INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & Howell Information Company 300 North Zeeb Road, Ann Arbor MI 48106-1346 USA 313/761-4700 800/521-0600 CHARACTERIZATION OF THE PSEUDOMONAS AERUGINOSA PILT GENE WHICH IS REQUIRED FOR TWITCHING MOTILITY AND PILUS FUNCTION
DISSERTATION
Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate
School of The Ohio State University
By Patricia Lynn Antalis, B. S.
* * * * *
The Ohio State University
1996
Dissertation Committee: Approved by Kathleen E. Kendrick, Adviser Aldis Darzins Neil R. Baker Adviser Charles J. Daniels Department of Microbiology UMI Number: 9630843
UMI Microform 9630843 Copyright 1996, by UMI Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.
UMI 300 North Zeeb Road Ann Arbor, MI 48103 DEDICATION
To my husband, and best friend, George ACKNOWLEDGMENTS
I would like to express my appreciation to each of my committee members for their guidance, especially my advisers Dr. Aldis Darzins and Dr.
Kathleen Kendrick. Thanks also go to Lynn O'Donnell and Mary Russell for their camaraderie, advice and wit. I am indebted to family for their continual support, especially my sister and brother-in-law for their generosity and humor during this last year. I am most grateful to my husband for his patience and for his reminding me of what is truly important. VITA
November 12,1966 ...... Bom - Wheeling, West Virginia
1989...... B. S. in Biology, The Ohio State University, Columbus, Ohio
1987 ...... Research Assistant, Department of Zoology, The Ohio State University
1987 -1989...... Research Assistant, Department of
Entomology, The Ohio State University
1989 - present...... Graduate Teaching Associate, Department
of Microbiology, The Ohio State University
FIELDS OF STUDY
Major Field: Microbiology
Studies in Bacterial Gene Regulation Dr. K. E. Kendrick and Dr. A. Darzins, Advisers TABLE OF CONTENTS
DEDICATION......
ACKNOWLEDGMENTS......
VITA......
TABLE OF CONTENTS......
LIST OF FIGURES......
LIST OF TABLES......
LIST OF ABBREVIATIONS......
CHAPTER I: Introduction ......
Role of Pseudomonas aeruginosa pili in adherence
Structure of P. aeruginosa pilin......
P. aeruginosa pilus biogenesis ......
Occurrence of type IV pili in other bacteria ......
Pilus retraction ......
Twitching motility Goals of this project...... 39
CHAPTER II: Materials and Methods...... 40 Media and culture conditions ...... 40 Bacterial strains, plasmids, and bacteriophages...... 41 Phage sensitivity and twitching motility assays...... 46 Electron microscopy techniques ...... 47
Mapping of the HOD1 phage-resistance m utation...... 48 LDgo studies ...... 48
Standard DNA isolation and manipulation techniques ...... 49 Transfer of plasmid DNA intoPseudomonas ...... 52
Generation of nested deletions for DNA sequence analysis ...... 54 DNA hybridization methods ...... 55
Standard methods for protein analysis...... 58 Overexpression ofpilT in E.coli and Pseudomonas using T7
promoter-based systems...... 60 RNA isolation and manipulation ...... 61 PCR conditions for generation of SI nuclease probes ...... 66
Enzyme assays ...... 67 Generation of random mutations in p ilT...... 69
CHAPTER III: pilT is Required for Phage Sensitivity and Twitching Motility in P. aeruginosa P A O l...... 71
Introduction ...... 71
vi R esults ...... 73
Isolation of phage-resistant mutants of PAOl defective in pilus biogenesis ...... 73
Complementation of the hyperpiliated m utants...... 76
Conjugations! mapping of the HOD1 phage-resistance
locus ...... 80
Overexpression ofpilT in E.coli and Pseudomonas using T7 promoter-based system s ...... 82 Nucleotide sequence of the pilT open reading frame...... 90 Disruption of pilT in PA O l...... 90
Characterization of the mutated pilT allele from HOD1 95
Generation of pilT mutants by using random mutagenesis . 99
Virulence of HOD1 in a mouse infection model ...... 104 Discussion ...... 107
CHAPTER IV: Construction and Analysis of pilU and pilT pilU Mutants
in P. aeruginosa P A O l ...... 130
Introduction ...... 130
R esults ...... 132
Introduction of pilU into the pilT mutant HOD1 ...... 132
Construction and complementation analysis of a
pilU mutant ...... 133 Construction and complementation analysis of a
pUTpilU double m u ta n t ...... 140
Discussion...... 145
CHAPTER V: Expression ofpilT is Differentially Regulated at the Transcriptional Level...... 147
Introduction ...... 147
Results ...... 149
Northern analysis of the pilT transcript...... 149
pilT expression does not require the alternative sigma factor RpoN...... 152
Identification of promoter activity in thepilT pilU intergenic region...... 152
Identification of the region required for pilT promoter
activity...... 155
Quantitative analysis of pilT expression on solid and liquid m edia ...... 162
Mapping of the transcription start sites ofpilT and
analysis of transcript populations...... 168
Identification of potential regulatoiy regions upstream ofp ilT...... 182
Discussion ...... 183
viii LITERATURE CITED...... 192
APPENDICES...... 212
A. Maps of cloning vectors used in this study ...... 212
B. List of plasmids constructed for this study ...... 231
C. List of oligonucleotides used in this stu d y ...... 237
D. Restriction maps of the PAOlpilT and pilU loci...... 239
E. Dendrogram of PilT homologs...... 242
ix LIST OF FIGURES
Figure Title Page 1. Structure of P. aeruginosa PAOl prepilin protein...... 10 2. Pseudomonas aeruginosa PAOl genetic map and location ofpH lo ci...... 17
3. Schematic representation of pilus assembly in P. aeruginosa...... 24 4. Twitching motility by Pseudomonas aeruginosa PAO1 ...... 33 5. Colony morphology ofP. aeruginosa PAOl and the pilT mutant HOD1 on LB agar ...... 74
6 Complementation of the hyperpiliated mutant HOD1...... 78 7. Plasmids used for the expression studies of the HOD1- complementing locus in E. coli BL21...... 83 8. Expression of the HOD 1-complementing locus ipilT) in E.coli BL21...... 85 9. Expression of the HOD 1-complementing locus ipilT) in the P. aeruginosa T7-expression strain ADD1976 ...... 88
10. The nucleotide sequence of the 1.5 kb Pstl-Kpnl fragment containingpi/71...... 91 11. Construction of the pilT: :pBR322 disruption mutant PAO-PT...... 93 12. Identification of the HOD1 pilT allele...... 96
13. Western analysis of PilT in cell lysates from P. aeruginosa . . . 100
x 14. Virulence of PAOl and HOD1 in A.BY/SnJ mice ...... 105 15. Partial alignment of the highly conserved regions of the putative nucleotide-binding site in PilT and its related proteins..-...... I l l
16. Schematic alignment of the PilT homologs ...... 113 17. Cloning scheme followed in the construction of the pilU gene-replacement plasmid pPT3547...... 134
18. Southern analysis of PAO-PU ...... 138 19. Southern analysis of PAO-TU ...... 142 20. Northern hybridization analysis of pilT expression in PA O l...... 150 21. Northern hybridization analysis of the pilT transcript in an RpoN m utant ...... 153 22. 13-galactosidase activity of a pilUwlacZ transcriptional plasmid fusion in PA O l...... 156 23. Mapping of the P.aeruginosa pilT promoter region using the luxAB reporter genes...... 159 24. 6-galactosidase activity in a pilT::lacZ transcriptional fusion in P. aeruginosa PA O l...... 163
25. Comparison of thepilT promoter activity in P. aeruginosa PAOl grown on solid and liquid media by using a pilT::lacZ transcriptional fusion ...... 166 26. The effect of growth medium on the level of the pilT transcript in P. aeruginosa PA O l...... 169
27. Low resolution SI nuclease mapping of the approximate location of the pilT transcription initiation site 172 28. High resolution SI nuclease mapping of the transcription initiation sites of P. aeruginosa pUT...... 174 29. Nucleotide sequence of the region upstream of p ilT ...... 176 30. High resolution SI nuclease analysis of the two pilT transcripts during and after growth on LB agar...... 180
xi 31. Circular map of pBR322 ...... 213
32. Circular map of pR01614 ...... 215 33. Circular map of pUCP18...... 217 34. Circular map of pNotl9...... 219 35. Circular map of pMob3...... 221 36. Circular map of pQF70 ...... 223 37. Circular map of pQF50 ...... 225 38. Circular map of pADD621...... 227 39. Circular map of pEB12...... 229
40. Restriction map of thepilT and pilU loci...... 240 41. Dendrogram of PilT homologs...... 243
xii LIST OF TABLES
Table Title Page
1. Local and disseminated infections caused by Pseudomonas aeruginosa ...... 3 2. Pseudomonas aeruginosa pillo d ...... 15 3. Bacteria displaying type IV pili...... 26 4. Bacterial strains and bacteriophages used in this study ...... 42
5. Plasmid vectors used in this study ...... 44 6. Mapping of the HOD 1 phage-resistance mutation...... 81
7. pilT mutations generated using an E. coli mutator strain . . . . 103 8. Homologs of P. aeruginosa type IV pilus biogenesis proteins .. 118
9. Phenotypic analysis of P. aeruginosa pilT and pilU mutants . 144
10. Plasmids constructed for this study ...... 232
11. Oligonucleotides constructed for this study ...... 238
xiii LIST OF ABBREVIATIONS
Ap Ampicillin BEC buccal epithelial cells bp base pairs cAMP cyclic adenosine monophosphate Cb carbenidllin cfu colony forming units cm centimeters Cm chloramphenicol cpm counts per minute dATP deoxyadenosine triphosphate dCTP deoxycytidine triphosphate dGTP deoxyguanosine triphosphate DNA deoxyribonucleic add dNTP deoxynucleoside triphosphate dTTP deoxythymidine triphosphate ECR elastin-congo red EM electron microscopy EPEC enteropathogenicE. coli GCG Genetics Computer Group hr hours IPTG isopropyl fi-D-thiogalactopyranoside kb kilobases kDa kdlodaltons Km kanamycin L liters LB Luria-Bertani LD lethal dose mA milliamperes mAb monoclonal antibodies mCi millicuries mg milligrams min minutes ml milliliters mm millimeters mM millimolar MW molecular weight nm nanometers nt nucleotides PAGE polyacrylamide gel electrophoresis PBS phosphate-buffered saline PCR polymerase chain reaction pfu plaque forming units PIA Pseudomonas isolation agar RNA ribonucleic add rpm revolutions per minute SDS sodium dodecyl sulfate sec seconds Tc tetracycline TEC tracheal epithelial cells Tw twitching UTL untranslated leader sequence V volts vol volume wt weight
XV CHAPTER I
INTRODUCTION
Pseudomonas aeruginosa is a Gram-negative, rod-shaped bacterium that is widely distributed in soil and aquatic ecosystems. The most frequent sources of this bacterium are water reservoirs that have been polluted with animal or human sewage, but P. aeruginosa can also be isolated from hospital bathroom fixtures, humidifiers, swimming pools or hot tubs, and medical or dental equipment that utilizes water (Botzenhart and Riiden, 1987; Rhame, 1980; Van Saene et al., 1989). P. aeruginosa colonizes a number of plants and vegetables, an association that usually results in pernicious effects on the plant (Lebeda et al., 1984). This organism is also capable of degrading a vast variety of natural and synthetic organic compounds and is inherently resistant to many conventional antimicrobial agents.
The ability of P. aeruginosa to colonize diverse or unfavorable environments has contributed to its efficacy as a pathogen. However, healthy individuals are rarely infected by P. aeruginosa, which does not display a specificity for particular tissues. Instead, this bacterium colonizes tissues that have been previously damaged by disease or injury, making P. aeruginosa
1 a significant opportunistic pathogen of immunocompromised individuals (Table 1). In fact, P. aeruginosa is the third most common cause of nosocomial infections, following Escherichia coli and Staphylococcus aureus, and results in approximately 10% of all hospital-acquired infections (Bellido and Hancock, 1993).
P. aeruginosa infections are exacerbated by the high level of resistance of this bacterium to many common antibiotic therapies and by the multitude of virulence factors produced by this bacterium (Bellido and Hancock, 1993). P. aeruginosa produces both cellular and extracellular virulence factors that contribute to the ability of this bacterium to cause sepsis, extensive tissue necrosis, microabscess formation, and destructive vascular lesions (Plotkowski et al., 1996). The extracellular virulence factors produced by P. aeruginosa include neutral and alkaline proteases, elastase, phospholipase C and rhamnolipid hemolysin (Nicas and Iglewski, 1986; Steadman et al., 1993). P. aeruginosa also secretes two adenosine diphosphate ribosyl transferase proteins, toxin A and exoenzyme S (Iglewski et al., 1978; Liu et al., 1961).
Although there have been conflicting reports concerning the function of exoenzyme S as an adhesin, it is now accepted that this protein is not an adhesin (N. R. Baker, personal communication; Baker et al., 1991; Kudoh et al., 1994; Kulich et al., 1993).
P. aeruginosa also produces several cell-associated virulence factors which are involved in adherence to mammalian cells: flagella, alginate, and pili Table 1. Local and disseminated infections caused by Pseudomonas aeruginosa.
Infection Predisposition respiratory infections cystic fibrosis; tracheostomy; tracheal intubation organ transplant recipients; corticosteroid treatment; bacteremia chemotherapy-induced neutropenia (hematologic malignancies) wound infections bum wounds; war wounds ocular infections trauma to cornea; extended wear contact lenses swimmer's ear conditions of injury or moistness (e.g. swimming) malignant external otitis advanced diabetes peritonitis continuous ambulatory peritoneal dialysis patients urinary tract infections urinary catheter in spinal cord injury patients
skin infections acquired from contaminated hot tubs, pools; damaged skin due to eczema, lesions meningitis shunts for head and neck malignancies osteomyelitis diabetes osteochondritis puncture wound of the foot endocarditis intravenous drug use; prosthetic heart valve (Ramphal and Pier, 1985; Woods et al., 1980, 1983). The P. aeruginosa flagellum was originally believed to contribute to virulence primarily by
mediating a directed motility (swimming). However, there is evidence that the flagellum demonstrates adherence to glycolipids (S. Wahl, personal communication), and that a recently identified protein, FliO, is required for both flagellar-mediated motility and adherence (Simpsonet al., 1995).
Alginate is an exopolysaccharide that consists of alternating residues of guluronic acid and mannuronic acid and which contributes to respiratory duct obstruction in patients with cystic fibrosis (Hoiby et al., 1987). Mucoid strains of P. aeruginosa displayed greater levels of adherence to acid-treated mouse
tracheal cells than that observed with a nonmucoid strain (Ramphal and Pier, 1985). In addition, Baker and Svanborg-Ed6n (1989) reported that anti-
. alginate antibodies inhibited adherence of mucoid strains of P. aeruginosa to tracheal epithelium.
In addition to flagella, alginate, and pili (discussed below), another type of
P. aeruginosa adhesin was recently identified by Plotkowski et al. (1996), who detected two laminin-binding proteins in the outer membrane of P. aeruginosa PAK. Many bacterial and fungal pathogens have surface-exposed adhesins
that bind laminin, a glycoprotein found in the basement membranes of most epithelia. These investigators propose that binding to laminin by P. aeruginosa may be of significance in damaged epithelium with exposed basement membranes (Plotkowski et al., 1996). B ole of Pseudomonas aeruginosa pili in adherence Pili have been demonstrated to be one of the most important adhesins of P. aeruginosa that mediate adherence to mammalian epithelial cells (Irvin et al., 1989; Paranchych et al., 1986; Ramphal and Pyle, 1983; Ramphalet al., 1984; Woods et al., 1981). The biosynthesis of these cell-surface appendages is the focus of this study; therefore, the significance of pili to the pathogenesis of P. aerugnosa is discussed in detail here.
P. aeruginosa pili (also referred to as fimbriae) belong to the class type IV pili, which includes pili that are assembled by a variety of gram-negative pathogens. In addition to contributing to adherence, the pili of P. aeruginosa
act as receptors for pilus-specific bacteriophages (Bradley and Pitt, 1974) and mediate a surface-dependent motility known as twitching (Bradley, 1980a).
The pili of P. aeruginosa are chromosomally encoded, flexible, polarly located filaments with diameters of 5.4 nm and average lengths of 2.5 pm
(Bradley, 1972a; 1980; Folkhard et al., 1981; Weiss, 1971). These pili are retractile, hollow cylinders composed of a single structural subunit protein, pilin, that has a molecular weight of 15 kDa (Bradley, 1972b, Paranchych et al., 1986; Watts et al., 1983). The pilin subunits are assembled in a helix, five subunits per turn, with each turn having a pitch of 4.1 nm (Watts et al., 1983).
The number of pili assembled per pole, usually 1-20, varies in each strain and throughout growth (Bradley, 1972a; Paranchych et al., 1986). Maximal piliation is seen during exponential growth and several investigators have reported that cultures must be grown on solid medium or in static broth to prevent the pili from being sheared from the cells (Bradley, 1972a; P. Castric, personal communication; Speert et al., 1986; Strom and Lory, 1993). However, Bradley (1972a) performed his EM analysis of pili using P.
aeruginosa cultures that were grown, with shaking, in liquid medium. Thus, the requirement of static cultures for piliation is not clear and is seldom mentioned in published studies. Most investigators, though, purify pili from cells that are cultivated on solid medium.
Woods et al. (1980) were the first to demonstrate that P. aeruginosa pili contribute to the adherence of this bacterium to mammalian cells. These investigators used human buccal epithelial cells (BEC) that had their surface fibronectin removed in an in vitro adherence assay and found that
preincubation of the buccal cells with purified pili resulted in a dose-dependent
decrease in the adherence of P. aeruginosa cells (Woods et al., 1980). The ability of the free pili to compete with intact bacterial cells for adherence to
BEC suggested that pili are involved in P. aeruginosa adhesion (Woods et al., 1980). P. aeruginosa pili have also been shown to bind to hamster tracheal epithelial cells (Baker and Marcus, 1982; Baker and Tao, 1980; Marcus and
Baker, 1985) and mouse murine epidermal cells in healthy and bumed-mouse models (Sato et al., 1988). Doig et al. (1988) demonstrated that pili are responsible for attachment to both BEC and human tracheal epithelial cells (TEC), and that the addition of purified pili to the binding assays competetively inhibited the adherence of the bacterial cells.
Once it became clear that the P. aeruginosa pilus was an adhesin,
investigations began to focus on the identification of the nature of the pilus adhesin and of the eukaryotic cell receptor(s) that it recognized. Irvin et al. (1989) established that the carboxyl terminus of P. aeruginosa PAK pilin binds
specifically to both BEC and TEC and that synthetic peptides with sequences identical to that of the pilin C-terminus can compete with purified pilin in binding assays. Thus, P. aeruginosa pilin can function by itself as an adhesin, in contrast to the E. coli Pap pilin, which requires a pilus-associated protein for binding to specific epithelial receptors (De Reeet al., 1987; Lund et al., 1987).
The adhesin properties of the pilin C-terminus have been attributed to a region containing two cysteine residues that form a disulfide loop. Irvin et al. (1989) proposed that the epithelial cell-binding domain of the pilus is contained within the sequence of this disulfide loop. To support their theory, these investigators demonstrated that monoclonal antibodies (mAb) specific for this region of pilin inhibited the adherence of P. aeruginosa PAK to BEC, whereas mAb directed against other regions of pilin had no inhibitory effects (Doig et al., 1990; Lee et al., 1989). Farinha et al. (1994) replaced the amino acid residues that form the disulfide loop in PAOl pilin to evaluate the contribution of this domain to adherence to human pneumocyte A549 cells. Compared to PAOl, the mutant strain displayed a 10-fold decrease in adherence and also displayed a significant decrease in virulence in the A.BY/SnJ mouse infection model
(Farinha etal., 1994). These results supported the idea that the C-terminal disulfide loop domain contains the pilus adhesin.
The receptors for pilus attachment to epithelial cell surfaces are gangliosides, the carbohydrate-rich sphingolipids that contain sialic acid and are major components of eukaryotic membranes. In vitro studies have shown that pilin binds to the glycosphingolipids asialo-GMi and asialo-GM2 (Gupta et al., 1994; Lee etal., 1994; Sheth, 1994). The specific carbohydrate receptor sequence of these glycosphingolipids that is recognized by the pilus adhesin was identified as GalNAcB(l-4)Gal [N-acetylgalactosamine-(61-4)-galactose] (Krivan et al., 1988; Sheth, 1994). Lee et al. (1994) used colloidal gold-coated asialo-GMi to demonstrate that the C-terminal disulfide epitope is exposed only at the pilus tip, despite the fact that this domain is present in each pilin subunit.
Structure of P. aeruginosa pilin The amino acid sequence of P. aeruginosa pilin varies from one species to another, and seven different pilins have been identified (Irvin, 1993). The central region of each of these pilins is the immunologically dominant region that varies among strains, whereas the C-terminal region containing the pilus adhesin is semi-conserved (Fig. 1; Sastry et al., 1985b; Watts et a l., 1983b). The N-termini of these pilins contain highly conserved, hydrophobic stretches of 29 amino acids (Fig. 1; Sastry et al., 1985b; Watts et al., 1983b).
P. aeruginosa pilin, encoded by pilA, is initially produced as a larger precursor protein containing a positively charged leader peptide that is six residues long in the majority of P. aeruginosa strains (Nunn et al., 1990; Pasloske et al., 1985; Sastry et al., 1985). The leader peptide is cleaved from prepilin during export across the cytoplasmic membrane by the prepilin leader peptidase, PilD (Nunn and Lory, 1991; Strom and Lory, 1992; Stromet al., 1993). Following cleavage of the leader sequence, PilD methylates the phenylalanine residue that is exposed at the N-terminus of the mature pilin (Nunn and Lory, 1991; Strom and Lory, 1992). This N-terminal phenylalanine residue is highly conserved in mature pilins and has resulted in the designation of NMePhe for the pili assembled from these pilins (Frost etal., 1978). These prepilins also contain an invariant glycine residue at the -1 position; thus, the PilD consensus cleavage site lies between the conserved glycine and phenylalanine residues (Fig. 1; Nunn and Loiy, 1991; 1992).
Substitution of the phenylalanine residue does not always affect PilD processing of prepilin. This was demonstrated by Strom and Lory (1991; 1992), who replaced the phenylalanine in three mutants with a methionine, serine, or asparagine, and all three mutants produced mature pilin. The serine residue at this position (+1) was not methylated, and yet pili were still Figure 1. Structure of P. aeruginosa PAOl prepilin protein. The prepilin proteins of P. aeruginosa are highly homologous at the amino-terminus, including the leader peptide (black box), and are moderately similar between residues 31-55 (gray box). The remainder of the protein is variable (white box), although there is a conserved region (striped box) at the carboxyl terminus
which contains two highly conserved cysteine residues believed to form a disulfide-loop adhesin domain (S—S) (Irvin et al., 1989). The central portion of the protein represents the immunodominant region. The PilD cleavage site, which is also conserved, occurs between a glycine (G) residue and the phenylalanine (F) residue that is subsequently methylated by PilD (Nunn and Lory, 1991; 1992). Two conserved tyrosine residues (Y) implicated in pilin
subunit-subunit interactions are also shown (Watts et al., 1983).
10 Processed by PilD
Conserved Variable
l_J ______I______I______I______I______I______I______i Residue -5 0 20 40 60 80 100 120 144
Figure 1 12
assembled (Strom and Lory, 1991). In contrast, a glycine substitution resulted in processed pilin subunits that were 50% N-methylglycine but were not assembled into pili (Strom and Lory, 1991). Thus, the role of the N- methylation of pilin is not clear. However, substitution of the invariant glycine residue at position -1 or of a conserved glutamate residue at +5 resulted in a lack of processing (Strom and Lory, 1991).
In addition to its role as the prepilin leader peptidase, PilD is also
essential for processing four components (XcpT,U,V,W) of the P.aeruginosa general protein secretion apparatus and has the alternative designation, XcpA (Bally et al., 1992; Nunn and Lory, 1992, 1993; Strom et al., 1991, 1993). Secretion of extracellular proteins by P. aeruginosa occurs by one of two pathways: the general secretory pathway, which is used for proteins with leader peptides, or the sec-independent pathway that is used to secrete proteins lacking leader sequences (Lory, 1992; Pugsley, 1993). PilD is a component of the general secretory pathway, which is a two-step process (Nunn and Lory, 1992).
In the first step of the general secretory pathway, which involves the translocation of the precursor protein across the cytoplasmic membrane, Sec proteins interact with the N-terminal signal sequence of the secreted protein (Pugsley, 1993). The signal sequence is then removed by the leader peptidase, and the protein is transported from the periplasm across the outer membrane by the Xcp protein components (Ballyet al., 1992 ; de Groot et al., 1991). 13
XcpT, XcpU, XcpV, and XcpW are processed by PilD as they are translocated across the cytoplasmic membrane and then assemble a putative secretory apparatus required for the secretion of extracellular proteins (Bally et al., 1992). PilD mutants, which are deficient in the secretion of extracellular proteases, accumulate these four Xcp proteins in the periplasm (Nunn and Lory, 1992; Strom et al., 1991).
P. aeruginosa pilus biogenesis Following cleavage and methylation by PilD, the mature pilin subunits are assembled into a pilus by a mechanism that is not understood. The conserved hydrophobic N-terminus of pilin subunits has been implicated in subunit-subunit interactions on the basis of experiments directed at two conserved tyrosine residues (Tyr24 & Tyr27)within this region (Fig. 1; Watts et al., 1983). These studies revealed that the region containing these tyrosine residues is not accessible to external solvents in native pili but is exposed in the pilin dimers that result from dissociation of the pili with octyl glucoside (Watts et al., 1983). This suggests that these residues are buried as a consequence of subunit interactions in the assembled pilus, yet it is not understood how the subunits are brought together during the assembly process.
The specific location of pilus assembly is also not known, although one possibility is that the pilus is assembled in the cytoplasmic membrane and thereby spans the entire cell wall in its functional form. A second possibility is 14 that the pilus is anchored in the outer membrane. Cellular fractionation studies aimed at localizing pilin pools have not clarified these questions. Mature, unassembled pilin was localized to both cytoplasmic and outer membrane fractions by Watts el al. (1982) and Nunn et al. (1990); however M.A. Russell (personal communication) detected a pilin pool in the cytoplasmic membrane only.
To date twenty-one genes in addition to pilA and pilD are known to be involved in pilus biogenesis and function in P. aeruginosa (Table 2) and are clustered at four unlinked loci on the chromosome (Fig. 2). The majority of these genes were classified as pilus biogenesis genes on the basis of their m utant phenotypes (Table 2), but the functions of only a few of the respective gene products are well established.
PilR and PilS are members of the family of two-component signal transduction proteins and control expression of the pilin subunit gene, pilA (Hobbs et al., 1993; Ishimoto and Lory, 1992). PilR is a transcription activator that binds a tandem array of consensus sequences, similar to the NifA recognition sequence, approximately 100 nt upstream of the pilA promoter
(Jin etal., 1994). The presence of four essential PilR-binding sites prompted these investigators to propose that PilR may form a multimer or bind DNA cooperatively. Mutations in pilR result in an inability to transcribe pilA and Table 2. Pseudomonas aeruginosa loci. pit 15
Gene Relevant Mutant Phenotype® Reference |
pilA Pasloske et al., 1 no production of pilin subunit;
pilN Martinet al., nonpiliated; (J^jTw- 1995
pilO Martin etal., nonpiliated; ({^jTw- 1995 etal., pilP Martin nonpiliated; (JjH; Tw- 1995 Martin etal., pilQ nonpiliated; (j^jTw- 1993 Ishimoto and pilR no expression ofpilA; (J^; Tw- Lory, 1992 16 Table 2. (Continued)
Gene Relevant Mutant Phenotype® Reference | | pilS no expression ofpilA Hobbs et al., 1993 This study; pilT hyperpiliated;
17 p ilA A C ,D pUV pilR ,S pilE llv-226 pilMyN,0,P,Q thlsl proB hls-4 '‘Cys-59Xjy f 0* 7570 * c'.''vv/y s - 12 5 7 LprgB pur-67, 1(TX 11 \,^trpA,B
P. aeruginosa X-met-28 pilTyU cys-54 PAOI pilG,HJ,J,K,Ll,L2
trpC,D roA
cat A
amIE met9001 leu-10 p ilZ
Figure 2 00 19 thus the absence of pili (Ishimoto and Lory, 1992). In addition to PilR, the alternative sigma factor RpoN (a54) is also required for transcription ofpilA
(Ishimoto and Lory, 1989).
PilS shares homology with the histidine kinase sensor proteins of the two-component regulatory family (Hobbs et al., 1993; Boyd et al., 1994). As in other two-component systems, an environmental signal presumably activates PilS, which then autophosphorylates at a conserved histidine residue and transfers the phosphate to the response regulator, PilR. Boyd and Lory (1996) reported that PilS has two potential input domains and may therefore respond to two environmental signals. These investigators also mentioned that PilS might respond to the cell division cycle to maintain the unipolar position of the pili on daughter cells (Boyd and Lory, 1996). They supported this hypothesis by the fact that PilS displays a high structural homology with DivJ, which is a histidine kinase involved in cell-cycle regulation in Caulobacter crescentus (Boyd and Lory, 1996). However, no investigators have identified the conditions that repress pilA expression or the signal(s) that is recognized by PilS.
RIB and R1C are both required for assembly of pilin monomers into the pilus, although they appear to have distinct functions (Nunn et al., 1990).
Little is known about R1C except that its predicted amino acid sequence implies that it is an integral cytoplasmic membrane protein (Nunn et al., 20
1990). PilB is a putative nucleotide-binding protein that may provide energy for pilus assembly by ATP hydrolysis (Turner et al., 1993). Alternatively PilB may modify other components of the pilus assembly machinery by phosphorylation (Turner et al., 1993). PilB is presumably a cytoplasmic protein that associates with the inner membrane (Turner et al., 1993), possibly by interacting with PilC (Hobbs and Mattick, 1993). There is no evidence, however, that PilB and PilC actually do interact.
PilE and PilV display significant sequence identity to prepilins, with the N-termini of these proteins displaying the highest homology (Aim and Mattick, 1995; Russell and Darzins, 1994). Both PilE and PilV contain signal sequences like that of PilA and both are processed by PilD (Aim and Mattick, 1995; Russell and Darzins, 1994). Several functions have been proposed for these proteins on the basis that their N-termini are homologous to the region of pilin that is involved in subunit-subunit interaction of pilin (Aim and Mattick, 1995; Russell and Darzins, 1994; Watts et al., 1983). This homology suggests that PilE and PilV may interact with pilin at some stage of pilus assembly. PilE and PilV may interact with pilin during export to allow the pilin subunits to be presented in the correct conformation for assembly or to assure that assembly occurs at the appropriate pole (Aim and Mattick, 1995; Russell and Darzins, 1994). Another possibility is that PilE and PilV somehow anchor the pilus to the membrane (Russell and Darzins, 1994). Studies examining the ability of PilE to interact with pilin are in progress (M. Russell, personal communication). 21
PilM, PilN, PilO and PilP were recently identified, and no functions have been proposed for these proteins (Martinet al., 1995). The only information known about these proteins is provided by their predicted amino acid sequences and their homologies with previously characterized proteins. PilM includes sequence motifs found in the cell division protein FtsA from E. coli and
B. subtilis (Martin et al., 1995). PilN and PilO have internal hydrophobic domains, suggesting that these proteins may be located in the cytoplasmic membrane (Martin etal., 1995). PilP contains a hydrophobic leader sequence that is characteristic of lipoproteins (Martinet al., 1995).
The pilQ gene is downstream of the pilM operon, and evidence suggests that pilQ is independently transcribed (Martin et al., 1993). PilQ displays amino acid sequence homology to an outer membrane antigen of Neisseria gonorrhoeae and contains an N-terminal hydrophobic signal sequence (Martin etal., 1993). On this basis Martin et al. (1993) proposed that PilQ is an outer membrane protein that potentially forms a porthole that allows the passage of the pilus through the outer membrane. However there is no evidence to support this hypothesis.
The most recent pilus biosynthesis gene to be identified is pilZ, which encodes a type of protein that had not been previously represented in sequence databases (Aim et al., 1996). Since a pilZ mutant is nonpiliated but able to form pools of processed pilin in the membrane, PilZ is probably required for the export and assembly steps of pilus biosynthesis (Aim et al., 1996). 22
Darzins (1993, 1994, 1995) identified an operon of pilus biosynthesis genes which encode proteins that are homologous to the chemotaxis proteins of
enteric bacteria and Myxococcus xanthus. The putative protein products of these genes, PilG-K, are believed to be components of a signal transduction pathway that controls pilus synthesis and twitching motility (Darzins, 1993, 1994, 1995). Interestingly, apilK mutant displays the wild type phenotype with regards to pilus biosynthesis, twitching motility, and phage sensitivity (Darzins, 1995). The significance of the PilG-K proteins will be discussed further in Chapter V.
Downstream of the pilG operon are two genes designated p ilLl and pilL2 (J. S. Mattick, personal communication). The products of these genes are also homologous to chemotaxis proteins of the enteric bacteria, implicating the PilL proteins in the signal transduction pathway proposed by Darzins
(1995; J. S. Mattick, personal communication).
Mutants of either pilT or pilU are hyperpiliated, although this is not the result of an increase in pilin expression (this study; Whitchurch et al., 1990; Whitchurch and Mattick, 1994a). Despite the presence of pili on these mutants, neither demonstrates twitching motility (this study; Whitchurch et al., 1990; Whitchurch and Mattick, 1994a). The pilTge ne is the subject of this study and will be addressed in further detail, along with pilU, throughout subsequent chapters. 23
Hobbs and Mattick (1993) presented a model for P. aeruginosa pilus assembly that incorporates the putative functions of a few of the better characterized Pil proteins (Fig. 3). This model proposes that the pilus is assembled and anchored in the cytoplasmic membrane, where PilC contributes to the assembly process. The model also suggests that an association of PilB with PilC couples the energy from ATP hydrolysis to pilus assembly.
Occurrence of Type IV pili in other bacteria
The pili of P. aeruginosa belong to the class of type IV pili that is expressed by a number of diverse, pathogenic gram-negative bacteria (Table 3). Many investigators have demonstrated that these pili contribute to the virulence of these pathogens (Britigan et al., 1985; Doig et al., 1988; Jackman and Rosenbusch, 1984). The bacteria in this group share similarities in the amino acid sequences of their pilin subunit proteins, including the presence of an N-methylated amino acid as the first residue of the mature protein. The pilus assembly components among these organisms are also conserved, as is the mechanism of transcriptional regulation of the genes encoding their pilin subunits. This class of pili is also divided into subgroups according to the presence of a short, positively charged leader peptide. Group A pilins, which indude P. aeruginosa PilA, are synthesized as prepilins that are processed and methylated by a leader peptidase. The cleavage site in all of these prepilins is flanked by a conserved glyrine (-1) and phenylalanine (+1) (Strom and Lory, 1993). Figure 3. Schematic representaion of pilus assembly in Pseudomonas aeruginosa. The potential roles of several Pil proteins are included in this model (adapted from Hobbs and Mattick, 1993), which proposes that PilB and PilC associate to couple energy generated by ATP hydrolysis with the assembly of the processed pilin subunit. PilQ is postulated to be an outer membrane protein which forms a channel through which the pilus protrudes. OM, outer membrane; CM, cytoplasmic membrane.
24 JVrSj-v.-A v . : r .‘ • .V > VV.-.-.-a .’ ;.'V PilD/XcpA rvv- ■.•I
ATP?
prepilin
Figure 3 TableS. Bacteria displaying type IV pili.
Bacterium Pathogenic Significance R eference Group A Pseudomonas aeruginosa opportunistic pathogen Frost et al., 1978 Neisseria gonorrhoeae gonorrhea, conjunctivitis Meyer et al., 1984 Neisseria meningitidis meningitis, pneumonia, arthritis, urethritis Potts and Saunders, 1988 Dichelobacter CBacteroides ) nodosus ovine footrot McKern et al., 1983 Moraxella nonliquefaciens conjunctivitis, endophathalmitis T0njum et al., 1991 Moraxella bovis infectious bovine keratoconjunctivitis Marrs et al., 1985 Moraxella lacunata conjunctivitis, endophathalmitis Marrs et al., 1990 Rao and Progulske-Fox, Eikenella corrodens infections from human bites 1993 Branhamella catarrhalis bronchopulmonary infections, otitis media Marrs and Weir, 1990 Group B Escherichia coli (BFP)* childhood diarrhea Donnenberg et al., 1992 Vibrio cholerae (TCP)b Asiatic cholera Faast et al., 1989
* BFP, bundle-forming pili of enteropathogenic (EPEC) strains b TCP, toxm-coregulated pili
to os 27
Group B pilins contain several deviations from the conserved sequences in the Group A pilins: the leader sequences are longer in Group B prepilins; the first residue in the mature pilin is not phenylalanine; and the C-termini do not contain the short region of homology found in the Group A pilins (Fig. 1; Strom and Lory, 1993). However, the pilins in both groups retain the pair of
conserved cysteines at the carboxyl terminus that presumably form the intrachain disulfide bond which may act as the pilus adhesin (Strom and Lory, 1993). There are currently only two bacteria known to produce Group B pilins: Vibrio cholerae (Faast et al., 1989) and enteropathogenicE. coli (EPEC) (Donnenberg et al., 1992). The V. cholerae pilus has been designated a toxin
coregulated-pilus because its expression is coupled with cholera toxin production (Miller etal., 1987). The bundle-forming pili of EPEC pili are unique in that they are encoded by a plasmid (Donnenberg et al., 1992).
As with the P. aeruginosa pilA gene, expression of the pilin subunit gene pilE in N. gonorrkoeae requires the alternative sigma factor RpoN, and RpoN consensus promoters have been identified upstream of other type IV pilin
subunit genes as well (Ishimoto and Lory, 1989; Fyfe et al., 1995). The accessory proteins required for pilus biosynthesis are also highly conserved in
type IV-piliated bacteria. This is evident in the ability of P. aeruginosa to assemble pili encoded by plasmid-bome copies of the pilin subunit genes from D. nodosus (Elleman et al., 1986; Mattick et al., 1987) and M. bovis (Beard et al., 1990; Elleman et al., 1990). In contrast, E. coli is able to express these heterologous pilin genes but cannot assemble the pilin subunits into pili, 28 suggesting that the E. coli pilus biogenesis system is different from the type IV pilus assembly system (Elleman et al., 1990; Marrs et al., 1985). Interestingly, expression of heterologous pilin genes in P. aeruginosa inhibited production of native pilin subunits, suggesting that a feedback pathway may be involved in the regulation of P. aeruginosa pilA expression (Mattick et al., 1987).
Some variations are seen, however, among the type IV pilus biosynthesis systems. N. gonorrhoeae is unique among this class of bacteria in that it undergoes phase- and antigenic variation of its pilin, which contributes to its virulence (Jonsson et al., 1991). Another novel trait is that gonococcal pilin can be processed either before or after the hydrophobic region at the amino terminus, which results in the production of two types of pilin: a full- length pilin that can be assembled or a truncated soluble (S) pilin that is secreted (Haas et al., 1987).
In addition to acting as adhesins, type IV pili appear to mediate several other processes in these bacteria. The presence of pili on some of the bacteria in this class, including E. corrodens and Moraxella spp., results in the formation of colonies that spread and form small depressions (agar corrosion) in solid media (B0vre and Fr0holm, 1972; Rao and Progulske-Fox, 1993). Pili were initially implicated in this process because nonpiliated mutants of M. bovis do not exhibit agar corrosion (B0vre and Fr0holm, 1972; Chandler et al., 1985).
However, other investigators have suggested that agar corrosion is likely due to an extracellular product produced by these bacteria and that the 29 nonpiliated, noncorrosive mutants may have also been deficient in secretion of a putative agar-corroding enzyme (Chandler etal., 1985). Interestingly, the agar-corroding phenotype ofM. nonliquefaciens is subject to phase variation, as this bacterium forms either spreading, corroding colonies (SC) or smooth, non-corroding colonies (N) (T0njum et al., 1991). The cells that form SC colonies are usually piliated while those forming N colonies are not (T 0njum et al., 1991).
Type IV pili are believed to mediate natural competence of N. gonorrhoeas, N. meningitidis, E. corrodens, and Moraxella spp. (Biswas etal.,
1977; B 0vre and Fr0holm, 1972; Rao and Progulske-Fox, 1993; T 0njum et al.,
1993). However the exact role of pili in competence is not clear, since competence ofN. gonorrhoeas is not affected by the addition of pilus-specific antiserum or purified pilin (Mathis and Scocca, 1984).
All bacteria with type IV pili demonstrate twitching motility, which is believed to be mediated by the retraction of type IV pili. P. aeruginosa has been the model system for the study of type IV pilus retraction and twitching motility, which are discussed in detail below.
Pilus Retraction The hypothesis that type IV pili retract is founded primarily on circumstantial evidence gathered by Bradley, who demonstrated that P. 30
aeruginosa pili act as receptors for certain bacteriophages and that the average lengths of the pili were reduced following adsorption to the
bacteriophage PP7. In addition, the adsorbed phages were usually present at the bases of the wild type PAOl pili but were randomly distributed along the pili of the phage-resistant mutant PA068. Bradley (1972c) concluded that each PAOl pilus had retracted into the cell until the adsorbed phage contacted the cell surface. In contrast, the PA068 pili must not have been capable of retraction, causing the phage particles to remain at their original sites of adsorption. Bradley (1972c) was the first to propose that pilus retraction was required for infection with pilus-specific phages.
Bradley (1972c) also noted that if PAOl cells had not been treated with pilus-specific antiserum prior to negative staining or shadowcasting
procedures, pili were no longer present on the cells. However, small black dots could be seen on the bacterial surface near the poles of these cells, in the region where the pili were expected to be present (Bradley, 1972c). Bradley
postulated that in the absence of antibodies, which apparently prevented retraction, the staining chemicals or the mounting of the cells to the EM grid
stimulated the pili to retract (1972c). He concluded that pili, although shortened, still remained outside the cell following phage adsorption because the bound phage particle prevented the pili from being fully retracted into the cell upon manipulation for EM. Thus, according to Bradley (1972c), pilus retraction is stimulated by phage adsorption and possibly by exposure of the cell to certain chemicals or to a solid surface. 31
Treatment of PAOl with pilus-specific antiserum not only prevented shortening of the pili, it also resulted in a 20-fold increase in the number of pili on each cell relative to the number present on untreated PAOl (Bradley, 1972c). Subsequent studies have supported this earlier observation and shown that inhibition of retraction whether it occurs genetically via a mutation in a gene required for retraction, or physically by the addition of anti-pilin antibodies, results in hyperpiliation (Bradley, 1974; this study; Whitchurch et al., 1990; Whitchurch and Mattick, 1994a).
The evidence for retraction is indirect, and the mechanism by which retraction occurs has not been elucidated. Bradley (1972b,c) suggested that the pilin subunits may dissociate by a depolymerization mechanism that occurs at the base of the pilus. A similar pilus retraction model was proposed for the E. coli F-pilus (Novotny and Fives-Taylor, 1974). The plasmid-encoded
F-pili ofE. coli promote cell-to-cell contact required during bacterial conjugation and act as receptors for bacteriophages (Burke et al., 1979; Frost et al., 1985).
Evidence for the retraction of F-pili includes electron microscopy observations that the pili quickly disappear from the cell surface following phage infection, treatment of the cells with cyanide or arsenate, or exposure to high temperatures (46°-50°C), whereas free pili cannot be detected in the medium (Burke et al., 1979; Novotny and Fives-Taylor, 1974, 1978). As with type IV pili, the mechanism of retraction of the F-pilus has not been determined. 32
Twitching Motility Twitching motility (Fig. 4) was initially observed by Lautrop (1961) as a spreading zone surrounding colonies of the nonflagellated Acinetobacter calcoaceticus. At the time only two types of surface motility had been characterized: swarming, which is flagellum-dependent, and gliding, which is flagellum-independent (Henrichsen, 1983). Henrichsen (1972) reported that twitching bacteria move primarily as individual cells, although small motile clusters of cells can sometimes be observed, and Henrichsen described these individual movements as intermittent and jerky. In addition, Henrichsen (1983) reported that the direction of movement is random, as is the frequency of the jerky motions.
Although twitching motility initially appeared to be a form of gliding motility (Lautrop, 1961), these two modes of surface motility are distinct.
Unlike twitching cells, gliding cells demonstrate a smooth progression of movement in the direction of their longitudinal axis and only occasionally stop or reverse their direction of movement (Hogkin and Kaiser, 1979a; Lapidus and
Berg, 1982). Growing myxobacterial cells do not normally glide independently, instead they associate into loose aggregates that exhibit cooperative movement (Hodgkin and Kaiser, 1979b).
As additional evidence that twitching and gliding are separate phenomena, Henrichsen (1983) reported that the average speed of progression observed with each of these forms of motility differs. The progression of Figure 4. Twitching motility by Pseudomonas aeruginosa PAOl. Strains were stab-inoculated to the bottom of a 1.0% LB agar plate. P. aeruginosa strain
PAOl exhibits twitching motility and forms a circular motility zone surrounding the point of inoculum (A), whereas a nontwitching mutant strain does not produce a motility zone (B). The smaller, white zones are the cells that are growing on the surface of the medium.
33
35
twitching cells is proportional to culture growth, whereas the spreading zone formed by gliding cells correlates with the average speed of the movements of the cells (Henrichsen, 1983). This observation prompted Henrichsen (1983) to conclude that gliding cells move actively, in contrast to twitching cells which are being moved by the physiochemical forces of their environment.
Henrichsen and Blom (1975a,b) first noted that the bacteria known to twitch possess polar pili. On this basis Henrichsen (1975a) systematically
surveyed more than 50 species of gram-negative and gram-positive bacteria in an attempt to characterize the occurrence of twitching motility. All of the strains that displayed twitching motility were gram-negative rods, both aerobic and anaerobic, that possessed polar pili (Henrichsen, 1975a). Twitching motility was later observed in one gram-positive bacterium, Streptococcus sanguis (Henriksen and Henrichsen, 1975), but little information is included
about this bacterium in reports concerning twitching motility.
Henrichsen (1975b,c) examined the effects of numerous environmental factors on the twitching motility of several strains of Acinetobacter calcoaceticus and determined that the availability of water on the agar surface is a crucial requirement for this surface-dependent motility. Maximal twitching, determined by colony diameter, occurred on thick (30 ml) agar plates that were slightly dried and incubated at high humidity (Henrichsen, 1975b).
When actively twitching cells were transferred to the surface of a freshly dried plate, twitching was halted unless a drop of liquid was also placed on the agar 36
(Henrichsen, 1975c). Decreases in the surface tension of the water phase of the medium, instilled by the addition of Tween 80 or sodium taurocholate,
resulted in decreased levels of twitching (Henrichsen, 1975c). On this basis, Henrichsen (1975b,c) proposed that twitching occurs in a film of liquid that surrounds the bacterial colony and permits spreading over the agar surface.
Thus, a decrease in surface tension inhibits the formation of a sufficient layer of liquid surrounding the colonies, and twitching motility cannot occur.
Twitching was also inhibited when Henrichsen substituted agarose for agar; thus, he proposed that the hydrophobic surface of agarose does not allow the formation of a sufficient film of liquid around the bacterial colonies,
supporting his theory that some liquid is required for twitching motility (Henrichsen, 1975b). Henrichsen (1975b) also reported that twitching cells demonstrate an affinity for the air-water interface when a drop of water was placed on a spreading zone. Actively twitching cells were seen floating on top of the water drop, while nontwitching cells remained within the water drop. Collectively, these studies implied that cells within liquid cannot twitch.
Instead, cells need to exist at an air-water interface, which occurs on solid surfaces, for twitching motility to occur.
To determine if twitching was influenced by changes in the electrokinetic potential of the cells, Henrichsen (1975b) altered the ionic strength of the medium by the addition of various concentrations of different ions to the medium. The results from these experiments were inconclusive, 37
though, because many of the compounds inhibited growth. In the same study Henrichsen tested the effect of pH and reported that twitching was more pronounced on medium with a pH of 9.0 than at a neutral pH; however, he considered the results to be inconclusive due to deviations in the data. The brand of agar used in the medium also affected twitching, and a concentration
of 1-2% agar was optimal. Lastly, Henrichsen (1975c) reported that twitching
is reversibly inhibited by a temperature shift to 4°C, suggesting that only
growing cells twitch.
When growth of A. calcoaceticus colonies was monitored from the single cell stage by microscopy, twitching was not observed until a minimum of 50- 100 cells was present (Henrichsen, 1975c). Henrichsen also noted that other strains did not twitch until many more cells were present and postulated that a minimum number of cells is necessary for twitching to occur.
Henrichsen (1975c) theorized that the pili on twitching bacteria could confer an affinity for the air-water interface if the pili themselves were
hydrophobic. It is now known that the N-terminus of the pilin subunit is hydrophobic, which supports Henrichsen’s early theory. However, Henrichsen (1975c) did not think that pilus retraction was involved in twitching, despite his
belief that pili mediate twitching motility. Indeed, Henrichsen (1983) denounced Bradley’s theory that pilus retraction pulls the cell outward, instead maintaining that the outermost cells are not pulled by the pilus but are instead pushed by the innermost cells of the colony. 38
Bradley (1980a) reported that the strains of P. aeruginosa used by Henrichsen did not have retractile pili, and that Henrichsen’s observations may not apply for strains that produce retractile pili. Bradley did not, however, provide evidence that Henrichsen’s strains produced non-retractile pili. To better understand twitching in P. aeruginosa , Bradley used the well- characterized strains PAOl and PAK. Bradley (1980a) also claimed that twitching could best be observed microscopically within the scattered, single layer of cells along the edges of colonies, as opposed to Henrichsen’s method of simply measuring colony diameter.
Bradley (1980a) observed that strains without pili and strains with non- retractile pili were not capable of twitching. Antipilus antiserum and pilus- specific bacteriophages, both shown to inhibit pilus retraction (Bradley, 1972c), prevented twitching motility of P. aeruginosa when either was added to the agar medium (Bradley, 1980a). On the basis of Bradley’s studies, it is now generally accepted that pilus retraction is required for twitching motility in P. aeruginosa, although the actual mechanism by which retraction mediates twitching motility has not been elucidated. It is not known if twitching motility occurs in the environment or during infection with type IV piliated bacteria as this type of motility has only been observed in the laboratory setting. 39
Goals of this project This project was initiated as part of a large-scale investigation to identify and characterize genes that are required for type IV pilus biosynthesis in Pseudomonas aeruginosa PAOl. Basic genetic techniques resulted in the isolation of a large number of mutants that were defective in pilus biosynthesis or function. Experiments for this project were designed to characterize one of these mutants with respect to its deficiencies in pilus biosynthesis. These studies resulted in the identification of pilT, which is required for pilus function.
Mutants were used to study the role of pilT in the pilus-mediated processes of virulence, twitching motility and phage sensitivity.
The remainder of this project focused on the regulation of p ilT expression in an attempt to elucidate the environmental signals that control pilus biosynthesis and function. The objectives of these experiments were to identify the p ilT transcript, the pilT promoter, and regulatory elements involved in pilT transcription. CHAPTER H
MATERIALS AND METHODS
Media and Culture Conditions P. aeruginosa and E. coli strains were routinely grown in Luria-Bertani
(LB) medium (Gibco BRL, Gaithersburg, MD) at 37 °C. General purpose solid media contained agar at a final concentration of 1.5% (wt/vol). Liquid cultures were routinely grown in LB at 37°C with shaking at 300 rpm in a New
Brunswick floor incubator (New Brunswick Scientific, Edison, NJ). Clarke minimal medium was prepared as reported by Brammer and Clarke (1964) except that the trace element solution was omitted. Glucose was added to the medium to a final concentration of 50 mM and the appropriate amino acids were added to a final concentration of 1 mM. The antibiotic concentrations ordinarily used for the growth of E. coli were as follows: 100 |ig/ml ampicillin; 40 pg/ml kanamycin; 25 pg/ml tetracycline. For selection of drug-resistant P. aeruginosa transconjugants or transformants, Pseudomonas Isolation Agar (PIA; Gibco BRL) was supplemented with carbenicillin (300 |ig/ml) or tetracycline (300 |ig/ml).
40 41
Bacterial Strains, Plasmids, and Bacteriophages All bacterial strains, plasmids, vectors and bacteriophages that were used in this study are listed in Tables 4 and 5. Appendix B contains a list of those plasmids that have been constructed for this study. Spontaneous phage-resistant pil mutants of PAOl used in this study were isolated by
mixing a 1:10 dilution of an overnight culture of PAOl with an excess of B3 cts or D3112 cts phage (102-103 pfu/cell). The mixtures were plated onto LB agar at a volume of 200 pi per plate, and plates were incubated at 42°C. Surviving colonies were streaked onto individual LB agar plates and tested for phage sensitivity (described below).
Bacteriophage lysates were prepared by thermoinduction (Roncero et al., 1990) from P. aeruginosa lysogens of B3 cts3, D3112 cts, and F116L cts53 containing temperature sensitive mutations in their repressors (cts).
Lysogenic strains were grown overnight at 30 °C with shaking in LB
supplemented with 1 mM MgS0 4 . Cultures were then diluted 1:100 in LB medium and incubated at 30°C until they reached mid-exponential phase, at which point they were shifted to 42°C and incubated for 2.5 hr or until lysis had occurred. Chloroform was added to a final concentration of 1% to the lysates to kill any remaining cells, and cellular debris was removed by centrifugation at 2000 x g. Lysates were titered in LB top agar containing 0.7% agar (wt/vol) and 1 mM MgS(>4 and were stored for approximately one year at 4°C without a decrease in titer (A. Darzins, unpublished research). 42
Table 4. Bacterial strains and bacteriophages used in this study.
Source or Strain/Phage Description8 Reference Escherichia coli strains endA hsdR17(re-niB-) supE44 DH5a thi-1 recAl gyrA(Nalr) relAl Han ah an, 1983 MlacZYA-argF) §80lacZAM15 Aigpt-proA) leuB6 thi-1 lacYl HB101 hsdS20 recA rpsL20(StrO ara-14 Boyer and Roland- 1 galK2 xyl-5 mtl-1 supE44 mcrBb Dussoix, 1969 1 RP4 2-Tc::Mu-Km::Tn7 pro S17-1 r* m+; Tpr Smr Simon et al., 1983
F- ompT re-niB-; X lysogen carrying BL2KDE3) phage T7 RNA polymerase gene Studier and Moffatt, under control of Piacuvs 1986 end A1 gyrA96 thi-1 hsdR17 Epicurian Coli supE44 relAl lac mutD5 mutS Stratagene XLl-Red mutT Tn 10 (Tet)r Pseudomonas aeruginosa strains PAOl prototroph; wild type pilus Holloway et al., I production and fimction 1979 Darzins and I CD10 D3112 cts lysogen of PA04141 Casadaban, 1989b | Darzins and 0 I ADD1193 B3 cts3 lysogen of PAOl Casadaban 8 (unpublished data) PAS246 F116L cts53 lysogen of PAOl R.V. Miller PA01::mini D180 (laclq lacUV5 ADD1976 Brunschwig and T7 RNA polymerase) Darzins, 1992 II HOD1 frameshift mutation in pilT This study | 1 PAO-PT pilT::pBB322 This study | 43
Table 4. (Continued)
Source or I Strain/Phage Description® Reference | Pseudomonas aeruginosa strain s PAO-PU Tcr; ApilU'.tet This study PAO-TU Tcr; pilT- ApilUtet This study Ishimoto and Lory, PAK-N1 Tcr; rpoN::tet 1989 ilv-226 his-4 lys-12 met-28 trp-6 PA0222 proA82 Nalr Haas, 1983 pur-67 thr-9001 cys-59 PAO1042 proB65 Nalr H. Matsumoto Bacteriophages
D3112 cts cts (temperature-sensitive Krylov et al., repressor) 1980 cts (temperature-sensitive B3 cts-3 repressor) A. Darzins
F116L cts53 cts (temperature-sensitive Krishnapillai, repressor) 1971
® Abbreviations: Smr, Nalr, Tpr and Tcr designate resistance to streptomycin, naladixic acid, trimethoprim, and tetracycline, respectively 44
Table 5. Plasmid vectors used in this study.
Description® Source o r 0 Plasm id Reference | Escherichia coli vectors I pUC 18/19 ColEl ori; cloning vector; Apr Yanisch-Perron et al., 1985 pK19 pUC19 derivative; cloning vector; Kmr Pridmore, 1987 pBR322 ColEl ori; cloning vector; Apr Tcr Sutcliffe, 1979 ColEl ori; Tra+ (RK2); helper Figurskiand pRK2013 plasmid for triparental matings; Kmr Hehnski, 1979 expression vector, A. Darzins, pADD621 T7 <|> 10 promoter; Apr unpublished data derivative of pUC19; gene pNotl9 replacement vector; Apr Schweizer, 1992 H Pseudomonas aeruginosa vectors || FP2 Tra+, chromosome mobilization ability Holloway et al., 1979 broad-host-range derivatives of PUCP18/19 pUC18/19; with1.8 kb stabilizing Schweizer, 1991 fragment from pRO 1614; Apr broad-host-range cloning vector; Olsen etal., I I pR01614 derivative of pBR322; Tcr Apr 1982 I D. Helinski and pTJS140 broad-host-range cloning vector; repRK2 oriT reppMBilac' IPOZ; Apr T. Schmidhauser Brunschwig and pEB12 expression vector; T7
Table 5. (Continued)
Source or 1 Plasm id Description* Reference |
pMob3 sacB oriT cassette for gene replacement; Kmr Cmr Schweizer, 1992 broad-host-range lacZ promoter- Farinha and pQF50 probe vector; Apr Kropinski, 1990 broad-host-range luxAB promoter- Farinha and pQF70 probe vector; Apr Kropinski, 1990 a Abbreviations: Apr, Tcr, Kmr, Cmr, Nalr, Hgr: designate resistance to ampicillin, tetracycline, kanamycin, chloramphenicol, nalidixic acid and mercury, respectively. Tra+> self-transmissable. 46
Phage Sensitivity and Twitching Motility Assays Phage sensitivity spot assays were performed on solid media as described by Darzins (1993). The strain to be assayed was streaked onto LB agar containing the appropriate antibiotic. A 10 pi drop of each phage lysate
(109-1010 pfu/ml) was spotted onto the streak and allowed to dry. The plate
was incubated overnight at 37 °C and then examined for zones of clearing that
denoted phage sensitivity. HOD1 was also assayed for the ability to simply adsorb the bacteriophage D3112 cts by using the procedure of Roncero et al. (1990).
Quantitative phage sensitivity assays involved infecting Pseudomonas
strains with varying dilutions of the bacteriophage in top agar. Serial dilutions
were prepared for the bacteriophage using LB broth containing 1 mM MgS0 4 .
A 0.1 ml aliquot of each phage dilution was added to 4 ml LB top agar (0.7%
wt/vol) containing 1 mM MgSC>4 and 0.1 ml of an overnight culture of
Pseudomonas. The inoculated tubes of top agar were mixed and poured onto LB agar plates, which were incubated overnight at 37°C. The number of plaques observed with the mutant strains, in addition to plaque morphology,
was compared to that obtained with the wild type strain, PAOl.
Twitching motility assays were also carried out as described by Darzins (1993). Either a freshly isolated colony or a broth-grown culture was used for twitching assays. A sterile inoculating needle was used to stab the culture to the bottom of a polystyrene petri dish containing approximately 11 ml of LB 47 solidified with 1% (wt/vol) agar. The plate was incubated at 37°C for 24 hr then observed for the presence of a thin, circular motile zone surrounding the point of inoculation and occurring between the medium and the bottom of the petri dish. The diameter of the zone was recorded and results were compared to those obtained from the wild type strain PAOl. The plate was reincubated for another 24 hr period and observed again for any changes in the shape or size of the motile zone.
Electron Microscopy Techniques For visualization of pili by electron microscopy (EM), cells were prepared according to the gentle wash-off method of Henrichsen (1983). Cultures to be examined were grown overnight at 37 °C in 3 ml of LB broth and diluted 1:10 in fresh LB, and 100 fil was plated onto LB agar. Agar plates were incubated at
37°C for 5 hr to allow for formation of pili. Approximately 1 ml of phosphate- buffered saline (PBS) [2.2 mM KH 2PO4 , 4.2 mM Na2HPC>4 , 145 mM NaCl] was added to the surface of each plate, which was gently agitated to resuspend the cells. Carbon-coated 300-mesh copper grids (Polysciences, Inc., Warrington, PA) were floated in each cell suspension for 5 min, gently blotted, and air-dried. Grids were shadow-cast with germanium and observed under a Zeiss 10C electron microscope at the Electron Microscopy Facility of The Ohio State University, Department of Microbiology (Columbus, OH). 48
Mapping of the HOD1 phage-resistance mutation The phage-resistant mutation in HOD1 was mapped using the chromosome mobilizing (Cma) plasmid FP2 (Holloway et al., 1979), following the method described by Darzins et al. (1985). The donor, HOD1 harboring FP2, and the recipient naladixic acid-resistant auxotrophic strains PA0222 and PAO1042 were grown in LB broth to mid-exponential phase (Aeoo approximately1 .0 ) at 37°C. Uninterrupted matings were conducted by mixing
5 ml of the donor with each 5 ml of each recipient and reincubating at 37 °C for
90 min without shaking. Cells were collected by centrifugation and resuspended in 5 ml of saline (0.85% NaCl). Aliquots of 100 |il were plated onto Clarke minimal medium containing the appropriate amino acids and 600 pg/ml of naladixic acid to select against the donor. Following a 2-3 day incubation at
37°C, resulting prototrophic colonies were chosen randomly from each plate and assayed for resistance to phages D3112 cts, B3 cts-3, and F116L cts53.
The coinheritance frequency of phage resistance with each auxotrophic marker was determined.
L D 5 0 Studies
Virulence studies with HOD 1 were performed by Linda Glasier in the laboratory of Dr. W. Paranchych at the University of Alberta (Edmonton,
Alberta, Canada) according to the procedure described by Farinha et al. (1994). A brief description of the experimental procedure is provided here. The mice 49
used for these studies were congenic A.BY/SnJ mice (Pennington and Williams, 1979). These mice were not immunosuppressed prior to the virulence studies because the LD go values for P. aeruginosa are low in this strain of mouse; thus
increases in LD50 values of less virulent strains could be readily determined.
To determine the LD50 for the mutant strain HOD1, eight-week old mice were
injected intraperitonealy with a variety of concentrations of HOD1 or of the parent strain, PAOl. The mice were observed every hour for 16 to 48 hr following challenge. A population of ten mice were used for each challenge, and the number of survivors remaining at 48 hr was recorded.
Standard DNA Isolation and Manipulation Techniques Digestion with restriction enzymes (Gibco BRL), T4 DNA ligation reactions (New England Biolabs, Beverly, MA), transformation of E. coli and
quantitation of DNA were carried out according to standard techniques
(Maniatis et al., 1982). Plasmid DNA preparations fromE. coli were prepared by the rapid boiling method (Holmes and Quigley, 1981) and from P. aeruginosa
by the alkaline lysis procedure (Bimboim and Doly, 1979). A modified version of this procedure was used for preparing plasmid DNA for nucleotide sequencing determination. In this procedure, 160 pi of 5% cetyltrimethyl- ammonium bromide (CTAB; Sigma Chemical Co., St. Louis, MO) was added to the samples following the removal of the chromosomal DNA. A 15 min 50
centrifugation (15,000 rpm) precipitated the plasmid DNA, which was then dissolved in 400 pi of 1.2 M NaCl and re-precipitated with 95% ethanol.
Electrophoresis of DNA was carried out on agarose gels in IX Loening
(1967) buffer [36 mM Tris-HCl (pH 7.7), 30 mM NaH 2P 0 4, 1 mM EDTA].
Following electrophoresis, the DNA was visualized by staining the agarose gels
in ethidium bromide (ca. 1 pg/ml). DNA restriction fragments were routinely purified from agarose gels using the GeneCleanll kit (Bio 101 Inc., La Jolla, CA) prior to use in cloning and probe preparation.
Genomic DNA was isolated from P. aeruginosa using a modification of
the procedure of Woo et al. (1992). Cultures were grown overnight in 2 ml LB. Following centrifugation, the bacterial pellets were washed in 1 ml of TN buffer [10 mM Tris-HCl (pH 8.0), 10 mM NaCl] and suspended in 135 pi of TN. Cells
were lysed by the addition of 135 pi of TN-2% Triton X-100 (Sigma) and 30 pi of
5 mg/ml lysozyme (Sigma). Samples were incubated for 30 min at 37°C,
followed by the addition of 15 pi proteinase K (20 mg/ml; Sigma). After
incubation for 2 hr at 65°C, 100-200 pi of TE [1 M Tris-HCl (pH 8.0), 0.5 M
EDTA] was added to decrease the viscosity of the samples. As suggested by Woo et al. (1992), 8.4 pi of DNA preparation was used in a 50 pi restriction digest reaction. Samples were stored at -20°C. Reactions for determination of the nucleotide sequence were generated via the dideoxy chain termination method (Sanger et al., 1977) using the 7-
deaza-dGTP Sequenase 2.0 kit (U.S. Biochemical, Cleveland, OH) to reduce compressions. Double-stranded plasmid DNA was denatured under alkaline conditions by the addition of 2 pi of 2 M NaOH to 18 pi (approximately 3-5 pg)
of the template and incubation at 75°C for 5 min. The sample was
immediately transferred to ice and chilled for 5 min. To precipitate the
denatured DNA, 7 pi of water, 6 pi of 3 M sodium acetate, and 75 pi of 95%
ethanol were added to the sample, which was incubated at 4°C for an additional
5 min. The sample was centrifuged for 20 min, the supernatant was removed, and the DNA-containing pellet was dried under vacuum in a Savant Speed-Vac (Savant Instruments, Inc., Farmingdale, NY). The procedure suggested by U.S. Biochemical was then followed from this point with several modifications. Universal ml3/pUC primers (Appendix C; New England Biolabs) were
annealed to plasmid DNA templates for 30 min at 42°C. The reaction was
radiolabelled with [a-35S]-dCTP (Amersham Corp., Arlington Heights, IL)
during a 3 min extension step at room temperature. Termination reactions were shortened to 2.5 min to aid in reducing the number of compressions and were conducted at 48-50°C. Reaction mixtures were fractionated on 6% acrylamide [5.7 % acrylamide, 0.3% bis-acrylamide], 8 M urea gels that were equilibrated in modified TBE buffer [130 mM Tris-HCl, 45 mM boric acid, 2.5 mM Na2-EDTA] at 80 Watts. Following electrophoresis, the gels were soaked for 30 min in 1 0 % acetic add, 12% methanol to remove the urea. Gels were 52
then transferred onto Whatman 3MM chromatography paper, covered with
plastic wrap, and dried under vacuum for 45 min at 80°C on a Savant Gel
Dryer (Savant Instruments, Inc.). The plastic wrap was removed from the dried gel, which was then exposed to XR-100 X-ray film (BioWorld, Dublin, OH) at room temperature. The computer software program MacVector 4.1.4
(International Biotechnologies, Inc., New Haven, CT) was used for sequence analysis.
Transfer of Plasmid DNA into Pseudomonas Plasmid DNA was transferred into Pseudomonas by either one of two methods: conjugation or transformation. Plasmids with a broad-host-range
replicon and an origin of transfer (oriT ) were mobilized from E. coli DH5a into
Pseudomonas by triparental matings using the helper strain E. coli
HB101/pRK2013 which provided the transfer functions (Figurski and Helinski, 1979). Matings were performed on solid medium using either patch matings or
filter matings. For patch matings, a sterile toothpick was used to mix several colonies of each parent strain on an LB plate which was then incubated
overnight at 37°C. Cells were removed from the agar surface, resuspended in 2
ml PBS, and serially diluted prior to being plated onto PIA containing the appropriate antibiotic. Filter matings were performed using cultures of each parent that had been grown to mid-exponential (Agoo approximately1 .0 ) phase in LB broth without any antibiotics. A 0.5 ml aliquot of each parent was 53
filtered through a 25 mm MF-type HA 0.45 pm filter (Millipore, Marlborough, MA) by using a 5 ml syringe. Sterile forceps were used to transfer the filter to
an LB plate which was then incubated overnight at 37°C. The cells on the
filter were resuspended by vortexing in 2 ml PBS and were plated on PIA containing the appropriate antibiotic.
Plasmids lacking an oriT are not mobilizable, therefore, they were introduced into Pseudomonas using a modified version of the transformation method of Olsen et al. (1982; S. Wahl, personal communication). Competent
Pseudomonas cells were prepared from late-exponential cultures grown to an Aeoo of 1.0-1.7. Cells were pelleted by centrifugation at 5000 rpm for 5 min,
resuspended in 0.5 ml of 0.1 M CaCl2 for each ml of culture, and incubated on
ice for 5 min. Cells were collected by centrifugation, resuspended in 0 .1 ml of
0.1 M CaCl2 for each ml of culture, and incubated for 20 min on ice. An equal
volume of 0.1 M CaCl 2 and 50% glycerol was added to the cells, which were
frozen in a dry ice-ethanol bath and stored at -70°C. Approximately 2 pg of plasmid DNA was used for transformation of each 300 pi aliquot of competent
cells, which were then incubated on ice for 1 hr. Cells were heat-shocked for 3 min at 49°C, placed on ice for 5 min, and grown in 2 ml LB for 1.5-3.0 hr. Prior to plating, outgrowth cultures were concentrated by centrifugation and resuspension of the pellet in 400 pi PBS. The concentrated cultures were plated on PIA containing the appropriate antibiotic. 54
Generation of Nested Deletions for DNA Sequence Analysis Unidirectional nested deletions were generated from the 5' end of the
PAOl and HOD1 pilT alleles with exonudease III (Gibco BRL) according to the procedure described by HenikofF (1987). Restriction endonuclease sites with blunt or 5' -protruding ends following cleavage are the preferred substrates of this enzyme. Those endonuclease sites that have 3' protruding termini are protected from the exonuclease activity of exonuclease III. Thus, digestion of pPT2144 and pPT3734 (Appendix B) withP stl, which generates 3' overhanging ends, would protect the pK19 vector DNA from digestion by exonuclease III. Digestion with Xba I, which leaves a 5' overhang, would permit the unidirectional deletion of the p ilT inserts in these plasmids. Therefore, 10 pg of either pPT2144 or pPT3734 was digested withPstl and Xbal and then subjected to exonuclease III digestion by following the method of HenikofF (1987). Aliquots of 5 pi were removed from the reaction at 30 sec intervals for a duration of 6 min. In order to create blunt ends, the deleted samples were treated with SI nuclease and Klenow fragment (Gibco BRL). A
10 pi aliquot of each sample was analyzed by agarose gel electrophoresis to determine the extent of DNA deleted from each. The remainder of those samples that were appropriately deleted were religated and introduced intoE. coli DH5a. Plasmid minipreparations of the resulting kanamycin-resistant colonies were used for sequence determination. 55
DNA Hybridization Methods Southern hybridization was performed according to the method of Southern (1975) on DNA that had been digested previously with restriction endonucleases and separated by electrophoresis on 0.7% agarose gels. Prior to transfer, the gel was exposed to three solutions for 4 min each: 0.25 N HC1 to depurinate the DNA; 1.5 M NaCl, 0.5 M NaOH to denature the DNA; and 1 M Tris, 2 M NaCl (pH 5.0) to neutralize the DNA. The DNA was transferred by vacuum blotting from the agarose gel to a Nytran nylon membrane (Schleicher and Schuell, Keene, NH) using the Pharmacia VacuGene apparatus (Uppsala,
Sweden). Transfers were conducted by using a vacuum pump at 40 cm H 2O for 2 hr in 20X SSC [3 M NaCl, 0.3 M Na3-Citrate, pH 7.0]. Following transfer, the nylon membrane was baked at 80 °C in a vacuum oven (NAPCO Scientific,
Tualatin, OR) for 30 min to immobilize the DNA.
Prehybridization and hybridization buffers were prepared with IX SSPE
[10 mM sodium phosphate (pH 7.7), 0.18 M NaCl, 1 mM EDTA] and were added to the membranes at a volume of 0.1 ml per cm 2. Membranes were incubated in prehybridization buffer [ 6X SSPE, 5X Denhardt's, 0.5% SDS, 0.1 mg/ml denatured salmon sperm DNA; Sigma] for 2 hr at 65°C. The prehybridization buffer formulation was also used for hybridization of labelled
DNA probe to the membrane in an overnight incubation at 65°C. 56
The radiolabelled probe was prepared from 50 ng denatured plasmid
DNA or agarose gel-purified DNA fragment and 25 pCi [a-32P]-dCTP
(Amersham Corp.) using the Random Primed DNA Labeling Kit (U.S. Biochemical) according to the manufacturer's instructions. Following the labelling step, unincorporated nucleotides were removed by using a Sephadex- 50 spun column following the procedure described by Maniatis et al. (1982).
The amount of label incorporated into each probe was determined by measuring the cpm of 1 pi in a Beckman Model LS6800 Liquid Scintillation Counter (Beckman Instruments, Inc., Palo Alto, CA). Probes were denatured by boiling for 5 min and were then added to the hybridization buffer such that
the total number of cpm was 1 x 105.
Following hybridization, signals on the membranes were monitored with
a Ludlum survey meter (Measurements, Inc., Sweetwater, Texas) as the membranes were washed under increasingly stringent conditions. Wash conditions were as follows: (1) two 10 min washes at room temperature in 150
ml of 5X SSPE, 0.2% SDS; (2 ) two 10 min washes at 37°C in 150 ml of IX
SSPE, 0.5% SDS; (3) a 1 hr wash at 65°C in 150 ml of 0.1 X SSPE, 0.5% SDS.
Wet membranes were wrapped in plastic wrap and exposed to X-ray film at-
70°C in the presence of an intensifying screen (Lightning Plus; DuPont NEN,
Wilmington, DE). 57
Colony hybridizations (Grunstein and Hogness, 1975) were used to detect a clone of the HOD1 pilT allele from a subgenomic library in E. coli
DH5a. Genomic DNA that had been isolated from HOD1 (as described above) was digested overnight with Bam'Hl and then separated by agarose gel electrophoresis. The DNA fragments that were approximately 5.0 - 7.0 kb in size were purified from the agarose, ligated into the BarriHl site of pUC18 and transformed into E. coli DH5a. Approximately 600 of the resulting ampicillin- resistant colonies were patched onto master plates of LB agar containing ampicillin and onto sterile Nytran nylon filter disks (Schleicher and Schuell) that had been placed on the surface of LB agar containing ampicillin. Following overnight incubation at 37°C, cells were treated as recommended by
Schleicher and Schuell. Cells were lysed directly on the filters by placing the disks on Whatman 3MM chromatography paper that had been soaked in 0.5 N
NaOH. This 5 min step was performed in an aluminum pan that had been placed over a glass dish containing boiling water. Following lysis, the disks were neutralized on Whatman 3MM paper that had been soaked in 1 M Tris- HCl (pH 8.0) and then on Whatman 3MM paper soaked in 1 M Tris-HCl (pH
8.0), 1.5 M NaCl for 5 min at each step. To reduce background, the disks were then placed onto Whatman 3MM paper that had been saturated with 10%
SDS for 3 min. Membranes were washed in 2X SSC and baked for 30 min at
80°C to immobilize the DNA. Prehybridization was performed for 2 hr at 65 °C in 4X SSPE, 5X Denhardt’s reagent, 1% SDS, and 0.2 mg/ml of sheared, denatured salmon sperm DNA. Hybridization was performed overnight at 58
65°C in the same buffer used for prehybridization. The probe was prepared by labelling pPT2154 (Appendix B) following the procedure provided above for the preparation of probes for Southern hybridization. Membranes were washed twice for 15 min at room temperature in 2 X SSPE, 0.2% SDS. Two additional washes of 15 min each were performed at room temperature in IX SSPE, 0.1% SDS. Membranes were wrapped in plastic wrap and exposed to X-ray film at
-70°C.
Standard Methods for Protein Analysis Protein quantitation was routinely carried out according to the method of Bradford (1976) by using bovine serum albumin as the protein standard.
Radiolabelled proteins generated in T7 expression studies (described below) were fractionated using sodium dodecylsulfate - polyacrylamide gel electrophoresis (SDS-PAGE) following the procedure of Dreyfuss et al. (1984).
Following electrophoresis, the radioactive gel was treated by fixing in 7% acetic acid for 30 min, washing in water for 30 min, and soaking in the fluor sodium salicylate (1 M) for 30 min. The gel was then laid on 3MM Whatman chromatography paper, covered with plastic wrap, and dried on a Savant Gel
Dryer (Savant Instruments, Inc.) for 1 hr at 80 °C. Dried gels were exposed at -
70 °C to XR 100 X-ray film in cassettes containing intensifying screens
(Dupont NEN). 59
For Western analysis, a semi-dry electroblotter (Trans-Blot SD; Bio-Rad Laboratories, Richmond, CA) was used to transfer proteins from SDS-PAGE gels to nitrocellulose membranes (Schleicher and Schuell) according to the manufacturer’s suggested procedure. Gels were transferred for 30 min at the recommended current of 3 mA/cm2 for large gels and 5.5 mA/cm 2 for minigels in transfer buffer that was prepared according to the formulation described by Towbin et al. (1979) [25 mM Tris, 192 mM glycine, 20% methanol (pH 8.3)].
Nitrocellulose membranes were blocked for 45 min at room temperature in 2 % skim milk, IX TBS (10 mM Tris, 0.5 mM NaCl, pH 7.5). Anti-gonococcus PilT polyclonal antiserum (a gift from M. Koomey, the University of Michigan, Ann Arbor, Michigan) was diluted 1:1000 in 100 ml of 1% skim milk, IX TBS and incubated overnight with the membranes at 4°C. The primary antiserum was rinsed from the blots with three 10 min washes at room temperature in 0 .1%
Tween-20 (Sigma), IX TBS with continual rotation (Scientific Products Variable Speed Rotator, McGaw Park, IL). Blots were incubated for 3 hr at room temperature with alkaline phosphatase-conjugated, goat anti-rabbit IgG (Bio-Rad Laboratories) diluted 1:3000 in 1% skim milk, IX TBS and then washed as described above. Membranes were developed using the Bio-Rad alkaline phosphatase color development reagents BCIP (5-bromo-4-chloro-3- indoyl phosphate p-toluidine salt) and NBT (p-nitro blue tetrazolium chloride), which produce an insoluble purple product. 60
Overexpression ofpilT in E. coli and Pseudomonas using T7 Promoter- Based Systems Specific details for the construction of plasmids used for T7 promoter- directed expression ofpilT in E. coli and P. aeruginosa are provided in Chapter III. The E. coli T7 expression strain BL21 (DE3) (Studier and Moffatt, 1986)
was transformed with each of the expression plasmids, pADD1408, pADD1409, pADD1410, and pADD1411. Bacterial cultures were induced according to the procedure described by Tabor and Richardson (1985).
Overnight cultures of these E. coli strains were grown at 37 °C with shaking in
2 ml LB containing ampicillin. Two dilutions were prepared for each starter culture, with each dilution consisting of 0.4 ml of the overnight culture diluted
in 6 ml of LB. Cultures were shaken at 37°C until the Aeoo of each reached 0.5,
at which time two 2 ml aliquots were removed from each culture and placed into new tubes. Isopropyl B-D-thiogalactopyranoside (IPTG; Sigma) was added
to one tube in each pair at a final concentration of 1 mM, and all cultures were
reincubated for 1 hr at 37°C. Rifampicin (50 mg/ml in N ', N'-dimethyl-
formamide; Sigma) was added to all tubes to a final concentration of 200 pg/ml
and cultures were incubated at 37 °C for another 30 min. The proteins were
radiolabelled by the addition of 2 pi of Tran 35S-label [ 35 S-methionine (70%) and 35S-cysteine (20%) at 8-12 mCi/ml; ICN Biomedicals, Inc, Costa Mesa, CA] to
each culture and incubation with shaking for 30 min at 37°C. Cells were harvested by centrifugation in a microcentrifuge, washed with PBS and resuspended in 50 pi of Laemmli lysis buffer (1970). 61
The procedure for overexpression of pilT in Pseudomonas aeruginosa was similar to that given above for E. coli but with the following modifications.
The pilT expression plasmid pPT2296 was introduced into the T7 expression strain ADD1976 (Brunschwig and Darzins, 1992) by a triparental mating. The labelling step was shortened to 2 0 min, and harvested cells were resuspended in 25 pi PBS prior to the addition of 25 pi of 2X Laemmli (1970) lysis buffer.
RNA Isolation and Manipulation Cultures used for the isolation of RNA were grown overnight in LB broth at 37°C and then diluted 1/100 in LB broth. The diluted culture was grown at 37°C with shaking and 50 ml samples were removed for RNA isolation. Agar- grown cells were cultured on 75 mm cellophane disks (Carriage Avonmouth, Bristol, England) that had been placed on the surface of an LB agar plate. The cellophane disks had been previously treated as described by Maniatis et al. (1982) for the preparation of dialysis membrane except that EDTA was eliminated from the treatment. Disks were layered individually between moist pieces of Whatman 3MM filter, autoclaved in a covered glass dish, and stored at 4°C. The following day, each disk was aseptically transferred to the surface of an LB agar plate. One hundred fifty microliters of the diluted starter culture was inoculated onto the surface of the disk and the culture was incubated at
37°C. One culture was sacrificed for each Aeoo reading and generally eight to ten plates were used for the isolation of each RNA sample. Cells were collected 62
from the plates by vortexing the cellophane disks in a constant volume of LB broth. These cell suspensions were used for determining the Aeoo and for
isolation of RNA.
All glassware used for RNA isolation and manipulation was baked
overnight at 185 °C to remove RNases. Cells were collected by centrifugation
in a Beckman JA-20 rotor (Beckman Instruments, Inc.) at 6000 rpm for 10
min at 4°C. Cell pellets were washed with 5 ml of ice cold lysis buffer (50 mM
Tris-Hcl, pH 7.0) and suspended in another 5 ml of room temperature lysis buffer. To lyse the cells, 1 ml of 20% SDS was added to the samples which
were then incubated at 56°C for 5 min in polypropylene culture tubes (United
Laboratory Plastics, Fenton, MO). Since SDS inhibits RNases, no diethyl pyrocarbonate (DEPC) was added to any of the solutions used for the isolation of RNA. Solid cesium chloride (4 g) was dissolved in each sample and an
additional 5 ml of lysis buffer was added. Samples were then centrifuged in a Beckman JA-20 rotor for 10 min at 11,000 rpm to remove the cellular debris. The clear supernatants were then layered gently onto 2.5 ml of 5.7 M CsCl in
open-top, thin-wall, PPCO 14 x 89 mm Nalgene ultracentrifuge tubes (Nalge Company, Rochester, NY). When supernatants were too viscous to pipet due to the presence of large quantities of chromosomal DNA, they were sheared through an 18-gauge hypodermic needle and then recentrifuged as above. Samples were centrifuged for 12-20h in a Beckman SW 41 rotor at 35,000 rpm at 15°C. The solutions containing DNA, protein, and cesium chloride were then 63
carefully removed from the ultracentrifuge tubes. The RNA pellets were resuspended in 300 jxl of water, extracted with chloroform-isoamyl alcohol (24:1), and precipitated by adding 1/10 volume of 3 M sodium acetate and 2.5 volumes of ethanol. Samples were stored as ethanol precipitates for several
months at -20°C. To collect the RNA, samples were centrifuged for 12 min,
supernatants were removed, and RNA pellets were resuspended in water.
These samples were stored at 4°C for no longer than one month.
All solutions used for RNA electrophoresis, transfer and hybridization for Northern analysis were treated with 0.2 % diethyl pyrocarbonate (Sigma)
and autoclaved to inactivate RNases. RNA samples were fractionated on 1.1% agarose gels containing 0.66 M formaldehyde (Ausubel et al., 1987) at
150V in 0 .2 M MOPS (3-[N-morpholino]propane-sulfonic acid; Sigma), 10 mM EDTA, and 50 mM sodium acetate. RNA was transferred by vacuum blotting
to Nytran nylon membrane (Schleicher & Schuell) by using 10X SSC as the
transfer buffer. Following transfer, blots were baked at 80°C for 30 min in an
oven. Prehybridizations and hybridizations were carried out at 65 °C as
described in the Schleicher & Schuell manual with the following modifications. Prehybridization solutions consisted of 5X Denhardt’s reagent, 5X SSC, 0.1% SDS, and 50 pg/ml denatured salmon sperm DNA. This same formulation was used for the hybridization solution except that the concentration of Denhardt’s reagent was lowered to 2X. A 1.5 kb Pstl - Kpnl fragment from pPT3910
(Appendix B) that contains the entirep ilT gene served as the probe for 64
Northern hybridizations. Probe fragments were labeled with [a- 32P]-dCTP
(Amersham Corp.) using the Random Primed DNA Labeling Kit (U.S. Biochemical). Blots were exposed to XR 100 X-ray film at -70°C in the presence of an intensifying screen.
Mapping of the 5' end of thepilT transcript was performed with SI nuclease (Gibco BRL) as described by Maniatis et al. (1982) with the following
specifications. Each SI reaction contained approximately1 x 105 cpm of probe (prepared as described below) and with 50-200 [ig of RNA. RNA and probe mixtures were lyophilized under vacuum in a Savant SpeedVac Concentrator (Savant Instruments, Inc.) and then resuspended in 30 pi of SI hybridization buffer [40 mM PIPES (pH 6.4), 1 mM EDTA (pH 8.0), 0.4 M
NaCl, 80% formamide]. Samples were denatured at 85-90°C for 10 min and
then allowed to cool slowly to 55°C over a 2 hr period in a heating block (VWR
Scientific, Boston, MA). Following an overnight hybridization at 55°C, 294 |il of
SI reaction buffer prepared according to the Gibco-BRL formulation [50 mM
NaCL, 30 mM Na-acetate (pH 4.6), 1 mM Zn-acetate, 5% glycerol] was added to each sample. In addition, 3 pi of denatured salmon sperm DNA (1 mg/ml) and 2 pi of diluted (1:10) SI nuclease (Gibco BRL) were added (80-300 units/reaction), and samples were incubated at 37°C for 30 min. Following extraction with phenol-chloroform and ethanol precipitation, samples were resuspended in 6 pi of water and 5 pi of loading buffer (95% formamide, 20 mM EDTA, 0.05% bromophenol blue, 0.05% xylene cyanol FF). 65
The DNA sequence ladders used in the SI nuclease studies were generated using the standard sequencing methods described earlier except that
they were labeled with [oc- 32P]-dATP (Amersham Corp.) that had decayed
approximately 3-4 half-lives. The use of extensively decayed label yielded a sequencing ladder with an intensity that was similar to the weak signals of the protected fragments generated in the SI nuclease reactions. Additionally, sharper bands were obtained with the sequence ladder following electrophoresis and autoradiography. Plasmid pPT3244 (Appendix B) was used as the template DNA and either oligonucleotide 85 or 146 (described below) was used in conjunction with a universal primer. SI nuclease and DNA sequencing reactions were heated to 90°C for 5 min prior to being fractionated on 5% acrylamide, 8 M urea gels at 80 Watts in TBE running buffer. Wet gels were transferred to Whatman 3MM paper, wrapped in plastic wrap and exposed to X-ray film.
Low-resolution SI nuclease protection assays were carried out as above with the following modifications. A molecular weight ladder was generated from
Saw3AI-digested pUC18 DNA that was end-labelled with [y-32P]-ATP
(Amersham) in the same manner as the PCR probe fragments were labelled (described below). Gels were analyzed using Instantlmager 2024 Electronic
Autoradiography (Packard Instrument Co., Inc., Meriden, CT) and were then exposed to X-ray film. 6 6
PCR Conditions for Generation of SI Nuclease Probes The polymerase chain reaction (PCR) was used to generate DNA fragments that were to be labelled for use as probes for SI nuclease protection studies. Two oligonucleotides (Ransom Hill Bioscience, Inc., Ramona, CA) were chosen for probe synthesis: (1) 5 ' -CACTGGAACAGGAAGATGGC-3' which was designated oligonucleotide 85; and (2) 5' -GATGAGCCAGGGTGCTTCC-3' which was designated oligonucleotide 146. Each of these oligonucleotides was used in conjunction with the M13 universal forward 17-mer primer (Ransom
Hill Bioscience, Inc.) to synthesize DNA probe fragments by using Deep Vent polymerase (New England Biolabs). PCR reactions were conducted in a volume of 100 pi and contained 50 pmol of each primer, 40 ng of the plasmid template DNA pPT3244 (Appendix B), 100 pM each of dATP, dCTP, dGTP, dTTP, and 2 U of Deep Vent polymerase in the buffer provided with the enzyme [10 mM KC1, 20 mM Tris-HCl (pH 8 .8 at 25°C), 10 mM (NH^SO^ 2 mM MgS0 4 , 0.1% Triton X-100]. The reaction mixtures were overlayed with
20 pi of mineral oil (U.S. Biochemical) to prevent evaporation. An initial 5 min incubation at 96°C denatured the template and was followed by 30 cycles of amplification under the following conditions: 1 min denaturation at 96°C, 1 min annealing at 60°C, and extension for 20 sec at 72°C. An additional 5 min extension cycle at 72°C followed the 30 amplification cycles to allow for complete polymerization. Electrophoresis on a1 .2 % agarose gel was used to analyze the reactions for the appropriately sized product. A 324 bp fragment was produced when oligonucleotide 85 was used as a primer, and a 402 bp 67
product was produced when oligonucleotide 146 was used in the PCR reaction. The desired fragment was purified from the agarose gel and end-labelled by using T4 polynucleotide kinase (Gibco BRL). Labelling reactions included 10 pmol of 5' ends of PCR-generated DNA, 10 U of T4 polynucleotide kinase
(Gibco BRL) and 50 pCi of [y - 32P]-ATP (Amersham) in IX forward reaction
buffer [100 mM KC1, 70 mM Tris-HCl (pH 7.6), 10 mM MgCl2, 1 mM 2 -
mercaptoethanol]. The phosphorylation step was performed at 37°C for 15
min and then stopped with the addition of EDTA to a final concentration of 5 mM. Water was added to the reactions to increase the volume to 80 pi, and the reactions were then centrifuged in Sephadex-50 spun-columns as described by Maniatis et al. (1982) to remove the unincorporated nucleotides. A 1 pi aliquot of each probe was counted in a liquid scintillation counter (Beckman Model LS6800) to determine the incorporation of label.
Enzyme Assays
Assays for general protease activity and elastolytic activity were performed using LB agar with agar overlays containing skim milk and elastin- congo red (ECR), respectively. Skim milk (Difco Co.) was added to sterile, molten LB agar at a concentration of 3.33% (wt/vol), and 5 ml of this suspension was overlaid onto each LB agar plate. ECR (Sigma) overlay plates were prepared in the same manner, with the ECR being added to the molten LB at a concentration of 2.5 mg/ml. Strains to be assayed were patched onto 68 these plates and incubated at 37°C for 24 hr, at which time they were observed for zones of clearing around each patch. The diameters of the zones of clearing were measured and cultures were reincubated for an additional 24 hr. Following this incubation, the zones were again measured, and all values were compared to those obtained from the wild type strain, PAOl.
Relative luminescence of E. coli and P. aeruginosa strains harboring luxAB transcriptional fusion plasmids was determined by an in vivo luciferase assay adapted from the method of Peabody et al. (1989). Growth conditions for the in vivo luminescence assay on solid medium were similar to those used for the twitching assay to ensure that conditions were favorable for expression from the p ilT promoter. Overnight cultures of the transformants were inoculated by stabbing onto 1% LB agar plates, which were then incubated at
37°C. At various intervals throughout exponential growth, the plates were removed from the incubator for luciferase assays which were performed in a dark room. The aldehyde substrate n-decanal (Sigma), which is highly volatile and can be absorbed by the cells on the agar, was swabbed onto the inside of the petri dish lid. After a 5 min incubation at 37°C, plates were exposed to X- ray film for 4 min. Relative luminescence was determined by comparing the diameters of the circular zones on the X-ray film that resulted from bacterial luminescence. Liquid cultures were visually screened for luminescence following the addition of 10 pi of a sonicated suspension of0 .1% n-decanal. 69
P. aeruginosa strains containing lacZ transcriptional fusion plasmids were assayed for B-galactosidase activity by the method of Miller (1972) with the following modifications. Cells to be assayed were lysed by the addition of 20 pi of 0.1% SDS and 40 pi chloroform followed by a 5 min incubation at room temperature. Prior to determination of the A420 , each reaction was centrifuged for 10 min in a microcentrifuge to remove cellular debris, thus eliminating the need to obtain the A 550 to correct for light scattering. The molar extinction coefficient for o-nitrophenol under the given assay conditions is 4500 M-i. Specific activity refers to nmol o-nitrophenol produced per min per amount of cells.
Specific Activity = 1000 x ------ft4.?-0 x 1---- min*mlxA6oo 4.5 (pmol/ml)
Generation of Random Mutations in pilT
Epicurian Coli XLl-Red, a mutator strain of E. coli, was purchased from Stratagene (La Jolla, CA) for random mutagenesis of pilT. This strain is deficient in three DNA repair pathways: 3 '- to 5'- exonuclease activity of
DNA polymerase III (mutD\ Scheuermann etal., 1983); error-prone mismatch repair (m utS; Radman et al., 1980); and hydrolysis of 8 -oxo-dGTP (mutT\ Cox,
1976). Expected mutations using this strain include base substitutions and frameshifts. Plasmid pPT3736 (Appendix B), which contained pilT in the shuttle vector pUCP18, was introduced into Epicurian coli XLl-Red using the following procedure for transformation. A 1:10 dilution of fi-mercaptoethanol 70
(14.2 M stock; Sigma) was prepared and 1.7 pi was added to a 100 pi aliquot of
t competent E. coli cells (final concentration 25 mM) in a 15 ml polypropylene
tube. The cells, which were gently swirled every 2 min, were incubated on ice for 10 min. Following the addition of 50 ng plasmid DNA, the cells were incubated on ice for an additional 30 min. The cells were heat shocked for 45 sec in a 42 °C water bath and then incubated on ice for 2 min. LB (0.9 ml) was added to the cells and the polypropylene tube was incubated at 37°C for 1 hr with shaking at 300 rpm. Cells were plated on LB agar containing ampicillin and incubated overnight at 37°C. Colonies (>200) were scraped from the agar surface of the plates, inoculated in 10 ml LB containing ampicillin (100 pg/ml), and incubated at 37°C for 12 hr with shaking. A 0.1 ml aliquot of this culture, designated passage one, was used to subculture another 1 0 ml culture, designated passage two. The remainder of the passage one culture was used for plasmid isolation, and these steps were repeated for seven passages.
A control, designed to mutate the lacZ gene of pUC18, was performed to determine the mutation frequency of Epicurean coli XLl-Red. On the basis of the results obtained with the control experiment, the pPT3736 target DNA was recovered from the mutator strain following four or five passages. These minipreparations were introduced into the p ilT mutant HOD1 and transformants were selected on PIA agar containing carbenidllin. Colonies were screened for altered twitching motility using the motility stab assay described above. Colonies that could not twitch, or formed a motility zone different from that produced by PAOl, were chosen for further analysis. CHAPTER m
PILT IS REQUIRED FOR PHAGE SENSITIVITY AND TWITCHING MOTILITY IN P. AERUGINOSA PAOl
Introduction
At the beginning of this study the pil genes that had already been identified in P. aeruginosa were demonstrated to be necessary either for the production of the pilin subunit (Johnson et al., 1986; Pasloske et al., 1985) or for the assembly of the pilus (Nunn et al., 1990). Several hyperpiliated strains harboring nonretractile pili had been identified (Bradley, 1974,1980a; Johnson and Lory, 1987), suggesting that these strains carried a mutation in one or more genes that were required for pilus retraction; however, the mutations occurring in these hyperpiliated strains were never characterized. Additionally, no genes in any other type IV-piliated bacterium had yet been demonstrated to be involved in pilus retraction.
The inability of nonretractile pili to mediate either phage sensitivity or twitching motility in P. aeruginosa provides an easily identifiable phenotype that aids in the isolation of strains that are mutated in genes encoding
71 72 proteins required for pilus retraction. However, since nonpiliated strains are also rendered phage-resistant and non-twitching, mutants must be screened for the extent of piliation in order to differentiate those which are hyperpiliated from those lacking pili altogether.
This project focused on the identification and characterization of the pilT gene which is essential for phage sensitivity and twitching motility, thus providing indirect evidence thatpilT is required for pilus retraction. The goals of the initial portion of this study were to:
(1) identify mutants of PAOl that were defective in their ability to assemble functional pili by isolating spontaneous phage-resistant mutants which were unable to twitch,
(2) screen these mutants to identify those which were hyperpiliated,
(3) characterize the mutations in these hyperpiliated strains to
identify the gene or genes required for pilus retraction in PAOl, and
(4) identify the produces) of the gene(s) involved in retraction. 73
Results Isolation of phage-resistant mutants of P. aeruginosa PAOl defective in pilus biogenesis Following infection of PAOl with excess D3112 cts, B3 cts-3, or F116L cts-53, phage-resistant colonies appeared at a frequency of approximately 1 x 10-5 per cell. Thirty-seven of these isolates were chosen randomly from a total of nine separate experiments for further analysis. Fifteen of the thirty-seven mutants were isolated on the basis of B3 cts-3 resistance, nineteen by D3112 cts resistance, and three by F116L cts-53 resistance. Subsequent phage sensitivity assays of these mutants revealed that all 37 mutants were resistant to all three bacteriophages. Moreover, none of the mutants was capable of twitching motility; these strains formed small compact colonies in contrast to the large, spreading colonies produced by PAOl (Fig. 5). This difference in colony morphology was readily visualized without magnification when cells were grown on LB agar.
Electron microscopy was used to determine the degree of piliation of each phage-resistant mutant. Shadowcasts of the mutants (data not shown) revealed that all but four of the 37 mutants were nonpiliated. The four remaining mutants were clearly hyperpiliated with respect to the parent strain, PAOl, and their pili appeared to be longer. The appearance of these hyperpiliated mutants was consistent with that of previously reported hyperpiliated mutants that had been viewed using electron microscopy (Bradley, 1972c, 1974). Furthermore, immunoblot studies using anti-pilin 74
Figure 5. Colony morphology of P. aeruginosa PAOl and the p ilT mutant, HOD1, on LB agar. The wild type strain PAOl (P) is capable of twitching motility and forms flat, spreading colonies with irregular edges. HOD1 is hyperpiliated with non-retractile pili that render the mutant incapable of twitching motility. As a result, HOD1 (H) forms smaller, dome-shaped colonies with smooth edges. Total magnification is 50 X. 75 antiserum, performed by others in this laboratory, confirmed that the extracellular pilin level for HOD1 was greater than that seen with PAOl (A.
Darzins, unpublished data; M. Russell, unpublished data). In this laboratory, pili are routinely isolated from cells according to the procedure of Roncero et al. (1990). In this procedure, pili are sheared from cells by vortexing cultures which have been adjusted to an A 600 of 5.0. These pili preparations are then concentrated and used for immunoblot analysis.
Of the four hyperpiliated mutants isolated in this study, three had been isolated by infection with D3112 cts and were designated HOD1, HOD3, and
HODIO. The fourth mutant had been isolated by selecting for resistance to B3 cts-3 infection and was designated HOB6. To characterize the nature of the mutations in these four mutants, each was examined for a loss of prototrophy and a decrease in extracellular protease activity. Because all four mutants grew on Clarke glucose minimal medium, they were classified as prototrophs. When compared to PAOl, none of the mutants displayed decreases in elastolytic or general protease activity when grown on elastin-congo red or skim milk medium, respectively; thus, the mutation(s) present did not affect general extracellular protease secretion. Lysates of D3112 cts, B3 cts-3, and F116L cts-53 that had been preadsorbed to the four mutants were titered using PAOl according to the method of Roncero et al. (1990). These titers (data not shown) were comparable to that obtained following preadsorption to
PAOl, indicating that the hyperpiliated mutants adsorbed the phages as well as PAOl. 76
Complementation of the hyperpiliated mutants Cloning of the pil locus capable of complementing the hyperpiliated mutants was facilitated by mobilizing a JHmdlll-generated cosmid library of PAOl chromosomal DNA into HOD1, HOD3, and HOB6. The PAOl library, which had been constructed earlier by Darzins (unpublished data; using the method of Darzins and Chakrabarty; 1984), was mobilized from E. coli S17-1 into each mutant recipient using filter matings. Tetracycline-resistant transconjugants were visually screened for the spreading colony morphology which represented the restoration of the ability to twitch. Ten HOD1, one HOD3, and two HOB6 twitching transconjugants were identified on this basis.
The thirteen twitching colonies were assayed for restored sensitivity to phage D3112 cts and all were phage-sensitive. Minipreparations of the recombinant cosmids were isolated from six of these transconjugants, representing all three m utant loci. All six cosmid constructions harbored a 22 kb H indlll insert of PAOl chromosomal DNA. Subsequent digestions with various restriction endonucleases revealed identical restriction patterns for these six constructions. Three of these cosmids, designated as HOD1.1+,
HOD3.1+ and HOB6.1+, were introduced into E. coli DH5afor subsequent mating with the original phage-resistant mutants. Each of the three plasmids restored phage sensitivity to all three mutants. HOD 10, the fourth hyperpiliated mutant that was isolated, showed twitching and phage- sensitivity upon introduction of HOD 1.1+, HOD3+, or HOB6+. Thus, the 22 kb 77
H indlll fragment from PAOl was capable of restoring both twitching motility and phage sensitivity to each of the four hyperpiliated mutants. To simplify my experiments, I chose to use only HOD1 for further analysis.
The 22 kb H indlll fragment was subcloned to a smaller fragment that was still capable of complementing HOD1. To do this I digested the plasmid with BamHl because this enzyme recognized a site in the vector that was adjacent to the insert and because it was likely that there were multiple Bam H l sites in the 22 kb fragment. This step was designed to delete the majority of the insert DNA so that only one BamHl fragment, which may contain the gene(s) capable of complementing HOD1, would remain in the vector. I added T4 DNA ligase to the entire digest reaction and introduced the ligation mixture into E. coli DH5a. Minipreparations of recombinant DNA from twelve E. coli transformant colonies were digested with BamHl and the resulting restriction patterns examined by gel electrophoresis. One construction contained a 6.0 kb BamHl fragment. When this construction was introduced into HOD1, twitching motility and phage sensitivity were restored, suggesting that the HOD 1-complementing locus resided on this 6.0 kb
BamHl fragment. Additional subcloning and phenotypic complementation steps, which are summarized in Fig. 6, resulted in the identification of a 1.5 kb Pstl-Kpnl fragment that retained the ability to complement HOD1 as well as Figure 6. Complementation of the hyperpiliated mutant HOD1. The 6.0 kb BamHl fragment was subcloned from the 22 kb H indlll chromosomal library of PAOl genomic DNA that had complemented HOD1. The 6.0 kb BamHl fragment was subcloned to a 1.5 kb Pstl-Kpnl fragment that retained the ability to restore phage sensitivity and twitching motility to HOD1. Abbreviations for restriction endonuclease sites: B, BamHl; K, Kpnl\ S, SaZI; P, Psf I; Bg, Bglll; X, .X7toI.
78 Plasmid Subcloned Fragment of PAOl DNA Complementation
B Bg XPXSK B pPT1916 i I I I i I i | +
PPT1920 I------1 PPT1921 ______I I I______I pPT1999 ______I | +
PPT2001______I______I +
1 kb
Figure 6 80
HOD3, HOD 10 and HOB6. Consequently, all four hyperpiliated mutants probably contained mutations in the pil locus residing on this'1.5 kb Pstl-Kpnl fragment.
Copjugational mapping of the HOD1 phage-resistance locus
The P. aeruginosa pil genes that had been identified prior to this study had been mapped to several different regions on the PAOl chromosome (Farinha et al., 1993; Nunn et al., 1990). To determine whether the HOD1 pil mutation mapped to one of these previously identified regions or to another, yet unidentified, pil locus on the chromosome, FP2-mediated conjugational mapping was conducted. The HOD1 mutation was 100% linked with the met and trp markers, which are located at 20-23 minutes on the chromosome, and 97% linked with the lys and met pair of markers located between 10-20 min (Table 6). Further mapping experiments that selected individually for the met and trp markers revealed that the HOD1 mutation was tightly linked to the met-28 allele at approximately 20 min on the PAOl chromosome. The location of the HOD1 locus is therefore unlinked to the pilin subunit gene ipilA) and the accessory genes (pilB, C ,D) required for pilus biogenesis that had already been identified and mapped to the 75 min region of the chromosome (Holloway et al., 1994). Table 6. Mapping of the HOD1 phage-resistance mutation.
M arker Donor* R ecipient Coinheritance M arkers Selected Location^ Frequency0 HOD1/FP2 PAO1042 pro+ cys+ pur+ 66-71 0% (0/76) HOD1/FP2 PA0222 ilv+ his+ lys+ met+ trp+ 0-23 98% (58/59) HOD1/FP2 PA0222 ilv+ his+ lys+ met+ 0-20 92% (72/78) HOD1/FP2 PA0222 ilv+ his+ 0-7 7% (6/85) HOD1/FP2 PA0222 his+ lys+ 7-10 10% (10/101) HOD1/FP2 PA0222 lys+ met+ 10-20 97% (60/62) HOD1/FP2 PA0222 met+ trp+ 20-23 100% (69/69) HOD1/FP2 PA0222 met+ 20 86% (25/29) HOD1/FP2 PA0222 trp+ 23 62% (31/50)
« FP2, chromosome mobilization ability (Tra+> b Marker locations correspond to minutes on the PAOl chromosome (Holloway et al., 1994) c Ratios provided in parentheses indicate the number of phage-resistant transconjugants per number of transconjugants assayed. 82
Overexpression ofpilT in E. coli and Pseudomonas using T7 promoter- based systems To identify the gene product(s) encoded by the 1.5 kb Pstl-Kpnl fragment, expression studies utilizing the phage T7 promoter were conducted in E. coli and P. aeruginosa. The ends of the 1.5 kb Pstl-Kpnl fragment were made blunt with the Klenow fragment of DNA polymerase I and ligated into the S m a l site of the T7 expression vector pADD621 (Appendix A) in both orientations with respect to the T7 promoter (forming pADD1408 and pADD1409; Fig. 7). The two plasmids were introduced into E. coli BL21 for induction, labelling, and SDS-PAGE analysis. A unique insert-specific protein of approximately 37 kDa was detected in the cell lysate of BL21/pADD1409 but not in that of BL21/pADD1408 (data not shown). Thus, the 1.5 kb Pstl-
Kpnl fragment was apparently transcribed in the direction from Pstl to Kpnl and encoded one protein of Mr approximately 37 kDa.
To confirm the above conclusion, two additional plasmids were constructed by deleting portions of the Pstl-Kpnl fragment. The plasmid pADD1410 contained the 1.1 kb Pstl-Sall fragment ligated in the Pstl and Sail sites of pADD621; therefore, the 0.39 kb Sall-Kpnl region was absent from the proposed 3' end of the insert (Fig. 7). In a similar manner, pADD1411 contained the 0.9 kb Xhol-Kpnl fragment (Kpnl blunted) in the Sall-Sacl sites (SacI blunted) and thus lacked the 0.6 kb Pstl-Xhol fragment believed to contain the 5' end of the coding sequence. The 37 kDa protein was detected when pADD1409 had been induced with IPTG (Fig. 8). When the 0.6 kb Pstl- Figure 7. Plasmids used for the expression studies of the HOD1- complementing locus in E. coli BL21. The location of the open reading frame which was later identified as pilT is represented by the thicker black arrow.
The direction of transcription initiating at the T7 promoter (Pt7) is indicated for each plasmid by the smaller arrow. Pertinent restriction endonuclease sites are also shown.
83 P la s m id
P stl Kpnl pADD1408 ■— ^ " PT7 Pstl Kpnl pADD1409
Pstl pADD1410 I—
Kpnl pADD1411
00 Figure 8. Expression of the HOD 1-complementing locus (pilT) in E. coli BL21. An autoradiograph of 35S labelled proteins separated by SDS-PAGE is shown. Lanes: 1, the expression vector pADD621; 2, pADD621 following addition of
IPTG; 3, pADD1409; 4, pADD1409 following addition of IPTG; 5, pADD1411; 6, pADD1411 following addition of EPTG; 7, pADD1410; 8, pADD1410 following addition of IPTG; 9, molecular mass markers (sizes, in kilodaltons, are indicated at right). Arrows at the right denote the 37 kDa protein (PilT) and its 27 kDa truncated form.
85 8 9
- 2 0 0
- 9 7 -6 8
-4 3 37
- 2 9 ■27
18
■ 14
00 Figure 8 05 87
Xhol fragment was deleted from the insert (pADD1411), no unique protein band could be detected. A 27 kDa protein was detected when the 0.39 kb Sall-
Kpnl was deleted from the insert (pADD1410). Expression of thePstl-Kpnl fragment therefore occurred in the direction from the Pstl site to the Kpnl site.
To determine if the putative gene residing on the 1.5 kb Pstl-Kpnl fragment encoded a 37 kDa protein when expressed in P. aeruginosa , the 1.5 kb fragment was removed from pPT2144 (Appendix B) by digestion withH indlll and Sm al and ligated into the similarly digested pEB12 (Appendix A). The fragment was oriented such that transcription driven from the T7 promoter would progress in the direction from Pstl to Kpnl. This plasmid, designated pPT2296, was mobilized into Pseudomonas aeruginosa ADD1976, a derivative of PAOl which contains the T7 RNA polymerase gene in the chromosome. Following induction with IPTG and .radiolabelling, a whole-cell lysate was analyzed by SDS-PAGE (Fig. 9). Lysates of uninduced ADD1976/pPT2296 and induced ADD1976/pEB12 were included as controls. The 37 kDa protein was detected in the lysates of both uninduced and induced ADD1976/pPT2296 and was not detected in the vector control. The amount of the 37 kDa protein present in the sample from the induced culture did not, however, appear to be greater than that seen in the uninduced culture. This lack of induction is likely due to a problem inherent to pEB12, since some other researchers have also been unable to see induction while using this plasmid (A. Darzins, personal communication; L. O’Donnell, personal communication). Figure 9. Expression of the HOD 1-complementing locus (pilT) in the P. aeruginosa T7-expression strain ADD 1976. The radiolabelled proteins that
were separated by SDS-PAGE are shown in this autoradiograph. Lanes: 1, pEB12 following addition of IPTG; 2, pPT2296; 3, pPT2296 following addition of IPTG. Molecular mass markers are included (MW), and their respective sizes
(in kDa) are indicated at the left. A 37 kDa polypeptide (PilT) is indicated by the arrow at the right.
88 Figure 9 90 Nucleotide Sequence of the pilT open reading frame
The nucleotide sequence of the 1.5 kb P stl-K p nl fragment was determined using nested deletion constructions and ml3/pUC universal primers (Fig. 10). The nucleotide sequence of this locus was simultaneously determined by Whitchurch et al. (1990). A putative open reading frame that encodes a 37 kDa protein was identified by Whitchurch et al. (1990) and was designated pilT because it was required for twitching motility.
Disruption of pilT in PAOl
A disruption of the pilT chromosomal locus was constructed to confirm that the mutant phenotype seen with HOD1 was due to a mutation that was in some way affecting pilT expression or PilT function. The 0.3 kb EcoKl fragment from the internal region of the pilT gene was ligated into the EcoRl site of the suicide vector, pBR322, to form pPT2381. Plasmid pPT2381 was introduced into E.coli DH5a and then introduced into PAOl by a triparental filter mating. Transconjugant colonies displayed resistance to both tetracycline and carbenicillin, indicating that pPT2381 had integrated into the chromosomal pilT locus by a single recombination event (Fig. 11). However, Southern analysis was not used to confirm this integration. Twitching motility and phage-sensitivity assays revealed that the null mutant, designated PAO- PT, exhibited the same non-twitching, phage-resistant phenotype as HOD1. Figure 10. The nucleotide sequence of the 1.5 kb Pstl-Kpnl fragment containing pilT. The sequence identified in this study is identical to that identified by Whitchurch et al. (1990). The predicted amino acid sequence of PilT and the putative ribosome binding site (boxed) which were identified by Whitchurch et al. (1990) are also shown. Restriction endonuclease sites that are referred to in other parts of this study are shown.
91 92
Pa tl CTGCAGGTAGTTTTCGCCGAAGTCGCGAAGGCCGGCGGCGTGCGCCTCGCGCACCGCGGCGGCGGGCTTGGTCTTGCTCA 80 S a p I CGGCGAGCAGGCCGACCGTGGCCGGATCGCGCCCCGCAGCTTGCGCTGCCTCACGGATGCGCGCGGCAACCTTTGCAATA 1 6 0
TTCTCTGCTATCGTGGACATTACCTGCGCCCTATGGAAGTGCGGTCGGCGTCGGCTCGACGCCCTCGAAACCGCCGGCCT 2 4 0 Mlu I CGATGGCCATCTTCCTGTTCCAGTGTCGAGGTTCCGCGTCTGGCGCAGGACACCACGCGTGTCCTGGAAGCACCCTGGCT 3 2 0 p l l T-> CATCCGGTGTTTTCCTTGTCCGAGACGGCGGCTTTGGCGGCATTCTACCTGCTTGCTTGTAATTG(fGGAGfrcCCATGGAT 4 0 0 M D 2
ATTACCGAGCTGCTCGCCTTCAGTGCCAAACAGGGCGCTTCGGACCTGCACCTCTCCGCCGGCCTGCCACCCATGATCCG 4 8 0 ITELLAFSAKQGASDLHLSAGLPPMIR 2 9
GGTGGATGGCGATGTACGCCGGATCAACCTGCCACCGCTGGAACACAAGCAGGTGCATGCGCTGATCTACGACATCATGA 5 6 0 VDGDVRRINLPPLEHKQVHALIYDIM 5 5 JEcoRX Xhol ACGACAAGCAGCGCAAGGACTTCGAGGAATTCCTCGAGACCGACTTCTCCTTCGAGGTGCCGGGCGTCGCGCGTTTCCGG 6 4 0 NDKQRKDFEEFLETDFSFEVPGVARFR 8 2
GTCAACGCCTTCAACCAGAACCGTGGCGCCGGCGCGGTATTCCGGACCATTCCCTCCAAGGTACTGACCATGGAGGAGCT 7 2 0 VNAFNQNRGAGAVFRTIPSKVLTMEEL 1 0 9
TGGCATGGGAGAAGTGTTCAAACGTGTTTCAGACGTCCCGCGCGGGTTGGTACTGGTCACCGGGCCGACCGGTTCGGGCA 8 0 0 GMGEVF KRVSDVPRGLVLVT GPTGSG 1 3 5
AGTCCACCACCCTGGCGGCGATGCTCGATTACCTGAACAACACCAAGTACCACCACATCCTCACCATCGAGGACCCGATC 8 8 0 KSTTLAAMLDYLNNTKYHHI LT IEDP I 1 6 2 .EcoRI GAATTCGTCCACGAATCGAAGAAGTGCCTGGTCAACCAGCGCGAGGTGCATCGCGACACCCTCGGCTTCAGCGAAGCGCT 9 6 0 EFVHESKKCLVNQREVHRDTLGFSEAL 1 8 9
GCGCTCGGCGCTGCGGGAGGACCCGGACATCATCCTGGTCGGCGAGATGCGCGACCTGGAAACCATCCGCCTGGCCCTGA 1 0 4 0 RSALREDPDI ILVGEMRDLETIRLAL 2 1 5 Sail CCGCGGCGGAGACCGGCCACCTGGTATTCGGCACCCTGCACACCACCTCGGCGGCGAAGACCATCGACCGGGTGGTCGAC 1 1 2 0 TAAETGHLVFGTLHTTSAAKTIDRVVD 2 4 2
GTGTTCCCGGCCGAGGAAAAGGCCATGGTTCGCTCGATGCTCTCCGAGTCGCTGCAATCGGTGATCTCGCAGACCCTGAT 1 2 0 0 VFPAEEKAMVRSMLSESLQSVISQTLI 2 6 9
CAAGAAGATCGGCGGCGGCCGGGTGGCGGCCCACGAGATCATGATCGGCACCCCGGCGATCCGCAACCTGATCCGCGAGG 1 2 8 0 KKIGGGRVAAHEIMXGTPAIRNLIRE 2 9 5
ACAAGGTCGCGCAGATGTATTCGGCGATCCAGACCGGCGGCTCGCTGGGCATGCAGACCCTCGACATGTGCCTCAAGGGC 1 3 6 0 DKVAQMYSAIQTGGSLGMQTLDMCLKG 3 2 2
CCTGGTCGCAAGGGCCTGATCAGCCGCGAGAACGCCCGCGAGAAGGCGAAGATCCCGGAAAACTTCTGATCCTGGCGCCG 1 4 4 0 L V AKGLXSREN AREKAKXPENF • 3 4 4 JCpnl ATCCGCCGCGCTTCGCCCGAATCCCGGAACTGCGTCTAGGATTCAAGAGCGGGAGAGCGGTACC 15 04
Figure 10 Figure 11. Construction of the piZT::pBR322 disruption m utant PAO-PT.
Plasmid pPT2381, which contains a 300 bp EcoRI fragment internal to pilT in the suicide vector pBR322, was introduced into P. aeruginosa PAOl. A single recombination event integrated pPT2381 into the chromosome, forming the pilT mutant PAO-PT.
93 94
ORI
Tc pPT2381
Ap
p ilT
I 1 i\ '\ i , PAOl — Chromosome pilT
PAO-PT ”)___ Bl—p i i i i i Chromosome pilT pilT
Figure 11 95
Characterization of the mutated pilT allele from HOD1 To characterize the HOD1 mutation by sequence analysis, I cloned the mutated p ilT allele. Because pilT resides on a 6.0 kb BamHl fragment, a mini-library containing 5.0-7.0 kb chromosomal BamHl fragments from HOD1
was introduced into E. coli DH5a and used to clone the HOD1 pilT allele. The
1.5 kb Pstl-Kpnl PAOl p ilT fragment from pPT2154 was used to probe 600 recombinant E. coli colonies by using colony hybridization. One colony
contained a recombinant plasmid that hybridized to the probe. As expected, the plasmid isolated from this colony contained an insert of approximately 6.0 kb. The 3.0 kb Pstl-BamHl fragment, containing the HOD1 pilT allele, was
subcloned into the Pstl and BamHl sites of pUC18 to form pPT3729. Results from restriction mapping studies of pPT3729 digested with EcoRl were as expected for the pilT done, confirming that the appropriate fragment had been isolated from HOD1.
The 1.5 kb Pstl-Kpnl fragment containing the HOD1 pilT was subcloned
from pPT3729 and ligated into the Pstl and Kpnl sites of pK19. This plasmid,
designated pPT3734, was used to generate nested deletions for DNA sequence determination. The sequence obtained revealed the presence of an additional C:G base pair inserted in the HOD1 pilT open reading frame 233 nt upstream of the putative translation stop codon (Fig. 12). This insertion should result in a frameshift within pilT, which would manifest itself by encoding a 40 kDa protein as opposed to the wild type 37 kDa PilT. Figure 12. Identification of the HOD1 pilT mutation. The nucleotide sequence and putative amino acid sequence of the 3 ' end of the wild type P. aeruginosa PAOl pilT gene is shown along with the nucleotide sequence changes identified in HOD1 and the three other characterized pilT mutants. Numbering of the nucleotide sequence is identical to that used in Figure 10. The HOD1 pilT mutation was identified as a single base pair (C:G) insertion between nt 1195-
1197, causing a frameshift mutation which would extend the pilT open reading frame by 50 nt. The predicted amino acid sequence which would result from this frameshift mutation is shown below the amino acid sequence of the PAOl PilT protein. The HOD1 mutation lies in the same region of pilT as the three other characterized p ilT mutations (Whitchurch and Mattick, 1994). The mutants PA02001.2 and K2.2 both contain deletions in pilT. The nucleotides affected are indicated, as are the resulting changes in the predicted amino acid sequences of these mutants. The mutant R364 contains a y8 transposon insertion after nucleotide 1073.
96 3' end pilT-> GCGCTGCGCTCGGCGCTGCGGGAGGACCCGGACATCATCCTGGTCGGCGAGATGCGCGACCTGGAAACCATCCGC 1 0 3 1 A L |R S A l | REDPD IILVGEMRDLETIR PA02001.2 R3 64 in-frame A <1 CTGGCCCTGACCGCGGCGGAGACCGGCCACCTGGTATTCGGCACCCTGCACACCACCTCGGCGGCGAAGACCATC 1106 LALTAAETGH LVFGTLHTT SAAKTI A S a il 4 GACCGGGTGGTCGACGTGTTCCCGGCCGAGGAAAAGGCCATGGTTCGCTCGATGCTCTCCGAGTCGCTGCAATCG 1181 DRVVDVFPAEEKAMVRSMLSESLQS RPWF ARCSPSRCNR C 1C2.2 Fram eshift 4 GTGATCTCGCAGACCCTGATCAAGAAGATCGGCGGCGGCCGGGTGGCGGCCCACGAGATCATGATCGGCACCCCG 1256 V I S QTLI KKIGGGRVAAHEIMI GTP * PDQEDRRRPGGGPRDHDRHP B0D1 Fram eshift
GCGATCCGCAACCTGATCCGCGAGGACAAGGTCGCGCAGATGTATTCGGCGATCCAGACCGGCGGCTCGCTGGGC 1331 AIRNLIREDKVAQMYS Al QTG GSLG GDPQPDP RGQGRADVFG DPDRR LAG
ATGCAGACCCTCGACATGTGCCTCAAGGGCCTGGTCGCCAAGGGCCTGATCAGCCGCGAGAACGCCCGCGAGAAG 1406 M Q TLDMCLKGLVAKGLISRENAREK HADPRHVPQGPGRQGPDQPRERPRE
GCGAAGATCCCGGAAAACTTCTGATCCTGGCGCCGATCCGCCGCGCTTCGCCCGAATCCCGGAACTGCGTCTAG 1 4 8 1 AKIPENF* GEDP GKLLI LAPIRRASPE SRNCV *
Figure 12 CO 98
I had identified the H0D1 mutation after having determined the nt sequence 800 nt from the 3 ' end of the 1.5 kb Pstl-Kpnl fragment. Thus, another mutation may have existed elsewhere in the pilT open reading frame or in the promoter region that might be responsible for the phage-resistant, non-twitching phenotype. Restriction fragments were exchanged between the pilT alleles of HOD 1 and PAOl to confirm that the identified insertion was the mutation responsible for the phenotype of HOD1. The 0.39 kb Sall-K pnl fragment spanning the 3 1 end ofpilT was exchanged between the HOD1 pilT
allele and the wild type PAOl p ilT allele. The HOD1 0.39 kb Sall-K pnl fragment containing the insertion mutation was ligated into the Sail and Kpnl sites of pPT3738 forming pPT3799. Similarly, the equivalent wild type 0.39 kb
Sall-Kpnl fragment was purified from pPT3738 and ligated into pPT3733, forming pPT3803. Thus, the final two constructions both contained the 1.5 kb Pstl-Kpnl pilT gene in pUCP19. However, pPT3799 contained the wild type
allele with the exception of the 0.39 kb Sall-Kpnl from HOD1, and pPT3803 contained the HOD1 allele with the exception of the PAOl Sall-Kpnl region.
Carbenicillin-resistant transformants of HODl/pPT3803 displayed restored twitching and phage sensitivity; however, pPT3799 did not complement HOD1. The 0.39 kb HOD1 Sall-Kpnl fragment was responsible for the inability of pPT3799 to restore twitching motility and phage-sensitivity to HOD1. Any unidentified mutation potentially present elsewhere in the HOD1 pilT gene has no effect on twitching motility or phage sensitivity. 99
The H0D1 p ilT gene product had a predicted mass of 40 kDa as a consequence of the frameshift mutation. To confirm this prediction, antiserum (from M. Koomey, University of Michigan) raised against the homologous Neisseria gonorrhoeae PilT protein (discussed below) was used to detect this 40 kDa protein in a crude extract of HOD 1. Antiserum specific for P. aeruginosa
PilT was not available; therefore, the anti-gonococcal antiserum was tried since the PilT proteins from these two bacteria share 67% amino acid identity (Brossay et al., 1994). Cell lysates from PAOl, PAO-PU (a pilU m utant discussed in Chapter IV), and two pilT mutants, HOD1 and the pt7T::pBR322 disruption mutant PAO-PT (described earlier), were examined using Western blot analysis (Fig. 13). The anti-gonococcal PilT antiserum recognized the P. aeruginosa PilT protein, and a band representing the 37 kDa PilT protein was detected in the PAOl and PAO-PU lysates. A cross-reacting protein of approximately the same size as PilT was also detected in all four lysates. The wild type 37 kDa PilT protein was not observed in either of the pilT mutants. A unique 40 kDa protein was detected in the HOD1 lysate, supporting nucleotide sequence analysis. No unique protein was evident in the PAO-PT lysate, consistent with my hypothesis that this strain was a null mutant as a consequence of the integration of pBR322 into the chromosomal pilT gene.
Generation of pilT mutants by using random mutagenesis Since the four characterized pilT mutations are mutated in the same region, I wanted to randomly mutagenize pilT to try to obtain mutations in other regions of this gene. A mutator strain of E. coli (jnutD m utS mutT) was Figure 13. Western analysis of PilT in cell lysates from P. aeruginosa. Rabbit anti-gonococcal PilT antiserum was used to detect Pseudomonas PilT. Lanes:
1, gonococcal PilT; 2, the pilU mutant PAO-PU; 3, PAOl; 4, HOD1; 5, PAO-PT.
The 37 kDa wild type PilT (bottom arrow at left) and the 40 kDa mutant HOD1
PilT (upper arrow at right) were detected. No PilT protein was detected in the null mutant PAO-PT.
100 Figure 13 102
used for this mutagenesis in an attempt to isolate random point mutations in pilT. Plasmid pPT3736, which contains pilT in the shuttle vector pUCP18,
was propagated in the mutator strain for four or five passages and introduced into HOD1. Mutations affecting the function of PilT could be identified by an inability to complement HOD1 to wild type levels of twitching motility. Twitching was used as the primary screen in an attempt to isolate mutants that differed from those isolated previously using phage resistance as the primary screen. I screened 200 transformants for twitching motility, and six were unable to twitch or produced a motility zone that was smaller than the twitching zone of PAOl (Table 7). The plasmid-borne p ilT gene was presumably mutated in each of these six colonies, preventing complementation
of the HOD1 mutation. These six mutants were also quantitatively tested for sensitivity to phage D3112 cts, and mutants 4.2 and 5.4 formed approximately half the number of plaques seen with PAOl. In addition, the plaques formed by
the two mutants were smaller and more turbid than that seen with PAOl.
Mutants 4.2, 5.3, and 5.4 display novel pilT phenotypes.
Minipreparations of the mutant alleles were digested with restriction endonucleases and analyzed by gel electrophoresis to confirm that the mutated pilT alleles were intact. I identified the approximate location of the mutations in 4.2 and 5.2 by systematically replacing restriction fragments in these m utant pilT alleles with the analogous restriction fragment from the PAOl wild typepilT allele. In separate experiments, I replaced the 0.6 kb Pstl-Xhol fragment and the 0.9 kb Xhol-Kpril fragment (Appendix D) of the mutatedpilT 103
Table 7. p ilT mutations generated using an E. coli mutator strain.
M utant Tw itching Sensitivity to Location of 1 Designation8 Motilityb D3112c mutation in pilT* |
4.1 - resistant N.D. I
4.2 - decreased sensitivity Xhol-Kpnl |
5.1 - resistant RD.
5.2 - resistant Xhol-Kpnl | 5.3 decreased resistant N.D. I 5.4 decreased decreased sensitivity RD. 1 a The first digit in each mutant designation indicates the number of passages through the mutator strain. b Twitching motility was classified as decreased when the motility zone was smaller than that produced by PAOl; (-), unable to twitch. c Mutants with decreased phage sensitivity formed approximately 50% fewer plaques with D3112 than that seen with PAOl; plaques in these mutants were also smaller and more turbid. «* N.D., not determined. 104 alleles with the fragment from the wild type allele and determined if these hybrid plasmids were capable of complementing HOD1 . Both mutations localized to the 0.9 kb Xhol-Kpnl fragment that spans the majority of thepilT open reading frame. I have not determined the locations of the four remaining pilT mutations.
Virulence of HOD1 in a mouse infection model To determine whether the altered piliation of HOD 1 affected virulence in addition to twitching motility and phage sensitivity, LD50 studies were performed using the A.BY/SnJ mouse infection model (Farinha et al., 1994).
The virulence studies were conducted by Linda Glasier at the University of Alberta (Alberta, Edmonton, Canada). One trial was performed in which ten non-immunosuppressed A.BY/SnJ mice were injected with various concentrations of the mutant strain HOD1 and the wild type parent strain PAOl for each challenge. The concentrations of bacteria injected in each challenge and the resulting mortality data are presented in Figure 14. The
LD50 values were determined from the graph and were approximately1 x 1 0 6 colony forming units (cfu) per mouse for PAOl and 2 x 106 cfu per mouse for
HOD1. Thus, PAOl was approximately twice as virulent as HOD1 when each bacterium was injected intraperitonealy into A.BY/SnJ mice. However, the statistical significance of the difference between these LD 50 values cannot accurately be determined from this individual experiment. Figure 14. Virulence of P. aeruginosa PAOl and HOD1 in A.BY/SnJ mice.
Groups of ten non-immunosuppressed mice were injected intraperitonealy with various concentrations (cfu) of each bacterium and observed for 48 hr for mortality. LD5 0 values were extrapolated from the graph. Squares, PAOl; circles, HOD1.
105 % M ortality 100 20 40 60 80 0 105 tralDoe( rmo se) ou m er (p ose D l cteria a B Figure 14 Figure 108 107
Discussion
The initial experiments performed in this study proved successful in that I was able to isolate a mutant of P. aeruginosa PAOl that was defective in its ability to form functional pili. This mutant, which was designated HOD1, was classified as a hyperpiliated strain that produced nonretractile pili. The pili assembled by HOD1 are nonfunctional in that they are unable to mediate either twitching motility or infection by pilus-specific phages. In addition,
HOD1 appears to be slightly less virulent than the parent strain, PAOl. Phenotypic complementation studies using HOD1 resulted in the identification of a novel gene involved in type IV pilus retraction. This same gene was
simultaneously characterized by Whitchurch et al. (1990), who designated it pilT.
The high number of spontaneous phage-resistant mutants obtained in this study is likely due to the large number of pil genes that are required for pilus biogenesis in P. aeruginosa PAOl. Twenty-three genes have been identified to date as being required for pilus biogenesis and function and have been mapped to four different regions on the PAOl chromosome (Fig. 2). On the basis of transposon mutagenesis studies, Hobbs et al. (1993) have proposed that there may be as many as 40 genes involved in pilus biogenesis in P. aeruginosa. 108
The pilT gene was the first pil gene to be mapped to the 20 min region of the PAOl chromosome. Seven other pil genes have since been mapped near this region on the chromosome, including a second gene designated pilU, which like pilT is required for pilus retraction and twitching motility (Whitchurch and Mattick, 1994a). The significance of the pilU gene will be discussed in more
detail in Chapter IV. Approximately 10 kb downstream from thepilT pilU genes are an additional seven pil genes (pilG, H, I, J, K, LI, L 2) that encode part of a signal transduction pathway that controls pilus production and twitching motility (Darzins, 1993, 1994, 1995; J. Mattick, personal communication). The products of these particular genes are homologous to the chemotaxis proteins of enteric bacteria (Che) and the gliding bacterium Myxococcus xanthus (Frz) (Darzins, 1993,1994, 1995). No direct involvement between any of these chemotaxis-like Pil proteins and PilT has been established.
HOD1 , HOD3, HODIO, and HOB6 were capable of adsorbing the pilus- specific bacteriophages yet were phage-resistant. This indicates that the phage receptors on the pili are intact and suggests that the absence of pilus retraction prevents the phages from infecting the cells. I have not directly demonstrated that the pili on the hyperpiliated mutants in this study are non- retractile; however, my hypothesis is supported by the early studies of Bradley (1974). In these studies, Bradley (1974) utilized electron microscopy to demonstrate that phage particles were adsorbed randomly along the lengths of the pili present on the hyperpiliated mutant K/2PfS. Conversely, the phage 109 particles that had been adsorbed by the wild type pili present on strain K were attached to the cell surface at the base of the pili. Bradley (1974) hypothesized that the phage particles adsorb to the sides of the wild type pili and are pulled to the cell surface by pilus retraction. The phages adsorbed to the non-retractile pili could not be drawn to the surface and therefore remained at the point of initial adsorption. Thus, injection of the phage DNA into the bacterial cell could not occur, rendering the cell phage-resistant. Bradley (1972b,c) reported similar results in his studies involving adsorption of the
RNA phage PP7 to the wild type strains PAOl and PA038 and to the hyperpiliated mutants of these strains, PA068 and PA01264, respectively. On the basis of Bradley’s studies, it is generally accepted that hyperpiliated mutants assemble nonretractile pili. An experiment designed to radiolabel assembled pilin subunits could prove very useful in directly demonstrating that wild type pili retract and that the pili assembled on hyperpiliated mutants cannot retract.
P. aeruginosa pilT encodes a protein with an apparent molecular weight of 37 kDa (this study; Whitchurch et al., 1990). The predicted amino acid sequence of PilT (Whitchurch et al., 1990) contains the nucleotide-binding consensus sequence GXXGXGKT/S known as a Walker box A (Walker et al.,
1982). This motif is believed to form the P-loop in which the conserved lysine (K) residue contacts the phosphoryl group of the bound nucleotide (Possot and Pugsley, 1994). Whitchurch et al. (1990) reported that the predicted PilT amino acid sequence is homologous to several proteins (discussed below) that 110
also contain Walker box A sequences (Fig. 15). These proteins exhibit
homology across their entire amino acid sequences, but they are particularly highly conserved in the region containing the Walker box A (Fig. 16). A profile sequence of PilT and its homologs was used to search protein data bases via BLAST and FASTA in an attempt to identify additional homologous proteins and potentially novel motifs (this study). Although no additional amino acid sequence motifs were identified for the HIT group of homologs, a dendrogram including recently identified homologs was compiled (Appendix E).
A second highly conserved nucleotide-binding motif was identified in most of the PilT homologs, yet it is not well conserved in PilT. Whitchurch et al. (1990) reported that this second conserved region corresponds to the Walker box B motif (R/KXXXGXXXL-hydrophobic-hydrophobic-hydrophobic-hydro- phobic-D) which is found occasionally in proteins with nucleotide-binding capabilities (Walker et al., 1982). The presence of a Walker box B in this set of proteins was also reported by Lauer et al. (1993). However, Possot and Pugsley (1994) identified a different region in these proteins (Fig. 15) as the
Walker box B. These authors instead classify Whitchurch's Walker box B as one of two aspartate boxes in these proteins. The two aspartate boxes consist of two short motifs (TXEDPXE and RXXPDXXXXGEI/ MRD) that each contain at least one aspartate residue. All of the PilT protein homologs thus far identified contain these consensus sequences (Fig. 15; Possot and Pugsley, 1994). The Walker box B identified by Possot and Pugsley is in better agreement with the conserved box B sequence published by Walker et al. Figure 15. Partial alignment of the highly conserved regions of the putative
nucleotide-binding site in PilT and its related proteins (modified from Possot
and Pugsley, 1994). The Walker box A, aspartate (Asp) boxes, Walker box B,
and the four conserved cysteine residues are denoted by boxes. The Walker box B that had been identified by Whitchurch et al., is indicated by dots (•).
Dashes (-) within the sequences represent gaps that were introduced to improve alignments. Numbers correspond to the amino acid sequence of P. aeruginosa PilT given in Figure 10. Underlined residues correspond to the locations of the PilT mutations which were shown in Fig. 12. The K2.2 mutation, which is not shown here, is located in the gap between residues
243-266. The proteins aligned here are: Klebsiella oxytoca PulE (Possot et al.,
1992); P. aeruginosa XcpR (Bally et al., 1992); Erwinia chrysanthemi OutE
(Lindeberg and Collmer, 1992); Erwinia carotovora OutE (Reeves et al., 1993);
Xanthomonas campestris XpsE (Dums et al., 1991); Aeromonas hydrophila
ExeE (Jiang and Howard, 1992); Vibrio cholerae EpsE (Sandvist et al., 1993); P. aeruginosa PilB (Nunn et al., 1990) V. cholerae TcpT (Kaufman et al., 1993;
Ogierman et al., 1993); P. aeruginosa PilT (Whitchurch etal., 1990; this study);
Neisseria gonorrhoeae PilT and PilF (Lauer et al., 1993); B. subtilis ComGl
(Albano et al., 1989); Agrobacterium tumefaciens VirBll (Ward et al., 1988);
Bordetella pertussis OrfH (Weiss et al., 1993).
I l l 119 Walker box A Asp box Asp box 216 Ko PulE IARPHGIVLV! GPTGSGKS :tTL— l - YAALSRLDAR-ERN-ItTIEDPIE CELEG- IGQTQVNAKVDMTFARGLRAI1 RQDPDWLVGEIRD 5ETAQIAVQ Pa XcpR VRKPHGILLV' GPTGSGKT :TL—:t l - YASLTTLNDR-TRN-IljTVEDPIEitHLEG------IGQTQVNAKVDMTFARGLRAI! RQDPDWMVGEIRD CETAEIAVQ Ech OutE IHRPHGIILV' GPTASGKS ’TL- TVEDPIE fELEG- IGQTQVNPKVDMTFARGLRAI1 RQDPDWLVGEIRD 5ETAQIAVQ Eca OutE IHRPHGIILV! GPTGSGKS ?TL- TVEDPIE fELEG- IGQTQVNTKVDMTFARGLRAII RQDPDWLVGE I RD ETAQIAVQ Xc XpsE LEQPHGIMLV: GPTGSGKT :t l - TVEDPVE 'QIEG- INQIQAKPQIGLDFANALRSI! RQDPDIIMIGEMRD jETARIAIQ Ah ExeE IRKPHGIILV! GPTGSGKS ’TL- TVEDPIE fDLEG- VGQTQVNTKVDMTFARGLRAI1 RQDPDWMVGEIRD ETAQIAVQ Vc EpsE IKRPHGIILV GPTGSGKS :TL—t l - YAGLQELNSS-ERN-lflTVEDPIE DIDG------IGQTQVNPRVDMTFARGLRAI] RQDPDWMVGEIRD .ETAQIAVQ Ng PilF IHRPYGMVLV: GPTGSGKT CSL- TAEDPAE :NLPG------INQVNVNDKQGLTFAAALKSFI RQDPDIIMNGEIRD .ETADIAIK Pa PilB LKQPQGMILV] GPTGSGKT fSL- TAEDPVE :NLEG------INQVNVNPRQGMDFSQALRAFI RQDPDVIMVGEIRD ,ETAEIAIK Vc TcpT LNTSYGLFIVi GTTGSGKS ’SLK1 TVEDPVE LISGAQQSS IVADNDDKTKNPFADAVRSAI RRDPDVIMIGEIRD CPTVEALSS Pa PilT SDVPRGLVLV! GPTGSGKS :t l - VHES---- KKCLVNOREVHRDTLGFSF.AT.RSAI RF.DPDTILVGEMRD .ETIRLALT Pa PilO ALTKRGLVIF\ GATGTGKS :s l - ITIHQH---- QGCIVTQREVGLDTDSFEVALKNTI RQAPVDIMIGEVRS CETNDHAVA Ng PilT AESPRGMVLV: GPTGSGKS ’TL~ VHQS---- KKSLINQRELHQHTLSFANALSSA! REDPDVILVGEMRD ’ETIGLALT Bs ComGl LKHSHGMLIF' GPTGSGKT 'TL- TLEDPVEfRDED------VLQVQVNEKAGVTYSAGLKAII RHDPDMIILGEIRD VETAEIAVR At VirBll CWSRLTMLLC GPTGSGKT !MSK .VI-PHDNHVRLLYSKNGAG-LGAVSAEHLLQAS1 RMRPDRILLGEMRD JAAWAYLSE Bp OrfH AVQAGKAILW GQTGSGKT :lmn TIEDVRE jRLDPATNHVHLLYGTPTEGRTAAVSATELLRAAI RMAPTRILLAELRG 5EAFDFLQA
217 243/266 Walker box B 310 Ko PulE ASLTGHLVLSTLHTNSALGAISRLQDM QRLVRR] C >i C IQQEPA APGTALWQPRC C U C !FTG RGRTGIHELLLVDDfcVRMAIHRGENEV--TLIQQLGT Pa XcpR ASLTGHLVLSTLHTNSAIGAITRLVDM QRLVRV! C CEPYRA AAPPTLHRARC C !I C 1QHG! RGRTGIYELWFDDpMRSLIHNESSEQ— EMTRHART Ech OutE ASLTGHLVLSTLHTNSALGALSRLQDM QRLVRT! C C IQPYTI AAGTTLYHPGC C !l C IYSG RGRTGIHELLLIDD VRAAIHRGESEL— GIARMLGA Eca OutE ASLTGHLVLSTLHTNSALGALSRLQDM QRLVRT1 C C iQPQPV APGTLLHNPVCC >< C >FTG RGRIGIHELVLIND JVRAAIHRSDGEM— AIAQILGG Xc XpsE SALTGHLVLSTLHTNNAAGGITRLLDM QRLVRKLDLANAERYAA DGEIFLYRPRATAA-APTG' LGRTTIVEFLVMND ] LRRAVMRRAGMG— EIEQLARK Ah ExeE ASLTGHLVMSTLHTNTAIGAITRMRDM QRLVRT] C >1 IAPRPI APDQQVWRPVC C C IHTG RGRTGIHELWIDE WREAIHSASGEL— AIERLIRD Vc EpsE ASLTGHLVMSTLHTNTAVGAVTRLRDM QRLVRT] Cpi CEPYEA KEPLILYRATC >1 C IHKG RGRTGIHELLLVDD iLQELIHSEAGEQ— AMEKHIRA Ng PilF AAQTGHMVFSTLHTNNAPATLSRMLNM QRLLRR1 C (QEVER AKDWKLYGAVC Cj)I C IGQG KGRAGVYEVMPISE MQRVIMNNGTEV— GILDVAYK Pa PilB AAQTGHMVMSTLHTNSAAETLTRLLNM QRLARK] C CKEHDV IGTFKLYSPVC Cpi C C-NG! KGRVGIYEWKNTP HQRIIMEEGNSI— EIAEQARK Vc TcpT AVESGHYCLTTIHAGSWSVLQRLSGL q k l i p e : C QRAVFSANENCC IHSG1 KGRLLLLETLVPTV DLELVASENWVSLYRKYRERRF Pa PilT AAETGHLVFS1LHTTSAAKTIDRWDV OTLIKKI------GGRVAAHEIMIGTP ^IRNLIREDKVAQMYSAIQTGGS Pa PilU FAETGHLCLATLHANNANQALERIIHF QQLVPT ------DC KGRRAVIEVLLNTP iAADLIRKGEVHELKPLMKRSTE Ng PilT AAETGHLVFGTLHTTGAAKTVDRIVDV QNLLKTH------NGRVASHEILIANPfcVRNLIRENKITQINSVIQTGQA Bs ComGl AAMTGHLVLTSLHTRDAKGAIYRLLEF QRLVDLj[cj>^cJ:N------GCSSV'jcjcQSRI TRRASVYELL-YGK ILQQCIQEAKGNH— ANYQYQTL At VirBll WSGHPGSISTIHGANPIQGFKKLFSL aDVIIPFRAYEDVYEVGEIWLA Bp OrfH CASGHSGGISTCHAASADMALQRLTLM fIDIW------WERRAG 112 Figure 15 Figure 16. Schematic alignment of the PilT homologs (adapted from
Whitchurch, 1994). The white boxes represent the regions sharing the highest amino acid sequence similarity, with the first eight amino acids constituting the Walker box A motif. Adjacent dark gray boxes represent regions of high sequence similarity, while the regions depicted by the lighter gray boxes have a slightly lower similarity. The black boxes represent regions of low sequence similarity. Boxes filled with the same patterns contain regions with high similarity within that specific subgroup but that have a low degree of similarity to the other proteins. Dashed lines were included where gaps had been introduced to approve alignment. Unbroken lines represent regions of low sequence similarity. The proteins aligned here include KilB from the
IncP plasmid pRK2 (Motallebi-Veshareh et al., 1992), the N. gonorrhoeae proteins PilF and ORF4 (Lauer et al., 1993), and FimN from D. nodosus
(Johnston et al, 1995). Refer to Figure 15 for the descriptions and references of the remaining proteins.
113 114
# Amino Acids I I I I I I I I 0 100 200 300 400 500 600 650
P. aeruginosa PilT
N. gonorrhoeae PilT P. aeruginosa PilU N. gonorrhoeae ORF4 im iiH B. pertussis ORFH A. tumefaciens VirB-11 roooi IncP plasmids KilB KWN B. subtilis ComG-1 V. cholerae TcpT
P. aeruginosa PilB
N. gonorrhoeae PilF
D. nodosus FimN
E. carotavora OutE £ chrysanthemi OutE K. oxytoca PulE
A. hydrophila ExeE
V. cholerae EpsE
P. aeruginosa XcpR
X. campestris XpsE
Figure 16 115
(1982) than the region identified by the other investigators. The sequence of amino acids between these two boxes is highly conserved in all of the PilT homologs, although it displays no similarity to the corresponding sequence found in classical ATPases (Possot and Pugsley, 1994).
One of the proteins that Whitchurch et al. (1990) identified as a PilT homolog is the P. aeruginosa PilB protein. As discussed in Chapter I, pilB mutants are nonpiliated, despite the fact that they still produce the pilin subunit (Nunn et al., 1990). PilB is therefore thought to be an accessory protein that is required for the actual assembly of the pilus (Nunn et al., 1990).
Another PilT homolog is the P. aeruginosa PilU protein, which is required for twitching motility but not phage sensitivity (Whitchurch and Mattick, 1994a). PilT and PilU are believed to have related but unique roles in pilus function (Whitchurch and Mattick, 1994a).
Recently a functional homolog of the P. aeruginosa PilT protein was identified in N. gonorrhoeae (Lauer et al., 1993). This protein, also designated PilT, is required for twitching motility. Furthermore, like in P. aeruginosa, pilT mutants of N. gonorrhoeae are hyperpiliated (Lauer et al., 1993). Of the proteins that are homologous to the P. aeruginosa PilT protein, the gonococcal PilT exhibits the highest degree of similarity. The deduced amino acid sequences of the PilT proteins from these bacteria are 6 6.6 % identical and 81.4% similar, and the Walker box A motif is conserved in the gonococcal PilT protein (Brossay et al., 1994). Although the gonococcal PilT is hydrophilic, 116
Brossay et al. (1994) observed that the protein localizes to both the cytoplasm and the inner membrane. Brossay et al. (1994) postulated that because PilT lacks any potential membrane-spanning domains and yet partially localizes to the inner membrane, it is a cytoplasmic protein that associates with an as yet unidentified integral membrane protein.
PilT homologs are not limited to type IV pilus biogenesis proteins. Type IV Pil proteins are a subset of a larger, more extensive family of proteins which are involved in the formation of bacterial surface-associated protein complexes (Whitchurch et al., 1990; Hobbs and Mattick, 1993). This family includes proteins that are involved in the general secretion pathway of gram-negative bacteria, morphogenesis of filamentous bacteriophages, and DNA transport (Whitchurch et al., 1990; Hobbs and Mattick, 1993).
Some of the better characterized Pil-like proteins that are involved in extracellular protein secretion are the P. aeruginosa Xcp proteins (secretion of various exoenzymes including the virulence factors phospholipase C, exotoxin
A, and elastase) (Filloux et al., 1987; Lindgren and Wretlind, 1987; Wretlind and Pavlovskis, 1984), the Klebsiella oxytoca Pul proteins (secretion of pullalanase)
(Reyss and Pugsley, 1990), the Erwinia spp. Out proteins (secretion of plant- degrading enzymes) (Lindeberg and Collmer, 1992; Reeves et al., 1993), the Xanthomonas campestris Xps proteins (secretion of various extracellular enzymes including cellulases) (Dums et al., 1991; Hu et al., 1992), and the
Aeromonas hydrophila Exe proteins (secretion of the aerolysin toxin) (Hobbs and Mattick, 1993) (Jiang and Howard, 1992). Hobbs and Mattick (1993) also reported that type IV Pil proteins are homologous to proteins that are required for secretion in Pseudomonas solanacearum (Huang et al., 1992), Pseudomonas syringae (Allaoui et al., 1993), Shigella flexneri (Michiels et al., 1991), Yersinia spp. (Haddix and Straley, 1992), andV. cholerae (Kaufman et al., 1993;
Ogierman et al., 1993). In addition, at least nine proteins sharing homology to this extensive family have been identified in E. coli (Finch et al., 1986; Whitchurch and Mattick, 1994b; Whitchurch et al., 1990); however, the functions of these proteins are not yet known (Whitchurch and Mattick, 1994b). Pil homologs involved in DNA transport include proteins that are required for transporting DNA into cells such as the competence proteins of
Bacillus subtilis (Com) and Haemophilus influenzae, as well as transferring DNA out of the cell, as with the virulence (Vir) proteins of Agrobacterium tumefaciens (Hobbs and Mattick, 1993; Lessel et al.,. 1992).
Each of the above Pil-like systems assembles a surface-located protein complex that is responsible for transporting other macromolecules into and out of the cell and includes proteins with functions homologous to those of the Pil proteins (Hobbs and Mattick, 1993). For example, proteins that are homologous to pilin at the N-terminus have been identified in many of these systems (Table 8 ). The homology at the N-termini of these proteins includes a conserved signal sequence which is cleaved by a sec-dependent leader peptidase (reviewed in Chapter I; Bally et al., 1992; Hobbs and Mattick, 1993). Proteins Table 8. Homologs of P. aeruginosa type IV pilus biogenesis proteins.*
Proteins with prepilin-like Prepilin leader Nucleotide- System signal sequences peptidases binding proteins (CM* (C/CM) (CM) (OM)
Type IV pili® P. aeruginosa PilA PilD/XcpA PilB PilC PilQ PilE PilT PilV PilU
N. gonorrhoeae PilE PilD PilT PilF ORF4 V. cholerae TcpA TcpJ TcpT TcpE
E. coli EAF plasmid BfpA E. coli Incl plasmid PilV PilU
Protein Secretion P. aeruginosa XcpT XcpU XcpV XcpW K oxytoca PulG PulH Pull PuU PulO PulE PulF PulD Erwinia spp. OutG OutH OutI OutJ OutO OutE OutF OutD A. hydtrophiUa ExeG ExeH Exel ExeJ ExeE ExeF ExeD X. campestris XpsG XpsH XpsIXpsJ XpsE XpsF PefD HrpAl S.flexneri MriD Yersinia spp. YscC P. solanaixarum HrpA P. syringae HrpH V. cholerae EpsE Table 8. (Continued)
Proteins with prepilin-like Prepilin leader Nucleotide- System signal sequences peptidases binding proteins (CM)** (C/CM) (CM) (OM)
DNA transfer IncP plasmids KilB A tumefaciens Ti plasmid VirB-11 B. subtilis ComG-3 ComG-4 ComG-5 ComC ComG-1 ComG-2 H. influenzae ORFE
Bacteriophage P. aeruginosa Pf3 ORF430 E. coli M13 gene IV
Other C. perfringens PpdA E. coli PpdA PpdB PpdC HopD HopB HopC HopQ
* Modified from Hobbs and Mattick, 1993. t> The putative locations of the proteins in each class are given (CM, cytoplasmic membrane; OM, outer membrane; C/CM, cytoplasmic but probably associates with cytoplasmic membrane). c With the exception ofN. gonorrhoeae PilE, none of the other classical type IV pilin homologs is included. 120
homologous to the P. aeruginosa PilD leader peptidase have also been identified in these other systems and are responsible for processing the homologous pilin- like protein. In the case of P. aeruginosa, the PilD protein is the leader peptidase required for both pilus biosynthesis (Aim and Mattick, 1995; Nunn and Lory, 1991; M. Russell, unpublished data) and the Xcp pilin-like secretory proteins (Nunn and Lory, 1991). PilT homologs have been identified in several of these Pil-like systems, including P. aeruginosa XcpR, K. oxytoca PulE, A. tumefaciens VirB-1 1 , B. subtilus ComGl, and Erwinia chrysanthemi OutE
(Whitchurch et al., 1990).
Although no studies have focused on P. aeruginosa PilT, lucrative studies have been performed using several of the PilT homologs. In one of these studies, Turner et al. (1993) mutated the nucleotide sequences of P. aeruginosa pilB and xcpR in the codons for a conserved glycine residue found in the Walker boxes A. The alteration, which replaced a glycine residue with a serine residue, resulted in the loss of pilus biosynthesis and Xcp-mediated protein secretion, respectively (Turner et al. 1993). This putative nucleotide-binding sequence is, therefore, essential for the functions of P. aeruginosa PilB and XcpR. These same investigators, however, were unsuccessfiil in their attempts to show that PilB and XcpR, when overproduced in E. coli, were capable of binding ATP (Turner et al. 1993). The authors suggest that PilB and XcpR may be capable of binding ATP only when these proteins interact with other protein components in the pilus assembly or protein secretion machinery. Another possibility is that PilB and XcpR interact with ATP only during the actual 121
process of protein translocation across the membrane (Turner et al. 1993). Precedence for the first possibility involves E. coli UvrB, a DNA-repair protein
which contains a Walker box A but exhibits ATPase activity only when complexed with UvrA (Stephenset al., 1995).
Similar studies carried out with the PilT homolog VirB-11 from A. tumefaciens support the observations described above. Site-directed mutagenesis of the Walker box A sequence which replaced a conserved lysine residue with either an alanine or an arginine residue resulted in the loss of VirB- 11 function, as measured by the loss of virulence (Stephens et al., 1995). Interestingly, VirB-11 binds and hydrolyzes ATP and exhibits autophosphorylation in vitro (Christie et al., 1989).
One of the better studied PilT homologs is the K. oxytoca PulE protein.
PulE is a membrane-associated protein that contains an essential Walker box A (Possot and Pugsley, 1994). However, these same investigators observed that substitutions of either a conserved glycine or aspartate in the Walker box
B of PulE had essentially no effect on pullulanase activity. In accordance with the fact that PilU and the PilT proteins do not have well conserved Walker boxes B, this finding suggests that this motif may be vestigial in nature and therefore not essential for the function of these proteins.
Possot and Pugsley (1994) also investigated the importance of the conserved aspartate boxes. On the basis of the results of other studies with 122
nucleotide-binding proteins, Possot and Pugsley (1994) proposed that the aspartate boxes in the PilT family (Fig. 15) may help to stabilize the formation of the nucleotide-binding fold by interacting with Mg2+ ions. To test their hypothesis, they replaced one of the conserved aspartate residues in PulE with an asparagine residue; this change resulted in a reduction in pullulanase
secretion of approximately 78% (Possot and Pugsley, 1994).
Another characteristic shared by the majority of these putative nucleotide-binding proteins is the presence of four cysteine residues in the highly conserved central region between the two Walker boxes (FHg. 15) (Possot and Pugsley, 1994; Reeves et al., 1993). Possot and Pugsley (1994) have
suggested that these residues are involved in the formation of disulfide bonds that allow the formation of homodimers. Indeed, these investigators reported that PulE forms homodimers. These cysteine residues are not present in either of the two PilT proteins or in PilU (Lauer et al., 1993; Possot and Pugsley, 1994; Whitchurch and Mattick, 1994a), implying that these particular proteins may not function as dimers.
All of these data that were obtained using PilT homologs strongly suggest that PilT in both P. aeruginosa and N. gonorrhoeae also requires the conserved nucleotide-binding sequence for its role in pilus retraction. It is also likely that P. aeruginosa PilT, like its homologs, is a cytoplasmic protein that associates with the cytoplasmic membrane. I radiolabelled PilT, produced from a plasmid-bome copy ofpilT under the control of the T7 promoter, and 123 attempted to localize the radiolabelled PilT in cellular fractions of PAOl. However I could not detect any radiolabelled PilT using SDS-PAGE anlysis. An association between PilT and the cytoplasmic membrane would presumably occur by interaction with another Pil protein which is an integral membrane protein. Such an interaction has been proposed for E. coli TraC, which is a putative nucleotide-binding protein required for the assembly of the F-pilus. TraC normally associates with the inner membrane, but in the absence of other Tra proteins, TraC is detected in the cytoplasm.
The P. aeruginosa PilB and PilT proteins probably act antagonistically to each other: PilB is required for pilus assembly (Nunn et al., 1990) while PilT is required for retraction, a process that may occur by a depolymerization mechanism similar to the model proposed for theE. coli F-pilus (Marvin and Hohn, 1969; Novotny and Fives-Taylor, 1974). The requirement for the
Walker box A in PilB and presumably also in PilT suggests that these two proteins provide the energy required for assembly and retraction. Another possibility is that these proteins phosphorylate themselves or some other Pil protein in a modification step that is required for pilus assembly or retraction.
The nature of the HOD1 mutation is interesting with respect to the types ofpilT mutation that have already been characterized by Whitchurch and Mattick (1994a). As shown in Figure 12, all four of the mutations were clustered near the Sail restriction endonuclease site, toward the 3' end of the coding sequence. The locations of these same mutations with respect to the 124
conserved amino acid motifs of PilT are shown again in Figure 15. All of the mutations occur in the highly conserved central region of the PilT protein that includes sequences necessary for the function of several PilT homologs. The clustering of these mutations may simply be due to the fact that this particular area of the protein is required for PilT function, making this the most likely region to find mutations that render the protein nonfunctional. However, PilT function can be assessed by two different phenotypic characteristics: twitching motility and phage sensitivity. All four pilT mutants were isolated or identified on the basis of their resistance to pilus-specific bacteriophages (this study; Bradley, 1972a, 1980a; Whitchurch and Mattick, 1994a). This initial selection and screening may have biased the type of pilT mutation that has been obtained, thus preventing the detection of potential non-twitching, phage- sensitive mutants. It is possible that this region of PilT is required for the mediation of both twitching motility and phage infection, while other regions of
PilT may be required only for twitching. However, Whitchurch et al. (1990) isolated six 5A transposon mutants of pilT, containing the transposon insertions throughout most of the open reading frame and upstream region, that are all phage-resistant and unable to twitch. This implies that there is not one region of PilT which is required for phage sensitivity. These transposon mutants are likely to be null mutants, and less drastic pilT mutations, such as point mutations, may not share the phenotype of ap ilT null mutant. For example, point mutations may not affect pilus retraction profoundly enough to allow the cells to survive the initial selection for phage resistance. 125
Indeed, the random mutagenesis studies described were designed such that the inital screen for mutants involved alterations in twitching motility. Using this primaiy screen, I isolated sixpilT mutants, three of which displayed the expected non-twitching, phage-resistant phenotype. The three remaining mutants have phenotypes not previously seen with a p ilT mutant. The mutation in one of these novel mutants, 4.2, has been localized to the pilT coding sequence and results in a loss of twitching. However, 4.2 does not display a wild type level of phage resistance, which is evident by the fact that D3112 cts forms approximately 50% less plaques with 4.2 than with PAOl. In addition, the plaques formed with 4.2 are smaller and more turbid than those formed with PAOl. Conversely, the mutation in 5.3 resulted in a complete loss of phage sensitivity but did not completely abolish twitching motility. Mutant 5.4 retained both twitching motility and phage sensitivity, but displayed an approximately 50% decrease in sensitivity to phage D3112 when compared to
PAOl. Darzins (1993) also observed differential phenotypes with two mutant alleles of pilG, which is required for pilus biogenesis: a spontaneous pilG mutant was resistant to D3112, yet sensitive to B3 and F116L, and a pilG insertion mutant was resistant to all three phages. Darzins (1993) proposed that the spontaneous pilG mutant may be capable of assembling a pre-pilus complex that permits D3112 infection. However, the analysis of thep ilT mutants described here is still in a preliminary stage; thus, it is difficult to speculate on the significance of these novel phenotypes to PilT structure and 126
function. Further studies should include sensitivity assays using B3 and
F116L, nucleotide sequence analysis, EM analysis to determine if the mutants are hyperpiliated, and isolation of additional mutants.
The use of pilus-specific bacteriophages in classifying P. aeruginosa pil mutants may prove to be more complex than initially believed. Contradictory results obtained by other investigators with several pil mutants make it difficult for me to interpret the varying degrees of phage sensitivity displayed by the above pilT mutants. Whitchurch and Mattick (1994a) examined their transposon library of putative pili mutants that had been isolated on the basis of altered colony morphology and observed that the majority of the 147 mutants display either exclusive resistance or exclusive sensitivity to four bacteriophages. However, more than 25% of the mutants exhibited sensitivity only to some of the phages or exhibited various degress of phage sensitivity when compared to wild type levels (Whitchurch and Mattick, 1994a).
Mutations in pilGJ^J,and U eliminate twitching and result in various patterns of phage sensitivity (Darzins 1993-4; Whitchurch and Mattick, 1994a), indicating that the requirement of functional pili for both twitching motility and phage sensitivity is not consistent.
The decrease in virulence demonstrated by HOD1 in the A.BY/SnJ mouse infection model implies that pilus retraction is required for the maximal virulence of P. aeruginosa. One model of P. aeruginosa virulence proposes that pilus retraction allows the bacterial cell, after having already established initial attachment to the host tissue via the pilus tip adhesin, to be pulled into closer contact with the host cell (A. Darzins, personal communication). This would
allow for a more substantial interaction between the mucosal cell surface and secondary bacterial adhesins. This model precludes any direct involvement of twitching motility in virulence, making pilus retraction simply a common
factor to both virulence and twitching. An alternative possibility is that twitching motility actually promotes virulence by conferring upon the bacterium the ability to move from one site, thereby disseminating the infection. The actual involvement of retraction and twitching motility in the
establishment of infection could entail either or both of the above scenarios. Virulence studies with the pilT mutants 4.2 and 5.4 could provide valuable information for determining the individual roles of twitching motility and pilus retraction for virulence. The pili produced by 4.2 and 5.4 might still be retractile since both mutants are sensitive to phages, albeit at decreased
levels from that of the wild type. But mutant 4.2 cannot twitch, while 5.4
exhibits decreased levels of twitching from that of PAOl. These mutants may display varying degrees of pathogenicity which could be attributed to the differences in their abilities to twitch.
The role of twitching motility in virulence was investigated by Depiazzi
and Richards (1985), who demonstrated that strains of D. nodosus with decreased abilities to twitch induce less virulent forms of ovine foot rot. However, since these strains were not characterized with respect to the number of pili present on the cells or to the ability of the pili to retract 128 normally, the individual contribution of twitching to virulence remains difficult to determine. Hazlett et al. (1991) have shown that three hyperpiliated m utants of P. aeruginosa strains PAK and PAOl are not infective when applied to the injured corneas of mice. The decreased virulence of these mutants was not simply due to a decrease in adherence because two of the three mutants displayed no significant decrease in binding to an in vitro ocular organ culture model. Saiman et al. (1990) observed that the P. aeruginosa hyperpiliated strains PA02001.2 (thepilT mutant identified in Fig. 12) and
PAK/PR1 bound to bovine trachea epithelial cell monolayers 20% less avidly than their respective parent strains, PAOl and PAK. It is tempting to speculate that the reason HOD 1 remains somewhat virulent is that its nonretractile pili are still capable of mediating initial attachment yet cannot retract to allow more secure adherence. If this assumption is correct, then repetitions of the LD50 study with a nonpiliated mutant of PAOl should result in an even higher LD5 0 , indicating that the presence of nonretractile pili still contributes to virulence.
The mouse infection model used here is only one of numerous methods used to study virulence and is probably one of the least informative since P. aeruginosa rarely colonizes the peritoneum. Additional models available for studying the role of pili in the pathogenesis of P. aeruginosa include the lungs of neonatal mice (Tang et a l., 1995), burned mice (Sato et al., 1988), and scratched corneas of mice (Hazlett et al., 1991). An effective model for further studies involving HOD1 may be the neonatal mouse model of pulmonary infection that has been used successfully by Tang et al. (1995) to monitor pneumonia, bacteremia, and mortality caused by piliated and nonpiliated strains of P. aeruginosa . The use of neonatal mice is beneficial for examining the effect of virulence factors, such as pili, that are involved in the establishment and early stages of lung infection. Also, the ability to monitor the dissemination of infection may be useful for examining strains that vary in their ability to twitch. CHAPTER IV
CONSTRUCTION AND ANALYSIS OF PILU AND PILT PILU MUTANTS IN P. AERUGINOSA PAOl
Introduction
As mentioned in Chapter III, Whitchurch and Mattick (1994a) identified the P. aeruginosa PAK pilU gene which is located 171 bp downstream of pilT and is required for twitching motility but not for phage sensitivity. Three transposon insertion mutants in pilU, which had been isolated on the basis of altered colony morphology, were tested for sensitivity to phages B3, D3112, Pfl, and P04 (Whitchurch and Mattick, 1994a). Two of the mutants were sensitive to all four phages, while the third mutant, designated S40, demonstrated a differential pattern of phage sensitivity. S40 is resistant to Pfl, and infection of S40 with P04 resulted in smaller, more turbid plaques than those formed by the parent strain (Whitchurch and Mattick, 1994a).
Other than this differential sensitivity to pilus-specific phages, the pilU mutant phenotype was similar to that displayed by the characterized pilT mutants: hyperpiliated and unable to twitch. However, the fact that their pilT mutants displayed a resistance to all pilus-specific phages tested prompted
130 131
Whitchurch and Mattick (1994a) to conclude that PilT and PilU perform separate functions in pilus retraction, despite their high degree of homology (65% similar, 39% identical) and similar mutant phenotypes.
However, the isolation of additional p ilT mutants revealed that mutations in this gene also displayed the differential sensitivity recently observed with other pil mutants. In an attempt to better understand the relationship between pilT and pilU , I constructed and analyzed two pilU mutants. The goals of these studies were to:
(1) clone pilU from P. aeruginosa PAOl and determine if pilU is capable of suppressing the HOD1 pilT frameshift mutation,
(2) construct a pilU mutant in the PAOl background and determine
if the pilT gene is capable of suppressing this mutation, and
(3) construct and characterize a pilT pilU double mutant of PAOl. 132
Results
Introduction of pilU into the pilT mutant HOD1 Since PilT and PilU are homologous (65% similar and 39% identical) and their functions appear to be quite similar, I wanted to determine if either protein, when present at higher than normal levels, could be substituted for the other. I initiated these studies by determining if either of my existing pilT mutants could be suppressed by multiple copies of the pilU gene. Dr. J. S.
Mattick (personal communication) provided me with a restriction map of the putative 1.3 kb pilU gene (Appendix D) and its location with respect topilT. The 22 kb PAOl H indlll library clone which I had used to clone pilT (refer to Chapter III) also contained the entire pilU locus. Plasmid pPT2937, which was constructed during the HOD1 complementation studies, contains a 3.0 kb Pstl-BamHI subclone of the 22 kb H indlll fragment which includes pilU in addition to pilT. The pilU gene was removed from this plasmid as a 1.5 kb
Kpril-BamYll fragment and ligated into the K pnl and Bam H l sites of the shuttle vector pUCP18, forming pPT3813. Plasmid pPT3813, which contained the pilU gene in the same orientation as the lacZ promoter, was introduced into the pilT mutant HOD1. Neither phage sensitivity nor twitching motility was restored to HOD1 by pPT3813. According to Schweizer (1991), plasmid pUCP18 is present in P.aeruginosa at 10 to 25 copies per cell. Therefore,pilU is not able to suppress this pilT frameshift mutation in trans, even when 10 to 133
25 additional copies of pilU are introduced into HOD1. The pilU gene inpPT3813 is expressed since subsequent studies (described below) revealed that this plasmid is capable of complementing apilU mutant.
Construction and complementation analysis of a pilU mutant To determine whether pilT is capable of complementing a pilU mutant, I needed to first construct a pilU mutant of PAOl. A mutation in the PAOl pilU gene was generated using a gene replacement system designed by
Schweizer (1992). This system employs a mobilizable ColEl-based plasmid oriT which cannot replicate in Pseudomonas and which carries the counter- selectable B. subtilis sacB marker, allowing for the selection of double recombinants. The oriT and sacB reside on a portable mobilization (MOB) cassette, which also contains a gene encoding resistance to chloramphenicol.
The steps followed in the construction of the pilU gene replacement plasmid, pPT3547, is summarized in Fig. 17. Two alterations were made in pilU during the construction of pPT3520: a 0.55 kb fragment was deleted from the middle of the pilU gene, and the tet gene was inserted into pilU so that tetracycline-resistance could be used to select for single recombinants. The two remaining regions of the pilU gene which flanked the tetracycline- resistance marker were each approximately 0.5 kb in size. DNA fragments of this length have been successfully used for recombination by other investigators in this laboratory who have utilized the Schweizer gene replacement system (A. Darzins and M. Russell, personal communications). Figure 17. Cloning scheme followed in the construction of the pilU gene- replacement plasmid, pPT3547. ThepilU gene was removed from pPT2937 as a 1.57 kb Kpnl-BamHl fragment, the ends were made blunt, and the fragment
was ligated into the blunted Afar I and S p h l sites of the gene replacement vector pNotl9 (Schweizer, 1992), forming pPT3519. Plasmid pPT3519 was
then digested with N arl and Sphl, which cut within the pilU open reading frame, and the overhanging ends were blunted. The tetracycline-resistance marker from pBR322 was then ligated into the Narl and Sphl sites as a 1.4 kb EcoRl-Aval blunt-ended fragment, producing pPT3520. The final cloning step involved the introduction of the MOB cassette, removed from pMob3 as a 5.8 kb Notl fragment, into the Notl site of pPT3520. This final plasmid, pPT3547, was introduced into PAOl and HOD1 to generate the pilU gene replacement mutants of PAOl and HOD1.
134 NoU 135
Ap pNotlS Seal
Narl pPT 2937 Narl laeZa P»U pilV 7.57 k b
pUT Digest with 'Sphl PIm t BamBl + Kpnl p ilU (blunt) BamHI‘ Kpnl
/ Digest with / Narl + Sphl EcoRI
— < “ Ap To Digest with pBR322 EcoRI+Aval (blunt) NoU 4.38 k b Aval
p P T ssao
NoU Digest pMobS with Notl
remove MOB cassette as Notl fragment «acR
pPT3547
Figure 17 Nod' 136
Following the insertion of the MOB cassette into pPT3520, restriction endonuclease mapping studies revealed the orientation of the MOB cassette and that the pilU and the tetracycline-resistance genes were both in the opposite transcriptional orientation of the plasmid-encoded 6-lactamase gene (Fig. 17).
Plasmid pPT3547 was mobilized into PAOl from E. coli DH5a using a
variation of the standard triparental filter-mating procedure described in
Chapter II. Approximately 20-fold more donor cellsE (. coli DH5cx/pPT3547)
were used for this particular mating in an attempt to increase the number of recombinant Pseudomonas colonies. The resulting tetracycline-resistant colonies were single-crossover recombinants in which the suicide plasmid
pPT3547 had been integrated into the PAOl chromosomal pilU gene. Eighty- two of these colonies were patched onto PIA agar containing tetracycline and 5% (wt/vol) sucrose. Since the sacB gene is lethal in Gram-negative bacteria that are grown in the presence of sucrose, only recombinants that had excised the plasmid, while maintaining thepilU::tet insertion mutation, could grow on this medium (Schweizer, 1992). Fifty-eight (71%) of the colonies patched grew on this medium, and I restreaked two colonies in order to confirm their ability to grow in the presence of the sucrose. 137
One of the colonies was designated PAO-PU, and I used Southern analysis to confirm that recombination had occurred at the pilU locus on the chromosome. Chromosomal DNA preparations that had been isolated from PAOl and PAO-PU were digested with 2?coRI or H indlll and were separated by agarose gel electrophoresis. The radiolabelled 1.5 kb EcoBl-BamHl pilU fragment (Appendix D) was hybridized to the genomic DNA preparations using Southern analysis (Fig. 18). The wild type pilU gene is located on a 2.4 kb jEcoRI fragment that lies within a 22 kb H indlll fragment. The insertion of the tet cassette within pilU was expected to increase the size of this EcoRI fragment to 3.27 kb. Furthermore, the mutated pilU gene contains an additional HindlU cleavage site that is located in tet; thus, digestion of PAO-PU DNA with Hindlll would be expected to result in two pilU regions flanking the tet gene and result in pilU sequence residing on both 20 kb and 2 kb fragments. I observed the expected restriction patterns and concluded that PAO-PU is an isogenic pilU gene replacement mutant of PAOl. Since I deleted 0.5 kb from the internal portion ofpilU in addition to inserting the tet gene, I expected PAO- PU to be a null mutant of pilU. PAO-PU was not able to twitch and was sensitive to the phages D3112 cts, F116L cts 53, and B3 cts-3. The 1.5 kb Kpnl-BamKl wild typepilU fragment was ligated into the similarly digested plasmids pUCP18 and pUCP19, forming pPT3813 and pPT3753, respectively.
The pilU gene was inserted in the same transcriptional orientation as the lacZ promoter in pPT3813 and in the opposite orientation of this promoter in pPT3753. Both plasmids restored twitching motility to PAO-PU. Figure 18. Southern hybridization analysis of PAO-PU. Chromosomal DNA isolated from P. aeruginosa PAOl and the pilU mutant PAO-PU was probed using the radiolabelled 1.5 kb Kpnl-BamHl fragment containing the pilU gene. Lanes include genomic DNA from PAOl digested with JEcoRI (1) or H indlll (3), and from PAO-PU digested with Ecoffl (2) or H indlll (4). A DNA molecular weight ladder was included on the agarose gel but is not shown. Sizes (kb) of the bands are indicated.
138 139
12 3 4
22 kb 20 kb
K ' ■*
3.27 kb -
2.4 kb - - 2.0 kb
Figure 18 140
To determine if multiple copies ofpilT could suppress the pilU mutation in PAO-PU, a plasmid-borne copy ofpilT was introduced into PAO-PU. The plasmid pPT3738 consisted of the 1.5 kb Pstl-Kpnl pilT fragment ligated into the Pstl-Kpnl sites of pUCP19. Carbenicillin-resistant transformants of PAO- PU harboring pPT3738 were assayed for twitching motility and phage
sensitivity. All transformants of PAO-PU/pPT3738 which were assayed remained incapable of twitching and were sensitive to the phages D3112 cts, F116L cts 53, and B3 cts-3. Thus, multiple copies of p ilT are not able to suppress the pilU mutation.
Construction and complementation analysis of a pilT pilU double mutant The pilU mutant phenotype was quite revolutionary with respect to its implications on the current theory of retraction and phage sensitivity since this phenotype was the first to unlink twitching motility and phage sensitivity. It is possible, however, that the true phenotype of a pilU m utant is the same as that of a pilT mutant (nontwitching, phage-resistant), and that PilT in the pilU mutant was partially masking the pilU mutation by allowing phage sensitivity. To try to answer this question, I constructed and analyzed a pilT pilU double mutant of PAOl to determine if the lack of functional PilT resulted in a different phenotype for thepilU mutant. 141
The pilT pilU double mutant was constructed by introducing the pilU gene-replacement plasmid pPT3547 into thepilT mutant HOD1 following the same procedure described above for the construction of PAO-PU. Thus, the wild typepilU allele was replaced with the mutated pilU'.tet fragment, forming the double mutant PAO-TU. Southern hybridization was performed using the same probe and restriction enzymes described above, and the results (Fig. 19) confirmed that the pilU gene in HOD1 had been replaced with the mutated pilUitet fragment.
The phenotype of thepilT pilU mutant strain is the same as that of a pilT mutant: resistant to the phages D3112 cts, F116L cts 53, and B3 cts-3 and unable to twitch (Table 9). Complementation studies were performed with
PAO-TU using the pUCP-based plasmids generated for the studies described earlier. The pilT and pilU genes were introduced separately into PAO-TU on plasmids pPT3738 and pPT3813, respectively. Plasmid pPT3738 restored phage sensitivity but not twitching motility to PAO-TU,thus complementing the p ilT mutation but not the pilU mutation (Table 9). Plasmid pPT3813 restored only phage sensitivity to the double mutant, thereby complementing only the pilU mutation. When the 3.0 kb Psfl-BamHI fragment containing both pilT and pilU (Appendix D) was introduced into the mutant on plasmid pPT2929, PAO-TU displayed the wild type phenotype with respect to twitching and phage sensitivity (Table 9). Figure 19. Southern hybridization analysis of PAO-TU. Chromosomal DNA isolated from P. aeruginosa HOD1 and the pilT pilU mutant PAO-TU was probed using the radiolabelled 1.5 kb Kpnl-BamHl fragment containing the pilU gene. Lanes include genomic DNA from HOD1 digested with EcoRl (1) or Hindlll (2), and from PAO-TU digested with HcoRI (3) or H indlll (4). Sizes
(kb) are shown next to each fragment. DNA molecular weight markers were included during electrophoresis but are not shown.
142 143
12 3
22 kb - - 2 0 kb
mm 3.27 kb 2.4 kb - 2.0 kb
Figure 19 Table 9. Phenotypic analysis of P. aeruginosa pilT and pUU mutants.
Mutant Phenotype with Phenotype with Phenotype with Mutant Phenotypes pPT3738b pPT3813c pPT2929d
HOD1 Tw- phageR Tw+ phageS Tw- phageR Tw+ phageS
PAO-PU Tw- phageS Tw- phageS Tw+ phageS Tw+ phages
PAO-TU Tw- phageR Tw- phageS Tw- phageR Tw+ phages a Abbreviations: Tw+, twitching; Tw-> nontwitching; phageR, resistant to D3112, B3, and F116; phageS, sensitive to D3112, B3, and F116. b pPT3738 contains the PAOl pilTgene c pPT3813 contains the PAOlpilU gene d pPT2929 contains the PAOlpilT and pilU genes 145
Discussion
I have constructed a pilU null insertion mutant, in addition to a pilT p i l U double mutant, and characterized both with respect to twitching motility and phage sensitivity. The phenotype of thepilU mutant, in
the PAOl background, correlates with that observed by Whitchurch and Mattick (1994a) using PAK pilU mutants. Although neither mutant was viewed by electron microscopy, both are presumed to be hyperpiliated on the basis that their pili are not retracting and that the p ilU mutants of Whitchurch and Mattick were hyperpiliated. The results indicate that neither pilU or p ilT is capable of suppressing a mutation in the other, even when present in multiple copies. This supports the hypothesis of Whitchurch and Mattick (1994a) that PilT and PilU perform similar but separate functions in pilus retraction.
The pilT pilU double mutant constructed in this study displays the same phenotype as ap ilT mutant. Because the pilT mutant phenotype masks that of a pilU mutant, it is impossible to determine from the studies described here if PilT has any effect on pilU expression or PilU function. One preliminary experiment indicated that apilU::luxAB transcriptional fusion confers luminescence to HOD1, suggesting that pilU is indeed expressed in a p ilT mutant (data not shown). However, I did not pursue this particular investigation and believe that more experiments of this nature should be 146
completed before a conclusion can be made. Antiserum raised against PilU would also be beneficial in determining if this protein is present in a pilT mutant.
The fact th at the 1.5 kb Kpnl-BamHl pilU fragment complements
PAO-PU suggests that this fragment contains a pilU promoter, thus, pilT and p ilU are probably not co-transcribed. However, it is possible that transcription ofpilU, even when inserted in the opposite orientation of thelacZ
promoter, is under the control of another plasmid-bome promoter. Whitchurch and Mattick (1994a) identified several potential inverted-repeat sequences in the p ilT pilU intergenic region which may function in termination of pilT
transcription. Additional studies that focused on the possibility that pilT and pilU might be cotranscribed are described in Chapter V. CHAPTER V
EXPRESSION OF PILT IS DIFFERENTIALLY REGULATED AT THE TRANSCRIPTIONAL LEVEL
Introduction
Although no other investigators were pursuing the role of the P. aeruginosa PilT protein, many were continuing to study the PilT homologs. I decided to focus my investigation on expression of the PAOl pilT gene. The cycling of pilus assembly and retraction that presumably occurs during twitching likely involves at least one type of signal transduction mechanism to control the expression of the necessary genes. Evidence for this hypothesis is based on the finding that expression of pilA is regulated by RpoN and the two- component sensor-regulator proteins PilS and PilR (Boyd et al., 1994; Hobbs et al.% 1993; Ishimoto and Lory, 1989, 1992). As discussed in Chapter I,pilA may be differentially expressed in response to an environmental signal that indicates a need for pilus assembly or retraction. This notion is supported by the knowledge that inhibition of pilus retraction somehow induces hyperpiliation (Bradley, 1972c), which suggests that pilA expression increases when retraction is prevented. Indeed, Johnson and Lory (1987) detected
147 148 increased levels of pilA transcription in P. aeruginosa hyperpiliated mutants. They proposed thatpilA expression increased in response to a depletion of pilin
(or another Pil protein) in the putative membrane pool of pilin due to the lack of pilus retraction. Unfortunately the mutations in these hyperpiliated strains were not characterized. It is unlikely, though, that the mutations were in pilT or pilU, because Whitchurch and Mattick (1994a) reported that levels of pilA transcript were decreased by 30-80% in pilT and pilU mutants. Taken together, these findings suggest that a complicated regulatory pathway must be involved to coordinate pilus assembly and retraction.
This segment of the project was directed toward the study of pilT expression in an attempt to further elucidate the role of this gene in pilus retraction. This set of experiments was designed to:
(1) determine if pilT is cotranscribed with pilU,
(2) map the pilT transcription initiation site to identity a putative pilT
promoter and regulatory region, and
(3) determine ifpilT expression is differentially regulated. 149
Results
Northern analysis of the pilT transcript Total cellular RNA was isolated from P. aeruginosa PAOl that had been grown under standard conditions in liquid LB medium and the RNA was examined for degradation using agarose gel electrophoresis. When I probed this RNA preparation with the 1.5 kb Pstl-Kpnl fragment that contains the entire p ilT gene, I was not able to detect the p ilT transcript. I repeated this experiment several times using freshly prepared RNA samples that had been isolated from various stages of growth, yet I still did not detect a signal. To confirm that I had performed the procedure correctly, I probed these blots with the PAOl lasA gene and detected the las A transcript (data not shown). I hypothesized thatpilT may only be expressed under the conditions that are
conducive to twitching motility. Since twitching is a surface-dependent motility (Bradley, 1980; Henrichsen, 1972, 1983), I repeated my Northern hybridization using RNA isolated from PAOl cells grown on LB agar medium.
A weak signal representing the 1.4 kb pilT transcript was present only in the lane containing the RNA isolated from agar-grown cells (Fig. 20). Figure 20. Northern hybridization analysis of pilT expression in PAOl. The 1.5 kb Pstl-Kpnl pilT probe was prepared by radiolabelling as described in Chapter II. The probe was hybridized to 20 |ig of RNA isolated from P. aeruginosa
PAOl grown either on solid (Lane 1) or in liquid (Lane 2) medium and fractionated on a 1.1% agarose gel containing 0.66 M formaldehyde. The 1.4 kb pilT transcript, which was detected only in RNA prepared from cells grown on solid medium, is denoted by the arrow at the left. The sizes of the molecular weight standards, in kb, are indicated to the right.
150 151
1 2
9.49 7.46 4.40
2.37 pilT 1.35
Figure 20 152 pilT expression does not require the alternative sigma factor RpoN
When Whitchurch et al. (1990) published the nucleotide sequence of pilT, they identified a sequence upstream of pilT which shared homology to bacterial promoters that require the alternative sigma factor RpoN (o54). The possibility that pilT expression is RpoN-dependent was strengthened by the result of the Northern hybridization, which suggested that pilT expression may be activated by an environmental stimulus associated with growth on solid medium. Indeed, RpoN-dependent promoters require activator proteins that respond to environmental or metabolic signals (Merrick, 1993).
PAK-N1 contains a null mutation in rpoN due to the insertion of the tet gene (Ishimoto and Lory, 1989). This mutant was confirmed to be a tetracycline-resistant, non-piliated, glutamine auxotroph prior to the Northern experiment described here. The 1.4 kb p ilT transcript was detected by Northern analysis in an RNA preparation from agar-grown PAK-N1 cells and in the control wild type strain, PAOl (Fig. 21). Thus, the expression ofpilT in
P. aeruginosa does not require the alternative sigma factor a54, and the putative promoter sequence identified by Whitchurch et al. (1990) was not the actual pilT promoter.
Identification of promoter activity in the pilT pilU intergenic region The 1.4 kb pilT transcript is apparently not large enough to also include the downstream pilU message. To confirm that the pilU gene is expressed Figure 21. Northern hybridization analysis of pilT expression in an rpoN mutant. The radiolabelled 1.5 kb Pstl-Kpnl pilT fragment was hybridized to
17 |ig of RNA isolated from agar-grown cultures of P. aeruginosa PAOl (lane 1) and the rpoN mutant PAK-N1 (lane 2). An RNA molecular weight ladder (not shown) was used to determine that the transcript was 1.4 kb. The pilT transcript, indicated by the arrow to the left, was present in both cultures.
153 Figure 21 155
from its own promoter, I used lacZ as a reporter gene to detect promoter activity in the pilT pilU intergenic region. The 0.4 kb Narl-Sacll fragment
representing the entire intergenic region in addition to 0.22 kb of the putative pilU open reading frame (Appendix D) was cloned into pUC18, forming plasmid pPT3341. This fragment was now flanked by convenient restriction
endonuclease sites which facilitated cloning into the lacZ promoter-probe vector pQF50 (Appendix A; Farinha and Kropinski, 1990) as HindUl-BaniHla fragment (pPT3344). This plasmid contained thepilU insert in the same
orientation as the lacZ gene. A second plasmid, pPT3496, was constructed in a similar fashion and contained the 0.4 kb Narl-Sacll fragment in the opposite orientation. PA01/pPT3344, PA01/pPT349 and PAOl/pQF50 were grown in
LB broth containing carbenicillin (50 pg/ml) and were assayed for B- galactosidase activity at 2 hr intervals. The 0.4 kb iVarl-SacII fragment demonstrated promoter activity when present in the same transcriptional orientation as the promoterless lacZ reporter gene (PA01/pPT3344; Fig. 22).
Interestingly, activity first appeared during the transition into stationary
phase. The presence of promoter activity in this region suggests thatpilU is
expressed from its own promoter and supports the results of the Northern hybridization which indicated that pilT and pilU are not cotranscribed.
Identification of the region required for pilT promoter activity To map the region of the p ilT promoter I used the promoter-probe vector pQF70, which contains the promoterless luxAB genes from Vibrio harveyi (Appendix A; Farinha and Kropinski, 1990). Because luciferase Figure 22. B-galactosidase activity of a pilU.lacZ transcriptional fusion in P. aeruginosa PAOl. Cultures were grown in liquid LB and assayed in triplicate, with data representative of one set of assays shown here. Activity values represent average ± standard deviation. Open symbols, absorbance at 600 nm; closed symbols, specific activity of J3-galactosidase. The cultures assayed were PAOl/pQF50 (circles), PA01/pPT3344 (squares), and PA01/pPT3496 (triangles).
156 M O 100 Specific Activity Absorbance (600 nm) o o o © i-» to o O 00 §• *1 H S3-
Figure 22 158
activity is easily detected in cells that are growing on solid medium, I chose to use luxAB as the reporter gene instead of lacZ. E. coli cells producing B- galactosidase are easily detected as blue colonies on solid medium containing X- gal (5-bromo-4-chloro-3-indoyl-B-D-galactopyranoside); however, this blue color is difficult to visualize in PAOl due to the presence of blue-green pigments produced by this bacterium.
The 1.0 kb Xhol fragment (Fig. 23A), which was likely to contain the pilT promoter since it included the 0.8 kb region upstream of the putative pilT
start codon, was the largest fragment used to construct the fusions. I removed this fragment from pPT2971 (Appendix B) by digestion with BarriHI-
H in d lll and ligated it into the BamHI-flmdlll sites of pQF70, forming pPT3018. PAOl/pPT3018 was stab inoculated into LB agar and assayed for luciferase production as described in Chapter II.
PAOl/pPT3018 was luminescent within several minutes of the addition of the substrate decanal. This luminescence was detectable in a dark room, as well as on X-ray film. Smaller regions of this 1.0 kb Xhol fragment were
systematically tested for promoter activity in PAOl (Fig. 23A). Because these assays were not quantitative, I simply recorded the presence or absence of luminescence for each strain. The 0.3 kb Pstl-M lul fragment (in pPT3319) was the smallest fragment to emit light. This Pstl-Mlul fragment (pPT3319) also displayed promoter activity inE. coli DH5a (Fig. 23B). Figure 23. Mapping of the P. aeruginosa pilT promoter region using the luxAB reporter genes. The restriction maps (Panel A) indicate thepilT fragments that were ligated into the promoter-probe vector pQF70 (Farinha and
Kropinski, 1990) to form pilT::luxAB transcriptional fusions. These plasmid fusions were introduced into P. aeruginosa PAOl and examined for luminescence. Luciferase activity was detected by the ability of the fusion strain to expose X-ray film which had been placed under the petri dish during the assay (Panel B). Cultures in Panel B: 1, PAOl/pQF70; 2, PA01/pPT3245;
3, PAOl/pPT3489; 4, DH5ot/pQF70; 5, PAOl/pPT3016; 6, PA01/pPT3323; 7,
PA01/pPT3319; 8, DH5a/pPT3319. Each fusion was classified simply as luminescent (+) or non-luminescent (-) (Panel A). Abbreviations for restriction endonuclease sites (Panel A): X, Xho I; P, Pstl; M, Mlul; S, Sspl.
159 Relative P lasm id Fragment used to construct luxAB fusion Luminescence
X P S M I______I______I_____ L_ piiT
X PPT 3016
P X PPT3245______i______j
PPT 3323
M PPT3489
P M p P T 3319 i i
100 bp
Figure 23 Figure 23 (Continued)
B.
3 4
7 8 • •
03 162
I also assayed E. coli cells containing a luxAB transcriptional fusion to the pilT pilU intergenic region. The inability of this strain to luminesce indicated that the putative pilU promoter is not recognized in E. coli DH5a
(data not shown). Not only does this result support other data which indicated that p ilT and pilU possess unique promoters, but also it suggests that the expression of these genes requires different transcription factors.
Quantitative analysis of pilT expression on solid and liquid media A liquid culture of PA01/pPT3319 was visually luminescent as were each of the other pilT constructions which had displayed promoter activity on solid medium. When a luminometer was used to quantitate luminescence, the results were extremely erratic and rather inconclusive. Because pilT promoter activity was not readily quantifiable as fusions to luxAB, I constructed a pilT::lacZ transcriptional fusion by ligating the 0.3 kb Pstl-Mlul pilT promoter fragment into pQF50. This plasmid, pPT3493, contained thepilT promoter in the same orientation as the promoterless lacZ gene. The 0.3 kb Pstl-Mlul fragment was also ligated into pQF50 in the opposite orientation, forming pPT3642, which was used as a negative control. I introduced these plasmids into PAOl and performed £-galactosidase assays on these cultures.
Assays of LB broth-grown cultures verified that promoter activity was evident in the 0.3 kb Pstl-M lul fragment (Fig. 24): fi-galactosidase activity Figure 24. B-galactosidase activity in a pilTilacZ transcriptional fusion in P. aeruginosa PAOl. The pilT::lacZ fusion was constructed using the 0.3 kb Pstl- M lul fragment thought to contain the pilT promoter on the basis of studies using pilT::luxAB fusions. Cultures were grown in liquid LB and assayed in triplicate, with data from one experiment shown. Activity values represent average ± standard deviation. Open symbols, absorbance at 600 nm; closed symbols, specific activity of B-galactosidase. The cultures assayed were PAOl/pQF50 (circles), PA01/pPT3493 (squares), and PA01/pPT3642 (triangles).
163 Absorbance (600 nm) 0.01 11 c .1 0 10 4 0 Tim e (hr) e Tim Figure 24 Figure 8 12 16 20 50 200 100 250 300 150 350 400
Specific Activity 164 165 was found exclusively in the PA01/pPT3493 culture. Also noteworthy was that the appearance of activity in stationary phase was similar to the pattern seen for the pilU promoter. After growth on solid and liquid LB medium, the levels of enzyme activity in PA01/pPT3493 were unexpectedly similar in both cultures (Fig. 25). The luciferase assays had indicated that the pilT promoter was active in liquid cultures, but because of the Northern hybridization result, I had expected the p ilT promoter to be less active in liquid culture. This discrepancy could be due to a titration effect as a consequence of there being approximately 13 copies of the fusion plasmid per genome (Farinha and Kropinski, 1990). Alternatively the 0.3 kb Pstl-Mlul fragment may contain the pilT promoter but lack potential regulatory regions which are responsible for the differential expression ofpilTin PAOl.
To determine if the region of DNA between the Mlul site and the start of the pilT gene was required for the differential expression of pilT, I constructed a fusion of the 0.59 kb Pstl-Xhol pilT fragment with lacZ (pPT3973). A limited number of B-galactosidase assays of PAOl/ pPT3973 growing on both solid and liquid medium showed that the specific activities were not affected by the growth medium (data not shown). Although these results were obtained from one preliminary set of assays, they suggest that the lack of regulation seen with the plasmid fusions could be due to the titration of a transcriptional factor that is required for the differential expression of p ilT . To demonstrate Figure 25. Comparison of the pilT promoter activity in P. aeruginosa PAOl grown on solid and liquid media by using a pilT:lacZ transcriptional fusion. The pilT promoter fragment used in this plasmid fusion (pPT3493) is the 0.3 kb P stl-M lul fragment. The culture was grown either in liquid LB or on nitrocellulose filters placed on LB agar and was assayed in triplicate, with data from one experiment shown. Activity values represent average ± standard deviation. Open symbols, absorbance at 600 nm; closed symbols, specific activity of fi-galactosidase. Circles signify the PA01/pPT3493 culture grown on solid medium, and squares denote PA01/pPT3493 grown in liquid medium.
166 167 to Ol o to o H* cn H* o o SpecificActivity Absorbance run) (600 cn o © o M o to 00 o f r ft
Figure 25 168
this, I attempted to insert a single-copy of a pilT::lacZ transcriptional fusion into the PAOl chromosome. Unfortunately, I was not successful in my
numerous attempts to construct a pilT::lacZ chromosomal fusion.
At this point, the regulated expression ofpilT was not clear: the pilT
transcript was detected exclusively in cells grown on solid medium, but pilT promoter activity was equivalent in liquid and solid media. To try to resolve this conflict, I used low resolution SI nuclease protection to quantitate the pilT
transcript in P. aeruginosa PAOl cultures grown on solid and liquid LB. Total cellular RNA was isolated from these cultures during exponential growth, at the transition between exponential and stationary phase, and after several hours in stationary phase.
The amount of pilT transcript increased during growth (Fig. 26) and was
most abundant during stationary phase. The amount of each protected fragment in the remaining samples was normalized to that in RNA that was isolated from stationary phase cells on solid medium (Fig. 26). These values
revealed that the pilT message is consistently present at an approximately three-fold greater level in cells grown on solid medium.
Mapping of the transcription start sites of pilT and analysis of the transcript populations The sequence of the pilT promoter might reveal information regarding the regulation of pilT transcription. For example, the promoter region may 169
Figure 26. The effect of the growth medium on the level of the pilT transcript in P. aeruginosa PAOl. Low resolution SI nuclease analysis was used to detect the pilT message in RNA isolated from cells grown on solid or liquid LB medium. The RNA samples were isolated at several times during growth and stationary phase, as indicated above each lane. Each reaction was performed with 100 }ig of RNA, and the entire reaction was loaded onto the gel. The Instant Imager 2024 (Packard Instrument Co.) electronic autoradiography measured the counts per minute (CPM) of each protected DNA fragment, and the relative amounts are presented below the figure. solid liquid medium medium
•&P jS ^
m m
Relative CPM: 0.65X 0.97X 1.00X 0.19X 0.32X 0.33X
Figure 26 contain a consensus sequence recognized by a sigma factor other than RpoN. To identify the transcription initiation site ofpilT and thus identify a potential p ilT promoter sequence, I first used low resolution SI nuclease protection
analysis. Experiments with pilT::luxAB transcriptional fusions revealed that the pilT promoter resided on the 0.3 kb Pstl-Mlul fragment, near the Sspl site. Hybridization of the 0.59 kb Pstl-Xhol DNA fragment to the pilT message,
followed by digestion with SI nuclease, would result in a protected fragment predicted to be between 300 and 450 nt in size. Indeed, a protected fragment of approximately 450 nt was detected (Fig. 27). Consistent with my earlier results, the protected fragment was present at a higher level in the reaction which contained RNA from cells grown on solid medium compared to that seen with RNA from a broth culture.
High-resolution SI nuclease protection assays definitively mapped the 5' end of the pilT transcript. Two transcription start sites differing in location by 5 nt were detected for pilT using RNA isolated from surface-grown cells.
(Fig. 28). These same 5 ' ends were mapped for the transcript in liquid-grown cells, and results were reproducible using oligonucleotide 146. The two pilT transcription start sites are located 211 and 216 nt upstream of the putative p ilT start codon (Fig. 29), resulting in unusually long untranslated leader regions. High-resolution SI nuclease analysis also revealed that the two transcription initiation sites for a plasmid-borne copy ofpilT were the same in
E. coli DH5a as in Pseudomonas. This indicates that the luciferase activity of
171 Figure 27. Low resolution SI nuclease mapping of the approximate location of
the pilT transcription initiation site. The 590 bp Pstl-Xhol fragment, which was radiolabelled at the Xhol 5' overhanging end (*), was hybridized to 50 |ig of RNA isolated from exponentially growing cultures of P. aeruginosa PAOl. The full-length, undenatured probe was detected in RNA isolated from LB agar- grown (A) and LB broth-grown (B) cultures. A 450 nt protected fragment was detected in both samples, although it was present at a very low level in the broth sample (B). The size of the protected fragment indicated that the pilT transcription initiation site was near theSspl restriction endonuclease site.
172 173
Xhol PstL Sspl Mful I I I___ vS/SysSs///< sS/SffSSA J/fa/s/f////SSfSsysfssfSJSi'S/*?s^f/J$ffi
* 590 bp DNA probe
MW B 9 4 3 - 5 8 5 - probe protected fragment 3 4 1 -
258-
153 —
Figure 27 Figure 28. High resolution SI nuclease mapping of the transcription initiation sites of P. aeruginosa pilT. Oligonucleotide 85 (Appendix C) was used as a PCR primer to synthesize a DNA fragment that was complementary to thepilT transcript. The fragment was labelled at the downstream end and hybridized to 100 |ig of RNA isolated from PAOl grown on LB agar. This same oligonucleotide and pPT3244 were used as the primer and template, respectively, to generate a sequence ladder (A, C, G, T) of the antisense strand. Two protected fragments, which differ in size by only 5 nt, were detected and are identified as T1 and T2. The sequence of the sense strand is given at the left, with the two underlined nucleotides representing the 5' ends of each protected fragment.
174 Figure 28 Figure 29. Nucleotide sequence of the region upstream of pilT. The putative pilT translational start codon (Whitchurch et al., 1990) is designated by the abbreviation Met . The 5' ends of the two transcripts are underlined and denoted as +li and +I2 . The -35 and -10 regions of a putative o 70-like promoter sequence are identified by the filled gray boxes. The putative ct54- dependent promoter (GG-N10-GC) which was identified by Whitchurch et al.
(1991) is doubly underlined. A second a70-like promoter sequence (underlined with filled circles) is present immediately upstream ofpilT , although there is no evidence that this sequence functions as a promoter. A potential regulatory region consisting of imperfect inverted 15 nt repeats and occurring upstream of the pilT promoter region is designated with converging half arrows. The 2 nt that were deleted from the inverted repeats in pPT3736ASacII are underlined with dots and a delta (A). Also shown are the nt complementary to oligonucleotides 85 (dashed box) and 146 (solid box), which were used for the high resolution SI nuclease protection studies.
176 P a t I ^ ^ S a c II 65 CTG CAGGT AGTTT TCGCC GAAGT CGCGA AGGC CGGC£GCGTG CGCCT CG£GC ACCGC GGCGG CGG T
130 GCT TGGTC TTGCT CACGG CGAGC AGGCC GACC GTGGC CGGAT CGCGC CCCGC AGCTT GCGCTGCC
-3 5 S s p l -10 +li +I2 195 TCACGGAT GCGCG TGCAATATT CTCTG C ^ M ^ g GGA C ATTA P.P.TGP GCCCT ATG
260 GAAGTGCG GTCGG CGTCG GCTCG ACGCC CTCGAAACC GCCGG CCTCG ATdGC CATCT TCCTG TTC ...... f ...... ■n'li**" "iiiwir^^iwinnnrrruvunwi
Mlu I 325 CAGTGjrCG AGGTT CCGCG TCTGG CGCAG GACACCACG CGTGT CC3|GGAAGCACCCTGGCTCATCb
390 GGT GTTTT CCTTG TCCGA GACGG CGGCT TTGG CGGCATTCTA CCTGC TTGCT TGTAA TTGGG GAG • ••••• •••••«
p i.1 T - - > TCCCATG Met
Figure 29 178 the pilT::luxAB fusion in E. coli DH5a (reported above) was due to the pilT promoter and not to an arbitrary DNA sequence which was recognized as a promoter inE. coli.
Although I identified two pilT transcript initiation sites, I could only discern one putative promoter sequence that was present immediately upstream of the two transcription start sites (Fig. 29). This promoter sequence was identified on the basis of its homology to the consensus sequence of the o70-like promoters found in streptomycetes and pseudomonads
[TTGAC(Pu) - TAT(Pu) (Pu)T] (Deretic et al., 1989; Strohl, 1992; L. O’Donnell, unpublished data). The putative pilT promoter identified here has a near consensus -10 sequence, but the -35 sequence is not well conserved. Even so, there are no sequences that resemble other classes of Pseudomonas promoters in this region, including those recognized by the 1995), and the aE-analog AlgU (environmental stress; Deretic et al., 1994; Hershberger etal., 1995). I also eliminated the possibility that the motility sigma factor RpoF (cy28;FliA) is required for pilT expression, since Stambach and Lory (1992) reported that piliation of afliA mutant is normal. There is a sequence that correlates well with the o70-like promoter consensus sequence and which is immediately upstream of the putative 179 ribosome binding site of pilT. However, all of the data presented in this study indicate that this sequence does not function as a promoter for pilT. Low- resolution SI nuclease analysis (presented above) was carried out using a DNA probe that spanned the region containing this second promoter-like sequence and no transcription initiation sites were detected in this region. In addition, Whitchurch et al. (1990) have isolated mutants that contain transposon insertions in the region that I have identified as the UTL. These mutants display the phenotype of apilT mutant, suggesting that this region is required for p ilT expression. Furthermore, the fact that the upstream promoter spans theS sp l site can be used to explain why I did not detect promoter activity in transcriptional fusions with either the 0.16 kb Pstl-Sspl fragment or the 0.14 kb Sspl-Mlul fragment. Since I knew that the total amount of pilT transcript increased during growth, I decided to use high resolution SI nuclease analysis to determine if the levels of both pilT transcripts increased during growth. The amount of the longer pilT transcript (Tl) increased during growth on solid medium, although it was difficult to determine if the amount of Tl continued to increase once the culture had reached stationary phase (Fig. 30). The shorter transcript (T2) was most abundant during exponential growth (Fig. 30). I obtained similar results in subsequent experiments using RNA isolated at equivalent stages from broth-grown cells. The relative levels of each transcript could not be quantitated because the signals were too weak to be detected accurately by an electronic imager. Figure 30. High resolution SI nuclease analysis of the two pilT transcripts during and after growth on LB agar. Oligonucleotide 85 was used to synthesize the DNA probe fragment and the sequence ladder (A, C, G, T). Arrows to the right of the autoradiograph denote the longer (Tl) and shorter (T2) transcripts. The sequence of the sense strand is given at the left with the two underlined nucleotides representing the transcription start sites. Reactions were carried out with 100 pg of RNA isolated from mid-exponential growth (Lane 1), late- exponential growth (Lane 2), transition (Lane 3), and stationary phase (Lane 4) cultures. 180 onoHnn>HHi>n> C T 1234ACGT Figure 30 Figure 181 182 Identification of potential regulatory regions upstream of pilT I searched for potential regulatory sequences upstream of the pilT promoter and in the sequence encoding the 5' untranslated region of the mRNA. The Genetic Computer Group (GCG) sequence analysis software package (University of Wisconsin) identified thirteen potential stem-loop structures in the untranslated leader sequence (data not shown). In addition, there are multiple regions that could be viewed as direct repeats. The only potentially significant sequence that I identified upstream of the putative -35 promoter sequence is a 15 bp inverted repeat (Fig. 29). Within the inverted repeat there is a S adi restriction endonuclease site that I exploited to mutate this region (Fig. 29). I digested plasmid pPT3315 (Appendix B) with SacII, which cut this plasmid only at the site within the inverted repeat, and removed the 2 bp 3' overhang using T4 DNA polymerase. After I religated the plasmid and introduced it into E. coli by transformation, I redigested minipreparations from the transformants with SacII to confirm that the site had been altered. I removed this mutated fragment from pPT3315 as a 0.59 kb Pstl-Xhol fragment and ligated it into the Pstl-Xhol sites of pPT3736, thereby replacing the wild type pilT inverted repeat with the deleted version. I introduced this plasmid into HOD1 and assayed these transformants for twitching motility. Plasmid pPT3736ASacII complemented the pilT mutation and restored twitching motility to HOD1, indicating that deletion of these 2 bp does not eliminate pilT expression. 183 Discussion The P. aeruginosa PAOl pilT gene is expressed independently ofpilU and likely requires a different transcription factor. This conclusion supports the hypothesis of Whitchurch and Mattick (1994a), who proposed that PilT and PilU, although homologous, do maintain separate functions in twitching motility. The results that I have described here also strongly suggest that expression ofpilT is regulated at the transcriptional level. The amount of the pilT message is approximately three-fold higher throughout growth on solid LB medium relative to the amount detected in a broth-grown culture. It is tempting to speculate that this represents differential expression ofpilT, which is regulated either by a transcriptional activator on solid medium or by a repressor in liquid medium. However, there is the possibility that the p ilT transcript is simply less stable under liquid growth conditions. Since I could not integrate a single copy of a pilT::lacZ transcriptional fusion into the PAOl chromosome, I do not know whether the pilT promoter is differentially regulated. The apparent titration effect seen with plasmid-bome pilT::lacZ fusions, however, suggests that a transcription factor is required for the differential expression of pilT. The non-consensus -35 sequence also suggests the requirement for a transcription factor. There is precedence for transcriptional regulation in response to growth on solid medium. In Vibrio parahaemolyticus, differential regulation results in a 184 switch in the type of motility used by the bacterium. The polar flagellar system in this organism mediates swimming in liquid medium, whereas the lateral flagellar system mediates swarming during growth on surfaces and in viscous environments (McCarter et al., 1988; McCarter and Wright, 1993) Despite the existence of this similar regulatory system in V. parahaemolyticus, further studies of pilT expression should probably analyze the level of PilT protein in cells grown under the conditions used here. If pilT is differentially expressed at the transcriptional level, as I have hypothesized, then the concentration of PilT is expected to be approximately three times higher in cells grown on LB agar. I have studied p ilT expression using only LB medium, but a better understanding of the environmental stimulus (or stimuli) that influences pilT expression could be obtained by studying pilT expression during growth on other media. Physical, chemical, and hydrodynamic conditions at the solid- liquid interface on the surface of a solid medium are quite different from those in liquid culture (Fletcher, 1991). Bacterial cells may twitch in response to nutritional levels, nitrogen or oxygen concentrations, water availability, viscosity, or a number of other stimuli (Fletcher, 1991; Henrichsen, 1972, 1983; McEldowney and Fletcher, 1988). As discussed in Chapter I, early studies on several twitching strains of Acinetobacter calcoaceticus revealed that the most important factor for twitching is the amount of liquid available at the 185 agar surface, since conditions affecting this factor influenced twitching (Henrichsen, 1975b,c). Similar studies could be used to determine if pilT expression is affected by water availability. Perhaps a better understanding of the environmental stimulus(i) affecting twitching motility will be gained from studies of pilG,H,I,J,K,Ll, and L2 (Darzins, 1993, 1994, 1995; J. Mattick, personal communication). These genes encode proteins that are homologous to proteins in the enteric chemotactic signal-transduction pathway and in the ‘frizzy’ motility system of Myxococcus xanthus (Darzins 1993, 1994, 1995; J. Mattick, personal communication). The homology displayed by these Pil proteins to the Che and Frz proteins strongly suggests that a similar signal transduction system exists in P. aeruginosa which may be responsible for controlling pilus function and regulating twitching motility. The nature of the signal that this proposed pathway responds to is not known (Darzins, 1993-1995). The ‘frizz/ motility system of M. xanthus controls the direction of movement, and several mutants have been identified which form doughnut-shaped swirls at low cell densities (Zusman, 1982). Interestingly, Darzins (1994) reported that a pilH mutant, which remains capable of twitching, demonstrated a similar motility pattern when observed microscopically. Although these are certainly preliminary results, cell density should be considered a possible factor affecting twitching motility. Henrichsen (1975a,b) reported that twitching requires a minimum number of cells and occurs at the 186 leading edge of growth, where the cells are present in only a single layer (Bradley 1980; Darzins 1994; Henrichsen 1983). This phenomenon may be due more to cell growth than to cell density, since Henrichsen reported that only growing cells display twitching motility (1975a). However, the finding that pilT expression increases as cultures enter stationary phase suggests that twitching motility is occurring during stationary phase. This hypothesis correlates well with the idea that cell density may contribute to twitching motility. The 5 ' untranslated leader (UTL) of the pilT message is unusually long (ca. 215 nt), which suggests that it may be involved in regulation of pilT expression. Unfortunately, no hypotheses for the function of the pilT UTL can be proposed on the basis of conserved sequences since a BLAST search did not reveal any homology between the p ilT UTL and the UTL of other genes. However, the characterized functions of other UTLs should be kept in mind as possible roles for the pilT UTL. For example, the long (133 nt) 5 ' UTL of the E. coli ompA transcript functions as a growth-rate-regulated transcript stabilizer (Chen et al., 1991). The transcripts of the P. aeruginosa alginate biosynthesis genes algC and algD also exhibit long UTL (244 nt and 367 nt, respectively) (Fujiwara and Chakrabarty, 1994), although the functions of these two UTL are not known. There is, however, evidence that the algC UTL is important for translation efficiency (Fujiwara and Chakrabarty, 1994). The UTL of the pilT transcript may prove to have a function similar to one of those mentioned here. 187 My observation of two growth-phase dependent p ilT transcripts suggests that a second level of regulation is involved in p ilT expression, in addition to the differential expression observed on solid medium. It is possible that the smaller transcript is simply a processed form of the longer one and is not an independent transcript. However, the spacing between the putative promoter sequence and the second transcription start site is better conserved than that for the start of the longer transcript (W. Strohl, personal communication). One way to confirm that there are two transcription initiation sites would be to utilize the capping enzyme guanylyltransferase, which would cap both p ilT transcripts if they possess a triphosphate at their 5' termini. I surmise that p ilT encodes two independent transcripts. The SI nuclease mapping results were consistent and reproducible with multiple RNA samples, suggesting that degradation of T1 was not occurring. Also, the pattern of appearance of each transcript throughout cell growth is not consistent with what I would expect to see if T2 were a processed variant of Tl. In early exponential phase the level of T1 appears to be greater than that of T2 (Fig. 30), implying that Tl is independent of T2. The identical arrangement of transcription start sites was found in the E. coli galactose operon, which has two transcription start sites separated by 5 nt (Musso et al., 1977). When E. coli utilizes galactose as the sole carbon source, the g a lP l promoter is dominant and the galP2 promoter is blocked by the cyclic AMP- Cyclic AMP Receptor Protein complex (cAMP-CRP) (Musso et a l., 1977). 188 During growth on glucose, the gal repressor (GalR) prevents expression from g a lP l, but a basal level of expression continues from galP2 (Jin, 1994). Since the transcript originating fromgalP2 is more efficiently transcribed than the transcript originating from g a lP l , low levels of the galP 2 transcript are sufficient for production of the galE gene product (UDP-glucose-4-epimerase; Jin, 1994). Production of UDP-glucose-4-epimerase is necessary even during glucose utilization since this protein is required for the synthesis of polysaccharides and the cell wall (Jin, 1994). Perhaps the 5 nt difference in length between the two p ilT transcripts is similarly important for translation efficiency. The presence of only one putative o70-like p ilT promoter immediately upstream of the two transcription initiation sites makes it difficult to speculate as to why there are two transcription initiation sites. However, the apparent growth phase-dependent expression of Tl and T2 has prompted me to speculate that transcription ofp ilT requires both RpoD (a70) and RpoS (o38). P. aeruginosa PAOl RpoS is a functional and structural homolog of the E. coli stationary phase sigma factor, RpoS (Tanaka and Takahashi, 1994). Promoter sequences recognized by these two sigma factors are similar, and it has been suggested that RpoS is a second principal sigma factor (Tanaka et a l., 1993). In P. aeruginosa, rpoD and rpoS are reciprocally expressed: the amount of rpoD message decreases in stationary phase and the rp o S transcript level increases in stationary phase (Fujita et a l., 1994). I propose that during exponential growth, transcription ofp ilT is mediated primarily by RpoD, but that a switch to RpoS occurs as levels of RpoD decline in stationary phase. Since the p ilT message appears to be present at constant levels in stationary phase, it seems likely that RpoS is responsible for this p ilT expression. The low level of RpoS that is present during exponential phase probably also contributes somewhat to p ilT expression. It is possible that the two transcription start sites observed for p ilT are the result of differences in binding of RNA polymerase containing the two different sigma factors. The shorter T2, which is expressed only during cell growth, may be the result of transcription initiation by E g70. Conversely, levels of Tl increase in stationary phase, possibly representing the increasing level of 0 38 available for transcription initiation. Tanaka et al. (1993) demonstrated that certain housekeeping promoters in E. coli are recognized by both a7o and a 38, although they did not report any effect on the transcription start site. There are no reported RpoS mutants of P. aeruginosa , and an E. coli RpoS mutant may not be useful in determining whether p ilT expression requires g 38. In E. coli, the g 70 protein remains at a constant level throughout growth; thus, a requirement of RpoS forp ilT expression may be masked if g 70 is present. An E. coli rpoS mutant may however be useful in determining if RpoS is required for the transcription start site of Tl. 190 The growth phase-dependent expression reported here correlates well with the fact that several p ilT homologs are also growth phase-dependent. The P. aeruginosa xcp genes are organized as two divergent operons that are expressed at low levels during exponential phase, exhibit a burst in expression during the transition to stationary phase, and then undergo a rapid decrease to a basal level of expression (Akrim et a l., 1993). These two operons contain 11 xcp genes, including xcpR which encodes the PilT homolog. Expression of the E rw inia spp.outC-M operon is similarly induced in early stationary phase (Lindeberg and Collmer, 1992). The expression of p ilT during exponential growth coincides with previous reports that twitching motility occurs in growing cells (Henrichsen, 1975c). This observation does not, however, explain why p ilT expression continues into stationary phase. It would be interesting to determine if pilB is expressed in a similar fashion to p ilT , since these two proteins are probably required under the same conditions. I have proposed thatp ilT expression is transcriptionally regulated at two levels. Transcription ofp ilT is growth phase-dependent, increasing during exponential phase into stationary phase. This level of regulation may require a switch from one principal sigma factor, RpoD, to another, RpoS, during the transition into stationary phase. Further up in the hierarchy of p ilT regulation are the elements responsible for the differential expression ofp ilT on solid and 191 liquid media. A transcriptional regulator may act in response to some unknown environmental signal(s) to cause a three-fold increase in p ilT transcription. LITERATURE CITED Akrim, M., M. Bally, G.Ball, J. Tommassen, H. Teerink, A. Filloux, and A. Lazdunski. 1993. Xcp-mediated protein secretion in Pseudomonas aeruginosa: identification of two additional genes and evidence for regulation of xcp gene expression. Mol. Microbiol. 10:431-443. Allaoui, A., P. J. Sansonetti, and C. Parsot. 1993. MxiD, an outer membrane protein necessary for the secretion of the Shigella flexneri Ipa invasins. Mol. Microbiol. 7:59-68. Aim, R. A., and J. S. Mattick. 1995. Identification of a gene, pilV, required for type IV fimbrial biogenesis in Pseudomonas aeruginosa , whose product possesses a pre-pilin-Uke leader sequence. Mol. Microbiol. 16:485-496. Aim, R. A., A. J. Bodero, P. D. Free, and J. S. Mattick. 1996. Identification of a novel gene, p ilZ , essential for type 4 fimbrial biogenesis in Pseudomonas aeruginosa. J. Bacteriol. 178:46-53. Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A Smith, and K. Struhl. 1990. Current Protocols in Moloecular Biology. Greene Publishing Associates and Wiley Inter-Science, New York. Baker, N. R., and H. Marcus. 1980. Adherence of clinical isolates of Pseudomonas aeruginosa to hamster tracheal epithelium in vitro. Curr. Microbiol. 7:3540. Baker, N. R., V. Minor, C. Deal, M. S. Shahrabadi, D. A. Simpson, and D. E. Woods. 1991. Pseudomonas aeruginosa exoenzyme S is an adhesin. Infect. Immun. 59: 2859-2863. Baker, N.,and C. Svandborg-Ed6n. 1989. The role of alginate in the adherence of Pseudomonas aeruginosa. Antibiot. Chemother. 42: 72-79. Bally, M., A. Filloux, M. Akrim, G. Ball, A. Lazdunski, and J. Tommassen. 1992. Protein secretion in Pseudomonas aeruginosa: characterization of seven xcp genes and pocessing of secretory apparatus components by prepilin peptidase. Mol. Microbiol. 6:1121-1131. 192 193 Beard, M. K. M., J. S. Mattick, L. J. Moore, M. R. Mott, C. F. Marrs, and J. R. Egerton. 1990. Morphogenetic expression ofMoraxella bovis fimbriae (pili) in Pseudomonas aeruginosa. J. Bacteriol. 172: 2601-2607. Bellido, F. and R. E. W. Hancock. 1993. Susceptibility and resistance of Pseudomonas aeruginosa to antimicrobial agents, p. 321-348. M. Campa (ed.), In Pseudomonas aeruginosa as an opportunistic pathogen. Plenum Press, New York. Benvenisti, L., S. Koby, A. Rutman, H. Gilada, T. Yura, and A. B. Oppenheim. 1995. Cloning and primary sequence of the rpoH gene from Pseudomonas aeruginosa. Gene 155: 73-76. Bimboim, H. C., and J. Doly. 1979. A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucl. Acids Res. 7: 1513-1523. Biswas, J. D., T. Sox, E. Blakman, and P. F. Sparling. 1977. Factors affecting genetic transformation of Neisseria gonorrhoae. J. Bacteriol. 129: 983-992. Botzenhart, K., and H. Riiden. 1987. Hospital acquired infections caused by Pseudomonas aeruginosa. Antibiot. Chemother. 39:1-15. B0vre, K., and L. O. Fr0holm. 1972. Variation of colony morphology reflecting fimbriation in Moraxella bovis and two reference strains of M. nonliquefaciens. Acta. Path. Microbiol. Scand. Sect. B, 80: 649-659. Boyd, J. M., T. Koga, and S. Lory. 1994. Identification and characterization of PilS, an essential regulator of pilin expression in Pseudomonas aeruginosa. Mol. Gen. Genet. 243: 565-574. Boyd, J. M., and S. Lory. 1996. Dual function of PilS during transcriptional activation of the Pseudomonas aeruginosa pilin subunit gene. J. Bacteriol. 178: 831-839. Boyer, H. W., and D. Rolland-Dussoix. 1969. A complementation analysis of the restriction and modification of DNA in Escherichia coli. J. Mol. Biol. 41: 459-472. Bradford, M. M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein dye binding. Anal. Biochem. 72: 248-254. Bradley, D. E. 1972a. A study of pili on Pseudomonas aeruginosa. Genet. Res. 19: 39-51. Bradley, D. E. 1972b. Evidence for retraction of Pseudomonas aeruginosa RNA phage pili. Biochem. Biophys. Res. Commun. 47:142-149. 194 Bradley, D. E. 1972c. Shortening of Pseudomonas aeruginosa pili after RNA- phage adsorption. J. Gen. Microbiol. 72: 303-319. Bradley, D. E. 1974. The adsorption of Pseudomonas aeruginosa pilus- dependent bacteriophages to a host mutant with nonretractile pili. Virology 58: 149-163. Bradley, D. E. 1980a. A function of Pseudomonas aeruginosa PAO polar pili: twitching motility. Can. J. Microbiol. 26:146-154. Bradley, D. E. 1980b. Mobilization of chromosomal determinants for the polar pili ofPseudomonas aeruginosa PAO by FP plasmids. Can. J. Microbiol. 26: 155-160. Bradley, D. E. and T. L. Pitt. 1974. Pilus-dependence of four Pseudomonas aeruginosa bacteriophage with non-contractile tails. J. Gen. Virol. 23:1-15. Brammer, W. J., and P. H. Clarke. 1964. Induction and repression of Pseudomonas aeruginosa amidase. J. Gen. Microbiol. 37: 307-319. Britigan, B. E., M. B. Hayek, B. N. Doebbeling, and R. B. Fick. 1993. Transferrin and lactoferrin undergo proteolytic cleavage in the Pseudomonas aeruginosa- infected lungs of patients with cystic fibrosis. Infect. Immun. 61: 5049-5055. Brossay, L., G. Paradis, R. Fox, M. Koomey, and J. Hebert, 1994. Identification, localization, and distribution of the P ilT protein in Neisseria gonorrhoeae. Infect. Immun. 62: 2302-2308. Brunschwig, E., and A. Darzins. 1992. A two-component T7 system for the overexpression of genes inPseudomonas aeruginosa. Gene 111: 35-41. Burke, J. M., C. P. Novotny, and P. Fives-Taylor. 1979. Defective F pili and other characteristics of Flac and Hfr Escherichia coli mutants resistant to bacteriophage R17. J. Bacteriol. 140: 525-531. Chandler, R. L., K. Smith, and B. A. Turfrey. 1985. Exposure of bovine cornea to different strains of Moraxella bovis and to other bacterial species in vitro. J. Comp. Pathol. 95: 415-423. Chen, L.-H., S. A. Emory, A. L. Bricker, P. Bouvet, and J. G. Belasco. 1991. Structure and function of a bacterial mRNA stabilizer: analysis of the 5' untranslated region of ompA mRNA. J. Bacteriol. 173:4578-4586. Christie, P. J., J. E. Ward, Jr., M. P. Gordon, and E. W. Nester. 1989. A gene required for transfer of T-DNA to plants encodes an ATPase with autophosphorylating activity. Proc. Natl. Acad. Sci. USA 86: 9677-9681. 195 Cox, E. C. 1976. Bacterial mutator genes and the conrol of spontaneous mutation. Annu. Rev. Genet. 10: 135-156. Darzins, A. 1993. The pilG gene product, required for Pseudomonas aeruginosa pilus production and twitching motility, is homologous to the enteric, single domain response regulator CheY. J. Bacteriol. 175:5934-5944. Darzins, A. 1994. Characterization of a Pseudomonas aeruginosa gene cluster involved in pilus biosynthesis and twitching motility: sequence similarity to the chemotaxis proteins of enterics and the gliding bacterium Myxococcus xanthus. Mol. Microbiol. 11:137-153. Darzins, A. 1995. The Pseudomonas aeruginosa pilK gene encodes a chemotactic methyltransferase (CheR) homologue that is translationally regulated. Mol. Microbiol. 15: 703-717. Darzins, A., and M .J. Casadaban. 1989a. in vivo cloning of Pseudomonas aeruginosa genes with mini-D3112 transposable bacteriophage. J. Bacteriol. 171: 3917-3925. Darzins, A., and M.J. Casadaban. 1989b. Mini-D3112 bacteriophage transposable elements for genetic analysis of Pseudomonas aeruginosa. J. Bacteriol. 171: 3909-3916. Darzins, A., S.-K. Wang, R. I. Vanags, and A. M. Chakrabarty. 1985. Clustering of mutations affecting alginic acid biosynthesis in mucoid Pseudomonas aeruginosa. J. Bacteriol. 164: 516-524. de Groot, A., A. Filloux and J. Tommassen. 1991. Conservation of xcp genes, involved in the two-step protein secretion process, in differentPseudomonas species and other gram-negative bacteria. Mol. Gen. Genet. 229: 278-284. Depiazzi, L. J., and R. B. Richards. 1985. Motility in relation to virulence of Bacteroides nodosus. Vet. Microbiol. 10:107-116. de Ree, J. M., P. Schwillens, and J. F. van den Bosch. 1987. Monoclonal antibodies raised against PAP fimbriae recognize minor components) involved in receptor binding. Microb. Path. 2:113-121. Deretic, V., W. M. Konyecsni, C. D. Mohr, D. W. Martin, and N. S. Hibler. 1989. Common denominators of promoter control in Pseudomonas and other bacteria. Bio/Technology 7:1249-1254. Deretic, V., M. J. Schurr, J. C. Boucher, and D. W. Martin. 1994. Conversion of Pseudomonas aeruginosa to mucoidy in cystic fibrosis: environmental stress and regulation of bacterial virulence by alternative sigma factors. J. Bacteriol. 176: 2773-2780. 196 Doig, P., P. A. Sastry, R. S. Hodges, K. K. Lee, W. Paranchych, and R. T. Irvin. 1990. Inhibition of pilin-mediated adhesion of Pseudomonas aeruginosa to human buccal epithelial cells by monoclonal antibodies directed against pili. Infect. Immun. 58: 124-130. Doig, P., T. Todd, P. A. Sastry, K. K. Lee, R. S. Hodges, W. Paranchych, and R. T. Irvin. 1988. Role of pili in the adhesion ofPseudomonas aeruginosa to human respiratory epithelial cells. Infect. Immun. 5:1641-1646. Donnenberg, M. S., J. A. Giron, J. P. Nataro, J. B. Kaper. 1992. A plasmid- encoded type IV fimbrial gene of enteropathogenic Escherichia coli associated with localized adherence. Mol. Microbiol. 6:3427-3437. Drake, D. and T. C. Montie. 1988. Flagella, motility and invasive virulence of Pseudomonas aeruginosa. J. Gen. Microbiol. 134:43-52. Dreyfuss, G., S. A. Adam, and Y. D. Choi. 1984. Physical change in cytoplasmic messenger ribonucleoproteins in cells treated with inhibitor of mRNA transcription. Mol. Cell. Biol. 4:415-423. Dums, F., J. M. Dow, and M.J. Daniels. 1991. Structural characterization of protein secretion genes of the bacterial phytopathogen Xanthomonas cam pestris pathovar campestris: relatedness to secretion systems of other gram-negative bacteria. Mol. Gen. Genet. 229: 357-364. Elleman, T. C., P.A. Hoyne, and A. W. D. Lepper. 1990. Characterization of the pilin gene ofMoraxella bovis in Pseudomonas aeruginosa. Infect. Immun. 58: 1678-1684. Elleman, T. C., P.A. Hoyne, D.J. Stewart, N.M. McKern, and J.E. Peterson. 1986. Expression of pili fromBacteroides nodosus in Pseudomonas aeruginosa. J. Bacteriol. 168: 574-580. Faast, R., M. A. Ogierman, U. H. Stroeher, and P. A. Manning. 1989. Nucleotide sequence of the structural gene tcpA, for a major pilin subunit of Vibrio cholerae. Gene 85: 227-231. Farinha, M. A., B. D. Conway, L. M. G. Glasier, N. W. Ellert, R. T. Irvin, R. Sherburne, and W. Paranchych. 1994. Alteration of the pilin adhesin of Pseudomonas aeruginosa PAO results in normal pilus biogenesis but a loss of adherence to human pneumocyte cells and decreased virulence in mice. Infect. Immun. 62: 4118-4123. Farinha, M. A., and A. M. Kropinski. 1990. Construction of broad-host-range plasmid vectors for easy visible selection and analysis of promoters. J. Bacteriol. 172: 3496-3499. 197 Farinha, M. A., S.L. Ronald, A.M. Kropinski, and W. Paranchych. 1993. Location of the virulence-associated genes pilA, pilR, rpoN, fliA, fliC, ent, and fb p on the physical map ofPseudomonas aeruginosa PAOl by pulsed-field electrophoresis. Infect. Immun. 61:1571-1575. Figurski, D. H., and D. R. Helinski. 1979. Replication of an origin-containing derivative of plasmid RK2 dependent on a plasmid function providedin trans. Proc. Natl. Acad. Sri. USA 76: 1648-1652. Filloux A., M. Murgier, B. Wretlind, A. Lazdunski. 1987. Characterization of two Pseudomonas aeruginosa mutants with defective secretion of extracellular proteins and comparison with other mutants. FEMS Microbiol. Lett. 40: 159- 163. Finch, P. W., R. E. Wilson, K. Brown, I. Hickson, A. E. Tomkinson, and P. T., Emmerson. 1986. Complete nucleotide sequence of the Escherichia coli recC gene and of the thyA-recC intergenic region. Nud. Acids Res. 14:4437-4451. Fletcher, M. 1991. The physiological activity of bacteria attached to solid surfaces. Adv. Microb. Physiol. 32: 53-85. Folkhard, W., and D. A. Marvin. 1981. Structure of polar pili from Pseudomonas aeruginosa strains K and O. J. Mol. Biol. 149: 79-93. Frost, L. S., M. Carpenter, and W. Paranchych. 1978. N-methylphenylalanine at the N-terminus of Pseudomonas aeruginosa K pili. Nature 271: 87-89. Frost, L. S., B. B. Finlay, A. Opgenorth, W. Paranchych, J. S. Lee. 1985. Characterization and sequence analysis of pilin from F-like plasmids. J. Bacteriol. 164: 1238-247. Fujita, M., K Tanaka, H. Takahashi, and A. Amemura. 1993. Organization and transcription of the principal sigma gene rpoDA)( of Pseudomonas aeruginosa PAOl: involvement of a a32- like RNA polymerase in rpoDA gene expression. J. Bacteriol. 175: 1069-1074. Fujita, M., K. Tanaka, H. Takahashi, and A. Amemura. 1994. Transcription of the principal sigma-factor genes,rpoD and rpoS, in Pseudomonas aeruginosa is controlled according to the growth phase. Mol. Microbiol. 13:1071-1077. Fujiwara, S., and A. M. Chakrabarty. 1994. Post-transcriptional regulation of the Pseudomonas aeruginosa algC gene. Gene 146:1-5. Fyfe, J. A. M., C. S. Carrick, and J. K. Davies. 1995. The pilE gene of Neisseria gonorrhoeae MS11 is transcribed from a a70 promoter during growth in vitro. J. Bacteriol. 177: 3781-3787. 198 Grunstein, M., and D. S. Hogness. 1975. Colony hybridization: a method for the isolation of cloned DNAs that contain a specific gene. Proc. Natl. Acad. Sci. USA. 72: 3961-3965. Gupta, S. K., R. S. Berk, S. Masinick, and L. D. Hazlett. 1994. Pili and lipopolysaccharide ofPseudomonas aeruginosa bind the glycolipid asialo GMi. Infect. Immun. 62: 4572-4579. Haas, D. 1983. Genetic aspects of biodegradation by pseudomonads. Experientia 39:1199-1213. Haas, D., H. Schwarz, and T. F. Meyer. 1987. Release of soluble pilin antigen coupled with gene conversion in Neisseria gonorrhoeas. Proc. Natl. Acad. Sci. USA 84: 9079-9083. Haddix, P. L., and S. C. Straley. 1992. Structure and regulation of the Yersinia pestis yscBCDEF operon. J. Bacteriol. 174: 4820-4828. Hanahan, D. 1983. Studies on transformation of Escherichia coli with plasmids. J. Mol. Biol. 166: 557-580. Hazlett, L. D., M. M. Moon, A. Singh, R. S. Berk, and X. L. Rudner. 1991. Analysis of adhesion, piliation, protease production and ocular infectivity of several P. aeruginosa strains. Curr. Eye Res. 10: 351-362. HenikofF, S. 1987. Unidirection digestion with exonuclease III in DNA sequence analysis. Methods Enzymol. 155:156-165. Henrichsen, J. 1972. Bacterial surface translocation: a survey and a classification. Bacteriol. Rev. 36:478-503. Henrichsen, J. 1975a. The occurrence of twitching motility among gram- negative bacteria. Acta. Pathol. Microbiol. Scand. Sect. B 83:171-178. Henrichsen, J. 1975b. The influence of changes in the environment on twitching motility. Acta. Pathol. Microbiol. Scand. Sect. B 83:179-186. Henrichsen, J. 1975c. On twitching motility and its mechanisms. Acta. Pathol. Microbiol. Scand. Sect. B 83:187-190. Henrichsen, J. 1983. Twitching motility. Ann. Rev. Microbiol. 37: 81-93. Henrichsen, J. and J. Blom. 1975a. Correlation between twitching motility and possession of polar fimbriae in Acinetobacter calcoaceticus . Acta. Path. Microbiol. Scand. Sect. B, 83:103-115. 199 Henrichsen, J. and J. Blom. 1975b. Examination of fimbriation of some gram- negative rods with and without twitching and gliding motility. Acta. Path. Microbiol. Scand. Sect. B 83:161-170. Henriksen, S. D. and J. Henrichsen. 1975. Twitching motility and possession of polar fimbriae in spreading Streptococcus sanguis isolates from the human throat. Acta. Path. Microbiol. Scand. Sect. B 83:133-140. Hershberger, C. D., R. W. Ye, M. R. Parsek, Z.-D. Xie, and A. M. Chakrabarty. 1995. The a lg T (a lg U ) gene of Pseudomonas aeruginosa , a key regulator involoved in alginate biosynthesis, encodes an alternative sigma factor (o&). Proc. Natl. Acad. Sci. USA 92: 7941-7945. Hobbs, M., E. S. R. Collie, P. D. Free, S. P. Livingston, and J. S. Mattick. 1993. PilS and PilR, a two-component transcriptional regulatory system controlling expression of type 4 fimbriae in Pseudomonas aeruginosa. Mol. Microbiol. 7: 669-682. Hobbs, M., and J. S. Mattick. 1993. Common components in the assembly of type 4 fimbriae, DNA transfer systems, filamentous phage and protein- secretion apparatus: a general system for the formation of surface-associated protein complexes. Mol. Microbiol. 10: 233-243. Hodgkin, J., and D. Kaiser. 1979a. Genetics of gliding motility in Myxococcus xanthus (Myxobacterales): genes controlling movement of single cells. Mol. Gen. Genet. 171: 167-176. Hodgkin, J., and D. Kaiser. 1979b. Genetics of gliding motility in Myxococcus xa n th u s (Myxobacterales): two gene systems control movement. Mol. Gen. Genet. 171: 177-191. Hoiby, N., G. Doring, and P. O. Schoitz. 1987. Pathogenic mechanisms of chronic Pseudomonas aeruginosa infections in cystic fibrosis patients. Antibiot. Chemother. 39: 60-76. Holder, I. A., R. Wheeler, and T. C. Montie. 1982. Flagellar preparations fromP. aeruginosa : animal protection studies. Infect. Immun. 35: 276-280. Holloway, B. W., V. Krishnapillai, and A. F. Morgan. 1979. Chromosomal genetics of Pseudomonas. Microbiol. Rev. 43: 73-102. Holloway, B. W., U. Romling, and B. Tummler. 1994. Genomic mapping of Pseudomonas aeruginosa PAO. Microbiology. 140: 2907-2929. Holmes, D. S., and M. Quigley. 1981. A rapid boiling method for the preparation of bacterial plasmids. Anal. Biochem. 114:193-197. 200 Hu, N.-T., M.-N. Hung, S.-J. Chiou, F. Tang, D.-C. Chiang, H.-Y. Huang, and C.- Y. Wu. 1992. Cloning and characterization of a gene required for the secretion of extracellular enzymes across the outer membrane by Xanthomonas campestris pv. campestris. J. Bacteriol. 174: 2679-2687. Huang, H.-C., S. Y. He, D. W. Bauer, and A. Collmer. 1992. The Pseudomonas syringae pv. syringae 61 hrpH product, an envelope protein required for the elicitation of the hypersensitive response in plants. J. Bacteriol. 174: 6878- 6885. Iglewski, B. H., J. Sadoff, M. J. Bjom, and E. S. Maxwell. 1978. Pseudomonas aeruginosa exoenzyme S: an adenosine diphosphate ribosyl transferase distinct from toxin A. Proc. Natl. Acad. Sci. USA 75: 3211-3215. Irvin, R. T. 1993. Attachment and colonization of Pseudomonas aeruginosa'. role of the surface structures, p. 19-42. M. Campa (ed.), In Pseudomonas aeruginosa as an opportunistic pathogen. Plenum Press, New York. Irvin, R. T., P. Doig, K. K. Lee, P. A. Sastry, W. Paranchych, T. Todd, and R. S. Hodges. 1989. Characterization of the Pseudomonas aeruginosa pilus adhesin: confirmation that the pilin structural protein subunit contains a human epithelial cell-binding domain. Infect. Immun. 57: 3720-3726. Ishimoto, K. S., and S. Lory. 1989. Formation of pilin inPseudomonas aeruginosa requires the alternative sigma factor (RpoN) of RNA polymerase. Proc. Natl. Acad. Sci. USA 86:1954-1957. Ishimoto, K. S., and S. Lory. 1992. Identification of p ilR , which encodes a transcriptional activator of the Pseudomonas aeruginosa pilin gene. J. Bacteriol. 174: 3514-3521. Jackman, S. H., and R. F. Rosenbusch. 1984. In vitro adherence of Moraxella bovis to intact comeal epithelium. Curr. Eye Res. 3:1107-1112. Jiang, B., and S. P. Howard. 1992. The Aeromonas hydrophila exeE gene, required for both protein secretion and normal outer membrane biogenesis, is a member of a general secretion pathway. Mol. Microbiol. 6:1351-1361. Jin, D. J. 1994. Slippage synthesis at thegalP2 promoter ofEscherichia coli and its regulation by UTP concentration and cAMP-cAMP receptor protein. J. Biol. Chem. 269: 17221-17227. Jin, S., K. S. Ishimoto, and S. Lory. 1994. PilR, a transcriptional regulator of piliation in Pseudomonas aeruginosa , binds to a cis-acting sequence upstream of the pilin gene promoter. Mol. Microbiol. 14:1049-1057. 201 Johnson, K., and S. Lory. 1987. Characterization of Pseudomonas aeruginosa mutants with altered piliation. J. Bacteriol. 169:5663-5667. Johnson, K., M. L. Parker, and S. Lory. 1986. Nucleotide sequence and transcriptional initiation site of two Pseudomonas aeruginosa pilin genes. J. Biol. Chem. 261: 15703-15708. Johnston, J. L., S. J. Billington, V. Haring, and J. I. Rood. 1995. Identification of fimbrial assembly genes from Dichelobacter nodosus : evidence that fim P encodes the typelV prepilin peptidase. Gene 161: 21-26. Jonsson, A.-B., G. Nyberg, and S. Normark. 1991. Phase variation of gonococcal pili by frameshift mutation in pilC, a novel gene for pilus assembly. EMBO J. 10: 477-488. Kaufman, M. R., C. E. Shaw, I. D. Jones, and R. K Taylor. 1993. Biogenesis and regulation of the Vibrio cholerae toxin-coregulated pilus: analogies to other virulence factor secretory systems. Gene 126: 43-49. Krishnapillai, V. 1971. A novel transducing phage. Its role in recognition of a possible new host-controlled modification system in Pseudomonas aeruginosa. Mol. Gen. Genet. 114:134-143. Krivan, H. C., V. Ginsberg, and D. D. Roberts. 1988a. Pseudomonas aeruginosa and Pseudomonas cepacia isolated from cystic fibrosis patients binding specifically to gangliotetraosylceramide (asialo-GMi) and gangliotrio- sylceramide (asialo-GM2). Arch. Biochem. Biophys. 260:493-496. Krivan, H. C., D. D. Roberts, and V. Ginsberg. 1988b. Many pulmonary pathogenic bacteria bind specifically to the carbohydrate sequence GalNAcfil- 4Gal found in some glycolipids. Proc. Natl. Acad. Sci. USA 85:6157-6161. Krylov, V. N., V. G. Bogush, and J. Shapiro. 1980. Bacteriophage of Pseudomonas aeruginosa with DNA similar in structure to that of phage Mul. General description, localization of sites sensitive to endonucleases in DNA, and structure of homoduplexes of phage D3112. Genetika 16: 824-832. Kudoh I., J. P. Wiener-Kronish, S. Hashimoto, J. F. Pittet, D. Frank. 1994. Exoproduct secretions of Pseudomonas aeruginosa strains influence severity of alveolar epithelial injury. Am. J. Physiol. 267: L551-L556. Kulich, S. M., D. W. Frank, and J. T. Barbieri. 1993. Purification and characterization of exoenzyme S from Pseudomonas aeruginosa 388. Infect. Immun. 61:307-313. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 277:680-685. 202 Lapidus, I. R., and H. C. Berg. 1982. Gliding motility of Cytophaga sp. strain U67. J. Bacteriol. 151: 384-398. Lauer, P., N. H. Albertson, and M. Koomey. 1993. Conservation of genes encoding components of a type IV pilus assembly/two-step protein export pathway inNeisseria gonorrhoeas. Mol. Micrbiol. 8:357-368. Lautrop, H. 1961. Bacterium anitratum transferred to the genus Cytophaga. Int. Bull. Bacteriol. Nomencl. 11:107-108. Lebeda, A., K. Kudela, and Z. Jedlickova. 1984. Pathogenicity of Pseudomonas aeruginosa for plants and animals. Acta Phytopathol. Acad. Sci. Hung. 19: 271-284. Lee, K. K., P. Doig, R. T. Irvin, W. Paranchych, and R. S. Hodges. 1989. Mapping the surface regions of Pseudomonas aeruginosa PAK pilin: the importance of the C-terminal region for adherence to human buccal epithelial cells. Mol. Microbiol. 3:1493-1499. Lee, K. K., H. B. Sheth, W. Y. Wong, R. Sherburne, W. Paranchych, R. S. Hodges, C. S. A. Lingwood, H. Krivan, and R. T. Irvin. 1994. The binding of Pseudomonas aeruginosa pili to glycosphingolipids is a tip-associated event involving the C-terminal region of the structural pilin subunit. Mol. Microbiol. 11: 705-713. Lessel, M., D. Balzer, W. Pansegrau, and E. Lanka. 1992. Sequence similarities between the RP4 Tra2 and the Ti VirB region strongly support the conjugation model for T-DNA transfer. J. Biol. Chem. 267: 20471-20480. Lindeberg, M., and A. Collmer. 1992. Analysis of eight out genes in a cluster required for pectic enzyme secretion by Erwinia chrysanthemi: sequence comparison with secretion genes from other gram-negative bacteria. J. Bacteriol. 174: 7385-7397. Lindgren V., and B. Wretlind. 1987. Characterization of a Pseudomonas aeruginosa transposon insertion mutant with defective release of exoenzymes. J. Gen. Microbiol. 133:675-681. Liu, P. V., Y. Abe, and J. L. Bates. 1961. The roles of various fractions of Pseudomonas aeruginosa in its pathogenesis. J. Infect. Dis. 108:218-228. Loening, U. E. 1967. The fractionation of high-molecular-weight ribonucleic acid by polyacrylamide gel electrophoresis. Biochem. J. 102: 251-257. Lory, S. 1992. Determinants of extracellular protein secretion in gram- negative bacteria. J. Bacteriol. 174: 3423-3428. 203 Lund, B., F. P. Lindberg, B. I. Marklund, S. Nonnark. 1987. The PapG protein is the a-D-galactopyranosyl-(l-4)-B-D-galactopyranose-binding adhesin of uropathogenic Escherichia coli. Proc. Natl. Acad. Sci. USA 84:5898-5902. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Marcus, H., and N. R. Baker. 1985. Quantitation of adherence of mucoid and nonmucoid Pseudomonas aeruginosa to hamster tracheal epithelium. Infect. Immun. 47: 723-729. Marrs, C. F., G. Schoolnik, J. M. Koomey, J. Hardy, J. Rothbard, S. Falkow. 1985. Cloning and sequencing of a Moraxella bovis pilin gene. J. bacteriol. 163: 132-139. 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. Mol. Microbiol. 9:857-868. 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. Mol. Microbiol. 16:497-508. Marvin, D. A., and B. Hohn. 1969. Filamentous bacterial viruses. Bacteriol. Rev. 33: 172-209. Mathis, L. S. and J. J. Scocca. 1984. On the role of pili in transformation of Neisseria gonorrkoeae. J. Gen. Microbiol. 130: 3165-3173. Mattick, J. S., M. M. Bills, B. J. Anderson, B. Dalrymple, M. R. Mott, and J. R. Egerton. 1987. Morphogenetic expression ofBacteroides nodosus fimbriae in Pseudomonas aeruginosa. J. Bacteriol. 169:33-41. McCarter, L., M. Hilmen, and M. Silverman. 1988. Flagellar dynamometer controls swarmer cell differentiation of V. parahaemolyticus. Cell. 54:345-351. McCarter, L. L., and M. E. Wright. 1993. Identification of genes encoding components of the swarmer cell flagellar motor and propeller and a sigma factor controlling differentiation of Vibrio parahaemolyticus. J. Bacteriol. 175: 3361-3371. McEldowney, S., and M. Fletcher. 1988. Effect of pH, temperature, and growth conditions on the adhesion of a gliding bacterium and three nongliding bacteria to polystyrene. Microbial Ecology. 16:183-195. 204 Merrick, M. J. 1993. In a class of its own - the RNA polymerase sigma factor o54 (oN). Mol. Microbiol. 10: 903-909. Michiels, T., J.-C. Vanooteghem, C. Lambert de Rouvroit, B. China, A. Gustin, P. Boudry, and G. R. Comelis. 1991. Analysis of virC, an operon involved in the secretion of Yop proteins by Yersinia enterocolitica. J. Bacteriol. 173: 4994- 5009. Miller, J. H. 1972. Experiments in molecular genetics, p. 352-355. Cold Spring Harbor Laboratory. Cold Spring Harbor, NY. Miller, V. L., R. K. Taylor, and J. J. Mekalanos. 1987. Cholera toxin transcriptional activator ToxR is a transmembrane DNA binding protein. Cell 48: 271-279. Musso, R. E., R. Di Lauro, S. Adhya, and B. de Crombrugghe. 1977. Dual control for transcription of the galactose operon by cyclic AMP and its receptor protein at two interspersed promoters. Cell 12:847-854. Naczynski, Z. M., C. Mueller, and A. M. Kropinski. 1995. Cloning the genes for the heat shock response regulator (a32 homolog) from Pseudomonas aeruginosa. Can. J. Microbiol. 41:75-87. Nicas, T. I. and B. H. Iglewski. (eds.) 1986. Toxins and virulence factors of Pseudomonas aeruginosa , p. 195-213, In The Bacteria, Vol. X. Academic Press, New York. Novotny, C. P., and P. Fives-Taylor. 1974. Retraction of F pili. J. Bacteriol. 117: 1306-1311. Novotny, C. P., and P. Fives-Taylor. 1978. Effects of high temperature on Escherichia coli F pili. J. Bacteriol. 133:459-464. 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. J. Bacteriol. 172: 2911-2919. Nunn, D. N., and S. Lory. 1991. Product of the Pseudomonas aeruginosa gene pilD is a prepilin leader peptidase. Proc. Natl. Acad. USA 88:3281-3285. Nunn, D. N., and S. Lory. 1992. Components of the protein-excretion apparatus ofPseudomonas aeruginosa are processed by the type IV prepilin peptidase. Proc. Natl. Acad. Sci. USA 89:47-51. 205 Nunn, D. N., and S. Lory. 1993. Cleavage, methylation, and localization of the Pseudomonas aeruginosa export proteins XcpT,-U,-V,and-X. J. Bacteriol. 175: 4375-4382. Ogierman, M. A., S. Zabihi, L. Mourtzios, and P. A. Manning. 1993. Genetic organization and sequence of the promoter-distal region of the tcp gene cluster of Vibrio cholerae. Gene 126: 51-60. Paranchych, W., P. A. Sastry, D. Drake, J. R. Pearlstone, L. B. Smillie. 1985. Pseudomonas pili: studies on antigenic determinants and mammalian cell receptors. Antibiot. Chemother. 36:49-57. Paranchych, W., P. A. Sastry, K. Volpel, B. A. Loh, and D. P. Speert. 1986. Fimbriae (pili): molecular basis of Pseudomonas aeruginosa adherence. Clin. Invest. Med. 9:113-118. Pasloske, B. L., B. B. Finlay, and W. Paranchych. 1985. Cloning and sequencing of the Pseudomonas aeruginosa PAK pilin gene. FEBS Lett. 183: 408-412. Peabody, D. S., C. L. Andrews, K W. Escudero, J. H. Devine, T. O. Baldwin, and D. G. Bear. 1989. A plasmid vector and quantitative techniques for the study of transcription termination in E. coli using bacterial luciferase. Gene 75: 289- 296. Pennington, J. E., and R. M. Williams. 1979. Influence of genetic factors on natural resistance to Pseudomonas. J. Infect. Dis. 129: 396-399. Plotkowski, M.-C., J.-M Tournier, and E. Puchelle. 1996. Pseudomonas aeruginosa strains possess specific adhesins for laminin. Infect. Immun. 64: 600-605. Possot, O., and A.P. Pugsley. 1994. Molecular characterization of PulE, a protein required for pullalanase secretion. Mol. Microbiol. 12: 287-299. Pridmore, R. D. 1987. New and versatile cloning vectors with kanamycin- resistance marker. Gene 56: 309-312. Pugsley, A. P. 1993. The complete general secretory pathway in gram-negative bacteria. Microbiol. Rev. 57: 50-108. Radman, M., R. E. Wagner, Jr., B. W. Glickman, and M. Meselson. 1980. DNA methylation mismatch correction and genetic stability, p. 121-130. In M. Aledviv (ed.), Progress in Environmental Mutagenesis. Elsevier/North-Holland, Amsterdam. 206 Ramphal, R., and G. B. Pier. 1985. Role ofPseudomonas aeruginosa mucoid exopolysaccharide in adherence to tracheal cells. Infect. Immun. 47:1-4. Ramphal, R., and M. Pyle. 1983. Adherence of mucoid and nonmucoid Pseudomonas aeruginosa to acid-injured tracheal epithelium. Infect. Immun. 41: 345-351. Ramphal, R., J. C. Sadoff, M. Pyle, and J. D. Silipigni. 1984. Role of pili in the adherence of Pseudomonas aeruginosa to injured tracheal epithelium. Infect Immun. 44: 38-40. Rao, V. K., and A. Progulske-Fox. 1993. Cloning and sequencing of two type 4 (N-methylphenylalanine) pilin genes from Eikenella corrodens. J. Gen. Microbiol. 139: 651-660. Reeves, P. J., E. Whitcombe, S. Wharam, M. Gibson, G. Allison, N. Bruce, R. Barallon, P. Douglas, V. Mulholland, S. Stevens, D. Walker, and G. P. C. Salmond. 1993. Molecular cloning and characterization of 13 out genes from Erwinia carotovora subspecies carotovora : genes encoding members of a general secretion pathway (GSP) widespread in Gram-negative bacteria. Mol. Microbiol. 8: 443-456. Reyss, I., and A. P. Pugsley. 1990. Five additional genes in the pulC -0 operon of the Gram-negative bacterium Klebsiella oxytoca UNF5023 which are required for pullalanase secretion. Mol. Gen. Genet. 222:176-184. Rhame, F. S. 1980. The ecology and epidemiology ofPseudomonas aeruginosa, p. 31-51. In L. D. Sabath (ed.), Pseudomonas aeruginosa, the organism, diseases it causes, and their treatment. Hans Huber Publishers, Bern. Roncero, C., A. Darzins, and M. Casadaban. 1990. Pseudomonas aeruginosa transposable bacteriophages D3112 and B3 require pili and surface growth for adsorption. J. Bacteriol. 172:1899-1904. Russell, M. A., and A. Darzin. 1994. The pilE gene product of Pseudomonas aeruginosa, required for pilus biogenesis, shares amino acid sequence identity with the N-termini of type 4 prepilin proteins. Mol. Microbiol. 13:973-985. Saiman, L., K. Ishimoto, S. Lory, and A. Prince. 1990. The effect of piliation and exoproduct expression on the adherence of Pseudomonas aeruginosa to respiratory epithelial monolayers. J. Infect. Dis. 161:541-548. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequencing with chain- terminating inhibitors. Proc Natl Acad Sci USA 74:5463-5467. 207 Sastry, P. A., B. B. Finlay, B. L. Pasloske, W. Paranchych, J. R. Pearlstone, and L. B. Smillie. 1985a. Comparative studies of the amino acid and nucleotide sequences of pilin derived from Pseudomonas aeruginosa PAK and PAO. J. Bacteriol. 164: 571-577. Sastry, P. A., J. R. Pearlstone, L. B. Smillie, and W. Paranchych. 1985b. Studies on the primary structure and antigenic determinants of pilin isolated from Pseudomonas aeruginosa K. Can. J. Biochem. Cell Biol. 63: 284-291. Sato, H., K. Okinaga, and H. Saito. 1988. Role of pili in the pathogenesis of Pseudomonas aeruginosa bum infection. Microbiol. Immunol. 32:131-139. Scheuermann, R., S. Tam, P. M.J. Burgers, C. Lu, and H. Echols. 1983. Identification of the e-subunit of Escherichia coli DNA polymerase III holoenzyme as the dnaQ gene product: a fidelity subunit for DNA replication. Proc. Natl. Acad. Sci. USA 80: 7085-7089. Schweizer, H. P. 1991. Improved broad-hoat-range Zac-based plasmid vectors for the isolation and characterization of protein fusions in Pseudomonas aeruginosa. Gene 103: 87-92. Schweizer, H. P. 1992. Allelic exchange in Pseudomonas aeruginosa using novel ColEl-type vectors and a family of cassettes containing a portable oriT and the counter-selectable Bacillus subitilis sacB marker. Mol. Microbiol. 6:1195- 1204. Sheth, H,B.,KK Lee, W. Y. Wong, G. Srivastava, O. Hindsgaul, R. S. Hodges, W. Paranchych, and R. T. Irvin. 1994. The pili ofPseudomonas aeruginosa strains PAK and PAO bind specifically to the carbohydrate sequence BGalNAc(l-4)BGal found in glycosphingolipids asialo-GMl and asialo-GM2. Mol. Microbiol. 11: 715-723. Simon, R., U. Priefer, and A. Puhler. 1983. A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in gram negative bacteria. Bio/Technology 1: 784-790. Southern, E. M. 1975. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98:503-517. Speert, D. P., B. A. Loh, D. D. Cabral, and I. E. Salit. 1986. Nonopsonic phagocytosis of nonmucoid Pseudomonas aeruginosa by human neutrophils and monocyte-derived macrophages is correlated with bacterial piliation and hydrophobicity. Infect. Immun. 53:207-212. Starnbach, M. N., and S. Lory. 1992. The fliA (rpoF) gene of Pseudomonas aeruginosa encodes an alternative sigma factor required for flagellin synthesis. Mol. Microbiol. 6:459-469. 208 Steadman, R., L. W. Heck, and D. R. Abrahamson. 1993. The role of proteases in pathogenesis ofPseudomonas aeruginosa infections, p. 129-143. M. Campa (ed.), In Pseudomonas aeruginosa as an opportunistic pathogen. Plenum Press, New York. Stephens, K. M., C. Roush, and E. Nester. 1995. Agrobacterium VirBll protein requires a concensus nucleotide-binding site for function in virulence. J. Bacteriol. 177: 27-36. Strohl, W. R. 1992. Compilation and analysis of DNA sequences associated with apparent streptomycete promoters. Nucl. Acids. Res. 20: 961-974. Strom, M. S., and S. Lory. 1991. Amino acid substitutions in pilin of Pseudomonas aeruginosa. J. Biol. Chem. 266:1656-1664. Strom, M. S., and S. Lory. 1992. Kinetics and sequence specificity of processing of prepilin by PilD, the type IV leader peptidase ofPseudomonas aeruginosa. J. Bacteriol. 174: 7345-7351. Strom, M. S., and S. Lory. 1993. Structure-function and biogenesis of the type IV pili. Annu. Rev. Microbiol. 47: 565-596. Strom, M. S., D. Nunn, and S. Lory. 1991. Multiple roles of the pilus biogenesis protein PilD: involvement of PilD in excretion of enzymes from Pseudomonas aeruginosa. J. Bacteriol. 173:1175-1180. Strom, M. S., D. N. Nunn, and S. Lory. 1993. A single bifunctional enzyme, PilD, catalyzes cleavage and N-methylation of proteins belonging to the type IV pilin family. Proc. Natl. Acad. Sci. USA 90: 2404-2408. Strom, M. S., D. N. Nunn, and S. Lory. 1994. Posttranslational processing of type IV prepilin and homologs by PilD ofPseudomonas aeruginosa. Methods Enzymol. 235: 527-540. Studier, F. W., and B. A. Moffatt. 1986. Use of bacteriophage T7 RNA polymerase to direct the selective high level expression of cloned genes. J. Mol. Biol. 189: 113-130. Sutcliffe, J. G. 1979. Complete nucleotide sequence of the Escherichia coli plasmid pBR322. Cold Spring Harbor Symp. Quant. Biol. 43: 77-90. Tabor, S., and C. C. Richardson. 1985. A bacteriophage T7 RNA polymerase/ promoter system for controlled expression of specific genes. Proc. Natl. Acad. Sci. USA 82: 1074-1078. 209 Tanaka, K., Y. Takayanagi, N. Fujita, A. Ishihama, and H. Takahashi. 1993. Heterogeneity of the principal o factor inEscherichia coli : the rpoS gene product a 38 is a second principal a factor of RNA polymerase in stationary phaseEscherichia coli. Proc. Natl. Acad. Sci. USA 90: 3511-3515. Tanaka, K., and H. Takahashi. 1994. Cloning, analysis and expression of an rpoS homologue gene from Pseudomonas aeruginosa PAOl. Gene 150:81-85. Tang, H., M. Kays, and A. Prince. 1995. Role of Pseudomonas aeruginosa pili in acute pulmonary infection. Infect. Immun. 63:1278-1285. T0njum, T., C. F. Marrs, F. Rozsa, and K. B 0vre. 1991. The type 4 pili of Moraxella nonliquefaciens exhibits unique similarities with the pilins of Neisseria gonorrhoeae and Dichelobacter (Bacteroides ) nodosus. J. Gen. Microbiol. 137: 2483-2490. T0njum, T., S. Weir, K. Bpvre, A. Progulske-Fox, and C. F. Marrs. 1993. Sequence divergence in two tandemly located pilin genes of Eikenella corrodens. Infect. Immun. 61:1909-1916. Towbin, H., T. Staehelin, and J. Gordon. 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA 76: 4350-4354. Turner, L. R., J. C. Lara, D. N. Nunn, and S. Lory. 1993. Mutations in the consensus ATP-binding sites of XcpR and PilB eliminate extracellular protein secretion and pilus biogenesis in Pseudomonas aeruginosa. J. Bacteriol. 175: 4962-4969. Van Saene, H. K. F., J. C. Van Putte, J. J. M. Van Saene, T. W. Van de Gromde, and E. G. A. Warmerdam. 1989. Sink flora in a long-stay hospital is determined by the patient’s oral and rectal flora. Epidemiol. Infect. 102: 231- 238. Walker, J. E., M. Saraste, M. J. Runswick, and N. J. Gay. 1982. Distantly related sequences in the alpha- and beta- subunits of ATP synthase, myosin, kinases and other ATP-requiring enzymes and a common nucleotide binding fold. EMBO J. 1: 945-951. Watts, T. H., C. M. Kay, and W. Paranchych. 1983a. Spectral properties of three quaternary arrangements of Pseudomonas pilin. Biochemistry 22: 3640- 3646. Watts, T. H., P. A. Sastry, R. S. Hodges, and W. Paranchych. 1983b. Mapping of the antigenic determinants of Pseudomonas aeruginosa PAK polar pili. Infect. Immun. 42:113-121. 210 Watts, T. H., E. A. Worobec, and W. Paranchych. 1982. Identification of pilin pools in the membranes of Pseudomonas aeruginosa. J. Bacteriol. 152: 687- 691. Weiss, R. L. 1971. The structure and occurrence of pili (fimbriae) on Pseudomonas aeruginosa. J. Gen. Microbiol. 67:135-143. Whitchurch, C. B. 1994. Genes involved in the biogenesis and function of type IV fimriae. Ph.D. dissertation. University of Queensland, Queensland, Australia. Whitchurch, C. B., M. Hobbs, S. P. Livingston, V. Krishnapillai, and J. S. Mattick. 1990. Characterisation of a Pseudomonas aeruginosa twitching motility gene and evidence for a specialized protein export system widespread in eubacteria. Gene 101: 33-44. Whitchurch, C. B., and J. S. Mattick. 1994a. Characterization of a gene, pilU , required for twitching motility but not phage sensitivity in Pseudomonas aeruginosa. Mol. Microbiol. 13:1079-1091. Whitchurch, C. B., and J. S. Mattick. 1994b. Escherichia coli contains a set of genes homologous to those involved in protein secretion, DNA uptake and the assembly of type-IV fimbriae in other bacteria. Gene 150: 9-115. Woo, T. H. S., A. F. Cheng, and J. M. Ling. 1992. An application of a simple method for the preparation of bacterial DNA. BioTechniques 13: 696-698. Woods, D. E., D. C. Strauss, W. G. Johanson, and J. A. Bass. 1981. Role of fibronectin in the prevention of adherence of Pseudomonas aeruginosa to buccal cells. J. Infect. Dis. 143: 784-790. Woods, D. E., D. C. Straus, W. G. Johanson, and J. A. Bass. 1983. Factors influencing the adherence of Pseudomonas aeruginosa to mammalian buccal epithelial cells. Rev. Infect. Dis. 5: S846-S851. Woods, D. E., D. C. Straus, W. G. Johanson, V. K. Berry, and J. A. Bass. 1980. Role of pili in adherence ofPseudomonas aeruginosa to mammalian buccal epithelial cells. Infect. Immun. 29:1146-1151. Wretlind, B., and O. R. Pavlovskis. 1984. Genetic mapping and characterization of Pseudomonas aeruginosa mutants defective in the formation of extracellular proteins. J. Bacteriol. 158:801-808. Yanisch-Perron, C., J. Vieira, and J. Messing. 1985. Improved M13 cloning vectors and host strains: nucleotide sequences of the M13mpl8 and pUC19 vectors. Gene 33: 103-119. 211 Zusman, D. R. 1982. ‘Frizzy’ mutants: a new class of aggregation-defective developmental mutants of Myxococcus xanthus. J. Bacteriol. 150: 1430-1437. Appendix A Maps of Cloning Vectors Used in This Study 212 Figure 31. Circular map of pBR322 (Sutcliffe, 1979). This plasmid is mobilizable but cannot replicate in Pseudomonas. Restriction endonuclease sites used in this study are shown. 213 214 H indlll EcoRI lamHI Sail P str Ap Tc pBR322 4.36 kb Aval on Figure 31 Figure 32. Circular map of pR01614 (Olsen et a l., 1982). This plasmid is mobilizable and has a broad-host-range replicon. The fi-lactamase gene, tet gene, and restriction endonuclease sites used in this study are shown. 215 216 H in d lll EcoRI lamHI S ail P stI Ap Tc pR01614 6.2 kb broad-hoat-range atabilicing fragment PstI Figure 32 Figure 33. Circular map of pUCP18 (Schweizer, 1991). This plasmid is essentially pUC18 containing the 1.8 kb broad-host-range stabilizing fragment from pR01614. Thus, this plasmid can replicate in Pseudomonas but is not mobilizable. Plasmid pUCP19 (Schweizer, 1991) is analogous to pUCP18 but contains the polylinker in the opposite orientation. 217 broad host- Ap range replicon pUCP18 4.5 kb MyHindin // \ 'Sail PstI II BamHI MSmal Kpnl Figure 33 Figure 34. Circular map of pNotl9 (Schweizer, 1992). This plasmid was used in conjunction with pMob3 (Fig. 35) for gene replacement experiments. The 13- lactamase gene and restriction endonuclease sites used in this study are shown. 219 220 Notl Narl EcoRI SphI pNotl9 Hindll! 2.7 kb ori Figure 34 Figure 35. Circular map of pMob3 (Schweizer, 1992). This plasmid is the source of the 5.9 kb Notl mobilization (Mob) cassette used for gene replacement experiments. Relevant restriction endonuclease sites are shown in addition to the genes encoding resistance to kanamycin and chloramphenicol. 221 222 iNotl Km 80cR Notl pMobS 8acB 8.3 kb oriT Cm Figure 35 Figure 36. Circular map of pQF70 (Farinha and Kropinski, 1990). This is a broad-host-range promoter-probe vector that contains the luxAB reporter genes. Relevant restriction endonuclease sites and the J3-lactamase gene are shown. 223 224 Ap luxAB PQF70 6.2 kb Figure 36 Figure 37. Circular map of pQF50 (Farinha and Kropinski, 1990). This is a broad-host-range promoter-probe vector that contains the lacZ reporter gene. Relevant restriction endonuclease sites and the 13-lactamase gene are shown. 225 226 pQFSO 6.8 kb OripMBl orii6oo Ap Figure 37 Figure 38. Circular map of pADD621 (Darzins, unpublished data). This plasmid contains the T7 promoter (Bn) and the broad-host-range replicon from pR01614. A transcription terminator (T) is immediately downstream from the polylinker region. Relevant restriction endonuclease sites, the fi-lactamase gene, and the tet gene are shown. 227 228 iHindin Vr.ii \ \ .8* 1 1 Xbml . B am H l I .Sm al j/. Sael / / / e c o R I Tc1 oriV pADD621 6.5 kb oriT Figure 38 Figure 39. Circular map of pEB12 (Brunschwig and Darzins, 1992). This plasmid contains the T7 promoter (Pt7) and the broad-host-range RSF1010 replicon. Four transcription terminators (T) lie immediately downstream from the polylinker region. Relevant restriction endonuclease sites, the fi-lactamase gene, and the tet gene are shown. 229 230 pE B 12 10.2 kb oriV oriT rep R S F 1 0 1 0 Figure 39 Appendix B Plasmids Constructed for This Study Table 10. Plasmids constructed for this study. P lasm id V ecto r Description* 6.1 kb BamHl subclone of 22 kb HindlH PAOl library done; contains pilT and pilU; pPT1916 pR01614 complemented HOD1; inBarriHl site of pR01614; Apr pPT1916 w ith 2.4 kb Sail fragment deleted; deletes 312 bp from 3' end of pilT and pPT1920 pR01614 all of pilU; Apr pPT1916 with 1.0 kbX7ioI fragment deleted; deletes 196 bp from 5' end ofpilT in pPT1921 pR01614 addition to 804 bp of region upstream of pilT\ Apr pPT1999 pTJS140 3.0 kb Pstl-BamHl pilT pilU clone in Pstl-BamHl sites of pTJS140; Apr pPT2001 pTJS140 1.5 kb Pstl-Kpnl Cbl) pilT clone in Pstl-BamHl Obi) sites of pTJS140; Apr pADD1408 pADD621 1.5 kb PstI (bD-Xpnl (bl) pilT done in Smal of pADD621; [-]; Apr pADD1409 pADD621 1.5 kb PstI (bl)-Xpnl Obi) pilT done in Sm al of pADD621; [+]; Apr pADD1410 pADD 621 1.1 kb Pstl-Sall deleted pilT fragment in Pstl-Sall of pADD621; [+]; Apr pADD1411 pADD621 0.9 kb Xhol-Kpnl Obi) deleted pilT fragment in Sall-Sacl (bl) of pADD621; [+]; Apr pPT2144 pK19 1.5 kb PstI d>\)-Kpnl (bl) pilT done in BamHl (bl) site of pK19; [+]; Kmr pPT2145 pK19 1.5 kb PstI (bD-Kpnl (bl) pilT done in BamHl (bl) site of pK19; [-]; Kmr pPT2152 pK19 2.2 exonudease m-generated deletion of 5' end of pilT from Xbal of pPT2144; Kmr pPT2154 pK19 2.4 exonudease IE-generated deletion of 5' end of pUT from Xbal of pPT2144; Kmr pPT2155 pK19 3.1 exonuclease IE-generated deletion of 5' end of pilT from Xbal of pPT2144; Kmr pPT2160 pK19 4.4 exonuclease IE-generated deletion of 5' end of pilT from Xbal of pPT2144; Kmr pPT2161 pK19 5.1 exonuclease IE-generated deletion of 5' end of pilT from Xbal of pPT2144; Kmr 232 pPT2296 pEB12 1.5 kb pilT clone (HindHL-Smal from pPT2144) inHindlH-Smal of pEB12; Ap Table 10. (Continued) P lasm id V ector Description* pPT2381 pBR322 0.3 kb EcoBl fragm ent from pilT ORF in EcdBl site of pBR322; Tcr Apr pPT2937 pUCP18 3.0 kb Pstl-BamHl pilT pilU fragm ent in Pstl-BamHl sites of pUCP18; [-]; Apr pPT2939 pUCP19 3.0 kb Pstl-BamHl pilT pilU fragment in Pstl-BamHl sites of pUCP19; [+]; Apr pPT2971 pUC18 1.0 kbXftoI (bl) subdone of pilT promoter region inSail (bl) site of pUC18; [-]; Apr pPT2972 pUC18 1.0 kb .X?ioI (bl) subdone of pilT promoter region inSail (bl) site of pUC18; [+]; Apr 1.0 kbjftol (bl) subdone of pilT promoter region [asSmal-Hindlll from pPT2972] pPT3016 pQF70 in Smal-HindHl sites of pQF70; [+]; Apr 1.0 kb Xhol (bl) subdone of pilT promoter region [asSmal-Hindlll from pPT2971] pPT3018 pQF70 in Smal-HindlH sites of pQF70; [-]; Apr pPT3020 pUC18 0.56 kb iVarl (bl) subdone of pilU promoter region inH indi site of pUC18; [+]; Apr pPT3022 pUC18 0.56 kb iVarl (bl) subdone of pilU promoter region inHincll site of pUC18; [-]; Apr 0.56 kb IVarl subdone of pilU promoter region (asHindlll-BamHl from pPT3020) pPT3023 pQF70 in HindlH-BamHl sites of pQF70; [+]; Apr 0.56 kb Narl subdone of pilU promoter region (asHindlll-Smal from pPT3022) in pPT3035 pQF70 HindlH-Smal sites of pQF70; [-]; Apr 0.59 kb PstI QrtyXhol (bl) subdone of pilT promoter region in Smal site of pUC18; pPT3244 pU C l8 [+]; Apr 0.59 kb Pstl-Xhol (bl) subdone of pilT promoter region [as 0.63 kb Kpnl-Hindlll pPT3245 pQF70 from pPT3244] inKpnl-Hindlll sites of pQF70; [+]; Apr pPT3315 pUC18 0.59 kb Pstl-Xhol (bl) subdone of pilT promoter region in Sm al of pUC18; [-]; Apr 233 Table 10. (Continued) P lasm id V ector Description* 0.3 kb Pstl QAyMlul (bl) subclone of pilT promoter region inSmal-Pstl (bl) sites in pPT3317 pUCl8 pUC18 [pPT3244 with 0.32 kb Mlul-Pstl deleted, (bl), and religated; [+]; Apr 0.3 kb Pstl-Mlul subclone of pilT promoter region [asKpnl-Hindlll from pPT3317] pPT3319 pQF70 in Kpnl-Hindlll sites of pQF70; [+]; Apr 0.16 kb Pstl-Sspl subclone of pilT promoter region [as aKpnl-Sspl from pPT3317] in pPT3323 pQF70 Kpnl-Smal sites of pQF70; [+]; Apr 0.56 kb iVarl (bl) subdone of pilU promoter region [asHindUI-BamHI from pPT3327 pQF50 pPT3020] in HindHL-BamHl sites of pQF50; [+]; Apr 0.4 kb AfarI->SacII subdone of pilU promoter region in pUC18 [pPT3020 with 0.16 kb pPT3341 pUC18 SacII-Psfl deleted, (bl), religated]; [+]; Apr 0.4 kb Narl-Sacll subdone of pilU promoter region [asBamHl-HindUl from pPT3342 pQF70 pPT3341] inBamHl-HindUl sites of pQF70; [+]; Apr 0.4 kb iVarl-SacII subdone of pilU promoter region [asBamHl-HindUl from pPT3344 pQF50 pPT3341] inBamHl-HindUl sites of pQF50; [+]; Apr 0.16 kb MluI-SspI subdone of pilT promoter region [asSmal-HindJII from pPT3489 pQF70 pPT3323] inSmal-HindlH sites of pQF70; [+]; Apr 0.3 kb Pstl-Mlul subclone of pilT promoter region [asKpnl-Hindlll from pPT3317] pPT3493 pQF50 in Kpnl-HijidGl of pQF50; [+]; Apr 0.41sb Narl-SacH subdone of pilU promoter region in pUC18 [pPT3022 with 0.17 kb pPT3495 pUC18 SacU-BaniHl fragm ent deleted, (bl), religated]; [-]; Apr | 0.4 kb Narl-SacU subdone of pilU promoter region [asKpnl-Hindlll from pPT3395] | pPT3496 pQF50 in Kpnl-Hindm sites of pQF50; [-]; Apr | 234 Table 10. (Continued) Plasmid Vector Description^ 1.57 kb Kpnl (bl)-BamHI (bl) containing pilU cloned in the JVarl (bU-Sp/d (bl) sites of pPT3519 pN otl9 pNotl9; Apr pPT3519 with 0.55 kb Narl-SphI internal pilU fragment deleted; 1.2 kb Ter cassette pPT3520 pN otl9 [as EcciRL (bl)-Ayal (bl)] from pBH322 inserted into Narl (bl)-SpAI (bl) sites; Apr Ter 5.8 kb Mob cassette [Cmr sacB sacR oriT] firom pMob3 as a Notl fragment cloned in pPT3547 pN otl9 the Notl site of pPT3520; Apr Tcr Cmr 0.3 kb Pstl-Mlul subclone of pilT promoter region [asMlul(h\)-HindUl from pPT3642 pQF50 pPT3315] inSmal-HindHL sites of pQF50; [-]; Apr pPT3728 pUC18 6.0 kb BamHl done containing HOD1 pilT in BamHl site of pUC18; Apr 3.0 kb Pstl-BamKl firom pPT3728 in Pstl-BamHI of pUC18; contains HOD1pilT pPT3729 pUC18 and pilU; Apr pPT3731 pUC18 1.5 kb Pstl-Kpnl containing HOD1 pilT in Pstl-Kpnl sites of pUC18; Apr 1.5 kb HindHl-Kpnl firom pPT3731 in Hindlll-Kpnl sites of pUCP19; contains pPT3733 pUCP19 HOD1 pilT] Apr pPT3734 pK19 1.5 kb Pstl-Kpnl (bl) in BamHl (bl) site of pK19; contains HOD1 pi/T;[-]; Kmr pPT3736 pUCP18 1.5 kb Pstl-Kpnl in Pstl-Kpnl sites of pUCP18;containspilT\ [-]; Apr pPT3738 pUCP19 1.5 kb Pstl-Kpnl in Pstl-Kpnl sites of pUCP19; containspilT; [+]; Apr 30:2 exonudease IQ-generated deletion of 3' end of HOD1 pilT firomXbal site of pPT3749 pK19 pPT3734; Kmr 1.5:3 exonuclease Ql-generated deletion of 3' end of HOD1 pilT from Xbal site of pPT3750 pK19 pPT3734; Kmr 235 Table 10. (Continued) Plasmid Vector Description® pPT3753 pUCP19 1.5 kb Kpnl-BamHl fragment containing pilU in Kpnl-BamHl of pUCP19; [-]; Apr 2:3 exonuclease IH-generated deletion of 3' end of HOD1 pilT from Xbal site of pPT3796 pK19 pPT3734; Kmr 3:5 exonuclease Ill-generated deletion of 3' end of HOD1 pilT from Xbal site of pPT3797 pK19 pPT3734; Kmr pPT3798 pK19 0.22 kb Sall-Sphl subdone of HOD 1 pilT in the Sall-SphI sites of pK19; Kmr 0.39 kb Sall-Kpnl subdone of HOD1 pilT, from pPT3733, in SaU-Kphl sites of pPT3799 pUCP19 pPT3738; Apr 0.39 kb Sall-Kpnl subclone of PAOl pilT, firom pPT3738, in Sall-Kpnl sites of pPT3803 pUCP19 pPT3733; Apr pPT3813 pUCP18 1.5 kb Kpnl-BamHl fragment containing pilU in Kpnl-BamHl of pUCP18; [+]; Apr * Abbreviations: (bl), overhanging ends of a restriction site were removed and/or end-filled with the Klenow fragment of DNA polymerase I; [+], insert is in same transcriptional orientation as thelac promoter or as the promoterless luxAB or lacZ reporter gene; [-], insert is in opposite transcriptional orientation with respect to thelac promoter or the luxAB or lacZ reporter genes. Apr, Ter, Kmr, Cmr: designate resistance to ampicillin, tetracycline, kanamycin, and chloramphenicol, respectively. 236 Appendix C Oligonucleotides Used in This Study 237 Table 11. Oligonucleotides used in this study. j oligonucleotide* sequence hybridization site purpose NEB 17mer 5 ' -GTTTTCCCAGTCACGAC-3' flanks polylinker in forward primer pUC 18/19 and pK18/19ml3/pUC universal primer NEB 24mer 5 ' -AGCGGATAACAATTTCACACAGGA-3' flanks polylinker in reverse primer pUC 18/19 and pK18/19ml3/pUC universal primer 129 bp upstream of 85 5 ' -CACTGGAACAGGAAGATGGC-3 ' generate probe fragments by pilT start codon PCR; sequencing primers 146 5 ' -GATGAGCCAGGGTGCTTCC-3' 70 bp upstream of generate probe fragments by pilT start codon PCR; sequencing primers • Abbreviation: NEB, New England Biolabs. to CO 00 Appendix D Restriction Map of thepilT and pilU Loci 239 Figure 40. Restriction map of the pilT and pilU loci (this study; Whitchurch et al., 1990; Whitchurch and Mattick, 1994a). The 3.0 kb Pstl-BamHl fragm ent which contains the 1.0 kb p ilT gene and the 1.3 kb pilU gene is shown. Restriction endonuclease sites relevant to this study are shown. 240 E coR I N a rl E coR I pilT pilU Appendix E Dendrogram of PilT Homologs 242 Figure 41. Dendrogram of the PilT homologs. This figure was generated using the computer software program PILEUP. 243 244 IncP KilB RK2 KlbA unknown organism TrbB A tumefaciens TrbB A. tumefaciens V irB ll E. coli BfpD V. cholerae TcpT N. gonorrhoeae PilT P. aeruginosa PilT E. coli HofB E. coli hypothetical protein XL campestris XpsE E. chrysanthemi OutE XL pneumoniae PulE A. hydrophila ExeE VL cholerae EpsE P. aeruginosa XcpR E. coli YheG A hydrophila TapB P. aeruginosa PilB D. nodosus FimN N. gonorrhoeae PilF H. influenzae HofB V. cholerae MshE A salmonicida Sec protein B. subtilis ComGl E. coli BfpF Synechocystis sp. Cpn60