UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The TFP ATPase of haemophilus influenzae and DNA uptake Yang, Minghui 2005

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

Item Metadata

Download

Media
831-ubc_2005-0710.pdf [ 7.33MB ]
Metadata
JSON: 831-1.0092266.json
JSON-LD: 831-1.0092266-ld.json
RDF/XML (Pretty): 831-1.0092266-rdf.xml
RDF/JSON: 831-1.0092266-rdf.json
Turtle: 831-1.0092266-turtle.txt
N-Triples: 831-1.0092266-rdf-ntriples.txt
Original Record: 831-1.0092266-source.json
Full Text
831-1.0092266-fulltext.txt
Citation
831-1.0092266.ris

Full Text

THE TFP ATPASE OF HAEMOPHILUS INFLUENZAE AND DNA UPTAKE  by MINGHUI Y A N G  B . S c , Shandong University (P. R. China), 1985 M . S c , Shandong University (P. R. China), 1988  A THESIS SUBMITTED IN P A R T I A L F U L F I L M E N T OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE  in T H E F A C U L T Y O F G R A D U A T E STUDIES (Zoology)  T H E UNIVERSITY OF BRITISH C O L U M B I A October 2005  © Minghui Y a n g , 2005  ABSTRACT  Haemophilus influenzae R d is a Gram negative bacterium capable of taking up exogenous D N A and undergoing genetic transformation through natural competence. A s in most naturally competent bacteria, H. influenzae proteins related to type I V pili are required for natural competence. The pil operon of H. influenzae is composed of four open reading frames, pilA, pilB, pilC and pilD, coding for homologues of proteins involved in the assembly and function of type I V pili.  pilB is the second gene of the pil operon. It codes for the only H. influenzae homologue of Pseudomonas aeruginosa P i l B and P i l T , which are required for Type I V pili assembly and retraction, respectively. The deduced protein sequence of H. influenzae P i l B was analyzed and nucleotide-binding motifs conserved in the V r b B / G s p E family ATPases were identified. Alignment with the homologues in other naturally competent bacteria identified the most conserved regions in the C-termini of proteins in this family that are essential for ATPase activities. However, little is known with respect to the specific role of P i l B in natural competence and D N A uptake. T o find out H. influenzae P i l B ' s function in D N A uptake, a pilB knockout RR1150 was constructed by an insertion at the Bell site using a Tn903 K a n cassette. R  To allow expression of the downstream genes pilC & pilD in this strain, an additional C R E (Competence Regulatory Element) regulatory sequence and promoter were placed in the knockout RR1150, right before pilC. Natural competence was eliminated in RR1150, indicating that pilB is essential for natural competence. D N A binding and uptake assays showed that the defect of competence in RR1150 is at the  ii  level of D N A uptake. T o further investigate the specific function of pilB, the P. aeruginosa PA01 homologues pilB and pilTv/zre, introduced into the above H. influenzae p//J?-background. The two P. aeruginosa genes were strongly expressed in H. influenzae pilB and the sequences have no mutations, as indicated by real time P C R and D N A sequencing. However, D N A binding and uptake in H. influenzae pilB was not restored. The experimental data and the recently available evidence in related studies suggest that complementation failures could be attributed to either the speciesspecific interactions of Tfp proteins, or to the different ways by which the Tfp assembly/function systems work. Furthermore, since PilT-like ATPases were not found in any Pasteurellaceae species possessing type I V pili by B L A S T search and sequence analysis, I suggest that the energetics of Pasteurellaceae Tfp is different from the P i l B - P i l T system of P. aeruginosa Tfp, and that an as-yet-uncharacterized protein provides the driving force required for Tfp retraction in Pasteurellaceae.  iii  TABLE OF CONTENTS ABSTRACT  ii  TABLE OF CONTENTS  iv  LIST OF TABLES  vii  LIST OF FIGURES  viii  x  LIST OF ABBREVIATIONS ACKNOWLEDGEMENTS  xi  CHAPTER ONE  1  Introduction 1.1 Natural competence and transformation  ..2  1.2 Type I V pili and natural competence  4  1.3 Possible mechanism of Tfp mediated D N A binding and uptake  10  1.4 Natural competence of H. influenzae  12  1.5 H. influenzae pil operon and Tfp homologues  12  1.6 H. influenzae Tfp and transformation  14  1.7 Objective and approaches  18  CHAPTER TWO  19  Materials and methods 2.1 General methods  19  2.1.1 Strains, media and plasmids  19  2.1.2 H. influenzae transformation  22  2.1.2.1 Natural transformation of H. influenzae with double-stranded linear DNA..  22  2.1.2.2 Natural transformation of H. influenzae with plasmid D N A  22  2.1.2.3 Transformation of plasmid D N A into chemically-competent H. influenzae cells  23  2.1.3 E.coli plasmid transformation  23  2.1.4 D N A manipulations  24  2.1.4.1 Isolation of plasmid D N A  24  iv  2.1.4.2 Isolation of chromosomal D N A  24  2.1.4.3 G e l electrophoresis, gel extraction, restriction digestion and ligation ... 25 2.1.4.4 P C R amplifications  25  2.1.5 D N A binding & uptake assays  29  2.1.5.1 Total D N A binding and uptake  29  2.1.5.2 D N A binding and/or uptake  29  2.1.6 Real-time P C R  :  30  2.1.7 Nucleotide sequence analyses  31  2.2 Plasmid and strain constructions  31  2.2.1 Construction of H. influenzae plasmids  31  2.2.1.1 p P I L B and p P I L B : : K a n  31  R  2.2.1.2 p P I L C  34  2.2.1.3 p S U P I L B  35  2.2.1.4 p B K a n C R E C  :  37  2.2.2 H. influenzae knockouts  39  2.2.2.1 RR1137 (pilB::Kan )  39  R  2.2.2.2 Construction of RR1150 (A pilB::Kan ICREIpilC)  39  R  2.2.2.3 Construction of RR1136 ( K W 2 0 carrying p S U P I L B ) , RR1138 (pilB::Kan , R  p S U P I L B ) and RR1151 (A  pilB::Kan /CRE/pilC, R  pSUPILB)  40  2.2.3 Introduction of P. aeruginosa P A 0 1 pilB, pilT to the H. influenzae pilB  background  40  2.2.3.1 Cloning of P. aeruginosa pilT.  41  2.2.3.2 Cloning of P. aeruginosa pilB  43  2.2.3.3 Construction of p P A P B T and p P A P B T 2  46  2.2.3.4 Transforming the P. aeruginosa plasmids into H. influenzae pilB  47  CHAPTER THREE  50  Results a n d discussion 3.1 Sequence analysis and functional prediction o f / / , influenzae P i l B  50  3.2 Construction of H. influenzae pilB knockouts  54  v  3.2.1 Overview  54  3.2.2 Construction of p P I L B and p P I L B : : K a n  56  R  3.2.3 Construction of RR1137 (pilB::Kan )  56  R  3.2.4 Complementation assays and real time P C R revealed severe polarity of RR1137 pilB knockout  57  3.2.5 Construction of a non-polar pilB knockout RR1150  59  3.3 Natural competence was eliminated in RR1150  60  3.4 RR1150 pilB knockout was complemented by wild type pilB  60  3.5 Transformation defect of RR1150 is at the level of D N A uptake.,  ;  3.6 Introduction of P. aeruginosa pilB and pilT into H. influenzae pilB  3.6.1 Overview  66 68  68  3.6.2 Cloning of P. aeruginosa pilT  :.  72  3.6.2.1 Construction of p P A P T and p P A P T 2  72  3.6.2.2 Construction of p P A P B  73  3.6.2.3 Construction of p P A P B T and p P A P B T 2  74  3.6.2.4 Sequencing of the P. aeruginosa plasmids in E.coli  74  3.6.3 Construction of H. influenzae strains bearing both the pilB knockout and the P. aeruginosa clones  76  3.6.4 Sequences of the P. aeruginosa plasmids in H. influenzae pilB have no mutations  '.  76  3.6.5 The P. aeruginosa genes were strongly expressed in H. influenzae pilB  77  3.7 The transformation defect of H. influenzae pilB could not be restored by P. aeruginosa pilB and pilT  80  3.8 The defect of D N A binding and uptake of H. influenzae pilB could not be complemented by P. aeruginosa pilB and pilT.  81  3.9 Discussion  82  References  91  vi  LIST OF TABLES Table 1.1 Names given to Tfp & Tfp related proteins in different bacteria  6  Table 1.2 The homologues of the putative non-specific D N A binding protein in different bacteria  11  Table 2.1 Bacterial strains used in this study  20  Table 2.2 Plasmids used in this study  21  Table 2.3 Primers used in this study  27  Table 2.3 Primers used in this study (Continued)  28  Table 3.1 Alignment scores of H.  influenzae P i l B homologues in P. aeruginosa and  N. gonorrhoeae  50  Table 3.2 Expression level of pilC in the non-polar pilB knockout compared to that of the wild type  65  Table 3.3 Hypothesis about complementation of H.  influenzae pilB with P.  aeruginosa P A 0 1 pilB & pilT  Table 3.4 Constructed P.  69  aeruginosa plasmids  Table 3.5 Taxonomic relationships between  71  Haemophilus influenzae, Pseudomonas  aeruginosa, Nesseria gonorrhoeae and Aeromonas hydrophila  vii  71  LIST OF FIGURES Figure 1.1 The  H. influenzae pil operon  2  Figure 1.2 A n overview of natural competence and transformation  3  Figure 1.3 Type I V pili of P. aeruginosa  5  Figure 1.4 Proposed model of the Tfp-like D N A uptake apparatus in H.  influenzae, in  comparison with P. aeruginosa Tfp model  15  Figure 2.1 Construction of p P I L B  32  Figure 2.2 Construction of p P I L B : : K a n  33  R  Figure 2.3 Construction of p P I L C  '.  34  Figure 2.4 Construction of p S U P I L B  36  Figure 2.5 Construction of p B K a n C R E C and the strain R R 1 1 5 0  38  Figure 2.6 Cloning of P A 0 1 pilT and construction of p P A P T  42  Figure 2.7 Cloning of P A 0 1  44  pilB and construction of p P A P B  Figure 2.8 The strategy for P. aeruginosa P A 0 1 pilB cloning  45  Figure 2.9 Construction of p P A P B T and p P A P B T 2  47  Figure 2.10 Construction of H.  influenzae strains containing both pilB knockout and  P. aeruginosa pilB, pilT plasmids  48  Figure 2.11 Sequencing of P. aeruginosa P A 0 1 pilB in p P A P B Figure 2.12 Sequencing of P.  aeruginosa PA01 pilT  Figure 2.13 Sequencing of P. aeruginosa P A 0 1 pilB  49  in p P A P T  49  and pilT in p P A P B T and  pPAPBT2  Figure 3.1a Residues 251-400 of H.  49 influenzae P i l B  51  Figure 3.1b A m i n o acid alignment of C-terminal segments of P A 0 1 P i l B , PilT homologues  52  Figure 3.2 Genetic organization of K W 2 0 , RR1137 and RR1150  '.  55  Figure 3.3 Transformation tests of the pilB knockout in RR1137 complemented with wild type pilB  58  Figure 3.4 Construction of RR1150  61  Figure 3.5 Transformation tests of the pilB knockout in R R 1 1 5 0 complemented with wild type pilB  62  viii  Figure 3.6 The expected and observed sequences around the - 1 0 region of  pil  promoter in the w i l d type and in p P I L B , p S U P I L B and p P I L C  64  Figure 3.7 Relationships between the plasmids of this study and the strain RR1150  65  Figure 3.8 D N A binding and uptake of RR1150  67  Figure 3.9 Graphic presentation of the two putative coding sequences of P. aeruginosa pilB and pilT  71  Figure 3.10 The strains and plasmids with or without the -10 substitution  72  Figure 3.11 The sequence of p P A P B #6 at the ligation joint  75  Figure 3.12 Expression level of P.  aeruginosa pilB in the H. influenzae pilB  background  Figure 3.13 Expression level of P.  78 aeruginosa pilT'm the H. influenzae pilB  background  79  Figure 3.14 Transformation assays of the P.  aeruginosa pilB, pilT complemented H.  influenzae pilB  Figure 3.15 D N A binding assays of the P.  81 aeruginosa pilB, pilT complemented H.  influenzae pilB  82  ix  LIST O F ABBREVIATIONS  Amp  ampicillin  BHI  brain heart influsion (rich culture medium)  cDNA  complementary D N A  Cam  Chloramphenicol  cpm  counts per minute  dATP  deoxyadenosine 5'-triphosphate  dCTP  deoxycytocine 5'-triphosphate  dGTP  deoxyguanine 5'-triphosphate  DNA  deoxyribonucleic acid  DNase I  deoxyribonuclease I  dNTPs  deoxyribonucleoside 5'-triphosphate  dTTP  dioxythymidine 5'-triphosphate  EDTA  ethylenediaminetetraacetic acid  E value  expected value  IPTG  isopropyl-|3-D-thiogalactopyranoside  Kan  Kanamycin  LB  luria Bertani broth  MIV  "M-four", starvation medium for H. influenzae competence induction  mRNA  messenger R N A  NAD  nicotinamide adenine allele  NOV  Novobiocin  PA01  Pseudomonas aeruginosa P A 0 1  PCR  polymerase chain reaction  RNA  ribonucleic acid  RT-PCR  reverse transcriptase P C R  sBHI  B H I supplemented with hemin and N A D  Tfp  Type I V pili  Tris  Tris(hydroxymethyl)aminomethane  Xgal  5-bromo-4-chloro-3-indolyl-|3-D-galactoside  ACKNOWLEDGEMENT  I would like to express my gratitude to my supervisor, Dr. Rosemary Redfield, for her encouragement, guidance and financial support. Her expertise, vast knowledge, and work attitude added considerably to my graduate experience. I also appreciate her understanding, patience and assistance in writing reports (i.e. proposals, progress reports and this thesis). I would like to thank the other members of my committee, Dr. Patricia Schulte and Dr. Rachel Fernandez, for their counsel and assistance.  Special thanks goes out to Andrew Cameron, for his numerous help and the constructive discussions. He was always a source of information, a listening ear as well. I want to thank all of the current and former members of the Redfield lab for their help and friendship over the last few years. I had a great time with them.  I am grateful to my family, which is a great source of courage and venting of frustration during my graduate program. This study could not have been finished without their support.  xi  Chapter 1: Introduction  Haemophilus influenzae has evolved the ability to take up D N A from the extracellular environment using a genetically regulated process called natural competence. A s in most naturally competent bacteria (6, 12, 26, 58), proteins related to cell-surface filaments called type I V pili (Tfp) are required for D N A uptake in H. influenzae (23). The pil operon of H. influenzae contains genes coding for homologues of important Tfp proteins (Fig. 1.1), and previous studies have shown that this operon is essential for natural competence (23). However, the specific functions of these proteins in D N A uptake remain to be elucidated.  This study focused on the second gene of the pil operon, pilB. The objective was to investigate the specific function of H. influenzae P i l B in D N A binding and uptake. This is important because: (1) Expression of type I V pili has not been demonstrated in H. influenzae R d , despite the presence of Tfp in another H. influenzae strain 86-028NP (7). Type I V pili have been identified in many Gram negative bacteria including competent and non-competent bacteria. Most naturally competent bacteria express Tfp on the cell surface (except A. actinomycetemcomitans) (2, 26, 58). The presence of Tfp proteins but no protruding type I V pili suggests that the H. influenzae pil operon may have evolved to encode a D N A uptake apparatus. (2) Unlike Pseudomonas aeruginosa, Pseudomonas stutzeri, and Nesseria gonorrhoeae, whose Tfp systems are well studied, P i l B is the only Tfp-related ATPase of the G s p E family proteins in H. influenzae (4, 17, 32, 37, 66, 98). Studies on P i l B ' s function may provide insight into the specific mechanism of H.  1  influenzae'^ D N A uptake, and help as well to understand the evolutionary role of the Tfplike machinery in H. influenzae.  H10300  CRE  H10295  ^JH10299 H10298  ampD  '|'«tt>|:< pilA pilB  pilC  pilD  F i g . 1.1 The H. influenzae pil operon.  1.1 Natural competence and transformation Natural competence, defined as the ability to take up exogenous D N A from the extracellular milieu using a genetically programmed process, plays a central role in bacterial horizontal gene transfer. A s a general consequence of competence, natural transformation happens when D N A uptake is followed by recombination of the incoming D N A into the bacterial chromosome, and may result in a heritable change of genotype. Naturally transformable bacteria have been found in at least 40 species, distributed through all taxonomic groups (12).  The mechanism of natural competence has been intensively studied in some model organisms, including the Gram-negative bacteria Neisseria gonorrhoeae,  Pseudomonas  2  (6, 12).  N. gonorrhoeae and H. influenzae and their relatives share the property that  they preferably take up D N A containing a specific uptake signal sequence ( U S S ) that is distributed in their genomes in high frequency (20, 27, 92).  Other transformation  systems do not display such specificity.  The process of natural transformation can be envisaged as several sequential steps (Fig. 1.2): (1) D N A binds to the cell surface.  (2) taken up into the periplasm, where D N A is  inaccessible to exogenous DNase I (known as the DNase I resistant state), through the inner membrane and,  (3) passing  (4) sometimes foreign D N A recombining with any  chromosomal homologue on the chromosome (71).  Fig. 1.2 A n overview of natural competence and transformation.  ® D N A binding to the  putative receptor. © Initial uptake across outer membrane. (D D N A translocation across inner membrane. © Chromosomal displacement through homologous recombination.  3  1.2 Type IV pili and natural competence Type I V pili are filamentous appendages constituted of pilin subunits on the surface of bacterial cells, mediating adherence and twitching motility, a flagella-independent bacterial translocation over moist surfaces (58, 62). Twitching motility is achieved through the extension, tethering, and then retraction of type I V pili into the bacterium, thereby pulling the cells along. A schematic overview of a P. aeruginosa type I V pilus is shown in Fig. 1.3.  Tfp and Tfp-related proteins can be assigned to distinct groups including pilin and pilin proteins, ATPases, prepilin peptidases, secretins and other assembly factors (6, 12) (Table 1.1).  4  NDP+Pj  NDP+Pj  Fig. 1.3 Type I V pili of P. aeruginosa. After processing by the pre-pilin peptidase P i l D , the P i l A subunits associate via their hydrophobic stems to form a pilus with the help of P i l B and possibly other assembly factors. Thereby, the core of the pilus forms a continuous hydrophobic layer with the inner membrane. The assembled pili penetrate the outer membrane through a gated pore formed by the multi-subunit complex of P i l Q . The retraction is driven by P i l T via depolymerization of the pilus fibers.  5  Table 1.1 Names given to Tfp & Tfp related proteins in different bacteria Tfp proteins  Neisseria gonorrhoeae Bacillus subtiBs Haemophilus influenzae  Pseudomonas aeruginosa  Pilin (pili subunit) & Pilin like proteins (various functions)  PilE, ComP, PilEl, PilV, etc. PilA  Nucleotide binding proteins (generate forces)  (1) PilF (for Tfp biogenesis) (2) PilT (for Tfp retraction) (3) PilU (doesn't affect competence)  PilB  ComGA  (1) PUB (for Tfp biogenesis) (2) PilT (for Tfp retraction) (3) PilU (required for twitching motility)  Prepilin peptidase (prepilin processing) Tfp assembly factor  PilD  PilD  ComC  PilD  PilG  PilC  ComGB  PilC  Secretin (forms gated pore on outer membrane)  PilQ®  CornE (different fromcomEl)  None  PilQ  Pilitipprotein (adhesion)  PilC  None  ?  ?  Other Tfp assembly factors  PilM,N,0,P  2  ComA,B,C,D  CornGC, ComGD, CornGE, ComGG (Minor similarity to Pilin subunits)  PilAI,PilAn  PilMKO.P  Pilin proteins are the subunits of the actual type I V pilus (or pseudopilus). A type I V pilus is primarily made up of the major pilin, a single small protein subunit, usually termed P i l A or pilin (58). The common feature of Type I V pili from a wide range of Gram negative bacteria is the sequence similarities among the major pilins (29), which are highly conserved in the N-terminal region including a short positively charged leader sequence and a hydrophobic domain. In addition to the major pilin, some Tfp systems have several pilin-like proteins called minor pilins, which might be incorporated as well in the pilus fiber (12, 58,63).  Pilins are all synthesized as prepilins, whose leader sequence is removed by the prepilin peptidase P i l D to produce pilins. P i l D is a bifunctional enzyme carrying out both this cleavage and yV-methylation of the pilin subunit (85). It also cleaves a wide variety of substrates with prepilin-like leader sequences, including those involved in protein secretion (52).  The nucleotide-binding proteins are ATPases that use energy from A T P hydrolysis to drive the assembly or retraction of Tfp. They belong to a large superfamily of proteins, the V i r B / G s p E family, which function not only in the Tfp system but also in Type II and Type IV protein secretion (30,49, 63). Proteins in this family are characterized by their nucleotide-binding motifs in the C-terminal region and by their peripheral (e.g. not integral) membrane localization. Multiple Tfp-related ATPases have been identified in most Tfp systems, including the well-studied Tfp systems of P. aeruginosa and N. gonorrhoeae.  In P. aeruginosa, 3 Tfp related ATPases have been identified. P i l B  powers assembly of the pilus (90), while PilT is required for pilus retraction (94). Both  7  are essential for D N A uptake in naturally competent bacteria. The third ATPase, P i l U , is required for twitching motility, but P i l U mutants are competent for D N A uptake in N. gonorrhoeae (66).  Secretin proteins (PilQ) form the polymeric gated pore on the outer membrane, allowing passage of Tfp (41). Electron microscopic analyses revealed that P i l Q forms a dodecameric ring with a central cavity of 5-6.5 nm in diameter, which is wide enough for passage of an intact type IV pilus (~ 6 nm in diameter) and could easily accommodate the D N A double helix (~2.4 nm) (6,10, 12, 16). Therefore current model postulates that transforming D N A is taken up through the secretin pore left by depolymerization of the Type IV pilus (6, 26).  Other outer or inner-membrane proteins are also required for Tfp assembly and functions (e.g. P i l C , P i l M ) (62). But little is known about their specific functions.  Tfp related functions are mainly mediated by the process of pilus elongation and retraction. Elongation is by subunit addition to the base of the helical array of pilin subunits, and retraction is thought to involve subunit removal (58, 62). In P. aeruginosa (Fig. 1.3), after the prepilin signal peptides are cleaved by the prepilin peptidase P i l D , the P i l A subunits associate with each other to form a pilus through interactions between the conserved N-terminal a-helices, with the help of P i l B and other assembly factors (47). A continuous hydrophobic core is formed in the pillus filament and maximizes contact between subunits to provide extreme mechanical strength (18). The assembled pili penetrate the outer membrane through the gated pore formed by the multi-subunit  8  complex of secretin P i l Q (10). The retraction is driven by the ATPase P i l T via depolymerization of the pilus fibre at its base (60).  Type I V pili systems are evolutionally related to type II secretion systems (30, 63, 78). A subset of the components of type II secretion system share extensive sequence similarity with Tfp pilins and other Tfp proteins required for biogenesis of the type I V pilus (including P i l B , P i l C , P i l D , PilQ) (63). The pseudopilins of type II secretion systems are homologous to Tfp pilins, and are thought to function in secretion by forming a periplasmic pseudopilus, which might work as a piston, pushing specific substrates from the periplasm through the secretin pore in the outer membrane (41, 72). When the genes of the Klebsiella oxytoca pullulanase (Pul) secreton (a type II secretion system) were expressed in E. coli K 1 2 cells at high levels, a bundled pilus-like structure composed of the pseudopilin P u l G could be formed on the cell surface (81). Furthermore, in P. aeruginosa, the secretin P i l D is shared between type II secretion and type I V piliation system (84). Because of the similarities of the two systems, knowledge of Tfp systems have been extrapolated to understand the structure and function of Type II secretion systems, and vice versa (63, 67).  Significant involvement of Tfp-related proteins in D N A binding & uptake has been found in most of the naturally competent bacteria studied so far (6, 12, 58), except Helicobacter pylori, which takes up D N A through an unrelated type I V secretion system (82), and the related Campylobacter jejuni, whose D N A uptake uses a type II secretion system (95). In the bacteria requiring Tfp proteins for competence, mutants failing to express Tfp  9  homologues are unable to bind or take up D N A (26). The different names used for Tfp related proteins in different bacteria are listed in Table 1.1.  1.3 Possible mechanism of Tfp mediated DNA binding and uptake Tfp proteins and homologues appear to function in the first steps of transformation, e.g. D N A binding and uptake. Experimentally this is measured by the amount of D N A bound to the cell surface and taken up into a DNase I resistant state, respectively (2, 26).  The specific role of Tfp in D N A uptake remains to be elucidated. However, based on the structure model of Tfp (X-ray diffraction of pilus assembly, electrostatic surface analysis) and the experimental data obtained from the model Tfp systems of N. gonorrhoeae and P. aeruginosa (18, 19, 47, 91), possible mechanisms of Tfp-mediated D N A binding and uptake in Gram negative bacteria have been proposed as follows (taking the P. aeruginosa Tfp proteins as an example):  Electrostatic surface analysis of modeled pilus fibers of P. aeruginosa pilin monomers suggested that a band of positive charge may be a common feature of all type I V pili (18, 47, 91). These positively charged patches that coil around the Tfp fibres provide high affinity for double-stranded D N A to bind non-specifically to the cell surface (2, 47, 65). Recently, van Schaik et al. showed that P. aeruginosa Tfp can bind D N A nonspecifically and that the D N A binding appeared to be a function of the intact pilus (91). In N. gonorrhoeae and H. influenzae, whose D N A uptake prefer D N A with U S S , an asyet-unidentified sequence-specific receptor is also required (12). In N. gonorrhoeae, sequence-specific binding of D N A to the cell can be modulated by the expression of  10  different minor pilins (1, 2). Once D N A  is selected by the sequence-specific receptor, the  secretin channel is opened allowing passage of DNA. Type I V pilus does not have space for a D N A (47), the transforming D N A  Since the predicted structure of  channel in the center of the pilus (~ 10A)  is thought to be bound to the surface and taken up through  the secretin pore as the pilus retracts by pilus depolymerization (26). Retraction of pili is powered by P i l T , which provides energy through hydrolysis of A T P However, PilT may not be directly involved in D N A  DNA  uptake also involve the periplasmic D N A  (33, 39, 54, 60).  uptake (12, 37).  receptor (12), known in N. gonorrhoeae  as C o m E (Table 1.2) (13), which has homologues or partial homologues in both Gram positive and negative bacteria including B. subtilis and H. influenzae (43, 74). C o m E is a periplasm protein with D N A of D N A  binding activity. Chen and Dubnau proposed that binding  to the cell surface triggers the opening of the secretin channel in the outer  membrane and C o m E could also mediate this process (12). Once the transforming is taken up into the periplasm, C o m E delivers the D N A membrane and D N A  DNA  to the channel in the inner  is then translocated to the cytoplasm.  Table 1.2 The homologues of the putative non-specific D N A  binding protein in different  bacteria N a t u r a l competent bacteria putative non-specific binding protein  DNA  Neisseria gonorrhoeae (meningitidis) ComE  Haemophilus influenzae  Bacillus subtilis  ComEl  ComEA  11  1.4 Natural competence of H.  influenzae  H. influenzae becomes competent either when the cells enter stationary phase (spontaneous competence) or when cells are transferred from rich medium into starvation medium (induced competence) (35, 45, 71). The natural competence of H. influenzae is a tightly regulated progress. Most of the competence genes identified so far are under the transcriptional control of the Competence Regulatory Element ( C R E ) , which promotes transcription in response to depletion of energy or nucleotide pools (74).  1.5 H. influenzae  pil operon and Tfp homologues  Like most naturally competent bacteria, H. influenzae encodes proteins homologous to type IV pili (23). Since Tfp-mediated D N A uptake has been identified in most naturally competent bacteria, this raises the question whether H. influenzae's D N A uptake mechanism is similar to that of other bacteria.  The pil operon contains four open reading frames (pilA, pilB, pilC, pilD), which code for Tfp homologues (Fig. 1.1). L i k e most H. influenzae competence induced genes, a C R E i  consensus sequence is found 61.5 bp upstream of the transcription start of the operon (53).  A B L A S T search showed that P i l A is a prepilin peptidase dependent protein  homologous to the P. aeruginosa & N. gonorrhoeae Tfp subunits P i l A (PilE) (See Table I. 1 for different names of Tfp proteins). A l l these three proteins contain a signal sequence and a conserved region at the N-terminal (22, 59, 83, 99, 100). P i l B , a protein with nucleotide binding motifs, is a homologue of both P i l B (PilF) and P i l T , paralogous P. aeruginosa & N. gonorrhoeae proteins with putative ATPase functions (32, 60, 98).  12  The other two proteins, P i l C and P i l D , are homologues of P. aeruginosa& N. gonorrhoeae Tfp biogenesis factors P i l C (PilG) and prepilin peptidase P i l D respectively (25,48, 64, 89).  Previous studies showed that the H. influenzae pil operon is essential for natural competence. Insertions into the upstream regulatory region were found to eliminate competence (23). In addition, all the four genes of the pil operon are induced 100 - 150 fold as other competence genes do when the cells become competent (74). Mutations have been constructed in our lab by insertion of a m i n i - T n / 0 transposon into pilA and into the C R E binding site of the operon (Lab data: CM#889, CM#893). Transformation experiments showed that the pilA mutation eliminated natural competence, and insertion into the C R E site reduced the transformation frequency to background level.  In addition to the four Tfp genes in the pil operon, H. influenzae also has genes coding for homologues of most other Tfp-Iike proteins identified in other bacteria so far, including comA, comB, comC, comD, comE and HI0366 (coding for homologues of P.  aeruginosa P i l M , P i l N , P i l O , PilP, P i l Q and PilF) (31, 55, 56). P. aeruginosa, pilQ codes for the secretin protein that forms oligomeric rings in the outer membrane allowing passage of Tfp. The other genes pilM, pilN, pilO, pilP and pilF of P. aeruginosa are also  required for biogenesis of Type I V pili (56, 93), although the specific functions of these genes are not known. In H. influenzae, M i n i - T n i O insertions in the pilM and pilQ homologues comA and comE (different from comEl) prevent D N A binding & uptake and abolish transformation (88). These findings suggest that H. influenzae may share similar  13  mechanisms in D N A binding and uptake with other Gram negative bacteria that are naturally competent.  1.6 H. influenzae Tfp and competence Despite the existence of a general model for D N A binding & uptake in Gram negative bacteria, comparison with the Tfp and competence of N. gonorrhoeae and other bacteria reveals some atypical features of H. influenzae Tfp and competence.  The first is that no visible type IV pili have been identified on the cell surface of H. influenzae, unlike most Gram negative bacteria capable of D N A uptake, including N. gonorrhoeae, N. meningitidis, P. stutzeri and Acinetobacter sp. B D 4 1 3 . I hypothesize that in H. influenzae the Tfp-like machinery is maintained specifically for D N A uptake. Previous studies have shown that D N A uptake in N. gonorrhoeae requires only small amounts of pilin rather than long pilus fibers (33, 51). Furthermore, substantial increases in transformation efficiency over an isogenic, nonpiliated mutant N. gonorrhoeae were observed when limited amounts of three of the pilin variants were expressed (51). It was proposed that, although Tfp proteins are required in D N A uptake, fully assembled type IV pili may not be necessary (12). Therefore I hypothesize that, the Tfp proteins in H. influenzae form Tfp-like assemblies or pseudopili within the periplasm, as was also proposed by Chen and Dubnau (12). The pseudopili that serve as the D N A uptake machinery would need to be long enough to, penetrate the outer membrane through the pore constituted by C o m E (PilQ in Neisseria) on the outer membrane (Fig. 1.4). The  14  retraction of the pilus-like fiber within the periplasm would then allow the translocation of D N A across the outer membrane.  source?  Fig. 1.4 Proposed model of the Tfp-like D N A uptake apparatus in H. influenzae, in comparison with P. aeruginosa Tfp. After processed by the pre-pilin peptidase P i l D , the P i l A subunits associate with their hydrophobic stems to form a pilus-like fibre with the help of P i l B and possibly other assembly factors. The assembled pili-like fibres of H. influenzae are only long enough to penetrate the outer membrane through a gated pore formed by the polymeric complex of C o m E . Driven by P i l B , or perhaps another as yet unidentified protein, the retraction of the fibre through depolymerization of the P i l A subunits, allows D N A to be taken up into the periplasm.  15  The second unusual feature of H. influenzae'^ Tfp genes is that only a single gene pilA encodes a Tfp prepilin protein (26). Multiple pilin proteins have been identified in most naturally competent bacteria containing Tfp. The pilus is primarily composed ofthe major pilin, while minor pilins might also be present in the pilus assembly (12, 26). In naturally competent bacteria, minor pilins have various effects on D N A uptake. In N. gonorrhoeae, the Tfp structural protein P i l E (major pilin) is required for non-specific D N A binding. A homologous protein, ComP, is involved in sequence-specific D N A binding (99). A third protein, P i l V , whose mutation causes a hypertransformable phenotype, acts as an inhibitor of C o m P function (1). P. stutzeri also has a major pilin gene piNl  that is required for D N A uptake and a minor pilin gene piNW that is  antagonistic to transformation (36). In H. influenzae, however, the only other two pilinrelated genes (HI0938, HI0939) code for PulG-like proteins that are more similar to the pseudopilins in Type II secretion systems, than to Tfp pilins (61).  The third unusual feature of H. influenzae''s competence genes is the role of C o m E l , the protein thought to be functionally equivalent to N. gonorrhoeae C o m E . A s mentioned in Section 1.2.3, Neisseria C o m E and its homologues are periplasmic proteins capable of D N A binding and have been proposed to be the periplasmic D N A receptor (12, 26). The first identified protein of this family, B. subtilis C o m E A , is essential for D N A binding (43). In N. gonorrhoeae, a 4 x 10" fold reduction in transformation frequency was 4  reached when all four copies of the comE genes were deleted (13). In H. influenzae, however, C o m E A only shares homology at the C-terminal region with Neisseria C o m E ,  16  as indicated by a B L A S T search, and deletion of comEA only resulted in a 10 fold decrease in transformation frequency (lab data CM#957).  The absence of a second Tfp related GspE-like ATPase is the fourth unique feature of H. influenzae's competence. A s mentioned in Section 1.2, three Tfp-related ATPases P i l B , PilT and P i l U have been found in the well studied Tfp systems of N. gonorrhoeae, P. aeruginosa and P. stutzeri (17, 32, 37, 60, 66). Both P i l B and PilT are required for D N A uptake in N. gonorrhoeae.  P i l B homologues are known to function in Tfp biogenesis  (32, 90). PilT is responsible for Tfp retraction/depolymerization, through which D N A is thought to be taken up in naturally competent bacteria (98). Our B L A S T searches with the above well-characterized Tfp-related ATPases showed that P i l B is the only Tfprelated homologue of this family in H. influenzae. When searching H. influenzae's genome with P. aeruginosa P A 0 1 P i l B and PilT, no other protein was found with E values lower than 1.1 and 0.36. The possible explanations are discussed below.  Taken together, although Tfp proteins are required for natural competence in H. influenzae, the detailed mechanism of Tfp-mediated D N A uptake is unknown. Since there is only one homologue of the Tfp-related ATPase P i l B in H. influenzae, the possible explanations would be that either P i l B is required only for Tfp assembly, or functions for both Tfp biogenesis and D N A uptake via Tfp retraction. Therefore in the former case, an as-yet-uncharacterized protein, other than a PilT homologue, would power the retraction of the Tfp-like machinery, using either A T P or other energy source (e.g. proton motive force) or both.  17  1.7 Objective and approaches Based on the above considerations, pilB was chosen as the focus of this study. The predicted protein sequence of P i l B was analyzed and conserved nucleotide-binding motifs in the V r b B / G s p E family ATPases were identified (49). The sequence was aligned with P i l B , and PilT homologues, showing the most conserved C-terminal region in this family (see Section 1.2). T o experimentally investigate the specific function of P i l B in D N A uptake, a pilB knockout was constructed by insertion of a Tn905 K a n  R  cassette at the Bell site. T o examine the polarity of the knockout, a w i l d type pilB was introduced into the strain. The transformation capacity of the complemented pilB were tested. Real time P C R was used to examine the expression of the downstream genes. Subsequently, transformation assays and D N A binding & uptake experiments were performed to test the effect of the knockout on competence. T o investigate H. influenzae P i l B ' s specific function, plasmids containing either P. aeruginosa P A 0 1 pilB or pilT, or both, under control of the C R E and the promoter of the H. influenzae pil operon, were constructed and introduced into the H. influenzae pilB knockout. The sequence and the expression of P. aeruginosa pilB and pilT'm the knockout cells were verified by D N A sequencing and real time P C R , respectively. The complemented pilB knockout strains, thus constructed, were used to carry out transformation tests and D N A binding & uptake assays.  18  Chapter 2: Material and methods  2.1 General methods 2.1.1 Strains, media and plasmids The bacterial strains used in this study are listed in Table 2.1. A l l H. influenzae strains are descendants of the original R d strain (3, 96). H. influenzae strains were routinely grown in sBHI medium (Brain Heart Infusion supplemented with 10 u,g/ml hemin and 2 p,g/ml N A D ) at 37°C and shaken at 200 rpm (70). T o promote aeration, the culture volume did not exceed 20% of the flask capacity. When growing strains with plasmids or cassettes, sBHI was supplemented with novobiocin (Nov, 2.5 u,g/ml), kanamycin (Kan, 7.0 u,g/ml), or chloramphenicol (Cam, 1.0 u,g/ml). M I V starvation medium was used for preparation of competent cells (40). Transformed H. influenzae, cells were plated on sBHI agar with or without addition of appropriate antibiotics. The E. coli D H 5 a , T o p l O (Inv itrogen) and G M 2 1 6 3 ( N E B ) strains, grown in L B , were used for propagating plasmids. When necessary, the following antibiotics were used for E. coli: ampicillin (Amp) 100 u,g/ml; chloramphenicol (Cam) 20 u,g/ml; kanamycin (Kan) 10 p,g/ml. P. aeruginosa P A 0 1 was grown at 37°C in low salt L B (0.05% NaCl).  The plasmids used in this work are listed in Table 2.2. p G E M - T easy vector (Promega) was used for cloning of H. influenzae pilB. pWJC3 is the source of the Tn903 K a n  R  cassette (14). pSU20 was used as a shuttle vector between E. coli and H. influenzae (57).  19  Table 2.1 Bacterial strains used in this study  Strain  Genotype  Source o r Reference  H. influenzae KW20  wild-type  (3)  RR1137  pilB::Kan  This study  RR1136  KW20, pSUPILB  This study  RR1138  pilB::Kan ,  This study  RR1150  ApilB::Kan /CRE/pilC  RR1151  ApilB::Kan /CRE/pilC,  pSUPILB  This study  RR1156  t\pilB::Kan /CRE/pilC,  p P A P T (Pa pilT)  This study  RR1157  ApilB::Kan /CREIpilC,  p P A P B (Pa pilB)  This study  RR1158  ApilB::Kan /CRE/pilC, R  p P A P B T (Pa pilB,  This study  RR1159  ApilB::Kan /CRE/pilC,  p P A P T 2 (Pa pilT)  This study  RR1196  ApilB::Kan /CRE/pilC,  p P A P B T 2 (Pa pilB,  This study  R  R  pSUPILB  This study  R  R  R  R  Pa pilT) R  R  Pa pilT)  E. coli DH5a  supE44 recAl  (38)  GM2163  dam dcm  NEB  ToplO  / a c Z A M 1 5 , endkA, deoR, hsdR, mcrA,  Invitrogen  recAl Pa: P. aeruginosa P A O l  20  Table 2.2 Plasmids used in this study. Plasmid  Vector  pPILB pPILB::Kan  p G E M - T easy  Insert  pSUPILB  pSU20  P  pPILC  p G E M - T easy  P  p G E M - T easy  P  s  Construction of RR1137  This study  + pilA + pilB with K a n + part of pilC  Construction of RR1137  This study  + pilA + pilB + part of pilC  Polarity examination  This study  + pilC + part of pilD  Construction of R R 1150  This study  Construction of RR1150  This study  pi  P  pBKan(CREQ  Source or Reference  P + pilA + pilB +  p G E M - T easy  R  Application part of pilC R  7  pil  7  pil  + pilA + truncated pilB with K a n  R  + (P  pil  +  pilC)  5  pPAPB  pSU20  Hi P  pil  + Pa pilB  Complementation of RR1150  This study  pPAPT  pSU20  Hi P  pil  + Pa piW (Starts at A T G )  Complementation of RR1150  This study  pPAPBT  pSU20  Hi P  pil  + Pa pilB + H i P  Complementation of RR1150  This study  pPAPT2  pSU20  Hi P  pPAPBT2  pSU20  Hi P  pWJC3  N/A  Kan  pSU20  N/A  P G E M - T easy  N/A  pil  + Pa pilT (Starts at  ATG) p ; 7  + Pa pilB (Starts at T T G )  Complementation of RR1150  This study  pil  + Pa pilB + H i P  Complementation of RR1150  This study  (14)  N/A  Source of Tn90? K a n cassette Shuttle vector between H.  N/A  Cloning vector  R  pil  + Pa pilT (Starts at T T G )  R  (57)  influenzae and E. coli  Promega  N / A : Not applicable. P : C R E promoter of H. influenzae pil operon. H i : H. influenzae Pa: P. aeruginosa pU  21  2.1.2 H. influenzae  transformation  2.1.2.1 Natural transformation of H. influenzae with double-stranded linear DNA Standard procedures for H. influenzae natural transformation have been described by Goodgal (35). The cells were grown as described above. A t an O D  6 0 0  nm of 0.2-0.3, 10-  12 ml of culture were washed with and transferred to M I V medium in which they were shaken for 100 min at 37°C at lOOrpm to induce competence (40, 97). M A P 7 D N A , the genetically marked D N A from the strain M A P 7 , was used for standard transformation assays (71). After addition of 1.0 ng/ml M A P 7 D N A or other D N A needed to 1 ml of cells, the cells were shaken at 37°C for 15-30 min, then diluted and plated on s B H I agar. Novobiocin was used as the antibiotic marker in transformation tests unless otherwise noted. When selecting for chloramphenicol resistance, the transformation mixture was diluted with 2 volumes of s B H I and incubated at 37°C for 90 minutes to allow expression of the resistance gene (71). Transformation frequency was defined as the number of transformants per milliliter divided by the total number of cells per milliliter and was measured using an excess of chromosomal D N A (lfxg to 1 ml of competent cells).  2.1.2.2 Natural transformation of H. influenzae with plasmid DNA The method for transforming a plasmid into H. influenzae cells was described by Stuy (86) and performed with minor modifications. After incubation of 1-3 \ig plasmid D N A with 1 ml of M l V - i n d u c e d competent cells for 30 min at 37°C, glycerol was added to a final concentration of 30-32%. The cells were incubated at room temperature for 10 min, collected by centrifugation, resuspended in 1.5 ml of sBHI, and plated as usual. For  22  chloramphenicol selection, the cells were incubated at 37°C for 90 minutes before plating (71).  2.1.2.3 Transformation of plasmid DNA into chemically-competent H. influenzae cells The method used for making H. influenzae cells chemically competent was modified from Barcak et al. and W i l l i a m s et al. (8, 97). A t an O D ^ of 0.30 - 0.33, cells in 10 m l of culture were washed with 10 m l of 25 m M cold C a C l and collected by filtration. The 2  cells were resuspended in 2 m l of 75 m M ice-cold C a C l , of which 1 ml of cells was 2  pelleted again, and resuspended in 100 u,l of 75 m M ice-cold C a C l . T o the competent 2  cells thus prepared, 1.0-2.0 p,g of plasmid was added and the mixture was incubated on ice for 30 min. Heat-shock was performed at 37°C for 3 min, followed by addition of 1 ml of sBHI to the cells, which were then shaken at 37°C for 1.5 hours and plated on sBHI agar.  2.1.3  E. coli  plasmid transformation  E. coli competent cells were prepared using the filtration method described by Williams et al. (97). Once grown in L B broth to an O D  6 0 0  o f ~0.375, 10 ml of the cells was  collected by filtration, washed once with 10 ml of 100 m M ice-cold C a C l and 2  resuspended in 2 ml of 100 m M ice-cold C a C l . The cells were used immediately for 2  transformation. Frozen stocks of competent cells were prepared by adding 0.5 ml of 80% glycerol to 2 m l of above cell suspension containing 100 m M ice-cold C a C l , dividing 2  23  into 150  [il aliquots and storing at - 8 0 ° C . Transformation of E. coli was done by the  standard method described by Sambrook et al. (77).  2.1.4 DNA manipulations  2.1.4.1 Isolation of plasmid DNA Plasmid D N A was isolated and purified from  E. coli or H. influenzae using Qiaprep spin  miniprep kits (Qiagen) following the instructions of the manufacturer.  2.1.4.2 Isolation of chromosomal DNA Chromosomal D N A of  H. influenzae was extracted using the method described by Poje  and Redfield (70). 1.5 m l of a fresh overnight culture was pelleted and resuspended in cell resuspension solution (50 m M Tris HC1, pH7.4, 50 m M E D T A ) . The cell suspension was then mixed with 50 ^1 of 10% S D S and incubated at 50°C for ~10 minutes for complete lysis of the cells. The chromosomal D N A was extracted twice with phenol/chloroform and precipitated with 2 volumes of 95% ethanol in the presence of 15 m M N a C l . The fibrous clump of D N A resulting from the precipitation was retrieved by winding it onto the sealed tip of a Pasteur pipette and was washed with 1.0 ml of 70% ethanol dribbling down the pipette. The D N A was then air-dried for one hour. The dry D N A was resuspended in 200 ul of d H 0 or T E buffer (10 m M T r i s - H C L , pH7.4, 1 m M 2  E D T A ) . A few hours were allowed for full dispersion of the D N A .  The method for extraction of  P. aeruginosa PA01 chromosomal D N A was modified from  the online protocol of Goldberg lab (34). From an overnight culture of  P. aeruginosa  24  grown at 37°C in low salt L B broth (0.05% NaCl), 1ml of cells were pelleted and resuspended in 1ml T N E buffer (10 m M T r i s - H C l , p H 8.0, 10 m M N a C l , 10 m M E D T A , p H 8.0). The cells were pelleted again and resuspended in 135 u,l of T N E buffer. T o lyse the cells, 135 \il of T N E with 2% Triton X-100 was added followed by addition of 30 \il freshly prepared 5 mg/ml lysozyme and incubation at 37°C for 30 minutes. The cell lysate was treated with 15 [il of 20mg/ml proteinase K at 6 5 ° C for 2 hours. Phenolchloroform extraction was performed twice afterwards to purify the D N A . The subsequent procedures were the same as described above for H. influenzae D N A extraction.  2.1.4.3 Gel electrophoresis, gel extraction, restriction digestion and ligation Agarose and polyacrylamide gel electrophoresis were performed as described by Sambrook et al. (77) and Ausubel et al. (5). D N A fragments were extracted and purified from agarose gels using QIAquick G e l Extraction Kits (Qiagen) according to the instructions of the manufacturer. D N A manipulations with restriction endonucleases and T 4 D N A ligase were done according to the manufacturers' recommendations. The ligated products were transformed into E. coli D H 5 a , G M 2 1 6 3 ( N E B ) or Top 10 (Invitrogen) competent cells.  2.1.4.4 PCR amplifications Amplification of D N A by P C R was done in a PTC-150 M i n i C y c l e r ( M J Research). Taq D N A polymerase was used for most P C R amplifications. Pfu polymerase was used if products longer than 3.5kb were to be amplified or i f P C R products with no " A " overhang were required. For Inverse P C R of p P I L B : : K a n (a 6174 bp amplicon), the R  25  Expand Long Template P C R System (Roche) was used. The P C R primers used in this I  study are listed in Table 2.3. Primers were designed using the online program NetPrimer of Premier Biosoft International (44). Simulation of P C R reactions were carried out using Amplify 1.2 (28). D N A sequencing was performed by the U B C N A P S Unit (Nucleic A c i d Protein Service Unit).  For colony P C R , single colonies were picked up with 20-200 \x\ pipette tips and resuspended in water in P C R tubes. The cells were lysed at 9 5 ° C for 10 min and placed on ice afterwards for a few minutes. The remaining components were then added and P C R reaction was performed the same way as normal P C R .  26  \  Table 2.3 Primers used in this study. Name  Sequence  O r i g i n a l usage  PilBf PilBr  5'-GCTTGTGAGAATGCTAAACCAGA 5'-GGTCAGCAATCCTTTCTCAATATCT  Cloning of K W 2 0 pilB  pilCf pilCr  5-TTCTTGGTTTGCCACGTTTG 5'-CAAGCTGGCGTACCGCTTAA  Real time P C R of H. influenzae pilC  This study  pilDf pilDr  5 '-CACTTTTTTTGTTGAATAGAGGCAAA 5 '-GCGATTTACGTTGAATTATTTCCA  Real time P C R of H. influenzae pilD  This study  LPICf LPICr  5'-AATGACTAAAAAACTCTTTTATTATCAAGC 5 '-GTTGTCTAGAGGCAAAGTGCGGTTGAAATG  Cloning of H. influenzae pilC.  This study  Inverse P C R of p S U P I L B and pPILB  This study  Inverse P C R of pPILB:: K a n  This study  Source This study  Xbal. VecF Vec2R  5'-CATTTTCCTTTTATTAAGCCTTTGTTGG 5'-AGGTTCTAGAGGTCGAATCGGCGTGTATCA  Xbal Vec2F  5'-TTTATCTAGAAATCGCCCATCTTGTGGAAGTC  Xbal piBF2  5'-  CCAAACTAGTGAACCTCAACAAGCACTTACC  Spel piBR  5'-TCTTACTAGTGCCCTCCCAGATGAAAAC  R  Cloning of P. aeruginosa pilB coding sequence and 200 bp upstream  This study  Inverse P C R of p S U P I L B  This study  Spel Vec3F  5'-CCTTTTCTAGAGCCTTTGTTGGCAACGACATTGTG  Xbal Pro-1 OR  5' -AGGTTCTAGAGGTCGAATCGGCGTGTATCA  Xbal  27  T a b l e 2.3 Primers used in this study (continued). Name  Sequence  O r i g i n a l usage  piTF piTR  5-AATGGATATTACCGAGCTGCTCGCCTTCA 5 '-TCCTACTAGTGAATCCTAGACGCAGTTCC Spel  Cloning of P. aeruginosa pilT (Starts at A T G ) . The first " A " in p i T F before the start codon A T G is the last residue of the pil promoter  piTF2 piTR  5'-ATTGGGGAGTCCTATGGATATTACCGA 5 '-TCCTACTAGTGAATCCTAGACGCAGTTCC Spel  Cloning of P. aeruginosa pilT (Starts at T T G ) . The first " A " in piTF2 before the start codon A T G is the last residue of the pil promoter  This study  Kl  Amplification of a 997 bp fragment from pWJC3 containing K a n cassette  (61)  K2  5 '-GGATCCGGGGGGGAAAGCCACGTTTGT BamHl 5'-AGCCGCCGTCCCGTCAAGT  PFRTPilA PRRTPilA  5'-AGTTCTTTGAGTTGGCTGCTTTCT 5'-TCAGCCTGTGAAACCATTGC  Real time P C R of H. influenzae pilA.  MN #304  SpBEf SpBr  5'-CTCTAGTGCCCTCCCAGATGAAAA 5' -CAGTCGT ATCTCTGCTCGTCTCAA  Sequencing of p P A P B  This study  SpBf  5'-GGTAGAGTTCCTTCTGGTCCTCCTC  Sequencing of p P A P B  This study  SpBSf  5 '-CGGAACTGCGTCTAGGATTCACTA  Sequencing of p P A P B T and p P A P B T 2  This study  SpTSf  5'-TCCCAGTCACGACGTTGTAAAAC  Sequencing of p P A P B T and p P A P B T 2  This study  PaBf PaBr  5'-GCTCTTCCGACCTGCACTTC 5' - A G C GCAC C C GGTAGATC T T  Real time P C R of P. aeruginosa pilB  This study  PaTf PaTr  5'-CGGGACGTCTGAAACACGTT 5 '-ATGGAGGAGCTTGGCATGG  Real time P C R of P. aeruginosa pi IT  This study  Source This study  R  28  2.1.5 D N A binding & uptake assays D N A binding & uptake by competent H. influenzae cells was measured using Bglll or Hinfl fragments of K W 2 0 D N A end-labeled with a - P - d A T P (Amersham). K W 2 0 D N A 33  (~ 6.8 ng) was treated with Bglll or Hinfl in a 50 u.1 reaction, and 20 u.1 was used in a f i l l in reaction with Klenow enzyme (Roche) in the presence of 6.25 m M d G T P / d T T P / d C T P and 16 pmol of a - P - d A T P (40 u € i ) . The reaction was then chased with 0.625 m M 33  dNTPs at 37°C for 15minutes. Unincorporated nucleotides were removed using Sephadex G25 mini columns (Amersham), yielding 450 ul of 8.4 ng/ul labeled D N A (1.45 x 10 cpm/ng). 4  2.1.5.1 Total DNA binding plus uptake To measure the total D N A bound to the cell surface and taken up into the cells, P 3 3  labeled D N A (~ 100 ng) was added to 1 ml of M l V - i n d u c e d competent cells. After incubation for 5-10 min at 37°C, the cells were washed twice with 1ml of M I V medium, pelleted, and resuspended in 100 ul of M I V . The cells and 100 u.1 of the supernatants of each wash were separately used for scintillation counting. The total amount of D N A binding plus uptake was determined from the radioactivity associated with the cells.  2.1.5.2 DNA binding and/or uptake To measure D N A binding and uptake separately, P labeled D N A (~ 100 ng) was added 3 3  to 1 ml of M l V - i n d u c e d competent cells. After incubation for 5-10 min at 37°C, the sample was then split into two 0.5 ml fractions. One 0.5 ml fraction was transferred to ice immediately to stop D N A uptake. The other fraction was incubated with 10 u.1 of 100  29  ug/ml DNase I for 15 min at 37°C. The cells were washed twice with 1ml of M I V medium, pelleted, and resuspended in 100 u.1 of M I V . Subsequently, the cells and 100 ul of the supernatants of each wash were subjected to scintillation counting. The total amount of D N A bound to the cell surface and taken up into the cells was determined from the radioactivity of the DNase I-free cell fraction, while the amount of D N A taken up into the cells was determined from the radioactivity of DNase I-treated cell fraction. The amount of D N A binding is obtained by subtracting the above two numbers.  Scintillation counting of the samples was performed in 1ml of liquid scintillation cocktail (Amersham) in a Beckman L S I 8 0 I liquid scintillation system.  2.1.6 Real time P C R Real time quantitive P C R was performed using an A B I Prism 7000 Sequence Detection System. The primers used for the real time P C R of H. influenzae pilA, pilC, pilD and P. * aeruginosa pilB and pilT in H. influenzae were designed using Primer Express software supplied by Applied Biosystems, and are listed in Table 2.3. Total m R N A was extracted from M l V - i n d u c e d competent cells of H. influenzae using an RNeasy M i n i K i t (Qiagen) and was treated with DNase I using a DNA-free kit (Ambion) to remove possible trace of D N A . Reverse transcriptase P C R was carried out in the P T C - 1 5 0 M i n i C y c l e r ( M J Research) according to the protocol provided by the S Y B R Green P C R Master M i x and R T - P C R protocol (Applied Biosystems) to get c D N A . Real time P C R was then  i  performed using the c D N A as template in the A B I Prism 7000 Sequence Detection  30  System (Applied Biosystems). Data analysis was done using the software provided by the same company.  2.1.7 Nucleotide sequence analyses Multiple sequence alignments were performed using N C B I B L A S T programs ( N C B I ) and ClustalW ( E M B L - E B I ) . Restriction maps of D N A were obtained using D N A Strider (24).  2.2 Plasmid and strain constructions This section includes the overall strategy and the general procedures of plasmid and strain construction.  2.2.1 Construction of H.  influenzae  2.2.1.1 p P I L B a n d p P I L B : : K a n  plasmids  R  The scheme for construction of p P I L B is shown in Fig. 2.1. T o generate p P I L B , a fragment containing H. influenzae K W 2 0 pilB and flanking sequences was amplified from M A P 7 chromosomal D N A using Taq D N A polymerase and primers PilBf/PilBr. The 2326 bp P C R product was ligated to a pGEMT-easy vector (Promega), and transformed into G M 2 1 6 3 (dam, dcm) ( N E B ) . This host was chosen because the restriction enzyme Bell used in the next step is blocked by D a m methylation. Transformants were plated on L B plates supplemented with A m p i c i l l i n , X-gal and I P T G .  31  White or light blue colonies were selected. Transformants carrying the desired p P I L B construct were verified by restriction analysis using EcoRl, Dral and Sacll.  For p P I L B : : K a n construction, a 1466 bp fragment bearing a Tn903-derived K a n R  R  cassette was obtained from p W J C 3 by BamHX digestion and inserted at the Bell site of pilB on p P I L B , resulting in the plasmid p P I L B : : K a n (Fig.2.2). K a n R  R  transformants  carrying the right constructs were identified by restriction analysis with BspHl, Pvul, Nru I, and Sspl digestions.  F i g . 2.1 Construction of p P I L B . A fragment containing pilB and the flanking sequence was amplified from K W 2 0 chromosomal D N A and inserted into the p G E M - T easy vector. The fragment includes the C R E promoter of the pil operon, pil A, pilB and about  32  330 bp of the  pilC gene. The Bell site used for the later cassette insertion was at 1280bp,  so that there are at least l k b flanking region on either side for the future homologous recombination of the  pilB::Kan fragment into the chromosome. R  Kan  R  Fig. 2.2 Construction of p P I L B : : K a n . A 1466 bp fragment containing the K a n cassette R  R  was obtained with BamHl digestion of pWJC3 and inserted into the Bell site of pilB on p P I L B . The orientation of the K a n cassette is opposite to that of pilB. R  33  2.2.1.2 p P I L C The scheme for p P I L C construction is shown in Fig. 2.3. A fragment containing H. influenzae pilC was amplified from K W 2 0 chromosomal D N A and cloned into a vector containing a C R E promoter of the H. influenzae pil operon. The detailed procedures are described below.  F i g . 2.3 Construction of p P I L C . H. influenzae pilC was P C R amplified and inserted into a vector containing a C R E promoter of the H. influenzae pil operon, thus that pilC is under the control of a regulatory sequence of its own.  34  A 1384 bp fragment bearing the complete pilC gene and a small fraction of pilD was amplified by P C R using primer L P I C f / L P I C r . L P I C f was 5' phosphorylated and L P I C r was engineered with an Xbal site. Pfu D N A polymerase was used in the P C R to get a blunt ended fragment.  To generate a pSU20-based shuttle vector containing the C R E promoter of the H. influenzae pil operon (simplified as " p S U - C R E vector" below), an inverse P C R of p P I L B was performed using primers V e c F / V e c 2 R . The V e c 2 R primer included an added Xbal site. Pfu D N A polymerase was used in the inverse P C R to get blunt-ended amplicons. The Xbal cleavage of the 3661 bp P C R product generated a p S U - C R E vector of 3652 bp that was ligated to the above Xbal digested pilC fragment, with the pilC start codon fused at the normal location of the pilA start codon. The ligation mixture was first transformed into Top 10 competent cells (Invitrogen). C a m resistant transformants carrying the desired p P I L C construct were verified by restriction analysis with EcoRl, Hindlll, and Sspl. Miniprep D N A from one of the colonies carrying the right construct was then transformed into G M 2 1 6 3 competent cells (dam, dcm) ( N E B ) to permit Xbal digestion in the downstream construction (Section 2.2.1.4). Sspl digestion and Xbal/Spel double digestion were used to confirm the restriction pattern of p P I L C extracted from G M 2 1 6 3 cells. The sequence of p P I L C around the 5' ligation joint was examined by D N A sequencing using primer PilBr.  2.2.1.3 p S U P I L B To generate the plasmid p S U P I L B , the insert of p P I L B was released by EcoRl digestion and subcloned into EcoRl linearized pSU20 (Fig.2.4). The ligation mixture was  35  transformed into  D H 5 a ,  plated on L B plates supplemented with Cam, I T P G and X g a l .  White colonies bearing the desired p S U P I L B construct were verified by restriction analysis with EcoRl, Dral, Hindlll, and by colony P C R using primers P i l B f / P i l B r . The sequence of p S U P I L B in the C R E promoter region was examined by D N A sequencing using primer P i l B r .  F i g . 2.4 Construction of p S U P I L B .  36  2.2.1.4 p B K a n C R E C To construct the plasmid p B K a n C R E C , a fragment containing the pilC gene under the control of the C R E promoter (CREIpilC fragment) was subcloned into a vector containing a H. influenzae pilB knockout. The plasmid thus constructed contains a fragment bearing a C R E + a truncated pilB with a K a n insertion at the Bell + another C R E + pilC + part of R  pilD (the ApilB::Kan  R  I CREIpilC fragment) and can be used to generate a non-polar pilB  knockout (Fig. 2.5). The detailed procedures for p B K a n C R E C construction are described below.  To obtain a vector containing a H. influenzae pilB knockout, inverse P C R was performed using plasmid p P I L B : : K a n and primers Vec2F/Vec2R, each with an Xbal site at the end. R  Expand Long Template P C R system" (Roche) was used to amplify the 6174 bp fragment. The P C R product was gel extracted to remove non-specific amplifications, followed by Xbal digestion to get the 6160 bp vector containing a H. influenzae pilB knockout. The 1537 bp CREIpilC fragment was released from p P I L C by SpellXbal double digestions and ligated to the above vector . The ligation was transformed into T o p l O E. coli competent cells (Invitrogen), selecting for A m p / K a n double resistance.  Transformants with the desired p B K a n C R E C constructs bearing the  ApilB::Kan l R  CREIpilC fragment were screened by restriction analysis using Accl, EcoRl, Hindlll, and Sacll/Ndel double digestion.  37  F i g . 2.5 Construction of p B K a n C R E C and the strain R R 1 1 5 0 .  38  2.2.2 H. influenzae  knockouts  2.2.2.1 RR1137 (pilB::Kan ) R  To generate the pilB knockout strain RR1137, the insertional knockout of pilB constructed in p P I L B : : K a n was released from its vector by EcoRl digestion and R  transformed into K W 2 0 by natural transformation, replacing the w i l d type pilB on the chromosome. T o reduce the probability that one cell would take up multiple fragments of the pilB knockout, serially diluted D N A was used in the transformations. Colonies were chosen from the transformation in which the number of transformants was greatly reduced compared to the transformation using more D N A , suggesting that the amount of D N A used was low enough that no cells took up more than one fragment. K a n colonies R  were chosen from a transformation that used ~ 0.5 ng D N A and chromosomal D N A was extracted. Substitution of the wild type pilB with the knocked-out pilB was verified by P C R using chromosomal D N A and the primer pair P i l B f / P i l B r .  2.2.2.2 Construction of RR1150 (kpilB::Kan ICREIpilC) R  The overall strategy used to generate a non-polar H. influenzae pilB knockout is shown in Fig.2.5. The strategy was to place a regulatory sequence containing a duplicate copy of the pil C R E promoter immediately upstream of pilC, downstream of the insertionally inactivated pilB. T o do this, the fragment containing the pilB knockout, pilC with a C R E promoter, and flanking sequences (ApilB::Kan /CRE/pilC) R  was released by Mlul/Spel  double digestion of p B K a n C R E C and transformed into K W 2 0 by natural transformation. For the same consideration as in Section 2.2.2.1, limited amounts and serial dilutions of  39  D N A (5 u l , 0.5 u l and 0.05 ul to 1 ml of competent cells) were used in the transformation. Colonies were chosen from the transformation with ~2.5 ng D N A , in which C F U were significantly reduced. P C R was used to screen for colonies carrying the recombined fragment in the chromosome, using primer pairs P i l B f / P i l B r , P i l B f / P R R T P I L A and P i l B f / K 2 .  2.2.2.3 Construction of RR1136 (KW20 carrying pSUPILB). RR1138 (RR1137 complemented with pSUPILB) and RR1151 (RR1150 complemented with pSUPILB) For construction of strain RR1136 ( K W 2 0 carrying p S U P I L B ) , p S U P I L B was introduced into K W 2 0 by natural transformation. Since RR1137 (pilB::Kan ) and R R 1 1 5 0 R  (hpilB: :Kan I CREIpilC) are defective in natural transformation, the strains RR1138 R  (RR1137 complemented with p S U P I L B ) and RR1151 (RR1150 complemented with p S U P I L B ) , were obtained by introducing p S U P I L B into RR1137 and R R 1 1 5 0 using chemical transformation, selecting for Cam/Kan double resistance. Presence of p S U P I L B in the strains was verified by restriction analysis and colony P C R as described in the construction of p S U P I L B .  2.2.3 Introduction of P. aeruginosa pilB  PA01 pilB, pilT to the H.  influenzae  background  P. aeruginosa P A 0 1 pilB and pilT were amplified from chromosomal D N A and placed under the control of the C R E promoter of H. influenzae pil operon. The cloning strategies are shown in Fig.2.6 and Fig.2.7 and Fig. 2.8. Details are given below.  40  2.2.3.1 Cloning of P. aeruginosa pilT For construction of P. aeruginosa pilT plasmids p P A F T and p P A F T 2 , the two fragments containing pilT (starting at A T G and G T G 0  1 2  respectively), were amplified from P.  aeruginosa P A O l chromosomal D N A and cloned into a pSU20-based vector containing a C R E promoter of the PL. influenzae pil operon. T o do this, fragments containing P. aeruginosa pilT and pilTl were P C R amplified using Pfu D N A polymerase and primers piTF/piTR or piTF2/piTR, respectively (Fig. 2.6). Both p i T F and p i T F 2 were 5' phosphorylated and a Spel site was included in piTR. The amplified fragments start exactly at the first nucleotide of the start codon, and end at ~50 bp downstream of each O R F . The P C R products were digested with Spel, resulting in 1096bp and 1108bp fragments. T o get a pSU20-based vector containing a C R E promoter of the H. influenzae pil operon (simplified as p S U - C R E vector below), inverse P C R of p S U P I L B was performed using Pfu D N A polymerase and the primer pair V e c F / V e c 2 R . V e c F was 5' phosphorylated and V e c 2 R had a Spel site at the 5' end. The 3016 bp P C R product was purified by gel extraction, followed by Xbal digestion, resulting in a pSU-CRE-vector of 2481 bp.  p P A P T and p P A P T 2 were obtained by ligation of Spel-digested P. aeruginosa pilT fragments into Xbal-digested p S U - C R E vector. The ligation mixture was transformed into D H 5 a and plated on L B plates selecting for C a m resistance. Restriction analyses were applied to screen for transformants carrying desired constructs. Restriction enzymes EcoRl, Ncol and Xhol were used in analysis of p P A P T candidates, and Ncol, Pvull, Xhol were used first for analysis of p P A P T 2 candidates. Hinfl digestion was  41  performed subsequently to confirm that p P A P T 2 has the extra 12 bp not present in p P A P T . In addition, M n d l l l , a unique cutter of both constructs, was used as well to check the size of the two plasmids. Sequences of p P A P T and p P A P T 2 around the ligation joint and the promoter were examined by D N A sequencing, using the primer pair PilBr/piTR.  PA01 chromosome  Fig. 2.6 Cloning of P A 0 1 pilT and construction of p P A P T  42  2.2.3.2 C l o n i n g o f P .  aeruginosa pilB  The overall strategy for P. aeruginosa PA01 pilB cloning is shown in Fig.2.7 and F i g . 2.8. P. aeruginosa pilB was cloned under the control of its own R B S (Ribosome Binding Site) and a C R E promoter of the H. influenzae pil operon. In the final pilB clone, the ligation joint of P. aeruginosa pilB and the H. influenzae C R E promoter was designed to be downstream of the transcription start and upstream of the R B S . This was to minimize the impact of short deletions at the ligation joint. T o construct this, a fragment containing P. aeruginosa P A 0 1 pilB and 200bp upstream was first amplified from P. aeruginosa PA01 chromosomal D N A and cloned into a pSU20-based vector containing a C R E promoter of the H. influenzae pil operon, giving the plasmid p P A P B p r o . The R B S in the C R E promoter and most of the regulatory sequence of P. aeruginosa pilB except the R B S , were then removed by inverse P C R of p P A P B p r o using the primer pair pilBF3 & Pro-10. The final clone of P. aeruginosa pilB was obtained by self-ligation of the P C R product, such that P. aeruginosa pilB uses H. influenzae pilA's C R E promoter including the putative transcription start, and uses P. aeruginosa pilB's own R B S . T o ensure that the ligation point was not too closed to the transcription start and R B S , 6 bases downstream of H. influenzae pilA's transcription start and 6 bases upstream of P. aeruginosa pilB's R B S were included in the clone, so that in the final pilB clone, there should be 12 bases between the putative transcription start and R B S . The experimental procedures are described in more detail below.  43  Transcription staff Promoter \  Transcription start Promoter  '^•'xx  RBS  H0,-35)  CRE  f  (200 bp upstream start codon) PCR  (aoiobp)  Transcription start Promoter (-10 and-35)  Spel digestion  I  Xba\ digestion Xbs\  CR! S p s l  ^  Inverse PCR  pilB f r a g . (l95DbpT  regulatory sequence _pjBF3  owi RBS ^iliP' "5»H( ^ Transcription start \JT ptiB Promoter (-10 and-35) CRE Inverse PCR ll,  te  Transcription start ' Promoter (-10and-35] CRE" f L  pPAPBpro  ^  (4236bp)  1  owiRBS  p/JB  Transcription start. Promoter  Self (-10 and-35  ligation  Fig. 2.7 Cloning of PA01 pilB and construction of pPAPB.  pPAPB  (4236bp)  transcription -10 start | P. aeruginosa pjBS with GCGTflMl^GGACCTCffoTAGqC^ its own promoter PAOl^tflB's promoter  B: wJtutiizaepdA aiul its promoter  - p3B coding seq. — C l o n e d area  trariscription stall start  -10  RBS  ;Pndir^;finalclone jTT;iJ|[Jig||5Q  transcription start  : influenzae'•promoter  stm1 codon pdA coding seq.  H. inJhicuzAe'•promoter Cloned area  -10  ol P. aeruginosa pUB  start codon  RBS  RBS ^  ligation joint  start codon P. aeruginosa pilB coding seq.  Fig. 2.8 The strategy for P. aeruginosa P A O l pilB cloning, showing the region and sequences around H. influenzae pil A and P A O l pilB chosen for the cloning, and the predicted sequence of p P A P B around the ligation point.  To construct the plasmid p P A P B p r o , a fragment containing the pilB O R F and 200 bp upstream was amplified by P C R using the primer pair p i B F 2 / p i B R , each having a Spel site at the 5' end. The P C R product was digested with Spel, which produced a 1950 bp fragment. A pSU20-based vector containing the C R E promoter ( p S U - C R E vector) from H. influenzae pil operon was obtained by inverse P C R of p S U P I L B using the primer pair Vec2F/Vec2R. A n Xbal site was included at the 5' end of Vec2F. Xbal digestion of the P C R product produced a 2469 bp p S U - C R E vector, which was ligated to the 1950 bp pilB fragment to get the plasmid pPAPBpro. The ligation mixture was transformed into D H 5 a , selecting for C a m resistance. Constructs carrying desired construct were screened with restriction analysis using BamHl, Banl and PvwII. Sequence of p P A P B p r o around  45  the ligation joint and the promoter was examined by D N A sequencing, using primer PilBr.  p P A P B was obtained by removing the unwanted sequence in p P A P B p r o , using inverse P C R and religation of the P C R product. Pfu D N A polymerase and the primer pair piB3F7Pro-10, both 5' phosphorylated, were,used in the P C R reaction. The P C R amplification removed most of pilB's own regulatory sequence except its R B S (Ribosome Binding Site). Therefore the final P. aeruginosa pilB clone, which was obtained afterwards by self-ligation of the P C R product, should contain pilB with its own R B S and H. influenzae pilA's C R E and promoter in the upstream region (Fig. 2.7). The predicted sequence at the joint of the self-ligation is shown in Fig. 2.8. Self-ligation of the P C R product was transformed into D H 5 a , selecting for C a m resistance. Restriction enzymes BamHl, Banl and PvwII were used to examine colonies bearing the desired construct. The sequences of pilB and the C R E promoter in p P A P B were examined by D N A sequencing, using primer pairs SpBf/PilBr and SpBEf/SpBr.  2.2.3.3 C o n s t r u c t i o n of p P A P B T a n d p P A P B T 2 The scheme for p P A P B T and p P A P B T 2 is shown in Fig.2.9. T w o fragments containing the P. aeruginosa pilT gene (starting at A T G and G T G . , respectively) under the control 0  1 2  of the C R E promoter were released from p P A P T and p P A P T 2 by NhellSpel double digestion, and inserted into Spel digested p P A P B . The ligation mixtures were transformed into D H 5 a and spread on L B plates, selecting for C a m resistance. The constructs were examined by restriction analysis with BamHl, Ncol and Sacll.  46  F i g . 2.9 Construction of p P A P B T and p P A P B T 2 .  2.2.3.4 T r a n s f o r m i n g the  P. aeruginosa plasmids into H. influenzae pilB  Since H. influenzae pilB is deficient in natural transformation, the P. aeruginosa plasmids p P A P T , p P A P T 2 , p P A P B , p P A P B T and p P A P B T 2 were first transformed into w i l d type H. influenzae K W 2 0 , chromosomal D N A of the non-polar pilB knockout was then introduced into the resultant strains by natural transformation (Fig.2.10). Details are given below.  47  Fig. 2.10 Construction of H. influenzae strains containing both pilB knockout and P. aeruginosa pilB, pilT plasmids.  The 5 plasmids were first transformed into K W 2 0 , selecting for C a m resistance. Since the pilB plasmids ( p P A P B , p P A P B T and p P A P B T 2 ) contain multiple HindlU sites, they were treated with methylase M n d l l l (Takara) before transformation into K W 2 0 (~ 0.6 units/u.g) to protect them from digestion by H. influenzae Hindlll.  Presence of the P.  aeruginosa plasmids was verified by restriction analyses. The resultant strains were then transformed with the chromosomal D N A of RR1150. Limited amount of D N A (100 ng, 10 ng and 0.1 ng to 1 ml of competent cells) was used in the transformation (See Section 2.2.2.1). Kan/Cam double resistant transformants were chosen from the 0.1 ng transformation. Chromosomal organization was examined by P C R using primer pairs P i l B f / P R R T P I L A and P i l B f / K 2 . Sequences of the P. aeruginosa inserts in the resultant strains were verified by D N A sequencing (Fig. 2.11, Fig. 2.12, Fig. 2.13). The sequencing primers used were: for p P A P B , SpBEf, SpBr, S p B f and P i l B r ; for p P A P T , and p P A P T 2 , p i T R and P i l B r ; and for p P A P B T and p P A P B T 2 , SpBsf, SpBr, SpBf, SpBsF, SpTsf and piTR.  48  Part II PilBr v  SpBr \ . ^ SpBEf  Parti pnmoter  ^  PAOl /ttfff  ^  Fig. 2.11 Sequencing of P. aeruginosa P A O l pilB in p P A P B .  PilBr  'SHE  Fig. 2.12  Fig. 2.13  PA01 /v//  Sequencing of P . aeruginosa P A O l pilT in p P A P T .  Sequencing of P. aeruginosa P A O l pi/fi an<ipilT in p P A P B T and p P A P B T 2 .  Since there are 2 C R E promoters in the plasmids, the primer P i l B r that anneals to the C R E promoter cannot be used. Therefore SpTSf and S p B S f were designed.  49  C h a p t e r 3: Results and discussion  3.1 Sequence analysis and functional prediction of H. influenzae PilB H. influenzae pilB (gene HI0298) is 1395bp long and encodes a predicted protein of 464 amino acids. Sequence alignments showed that P i l B is more similar to homologues of P. aeruginosa P i l B (required for Tfp biogenesis), and less similar to P i l T homologues (required for Tfp retraction). In a B L A S T P with H. influenzae P i l B , a number of Tfp assembly proteins (required for Tfp biogenesis) were obtained with high identities to P i l B , including P i l B (or HofB) in Pasteurellaceae Species, P i l B of Vibrio cholerae, P i l F of N. gonorrhoeae and N. meningitides, P i l B of Acinetobacter sp. A D P 1 . Following the Tfp biogenesis proteins, twitching motility proteins (PilT homologues, required for Tfp retraction) were found with much lower E-values (>10" °). Alignment scores of H. 3  influenzae P i l B homologues in P. aeruginosa and N. gonorrhoeae are shown in Table 3.1.  Table 3.1 Alignment scores of H. influenzae P i l B homologues in P. aeruginosa and N. gonorrhoeae ClustalW score  B L A S T E value  Identity to H. influenzae P i l B  PaPilB  40  4e-89  38%  NgPilF  35  2e-73  41%  PaPilT  21  3e-30  38%  NgPilT  17  8e-27  31%  50  P i l B and PilT homologues all belong to the P u l E / G s p E family (TrbB proteins). Members of this family are distinguished by conserved nucleotide binding and hydrolysis motifs (49). The nucleotide binding motifs of H. influenzae P i l B are indicated in Fig. 3.1a. F i g . 3.1b shows the sequence alignment of the C-terminal of some P i l B and PilT homologues. A l l the proteins share the three conserved domains related to nucleotide binding and A T P hydrolysis (Fig. 3.1a). In addition to the sharing of these domains, extended searching in the alignment also revealed some conserved regions shared by the PilT proteins but absent in the P i l B proteins (also see (66)). This is consistent with the different functions known for P i l B & PilT in P. aeruginosa and N. gonorrhoeae (4, 33). H. influenzae P i l B shares high homology with other P i l B proteins and does not contain the conserved regions unique to the PilT proteins. Therefore H. influenzae P i l B is expected to function as an ATPase in the assembly of the Tfp-like apparatus for D N A uptake.  * 2 7 2  251 -300 RVLSQPQGtilLVTGPTGSGKSISLYTAtQWLHTPDKHIMTAEDPIEJIELD IhhhxGPIFGSSKoirxhxxrThl jhhhhEDxxg Motif I  Motif II  301-350 GIlOSQJNPQra frExxxKRxxPDxhhhGEhRDxExhxh^ Motif III  351-400-^GHLVLSTLHtt'NDAISAISRLOOLGIQQYEIKNSLLLVIAQRLVRKLCSKC ^xhhhx&JHl  F i g . 3.1a Residues 251-400 of H. influenzae P i l B showing the 3 nucleotide binding motifs predicted by Lessl and Lanka (49). h represents hydrophobic amino acids, o marks polar ones, and x any other amino acids. Highlighted capital letters represent residues that are strictly conserved in at least 12 of the 14 proteins analyzed by Lessl and Lanka. The sequence of P i l B fits all the three motifs except residue I l e  272  (underlined)  that should be a Thr (the 13 residue) in Motif I. th  51  200 HiPilB AaPilB PaPilB AcPilB NgPilF EcHofB PaPilT AcPilT NgPilT  HiPilB AaPilB PaPilB AcPilB NgPilr EcHofB PaPilT AcPilT NgPilT  QPQPLISKIFANRIISRL •fiVQNTVPLSIASRLISRI H EVAKPPIQ1ASRI SARI R LIATPPLQLATRLASRL REWOPPIAVRGOIASRI PLPDVSPDAGVALTARL  .AKlilSENRLPQpGRWdfKTTFS-DI . AKlJo I S E T RLPQ D G R f J F K T T F 8 - D ' I A G L 0 I S E R RKPO D G R I K M R V S K T - K S I ISOJMDISEKRVPQDGRIKL K M S K S - K T I I S R L DISEK RIPC D G R M Q L TFQKGGKPV .GNLIJI AEHRLPQDGOFITVELAGN—J  RLSTLPTHWBEKI RLSTJJATPFDEKJ R VN T L PTLWJG E K I RVNSLPTLFGEKI RJVSI%|PTLFGEKV IATLPCRGGEKV  INLPJPLEH! INLCAMDH|  2  d  ilHLEEMS 250 LSTSE1 GMTENQQQAFQRVLSg l L V T G P T G S G K S I Sjl L A F G E L GMTES0.QORFRRAL.Sd I L V r G P T G S G K & 161 M G I DA L G Y E E D Q K E L Y L A A L K Q ILVrGPTGSGKIVSL L G I D A L G Y E P E 0 . K A L F M E A L N 1 IdGMLL I r G P T G S G K I V S I S D A A S U H I D g L G F E PFQKKLLLEAIHRjPYjGMV L V rGPTGSGK I V E I  t  NKPVE NRPVQ SSSAQ PASAH  JVLV -VLV >]ICDJY|PWGfLVliV :IAESPWCMV L V  QVGQMjgVjNMGMQPLQLADFAHAL GHG GLG  EVjF KI|F| — s :  300 lEjlJELDG— EIQLPG— VEINLEG-17 E I N L E G — MEINLPG— VE1PIAG ]  K-TPDKHIM1AI S-TSEKHIMIAI SLNI^f-TTDINISIAI K j - T E H A N1 SflAJ -TESVNIfl H-TADIHIC  TGPTGSGK rGPTGSGK TGPTGSGK rGPTGSGK  m  IfHHJ  350 HiPilB AaPilB PaPilB AcPilB NgPilF EcHofB PaPilT AcPilT NgPilT  SEjSAKljAJLRAAQTGHLVLSTLHXNDAI EEGAAHRLRAAQrGEJLVLSrLHIND, LEIAEiniKADSrGHMVMSTLHlNS  S^l|steJdQL|G I0RH UPAF VASF UAPF VARH  [. E i A E i n I K A A o ran L v Hsfruii N: L E I AD I H I K A A E rCK M V F S T L H 1 N N ; S R I J A E I A I K A A 8 TOH L V L S J T L H I J N S R L A L I A A E TGH L V  FS TIM' RtALIAAlrGKLVFGTLH  GMAHIJAAlETrGHlJvrOTLH  1  400 HiPilB AaPilB PaPilB AcPilB NgPilF EcHofB PaPilT AcPilT NgPilT  BIKHSl  VIAQRWVRKLCSKCGG-  - N L A N S C D C H - Q G Y R 5 R I G V YQFLJH -  EIDGSLLLVIACRLVRKRCQKCGG  -NATRFCDCH-OGYK BRVGVYQFLOPNN L A T BV NL 1 1 A J R L A R K L C S H C K K E H E V P - R E T L L H E G F P E D K I G T — F K L Y S P V G C D H C K - N G Y K J G R J v j d l J Y E V ^ K N T P A NIATBVNL VIAQRL  ARRXCSQCKRPIQVP-ERSLLEMGFTPEDLAQPEFQIFEPVGCHDCR-EGYKSRVGIYEVMKITPE  N I A S BV S L IMA J R L L R R L C S S C K Q E V E R P S A S A L K E V G F T D E D L A K - D W K L Y G A V G C D R C R G Q G YK 5 R A S V Y E V M P I S E E MLSSA1TLVIACRLVR KLCPHCRRQOGEP Mil 1  IHIPDNVWPSPLPHWOAPGCVHCY-HGrYSRTALFEVLPITPV  'iio"^  ML!  E " El!3?H B  m  ^IPISHTHD 450  -WQQNDYeTBFKiejRASGlZKVSOBSrDEKEIERVLGKNL  HiPilB AaPilB  LQH POG YETpFjAD^GQS A L E K L K D H T p D L T E I Q R V L G Q T H D  PaPilB  LQRIIMEE  AcPilB  ISKIIMED  NgPilF  MQRVIMNK-  GNSIEI AEQARKEGFpDiRTSGLLKAMQalfTSLEEVNRVTKD G N A L E I A A T A E T L GF NM L R R S G L K K V M u G J V T S L O E I N R V T S E  EcHofB  G T E V G I L D V A Y K E G W VD L R R A G I L K I M Q q i T S L E E V T A N T N D I R Q L I S A N — T D V E S L E T H A R ! lAbMRT^ENGCLAVEgflLfTTFEELIRVLGMPHGE--  PaPilT  II  AcPilT  SRENAREKAjKI Pt LPTDAVBHSIPI  NgPilT  Fig. 3.1b A m i n o acid alignment of C-terminal segments of P A 0 1 P i l B and PilT homologues, accomplished by using ClustalW. The numbering is based on the H. influenzae P i l B sequence. Conserved residues are enclosed in boxes, and identical residues shaded. The alignment begins in the regions corresponding to residue 150 of H. influenzae P i l B where the sequences begin to align well. The boxes and shaded region within the last three sequences indicate the conserved regions unique to PilT family proteins. The following proteins are shown: H i P i l B (H. influenzae P i l B ) , A a P i l B (Actinobacillus actinomycetemcomitans P i l B ) , PaPilB (P. aeruginosa P i l B ) , A c P i l B (Acinetobacter sp. Strain B D 4 1 3 P i l B ) , N g P i l F (N. gonorrhoeae P i l F ) , E c H o f B (E. coli HofB), PaPilT (P. aeruginosa PilT), A c P i l T (Acinetobacter sp. Strain B D 4 1 3 PilT), NgPilT (N. gonorrhoeae PilT).  52  B L A S T searches with the well characterized nucleotide binding proteins P i l B , PilT and P i l U of P. aeruginosa and N. gonorrhoeae indicated that P i l B is the only Tfp-related ATPase in H. influenzae (Section 1.6). Efficient D N A uptake by H. influenzae in the absence of a PilT homologue indicates that D N A uptake does not require a PilT-like ATPase. This has several possible explanations. First, P i l B may function in both assembly and reaction of the Tfp-like apparatus. In this case, P i l B may drive the two opposite processes (assembly/polymerization and retraction/depolymerization) under different conditions. The second explanation is that P i l B only powers Tfp assembly. In this case, a non-VrbB/GspE-like protein could power the retraction of the Tfp-like machinery by either A T P hydrolysis or other process that can generate energy. Alternatively, H. influenzae might need no other proteins for Tfp retraction; that is, Tfp might automatically depolymerize and retract when the assembly protein P i l B is not active. Thus the transforming D N A would be taken up during this process. The two alternative hypotheses (whether P i l B is required for both Tfp assembly and retraction, or only required for Tfp assembly), can be tested by complementing a H. influenzae pilB knockout with pilB/pilT genes from a well-characterized Tfp system, which is the major approach of this study.  In this analysis I have assumed that pilus retraction is needed for D N A uptake by H. influenzae. The validity of this assumption is reconsidered in the discussion.  53  3.2 Construction of  H. influenzae  pilB  knockouts  3.2.1 Overview Two H. influenzae pilB knockouts, RR1137 and RR1150, were constructed in this study. As described below, inserting a K a n cassette at the Bell site of the pilB O R F generated R  the pilB knockout strain RR1137. However, complementation experiments and real time PCR  showed that the pilB knockout in RR1137 is seriously polar to the downstream  genes pilC and pilD (i.e. a mutation or knockout in one gene reduces the expression of the downstream genes) (more details below). This polarity was eliminated by placing an additional C R E regulatory sequence and a promoter before pilC in the pilB::Kan  R  background, giving strain RR1150. The relevant genotypes of RR1137 and R R 1 1 5 0 are illustrated in Fig.3.2.  Construction of the necessary plasmids is described in detail in the Methods. The plasmids used for generation of RR1137 were p P I L B and p P I L B : : K a n . The plasmids R  used for generation of RR1150 were p P I L B : : K a n , p P I L C and p B K a n C R E C . p S U P I L B R  was used to test polarity of the pilB mutations in both RR1137 and RR1150.  54  RR1137 pilA  plB::KsrP  pMC pMD  F i g . 3.2 Genetic organization of K W 2 0 , RR1137 and RR1150. P  pil  = C R E promoter of  the pil operon.  55  1  3.2.2 Construction of pPILB and pPILB::Kan  R  To investigate H. influenzae pilB's function, a pilB knockout was generated using a Tn903 K a n insertion. First, a plasmid carrying the pilB gene was constructed. A 2326 R  bp fragment encompassing H. influenzae pilB O R F and the flanking sequences was P C R amplified and cloned into a p G E M - T easy vector, resulting in p P I L B .  To knock out pilB, a K a n cassette was inserted into the pilB gene on p P I L B at the Bell R  site. The knockout plasmid was designated p P I L B : : K a n . R  3.2.3 Construction of RR1137  (pilB::Kan ) R  The knockout was then introduced into H. influenzae K W 2 0 by natural transformation, replacing the wild-type copy of pilB. T w o colonies carrying the desired chromosomal organization were examined for exponential growth rate in s B H I and the ability to develop competence using a chromosomal D N A  transformation assay. W i l d type K W 2 0  was used as the control. N o discernable difference of growth rate was observed between the strains (data not shown). Since H. influenzae pilB codes for a nucleotide binding protein predicted to be required for the biogenesis and/or function of the Tfp-like machinery, the knockout of pilB was expected to give no transformation due to lack of a Tfp-like machinery. N o transformants were obtained from either isolate, giving transformation frequencies less than 9.5 x 10" , which was 8.9 x 10 lower than that of the 9  5  wild type. One colony was frozen as strain RR1137.  56  The elimination of transformation could have resulted either from the disruption of pilB itself pr from polarity, which significantly reduced expression of the downstream genes pilC and pilD, another two genes expected to be required for Tfp biogenesis (Section 1.5). In the later case, the lack of competence in RR1137 could not be attributed to the knockout of pilB. The polarity of the pilB knockout in RR1137 was examined below.  3.2.4 Complementation assays and real time P C R revealed severe polarity of RR1137  pilB knockout  T o assess whether the pilB knockout in RR1137 was polar on the downstream genes, a plasmid expressing a w i l d type pilB was needed. Full complementation of the knockout by this plasmid would imply no polarity.  T o generate a plasmid that can replicate in H. influenzae and contains a w i l d type pilB, the 2346 bp EcoRl fragment of p P I L B containing pilB was subcloned into the H. influenzaelE. coli shuttle vector pSU20, giving p S U P I L B . p S U P I L B was then introduced into RR1137 by chemical transformation, as RR1137 is defective in natural transformation. The plasmid-carrying knockout strain RR1138 was then used in the transformational complementation assays. RR1137 and K W 2 0 were included as negative and positive controls. Due to addition of Cam to keep p S U P I L B , growth of RR1138 was slightly slower than that of RR1137 and K W 2 0 (data not shown). Therefore another strain RR1136 (wild type strain containing p S U P I L B ) was included as a positive control in the second assay. O n average of the 4 transformation assays, introduction of p S U P I L B into RR1137 only restored transformation frequency to 1.75 x 10" , which is about 1800 6  57  fold lower than that of the wild type K W 2 0 (Fig. 3.3). This suggests that the pilB knockout in RR1137 may be strongly polar on pilC and pilD.  1.0E-02 1.0E-03  c 0 1.0E-04 3 CT 01 1.0E-05 c o  1.0E-06  m  4 J  E 1.0E-07  I  1.0E-08  jo  1.0E-09  (A e  H  1.0E-10 KW20  RR1137  RR1138  RR1136  3£x103  9£x10^  IBxIO*  63x10^  F i g . 3.3 Transformation tests of the pilB knockout in RR1137 complemented with wild type pilB.  58  To confirm that lack of complementation was due to polarity, real time P C R was used to measure the m R N A levels of pilC and pilD. Since the real time P C R primer pair for pilA was already available in the lab (MN#403), and p S U P I L B contains pilA in addition to pilB, the pilA gene, instead of pilB was analyzed to examine the expression of the genes on p S U P I L B . The gene murG, whose expression is largely constant (74), was used as an internal control. The result showed that the average expression levels of pilC and pilD in RR1137 were only 6.3% and 5.8%, respectively, of the same genes in the wild type K W 2 0 , indicating that the pilB knockout in RR1137 has serious polar effect on the downstream genes pilC and pilD. Thus I could not conclude that its competence defect is due to lack of P i l B .  3.2.5 Construction of a non-polar  pilB  knockout RR1150  T o circumvent the polar effect of the pilB insertional knockout on pilC and pilD, a competence inducible C R E promoter (see Section 1.3) was inserted in front of pilC (Fig. 2.5), downstream ofthe K a n insertion in pilB. The strain constructed using this strategy R  was designated RR1150.  To add an additional C R E promoter before pilC, a 1384 bp fragment containing pilC coding sequence was cloned into a pSU20 based vector containing a C R E promoter (Fig. 2.3) to get the plasmid p P I L C . Subsequently, the pilC gene, together with the C R E promoter right before it, was subcloned into a vector containing a truncated pilB with a Tn903 K a n insertion at the Bell site (Fig. 2.5). The resultant plasmid p B K a n ( C R E C ) 5 R  thus constructed contains a fragment bearing a truncated pilB with a K a n + C R E R  59  promoter + pilC (Fig.  3.4). The fragment was then released by  digestion and introduced into  Mlul/Spel double  H. influenzae K W 2 0 by natural transformation, giving the  strain RR1150.  3.3 Natural competence was eliminated in RR1150 Growth and natural competence of RR1150 were tested using the w i l d type K W 2 0 as a control. Although the integrity of the  pilB coding sequence was disrupted by the K a n  R  insertion, R R 1 1 5 0 did not exhibit discernable difference in growth rate from that of the wild type (data not shown). L i k e RR1137, the transformation frequency was reduced to background level 5.6 x 1 0 , more than 10 fold lower than that of the wild type K W 2 0 9  5  (9.0 x 10" ). 3  3.4 RR1150 pilB knockout was complemented bv wild type To confirm that insertion of a C R E  promoter before  pilB  pilC eliminated polarity, the pilB  knockout in R R 1 1 5 0 was tested for complementation by the /?//5-expressing plasmid p S U P I L B . A strain RR1151 was constructed by transforming plasmid p S U P I L B into RR1150. Transformation assays were performed using R R 1 1 5 0  (ApilB::Kan /CRE R  IpilC), RR1151 (RR1150 complemented with p S U P I L B ) , K W 2 0 , and RR1136 ( K W 2 0 carrying p S U P I L B ) . The result showed that RR1151, with the w i l d type  pilB introduced  on the plasmid p S U P I L B , had an average transformation frequency of 6.2 x 10" , which is 4  only 6.6 fold lower than the average transformation frequency of 4.08 x 1 0 of K W 2 0 3  (Fig.  3.5).  60  F i g . 3.4 Construction of RR1150. In construction of p B K a n C R E C , restriction analysis showed that a plasmid p B K a n ( C R E C ) containing n  >5 tandem insertions of the 1537 bp CREC/pilC fragment was obtained instead of p B K a n C R E C . Since p B K a n ( C R E C ) carries the final n  fragment for a non-polar pilB knockout construction, it was used to generate the strain RR1150.  61  1.00E-02 g< C  1.00E-03  o 2.  l.OOE-04  a c  1.00E-05  «  1.00E-06  ti  1.00E-07  £  1.00E-08  or o  1 o c ra  1.00E-09 KW20  43x103  RR1150 6A»W-  RR1151 9  tirixlft  RR1136 4  (SLIXIO"  3  F i g . 3.5 Transformation tests of the pilB knockout in RR1150 complemented with wild  typtpilB.  Although the transformation frequency of the complemented pilB strain RR1151 is close to that of the wild type, it is still considerably lower than expected for a full complementation. T o examine whether the regulatory sequence added next to the 5 ' of pilC is correct, the region of C R E and the promoter in p P I L C was sequenced using the primer PilBr. A n A - t o - G substitution was detected in the - 1 0 area of the putative promoter (46), which could explain the low complementation. The substitution is also expected to be present in the C R E promoter of pilC in RR1150. This substitution may  62  have reduced the expression of the downstream gene pilC and pilD, which was supported by the later real time P C R that showed the expression of pilC in R R 1 1 5 0 was about 3.7 fold lower than that of K W 2 0 (Table 3.2). Thus the incomplete complementation observed in RR1151, was still caused by polarity, although indirectly. Nevertheless, introduction of a wild type pilB into RR1150 has restored the transformation more than 1 0 fold, suggesting the pilB knockout in RR1150 is sufficient to account for the s  transformation deficiency of RR1150. This strain was used for later experiments and for construction of derivative strains.  Later sequencing analyses indicate that the A - t o - G substitution found in the - 1 0 segment of p P I L C was originated from p P I L B , the very first clone of H. influenzae pilB fragment (Fig. 3.6, Fig. 3.7). Therefore the pPILB-derived p S U P I L B , used for complementation of RR1150, should have the substitution as well. This was confirmed later by D N A sequencing of p S U P I L B .  63  Expected promoter sequence ofpUA from TIGR (wild type)  °s^n^^inpMLB  T  sequenced pSUPnS  TTTT^KggGTCGTTGCC^^CARAGGCTTAATARAflGGARARTGfiJCTGAAATTAA,C -10 transcription RBS start start codon  TTTTEjffiQiiffiGTCGTTCXJC^C^^ -10 transcription start  RBS  start codon  TTTQBiCgilffllGTCGTT^ -10 transcription start  RBS  start codon  RBS  start codon  -10  traiiscriptioii start  F i g . 3.6 The expected and observed sequences around the - 1 0 region of pil promoter in the wild type and i n p P I L B , p S U P I L B and p P I L C , showing the A - t o - G substitution.  64  KW20 chromosome  H. Influenzae plasmids pPILB::kanR  pilB::Kan  Chromosome „  R  BKanCRE'  pBKan(CREC)5  CRE pPILB  promoter^  R11S0  CRE  CRE  promoter  promoter  _  pPILC  _  pSA Tiuncated \pHC  pilD  From pPILB::Kisn  /jtffffrag.  CRE  iromot er  pPAPBpro  pPAPT2  CRE  CRE +  PA/#/2  CRE  pPAPBT  FrompPILC  CRE  iromot er  pPAPT  k  R  wild type pt/B ( C o m p l e m e n t a t i o n )  pSUPILB  CRE  pPAPB  PA/J«?^  PPAPBT2  P. aeruiHftosa.plasmids Fig. 3.7 Relationships between the plasmids of this study and the strain RR1150.  Table 3.2 Expression level of pilC in the non-polar pilB knockout compared to that of the wild type.  pilC expression in competent KW20  Real Time PCR I  Real Time PCR II  129  146  Average pilC expression in 33 RR1150 complemented with Pa  42.8  pilB, T  pilC expression in RR1150 relative to that of KW20  Average 1 : 3.7  Note: (1) The quantities of the expression were normalized by the gene murG, the expression of which is constant. (2) The pilC expression of the knockout came from the expression data of the P. aeruginosa pilB, 7 complemented RR1150.  (3) Pa = P. aeruginosa  65  3.5 Transformation defect of RR1150 is at the level of DNA uptake A s mentioned in the Introduction, natural competence includes steps in which D N A is bound, taken up, translocated across the outer and inner membranes and sometimes recombined with its homologous counterpart in the genome. A defect at any of these steps could be responsible for the transformation defect of RR1150. However, D N A binding and uptake would be expected to be normal i f the defect occurs in the downstream steps such as translocation or recombination. A s mentioned in Section 3.1,  H. influenzae P i l B was predicted as a Tfp-related ATPase responsible for assembly and/or retraction of the putative Tfp-like apparatus, which in other bacteria acts as D N A uptake machinery. R R 1 1 5 0 is expected to be deficient in D N A binding and uptake. T o test this, D N A binding and uptake assays were performed for RR1150, using P-end33  labeled Hinfl fragments of K W 2 0 D N A . M l V - i n d u c e d K W 2 0 and RR1151 (RR1150 complemented with p S U P I L B ) were used as positive controls for D N A binding and uptake, and non-competent K W 2 0 cells (in exponential phase) were used as a negative control.  According to the current model, Type I V pili (or pseudopili) provide the structural basis for D N A binding, whereas retraction of Tfp drives D N A uptake. In H. influenzae, P i l B is predicted to be required for assembly, or for both assembly and retraction of the Tfp-like machinery. In either case, knocking out pilB would result in a failure in Tfp assembly. Therefore no D N A binding & uptake is expected in RR1150. The assays of D N A binding plus uptake showed that the amount of D N A associated with R R 1 1 5 0 cells was ~  66  500 fold lower than that of the wild type K W 2 0 , and was not significantly different form that of the negative control (Fig. 3.8). This result indicates that R R 1 1 5 0 can neither bind nor take up D N A . The D N A binding and uptake of RR1151 (RR1150 complemented with p S U P I L B ) is ~7 fold lower than that of the wild type, but ~57 fold higher than that of RR1150. The defect in RR1150 suggests that the pilB knockout strain was unable to assemble a Tfp-like apparatus, thus leading to the strongly decreased D N A binding and uptake.  E  a.  1.00E+06  w  a 3 c  1.00E+05  QJ ro  a  •a 1 . 0 0 E + 0 4 c  3  o .o Q  1.00E+03  Q.  m m ro 4-1  o  H  1.00E+02 KW20  RR1150  RR1151  Neg. control  F i g . 3.8 D N A binding and uptake of RR1150. After incubation in M I V medium for 100 min, cells were incubated with c t - P - d A T P labeled K W 2 0 D N A . DNase I was not added 33  so that total cell-associated D N A ( D N A bound or taken up) could be measured. Competent K W 2 0 and RR1151 (RR1150 containing p S U P I L B ) were used as the positive controls, and non-competent K W 2 0 (cells in log-phase) was used as the negative control.  67  3.6 Introduction of P. aeruginosa  pilB and pilT into the H.  influenzae  pilB knockout A s mentioned above, P i l B is the only H. influenzae R d ATPase-like protein in the P i l B , PilT and P i l U family. If all the bacteria that have Tfp or Tfp-like machinery share similar mechanism in D N A binding and uptake, either H. influenzae P i l B performs the functions of both Tfp biogenesis (required for D N A binding) and retraction (required for D N A uptake), or P i l B only functions in Tfp assembly and another as-yet-uncharacterized protein is responsible for Tfp retraction.  3.6.1 Overview Complementation of pili genes and type II secretion genes with foreign homologues has been reported (21, 30, 42, 68). P. aeruginosa P i l B / P i l T and N. gonorrhoeae P i l F / P i l T are the only well-characterized ATPases systems required for Tfp biogenesis/retraction. Both P. aeruginosa P i l B and N. gonorrhoeae P i l F share high similarity with H. influenzae P i l B (identities 38% and 4 1 % , B L A S T E value 4e-89 and 2e-73) (see Table 3.1). Since P. aeruginosa P A O l P i l B and PilT are a closer homologues of H. influenzae PilB than N. gonorrhoeae P i l F and PilT, they were chosen for complementation of the H. influenzae pilB knockout R R 1 1 5 0 and for identification of the specific role of H. influenzae P i l B in D N A uptake. If H. influenzae P i l B is needed for both pseudopilus elongation & retraction, complementation of the pilB knockout would not be observed unless both P. aeruginosa pilB and pilT have been introduced (Table 3.3). If P. aeruginosa P i l B can complement the assembly defect of H. influenzae pilB, then introduction of P. aeruginosa pilB might restore D N A binding but not uptake in RR1150.  68  On the other hand, i f another H. influenzae protein is responsible for Tfp retraction, introduction of P. aeruginosa pilB would be able to restore both D N A binding and uptake in the H. influenzae pilB knockout. However, i f P. aeruginosa pilB and pilT cannot  interact with the H. influenzae Tfp genes, no complementation would be observed.  Table 3.3 Hypothesis about complementation of H. influenzae pilB with P. aeruginosa P A O l pilB & pilT  Introduced P. aeruginosa gene P. aeruginosa pilB and pilT  able to complement  Transformation  DNA DNA binding uptake  Hypothesis I H. influenzae pilB  pilB  -  +  pilT  -  -  -  pilB + pilT  +  +  +  H. influenzae pilB  pilB  +  +  +  required only for Tfp assembly  pilT  -  -  -  pilB + pilT  +  +  +  pilB  -  -  -  -  -  -  required for required for both assembly and retraction Hypothesis II  P. aeruginosa pilB and pilT cannot  interact with H. proteins  influenzae  Tfp  pilT pilB + pilT  69  P. aeruginosa P A O l pilB and pilTv/txt  cloned using similar strategies, placing each gene  under the control of the C R E promoter of the H. influenzae pil operon (Section 2.2.3). However, two possible start codons are found in the 5' of both P. aeruginosa pilB and pilT (Fig. 3.9) (87). Sequence alignment of P i l B homologues indicated that A T G , rather 0  than G T G  1 1 7  , is the real start of translation of pilB, because the region from 0-117 bp is  well conserved in homologues. Analysis for PilT, however, could not identify the true start codon of pilT from the two codons A T G and G T G . , because only 4 amino acids 0  1 2  were involved. The sequences upstream of both pilT putative start codons were also examined for ribosome binding sites. A n obvious consensus sequence for a possible ribosome binding site could not be found. Consequently, two pilT plasmids were constructed, each using one of the candidate start codons.  Three types of P. aeruginosa plasmids were generated, containing pilB ( p P A P B ) , pilT (pPAPT and p P A P T 2 ) , or both ( p P A P B T and p P A P B T 2 ) , and were transformed into the H. influenzae pilB knockout RR1150.  Plasmids containing P. aeruginosa pilB and pilT  are listed in Table 3.4.  70  Putative coding seq. 1 (1701 bp) ATGo GTG117  ^ ~  Putative coding seq. 11(1584 bp)  Putative coding seq. 1 (1035 bp)  PA01 pilT:  A T G o  ^  T T G -12  W  Putative coding seq. 11 (1047 bp)  F i g . 3.9 Graphic presentation of the two putative coding sequences of P. aeruginosa pilB and pilT.  Table 3.4 Constructed P. aeruginosa plasmids. No.  plasmids  P. aeruginosa gene  1  pPAPB  pilB  2  pPAPT  pilT  Starts at A T G  3  pPAPT2  pilT  Starts at T T G  4  pPAPBT  pilB and pilT  pilT starts at A T G „  5  pPAPBT2  pilB and pilT  pilT starts at T T G .  Start codon  0  1 2  1 2  71  3.6.2 Cloning of  P. aeruginosa  pilT  3.6.2.1 C o n s t r u c t i o n of p P A P T a n d p P A P T 2 Fragments of 1100 bp (pilT, from start codon A T G ) and 1112 bp (pilTZ, from G T G , ) 0  2  containing either possible P. aeruginosa pilT coding sequences were amplified from chromosomal D N A (Fig.2.6) and cloned into a pSU20 based vector containing a C R E promoter of the H. influenzae pil operon, to get the plasmids p P A P T and p P A P T 2 .  A s indicated by D N A sequencing, an A - t o - G substitution in the putative - 1 0 region of the promoter was found as well in both p P A P T and p P A P T 2 , as was found in the p P I L C construct (Section 3.4) (Fig. 3.6, Fig. 3.7 and Fig. 3.10). A s shown in the polarity examination of R R 1 1 5 0 (Section 3.4), the A - t o - G substitution did not cause dramatic decrease in the expression of the downstream gene, because the transformation frequency of the complemented strain RR1151 (RR1150 complemented with p S U P I L B ) was only 6.6 fold lower than that of the wild type but was more than 10 fold higher than that of the 5  pilB knockout. The p P A P T and p P A P T 2 constructs were used afterwards for construction of p P A P B T (containing P. aeruginosa pilB and pilT that starts at A T G ) , 0  p P A P B T 2 (containing P. aeruginosa pilB and pilT that starts at G T G ) and for 1 2  complementation of the H. influenzae pilB knockout.  72  non-polar ti.  Wild type  H. influenzae'  pSUPILB  ti. influenzaepilB  influenzaepiiB  £51  pPAPT2  H. influenzae p$C IJI _ i *Jtid»  pPAPT  P. aeruginosa piiB  SSL  pilC  l3h  pPILC  ti. influenzae pilB pPAPB  pflB  P. aeruginosa pilT pPAPBT  —i-sx^>—1^— PapilT PapflB  P. aeruginosapflT  pPAPBT2 PapilT PapilB  F i g . 3.10 The strains and plasmids with or without the A - G substitution in the putative  - 1 0 segment (indicated by x ). Pa = P. aeruginosa.  3.6.2.2 C o n s t r u c t i o n o f p P A P B The strategy used to construct p P A P T and p P A P T 2 was initially also used to construct the  P. aeruginosa pilB plasmid p P A P B . However, sequencing of the constructs revealed  that deletions occurred at the ligation joint of the end of the  pilA promoter and start of  pilB. The problem was not resolved by several modifications in experimental procedures. Therefore toxicity of P.  aeruginosa pilB to the host cells DH5ct was suspected. The  original cloning strategy was modified by shifting the breakpoints in the promoter and in  pilB (Fig. 2.7, Fig. 2.8). Instead of cloning only the coding sequence of pilB, a fragment extending 200 bp upstream was amplified from  P. aeruginosa chromosomal D N A and  73  inserted into a pSU20 based vector containing a shortened version of H. influenzae pilA C R E promoter (simplified as p S U - C R E vector below). The unwanted P. aeruginosa pilB promoter was then removed, so that the final plasmid p P A P B contained the P. aeruginosa pilB gene and R B S under the H. influenzae pilA promoter.  The sequence of pilB and the upstream C R E promoter in p P A P B #6 was examined by D N A sequencing using primers SpBEf, SpBr, SpBf and P i l B r (See Section 3.6.2.4 below). p P A P B #6 was used for heterologous complementation tests and for construction of p P A P B T , p P A P B T 2 , as described below.  3.6.2.3 Construction of pPAPBT and pPAPBT2 To get plasmids that express both P. aeruginosa pilB and pilT, the inserts of p P A P T and p P A P T 2 were released with Nhel/Spel double digestion, and subcloned into Spellinearized p P A P B (Fig. 2.9). The resultant plasmids were designated p P A P B T and p P A P B T 2 , respectively.  3.6.2.4 Sequencing of the P.  aeruginosa  plasmids in E.  coli  A l l five P. aeruginosa plasmids (pPAPT, p P A P T 2 , p P A P B , p P A P B T and p P A P B T 2 ) were examined by D N A sequencing after they were constructed in DH5ct. A s in p S U P I L B , the source of the C R E promoter, both p P A P T and p P A P T 2 bears the same A to-G substitution that tuned the original putative - 1 0 from C A C A A T to C A C G A T . For unknown reason, however, the - 1 0 area in p P A P B was restored to wild type C A C A A T , although the C R E promoter originated from p S U P I L B (Fig. 3.5, F i g . 3.6). 2 bases ( C C ) were also lost at the joint of the ligation in p P A P B (Fig. 3.11). However, as mentioned in  74  Section 2.2.3.2, the number of bases between the transcription start and R B S was determined arbitrarily in p P A P B , because the sequence in this area plays no significant role in the initiation of gene expression. The 2 bases deletion should not significantly affect the expression of pilB.  The other two plasmids p P A P B T and p P A P B T 2 that are derived from p P A P B and p P A P T , consequently, both of the pilT promoters bear the substitution, whereas both of the pilB promoters are wild type (Fig. 3.9). Nevertheless, the rest of the sequences of all the 5 plasmids have no changes.  Deleted bases ,  p  UwPA^ done U  Sequence of the resultant PA pilB clone  transcription / start -10 start I / RBS codon ' ^ - ^ E S S B ^ S G T v . G G T T. GCCSftC AAAGGST.T C5^S^®T T C C TT.CC C G^ISSAC G AC AG H. infhien&e promoter . pilB coding seq. Sequence from H. mfhienzae ' > Sequence from P A O l ligation joint  I  T  T T  rransmptioii™ start -IQ . start . RBS codon TBlC^^GT-CGT-tGCC«ftGAAAGp.T CTiSCSMT T.CCTT CCCQATClAACGACAG hi. influenzae, promoter I piiB coding seq; restoi-ed base  F i g . 3.11  The sequence of p P A P B #6 at the ligation joint, showing the C C 2 base  deletion and the restored wild type - 1 0 sequence.  75  3.6.3 Construction of H. influenzae knockout and the  P. aeruginosa  strains bearing both the  pilB  plasmids  The non-polar H. influenzae pilB knockout in RR1150 prevents transformation, so cells carrying both the pilB knockout and potentially complementing plasmids were constructed indirectly by first transforming the plasmids into the wild H. influenzae strain K W 2 0 and then introducing the chromosomal pilB knockout by transformation (Fig. 2.10). The resultant strains are R R 1 1 5 6 (RR1150 complemented with p P A P T ) , RR1157 (RR1150 complemented with p P A P B ) , RR1158 (RR1150 complemented with p P A P B T ) , RR1159 (RR1150 complemented with p P A P T 2 ) and RR1197 (RR1150 complemented with p P A P B T 2 ) .  Because of the severe interference caused by the M n d l l l restriction system in H. influenzae, great difficulties were experienced in the earlier attempts to transform the plasmids p P A P B , p P A P B T and p P A P B T 2 into H. influenzae (data not shown), as these three plasmids contain a number of M n d l l l sites. T o prevent the digestion of the plasmids by M n d l l l , they were treated with methylase Hindlll before transformation into KW20.  3.6.4 Sequences of the P. aeruginosa  plasmids in H.  influenzae  pilB  have  no mutations All the 5 P. aeruginosa plasmids in the H. influenzae pilB background were resequenced. N o changes were found compared to the plasmids from E. coli (Section 3.6.1.4).  76  3.6.5 The P. aeruginosa  genes were strongly expressed in H.  influenzae  pilB To examine the expression of the P. aeruginosa pilB, pilT genes in H. influenzae, real time P C R was performed for the complemented H. influenzae pilB knockouts. T o induce CRE-regulated genes, exponentially growing cells were transferred into M I V  and  incubated as usually done for competence induction. The expression of P . aeruginosa pilB and pilT was measured by real time P C R of c D N A reverse transcribed from samples of the above M I V  RNA  cells. The result showed that both pilB and pilT were strongly  expressed in H. influenzae pilB (Fig.  3.12 and Fig.  3.13).  77  l.OOE+03  ^ O c o "55  1.00E+02  2  l.OOE+01  (A  &  X  OJ OJ  > •5  l.OOE+OO  l.OOE-01 wt pilC  50/pB  50/pBT  50/pBT2  F i g . 3.12 Expression level of P. aeruginosa pilB. in the H. influenzae pilB background relative to the expression of H. influenzae pilC, assuming the expression of pilC is 1. The expression level of P. aeruginosa pilB was compared to that of H. influenzae pilC in the same strain (e.g. the expression of P. aeruginosa pilB in RR1158 was compared to the pilC expression in RR1158 according to the difference in the cycle numbers). In order to know the expression of P. aeruginosa pilB relative to a wild type H. influenzae competence gene, the expression of pilC was normalized to the wild type level by subtracting 1.87 from its cycle number, because the expression of pilC in the pilB knockout is ~3.7 fold lower than that of K W 2 0 (Section 3.4). Assuming ~ 2 fold/cycle difference of P C R amplification, the cycle number difference between RR1150 pilC and a wild type pilC expression was obtained by the formula 2 difference, d  e x p  d C l  =d  e x p  (dCt = cycle number  = difference in expression).  78  1.00E+03  0- 1.00E+02  co  '35  (A OJ l.OOE+01  o. x L.  OJ OJ  > iS  l.OOE+00  l.OOE-01 wtpilC  50/pT  50/pT2  50/pBT  50/pBT2  P. aeruginosa pilT in the H. influenzae pilB background relative to the expression of H. influenzae pilC, assuming the expression of pilC is 1. The expression level of P. aeruginosa pilB was compared to that of H. influenzae pilC in the same strain (e.g. the expression of P. aeruginosa pilT in RR1156 was compared to the pilC expression in RR1156 according to the difference in the cycle numbers). The expression of pilC has been normalized to the level of pilC in a w i l d type. See F i g . 3.11 F i g . 3.13 Expression level of  for explanation.  79  3.7 The transformation defect of H. influenzae restored bv  P. aeruginosa  pilB  and  pilB  could not be  pilT  If P. aeruginosa pilB and pilT are together functionally equivalent to H. influenzae pilB,  and can interact with other H. influenzae Tfp genes, the defect of pilB should be restored effectively by introduction of P. aeruginosa pilB, pilT into pilB. Transformation assays were employed using the strains RR1156 (RR1150 complemented with p P A P T ) , RR1157 (RR1150 complemented with p P A P B ) , RR1158 (RR1150 complemented with p P A P B T ) , RR1159 (RR1150 complemented with p P A P T 2 ) and RR1197 (RR1150 complemented with p P A P B T 2 ) . K W 2 0 , R R 1 1 5 0 (ApilB::Kan /CRE/pilC) R  and RR1151 (RR1150  complemented with p S U P I L B ) were used as controls. N o complementation was observed in any of the constructs (Fig. 3.14). In the several transformation tests performed, the non-polar pilB knockout complemented by the plasmid copy of wild type H. influenzae pilB, R R 1 1 5 1 , had a transformation frequency approximately 6.8 fold lower than that of K W 2 0 , confirmed the previous result in polarity examination of RR1150 (Section 3. 2.4). The transformation frequencies of the H. influenzae pilB knockout complemented with P. aeruginosa pilB and pilT, however, were all below 8.5 x 10" , 9  which are not significantly different from that of the uncomplemented strain  RR1150 ( s 8 . 1 x 10" ). 9  80  l.OOE+00  &  1.00E 01  e l.OOE 02 3 cr l.OOE 03 c l.OOE 04 o ro l.OOE 05 c c  £  l.OOE 06  c l.OOE 07 KS k. 1-  l.OOE 08 '8  l.OOE 09 KW20  4J6xlD-3  pilB  44xlO*  ii'  pilB/Hi pILB pilB/PA pilB pilB/PA PilT  74xl0-»  5SxlO*  8.8x10*  pilB/PA pilT2  pilB/PA pilBT  pilB/PA pilBT2  5.5xlO-»  1J8X10-8  54x10-10  F i g . 3.14 Transformation assays of the P. aeruginosa pilB, pilT complemented H. influenzae pilB. The numbers below the figure are the average transformation frequencies.  3.8 The defect of DNA binding and uptake of H.  influenzae pilB could  not be complemented bv P. aeruginosa pilB and pilT If introduction of P. aeruginosa genes only restores the Tfp biogenesis function of H. influenzae pilB, but not the retraction of the Tfp-like machinery, then the cells containing the above constructs might only bind D N A but not take it up. T o examine this, D N A binding experiments were performed using P - d A T P end-labeled M n f l fragments of 3 3  K W 2 0 D N A . None of the five complemented H. influenzae pilB strains had detectable D N A binding (Fig. 3.15). The radioactivities associated with the cells of the five strains and RR1150 were at similar levels, which all were more than 10 fold lower than that 2  associated with K W 2 0 cells.  81  ^> E  1.00E+06  a u  V ro a 3  TJ  c  1.00E+05  1.00E+04  ro  c •a c  < z Q  1.00E+03  1.00E+02 KW20  RR1150  RR1151  50/pB  50/pT  50/pT2  50/pBT  50/pBT2 Negative control  F i g . 3.15 D N A binding assays of the P. aeruginosa pilB, pilT complemented H.  influenzae pilB.  3.9 Discussion The pil operon is required for natural transformation in H. influenzae R d (23). T o investigate the specific mechanism of Tfp mediated natural competence, the second gene of this operon, pilB, was chosen as the focus of this study. Sequence analysis predicts P i l B to be a G s p E family ATPase responsible for Tfp biogenesis (49). However, the absence of a P i l T homologue in H. influenzae led to several speculations about P i l B ' s function (Section 3.1).  Inactivation of H. influenzae pilB with a non-polar K a n insertion, eliminated natural R  competence. The defect was because the mutant was unable to bind and take up D N A . When the pilB knockout was complemented with the homologous genes pilB and pilT  82  from P. aeruginosa, neither transformation nor D N A binding and uptake were restored, although the P. aeruginosa genes were strongly expressed.  P. aeruginosa and N. gonorrhoeae are the only species where the roles of P i l B & PilT are well established. P. aeruginosa P i l B and PilT are closer homologues of H. influenzae P i l B than N. gonorrhoeae P i l F and PilT (Table 3.1, Table 3.5). Therefore they were chosen for complementation of H. influenzae pilB.  Complementation with a heterologous protein has been commonly used for investigation of a protein's function. The probability of producing a functional hybrid machinery depends on the phylogenetic distance between the organisms and the extent of the similarity between the proteins involved. In Type II secretion systems (T2ss), most components have exchangeable examples except G s p C (a component unique to T2ss) and GspD (a P i l Q homologue) (30). In Tfp systems, there are several reported cases of complementation by heterologous proteins, including pilin, peptidase and G s p E family ATPases (21,42, 68, 80). A cosmid clone bearing Pseudomonas syringae pv. Tomato DC3000 D N A was able to complement a mutation of the pilB gene in P. aeruginosa (75). In Aeromonas hydrophila, the gene tapB was identified to be a homologue of P. aeruginosa pilB by its ability to complement a pilB mutation in P. aeruginosa (68),  although A . hydrophila is not more closely related to P. aeruginosa than H. influenzae (Table 3.5). Another Tfp protein that can substitute the homologue in other species is peptidase (PilD homologues), probably because its function is solely enzymatic and P i l D does not need to interact with other proteins (21, 30, 68, 75). The most permissive T2ss  83  and Tfp components are pilins and pseudopilins, which can often be exchanged, more or less efficiently (30, 42, 80).  Table 3.5 Taxonomic relationships between Haemophilus influenzae, Pseudomonas  aeruginosa, Nesseria gonorrhoeae and Aeromonas hydrophila (adapted from N C B I ) . Bacteria  Class  Order  Neisseria gonorrhoeae  B etaproteobacteri a  Neisseriales  Aeromonas hydrophila  Gammaproteobacteria  Aeromonadales  Haemophilus influenzae  Gammaproteobacteri a  Pasteurellales  Pseudomonas aeruginosa  Gammaproteobacteria  Pseudomonadales  84  However, complementation does not often occur, despite the conservation of the Tfp systems and Type II secretion systems (30, 78), suggesting that there is specificity within the machinery itself. The situation of G s p E like ATPases appears to be especially complicated. In addition to the nucleotide binding motifs in the C-termini that are thought to perform the enzymatic function, the N-termini may interact with other proteins, thus determining species specificity (30). Sandkvist showed that T2ss V. cholerae E p s E (GspE homologue) is associated with and stabilized by the cytoplasmic membrane via interaction with E p s L (79). In P. aeruginosa, localization of P i l B to both poles of the cells requires P i l C , suggesting that P i l B and P i l C interact (15). Therefore, in addition to the phylogenetic distance of the two organism involved, the functional interactions between the introduced protein and other members of the Tfp machinery in the organism are also important constraints on heterologous complementation.  In P. aeruginosa, although the pilB mutation was complemented by Aeromonas hydrophila tapB, the two proteins P i l B and TapB are substantially more similar than the H. influenzae & P. aeruginosa PilBs. The failure of P. aeruginosa P i l B and PilT to complement the H. influenzae pilB knockout, therefore, is likely to be at least partly because the P. aeruginosa homologues could not interact with H. influenzae P i l C to form a functional hybrid Tfp machinery.  However, the result of a recent work and our sequence analyses suggest that the reason of the non-complementation of P i l B in this study may also be because the difference in the Tfp energetics of H. influenzae and P. aeruginosa.  Another H. influenzae strain has been  85  recently shown to express functional Tfp and twitching motility (7), the first report of type IV pili in H. influenzae. It was also shown that the Type I V pili in this strain are required for twitching motility, which in other Tfp systems requires pilus retraction. However, a PilT protein was not found in this PL. influenzae. This raised the question of which protein would drive the retraction required for twitching motility. Since a PilT is absent in H. influenzae R d as well, I was interested to see whether there is a PilT in A. pleuropneumoniae and P. multocida, the other two Pasteurellaceae species that possess type I V pili (76, 102). Similarly, a PilT was not found.  PilT proteins are required for Tfp retraction and Tfp mediated functions (e.g. adhesion, cell motility) in Gram negative bacteria including P. aeruginosa, P. stutzeri, N. gonorrhoeae, Myxococcus xanthus, Aquifex aeolicus, Synechocystis sp. P C C 6803 and  Ralstonia solanacearum (9, 37, 50, 94, 98, 101). The absence of a PilT homologue in Pasteurellaceae implies the presence of an as-yet-uncharacterized protein to provide the force for Tfp retraction, and a difference between energetics of Tfp in Pasteurellaceae species and other Gram negative bacteria.  On the other hand, however, the Pasteurellaceae species may not need a protein to drive Tfp retraction. In P. stutzeri, an insertion in the pilT gene resulted in a hyperpiliated mutant, which was able to bind D N A but could not take it up (37). This transformation defect was suppressed by replacement of the six C-terminal amino acids in the pilin subunit P i l A I with six histidine residues. It was hypothesized in the study that the pilTindependent transformation can promoted by mutant P i l A protein either as single molecules or as minimal pilin assembly structures in the periplasm which may resemble  86  depolymerized pili, and that these cause the outer membrane pores to open for D N A entry. Both of these two explanations have problems. Studies have shown that D N A binding is a function of intact pili (91). Therefore the D N A uptake machinery has to be a Tfp-like assembly that is at least long enough to penetrate to the outside of the outer membrane to reach D N A , although fimbriae may not be seen on the cell surface. The current model postulates that the assembled Tfp fimbriae are translocated across the outer membrane through the secretin pore of P i l Q (16). Since the predicted structure of Type IV pilus does not have space for a D N A channel in the center of the pilus (47), transforming D N A is thought to be transformed through the P i l Q secretin pore (6, 26). D N A would not be allowed to come in until the secretin pore become unoccupied by pilus disassembly. Therefore disassembly of the Tfp-like machinery into the periplasm would be required for transport of D N A . Consequently, a more plausible explanation of the observation in the above study would be that the mutation in P i l A I altered the association between the pilin subunits such that the pilus assembly was shorter and less stable at the base than a regular pilus and it spontaneously depolymerized when P i l B was not active, thus a PilT protein was not needed.  Similarly, the absence o f a PilT-like ATPase in Pasteurellaceae species, therefore, may also be because a P i l T protein is not needed. Sequence analysis showed some considerable differences between the prepilin protein P i l A o f nontypeable H. influenzae and other Type I V pilins (7). The leader peptides o f nontypeable H. influenzae prepilins are longer than that o f P. aeruginosa and N. gonorrhoeae prepilins, whereas the length o f the mature pilins in nontypeable H. influenzae are shorter than that in P. aeruginosa and  87  N. gonorrhoeae.  The sequence features ofH. influenzae pilins are also found i n the pilins  of A. pleuropneumoniae  and P. multocida (data not shown). The Tfp fibers of the  Pasteurellaceae species may be able to disassemble from the base spontaneously from the base when P i l B is not active, therefore a PilT protein would not be required and the pilT gene might have been lost during evolution.  In summary, the absence of a PilT protein in Pasteurellaceae may be either because anas-yet uncharacterized protein drives Tfp retraction, or a PilT is not required for pilus disassembly in Pasteurellaceae. T o test which of the above hypotheses is right, experiments can first be done in either of the two H. influenzae strains, to look for the putative protein that the first hypothesis predicts to drive the Tfp retraction in Pasteurellaceae species. According to the first hypothesis, this protein should be unique to Pasteurellaceae species. Moreover, i f Pasteurellaceae Tfp are disassembled at the base, the protein that drives retraction may not be in the outer membrane and would be either in the periplasm, or intrinsic or peripheral to the inner membrane. This as-yetuncharacterized protein could be an ATPase that does not belong to the G s p E family, or could use another energy source, such as proton motive force. Bremer, et al. showed that proton motive force is needed for D N A uptake by H. influenzae (11). Comparison analysis of the genomes of H. influenzae and P. aeruginosa can be performed to look for genes present in Pasteurellaceae but not in P. aeruginosa.  Sequence analysis including  B L A S T search and reserved domain analysis could be used to predict location and function of the proteins. Subsequently, High-throughput gene mutagenesis can be used. A gene whose knockout strain can bind D N A but cannot take it up in H. influenzae R d , or  88  a gene whose knockout can express Tfp but is deficient in twitching motility in the other H. influenzae strain that possess visible Tfp, would be the one responsible for Tfp retraction. A s a result, i f such a protein cannot be found, it would suggest that a protein may be unnecessary in Pasteurellaceae. However, more convincing evidence would be required to confirm it.  The lack of another G s p E family protein and of visible Tfp in H. influenzae R d , makes it difficult to test whether H. influenzae P i l B function both i n the assembly and retraction of the Tfp-like machinery or only function in assembly. Due to the failure of complementation by P. aeruginosa homologues, H. influenzae PilB's specific function in D N A uptake remains unidentified. However, sequence analysis indicates that H. influenzae P i l B is more related to P i l B proteins than to PilT proteins (67, 69). These suggest that H. influenzae P i l B only functions in the assembly of the Tfp-like D N A uptake machinery, although more convincing evidence is not available yet. Furthermore, the existence of functional type I V pili in some Pasteurellaceae in the absence of a PilT homologue suggests that a G s p E family ATPase may not be required for retraction of the type I V pili in this family (7, 76, 102).  In future experiments, a H. influenzae strain that can express visible pili would be useful to identify P i l B ' s specific function in Tfp assembly/retraction. It has been shown that the majority o f cells o f H. influenzae 86-028NP express Tfp only when growing under defined nutrient conditions at an alkaline p H (pH8.5 to 9.0) (7). H. influenzae R d might also do so under similar conditions. O n the other hand, the hypertransformable Sxy  89  mutant strain might be able to express visible Tfp as well, because pilin genes are hyperinduced in this strain (73, 74). If either of the above can express visible Type IV pili, electron microscopy and/or motility assays using agar slabs can be used to examine Tfp expression in the above strains and in the pilB knockout RR1150 (7). The presence of visible Tfp in the w i l d type H. influenzae or in the Sxy mutant, and the absence of Tfp in RR1150 would indicate that P i l B is responsible for Tfp assembly.  90  References 1.  Aas, F. E., C. Lovold, and M. Koomey. 2002. A n inhibitor of D N A binding and uptake events dictates the proficiency of genetic transformation in Neisseria gonorrhoeae: mechanism of action and links to Type I V pilus expression. M o l Microbiol 46:1441-50.  2.  3.  4.  5.  Aas, F. E., M. Wolfgang, S. Frye, S. Dunham, C. Lovold, and M; Koomey. 2002. Competence for natural transformation in Neisseria gonorrhoeae: components of D N A binding and uptake linked to type I V pilus expression. M o l Microbiol 46:749-60. Alexander, H. E., and G. Leidy. 1951. Determination of inherited traits of H. influenzae by deoxyribonucleic acid fractions isolated from type-specific cells. J Exp M e d 93:345-59. Aim, R. A., and J. S. Mattick. 1997. Genes involved in the biogenesis and function of type-4 fimbriae in Pseudomonas aeruginosa. Gene 192:89-98.  Ausubel, F. M., R. Brent, R. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl (ed.). 1994. Current protocols in molecular biology. John Wiley & Sons, Inc.  6.  7.  Averhoff, B., and A. Friedrich. 2003. Type I V pili-related natural transformation systems: D N A transport in mesophilic and thermophilic bacteria. A r c h Microbiol 180:385-93. Bakaletz, L. O., B. D. Baker, J. A. Jurcisek, A. Harrison, L. A. Novotny, J. E. Bookwalter, R. Mungur, and R. S. Munson, Jr. 2005. Demonstration of Type  8.  9.  IV pilus expression and a twitching phenotype by Haemophilus influenzae. Infect Immun 73:1635-43. Barcak, G. J., M. S. Chandler, R. J. Redfield, and J. F. Tomb. 1991. Genetic systems in Haemophilus influenzae. Methods Enzymol 204:321-42.  Bhaya, D., N. R. Bianco, D. Bryant, and A. Grossman. 2000. Type IV pilus biogenesis and motility in the cyanobacterium Synechocystis sp. PCC6803.  10.  Bitter, W., M. Koster, M. Latijnhouwers, H. de Cock, and J. Tommassen. 1998. Formation of oligomeric rings by X c p Q and P i l Q , which are involved in protein transport across the outer membrane of Pseudomonas aeruginosa.  11.  12. 13. 14.  15.  Bremer, W., J. Kooistra, K. J. Hellingwerf, and W. N. Konings. 1984. Role of the electrochemical proton gradient in genetic transformation of Haemophilus influenzae. J Bacteriol 157:868-73. Chen, I., and D. Dubnau. 2004. D N A uptake during bacterial transformation. Nat Rev Microbiol 2:241-9. Chen, I., and E. C. Gotschlich. 2001. C o m E , a competence protein from Neisseria gonorrhoeae with D N A - b i n d i n g activity. J Bacteriol 183:3160-8. Chen, W. J., L. Gross, K. E. Joho, and W. T. McAllister. 1992. A modified kanamycin-resistance cassette to facilitate two-codon insertion mutagenesis. Gene 111:143-4. Chiang, P., M. Habash, and L. L. Burrows. 2005. Disparate subcellular localization patterns of Pseudomonas aeruginosa Type I V pilus ATPases involved in twitching motility. J Bacteriol 187:829-39.  91  16.  Collins, R. F., L. Davidsen, J. P. Derrick, R. C. Ford, and T. Tonjum. 2001. Analysis of the P i l Q secretin from Neisseria meningitidis by transmission electron microscopy reveals a dodecameric quaternary structure.  17.  18.  19.  Comolli, J. C , A. R. Hauser, L. Waite, C. B. Whitchurch, J. S. Mattick, and J. N. Engel. 1999. Pseudomonas aeruginosa gene products P i l T and P i l U are required for cytotoxicity in vitro and virulence in a mouse model of acute pneumonia. Infect Immun 67:3625-30. Craig, L., M. E. Pique, and J. A. Tainer. 2004. Type I V pilus structure and bacterial pathogenicity. Nat Rev Microbiol 2:363-78.  Craig, L., R. K. Taylor, M. E. Pique, B. D. Adair, A. S. Arvai, M. Singh, S. J. Lloyd, D. S. Shin, E. D. Getzoff, M. Yeager, K. T. Forest, and J. A. Tainer. 2003. Type I V pilin structure and assembly: X-ray and E M analyses of Vibrio cholerae toxin-coregulated pilus and Pseudomonas aeruginosa P A K pilin. M o l Cell 11:1139-50.  20.  Danner, D. B., R. A. Deich, K. L. Sisco, and H. O. Smith. 1980. A n elevenbase-pair sequence determines the specificity of D N A uptake in Haemophilus transformation. Gene 11:311-8.  21. 22. 23.  24. 25.  de Groot, A., I. Heijnen, H. de Cock, A. Filloux, and J. Tommassen. 1994. Characterization of type I V pilus genes in plant growth-promoting Pseudomonas putida W C S 3 5 8 . J Bacteriol 176:642-50. De Reuse, H., and M. K. Taha. 1997. RegF, an SspA homologue, regulates the expression of the Neisseria gonorrhoeae pilE gene. Res Microbiol 148:289-303. Dougherty, B. A., and H. O. Smith. 1999. Identification of Haemophilus influenzae R d transformation genes using cassette mutagenesis. Microbiology 145 (Pt 2):401-9. Douglas, S. E. 1995. D N A Strider. A n inexpensive sequence analysis package for the Macintosh. M o l Biotechnol 3:37-45. Drake, S. L., and M. Koomey. 1995. The product of the pilQ gene is essential  for the biogenesis of type IV pili in Neisseria gonorrhoeae. 26. 27. 28. 29. 30.  31. 32.  33.  Dubnau, D. 1999. D N A uptake in bacteria. A n n u Rev Microbiol 53:217-44. Elkins, C , C. E. Thomas, H. S. Seifert, and P. F. Sparling. 1991. Speciesspecific uptake of D N A by gonococci is mediated by a 10-base-pair sequence. Engels, W. R. 1993. Contributing software to the internet: the Amplify program. Trends Biochem Sci 18:448-50. Fernandez, L. A., and J. Berenguer. 2000. Secretion and assembly of regular surface structures in Gram-negative bacteria. F E M S Microbiol Rev 24:21-44. Filloux, A. 2004. The underlying mechanisms of type II protein secretion. Biochim Biophys A c t a 1694:163-79.  Finkel, S. E., and R. Kolter. 2001. D N A as a nutrient: novel role for bacterial competence gene homologs. J Bacteriol 183:6288-93. Freitag, N. E., H. S. Seifert, and M. Koomey. 1995. Characterization of the pilF-pilD pilus-assembly locus of Neisseria gonorrhoeae. M o l Microbiol 16:57586.  Fussenegger, M., T. Rudel, R. Barten, R. Ryll, and T. F. Meyer. 1997. Transformation competence and type-4 pilus biogenesis in Neisseria gonorrhoeae—a. review. Gene 192:125-34.  92  34.  Goldberg, G. B. 1995, posting date. Goldberg lab standard protocols. Cayne, M . [Online.]  35.  Goodgal, S. H. 1982. D N A uptake in Haemophilus transformation. A n n u Rev Genet 16:169-92.  36.  37.  38. 39.  40.  41.  42.  Graupner, S., and W. Wackernagel. 2001. Pseudomonas  stutzeri has two  closely related pilA genes (Type I V pilus structural protein) with opposite influences on natural genetic transformation. J Bacteriol 183:2359-66. Graupner, S., N. Weger, M. Sohni, and W. Wackernagel. 2001. Requirement of novel competence genes pilT and pilU of Pseudomonas stutzeri for natural transformation and suppression of pilT deficiency by a hexahistidine tag on the type I V pilus protein PilAI. J Bacteriol 183:4694-701. Hanahan, D. 1983. Studies on transformation of Escherichia coli with plasmids. J M o l B i o l 166:557-80. Herdendorf, T. J., D. R. McCaslin, and K. T. Forest. 2002. Aquifex aeolicus PilT, homologue of a surface motility protein, is a thermostable oligomeric NTPase. J Bacteriol 184:6465-71. Herriott, R. M., E. M. Meyer, and M. Vogt. 1970. Defined nongrowth media for stage II development of competence in Haemophilus influenzae. J Bacteriol 101:517-24. Hobbs, M., and J. S. Mattick. 1993. Common components in the assembly of type 4 fimbriae, D N A transfer systems, filamentous phage and protein-secretion apparatus: a general system for the formation of surface-associated protein complexes. M o l Microbiol 10:233-43.  Hoyne, P. A., R. Haas, T. F. Meyer, J. K. Davies, and T. C. Elleman. 1992. Production of Neisseria gonorrhoeae pili (fimbriae) in Pseudomonas  aeruginosa.  J Bacteriol 174:7321-7. 43.  44. 45. 46.  47.  48.  49. 50.  Inamine, G. S., and D. Dubnau. 1995. C o m E A , a Bacillus  subtilis integral membrane protein required for genetic transformation, is needed for both D N A binding and transport. J Bacteriol 177:3045-51. International, P. B. 1994, posting date. NetPrimer. [Online.], Kahn, M. E., and H. O. Smith. 1984. Transformation in Haemophilus: a problem in membrane biology. J Membr B i o l 81:89-103. Karudapuram, S., and G. J. Barcak. 1997. The Haemophilus influenzae dprABC genes constitute a competence-inducible operon that requires the product of the tfoX (sxy) gene for transcriptional activation. J Bacteriol 179:4815-20.  Keizer, D. W., C. M. Slupsky, M. Kalisiak, A. P. Campbell, M. P. Crump, P. A. Sastry, B. Hazes, R. T. Irvin, and B. D. Sykes. 2001. Structure of a pilin monomer from Pseudomonas aeruginosa: implications for the assembly of pili. J B i o l Chem 276:24186-93. Koga, T., K. Ishimoto, and S. Lory. 1993. Genetic and functional characterization of the gene cluster specifying expression of Pseudomonas aeruginosa pili. Lessl, M., and E. Lanka. 1994. Common mechanisms in bacterial conjugation and Ti-mediated T - D N A transfer to plant cells. Cell 77:321-4. Liu, H., Y. Kang, S. Genin, M. A. Schell, and T. P. Denny. 2001. Twitching motility of Ralstonia solanacearum requires a type I V pilus system.  93  51. 52. 53. 54.  55.  Long, C. D., D. M. Tobiason, M. P. Lazio, K. A. Kline, and H. S. Seifert. 2003. Low-level pilin expression allows for substantial D N A transformation competence in Neisseria gonorrhoeae. Infect Immun 71:6279-91. Lory, S., and M. S. Strom. 1997. Structure-function relationship of type-IV prepilin peptidase of Pseudomonas aeruginosa—& review. Gene 192:117-21. Macfadyen, L. P. 2000. Regulation of competence development in Haemophilus influenzae. J Theor B i o l 207:349-59. Maier, B., L. Potter, M. So, C. D. Long, H. S. Seifert, and M. P. Sheetz. 2002. Single pilus motor forces exceed 100 pN. Proc Natl A c a d Sci U S A 99:16012-7.  Martin, P. R., M. Hobbs, P. D. Free, Y. Jeske, and J. S. Mattick. 1993. Characterization of pilQ, a new gene required for the biogenesis of type 4 fimbriae in Pseudomonas aeruginosa.  56.  Martin, P. R., A. A. Watson, T. F. McCaul, and J. S. Mattick. 1995. Characterization of a five-gene cluster required for the biogenesis of type 4 fimbriae in Pseudomonas aeruginosa.  57. 58.  59.  Martinez, E., B. Bartolome, and F. de la Cruz. 1988. pACYC184-derived cloning vectors containing the multiple cloning site and lacZ alpha reporter gene of pUC.8/9 and pUC18/19 plasmids. Gene 68:159-62. Mattick, J. S. 2002. Type IV pili and twitching motility. A n n u Rev Microbiol 56:289-314.  Mattick, J. S., M. M. Bills, B. J. Anderson, B. Dalrymple, M. R. Mott, and J. R. Egerton. 1987. Morphogenetic expression of Bacteroides nodosus fimbriae in Pseudomonas  60. 61.  62. 63. 64.  aeruginosa.  Merz, A. J., M. So, and M. P. Sheetz. 2000. Pilus retraction powers bacterial twitching motility. Nature 407:98-102. Molnar, S. 2004. Identification and characterization of D N A receptor candidates in Haemophilus influenzae. Thesis for Master of Science. The University of British Columbia, Vancouver. Nudleman, E., and D. Kaiser. 2004. Pulling together with type I V pili. J M o l Microbiol Biotechnol 7:52-62. Nunn, D. 1999. Bacterial type II protein export and pilus biogenesis: more than just homologies? Trends Cell B i o l 9:402-8. Nunn, D., S. Bergman, and S. Lory. 1990. Products of three accessory genes, pilB, pilC, and pilD, are required for biogenesis of Pseudomonas aeruginosa pili.  65.  Parge, H. E., K. T. Forest, M. J. Hickey, D. A. Christensen, E. D. Getzoff,  66.  and J. A. Tainer. 1995. Structure ofthe fibre-forming protein pilin at 2.6 A resolution. Nature 378:32-8. Park, H. S., M. Wolfgang, and M. Koomey. 2002. Modification of type I V pilus-associated epithelial cell adherence and multicellular behavior by the P i l U protein of Neisseria gonorrhoeae. Infect Immun 70:3891-903.  67.  Peabody, C. R., Y. J. Chung, M. R. Yen, D. Vidal-Ingigliardi, A. P. Pugsley, and M. H. Saier, Jr. 2003. Type II protein secretion and its relationship to  68.  Pepe, C. M., M. W. Eklund, and M. S. Strom. 1996. Cloning of an  bacterial type I V pili and archaeal flagella. Microbiology 149:3051-72. Aeromonas  hydrophila type I V pilus biogenesis gene cluster: complementation of pilus assembly functions and characterization of a type I V leader peptidase/N-  94  69.  70.  71.  72. 73. 74.  75.  76.  77.  78. 79.  80.  81.  82.  83. 84.  methyltransferase required for extracellular protein secretion. M o l Microbiol 19:857-69. Planet, P. J., S. C. Kachlany, R. DeSalle, and D. H. Figurski. 2001. Phylogeny of genes for secretion NTPases: identification of the widespread tadA subfamily and development of a diagnostic key for gene classification. Poje, G., and R. J. Redfield. 2003. General methods for culturing Haemophilus influenzae. In M . A . Herbert, D . W . Hood, and E . R. M o x o n (ed.), Haemophilus influenzae protocols. Human Press, Totowa, New Jersey. Poje, G., and R. J. Redfield. 2003. Transformation of Haemophilus influenzae. In M . A . Herbert, D . W . Hood, and E . R. M o x o n (ed.), Haemophilus influenzae protocols. Human Press, Totowa, New Jersey. Pugsley, A. P. 1993. The complete general secretory pathway in gram-negative bacteria. Microbiol Rev 57:50-108. Redfield, R. J. 1991. sxy-l, a Haemophilus influenzae mutation causing greatly enhanced spontaneous competence. Redfield, R. J., A. D. Cameron, Q. Qian, J. Hinds, T. R. Ali, J. S. Kroll, and P. R. Langford. 2005. A novel CRP-dependent regulon controls expression of competence genes in Haemophilus influenzae. J M o l B i o l 347:735-47. Roine, E . , D. M. Raineri, M. Romantschuk, M. Wilson, and D. N. Nunn. 1998. Characterization of type I V pilus genes in Pseudomonas syringae pv. tomato DC3000. M o l Plant Microbe Interact 11:1048-56. Ruffolo, C. G., J. M. Tennent, W. P. Michalski, and B. Adler. 1997. Identification, purification, and characterization of the type 4 fimbriae of Pasteurella muliocida. Infect Immun 65:339-43. Sambrook, J., E . F. Fritsch, and T. Maniatis. 1989. Molecular Cloning - A Laboratory Manual, Second Edition ed. C o l d Spring Harbour Laboratory Press, New York. Sandkvist, M. 2001. Biology of type II secretion. M o l Microbiol 40:271-83. Sandkvist, M., M. Bagdasarian, S. P. Howard, and V. J. DiRita. 1995. Interaction between the autokinase EpsE and E p s L in the cytoplasmic membrane is required for extracellular secretion in Vibrio cholerae. Sauvonnet, N., P. Gounon, and A. P. Pugsley. 2000. PpdD type I V pilin of Escherichia coli K - 1 2 can be assembled into pili in Pseudomonas aeruginosa. J Bacteriol 182:848-54. Sauvonnet, N., G. Vignon, A. P. Pugsley, and P. Gounon. 2000. Pilus formation and protein secretion by the same machinery i n Escherichia coli. Embo J 19:2221-8. Smeets, L. C , and J. G. Kusters. 2002. Natural transformation in Helicobacter pylori: D N A transport in an unexpected way. Trends Microbiol 10:159-62; discussion 162. Strom, M. S., and S. Lory. 1993. Structure-function and biogenesis of the type IV pili. Strom, M. S., D. Nunn, and S. Lory. 1991. Multiple roles of the pilus biogenesis protein P i l D : involvement of P i l D in excretion of enzymes from Pseudomonas aeruginosa. J Bacteriol 173:1175-80.  95  85.  86. 87. 88. 89.  90.  91. 92.  93.  Strom, M. S., D. N. Nunn, and S. Lory. 1993. A single bifunctional enzyme, P i l D , catalyzes cleavage and N-methylation of proteins belonging to the type I V pilin family. Proc Natl A c a d Sci U S A 90:2404-8. Stuy, J. H. 1979. Plasmid transfer in Haemophilus influenzae. J Bacteriol 139:520-9. TIGR 1999-2004, posting date, http://www.tigr.org/. [Online.] Tomb, J. F., H. el-Hajj, and H. O. Smith. 1991. Nucleotide sequence of a cluster of genes involved in the transformation of Haemophilus influenzae R d . Tonjum, T., N. E. Freitag, E. Namork, and M. Koomey. 1995. Identification and characterization of pilG, a highly conserved pilus-assembly gene in pathogenic Neisseria. M o l Microbiol 16:451-64. Turner, L. R., J. C. Lara, D. N. Nunn, and S. Lory. 1993. Mutations in the consensus A T P - b i n d i n g sites of X c p R and P i l B eliminate extracellular protein secretion and pilus biogenesis in Pseudomonas aeruginosa. J Bacteriol 175:49629.  van Schaik, E. J., C. L. Giltner, G. F. Audette, D. W. Keizer, D. L. Bautista, C. M. Slupsky, B. D. Sykes, and R. T. Irvin. 2005. D N A binding: a novel function of Pseudomonas aeruginosa type IV pili. J Bacteriol 187:1455-64. Wang, Y., S. D. Goodman, R. J. Redfield, and C. Chen. 2002. Natural transformation and D N A uptake signal sequences in Actinobacillus actinomycetemcomitans. J Bacteriol 184:3442-9. Watson, A. A., R. A. Aim, and J. S. Mattick. 1996. Identification of a gene, pilF, required for type 4 fimbrial biogenesis and twitching motility in Pseudomonas  94.  95.  96.  97.  aeruginosa.  Whitchurch, C. B., M. Hobbs, S. P. Livingston, V. Krishnapillai, and J. S. Mattick. 1991. Characterisation of a Pseudomonas aeruginosa twitching motility gene and evidence for a specialised protein export system widespread in eubacteria. Wiesner, R. S., D. R. Hendrixson, and V. J. DiRita. 2003. Natural transformation of Campylobacter jejuni requires components of a type II secretion system. J Bacteriol 185:5408-18. Wilcox, K. W., and H. O. Smith. 1975. Isolation and characterization of mutants of H. influenzae deficient in an adenosine 5'-triphosphate-dependent deoxyribonuclease activity. J Bacteriol 122:443-53. Williams, P., W. L. Hung, and R. J. Redfield. 1996. Cell transfer by filtration: evaluation of protocols for transformational competence. F E M S Microbiol Lett 137:183-7.  98.  Wolfgang, M., P. Lauer, H. S. Park, L. Brossay, J. Hebert, and M. Koomey.  99.  1998. PilT mutations lead to simultaneous defects in competence for natural transformation and twitching motility in piliated Neisseria gonorrhoeae. M o l Microbiol 29:321-30. Wolfgang, M., J. P. van Putten, S. F. Hayes, and M. Koomey. 1999. The comP locus of Neisseria gonorrhoeae encodes a type I V prepilin that is dispensable for pilus biogenesis but essential for natural transformation. M o l Microbiol 31:134557.  96  100. 101.  102.  Wu, S. S., and D. Kaiser. 1995. Genetic and functional evidence that Type I V pili are required for social gliding motility in Myxococcus xanthus. Wu, S. S., J. Wu, and D. Kaiser. 1997. The Myxococcus xanthus pilT locus is required for social gliding motility although pili are still produced.  Zhang, Y., J. M. Tennent, A. Ingham, G. Beddome, C. Prideaux, and W. P. Michalski. 2000. Identification of type 4 fimbriae in Actinobacillus pleuropneumoniae.  F E M S Microbiol Lett 189:15-8.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0092266/manifest

Comment

Related Items