Fis carries out some of the known functions of E. coli Fis, indicate that the functional role of Fis is conserved within these two families of bacteria. Fis is 49% identical to the carboxy terminus of the two-component nitrogen sensor protein NtrC, and Orfl is homologous to the NifR3 protein of unknown function from Azospirillum brasilense, Rhodobacter capsulatus and Rhizobium etli (134). Since nifR3 is cotranscribed with ntrC in these organisms, it has been suggested that the fis operon was acquired in the y proteobacteria by a single horizontal transfer event from the a proteobacteria prior to the divergence of the Enterobacteriaceae and Pasteurellaceae (134). In E. coli, the regulation of fis expression occurs primarily at the level of transcriptional initiation (24,144,147). Fis levels are dramatically increased - from less than 100 molecules per cell to 50,000 to 100,000 molecules per cell - within 75 to 90 minutes following transition of cells in stationary phase cells to exponential growth in rich media (24,143,144). Fis levels then decrease rapidly, so that less than 1% of peak concentrations of Fis remain as cells enter stationary phase (24). This pattern of fis 147 expression appears to be similar in other members of the Enterobacteriaceae (30,147). The fis promoter is regulated by the stringent response, by negative autoregulation, and by growth rate B(24,144,147,153,189). The growth rate regulation may be mediated by fluctuations in CTP pools (189). Although many functions are ascribed to Fis, its overall role is still unclear. Fis mutants grow slowly in rich medium and form filaments at elevated temperatures, presumably due to defects in DNA replication (67,142); however, despite its many diverse functions, Fis is not an essential protein (92,105). In addition, although Fis has been shown to transcriptionally activate many genes, it generally only does so in combination with other transcription factors, and in most cases Fis is not essential for the expression of these genes. For example, the transcription of stable RNA operons in fis cells is only reduced when the core promoters have been disabled by mutations or when RNA polymerase levels are low (143). On the other hand, Fis binding sites within the UAS of stable RNA promoters are conserved, indicating that Fis plays an advantageous role in their regulation (143). Several global functions of Fis have been proposed. One is that Fis allows cells to adjust quickly to conditions of rapid growth (142,184). Another is that Fis regulates supercoiling-dependent promoters and modulates the overall superhelicity of bacterial DNA (165,184). In the following section, I discuss the isolation and competence phenotype of fis mutants of H. influenzae. 148 4.2. Results 4.2.1. Tn916 mutagenesis and isolation of mutants with delayed spontaneous competence development Can transposon mutagenesis create H. influenzae null mutants that are hypercompetent? This was addressed by Tn926 mutagenesis of KW20, followed by attempts to select hypercompetent strains from amongst the pool of mutagenized cells. Tn916 was used to mutagenize the KW20 genome as previously described (101). The plasmid pAM120 (70) contains the large 18kb conjugative transposon Tn916 cloned into the pBR322-derived plasmid pGLl (Figure 4.1). Three independent 1 ml aliquots of MlV-competent KW20 cells were transformed with pAM120 (0.2 ug/ml; pools A, B and C). All tetracycline resistant transformants were expected to arise from conjugative transposition, since pBR322 cannot replicate in H. influenzae. I recovered transformants at a frequency of approximately 1 X 10~5, consistent with previously reported values (101). Approximately 10,000 pooled colonies from each independent transformation were pooled in 6 ml of sBHI and frozen in 1 ml aliquots for future use. No attempt was made to determine the randomness of the Tn916 insertion locations in the general pool of mutants. 149 1 T T T H S S a K J Figure 4.1. Restriction map of pAM120, showing position of the tetM gene (Gawron-Burke and Clewell, 1984). Thick line, pGLlOl DNA; thin line, Tn916 DNA; E, EcoRI; S, Seal; Sa, Sad and K; Kpnl. Adapted from Holland and Williams, 1992. Screens for hypercompetent mutants in early logarithmic growth were carried out as previously described (155). Two 1 ml aliquots of each pool of mutant cells (i.e. pools A, B and C) and KW20 (as a negative control) were thawed, centrifuged, and resuspended at an OD600 of 0.005 in sBHI (KW20) or sBHI containing tetracycline (mutant cells). Resuspended cultures were grown to an OD600 of 0.08-0.1, well before competence normally develops. At that time, 10 ml of each culture was removed and transformed with linearized pRRNovl (0.25 pg/ml). After 30 minutes at 37°C, DNasel was added, and cultures were incubated for an additional 10 minutes. Cells were then concentrated by centrifugation, and plated on sBHI containing novobiocin (KW20) or on sBHI containing novobiocin and tetracycline (mutants). Approximately equal numbers of Nov r transformants were obtained from KW20 and mutant cells. A total of 59 Tetr Nov r transformants was isolated from the original pools of Tetr cells. Each of these colonies was put through a second round of transformation to assay for hypercompetence. Colonies of KW20 (negative control) and RR699 (sxy-1) were assayed simultaneously. Individual colonies were inoculated into 2 ml sBHI, and grown to an approximate OD600 of 0.02-0.05. 1 ml of each culture was removed and transformed with DNA from RR520 (KanrNovsStrs; lOOOng/ml). After 30 minutes at 150 37°C, DNasel was added, and cultures were incubated for an additional 10 minutes. None of the colonies tested was found to be hypercompetent. Since none of the mutant strains was hypercompetent during early logarithmic growth in rich media, I then screened for mutants hypercompetent during late logarithmic growth. Because wild type cells are moderately competent at this stage of growth, I attempted to select for hypercompetent mutant strains by enriching for transformants at sub-saturating concentrations of DNA. The disadvantage of this approach is that it wouldn't necessarily enrich for regulatory mutants, and might instead result in the selection of mutants which are able to bind or take up DNA more efficiently, possibly in a nonspecific manner - for example, because of a cell wall alteration. 1 ml aliquots of Tn916-containing cells from pools A and B were thawed, centrifuged and resuspended in sBHI. Resuspended cells were grown to late logarithmic phase in sBHI (OD600 1-1.2), and transformed with linearized pRRNovl DNA (10 ng/ml). Independent pools of Tetr Nov1" colonies (4000-5000 colonies each) were pooled, grown to late logarithmic phase in sBHI, and transformed with sub-saturating concentrations of chromosomal DNA from RR514 (Strr Kan s Nov s; 1 ng/mL). Under these conditions, only a few Strr transformants were recovered for each of the two mutant pools; 52 Tetr Strr colonies were recovered from mutant pool A, and 82 colonies were recovered from mutant pool B. (No Strr colonies were recovered for KW20 transformed under similar conditions). Individual Tetr Strr colonies were then tested by colony transformation assays as described in Section 2.1.4.1. Colonies were transformed to kanamycin resistance with saturating amounts of chromosomal DNA from RR518 (Kan r Strs Nov s; 1000 ng/ml). Under these conditions, the majority of colonies tested were found to be 2 to 4-fold more competent than KW20. A typical result for such a colony assay is shown in Figure 4.2. 151 Figure 4.2. Results of colony transformation assays for KW20 (L7J), a KW20 pAM120 transformant that had not been selected with limiting DNA (H), and mutants from pools A and B ( i i ) which had been independently mutagenized and selected for by transformation with limiting DNA. Colony transformations were carried out as described in Section 2.1.3.1. Transformation frequencies are the mean frequencies of 4 individual colony transformations per strain. Error bars represent the range of the replicate data for 4 trials . 152 Seven of the mutant strains were also tested for spontaneous competence development in sBHI under saturating DNA conditions. Two classes of mutant phenotype were found for mutants derived from both pool A and pool B. One phenotypic class, represented by 2 of the isolates tested, had transformation frequencies that were approximately 1.5 to 2-fold higher than KW20 throughout all stages of growth in sBHI (data not shown). This class of mutant may have been isolated as a hypercompetent mutant because it transforms more efficiently than KW20 under conditions of limiting DNA (a hypothesis that has not been rigorously tested). However, because this type of mutation was unlikely to result from the disruption of a regulatory step of competence development, this class of mutant was not studied further. A second class of mutant phenotype was found in 5/7 of the isolates tested. Figure 4.3 shows a typical transformation frequency phenotype for this class of mutant phenotype. (Note: for this time course, large volumes of cells were transformed and concentrated in order to measure transformation frequencies during early logarithmic growth). This class of mutant strain exhibited a significant delay in the onset of spontaneous competence development, as shown in Figure 4.3. For time points prior to maximal competence, mutant B-l exhibited transformation frequencies that were reduced 10 to 100-fold compared to the wild type strain at an equivalent stage of growth. Interestingly, despite the delay in the onset of competence during exponential growth, this class of mutant achieved transformation frequencies in post-exponential growth that were are as high, or slightly higher than (by 1.5 to 2-fold), those of wild type cells. Four other independent isolates had an equivalent spontaneous competence phenotype to that of mutant B-l (data not shown). Mutant B-l exhibited no delay in the onset of MlV-induced competence (data not shown), indicating that the delay in competence is specific to cells in sBHI. Backcrossed strains of mutant B-l (i.e. KW20 transformed to Tet r with mutant chromosomal DNA) had a competence phenotype equivalent to that of mutant B-l, confirming that the transposon insertion in the 153 Time in sBHI Figure 4.3. Transformation frequencies (A) and growth (B) of KW20 (circles) and mutant B-l (squares) in sBHI. For early time points, large volumes (50 mL) of cells were transformed and concentrated for plating. Transformation frequencies were measured as described in Section 2.1.4.1. cfu/ml and transformation frequencies values are mean values of two replicate cultures. 154 original mutant strain was responsible for the observed mutant phenotype. This second class of mutant strain may also have been isolated because it transforms more efficiently than KW20 under limiting concentrations of DNA (again, this was not rigorously tested); however, the delay in the onset of competence development in the mutants was considered a potential regulatory phenomenon, so this class of mutant was studied further. 4.2.2. Mapping of the transposon hops: The Tn916 mutation causing delayed spontaneous competence is in the conserved fis operon. Do mutant strains with delayed hypercompetence have transposon insertions in the same genomic location? Which gene or region of the KW20 genome is mutated within these strains? These questions were addressed by Southern analysis of mutant genomic DNA, followed by cloning of the Tn926 insertion from one of the mutant strains. Southern analysis was carried out to determine if mutants with delayed spontaneous competence had transposon insertions in the same chromosomal location. Mutant chromosomal DNA was digested with EcoRl, which does not cut within the transposon, and with Kpnl and Seal, each of which has a single restriction site within the transposon (Figure 4.4.A). Digested chromosomal DNAs were run on a 6% agarose gel, and probed with a 1.8 kb, Digoxigenin-labeled, Kpnl-Hind\l\ fragment of pAM120 (Figure 4.4.A). pAM120 DNA was used as a positive control, and KW20 DNA was used as a negative control. Within the limits of resolution (i.e. within 10 % of the fragment length), all of the ten class 2 mutants had the same pattern, suggesting that they had transposon insertions within the same location. Figure 4.4.A shows a general restriction map deduced for the Tn916 insertions and the surrounding genomic DNA of class 2 mutants. 155 A . 15 20 tetM 30 J I ™ _ J _ -I 1 -K S Tn916 Probe B . tetM He He He He He He , i I_J i T K S pLBTnl Figure 4.4. (A). Restriction map of the Tn916 insertion in the chromosomes of mutants with delayed spontaneous competence, as determined by Southern analysis. The fragment of Tn916 used as a probe for Southerns and colony hybridizations is shown beneath the restriction map. (B) pLBTnl, containing an 18 kb subfragment of Tn916 and flanking mutant chromosmal DNA cloned into the unique Kpnl site of pSU20. I—I, KW20 DNA ;CZ3, Tn916 DNA; W, tetM; H i , pSU20 DNA. K, Kpnl; He,Hindi; H; Hindlll and S, Seal. 156 In order to map the precise location of the Tn926 insertion in mutant B-l, I cloned a 18 kb Kpnl chromosomal fragment from this mutant, which included most of Tn926 and the genomic DNA flanking the right side of the insertion (Figure 4.4.A). Mutant B-l chromosomal DNA was digested with Kpnl and run on a 6% agarose gel. Digested fragments of ~18 kb were size selected, excised from the gel, and purified as described in Section 2.1.5.3. Purified DNA was cloned into the unique Kpnl site of pSU20. Blue/white-screening was used to identify plasmids containing inserts, and white colonies were patched onto LB Cm plates and screened for the desired insertion by colony hybridizations, as described in Section 2.3.1. Three of 200 colonies tested had plasmids containing the desired insert, as confirmed by restriction digestion with Hindi and Kpnl. One of these plasmids, designated pLBTnl (figure 4.4.B), was used in subsequent studies. The insertion site of the Tn926 insertion in pLBTnl was determined by sequence analysis using the primer TnRl to sequence the genomic DNA flanking the transposon insertion (Figure 4.5). The resulting sequence was used to search the TIGR H. influenzae database (2), and located the Tn916 insertion of mutant B-l within orfl of them's operon. The insertion disrupts the carboxy terminus of the orfl gene product, 16 nucleotides upstream of the 5' end of the fis gene (Figure 4.5). The fis operon is conserved between H. influenzae and the Enterobacteriaceae (see Section 4.1.2). Orfl (HI#0979) and fis (HI#0980) of H. influenzae overlap, as they do in S. marcescens (30). In E. coli, there is no known function of the orfl protein, and mutations in this gene are polar on the expression of fis, since the two genes are transcribed from a common promoter (24, 144). Therefore, the Tn916 insertion might affect competence by lowering or eliminating fis expression. This was confirmed by the experiments discussed in the following section. 157 158 4.2.3. Mutations in fis cause a 500-fold reduction in transformation frequencies during spontaneous competence development. Does the Tn916 insertion in mutant B-l lower competence development by reducing or eliminating fis expression? This was addressed by cloning and mutating the orfl and fis genes, and examining spontaneous competence of the mutant strains. I used PCR to amplify a 2 kb KW20 genomic fragment containing the orfl and fis genes (described in Section 2.3.2). The PCR product was digested with Clal + Bglll (Figure 4.5) and cloned into the unique Clal and BamHI sites of pSU21 to create the plasmid pLBFl. I constructed fis and orfl null alleles by cloning a Sm r /Sp r cassette from the plasmid pKRP13 into the unique SnaBI and EcoRI sites, respectively, of pLBFl, to create the plasmids pLBF2 {fisv.spec) and pLBF3 (orfl::spec) (Figure 4.5). In pLBF2, the cassette in fis, which disrupts the a D helix, should eliminate DNA binding by the Fis protein. In pLBF3, the cassette in orfl, which is in the middle of the gene, should interfere with the normal function of Orfl. pLBF2 and pLBF3 were linearized and used to transform MlV-competent KW20 to Spec1". Figure 4.6 shows that the spontaneous competence phenotype of RR859 (orflv.spec) was similar to that of mutant B-l, in that RR859 transformation frequencies were several-fold lower than those of wild type cells during exponential growth, but maximal transformation frequencies were similar. The delay in competence in RR859 was not as severe as that of mutant B-l; the reason for this is unknown, but may have to do with the position of the respective insertions within orfl in these two strains. 159 Figure 4.6. Transformation frequencies (A) and growth (B) of KW20 (circles) and RR859 (orflv.spec; triangles) in sBHI. Transformation frequencies were measured as described in Section 2.1.4.1. Transfomation frequency and cfu/ml values were obtained fom measurements on single cultures. The growth rate of KW20 and RR858 (fis::spec) are compared in Figure 4.7. RR858 exhibited a much slower doubling time (45-50 minutes compared to 30 minutes for KW20). This decrease in growth rate was similar to the growth rate reduction observed in E. coli fis strains (142). Figure 4.8 shows spontaneous competence development of KW20 and RR858 in sBHI. The maximal transformation frequencies of RR858 were 500 to 1000 times lower than those observed for wild type cells. MIV transformation frequencies of RR859 and RR858 compared to KW20 are shown in Figures 4.9 and 4.10, respectively. Strains carrying the orfl null allele (RR859) had transformation frequencies that were approximately 10-fold lower than those of wild type cells (Figure 4.9). However, cell densities were also much lower in this strain, and may have limited the development of competence. Strains carrying the fis null allele (RR858) had transformation frequencies that were 2 to 5-fold lower than those of wild type cells. Because strains carrying the orfl null allele had a similar competence phenotype to that of mutant B-l, this provides additional evidence that the mutant phenotype of the Tn926 mutants results from a polar effect on fis expression. In E. coli, a cassette mutation within orfl was shown to be polar on fis expression (24); by analogy, transposon and cassette mutations within orfl of H. influenzae presumably have a moderate affect on competence development by reducing, but not eliminating, fis expression. However, since I haven't carried out any complementation analyses, I cannot rule out the possibility that the competence phenotypes of mutant B-l and RR859 could result from a lack of orfl gene product. The observation that RR858 has a 500 to 1000-fold reduction in transformation frequencies in sBHI implies that Fis plays an important role in competence development. 161 Figure 4.7. Growth (OD 6 0 0) of KW20 (circles) and RR858 {fisv.spec, triangles) in sBHI. Time in sBHI (min) Figure 4.8. Transformation frequencies (A) and growth (B) of KW20 (circles) and RR858 (fisr.spec; triangles) in sBHI. Transformation frequencies were measured as described in Section 2.1.4.1. For RR858, cfu/ml and transformation frequency values are mean values of two replicate cultures. 163 IO 8 — i 1 1 1 1 1 -m »/•> >/~> m i n m - h i n 0\ co r- CN Time in MIV (min) Figure 4.9. Transformation frequencies (A) and growth (B) of KW20 (circles) and RR859 (orflv.spec, triangles) in MIV. Transformation frequencies were measured as described in Section 2.1.4.1. For RR859, cfu/ml and transformation frequency values are mean values of two replicate cultures. i o 8 — i 1 1 1 1 1—1 ITt V~i IT) I/-) —i >/"> Q\ co t-- CS Time in MIV (min) Figure 4.10. Transformation frequencies (A) and growth (B) of KW20 (circles) and RR858 (fis::spec, triangles) in MIV Transformation frequencies were measured as described in Section 2.1.4.1. For RR858, cfu/ml and transformation frequency values are mean values of two replicate cultures. 165 4.2.4. Epistasis between fis and the competence regulatory genes cya and sxy I next examined whether Fis plays a regulatory role in activating competence development by testing possible epistatic relationships between fis and competence regulatory genes. I determined the effects cAMP addition on the competence of fis cells, and also examined the spontaneous competence phenotype of a sxy-1 fis::spec double mutant. The effect of addition of cAMP to RR858 cells growing in sBHI is shown in Figure 4.11. cAMP addition increased RR858 transformation frequencies only 5-fold during early logarithmic growth (wild type transformation frequencies were elevated 10,000-fold upon cAMP addition) and about 100-fold during late logarithmic growth. These results imply that the reduction in transformation frequencies exhibited by fis cells in sBHI was not due to the inability of cells to produce a sufficient amount of cAMP. This finding was confirmed by the observation that sugar fermentation by mutant B-l was equivalent to that of wild type cells (data not presented). Therefore, if Fis regulates competence, it does so at a step that is independent of regulation by cAMP/CRP. The spontaneous competence phenotype of RR873 (sxy-lfis::spec) is shown in Figure 4.12. Compare Figure 4.12 to Figure 3.4, which shows the spontaneous competence phenotypes of KW20 and strains carrying the sxy hypercompetence mutations. The presence of the sxy-1 mutation was able to raise transformation frequencies of fis cells by 30-fold; however, transformation frequncies of RR873 remained about 10-fold lower than those of wild type cells, and were approximately 500-fold lower than those normally exhibited by RR699 (sxy-1). These results indicate that the competence defect in the fis strain occurs downstream of, or is independent of, the effects of cAMP and Sxy on the regulation of competence, since spontaneous competence of fis cells is not fully restored to wild type levels by the 166 addition of cAMP, nor by the presence of excess amounts of Sxy. This issue is considered further in the Discussion. 167 10-03 Figure 4.11. Effect of ImM cAMP on transformation frequencies of KW20 and RR858 (fisv.spec) in sBHI. cAMP was. added to cells at approximately OD 6 0o 0.01. Transformation frequencies were measured as described in Section 2.1.3.1. For fis cells, transformation frequncies are the mean values of two r eplicate cultures, and error bars represent the range of the data. 'Early' refers to cells sampled in early logarithmic (OD60y 0.15 - 0.2) growth;' late' refers to cells sampled in late logarithmic (OD6000.7 -1.0) growth. '-'; no cAMP added;'+', cAMP added. T 1 1 1 8 8 8 8 CM C"> Time in sBHI (min) Figure 4.12. Transformation frequencies (A) and growth (B) of RR873 (fisv.spec; circles) and RR866 {sxy-1 fisv.spec, triangles). Transformation frequencies were measured as described in section 2.1.4.1. cfu/ml and transformation frequency values are mean values of two replicate cultures. 169 Chapter 5 : Discussion 5.1 Discussion of sxy results In Chapter 3 of this thesis, I described experiments investigating a simple model for competence development in H. influenzae. Sxy is suggested to function specifically in competence development by activating the transcription of competence genes. This model hypothesizes that levels of competence in H. influenzae are positively correlated with expression of sxy. Cells in MIV are hypothesized to be more competent than those in sBHI because sxy expression is higher in MIV than in sBHI. It was further suggested that maximal expression of sxy in sBHI was not achieved because the 5' sequence of sxy mRNA folds into an RNA secondary structure that negatively regulates sxy expression in sBHI, and that this inhibitory structure is precluded from forming in MIV. 5.1.1 Sxy function and evolution What is the function of Sxy? Sxy was suggested to be an activator of competence development because gene knockouts prevent competence, because multiple copies of sxy and point mutations in sxy cause elevated competence, and because some competence genes require Sxy for transcriptional activation (Section 3.1). More specifically, it was suggested that Sxy may be a transcription factor that binds at a conserved regulatory sequence (the CRE element) preceeding the promoters of several competence genes (200). However, the CRE site resembles the binding site for the CRP complex, and it has been suggested that the CRP/cAMP complex binds to and and activates transcription at promoters containing CRE sites (120). To address whether Sxy functions specifically in competence development, and to further address the precise role of Sxy in competence development, I searched for Sxy homologs amongst other eubacteria. Potential homologs were found in both 170 transformable and nontransformable members of the y division of the Proteobacteria, and within nontransformable members of the Enterobacteriacae, Pasteurellaceae and Vibrionaceae. The identification of Sxy homologs in nontransformable bacteria implies that Sxy is likely to have a cellular function other than competence development. Although some of the identified proteins share only weak overall similarity with Sxy (Blast E values ranged from 2e~5 to 3e"32; Table 3.1), these proteins are probably true Sxy homologs for several reasons. The first is that a minimum E value of 0.01 has been suggested as a cutoff score for the identification of true homologs El(91,141). In addition, the proteins are of similar size (ranging from 195 to 217 amino acids), and the shared amino acid similarity with Sxy extends over 60% of the proteins. The issue of whether the Sxy-like proteins identified in other bacteria are true Sxy homologs could be addressed further by cloning DNA encoding one or more of the Sxy-like proteins and determining whether they can complement sxy- cells of H. influenzae. Unfortunately, none of the Sxy homologs has an assigned function, and neither they nor Sxy itself contain identifiable motifs that allow for functional identification. However, I have identified protein motifs that are conserved between Sxy and its homologs; these motifs are likely to encode the functional domains of this family of proteins (Figures 3.6 and 3.7). My results do not confirm that Sxy or any of its homologs act as transcription factors. While some of the protein sequences contain amino acid sequences that are similar to helix-turn-helix motifs of known transcription factors, the similarity is not significant. The issue of whether Sxy can bind DNA alone or in combination with cAMP/CRP should be addressed by developing a protocol for Sxy purification and subsequent analysis of CRE binding via DNA bandshifts or Southwestern blotting experiments. 171 5.1.2. The correlation of sxy expression with competence development It is well established that an increase in the concentration of intracellular cAMP induces competence development. However, the maximal induction of competence that occurs when cells are transferred to starvation medium requires at least one other cAMP-independent signal. It has been proposed that a starvation-induced rise in the concentration of Sxy is sufficient to cause maximal competence in cells with elevated cAMP levels (Redfield, unpublished). This model predicts that Sxy is a key regulator of competence development, and that variations in the levels of natural transformation should be positively correlated with variations in the levels of Sxy within cells (Section 3.1.4.). If competence is limited by the availability of a specific activator, such as Sxy, what gene expression pattern do we expect in relation to competence development? Part of the answer depends upon how long it takes cells to develop genetic competence and express the cellular machinery required for transformation. For example, when 1 mM cAMP is added to cells in early logarithmic growth in sBHI, transformation frequencies rise and plateau in about 45 minutes, presumably because cells need this amount of time to synthesize and assemble the cellular machinery required for DNA uptake, translocation and/or recombination. In sBHI without added cAMP, transformation frequencies begin to rise when cells are at a density of approximately 5 X 108 cfu/ml, and become maximal when cells have reached a density of approximately 5 x 109 cfu/ml, requiring about 3 generations, or 90 minutes. The reason maximal transformation frequencies take this long to occur under these conditions is unclear, but may be because competence-inducing factors, such as Sxy, or cAMP, accumulate 172 gradually within cells. Cells transferred to MIV also require 90 to 100 minutes to develop maximal competence, probably for the same reason stated above. Considering the information above, we would expect the expression of a gene encoding an activator of competence to be maximal sometime prior to the time at which maximal transformation frequencies are observed within the population, and perhaps as early as transformation frequencies begin to rise. (Note: another complicating issue is that a base line of transformation has not been established. We generally assume that competence development does not occur until we observe the first transformants amongst a population of cells; however, the sensitivity of our standard transformation assay is such that transformation frequencies below about 10~9 are not measurable without the transformation and plating of very large volumes of cells). I used sxyv.lacZ gene fusions to indirectly measure the concentration of Sxy within cells at times when transformation frequencies are undetectable (< 10~8), during early logarithmic growth in sBHI; when transformation frequencies are moderate (10~8 to 10"4), during mid-logarithmic growth to stationary phase in sBHI, and when transformation frequencies are maximal (10"2), during transfer of cells to MIV. These results are summarized in Table 5.1. 5.1.2.1. The correlation of sxy expression and competence development in sBHI If we consider lacZ expression from sxyw.dacZ protein fusions, these data imply that Sxy concentrations increase in cells as they proceed from logarithmic growth in sBHI to post-logarithmic growth - LacZ activity from about 2 units of P-galactosidase activity at the earliest stages of growth measured, to almost 500 units for cells at the 173 o H • GO. O) o c a> o-a> .fc C O V3 CO E o (0 13 E "§ 2 X o x o Op b o o V o V I-03 •g <o o CO g 'co CD C a CD o o in 6 m co o 6 in o in 8 in 1-o m 6 o o o 6 o CO o in 6 in CO •3 .5? 8 i o in co 8 CO $ o o in in 6 8 o o 6 o CO o in w, 6 8 •a G> T— T-~ CO i 55 55 55 Q z o co Q z o o CO Q O 174 end of logarithmic growth (Figure 3.15 and Table 5.1). The observed increase in lacZ expression occurs in two distinct stages: LacZ activity is increased initially to about 200 units by the time cells are in mid-logarithmic growth in sBHI, remains at this level for several generations, and increases approximately 1.5 to 2-fold as cells reached the late stages of logarithmic growth. This apparent increase in Sxy concentrations within cells at the end of logarithmic growth compared to cells during early logarithmic growth in sBHI is consistent with the predictions of the model. However, it is surprising that lacZ expression from sxydacZ protein fusions is induced to moderately high levels many generations before cells achieve maximum transformation frequencies in sBHI. p-galactosidase production from sxy protein fusions are moderately high (-200 units) by the time cells reach a density of 5 X 107 cfu/ml in sBHI, even though transformation does not become detectable until several generations later, and maximal transformation frequencies in sBHI do not arise until cells reach a density of ~5 X 109 cfu/ml (see Figures 3.2 and 3.3). Overall, these data imply that there is not a simple, linear relationship between Sxy concentration and the level of competence achieved by cells within sBHI. How can we reconcile the observation that Sxy is an activator of competence but appears to be present at high concentrations well before cells achieve maximal transformation frequencies within sBHI? Why don't cells develop maximal competence at an earlier stage of growth? The most likely explanation is that the simple model for competence development is incorrect: although Sxy is absolutely required for competence development, Sxy is not limiting for competence development throughout most stages of growth in sBHI. For maximal competence development in sBHI, Sxy concentrations must reach a certain threshold level, but competence development is ultimately limited by another cellular factor, such as cAMP. Another possibility that cannot be presently ruled out is that Sxy activity is increased during exponential growth by a posttranslational modification such as phosphorylation. 175 In this modified version of the model for competence development in H. influenzae, the observed ~2-fold increase in Sxy concentrations observed near the end of logarithmic growth in sBHI may or may not play a functional role in increasing competence levels. I suggest that this moderate increase in Sxy concentrations does play a role in increasing competence levels, and, furthermore, that sxy mRNA secondary structure partially limits sxy expression during logarithmic growth in sBHI by inhibiting translational initiation at the sxy RBS (see below). However, although Sxy may be partially limiting for competence development until cells reach the later stages of exponential growth, it is clearly not the sole factor limiting competence development, as was suggested in the simple model for competence development. If, as I suggest, Sxy concentrations are not normally limiting for competence development in wild type cells growing in sBHI, how can we account for the observation that conditions which increase sxy expression, through sxy mutation or overexpression, can increase transformation frequencies above those normally observed for wild type cells in sBHI? The answer to this question may depend on the function of Sxy within wild type cells. The observation that moderate increases in sxy expression - for example, the 1.5 to 2-fold increase in sxy expression observed in the sxy hypercompetent mutants - can dramatically, i.e. 100-fold or more, increase competence is surprising. Because several competence genes contain CRE sites and are believed to be regulated in a coordinated manner, small changes in a transcription factor such as Sxy could have a large overall affect on the levels of competence. Alternatively, Sxy may not directly activate transcription of competence genes, since there is not a direct correlation between sxy expression and competence levels. In addition, hypercompetent mutants that carry mutations in the murE gene do not have increased sxy expression, implying that proteins other than Sxy may be involved in competence activation (Ma and Redfield, unpublished). It is possible that Sxy either regulates the activity or expression of a second gene product that activates transcription of 176 competence genes, or that Sxy is required to reduce the activity or expression of a gene product that represses transcription of competence genes. In either of these cases, cAMP/CRP is normally required to fully activate the transcription of competence genes, and maximal gene expression therefore does not occur until the end of logarithmic growth in sBHI, when cAMP concentrations increase. However, conditions that increase Sxy expression above that normally found in wild type cells increase competence because of the presence of more activator or less repressor than normally occurs within cells grown in sBHI. This model is discussed further in Section 5.1.6. 5.1.2.2. Correlation of sxy expression and competence development in MIV The simple model for competence development states that MIV transformation frequencies are higher than those in sBHI because Sxy concentrations are higher in MIV. Are Sxy concentrations increased as cells are transferred to MIV? Again, if we consider lacZ expression from sxy^vXacZ protein fusions, these data imply that Sxy concentrations are not higher in MIV than they are in stationary phase in sBHI (maximum P-galactosidase activities in both cases are approximately 400 to 500 units; Table 5.1 and Figure 3.15). The simplest explanation of these data is that an increase in sxy expression is not required for the increased levels of competence of MIV cells relative to those in the stationary phase in sBHI, therefore, suggesting that increased concentrations of neither Sxy nor cAMP (since the addition of ImM cAMP doesn't increase MIV competence) is responsible for the increased competence of cells in MIV compared to those in sBHI. Therefore, although both Sxy and cAMP are absolutely required for competence devlopment in MIV, the presence or absence of additional factors is/are required for competence development in MIV. This additional factor may be differentially expressed in MIV than in sBHI. The hypothesis that Sxy concentrations are not increased in MIV relative to sBHI is difficult to reconcile with the observation that 177 increasing sxy expression in cells in sBHI allows them to achieve competence levels normally observed only for cells in MIV. More information about the regulation of competence development in both types of media is required to address this issue. Another, less likely, possibility is also suggested by these data. If we examine lacZ expression from sxyw.dacZ protein fusions, we see that lacZ expression is induced much more rapidly upon transfer of cells to MIV compared to cells growing in sBHI. The increase in lacZ expression from sxy protein fusions - from 200 units to between 400 and 500 units of (3-galactosidase activity - occurs within 45 minutes for cells transferred to MIV. In contrast, the same increase in (3-galactosidase activity takes 120 minutes, about 2.5-fold longer, for cells in sBHI (Figure 3.15). An equivalent result for cAMP was reported by Macfadyen, who indirectly measured cAMP concentrations by the analysis of cAMP-dependent reporter strains (120). She found that cAMP-dependent LacZ reporter activity increased rapidly for cells transferred to MIV, but that the final activities of the reporter strains were not higher in MIV than in sBHI. These data suggest that the timing of increase in the concentrations of Sxy and cAMP may be important in determining the level of competence that is achieved by cells. A further extension of this hypothesis is that signals for competence development in sBHI and in MIV - for example, the depletion of a particular metabolite - are not different between the two media, but that the signals are stronger in MIV. For example, a decrease in the concentration of a metabolite may occur gradually over time in sBHI as the cellular density increases; however, the same metabolite may be drastically reduced or absent in MIV. Alternatively, a repressor of competence may be present at high concentrations in sBHI but may be absent or present in lower concentrations in MIV. 178 5.1.3. Sxy expression is mediated by transcriptional and posttranscriptional regulation. Transcriptional and translational regulation of sxy was investigated by comparing lacZ expression from sxyv.lacZ operon and protein fusions with fusions at the same positions (i.e. within sxy codon 89). These results are also summarized in Table 5.1. As cells proceed from early logarithmic growth to mid-logarithmic growth, lacZ expression from sxyv.lacZ operon and protein fusions increases to approximately 200 units of P-galactosidase activity (Table 5.1, Figure 3.15). Provided lacZ translation in logarithmic growth was equivalently low for both types of fusion, this implies that the increase in sxy expression at this stage of growth is due to an increase in available sxy mRNA, either because of transcriptional activation or because sxy mRNA stability or elongation is increased. Zulty and Barcak reported that sxy expression is decreased 3 to 4 -fold in a crp background (200), however, my results showed only a slight decrease (< 2-fold) of lacZ expression from sxyv.lacZ fusions in a cya background (Figure 3.12). In fact, lacZ expression from sxy operon fusions appears to be slightly decreased in the presence of 1 mM cAMP (Figure 3.13). The regulatory mechanism that acts at the regulation of sxy transcription and/or mRNA stability remains unidentified, but does not seem to occur in response to nutrient deprivation, since sxy mRNA levels appear to be high well before the end of logarithmic growth In sBHI, at about the same time that cells begin to develop maximum competence, the level of LacZ produced from protein fusions increases relative to lacZ produced from operon fusions (Figure 3.15, Table 5.1). Cells that are transferred to MIV also have increased LacZ production from protein fusions versus operon fusions (Figure 3.15). 179 These observations suggest that translational efficiency of sxy may increase both in post-logarithmic growth in sBHI and upon cell starvation in MIV. Although this evidence is not very convincing on its own, it is strengthened by other evidence, discussed in the next section, suggesting that sxy translation is inhibited by the formation of sxy RNA structure A. l or A.2, which may inhibit translational initiation at the sxy RBS. The simple model for competence development stated that sxy expression would be limited by sxy RNA secondary structure in sBHI at all growth stages. However, results presented here suggest that sxy RNA secondary structure may limit sxy translation only during logarithmic growth of cells in sBHI. 5.1.4. The 5' region of sxy mRNA folds into a secondary structure that negatively regulates sxy translation. The folding of sxy mRNA into the proposed RNA secondary structure was supported by in vitro RNase mapping of sxy RNA secondary structure as well as several lines of in vivo evidence. Results of sxy RNA structure mapping are summarized in Figure 3.33. Nuclease-sensitive sites for RNAse T l , RNAse T2, RNase CL3 and RNase Physarum M correlated well with the residues of sxy structure A in stem IA that were expected to be single-stranded based upon Mfold analysis. Structure mapping data was used in Mfold analysis to generate modified versions of sxy structure A, structures A. l and A.2 (Figure 3.35). Several independent lines of in vivo evidence confirm that sxy RNA secondary structure negatively regulates sxy expression. One piece of evidence came from the analysis of P-galactosidase production from cells carrying sxy\\::lacZ operon and protein fusions in which Stem IA of structure A is eliminated (Figures 3.18 and 3.19, Table 5.1). P-galactosidase production from sxy\y.:lacZ protein fusions was increased 5 to 7-fold relative to sxy^v.lacZ protein fusions, and P-galactosidase production from sxy\\::lacZ 180 operon fusions was increased 2-fold relative to sxyggdacZ operon fusions, in cells grown in sBHI. These results suggested that the major effect of eliminating the 3' half of Stem IA was to increase sxy translatability and disrupt the negative regulation of translation seen during logarithmic growth in wild type cells. However, the possibility remains that operon and protein fusions fused to codon 11 of sxy have increased expression because of an affect unrelated to RNA secondary structure. Gene fusions that disrupt Stem IA at other positions within sxy should be examined. In addition, sxy transcription and translation and transformation could be measured in a strain in which sxy is expressed from an inducible promoter placed at different positions within sxy relative to the location of the proposed inhibitory RNA secondary structure. A second piece of in vivo evidence comes from studies of the sxy-1 and sxy-3 mutations. These mutations are thought to cause hypercompetence by destabilizing sxy RNA secondary structure, thereby increasing sxy expression. lacZ fusion studies confirmed this prediction for the sxy-1 mutation. P-galactosidase production from sxy-1 operon fusions is approximately 1.5 to 2-fold higher than from sxy operon fusions (Figure 3.21 and Table 5.1). Preliminary Northern analysis of sxy message levels in wild type, sxy-1 and sxy-3 cells confirmed that sxy-1 and sxy-3 cells have increased amounts of steady state sxy mRNA relative to wild type cells (Figure 3.30). P-galactosidase production from sxy-1 protein fusions is 3-fold higher than from sxy protein fusions, for cells in sBHI at the end of logarithmic growth (Figure 3.22 and Table 5.1). Therefore, the sxy-1 mutation appears to increase sxy translation more than it increases sxy mRNA availability. However, sxy-1 protein fusion analysis indicates that this mutation only partially disrupts the negative control of sxy translation by sxy RNA secondary structure, because the sxyll fusion had a much larger effect on lacZ expression from protein fusions. The observed changes in the amounts of sxy mRNA and protein caused by the sxy hypercompetence mutations are small. However, it appears that conditions which only 181 moderately increase Sxy concentrations in sBHI dramatically increase cellular transformation frequencies, as previously discussed (Section 5.1.2.1). The indirect measurements of Sxy protein within sxy hypercompetent mutants and wild type cells should be independently confirmed by quantitation with an anti-Sxy antibody. Despite the Mfold prediction that sxy-1 RNA should fold into RNA structure B of Figure 3.20, RNase secondary structure mapping of sxy-1 RNA yielded results that are not significantly different than those obtained with the wild type message. These results don't preclude the possibility that sxy-1 folds into an RNA secondary structure other than secondary structure A. l or A.2 in vivo. For example, there could be some effect of polarity of transcription on message folding. In addition, it is possible that a trans-acting factor may be required to destabilize sxy RNA secondary structure, in vivo. It is also possible that the sxy-1 message does fold into sxy structure A . l in vivo, but the sxy-1 mutation destabilizes this structure relative to that of the wild type structure, without promoting the formation of a different RNA secondary structure. This is supported by the observation that negative regulation of sxy translation appeared to be only partially disrupted in sxy-1 protein fusions. The site-directed mutation sxy-7 introduces two additional stabilizing GC base pairs into the wild type sxy RNA secondary structure. I expected that this mutation would severely decrease sxy translational initiation. The sxy-7 mutation eliminates competence development (Figure 3.28), and reduces the amount of lacZ expression from sxy-7 operon fusions by approximately 3 to 4-fold (Figure 3.29, Table 5.1). Preliminary Northern analysis of sxy mRNA in the sxy-7 mutant showed that the sxy transcript is severely reduced or absent. The effect of the sxy-7 mutation on translation has not been measured yet. It seems likely that the primary effect of this mutation is a reduction in sxy transcriptional initiation, mRNA stability, or mRNA elongation rates. The sxy-7 mutation, located approximately 30 nucleotides downstream from the sxy transcriptional start site, is unlikely to influence transcriptional initiation from this 182 promoter. This mutation may reduce transcription from another sxy promoter, or it may decrease mRNA stability, or cause pausing or transcriptional termination by RNA polymerase. It is also possible that this mutation decreases translational initiation at the sxy RBS, and that the reduction in sxy mRNA in the sxy-7 mutant is an indirect effect of concomitant decreased sxy mRNA stability. It is crucial to determine whether the effect of the sxy-7 mutation on sxy mRNA levels is dependent on mRNA secondary structure A. l or A.2, or if the mutation affects sxy mRNA availability in a manner that is independent of secondary structure. For example, if the phenotype of the sxy-7 mutant depends on an intact sxy RNA secondary structure, gene expression from sxy-7\\::lacZ fusions should not be reduced relative to wild type fusions. Similarly, sxy-7 RNA would be expected to have altered RNA secondary structure, which could be tested by RNase structure mapping. The sxy-6 mutation (sxy-1 plus sxy-3) was constructed as a potential compensatory mutation. Strains carrying sxy-6 were no longer hypercompetent (data not presented), providing additional evidence that the hypercompetence mutations exert their effect by destabilizing the wild type secondary structure. Northern analysis of sxy mRNA levels from RR51 (sxy-6) confirmed that the amounts of sxy message were equivalent in sxy-6 and wild type cells (data not presented). However, lacZ expression from sxy-6::lacZ operon and protein fusions was approximately 1.5-fold higher than that from wild type cells. The reason for the discrepancy between f3-galactosidase measurements of strains RR856 and RR857 (sxy-6^::lacZ operon and protein fusions, respectively) versus sxy mRNA measurements and transformation of RR851 (sxy-6) is not known. The DNA sequences of RR856 and RR857 should be checked to rule out the presence of undesired mutations and to confirm the presence of both the sxy-1 and sxy-3 mutations in the genomic DNA of these strains (The sequence of RR851 has already been confirmed). In addition, one or two further compensatory mutations should be created to confirm the 183 sxy-6 competence phenotype. For example, a G+50 to A+50/C-34 to A-34 double mutation could be constructed. 5.1.5. Regulation of gene expression by RNA secondary structure: a review Above, I discussed several independent lines of in vivo and in vitro evidence suggesting that the 5' region of sxy mRNA can fold into an RNA secondary structure that regulates sxy translation. Below, I discuss well-established examples of how RNA secondary structure regulates gene expression. Most examples involve one of the following mechanisms: attenuation of transcription, regulation of mRNA stability, or regulation of translational initiation. I suggest a mechanism by which the formation of the sxy mRNA secondary structure controls translation by limiting ribosomal access to the sxy RBS, and that the efficiency of sxy mRNA translation indirectly affects message stability. 5.1.5.1. Sxy expression is probably not regulated by an attenuation mechanism. Many genes are regulated by the mechanism of transcriptional attenuation (111), which limits gene expression via the premature termination of transcription, usually in response to a particular metabolic signal. Transcriptional attenuation can be divided into four distinct classes (summarized in Table 5.2), each with its own representative genes and regulatory features. For the first three classes of attenuation, the regulation of transcription termination is achieved by the formation of competing RNA secondary structures, of which one structure serves as a rho-independent transcriptional terminator. Formation of the terminating RNA secondary structure depends on either the coupling of transcription with translation, or on the presence of a frans-acting RNA binding factor. Each of these mechanisms is intricate and will not discussed in further detail here [See reference (111) for a detailed discussion]. 184 Because mutations that were predicted to affect the stability of sxy RNA secondary structure affected p-galactosidase production from sxyv.lacZ operon fusions, but were not located in a region of sxy DNA expected to affect transcriptional activity at the sxy promoter, it seemed possible that structure A.l or A.2 might negatively regulate sxy expression by serving as transcriptional terminator or RNA polymerase pause site that limited sxy expression during logarithmic growth. However, I suggest that this type of mechanism is unlikely to affect sxy expression for the following reasons. Operons controlled by transcriptional attenuation contain long leader RNAs which encode small peptides (generally 30-40 amino acids) whose translation is required for the modulation of transcriptional attenuation. The sxy message does not appear to encode a leader peptide. In addition, sxy RNA structures A. l and A.2 lack the characteristic features of hairpin terminators, which usually consist of a small stable hairpin structure followed by a string of U residues (111). In the fourth class of transcriptional attenuation, the regulation of transcriptional termination occurs at a rho-dependent attenuator. Although no consensus sequence has been determined for rho termination, it is generally accepted that rho factor binds to unstructured mRNAs (158). Therefore, neither structure A. l nor A.2 is likely to serve as a site for rho-dependent transcriptional termination. However, H. influenzae does contain rho factor (HI#0295), and the possibility that sxy mRNA secondary structure A. l or A.2 contain a rho-dependent transcriptional termination site cannot be eliminated. The drastic reduction of sxy mRNA in the sxy-7 mutant is the only piece of evidence that supports the possibility that sxy structure A. l or A.2 may negatively regulate gene expression by a transcriptional attenuation mechanism. The sxy-7 mutation may act to increase the stability of structure A. l or A.2, thereby increasing the strength of transcriptional pausing or termination on sxy mRNA. 185 Table 5.2. Mechanisms of transcriptional attenuation Mechanism Operon Ribosome stalling histidine and tryptophan biosynthetic operons of E. coli Ribosome-coupling-dependent attenuation pyrimidine biosynthesis operon of E. coli Regulatory factor-dependent attenuation B-glucoside utilization operon of B. subtilis Rho-dependent attenuation tryptophan utilization operon of B. subtilis 186 5.1.5.2. Sxy RNA secondary structure is probably not a cleavage determinant for endonucleolytic mRNA decay. Many reviews have been published on the subject of the decay of bacterial messenger RNA El(16,49,110,139,149). Since I am considering regulation of sxy expression by a 5' structural determinant, I will focus on endonucleolytic cleavage events, the rate-determining steps in transcript decay (49). In E. coli, these steps are accomplished by one of two enzymes, RNase E or RNase III. Although mRNA decay has not been investigated in H. influenzae; identifiable homologs of both RNase E (62.3% similarity) and RNase III (80.2% similarity) are present (68), and I will assume that the pathways are conserved. RNase E plays a key role in the decay of bacterial mRNA El(16,49,110,130,139,1490. This enzyme is a component of a multiprotein complex, the degradasome, that is responsible for the majority of endonucleolytic mRNA decay in E. coli (49). The degradasome also contains PNPase (polynucleotide phosphorylase, a phosphate-dependent exonuclease), RhlB (a dead box helicase), the glycolytic enzyme enolase, and possibly the chaperone proteins DnaK and GroEL (34,132). The degradasome is proposed to bind to the 5' end of messenger RNA and the initial endonucleolytic cleavage event determines the rate-determining step of message decay, which proceeds very rapidly in an 'all or none' fashion (49). RNase E cleavage sites have been identified within many bacterial, phage and plasmid genes and operons, including the his, rpslI-dnaG-rpoD, pap, rplKAJLrpoBC, dicB, rpsO-pnp, unc, ribosomal protein S20, and rne messages of E. coli, RNA I of ColEl and T4 gene 32 (reviewed in (139). Despite the number of messages known to contain RNase E cleavage sites, a consensus site for cleavage has not been determined. It is currently 187 thought that RNase E binds to AU-rich single-stranded RNA and cleaves 5' to the dinucleotide A U 0(117,129). The role of mRNA secondary structure in substrate recognition by RNase E is unclear (49) . Most RNAse E cleavage sites are preceded or followed by a stable stem structure, suggested to stabilize local secondary structure so that the adjacent cleavage site remains single-stranded (49). On the other hand, there are many examples of messages that contain 5' RNA secondary structures that act as stabilizers against endonucleolytic mRNA decay (32,139,149). In this case, RNA secondary structures may sterically hinder substrate recognition by RNAse E (49). Therefore, RNA secondary structures may either enhance or interfere with endonucleolytic decay of mRNA by RNase E. Another endonuclease involved in mRNA decay is RNase III (50,138). Although the expression of some genes is regulated by RNase III cleavage, this enzyme does not play a general role in mRNA decay, because rnc mutants exhibit normal decay of bulk messenger RNA (19,177). However, there are some examples of messages whose stability is regulated by RNase III cleavage. These include the pnp, rnc, metY-nusA-infB, dicF, secE-nusG and rplKAJL-rpoBC genes and operons of E. coli, and also several plasmid and phage genes [reviewed in (50]. Unlike RNase E, RNase III binds to and cleaves within stretches of double- stranded RNA, including mRNA stem loops and antisense/sense RNA duplexes in vivo, and long stretches of synthetic double-stranded RNA in vitro. Although an RNase III cleavage consensus sequence has been proposed (109), many in vivo cleavage sites do not conform well to the consensus sequence (50). In general, RNase III in vivo cleavage sites occur within stems that are up to 20 bp in length and are interrupted by small bulges; the cleavage sites are symmetrical - two monomers of RNase III bind to substrate RNA and cleave on either side of the stem (50) . 188 Could sxy RNA structure A. l or A.2 be a site for endonucleolytic cleavage by RNAse E or RNase III? I suggest that neither RNA structure is a substrate for RNAse E cleavage, although the possibility that either structure stabilizes a nearby RNAse E cleavage site cannot be ruled out. Because RNase III is known to cleave double stranded RNA molecules containing internal bulges, and sxy structures A. l and A.2 match these criteria, it is also possible that RNase III cleaves within sxy mRNA secondary structure. However, I suggest that the main regulatory function of structure A. l or A.2 is not to destabilize sxy mRNA. If either structure served as a 5' recognition motif for endonucleolytic cleavage, we would expect that cleavage events within the 5' half of the sxy message would couple the chemical and functional decay of the sxy message, so that transcription and translation of the sxy message would be equally affected; therefore, this type of regulatory mechanism would not be expected to differentially affect sxy mRNA and protein levels. Data obtained with sxyv.lacZ operon and protein fusions showed clearly, for fusions in which the 3' half of the sxy message had been eliminated (Figures 3.18 and 3.19; Table 5.2), and for fusions containing the sxy-1 mutation, that elimination or destabilization of sxy RNA 5' secondary structure had a larger effect on protein fusions than on operon fusions. Overall, my results do not support a role for sxy structure A. l or A.2 in the regulation of sxy mRNA stability. One issue to be addressed is whether the observed differences in sxy mRNA levels in cells grown under different conditions are due to differences in mRNA stability. [For example, the ompA message is regulated by growth-rate-determined differences in mRNA stability (187)]. This can be addressed by measuring sxy mRNA stability throughout growth in sBHI and in MIV. A second issue to be addressed is whether the different amounts of sxy mRNA in different strains are due to differences in mRNA stability. 189 5.1.5.3. Sxy RNA secondary structure may affect translational initiation from the sxy ribosome binding site. mRNA secondary structure often affects translational initiation, the rate-controlling step in the regulation of translation El(52,53,59,74,75,127,128,171). The translational initiation region (TIR; 128) includes, at a minimum, the ribosome binding site (the Shine Dalgarno Sequence) and the initiation codon (75). Although the Shine Dalgarno sequence is thus far the only sequence other than the initiation codon to be shown unequivocally to participate in translational initiation, other sequences within the TIR may also affect the efficiency of ribosome binding (75,127,128). The TIR sequence is non-random from approximately -20 to +13 (where +1 corresponds to the first A of the initiation codon), which corresponds with the region of the mRNA to which the ribosome binds (75). In general, the TIR sequence is biased, for A residues, and most naturally-occurring ribosome binding sites contain unstructured RNAs to which the 30 S subunit of the ribosome can bind efficiently (75). In addition, mRNA regions other than the Shine Dalgarno sequence and initiation codon have been recognized as being important for interaction with 16S rRNA or with ribosomal subunits (127,128). Several models have been proposed for the initiation of translation [reviewed in (59)]. McCarthy has recently reviewed the pathway for translational initiation. Translational initiation requires the 30S ribosomal subunit, each of the three initiation factors IF1, IF2 and IF3, mRNA, the initiator tRNA (f-Met-tRNAfMet) and a GTP molecule. The first step in translational initiation is the binding of the 30S ribosomal subunit, complexed with the three initiation factors, to mRNA and f-Met-tRNAfM e t to generate a pre-ternary complex. A first-order rearrangement leads to the formation of a stable 30S complex (the ternary complex), which can either disassociate into its original components, or become transformed into a 70S initiation complex, as a result of the 190 binding of the 50S subunit. The 30S-50S association becomes irreversible following the ejection of the initiator factors and hydrolysis of GTP. The initiation complex cannot form unless the Shine Dalgarno sequence and the initiation codon are accessible [(53,75)]. It has been demonstrated that mRNA secondary structure can lower the efficiency of translational initiation; the best characterized example is the MS2 coat protein gene of phage R17, whose translational efficiency has been shown to be linearly proportional to the stability of a hairpin structure that sequesters the Shine Dalgarno sequence and start codon of this gene (53). There are many examples of phage and bacterial genes whose expression is controlled by RNA secondary structures that inhibit the initiation of translation. Most examples fall into one of two categories. In the first, an RNA secondary structure controls the expression of a single structural gene; examples include the MS2 coat protein; the lamB, rpoS, rpoH (sigma 32), trmD, and secA genes of E. coli; the trpE gene of B. subtilis, and the tnp (transposase) gene of IS10. In the second category, an inhibitory RNA secondary structure is responsible for coupling the translation of a downstream gene in an operon with the translation of an upstream gene; examples include the lysis peptide of the f2 and MS2 phages, the coat and replicase genes of the MS2 and QB phages, the erm genes of B subtilis, the cat genes of Gram-positive bacteria, and the uncDC, glnBA and rplIJL genes of E. coli. Several of the inhibitory RNA secondary structures in the systems described above resemble sxy RNA structures A. l and A.2, in that they are long-range secondary structures that involve base pairing between distal segments of the mRNA. These include the RNA secondary structures involved in the regulation of the sec A and rpoH genes of E. coli, and in the translational coupling of the glnB and gin A genes of E. coli and the coat and replicase genes of phages MS2 and QB. 191 There appears to be no general rule for describing the relationship between the translational efficiency of a particular mRNA species and its stability against nucleolytic degradation (32,49,110,149,150). For some mRNAs, alterations that affect translational efficiency have little effect on decay B(172,188). However, for a number of mRNAs - such as the lacZ and trp mRNAs of E. coli - mutational alteration of the TIR affects both the efficicency of translational initiation and the half life of mRNA (46,188, 196). For messages in which there is a positive correlation between translational efficiency and message half life, 5' determinants of stability may overlap with the TIR, such that the rate-limiting step of mRNA decay is at or near the RBS. This overlap may result in competition of the degradasome with the ribosome for binding to the mRNA. In this case, the initiation of mRNA decay may occur only when local RNA secondary structure forms, impeding binding of mRNA by 30S ribosomal subunits and preventing ribosome loading (32, 49,110,139, 149). Because the sxy-1 mutation and eliminating stem IA of sxy structures A . l and A.2 increase the expression of lacZ from protein fusions more than lacZ expression from operon fusions, I suggest that sxy RNA secondary structure regulates sxy expression primarily at the level of translational initiation. Furthermore, I suggest that the observed effects on sxy mRNA levels in these mutants may be a secondary effect of increased access of the H. influenzae degradative machinery to the 5' end of the sxy message under conditions where ribosomal loading efficiency has been reduced. As previously mentioned, this model presumes that there is no role for the regulation of sxy mRNA stability throughout growth in wild type cells. How might this hypothesis be tested in future experiments? One way would be to directly measure the relative amounts of Sxy protein produced by cells carrying sxy, sxy-1, sxy-6 and sxy-7 alleles, to ensure that the differences in protein levels predicted by 192 indirect f3-galactosidase measurements are correlated with direct measurements of protein levels. This experiment would also address the issue of whether more than one form of the Sxy protein is produced in KW20, as previously suggested (201). Before this experiment could be done, an anti-Sxy antibody would need to be generated, since there is not one currently available. An alternative would be to measure the translational efficiencies of the wild type and mutant TIRs using an RNase-free in vitro translational reporter system. If the translational efficiencies of the wild type and mutant sxy RNAs in such a system were correlated with in vivo results, this would provide strong evidence that sxy RNA secondary structure A regulates sxy expression at the level of translation, and, in addition, would indicate that no trans—acting factor is required for stabilizing or destabilizing the inhibitory RNA structure. Another way would be to perform toeprint assays on the wild type and mutant RNAs. This type of assay measures the in vitro formation of 30S ternary complexes, which have been shown to correlate well with in vivo translational efficiencies (75). 5.1.5.4. What is the signal for the increase in sxy mRNA translation during stationary phase in sBHI and in MIV? Since the concentration of Sxy appears to be a key determinant of the levels of competence achieved by cells, identification of the specific regulatory signals that control sxy expression should yield information about the signals that cause cells to develop competence. One type of sxy regulation appears to occur at the level of sxy mRNA availability, probably by transcriptional initiation at the sxy promoter. A second type of regulation, as suggested above, appears to occur at the level of translational efficiency, and is mediated by sxy mRNA secondary structure. It appears that translation of the sxy RBS increases when cells experience nutrient limitation, following 193 logarithmic growth in sBHI, and in MIV. What is the mechanism by which sxy RNA secondary structure inhibits mRNA translation at a specific stage of growth? Negative regulation of translational initiation often involves the binding of repressor proteins or antisense RNAs to mRNAs at or near the TIR [reviewed in (128) and (172)], sometimes involving recognition of mRNA secondary structure. Examples of genes whose translation is regulated by frans-acting proteins include ribosomal subunits (a, rif, L l l and spc) and the threonyl synthetase genes of E. coli, the DNA polymerase of bacteriophage T4, the coat protein gene of bacteriophage R17, and the early genes of bacteriophage T4. Most commonly, regulation occurs by auto-repression brought about by the binding of a protein encoded by the gene or operon. Examples in which antisense RNA mediates the negative regulation of translation include IS10 transposase, ompF of E. coli and the hok/sok plasmid maintenance system. There are several known mechanisms by which the initiation or elongation of translation is positively, rather than negatively, regulated [reviewed in (128,171)]. One mechanism of translational activation of E. coli and phage genes involves the regulated formation of a transcript containing a novel 5' end that somehow precludes the formation of an inhibitory RNA secondary structure. The formation of a message with a novel 5' end may occur by transcriptional activation of a second promoter, as described for the pyrBl and gal operons of E. coli and the lysozyme gene of bacteriophage T4, or by endonucleolytic cleavage by RNase III, as described for the bacteriophage T7 early genes and the X cIII gene (128). A second mechanism of translational activation occurs via the binding of trans-activating proteins. Translational regulation of the phage Mu com gene by binding of the mom protein provides the best example of a situation in which the binding of a protein to a message activates translation. A third mechanism for translational activation involves coupling of the initiation of translation from the RBS of one gene with translation of an upstream gene, as described above (Section 5.1.5.3). 194 Sxy translational initiation may be negatively regulated by the binding of a repressor protein that binds to sxy RNA structure A. l or A.2. In this scenario, the trans—acting regulatory factor would be expected to be present in high concentrations during exponential growth in sBHI. A precedent for this type of molecule is the Hfq protein of E. coli, which serves as an RNA binding 'chaperone' known to influence the mRNA stability and/or translational efficiency of several E. coli messages, including the 9S RNA, ompF, rpoS, mutH and mutS messages (71,185,187). The H. influenzae Hfq protein (HI#0411) is 92% identical to its E. coli homolog (68), and would therefore be expected to have a similar function. Alternatively an antisense RNA molecule may bind to the sxy TIR during early exponential growth. A second possibility is that sxy mRNA translation may be positively regulated. One such mechanism might be the binding of a frans-acting activator of translation, which we would expect in this case to be activated or more highly expressed only in the stationary phase of sBHI or in cells transferred to MIV. Alternatively, the activation of sxy translation could result from the formation of a novel transcript that doesn't include the inhibitory RNA secondary structure. Although this would be unlikely to occur as a result of RNAse III degradation, it is possible that such a transcript could be induced by transcriptional activation at a second sxy promoter. This suggestion is supported by the biphasic pattern observed for sxydacZ operon fusion expression. One outstanding issue is whether a trans acting factor is required to inhibit or activate sxy translation. RNA secondary structure mapping studies suggested that RNA structure A.lor A.2 can form in vitro without the presence of regulatory factors. However, this observation doesn't preclude the possibility that a frans-acting factor is required in vivo for stabilization or destabilization of this secondary structure. This issue could be easily addressed by examining translation of sxy and sxy mutant mRNAS in vitro in cell-free translation assays. In addition, proteins that bind to sxy mRNA during exponential growth (repressing factors) or in stationary phase (activating factors) of 195 sBHI could be identified by biochemical means, such as Northwestecr^ Jblotting or gel shift assays. Another outstanding issue is whether only the base-pairing of sxy structure A . l or A.2 is important, or whether the actual sequence of the structure is important because of interaction of sxy mRNA with trans-acting factors, ribosomal RNA, etc. To investigate this issue, further RNase or chemical mapping experiments could be carried out on wild type sxy and mutant RNAs. Another approach would be to try and identify additional sxy hypercompetence point mutations as previously described (155). The importance of sxy mRNA sequence could also be investigated by creating site-directed mutations which exchange GC for CG base pairs within Stem IA of structures A. l and A.2. If it appears that trans-acting factors are not required for the regulation of sxy translation, the issue of multiple start sites for sxy transcription as a means of controlling 5' message end formation could be investigated. Although the sxy transcriptional start site was identified using primer extension (200), it was identified only for cells in MIV. Therefore, primer extension should be carried out for cells growing in sBHI during the logarithmic phase of growth. The sxy promoter or promoter(s) could also be mapped by cloning a promoterless lacZ gene downstream of subcloned sxy DNA fragments and measuring P-galactosidase expression from cloned fragments throughout various stages of growth. 5.1.6. A modified model for competence development in sBHI. Overall, my results do not support the simple model for competence development presented in Section 3.1.4. Two versions of a modified model for the regulation of competence in wild type cells growing in sBHI are presented in Figures 5.1 and 5.2. respectively. In these models, there are two steps necessary for the activation of competence, both of which are required to increase the expression of late competence (com) genes. The first is the activation of a hypothetical com transcriptional activator 196 (Figure 5.1), or the inhibition of a hypothetical repressor of com transcription (Figure 5.2), by Sxy, either by posttranslational modification or at the level of gene expression. Modification of the activity levels of activator or repressor is accomplished by increasing Sxy concentrations within the cell by several mechanisms. During early logarithmic growth in sBHI, concentrations of Sxy are increased due to an increase in the relative amounts of sxy mRNA by an undefined mechanism, probably due to a cAMP-independent transcriptional activation of the sxy promoter. As cells enter stationary phase in sBHI, or are transferred to MIV, the translational efficiency of the sxy RBS is increased so that concentrations of Sxy are increased to levels that promote maximal competence development. However, increasing Sxy concentration is not normally sufficient to induce maximal competence development when cAMP concentrations are low. Maximal competence development also requires an increase in concentrations of intracellular cAMP to fully induce transcription of competence genes, and cAMP concentrations do not become high until cells have reached the end of logarithmic growth in sBHI. It should be noted that this model is purely hypothetical, and is based primarily on the observed patterns of expression of sxy and cAMP-dependent reporter genes. It has not been proven that cAMP/CRP directly activates the transcription of competence genes. Furthermore, the possibility that Sxy itself is transcription factor has not been ruled out. Moreover, candidate activator or repressor genes that may be regulated by Sxy have not been identified. In fact, there are currently no known repressors of competence. Although genetic screens have been carried out to identify repressors of competence [(155) and Chapter 4 of this thesis], these genetic screens were not extensive and, urthermore, knockouts of such repressors may prove lethal to cells. Further studies are required to determine if either the cAMP/CRP complex or Sxy can bind to CRE sequences upstream of competence genes. In addition, other gene products that interact with Sxy could be identified through genetic screens, or by protein-protein 197 C ^ J com t > ^ L ^ H _ _ l ^ Z l > Q Com ^ sxy mRNA B. Mid logarithmic growth in sBHI: transformation frequencies -10 " 6 198 C. Post logarithmic growth in sBHI: transformation frequencies - 1 0 - 4 Figure 5.1. Modified model for competence development in sBHI, version 1. A. Early logarithimic growth in sBHI. Both Sxy and cAMP are absent or very low in concentration. Hypothetical activator (A) is either not activated (as shown in the figure) or not expressed, because Sxy concentrations are low. Little or no transcription of com genes occurs. B. Mid logarithmic growth in sBHI. Sxy concentrations increase, due to an increase in the levels of sxy mRNA, thereby increasing the activity (as shown in the figure) or expression of competence activator (A*). However, cAMP concentrations are still low. Transcription of com genes increases but is still not maximal. C. Post logarithmic growth in sBHI. Sxy concentrations increase further, due to an increase in sxy mRNA translatability, and therefore more competence activator is available. cAMP concentrations are high. Transcription of comgenes increases, and cells achieve maximal competence in sBHI. CRE; competence regulatory element; A, hypothetical competence activator; ribosomes translating sxy mRNA. 199 A . Early logarithmic growth in sBHI: transformation frequencies <10_l B. Mid logarithmic growth in sBHI: transformation frequencies ~10" 6 v : ) 200 C. Post logarithmic growth in sBHI: transformation frequencies ~10" 4 Figure 5.2. Modified model for competence development in sBHI, version 2 . A. Early logarithimic growth in sBHI. Both Sxy and cAMP are absent or very low in concentration. Hypothetical competence repressor (R) is activated (as shown in the figure) or not transcriptionally repressed, because Sxy concentrations are low. Little or no transcription of the com genes occurs. B. Mid logarithmic growth in sBHI. Sxy concentrations increase, because of an increase in the levels of sxy mRNA, thereby reducing the activity (as shown in the figure) or expression of competence repressor. However, cAMP concentrations are still low. Transcription of com genes increases but is still not maximal. C. Post logarithmic growth in sBHI. Sxy concentrations increase further, due to an increase in translatability of sxy mRNA, and therefore less competence repressor is present or active. cAMP concentrations are high. Transcription of com genes increases, and cells achieve maximal competence in sBHI. CRE, competence regulatory element; R, hypothetical competence repr essor;^  ribosomes translating sxy mRNA. 201 interaction screens, using techniques such yeast two-hybrid screens or immunoprecipitation. In order for a model of competence development to be robust, it should be able to satisfy several independent observations about competence development in sBHI. The models presented in Figures 5.1 and 5.2 satisfy the following observations about competence development: 1. Sxy, cya, crp are absolutely required for competence development, because none of these gene products alone is able to fully activate the expression of competence genes. 2. The addition of 1 mM cAMP to cells in mid logarithmic growth in sBHI fully induces wild type levels of competence, not, as previously suggested, because cAMP activates sxy transcription, but because Sxy concentrations are already high at this stage of growth. However, under wild type Sxy concentrations, competence is not increased above wild type levels under these conditions because not enough Sxy is present to fully increase the activity of competence activator or to fully inhibit the activity of competence repressor. 3. Increasing Sxy concentrations above wild type levels increases competence above wild type levels, as long as cAMP is present, because enough Sxy is present to fully increase the activity of competence activator or to fully inhibit the activity of competence repressor. I have not extended the above models of competence development to cells in MIV. Studies presented in this thesis and elsewhere (120) suggest that neither an increase in Sxy nor cAMP concentrations is responsible for the higher competence levels achieved by cells in MIV relative to those in sBHI. Presently, it is not clear whether similar mechanisms of competence development occur under these different growth conditions. 202 5.2. Discussion of Fis results Using Tn926 mutagenesis, I identified a mutation that delayed spontaneous competence development of cells in sBHI (Figure 4.3). This mutation was mapped to the H. influenzae fis operon, which encodes the E. coli Orfl and Fis homologs (Figure 4.5). Subsequently, I created orfl and fis null mutations, and demonstrated that spontaneous competence of fis mutants was reduced approximately 500-fold relative to that of wild type cells (Figure 4.8). In E. coli, Fis has many known cellular functions. Presumably, Fis carries out similar functions in H. influenzae and E. coli, since Fis is highly conserved between H. influenzae and the enterics, and since the H. influenzae Fis protein is able to carry out some of the known functions of its E. coli homolog (30). In the following discussion, I suggest mechanisms by which Fis might affect natural transformation of H. influenzae. 5.2.1. Repressors of competence development were not identified by Tn916 mutagenesis. Initially, I looked for repressors of competence development by Tn916 mutagenesis and mutants that are hypercompetent during early logarithmic growth (Section 4.2.1). However, the desired mutants were not recovered. Recently, the site specificity of Tn916 insertion has been evaluated (89, 137). Hosking (1998) identified the sequence 5' TTTTT(N4_7)AAAAA 3' as a specific target for Tn916 insertion, and identified 167 such sites within the H. influenzae genome. These authors, and Nelson et al. (1997), noted that because of the bias of Tn926 insertion sites, and because many of the identified insertion sites did not occur in open reading frames in H. influenzae, Tn916 is not likely to represent an effective mutagenesis tool. The search for repressors of competence should be conducted again after mutagenizing KW20 more effectively, using cassette mutagenesis or in vitro Tn7 mutagenesis (58,83). 203 5.2.2. fis mutants are unlikely to have reduced transformation frequencies simply as a result of slow growth. H. influenzae cells carrying fis null mutations grew slowly, at about 75% of the rate of wild type cells (Figure 4.7). As discussed previously, E. coli fis cells also grow slowly. The competence reduction in fis mutants may be a nonspecific response of cells to growth rate reduction. However, a reduction in growth rate of KW20 is unlikely to account for a 500-fold decrease in transformation frequncies for several reasons. One is that other H. influenzae mutants with equivalent growth rates to fis mutants don't have drastic reductions in transformation frequencies. For example, a strain carrying a transposon mutation in atpA (ATPase), like/z's cells, grows with a doubling time of about 40 minutes, and yet its transformation frequency is reduced only 10-fold in MIV and 6 to 7-fold in sBHI (83). This suggess that a slower growth rate per se isn't sufficient to drastically reduce transformation frequencies. A second reason is that antibiotics and other chemical compounds tested for inhibition of competence development reduce H. influenzae cell viability without decreasing transformation frequencies (154). Therefore, it seems that the fis mutant phenotype is not explainable by a nonspecific inhibition of competence development due to reduced growth. 5.2.3. Fis may modulate a mechanistic step of natural transformation. In E. coli, Fis is known to bind to numerous sites in the bacterial genome, facilitating the role of this protein in DNA architecture, DNA replication and site-specific recombination (Section 4.1.2). Because of its structural role in binding to and bending DNA, Fis may have a mechanistic role in natural transformation, instead of regulating gene expression. This scenario seems unlikely, since fis expression in enteric bacteria is dramatically increased 204 upon nutritional upshift and is repressed during the stationary phase of growth. If the pattern of fis expression is similar in H. influenzae, there are predicted to be relatively few molecules (<100 dimers, by comparison with E. coli) of Fis present within competent cells. In addition, the lack of Fis causes a dramatic reduction in competence only in sBHI, which is unexpected for a gene with a mechanistic role in transformation. Nevertheless, mechanistic roles for Fis should not be ruled out. One such role for Fis may be the enhancement of homologous recombination by stabilization of recombinational intermediates. Although Fis is not known to be involved in homologous (RecA-mediated) recombination in E. coli, it does enhance both site-specific and illegitimate recombination. In site-specific recombination, Fis promotes DNA exchange by binding to enhancer sequences and stabilizing recombinational intermediates via direct protein-protein interactions between Fis and the site-specific recombinase proteins (42,93,104,146). Similarly, Fis is suggested to enhance illegitimate recombination by binding to and bending DNA to promote DNA exchange (166). Therefore, it is possible that Fis promotes homologous recombination during natural transformation by stabilizing RecA-DNA recombination intermediates. DNA binding, DNA uptake and phage recombination have not been measured in H. influenzae fis mutants. If Fis affects the homologous recombination step of transformation, we would predict that//s mutants would have wild type levels of DNA binding and uptake, would have reduced levels of phage recombination, and would possibly be hypersensitive to UV irradiation. Each of these parameters is easily testable. 5.2.4. Fis may directly or indirectly affect competence by transcriptional regulation In E. coli, Fis regulates the transcription of a number of genes, including those for stable RNAs (rRNA and tRNA) and for proteins involved in catabolic processes (Section 4.2.1). Since we study the regulation of transformation in H. influenzae, the most interesting 205 role for Fis in competence development would be that of a regulatory protein. However, because of the pleiotropic nature of fis null mutants in E. coli, a specific functional role for Fis in any process is difficult to identify. The issue of whether Fis plays a regulatory role in competence development can be addressed by further examining epistasis between fis and other known competence genes. In E. coli, Fis is known to affect the expression of several genes also regulated by CRP. Therefore, Fis could act in conjunction with CRP to transactivate competence genes. Three obvious candidate genes for Fis regulation include cya, sxy and the crp gene itself. However, Fis is unlikely to activate the transcription of cya, since the addition of exogenous cAMP failed to overcome the transformation defect of fis cells (Figure 4.11). Similarly, the presence of the sxy-1 mutation failed to raise the competence of fis cells more than 30-fold. The expression of late com genes, such as Apr A, rec-2 and comF should be compared in Fis+ and Fis- cells. Fis is also known to affect the regulation of promoters that are sensitive to the level of supercoiling, including stable rRNA operons. In H. influenzae, strains with mutations in topoisomerase I (topA) have severely reduced transformation frequencies (45). Perhaps Fis, in conjunction with topoisomerase I, affects competence development by controlling the level of superhelicity at competence- specific promoters. The interaction between TopA and Fis in competence development could be tested by examining the effect of overexpression of top A in a fis strain and vice versa. Since supercoiling-sensitive promoters are known to be induced by conditions of low oxygen availability, the fis mutant should be tested for competence following the aerobic-anaerobic shift procedure for competence induction to see if fis cells have reduced competence under these conditions. 206 5.2.5. Summary of Fis Results Null mutations in H. influenzae fis were found to have transformation frequencies which were reduced by 500-fold relative to those of wild type cells. Since fis null alleles are known to be pleiotropic, and since Fis is known to have many functions in E. coli, it is difficult to assign a specific functional role for Fis in the regulation of competence development. I have suggested several possibilities for the reduction of competence observed in fis mutants, including the possibilities that Fis is involved in stabilizing recombination intermediates, or that Fis activates the transcription of competence genes. 207 B i b l i o g r a p h y 1. eMotif Search. 1999. http://dna.stanford.edu/emotif/ 2. The Institute for Genomic Research H. influenzae Rd Genome Database (HIDB). 1999. http://www.tigr.org/tdb/CMR/ghi/htmls/SplashPage.html 3. 2ZIP. 1999. http://www.dkfz_heidelberg.de/tbi/services/2Zip/Zip_start.sh. 4. Baylor College of Medicine Search Launcher : Protein secondary Structure Prediction, http: //dot.imgen.bcm.tmc.edu:9331/seq-search/struc-predict.html 5. Baylor College of Medicine Search Launcher: Multiple Sequencce Alignments. 1999. http://dot.imgen.bcm.tmc.edu:9331/multi-align/multi -align.html 6. Blast Version 2. 1999. http://www.ncbi.hlm.nih.gov.blast/blast.cgi?jform=0 7. Helix Turn Helix Motif Prediction. 1999. http://pbil.ibcp.fr/cgi-bin/npsa_automat.pl?page=npsa_hth.html 8. Local Alignment of Multiple Alignments. 1999. http://www.blocks.fhcrc.org /blocks-bin/LAMA_search.sh?/ howard/servers/blockmkr/tmp/ 28613/ 28613.mblks 9. Mfold Server version 3.0. 1999. http://mfold2.wustl.edu/~mfold/rna/forml.cgi 10. Motif Alignment and Search Tool. 1999. http://www.blocks.fhcrc.org/blockmkr-bin/mast.sh?/howard/servers/blockmkr/tmp/28613/28613.mmast 11. NCBI Microbial Genomes Blast Databases. 1999. http://www.ncbi.nlm.gov/ BLAST/ unfinishedgenome.html 208 12. Afflerbach, H., Schroder, O. and Wagner, R. 1999. Conformational changes of the upstream DNA mediated by H-NS and FIS regulate E. coli RrnB PI promoter activity. /. Mol. Biol. 286: 339-53. 13. Albritton, W. L., Setlow, J. K., Thomas, M., Sottnek, F. and Steigerwalt, A. G. 1984. Heterospecific transformation in the genus Haemophilus. Mol. Gen. Genet. 193: 358-63. 14. Alexander, H. and Leidy, G. 1951. Determination of inherited traits of H.influenzae. J.Exp.Med. 93:345-359. 15. Alexander, H., Leidy, G. and Hahn, E. 1954. Studies on the nature of Hemophilus influenzae cells susceptible to heritable changes by desoxyribonucleic acids. /. Exp. Med. 99: 505-533. 16. Alifano, P., Bruni, C. B. and Carlomagno, M. S. 1994. Control of mRNA processing and decay in prokaryotes. Genetica 94:157-72. 17. Altschul, S. F., Madden, T. L., Schaffer, A. A., Zhang, J., Zhang, Z., Miller, W. and Lipman, D. J. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25: 3389-402. 18. Amin, N. and Peterkofsky, A. 1995. A dual mechanism for regulating cAMP levels in Escherichia coli. J. Biol. Chem. 270:11803-5. 19. Apirion, D. and Gitelman, D. R. 1980. Decay of RNA in RNA processing mutants of Escherichia coli. Mol. Gen. Genet. 177: 339-43. 20. Attwood, T. K., Beck, M. E., Bleasby, A. J., Degtyarenko, K., Michie, A. D. and Parry-Smith, D. J. 1997. Novel developments with the PRINTS protein fingerprint database. Nucleic Acids Res. 25: 212-7. 209 21. Augustin, L. B., Jacobson, B. A. and Fuchs, J. A. 1994. Escherichia coli Fis and DnaA proteins bind specifically to the nrd promoter region and affect expression of an nrd-lac fusion. /. Bacteriol. 176: 378-87. 22. Ausubel, F., Brent, R., Kingston, R., Moore, D., Seidman, J., Smith, J. and Stuhl, K. (Ed.). 1994. Current Protocols in Molecular Biology. Green/Wiley, New York. 23. Bailey, T. L. and Gribskov, M. 1998. Methods and statistics for combining motif match scores. /. Comput. Biol. 5: 211-21. 24. Ball, C. A., Osuna, R., Ferguson, K. C. and Johnson, R. C. 1992. Dramatic changes in Fis levels upon nutrient upshift in Escherichia coli. J. Bacteriol. 174: 8043-56. 25. Barany, F., Kahn, M. E. and Smith, H. O. 1983. Directional transport and integration of donor DNA in Haemophilus influenzae transformation. Proc. Natl. Acad. Sci. USA 80: 7274-8. 26. Barcak, G. J., Chandler, M. S., Redfield, R. J. and Tomb, J. F. 1991. Genetic systems in Haemophilus influenzae. Methods Enzymol. 204: 321-42. 27. Barcak, G. J., Tomb, J. F., Laufer, C. S. and Smith, H. O. 1989. Two Haemophilus influenzae Rd genes that complement the recA-like mutation rec-1. J. Bacteriol. 171: 2451-7. 28. Barouki, R. and Smith, H. O. 1985. Reexamination of phenotypic defects in rec-1 and rec-2 mutants of Haemophilus influenzae Rd. /. Bacteriol. 163: 629-34. 29. Bartolome, B., Jubete, Y., Martinez, E. and de la Cruz, F. 1991. Construction and properties of a family of pACYC184-derived cloning vectors compatible with pBR322 and its derivatives. Gene 102: 75-8. 30. Beach, M. B. and Osuna, R. 1998. Identification and characterization of the fis operon in enteric bacteria. /. Bacteriol. 180: 5932-46. 210 31. Beattie, K. L. and Setlow, J. K. 1971. Transformation-defective strains of Haemophilus influenzae. Nat. Nezv Biol. 231:177-9. 32. Bechhofer, D. 1993. 5' mRNA Stabilizers. In: Control of messenger RNA stability. Belasco, J. G. and Brawerman, G. (Ed.). Academic Press Inc., San Diego, California. 31-52: 31-52. 33. Berg, O. G. and von Hippel, P. H. 1988. Selection of DNA binding sites by regulatory proteins. II. The binding specificity of cyclic AMP receptor protein to recognition sites. /. Mol. Biol. 200: 709-23. 34. Blum, E., Py, B., Carpousis, A. J. and Higgins, C. F. 1997. Polyphosphate kinase is a component of the Escherichia coli RNA degradosome. Mol. Microbiol. 26: 387-98. 35. Bokal, A. J., Ross, W., Gaal, T., Johnson, R. C. and Gourse, R. L. 1997. Molecular anatomy of a transcription activation patch: FIS-RNA polymerase interactions at the Escherichia coli rrnB PI promoter. Embo J. 16:154-62. 36. Bokal, A. J. t., Ross, W. and Gourse, R. L. 1995. The transcriptional activator protein FIS: DNA interactions and cooperative interactions with RNA polymerase at the Escherichia coli rrnB PI promoter. /. Mol. Biol. 245:197-207. 37. Boling, M. and Setlow, J. 1969. Dependence of vegetative phage recombination among Haemophilus influenzae bacteriophage on the host cell. /. Virol. 4: 240-243. 38. Bornberg-Bauer, E., Rivals, E. and Vingron, M. 1998. Computational approaches to identify leucine zippers. Nucleic Acids Res. 26: 2740-6. 39. Botsford, J. L. and Harman, J. G. 1992. Cyclic AMP in prokaryotes. Microbiol. Rev. 56: 100-22. 211 40. Brendel, V., Bucher, P., Nourbakhsh, I. R., Blaisdell, B. E. and Karlin, S. 1992. Methods and algorithms for statistical analysis of protein sequences. Proc. Natl. Acad. Sci. USA 89: 2002-6. 41. Brenner, S. 1998. Practical database searching. In: Trends Guide to Bioinformatics. Patterson, M. and Handel, M. (Ed.). Elsevier Science, Cambridge, UK. 42. Bruist, M. F., Glasgow, A. C , Johnson, R. C. and Simon, M. I. 1987. Fis binding to the recombinational enhancer of the Hin DNA inversion system. Genes. Dev. 1: 762-72. 43. Caster, J. H., Postel, E. H. and Goodgal, S. H. 1970. Competence mutants: isolation of transformation deficient strains of Haemophilus influenzae. Nature 227: 515-7. 44. Chandler, M. S. 1992. The gene encoding cyclic AMP receptor protein is required for competence development in Haemophilus influenzae Rd. Proc. Natl. Acad. Sci. USA 89: 1626-1630. 45. Chandler, M. S. and Smith, R. A. 1996. Characterization of the Haemophilus influenzae top A locus: DNA topoisomerase I is required for genetic competence. Gene 169: 25-31. 46. Cho, K. O. and Yanofsky, C. 1988. Sequence changes preceeding a Shine-Dalgarno region influence trpE mRNA translation and decay. /. Mol. Biol. 204: 51-60. 47. Claret, L. and Rouviere-Yaniv, J. 1996. Regulation of HU alpha and H U beta by CRP and FIS in Escherichia coli. J. Mol. Biol. 263:126-39. 48. Clifton, S. W., McCarthy, D. and Roe, B. A. 1994. Sequence of the rec-2 locus of Haemophilus influenzae: homologies to comE-ORF3 of Bacillus subtilis and msbA of Escherichia coli. Gene 146: 95-100. 49. Coburn, G. A. and Mackie, G. A. 1999. Degradation of mRNA in Escherichia coli: an old problem with some new twists. Prog. Nucleic Acid Res. Mol. Biol. 62: 55-108. 212 50. Court, D. 1993. RNA processing and degradation by RNase III. In: Control of messenger stability. Belasco, J. G. and Brawerman, G. (Ed.). Academic press Inc., San Diego, California, pp 71-116. 51. Danner, D. B., Deich, R. A., Sisco, K. L. and Smith, H. O. 1980. An eleven-base-pair sequence determines the specificity of DNA uptake in Haemophilus transformation. Gene 11:311-8. 52. de Smit, M. H. and van Duin, J. 1990. Control of prokaryotic initiation by mRNA secondary structure. Prog. Nucl. Acid Res. 38:1-35. 53. de Smit, M. H. and van Duin, J. 1994. Control of translation by mRNA secondary structure in Escherichia coli. A quantitative analysis of literature data. /. Mol. Biol. 244: 144-50. 54. Deich, R. A. and Smith, H. O. 1980. Mechanism of homospecific DNA uptake in Haemophilus influenzae transformation. Mol. Gen. Genetics 177: 369-74. 55. Dennis, P. P. and Nomura, M. 1975. Regulation of the expression of ribosomal protein genes in Escherichia coli. J. Mol. Biol. 97: 61-76. 56. Dodd, I. B. and Egan, J. B. 1990. Improved detection of helix-turn-helix DNA-binding motifs in protein sequences. Nucleic Acids Res. 18: 5019-26. 57. Dorocicz, I., Williams, P. and Redfield, R. J. 1993. The Haemophilus influenzae adenylate cyclase gene: cloning, sequence and essential role in competence. /. Bacteriol. 175: 7142-7149. 58. Dougherty, B. A. and Smith, H. O. 1999. Identification of Haemophilus influenzae Rd transformation genes using cassette mutagenesis. Microbiology 145: 401-9. 213 59. Draper, D. 1996. Translational initiation. In: Escherichia coli and Salmonella cellular and molecular biology. Neidhardt, F. C. (Ed.). ASM Press, Washington, D.C. pp 902-908. 60. Dreiseikelmann, B. 1994. Translocation of DNA across bacterial membranes. Microbiol. Rev. 58: 293-316. 61. Drlica, K. and Rouviere-Yaniv, J. 1987. Histonelike proteins of bacteria. Microbiol. Rev. 51:301-19. 62. Dubnau, D. 1991. Genetic competence in Bacillus subtilis. Microbiol. Rev. 55: 395-424. 63. Dubnau, D. 1997. Binding and transport of transforming DNA by Bacillus subtilis: the role of type-IV pilin-like proteins-a review. Gene 192:191-8. 64. Ehresmann, C., Baudin, F., Mougel, M., Romby, P., Ebel, J. P. and Ehresmann, B. 1987. Probing the structure of RNAs in solution. Nucleic Acids Res. 15: 9109-28. 65. Elkins, C., Thomas, C. E., Seifert, H. S. and Sparling, P. F. 1991. Species-specific uptake of DNA by gonococci is mediated by a 10-base-pair sequence. /. Bacteriol. 173: 3911-3913. 66. Engler-Blum, G., Meier, M., Frank, J. and Muller, G. A. 1993. Reduction of background problems in nonradioactive northern and Southern blot analyses enables higher sensitivity than 32P-based hybridizations. Anal. Biochem. 210: 235-44. 67. Filutowicz, M., Ross, W., Wild, J. and Gourse, R. L. 1992. Involvement of Fis protein in replication of the Escherichia coli chromosome. /. Bacteriol. 174: 398-407. 68. Fleischmann, R. D., Adams, M. D., White, O., Clayton, R. A., Kirkness, E. F., Kerlavage, A. R., Bult, C. J., Tomb, J. F., Dougherty, B. A., Merrick, J. M. and et al. 1995. Whole-genome random sequencing and assembly of Haemophilus influenzae Rd. Science 269:496-512. 214 69. Franke, A. E. and Clewell, D. B. 1981. Evidence for a chromosome-borne resistance transposon (Tn916) in Streptococcus faecalis that is capable of "conjugal" transfer in the absence of a conjugative plasmid. /. Bacteriol. 145: 494-502. 70. Gawron-Burke, C. and Clewell, D. B. 1984. Regeneration of insertionally inactivated streptococcal DNA fragments after excision of transposon Tn916 in Escherichia coli: strategy for targeting and cloning of genes from gram-positive bacteria. J. Bacteriol. 159: 214-21. 71. Georgellis, D., Arvidson, S. and von Gabain, A. 1992. Decay of ompA mRNA and processing of 9S RNA are immediately affected by shifts in growth rate, but in opposite manners. /. Bacteriol. 174: 5382-90. 72. Gibson, B. W., Campagnari, A. A., Melaugh, W., Phillips, N. J., Apicella, M. A., Grass, S., Wang, J., Palmer, K. L. and Munson, R. S., Jr. 1997. Characterization of a transposon Tn926-generated mutant of Haemophilus ducreyi 35000 defective in lipooligosaccharide biosynthesis. /. Bacteriol. 179: 5062-71. 73. Gille, H., Egan, J. B., Roth, A. and Messer, W. 1991. The FIS protein binds and bends the origin of chromosomal DNA replication, oriC, of Escherichia coli. Nucleic Acids Res. 19: 4167-72. 74. Gold, L. 1988. Posttranscriptional regulatory mechanisms in Escherichia coli. Ann. Rev. Biochem. 57:199-233. 75. Gold, L., Pribnow, D., Scneider, T., Shinedling, S., Britta, S. S. and Stormo, G. 1981. Translational initiation in prokaryotes. Ann. Rev. Microbiol. 35: 365-403. 76. Gonzalez-Gil, G., Bringmann, P. and Kahmann, R. 1996. FIS is a regulator of metabolism in Escherichia coli. Mol. Microbiol. 22: 21-9. 215 77. Goodgal, S. H. and Herriott, R. M. 1961. Studies on transformations of Haemophilus influenzae: I. Competence. /. Gen. Physiol. 44:1201-1227. 78. Goodgal, S. H. and Mitchell, M. A. 1990. Sequence and uptake specificity of cloned sonicated fragments of Haemophilus influenzae DNA. /. Bacteriol. 172: 5924-8. 79. Goodman, S. D. and Scocca, J. J. 1988. Identification and arrangement of the DNA sequence recognized in specific transformation of Neisseria gonorrhoeae. Proc. Natl. Acad. Sci. USA 85: 6982-6. 80. Gosink, K. K., Gaal, T., Bokal, A. J. t. and Gourse, R. L. 1996. A positive control mutant of the transcription activator protein FIS. /. Bacteriol. 178: 5182-7. 81. Gosink, K. K., Ross, W., Leirmo, S., Osuna, R., Finkel, S. E., Johnson, R. C. and Gourse, R. L. 1993. DNA binding and bending are necessary but not sufficient for Fis-dependent activation of rrnB PI. /. Bacteriol. 175:1580-9. 82. Gwinn, M. L., Ramanathan, R., Smith, H. O. and Tomb, J. F. 1998. A new transformation-deficient mutant of Haemophilus influenzae Rd with normal DNA uptake. /. Bacteriol. 180: 746-8. 83. Gwinn, M. L., Stellwagen, A. E., Craig, N. L., Tomb, J. F. and Smith, H. O. 1997. In vitro Tn7 mutagenesis of Haemophilus influenzae Rd and characterization of the role of atpA in transformation. /. Bacteriol. 179:7315-20. 84. Gwinn, M. L., Yi, D., Smith, H. O. and Tomb, J.-F. 1996. Role of Two-Component Signal Transduction and the Phosphoenolpyruvate:carbohydrate Phosphotransferase Systems in Competence development of Haemophilus influenzae Rd. /. Bacteriol. 178: 6366-6368. 216 85. Haff ter, P. and Bickle, T. A. 1987. Purification and DNA-binding properties of FIS and Cin, two proteins required for the bacteriophage PI site-specific recombination system, cin. /. Mol. Biol. 198: 579-87. 86. Henikoff, S., Henikoff, J. G., Alford, W. J. and Pietrokovski, S. 1995. Automated construction and graphical presentation of protein blocks from unaligned sequences.,. Gene 163: GC17-26. 87. Herriott, R. M., Meyer, E. M. and Vogt, M. 1970. Defined nongrowth media for stage II development of competence in Haemophilus influenzae. J. Bacteriol. 101: 517-24. 88. Holland, J., Towner, K. J. and Williams, P. 1992. Tn916 insertion mutagenesis in Escherichia coli and Haemophilus influenzae type b following conjugative transfer. /. Gen. Microbiol. 138: 509-515. 89. Hosking, S. L., Deadman, M. E., Moxon, E. R., Peden, J. F., Saunders, N. J. and High, N. J. 1998. An in silico evaluation of Tn916 as a tool for generalized mutagenesis in Haemophilus influenzae Rd. Microbiology 144: 2525-30. 90. Hubner, P. and Arber, W. 1989. Mutational analysis of a prokaryotic recombinational enhancer element with two functions. Embo J. 8: 577-85. 91. Huynen, M. A. and Bork, P. 1998. Measuring genome evolution. Proc. Natl. Acad. Sci. U S A 95: 5849-56. 92. Johnson, R. C., Ball, C. A., Pfeffer, D. and Simon, M. I. 1988. Isolation of the gene encoding the Hin recombinational enhancer binding protein. Proc. Natl. Acad. Sci. U S A 85: 3484-8. 93. Johnson, R. C., Glasgow, A. C. and Simon, M. I. 1987. Spatial relationship of the Fis binding sites for Hin recombinational enhancer activity. Nature 329: 462-5. 217 94. Kahmann, R., Rudt, F., Koch, C. and Mertens, G. 1985. G inversion in bacteriophage Mu DNA is stimulated by a site within the invertase gene and a host factor. Cell 41:771-80. 95. Kahn, M., Maul, G. and Goodgal, S. 1982. Possible mechanism for donor DNA binding and transport in Haemophilus. PNAS 79: 6370-6374. 96. Kahn, M. E., Barany, F. and Smith, H. O. 1983. Transformasomes: specialized membranous structures that protect DNA during Haemophilus transformation. Proc. Natl. Acad. SciUSA80: 6927-31. 97. Kahn, M. E. and Smith, H. O. 1984. Transformation in Haemophilus: a problem in membrane biology. [Review]. /. of Membrane Biol. 81: 89-103. 98. Karlin, S., Mrazek, J. and Campbell, A. M. 1996. Frequent oligonucleotides and peptides of the Haemophilus influenzae genome. Nucleic Acids Res. 24: 4263-72. 99. Karudapuram, S. and Barcak, G. J. 1997. The Haemophilus influenzae dprABC genes constitute a competence- inducible operon that requires the product of the tfoX (sxy) gene for transcriptional activation. /. Bacteriol. 179: 4815-20. 100. Karudapuram, S., Zhao, X. and Barcak, G. J. 1995. DNA sequence and characterization of Haemophilus influenzae dprA+, a gene required for chromosomal but not plasmid DNA transformation. /. Bacteriol. 177: 3235-40. 101. Kauc, L. and Goodgal, S. H. 1989. Introduction of transposon Tn916 DNA into Haemophilus influenzae and Haemophilus parainfluenzae. J. Bacteriol. 171: 6625-8. 102. Keener, J. and Nomura, M. 1996. Regulation of ribosome synthesis. In: Escherichia coli and Salmonella cellular and molecular biology. Neidhardt (Ed.). ASM PRess, Washington, D.C. pp. 1417-1431. 218 103. Kleckner, N., Bender, J. and Gottesman, S. 1991. Uses of transposons with emphasis on TnlO. Methods. Enzymol. 204:139-80. 104. Koch, C., Ninnemann, O., Fuss, H. and Kahmann, R. 1991. The N-terminal part of the E.coli DNA binding protein FIS is essential for stimulating site-specific DNA inversion but is not required for specific DNA binding. Nucleic Acids Res. 19: 5915-22. 105. Koch, C., Vandekerckhove, J. and Kahmann, R. 1988. Escherichia coli host factor for site-specific DNA inversion: cloning and characterization of the fis gene. Proc. Natl. Acad. Sci. USA 85: 4237-41. 106. Koetsier, P. A., Schorr, J. and Doerfler, W. 1993. A rapid optimized protocol for downward alkaline Southern blotting of DNA. Biotechniques 15: 260-2. 107. Kondrashov, A. 1994. Sex and deleterious mutation. Nature 369: 99-100. 108. Kostrewa, D., Granzin, J., Stock, D., Choe, H. W., Labahn, J. and Saenger, W. 1992. Crystal structure of the factor for inversion stimulation FIS at 2.0 A resolution. /. Mol. Biol. 226: 209-26. 109. Krinke, L. and Wulff, D. L. 1990. The cleavage specificity of RNase III. Nucleic Acids Res. 18:4809-15. 110. Kushner, S. 1996. mRNA Decay. In: Escherichia coli and Salmonella. Neidhardt, F. C. (Ed.). ASM press, Washington, D.C pp 849-860. 111. Landick, R., Turnbough, C. L. and Yanofsky, C. 1996. Transcription attenuation. In: Escherichia coli and Salmonella cellular and molecular biology. Neidhardt, F. C. (Ed.). ASM Press, Washington D.C. pp 1263-1286. 112. Larson, T. G. and Goodgal, S. H. 1991. Sequence and transcriptional regulation of comlOlA, a locus required for genetic transformation in Haemophilus influenzae. J. Bacteriol. 173:4683-91. 219 113. Larson, T. G. and Goodgal, S. H. 1992. Donor DNA processing is blocked by a mutation in the comlOlA locus of Haemophilus influenzae. J. Bacteriol. 174: 3392-4. 114. Larson, T. G., Roszczyk, E. and Goodgal, S. 1991. Molecular cloning of two linked loci that increase the transformability of transformation-deficient mutants of Haemophilus influenzae. J. Bacteriol. 173: 4675-4682. 115. LeClerc, J. E and Setlow, J. K. 1974. Single-strand regions in the deoxyribonucleic acid of competent H. influenzae. J. Bacteriol. 122:1091-102. 116. Lethem, M. I., James, S. L., Marriott, C. and Burke, J. F. 1990. The origin of DNA associated with mucus glycoproteins in cystic fibrosis sputum. European Respiratory Journal 3:19-23. 117. Lin-Chao, S., Wong, T. T., McDowall, K. J. and Cohen, S. N. 1994. Effects of nucleotide sequence on the specificity of rne-dependent and RNase E-mediated cleavages of RNA I encoded by the pBR322 plasmid. /. Biol. Chem. 269:10797-803. 118. Lorenz, M. G. and Wackernagel, W. 1994. Bacterial Gene Transfer by Natural Genetic Transformation in the Environment. Micro. Rev. 58: 563-602. 119. Lunsford, R. D. 1998. Streptococcal transformation: essential features and applications of a natural gene exchange system. Plasmid 39:10-20. 120. Macfadyen, L. P. 1999. Regulation of intracellular cAMP levels and competence development in Haemophilus influenzae by a phosphoenolpyruvate phosphotransferase system. Ph. D. Thesis. University of British Columbia. Vancouver, Canada. 121. Macfadyen, L. P., Dorocicz, I. R., Reizer, J., Saier, M. H., Jr. and Redfield, R. J. 1996. Regulation of competence development and sugar utilization in Haemophilus influenzae Rd by a phosphoenolpyruvate:fructose phosphotransferase system. Mol. Microbiol. 21: 941-52. 220 122. Macfadyen, L. P., Ma, C. and Redfield, R. J. 1998. A 3',5' cyclic AMP (cAMP) phosphodiesterase modulates cAMP levels and optimizes competence in Haemophilus influenzae Rd. / . Bacteriol 180: 4401-5. 123. Mackie, G. A. 1992. Secondary structure of the mRNA for ribosomal protein S20. Implications for cleavage by ribonuclease E [published erratum appears in /. Biol. Chem. 1992 Apr 25;267(12):8702]. / Biol Chem 267:1054-61. 124. Mathews, D. H., Sabina, J., Zuker, M. and Turner, D. H. 1999. Expanded sequence dependence of thermodynamic parameters improves prediction of RNA secondary structure. J. Mol. Biol. 288:911-40. 125. McCarthy, D. 1989. Cloning of the rec-2 locus of Haemophilus influenzae. Gene 75: 135-43. 126. McCarthy, D. and Kupfer, D. M. 1987. Electron microscopy of single-stranded structures in the DNA of competent Haemophilus influenzae cells. /. Bacteriol. 169: 565-71. 127. McCarthy, J. E. and Brimacombe, R. 1994. Prokaryotic translation: the interactive pathway leading to initiation. Trends Genet. 10: 402-7. 128. McCarthy, J. E. and Gualerzi, C. 1990. Translational control of prokaryotic gene expression. Trends Genet. 6: 78-85. . 129. McDowall, K. J., Lin-Chao, S. and Cohen, S. N. 1994. A+U content rather than a particular nucleotide order determines the specificity of RNase E cleavage. /. Biol. Chem. 269:10790-6. 130. Melefors, O., Lundberg, U. and Von Gabian, A. 1993. RNA processing and degradation by RBNase K and RNase E. In: Control of Messenger Stability. Belasco, J. G. and Brawerman, G. (Ed.). Academic Press Inc., San Diego, California, pp 53-70. 221 131. Michod, R. E. and Wojciechowski, M. F. 1994. DNA repair and the evolution of transformation IV. DNA damage increases transformation. /. Evol. Biol. 7:147-175. 132. Miczak, A., Kaberdin, V. R., Wei, C. L. and Lin-Chao, S. 1996. Proteins associated with RNase E in a multicomponent ribonucleolytic complex. Proc. Natl. Acad. Sci. U S A 93: 3865-9. 133. Miller, J.H. 1992. A short course in bacterial genetics: a laboratory manual and handbook for Escherichia coli and related bacteria. Cold Spring Harbor Laboratory Press, New York, USA. 134. Morett, E. and Bork, P. 1998. Evolution of new protein function: recombinational enhancer Fis originated by horizontal gene transfer from the transcriptional regulator NtrC. FEBS Lett. 433: 108-12. 135. Muskhelishvili, G., Buckle, M., Heumann, H., Kahmann, R. and Travers, A. A. 1997. FIS activates sequential steps during transcription initiation at a stable RNA promoter. Embo f. 16: 3655-65. 136. Muskhelishvili, G., Travers, A. A., Heumann, H. and Kahmann, R. 1995. FIS and RNA polymerase holoenzyme form a specific nucleoprotein complex at a stable RNA promoter. Embo J. 14:1446-52. 137. Nelson, K. E., Richardson, D. L. and Dougherty, B. A. 1997. Tn926 transposition in Haemophilus influenzae Rd: preferential insertion into noncoding DNA. Microb. Comp. Genomics 2: 313-21. 138. Nicholson, A. W. 1996. Structure, reactivity and biology of double-stranded RNA. Prog. Nucleic Acid Res. Mol. Biol. 52:1-65. 139. Nierlich, D. P. and Murakawa, G. J. 1996. The decay of bacterial messenger RNA. Prog Nucleic Acid Res. Mol. Biol. 52:153-216. 222 140. Nilsson, G., Belasco, J. G., Cohen, S. N. and von Gabain, A. 1984. Growth-rate dependent regulation of mRNA stability in Escherichia coli. Nature 312: 75-7. 141. Nilsson, L., Vanet, A., Vijgenboom, E. and Bosch, L. 1990. The role of FIS in trans activation of stable RNA operons of E. coli. Embo J. 9: 727-34. 142. Nilsson, L., Verbeek, H., Hoffmann, U., Haupt, M. and Bosch, L. 1992. Inactivation of the fis gene leads to reduced growth rate. FEMS Microbiol. Lett. 78: 85-8. 143. Nilsson, L., Verbeek, H., Vijgenboom, E., van Drunen, C , Vanet, A. and Bosch, L. 1992. FIS-dependent trans activation of stable RNA operons of Escherichia coli under various growth conditions. /. Bacteriol. Y7A: 921-9. 144. Ninnemann, O., Koch) C. and Kahmann, R. 1992. The E.coli fis promoter is subject to stringent control and autoregulation. Embo J. 11: 1075-83. 145. Notani, N. and Goodgal, S. H. 1966. On the nature of recombinants formed during transformation in Haemophilus influenzae. }. Gen. Physiol. 49:197-209. 146. Osuna, R., Finkel, S. E. and Johnson, R. C. 1991. Identification of two functional regions in Fis: the N-terminus is required to promote Hin-mediated DNA inversion but not lambda excision. Embo J. 10:1593-603. 147. Osuna, R., Lienau, D., Hughes, K. T. and Johnson, R. C. 1995. Sequence, regulation, and functions of fis in Salmonella typhimurium. J. Bacteriol. 177: 2021-32. 148. Palmer, K. L. and Munson, R. S., Jr. 1995. Cloning and characterization of the genes encoding the hemolysin of Haemophilus ducreyi. Mol. Microbiol. 18: 821-30. 149. Petersen, C. 1992. Control of functional mRNA stability in bacteria: multiple mechanisms of nucleolytic and non-nucleolytic inactivation. Mol. Microbiol. 6: 277-82. 223 150. Petersen, C. 1993. Translation and mRNA stability in bacteria: a complex relationship. In: Control of messenger RNA stability. Belasco, J. G. and Brawerman, G. (Ed.). Academic Press Inc., San Diego, California, pp 117-145. 151. Pietrokovski, S. 1996. Searching databases of conserved sequence regions by aligning protein multiple-alignments [published erratum appears in Nucleic Acids Res. 1996 Nov 1;24(21):4372]. Nucleic Acids Res. 24: 3836-45. 152. Pifer, M. L. and Smith, H. O. 1985. Processing of donor DNA during Haemophilus influenzae transformation: analysis using a model plasmid system. Proc. Natl. Acad. Sci. USA 82:3731-5. 153. Pratt, T. S., Steiner, T., Feldman, L. S., Walker, K. A. and Osuna, R. 1997. Deletion analysis of the fis promoter region in Escherichia coli: antagonistic effects of integration host factor and Fis. /. Bacteriol. 179: 6367-77. 154. Ranhand, J. M. and Lichstein, H. C. 1969. Effect of selected antibiotics and other inhibitors on competence development in Haemophilus influenzae. J Gen. Microbiol. 55: 37-43. 155. Redfield, R. J. 1991. sxy-1, a Haemophilus influenzae mutation causing greatly enhanced competence. /. Bacteriol. 173: 5612-5618. 156. Redfield, R. J. 1993. Genes for breakfast: the have-your-cake-and-eat-it-too of bacterial transformation. J Hered. 84:400-4. 157. Reece, K. S. and Phillips, G. J. 1995. New plasmids carrying antibiotic-resistance cassettes. Gene 165:141-2. 158. Richardson, J. P. and Greenblatt, J. 1996. Control of RNA chain elongation and termination. In: Escherichia coli and Salmonella cellular and molecular biology. Neidhardt, F. C. (Ed.). ASM Press, Washington, D.C. pp 822-848. 224 159. Ross, W., Thompson, J. F., Newlands, J. T. and Gourse, R. L. 1990. E.coli Fis protein activates ribosomal RNA transcription in vitro and in vivo. Embo J. 9: 3733-42. 160. Rothstein, D. M., Pahel, G., Tyler, B. and Magasanik, B. 1980. Regulation of expression from the glnA promoter of Escherichia coli in the absence of glutamine synthetase. Proc. Natl. Acad. Sci. USA 77: 7372-6. 161. Rowji, P., Gromkova, R. and Koornhof, H. 1989. Genetic transformation in encapsulated clinical isolates of Haemophilus influenzae type b. / Gen. Microb. 135: 2775-82. 162. Sambrook, J., Fritsch, E. F., and Maniatis, T. 1989. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory Press, New York, USA. 163. Schmid, M. B. 1990. More than just "histone-like" proteins. Cell 63:451-3. 164. Schneider, R., Travers, A. and Muskhelishvili, G. 1997. FIS modulates growth phase-dependent topological transitions of DNA in Escherichia coli. Mol. Microbiol. 26: 519-30. 165. Setlow, J. K., Spikes, D. and Griffin, K. 1988. Characterization of the rec-1 gene of Haemophilus influenzae and behavior of the gene in Escherichia coli. J. Bacteriol. 170: 3876-81. 166. Shanado, Y., Kato, J. and Ikeda, H. 1997. Fis is required for illegitimate recombination during formation of lambda bio transducing phage. /. Bacteriol. 179: 4239-45. 167. Sisco, K. L. and Smith, H. O. 1979. Sequence-specific DNA uptake in Haemophilus transformation. Proc. Natl. Acad. Sci. USA 76: 972-6. 168. Smith, H. O., Danner, D. B. and Deich, R. A. 1981. Genetic transformation. [Review]. Ann. Rev. Biochem. 50: 41-68. 225 169. Smith, H. O., Tomb, J.-R, Dougherty, B. A., Fleischmann, R. D. and Venter, J. C. 1995. Frequency and distribution of DNA uptake signal sequences in the Haemophilus influenzae Rd genome. Science 269: 538-540. 170. Solomon, J. M. and Grossman, A. D. 1996. Who's competent and when: regulation of natural genetic competence in bacteria. Trends Genet. 12:150-5. 171. Springer, M. 1996. Translational control of gene expression in E. coli and bacteriophage. In: Regulation of gene expression in Escherichia coli. Lin, E. C. C. and Lynch, A. S. (Ed.). R.G. Landes Company and Chapman and Hall, Georgetown, Texas. 172. Stanssens, P., Remaut, E. and Fiers, W. 1986. Inefficient translation initiation causes premature transcription termination in the lacZ gene. Cell 44: 711-8. 173. Stewart, G. J. and Carlson, C. A. 1986. The biology of natural transformation. Ann. Rev. Microbiol. 40: 211-235. 174. Stuy, J. H. 1965. Fate of transforming DNA in the Haemophilus influenzae transformation system. /. Mol. Biol. 13: 554-70. 175. Stuy, J. H. 1989. Cloning and characterization of the Haemophilus influenzae Rd rec-2+gene. J. Bacteriol. 171:4395-401. 176. Stuy, J. H. and Walter, R. B. 1986. Effect of glycerol on plasmid transfer in genetically competent Haemophilus influenzae. Mol. Gen. Genet. 203: 296-9. 177. Talkad, V., Achord, D. and Kennell, D. 1978. Altered mRNA metabolism in ribonuclease Ill-deficient strains of Escherichia coli. J. Bacteriol. 135: 528-41. 178. Thompson,}. D., Higgins, D. G. and Gibson, T. J. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22: 4673-80. 226 179. Thompson, J. F., Moitoso de Vargas, L., Koch, C , Kahmann, R. and Landy, A. 1987. Cellular factors couple recombination with growth phase: characterization of a new component in the lambda site-specific recombination pathway. Cell 50: 901-8. 180. Tomb, J. F. 1992. A periplasmic protein disulfide oxidoreductase is required for transformation of Haemophilus influenzae Rd. Proc. Natl. Acad. Sci. USA 89:10252-6. 181. Tomb, J. F., Barcak, G. J., Chandler, M. S., Redfield, R. J. and Smith, H. O. 1989. Transposon mutagenesis, characterization, and cloning of transformation genes of Haemophilus influenzae Rd. J. Bacteriol. 171:3796-802. 182. Tomb, J. F., el Hajj, H. and Smith, H. O. 1991. Nucleotide sequence of a cluster of genes involved in the transformation of Haemophilus influenzae Rd. Gene 104: 1-10. 183. Tonjum, T., Bukholm, G. and Bovre, K. 1990. Identification of Haemophilus aphrophilus and Actinobacillus actinomycetemcomitans by DNA-DNA hybridization and genetic transformation. /. Clin. Microbiol. 28:1994-1998. 184. Travers, A. and Muskhelishvili, G. 1998. DNA microloops and microdomains: a general mechanism for transcription activation by torsional transmission. /. Mol. Biol. 279: 1027-43. 185. Tsui, H. C , Feng, G. and Winkler, M. E. 1997. Negative regulation of mutS and mutH repair gene expression by the Hfq and RpoS global regulators of Escherichia coli K-12. /. Bacteriol. 179: 7476-87. 186. Tzou, W. S. and Hwang, M. J. 1997. A model for Fis N-terminus and Fis-invertase recognition. FEBS Lett 401:1-5. 187. Vytvytska, O., Jakobsen, J. S., Balcunaite, G., Andersen, J. S., Baccarini, M. and von Gabain, A. 1998. Host factor I, Hfq, binds to Escherichia coli ompA mRNA in a growth 227 rate-dependent fashion and regulates its stability. Proc. Natl. Acad. Sci. USA 95: 14118-23. 188. Wagner, L. A., Gesteland, R. F., Dayhuff, T. J. and Weiss, R. B. 1994. An efficient Shine-Dalgarno sequence but not translation is necessary for lacZ mRNA stability in Escherichia coli. J. Bacteriol. 176:1683-8. 189. Walker, K. A., Atkins, C. L. and Osuna, R. 1999. Functional determinants of the Escherichia coli fis promoter: roles of - 35, -10, and transcription initiation regions in the response to stringent control and growth phase-dependent regulation. /. Bacteriol. 181: 1269-80. 190. Wilcox, K. W. and Smith, H. O. 1975. Isolation and characterization of mutants of Haemophilus influenzae deficient in an adenosine 5'-triphosphate-dependent deoxyribonuclease activity. Journal of Bacteriology 122:443-53. 191. Williams, P., Hung, W. L. and Redfield, R. J. 1996. Cell transfer by filtration: evaluation of protocols for transformational competence. FEMS Microbiol Lett 137:183-7. 192. Williams, P. M , Bannister, L. A. and Redfield, R. J. 1994. The Haemophilus influenzae sxy-1 Mutation is in a Newly Identified Gene Essential for Competence. Journal of Bacteriology 176: 6789-6794. 193. Wise, E. M., Alexander, S. and Powers, M. 1973. Adenosine 3':5'-cyclic monophosphate as a regulator of bacterial transformation. Proc. Natl. Acad. Sci. USA. 70: 471-474. 194. Wold, S., Crooke, E. and Skarstad, K. 1996. The Escherichia coli Fis protein prevents initiation of DNA replication from oriC in vitro. Nucleic Acids Res. 24: 3527-32. 228 195. Xu, J. and Johnson, R. C. 1995. Fis activates the RpoS-dependent stationary-phase expression oiproP in Escherichia coli. J. Bacteriol. 177:5222-31. 196. Yarchuk, O., Jacques, N., Guillerez, J. and Dreyfus, M. 1992. Interdependence of translation, transcription and mRNA degradation in the lacZ gene. /. Mol. Biol. 226: 581-96. 197. Yuan, H. S., Finkel, S. E., Feng, J. A., Kaczor-Grzeskowiak, M., Johnson, R. C. and Dickerson, R. E. 1991. The molecular structure of wild-type and a mutant Fis protein: relationship between mutational changes and recombinational enhancer function or DNA binding. Proc. Natl. Acad. Sci. USA 88: 9558-62. 198. Zuker, M., Mathews, D. H. and Turner, D. H. 1999. Algorithms and Thermodynamics for RNA Secondary Structure Prediction: A Practical Guide. In: RNA Biochemistry and Biotechnology. Barciszewski, J. and Clark, B. F.C. (Eds.). Kluwer Academic Publishers, Dordrecht, the Netherlands. 199. Zulty, J. and Barcak, G. 1993. Structural organization, nucleotide sequence, and regulation of the Haemophilus influenzae rec-l+ gene. /. Bacteriol. 175: 7269-7281. 200. Zulty, J. J. and Barcak, G. J. 1995. Identification of a DNA transformation gene required for coml01A+ expression and supertransformer phenotype in Haemophilus influenzae. Proc. Natl. Acad. Sci. USA 92: 3616-20. 229 Date: F r i , 9 Feb 2001 12:41:02 -0800 (PST) From: Lynda Johnston <lyndaj@interchange.ubc.ca> To: lib-cat4@interchange.ubc.ca Subject: Theses Your l i s t s are i n your mailboxes. You w i l l each have 12 or 13. Please r e f e r to your manuals as these are Doctorals and we haven't done them f o r awhile. The Asian L i b r a r y people w i l l get your l i s t s when Rudy comes over and the theses w i l l be mailed to you. Due date: FRIDAY MARCH 2, 2001 (the e a r l i e r they are done the more time Rudy and I w i l l have to solve problems). Date: Mon, 5 Feb 2001 07:01:30 -0800 (PST) From: Lynda Johnston <lyndaj©interchange.ubc.ca> To: lib-cat4@interchange.ubc.ca Subject: Theses Just to l e t you know F a l l 1999 Doctoral and Spring 2000 Doctoral these are here so you can d i g out your theses manuals and your f e l t pens (So you can mark the f o l d e r s ) . The l i s t s won't be comming u n t i l at l e a s t Feb.12th (next Monday) and you w i l l have about 3 weeks to do them and depending on your workload you might have 4 weeks but there are rumours that the F a l l 2000 master theses are looming i n the background so we are t r y i n g to get these ones done. Lynda 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0089846/manifest

Comment

Related Items