INFORMATION TO USERS
This manuscript has been reproduced from the microfihn 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
COMPARATIVE ANALYSIS OF PSEUDOMONAS AERUGINOSA
LASA AND LASD
DISSERTATION
Presented in Partial Fulfillment of the Requirements for the Degree Doctor of
Philosophy in the Graduate School of The Ohio State University
By
Sukjoon Park, B. S.
*****
The Ohio State University
1996
Dissertation Committee: Approved by Dr. Darrell R. Galloway
Dr. Charles J. Daniels
Dr. Donald H. Dean Advisor Graduate Program in Molecular, Dr. Gary E. Means Cellular and Developmental Biology ÜMX Number: 9710640
UMI Microform 9710640 Copyright 1997, 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 ABSTRACT
Pseudomonas aeruginosa is widely known as an opportunistic pathogen.
The organism produces many virulence factors including both proteolytic and non- proteolytic enzymes. Among the proteolytic virulence factors, pseudolysin (elastase) has long been considered to be of major significance due to its ability to digest elastin tissues. However, LasA has been the focus of research for the past several years particularly due to its ability to enhance the elastolytic activity of pseudolysin. In addition, LasA was characterized as a staphylolytic enzyme; LasA causes lysis of
Gram-positive organisms such as Staphylococcus aureus.
In this study, another protease, LasD, was purified and characterized from various strains of P. aeruginosa. LasD is a 23 kDa protease which shares many of the enzymatic properties of LasA, including the ability to lyse heat-killed staphylococci and hydrolyzing synthetic peptides such as pentaglycine. However,
LasD is distinct from the 22 kDa LasA for the following reasons: (i) the N-terminal sequence of LasD shares no homology with LasA or the LasA precursor sequences;
(ii) P. aeruginosa LasA negative mutant strains AD 1825 and FRD2I28 do not produce LasA yet produce LasD, and (iii) specific antibodies to each protease do not exhibit any cross-reactivity. Furthermore, the purified LasD shows strong
u staphylolytic activity only under alkaline pH conditions, while LasA exhibits staphylolytic activity over a broad pH range. The staphylolytic activity of both LasA and LasD may be related to hydrolysis of the pentaglycine bridge in the Gram- positive cell wall. To verify this possibility, various synthetic peptides containing glycine residues were analyzed and the results show that both LasA and LasD cleave peptides with internal diglycine, and the minimum size of the peptide is a tetramer such as tetraglycine.
In addition to the study on the substrate specificity of LasA and LasD, kinetic analysis of both proteases was also performed using acetylated pentaglycine as a substrate. The results indicate that the acetylated pentaglycine is better substrate for
LasA than LasD, thereby indicating another difference between LasA and LasD.
LasD cuts the acetylated pentaglycine more efficiently at higher pH consistent with the observation made using heat-killed S. aureus cells as a substrate.
Finally, the processing events of LasA from an inactive 41 kDa precursor form to an active 22 kDa form was studied primarily to elicit possible roles of other proteases, including LasD, in the processing event. Surprisingly, LasD was found to be involved in the processing of LasA in vitro, indicating LasA and LasD work together for greater efficiency. However, the fact that LasA negative strains still produce active LasD suggests that the possible processing event for activation of
LasD is independent of LasA.
m Dedicated to my parents
IV ACKNOWLEDGMENTS
I express my sincere appreciation to Dr. Darrell R. Galloway for his advice and guidance throughout these years. Without his support, it would not have been possible to come this far. I also thank Mrs. Galloway for inviting me to every holiday celebration for the last six years; I appreciate her willingness to make me a part of the Galloway family. Additional thanks is given to the other members of my committee: Drs. Charles Daniels, Donald Dean, and Gary Means for their encouragement. I would also like to thank Dr. John Lowbridge at the Ohio State
University Biochemical Instrumentation Center for his advise on HPLC. My sincere thanks to the former and present members of the Galloway Lab: Dr. John Peters,
Dr. Christine Sauer, Dr. Leena Hiremath, Dr. Scotty Walker, Dr. Christine Besson,
Dr. Xiang-Yang Han, Dr. Pil-Jung Kang, Monte Kendricks, Todd Kelly, Madhura
Pradhan, Kimberly Denis-Mize, and Susan Meyer. I especially thank Kimberly
Denis-Mize for patiently proofreading my draft over and over again. Most of all, I would like to thank my parents and my wife without whom I could not finish my long journey as a graduate student. Dad, you will never know how grateful I am that you encouraged and supported me for so long. VTTA
February 27,1964 ...... Bom - Seoul, Korea
1990...... B.S. Botany, Ohio University, Athens, Ohio
1990-96...... Graduate Teaching and Research Associate, Ohio State University, Columbus, Ohio
PUBLICATIONS
1. Sulqoon Park and D. R. Galloway (1995). “Purification and characterization of LasD: a second staphylolytic proteinase produced by Pseudomonas aeruginosa.'' Molecular Microbiology 16: 263-270.
2. Peters, J. E., S. J. Park, A. Darzins, L. C. Freck, J. M. Saulnier, J.M. Wallach and D. R. Galloway (1992). “Further studies on Pseudomonas aeruginosa LasA: analysis of specificity.” Molecular Microbiologv 6: 1155-1162.
FIELDS OF STUDY
Major Field: Molecular, Cellular and Developmental Biology
Minor Field: Biochemistry
VI TABLE OF CONTENTS
Page
Abstract...... ü
Dedication ...... iv
Acknowledgments ...... v
Vita...... vi
List of Tables ...... xi
List of Figures ...... xiii
List of Abbreviations ...... xvi
Chapters;
1. Literature Review ...... 1
A. Classification of proteases ...... 1
1. Serine proteases ...... 4 2. Cysteine proteases ...... 8 3. Aspartic proteases ...... 11 4. Metallopro teases ...... 12
B. Pseudomonas aeruginosa as an opportunistic pathogen ...... 19
C. Non-proteolytic Pseudomonas aeruginosa virulence factors ...... 22
1. Exotoxin A ...... 22 2. Exoenzyme S ...... 24 3. Phospholipase C ...... 24
D. Pseudomonas aeruginosa proteases ...... 25
1. Pseudolysin ...... 25 2. LasA...... 31 vii 3. Alkaline protease ...... 34 4. Lysine-specific protease ...... 35
E. Regulation of Pseudomonas aeruginosa proteases ...... 36
2. Materials and Methods ...... 44
A. Strains and plasmids ...... 44
B. Biochemical assays...... 44
1. Determination of elastolytic activity ...... 44 2. Determination of proteolytic activity and inhibition 47 3. Determination of staphylolytic activity and inhibition ...... 49 4. HPLC analysis of substrate specificity using synthetic peptides and p-insulin ...... 52 5. N-terminal sequence determination of LasD using PVDF membrane ...... 55 6. Peptide mapping by proteolysis ...... 60 7. Denaturing polyacrylamide gel electrophoresis ...... 61 8. Purification of P. aeruginosa pseudolysin ...... 64 9. Purification of LasA ...... 66 10. Purification of LasD ...... 67 11. Radiolabeling of LasA precursor proteins ...... 68
C. Assays for kinetic analysis ...... 74
1. Modification of pentaglycine by acétylation ...... 74 2. Verification of acetylated pentaglycine ...... 77 3. Kinetic analysis of enzymatic reactions ...... 77 4. Kinetic analysis of inhibitors ...... 79
D. Assays for genetic analysis ...... 80
1. Agarose gel electrophoresis ...... 80 2. Purification of DNA by Gene cleaning ...... 81 3. Isolation of P. aeruginosa genomic D N A ...... 82 4. Isolation of plasmid DNA by miniprep ...... 83 5. Overexpression of LasA and the LasA precursor in E. c o li...... 86 6. Production of competent cells ...... 92 7. Transformation of plasmid DNA ...... 92
E. Immunological and other assays ...... 95
1. Production of polyclonal antisera ...... 95 viii 2. Immunoblotting ...... 96 3. Fractionation of Pseudomonas aeruginosa...... 99
3. Purification and characterization of LasA and LasD ...... 101
A. Introduction ...... 101
B. Purification ...... 103
1. Purification of LasA and LasD ...... 103 2. Production of anti-LasD and anti-LasA precursor antisera ...... 113
C. Physical characteristics of LasA and LasD...... 114
1. pH and temperature optima of LasA and LasD ...... 114 2. Determination of N-terminal amino acid sequence of LasD ...... 117
D. Substrate specificity of LasA and LasD ...... 120
1. Staphylolytic activity of LasA and LasD ...... 120 2. Peptide substrate specificity of LasA and LasD ...... 123 3. Digestion studies with 3-casein ...... 133
E. Inhibitors of LasA and LasD ...... 138
1. Inhibitor study using heat-killed S. aureus cells as substrate...... 138 2. Inhibitor analysis using p-casein as a substrate ...... 147
F. Discussion ...... 151
4. Kinetic analysis of LasA and LasD ...... 155
A. Introduction ...... 155
B. Kinetic analysis of LasA ...... 158
1. Acétylation of pentaglycine ...... 158 2. Mathematical analysis of the enzymatic reaction ...... 159 3. Inhibition kinetics of LasA ...... 166
C. Kinetic analysis of LasD ...... 167
1. Mathematical analysis of the enzymatic reaction ...... 167 ix a) Alkaline pH condition ...... 170 b) Lower pH condition ...... 173 2. Inhibition kinetics of LasD ...... 173
D. Discussion ...... 174
5. Processing of LasA...... 180
A. Introduction ...... 180
B. Overexpression of the LasA precursor and active LasA ...... 182
C Processing of LasA by LasD ...... 189
D. Discussion ...... 200
6. Conclusions ...... 204
List of References ...... 212
X LIST OF TABLES
Table Title Page
1.1 Classification of proteases by mode of action on substrates ...... 2
2.1 Bacterial strains used in this study ...... 45
2.2 Plasmid vectors used in this study ...... 46
2.3 Preparation of inhibitors used in this study ...... 50
2.4 Synthetic peptides and proteases utilized in HPLC analysis ...... 53
2.5 HPLC methods used in this study for substrate specificity ...... 56
2.6 Recipe for a large double SDS polyacrylamide gel ...... 63
2.7 HPLC methods used in this study to purify LasD ...... 69
2.8 PCR conditions for overexpression of active LasA and the LasA precursor ...... 89
3.1 A typical purification table for pseudolysin ...... 104
3.2 A typical purification table for LasA ...... 109
3.3 A typical purification table for LasD ...... 110
3.4 Substrate specificity of LasD and LasA ...... 137
3.5 Purification time schedule for pseudolysin, LasA, and LasD ...... 152
4.1 Calculation of standard O.D .450 for pentaglycine ...... 165
4.2 Kinetic analysis of LasA and LasD using acetylated pentaglycine as a substrate ...... 179
XI 6.1 C onsm ed glycine residues nearthe active site serine residue in various families of serine proteases ......
XU LIST OF nOURES
Figures Title Page
1.1 Schematic representation of the steps involved in catalysis by a typical serine protease ...... 6
1.2 Schematic representation of the steps involved in catalysis by a typical cysteine protease ...... 9
1.3 Schematic representation of the steps involved in catalysis by a typical aspartic protease...... 13
1.4 Schematic representation of the catalysis of peptide bond cleavage carried out by a metalloprotease ...... 17
1.5 Proposed sequence of events involved in P. aeruginosa pseudolysin (elastase) secretion ...... 29
1.6 A model for the regulation of rpoS via a hierarchical quorum sensing cascade involving LasR, RhlR, and their cognate N-acylhomoserine lactones ...... 42
2.1 Modification of pentaglycine by acétylation ...... 76
2.2 Primers used to overexpress active LasA and the LasA precursor... 87
3.1 Chromatographic elution profile of the LasA and LasD proteases using the FPLC Mono-S cation exchange column 107
3.2 A flow chart representing the purification procedures of various proteins secreted by Pseudomonas aeruginosa...... Ill
3.3 SDS-PAGE and immunoblot analysis of the purified LasA and LasD proteases following final chromatographic separation ...... 115
3.4 Effect of pH on staphylolytic activity of LasA and LasD ...... 118
xiu 3.5 Temperature optima for staphylolytic activity of LasA and LasD ...... 119
3.6 Staphylolytic activity of LasA and LasD ...... 121
3.7 HPLC analysis of glycine standard ...... 125
3.8 HPLC analysis of pentaglycine digestion by LasA...... 127
3.9 Verification of diglycine and triglycine by spiking experiments 129
3.10 HPLC analysis of pentaglycine digestion by LasA ...... 131
3.11 HPLC einalysis of insulin p-chain digestion by various P. aerugmosa proteases ...... 134
3.12 Digestion of P-casein by LasA ...... 139
3.13 Digestion of P-casein by LasD ...... 141
3.14 Effect of inhibitors on staphylolytic activity of LasA ...... 143
3.15 Effect of inhibitors on staphylolytic activity of LasD ...... 145
3.16 Digestion of p-casein with either LasD or LasA in the presence of inhibitors ...... 148
4.1 Plot of initial velocity versus substrate concentration ...... 156
4.2 HPLC elution profiles of unmodified (A) and acetylated (B) pentaglycines ...... 160
4.3 Kinetic analysis of acetylated pentaglycine digestion by LasA 162
4.4 Lineweaver-Burk plot showing competitive inhibition of LasA by ophenanthroline ...... 168
4.5 Kinetic analysis of acetylated pentaglycine digestion by LasD 171
4.6 Lineweaver-Burk plot showing competitive inhibition of LasD by ophenanthroline ...... 175
5.1 Graphical representation of the T7 overexpression system ...... 183
5.2 Overexpression of the LasA precursor in BL21(DE3) ...... 184
xiv 5.3 Overexpression of the LasA precursor and active LasA inBL21(DE3)/pLysS...... 187
5.4 Involvement of various P. aeruginosa fractions in processing the LasA precursor...... 190
5.5 Immunoblot analysis of various proteases involved in LasA processing ...... 194
5.6 Immunoblotting showing the involvement of LasD in LasA processing ...... 197
5.7 Staphylolytic activity of the processed LasA precursor by various proteases ...... 202
6.1 Model illustrating possible roles of P. aerugmosa proteases 210
XV LIST OF ABBREVIATIONS
A Amps Abs. Absorbance Ab Antibody(ies) Amp 100 lOOpg/ml Ampicillin APS Ammonium persulfate bp Base pair(s) BSA Bovine serum albumin ca. Approximately CAPS 3-[cyclohexylamino]-l-propanesulphonic acid CblOO lOOpg/ml Carbenicillin CM Carboxymethyl q)m Counts per minute Cui Curies DEAE Diethylaminoethyl dd Double distilled DEPC Diethyl pyrocarbonated DFP Diisopropyl fluorophosphate DNA Deoxyribonucleic acid DNase Deoxyribonuclease dNTP Deoxyribonucleoside triphosphate DTT Dithiothreitol E. Escbericbia EDTA Ethylenediaminetetraacetate ELISA Enzyme-linked immunosorbent assay ES Enzyme-substrate complex
XVI F Farads Fig. Figure FPLC Fast protein liquid chromatography g Gram GTE Glucose/Tris/EDTA HEPES N-2-hydroxyethyIpiperazine-N’-2-ethanesuIfonic acid HPLC High pressure liquid chromatography hr Hour HRP Horseradish peroxidase HSL Homoserine lactone lEF Isoelectric focusing IPTG Isopropyl P-D-Thiogalactopyranoside k K j Io kb Kilobase kDa Kilodalton L Liter LB Luria-Bertani M Molar concentration m Meter ki- Prefix micro m- Prefix milli m c Minimal inhibitory concentration min Minute(s) n- Prefix nano a Ohms O.D. Optical density O.Dfioo Optical density at 600nm wavelength P. Pseudomonas PBS Phosphate-buffered saline PCR Polymerase chain reaction
xvu pi Isoelectric point PMSF Phenylmethylsulphonyl fluoride PVDF Polyvinylidene difluoride rpm Revolutions per minute S Substrate [S] Substrate concentration S. Staphylococcus SDS Sodium dodecyl sulfate SDS-PAGE Sodium dodecyl sulfate polyacrylamide gel electrophoresis sec Second(s) TAB Tris-acetate/EDTA TBS Tris-buffered saline TCA Trichloroacetic acid TE Tris-EDTA TEMED N, N, N, N-tetramethylethylenediamine TEA Trifluoroacetic acid TLCK Tosyl lysyl chloromethyl ketone TNBSA Trinitrobenzenesulfonic acid TPCK Tosyl phenylalanyl chloromethyl ketone TTBS Tween 20/TBS V Volts v/v Volume per volume wt. Weight w/v Weight per volume X-gal 5-Bromo-4-chloro-3-indolyl-P-D-galactoside
xvui CHAPTER 1
LITERATURE REVIEW
A. Classification of proteases
Proteolytic cleavage of proteins is one of the most frequent modifications of proteins. The study of proteolytic enzymes began early in this century and by 1940, several proteases such as pepsin, chymotrypsin, trypsin, and papain had been crystallized leading to an initial understanding of proteolytic mechanisms (Barrett,
1994). Since then, numerous proteases have been purified and characterized.
Proteases are currently classified using three major criteria: mode of action on substrates, structural similarities, and catalytic mechanisms (Barrett, 1994). In the first case, the classification of enzymes by the type of reaction they catalyze has led to the standards of enzyme nomenclature established by the International Union of
Biochemistry and Molecular Biology. According to Enzyme Nomendature 1992
(Nomenclature Committee of the International Union of Biochemistry and
Molecular Biology, 1992), class 3 contains the hydrolases, and subclass 3.4 contains the peptide hydrolases which is an official name for proteases (or peptidases).
Proteases are composed of two sub-classes: exopeptidase and endopeptidase (Table
1.1). The exopeptidases act on the near end of polypeptide chains and can cut at
1 Action^ Group
Exopeptidase Aminopeptidases • ^ o o o o o - : - Dipeptidyl-peptidases e-eto-oocy; T ripeptidyl-peptidases
Carboxypeptidases : k k > o o o ^ Peptidyl-dipeptidases :> - o o o o W # Dipeptidases #L
i - o o o J é * Omega peptidases
»# h X X > O j
Endopeptidases > O O O O O O j
^ Open circles represent amino acid residues and filled circles are the residues that are cleaved off by an enzyme. The filled squares indicate the blocked termini that provide substrates for omega peptidases.
Table 1.1: Classification of proteases by mode of action on substrates (modified from
Barrett, 1994). either N-terminus (aminopeptidases, dipeptidyl-peptidases, and tripeptidyl- peptidases) or C-terminus (carboxypeptidases and peptidyl-dipeptidases). In addition, some exopeptidases are specific for dipeptides (dipeptidases) and others remove terminal residues that are not linked by a-carboxyl to a-amino group (omega peptidases). Endopeptidases preferentially cut inner regions away from either termini. The oligopeptidases are subset of endopeptidases that cut oligopeptides smaller than proteins.
The second method for classification of proteases considers the structural similarities among proteases. This method depends on the evolutionary and structural similarities based on primary structures of proteases. By this method,
“families” of proteases are classified (Rawlings and Barrett, 1993) and all the members of a family have originated from a single protein by divergent evolution
(Reeck, de Haen eta l., 1987). For example, both trypsin and chymotrypsin belong to the chymotrypsin family (Family SI) and thermolysin and Bacillus stearotbermopbilusnçMXidl endopeptidase belong to the thermolysin family (Family
M4). To identify statistically significant similarities in amino acid sequence, programs such as FAST?, FASTA, TFASTA, and RDF are used (Dayhoff, Barker e ta l. 1983; Lipman and Pearson 1985; Pearson and Lipman, 1988). In addition to families, a “clan” is composed of a group of families for which there is evidence of an evolutionary relationship despite the lack of statistically significant similarities in the primary sequences (Rawlings and Barrett, 1993). For instance, both trypsin
(Family SI) and Staphylococcus endopeptidase (Family S4) belong to Clan SA since both of them require a His, Asp, Ser catalytic triad for activity even though there is no statistically significant similarity in their primary structures.
The third and the most widely used classification method for proteases is by catalytic mechanism. According to this method, there are five classes of proteases
(Barrett, 1994). The serine proteases have an active center serine involved in the catalytic reaction, the cysteine proteases have a cysteine residue in the active center, the aspartic proteases depend on two aspartic acids for their activity, and the metalloproteases require metal ions for activity. The fifth class is actually not a class but sum of all proteases of unknown catalytic mechanisms. In this method, classification is based on the effects of inhibitors; characteristics of each class will be discussed in the following sections.
1. Serine proteases
Serine proteases were among the first enzymes to be studied extensively and are found in a wide variety of organisms including eukaryotes, bacteria, and viruses.
Members of this class include exopeptidases, endopeptidases, oligopeptidases, and omega peptidases (Neurath, 1985; Rawlings and Barrett, 1994a). Classically, serine proteases are involved in a host of physiological processes; trypsin, for example is known for its digestive role. However, some serine proteases act as regulators through the proteolytic activation of precursor proteins (Neurath, 1984; Perona and
Craik, 1995). A well known example of this category is the trypsinogen processing by enteropeptidase to produce active trypsin (Huber and Bode, 1978). In addition. some serine proteases are known to be involved in cell differentiation in organisms such as D rosophila (Chasan and Anderson, 1989; Smith and DeLotto, 1994).
As mentioned earlier, serine proteases are so named because of the involvement of a serine residue in the active site. In addition to a serine residue, serine proteases from many different families contain a catalytic triad consisting of three amino acids; serine, aspartate, and histidine (Carter and Wells, 1988; Rawlings and Barrett, 1994a). For example, in Figure 1.1, Ser 195, His 57, and Asp 102 are involved in the catalytic reaction of chymotrypsin. In the absence of a substrate. His
57 is unprotonated. However, when Ser 195 carries out a nucleophilic attack on a substrate. His 57 accepts a proton from the hydroxyl group of the Ser 195 forming an tetrahedral intermediate (Bachovchin and Roberts, 1978). The role of Asp 102 is to stabilize the positively charged His 57 in the tetrahedral intermediate by forming a hydrogen bond with His 57. In addition, formation of the tetrahedral intermediate is made possible by the existence of hydrogen bonds between the oxyanion atom of the substrate and two main chain NH groups (Gly 193 and Ser 195 in this case). The proton held by the His 57 is then donated to the nitrogen atom of the substrate cleaving the peptide bond. As a result, the acyl enzyme intermediate is formed.
Finally, the déacylation reaction is accomplished when hydrolysis occurs by the addition of a water molecule thereby completing the proteolytic cleavage reaction.
Traditionally, inhibitors have been widely used to determine the catalytic classification of proteases. In the case of serine proteases, inhibitors such as 3,4- dichloroisocoumarin (3,4-DCI), phenylmethylsulfonyl fluoride (PMSF), and diisopropyl fluorophosphate (DFP) are most commonly used (Barrett, 1994). Figure 1.1; Schematic representation of the steps involved in catalysis by a typical serine protease'"^.
‘ Chymotrypsin is used as an example of a typical serine protease and, as a result, all numbers of amino acids follow those of chymotrypsin.
^ Modified from Beynon & Bond, 1989. Serine Protease Active Site -His 57
Ser 0 —H —H ------0 —C^^^Asp 102 195
/H \ / H -N—H—C------C -N-H—C C- c \ II \ II Substrate o Pr o
H H His 57 / \ N N— Scr 195 Ser^ 0 H-r4 H O '- A s p 102 \ 195 ' ^ Gly 193 / H / H -N-H-C ------C - N - H - C - c - Tetrahedral \ \ Intermediate 0
H H / \ N N— Ser 195 \ Gly 193 \ -His 57 X " . / X —H O — C Asp 102
/ H Acyl Enzyme ° H-N-H-C ------C- Intermediate \ I I H / ? ,• 0 HzO -N-H-C His 57 X \ \ /X Q—H N H ------O A s p 102 195 ^ II
/H -N-H-C C-OH \ II P. 0
Figure 1.1 However, not all serine proteases are inhibited by all serine protease inhibitors. For example, propyl oligopeptidase is inhibited by DFP but not by 3,4-DCI or PMSF
(Kalwant and Porter, 1991). In contrast, glutamyl endopeptidase from
Staphylococcus is inhibited only by 3,4-DCI but not by neither DFP nor PMSF
(Barrett, 1994). Therefore, it is essential to examine a protease with a wide variety of different protease inhibitors before classifying a protease based on catalytic mechanism.
2. Cysteine proteases
All known cysteine proteases (formally known as thiol proteases) require a cysteine residue in their active site for catalysis reactions. However, similar to serine proteases, the cysteine residue is not the only requirement for the activity. There are about 20 famihes of cysteine proteases known and all of them require a catalytic dyad of cysteine and histidine (Rawlings and Barrett, 1994b). For example, in the case of papain which is the best known cysteine protease, cysteine 25 and histidine
159 are the key residues of the active site (Figure 1.2). The mechanism of catalysis in cysteine proteases shares many similarities with those of serine proteases. First, cysteine 25 acts like serine 195 of chymotrypsin because the sulfur atom in the cysteine residue is the attacking nucleophile just like the oxygen atom of serine 195 in chymotrypsin. Second, histidine 159 of papain has exactly same function as histidine 57 of chymotrypsin. In both cases, the histidine residue is protonated when a substrate is present and, as a result, is involved in the formation of the tetrahedral intermediate. Third, two main chain residues, glutamic acid 19 and cysteine 25, are
8 Figure 1.2: Schematic representation of the steps involved in catalysis by a typical cysteine protease*’^.
‘ Papain is used as an example of a typical cysteine protease.
^ Modified from Beynon & Bond, 1989. Cysteine Protease Active Site y^His 159 /\ Cys S - H ------
/H V .H -N —H-C------C—N-H -C ----- C- \ \ His 159 Pi o Cys^S H-N
His 159 H,0. XX / - C Cys S-H N^^^^-H
/ H -N —H-C— -C-OH \ Pi O
Figure 1.2
10 also involved in the formation of the tetrahedral intermediate by forming hydrogen bonds with the substrate. The same function is carried out by glycine 193 and serine
195 in the case of chymotrypsin. Fourth, the peptide bond of the substrate is cleaved by protonation from histidine 159 to the substrate forming an acyl enzyme intermediate in the same manner as chymotrypsin. Finally, the déacylation reaction is achieved by the addition of a water molecule in both cases. Therefore, the only difference in the mechanisms of cysteine proteases and serine proteases is that cysteine proteases do not require an aspartic acid residue for stabilization of the tetrahedral intermediate.
Many of the cysteine proteases, including the papain and calpain families, are susceptible to rapid and irreversible inactivation by E-64 (Barrett, Kembhavi era/.,
1982; Parkes, Kembhavi era/, 1985). E-64 [N-(L-3-rran5carboxyoxiran-2-carbonyl)-
L-leucyl]-amino(4-guanidino)butane] is a peptide epoxide isolated from an extract of
Aspergillus japonicus. The inhibitory mechanism of E-64 is not clear but it is thought that the rrana epoxide group alkylates the active site cysteine residue (Durm,
1990). In addition to E-64, natural protein inhibitors known as cystatins and peptidyldiazomethanes are also known to irreversibly inhibit many cysteine proteases (Barrett, 1994).
3. Aspartic proteases
Aspartic proteases are formally known as acid proteases due to the low pH optimum of this group. The requirement for acidic conditions led to the conclusion that a carboxyl group is involved in activity (Hartsuck and Tang, 1972). As
11 compared to the serine proteases, there are relatively few aspartic proteases. These include the gastric proteases (pepsin, gastricsin, chymosin), renin, cathepsin D, and several proteases isolated from fungi (Tang, 1979; Polgar, 1989). Unlike serine and cysteine proteases, aspartic proteases catalyze the cleavage of peptide bonds without the use of nucleophilic attack by a serine or cysteine residue mainly because the carboxyl groups of the two aspartic acids are not potent nucleophiles (Hofmann,
Dunn e ta l., 1984). Instead, the nucleophilic attack is carried out by a water molecule and as a result, there is no intermediate covalent bond formation between the protease and the substrate (Polgar, 1989). As depicted in Figure 1.3, aspartic acid residues 32 and 215 are involved in the active site of porcine pepsin and both residues are close enough to share a hydrogen bond between two of their oxygen atoms. In addition to the two aspartic acid residues, water molecules are also involved in the formation of the tetrahedral intermediate and subsequent cleavage of the peptide bond (Dunn, 1990).
The best characterized examples of aspartic proteases such as pepsin, chymosin, cathepsin D, and renin are endopeptidases which are reversibly inhibited by pepstatin A (Barrett, 1994). Pepstatin A is a peptide-like compound (isovaleryl-
Val-Val-AHMHA-Ala-AHMHA)[AHMHA = (3S,4S)4-amino-3-hydroxy-6-methyl- heptanoic acid] secreted by 5 &v?/>ton 2 7 ce5 (Salvesen and Nagase, 1990).
4. Metalloproteases
Most metalloproteases contain zinc as an essential metal ion for their activity
(Jiang and Bond, 1992). The role of zinc in metalloproteases can be divided into
12 Figure 1.3: Schematic representation of the steps involved in catalysis by a typical aspartic protease''^.
‘ Porcine pepsin is used as an example of a typical aspartic protease and, as a result, all numbers of amino acids follows those of porcine pepsin.
Modified from Beynon & Bond, 1989.
13 Substrate / H —N —H —C ------z^C—N —H—C------C— I I I \ I I J ; o ., p,' o o Hv H / H » / -N —H-C ------C--N—H-C- -C - Asp 215 of A Asp 32 Pi “'"O,\ H Aspartic Protease Active Site
Asp 215 Asp 32 Tetrahedral Intermediate
/H /cr - N - H - c — qC I o Pi
O o' '' 0 ^ /
A%)215 Asp 32
Figure 1.3
14 four categories: catalytic, structural, regulatory, and noncatalytic (Vallee, 1983).
Zinc has a catalytic role when it is essential for the activity of the protease; carboxypeptidase and thermolysin are examples of this type. Zinc also can be considered structural when it is required only for structural stability of the protease.
Often, structural zinc stabilizes the quaternary structure of proteases. For example, a zinc atom is required to dimerize Badllus subtilis a-dimylast. Zinc also plays a regulatory role when it acts as either an activator or an inhibitor, but activity is still present in the absence of zinc ions. For instance, zinc acts as an activator of bovine lens leucine aminopeptidase and as an inhibitor of fructose-1,6-bisphosphatase
(Vallee, 1983). Finally, zinc has a noncatalytic role when zinc is not involved in any of the above three roles. For example, zinc is required for E. co/f alkaline phosphatase but the role of the zinc atom is not clear (Coleman and Gettins, 1983).
Therefore, in the absence of specific knowledge on how the metal acts, the requirement is just referred to as noncatalytic.
The coordination of the zinc atom in the active site of the protease has been demonstrated in several metalloproteases including bovine pancreatic carboxypeptidase A (Lipscomb, Hartsuck e ta l., 1969), Bacillus tbermoproteolyticus thermolysin (Holmes and Matthews, 1982), Pseudomonas aeruginosa pseudolysin
(Thayer, Flahertyeta l., 1991), and Bacilluscereusnexittal protease (Stark, Pauptit et al., 1992). The three bacterial proteases listed above are closely related to one another and contain a common HEXXH motif (Jiang and Bond, 1992). This
HEXXH motif has been conserved in many of the metalloproteases and, in some cases, there is another conserved glutamic acid remotely located from the HEXXH
15 motif (Vallee and Auld, 1990). As illustrated in Figure 1.4, the zinc ion attracts a water molecule and uses the water molecule as the fourth tetrahedral site. The other ligands to the zinc are histidine 142, histidine 146 and glutamic acid 166 in case of thermolysin (not shown in Figure 1.4) (Colman, Jansonius eta l., 1972). A water molecule is displaced when a carbonyl group of the substrate is attracted to the zinc ion, but still remains in the active site. The water molecule also interacts with glutamic acid 143 by a hydrogen bond. The carboxyl group of the glutamic acid serves as a general base to remove a proton and assist the attack of the same water molecule on the carbonyl group of the substrate. A proton is then transferred from glutamic acid 143 to the leaving nitrogen atom on the substrate. Therefore, glutamic acid acts as a proton shuttle just as the histidine residues in both the serine and cysteine proteases.
The design of metalloprotease inhibitors has primarily focused on chelating the zinc atom. Among many known metalloprotease inhibitors, ophenanthroline
(1,10-phenanthroline) is the most useful inhibitor because it has a greater affinity for zinc than calcium (stability constants 2.5 x 10* M ‘ and 3.2 M'% respectively). As a result, ophenanthroline can be used in the presence of 10 mM Ca^, which is required for the stability (or activity) of many proteases (Barrett, 1994). In addition to ophenanthroline, phosphoramidon [N-(a-L-rhamnopyranosyl- oxyhydroxyphosphinyl)-Leu-Trp] is also an effective inhibitor for some metalloproteases especially many bacterial metalloproteases of the thermolysin family (Umezawa, 1976). However, many other metalloproteases are not inhibited by phosphoramidon and, as a result, it is not a diagnostic inhibitor for all
16 Figure 1.4; Schematic representation of the catalysis of peptide bond cleavage carried out by a metalloprotease*’^.
* * Thermolysin from B. tbermoproteolyocus is used as an example in this figure.
^ Modified from Beynon & Bond, 1989.
17 -Zn- 7 His 231
H H i/H H cC Substrate Metalloprotease q Active site N^H
Glu 143 J
-Znr 7 His 231 I O / C = 0 H N ^ ^ - H I . H j/^ '-H O P
/ '■ ^ N - H 143
.0 C-H
Glu 143
Zn- His 231 y \ v ° H H > H O''
Glu 143
Figure 1.4 18 metalloproteases. Inhibition of a metalloprotease by an unspecific chelators such as
EDTA is unreliable because many proteases of other types, such as the members of the subtilisin family (serine proteases), require cations (especially Ca^) (Barrett,
1994).
B. Pseudomonas aerugiiiosa as an opportunistic pathogen
Pseudomonas sçeâes comprise a large group of Gram-negative microorganisms found in wide range of different environments. Among them,
Pseudomonas aeruginosa is a tremendously versatile organism which can utilize numerous compounds as carbon sources and can successfully colonize in a variety of environments that contain a very low level of nutrients (Rhame, 1980; Botzenhart and Doring, 1993). Therefore, given the ability of P. aem ginosa to adapt to many different environments, it is not surprising that P. aeruginosa is capable of causing a wide range of infections in humans. However, even though P. aeruginosa is the most frequently isolated nonfermentative bacillus found in clinical specimens, it is not an obligate pathogen but an opportunistic pathogen associated with infections of the immunocompromised host. In other words, P. aemginosa seldom results in pathogenicity in healthy humans, but is a significant cause of bum wound and nosocomial infections in immunocompromised patients. P. aemginosa is also involved in other infections such as skin infections, gastrointestinal infections, central nervous system infections, comeal infections, bone infections, and urinary tract infections (Pollack, 1990).
19 In addition to the various infections described above, P. aeruginosa is the most frequently reported pathogen among cystic fibrosis (CF) patients, cultured in specimens from 61% of all CF patients (Govan and Deretic, 1996). CF is the most common genetic disease among Caucasian populations, with an incidence of 1 in
2,500 births and carrier frequency of 1 in 25 (Bye, Ewig e ta l., 1994). The spectrum of bacteria associated with CF respiratory infection is comparatively restricted. The microbial pathogens most commonly reported include Stapbylococcus aureus and
Hemophilus inSuenzae eatXy in life and later, P. aeruginosa (Govan and Deretic,
1996). The reason why P. aeruginosa preferentially colonizes the respiratory tracts of CF patients remains obscure. However, considering the large number of virulence factors produced by P. aeruginosa, it is possible that any one or a combination of virulence factors are involved in the infection.
Cell-based structures are believed to contribute to the virulence of P. aeruginosa. Hazlett, eta l. suggested that pili act as adhesins allowing binding of P. aeruginosa to epithelial cells since pili negative mutants exhibited decreased virulence in animal models (Hazlett, Moon eta l., 1991). Flagella also appear to be involved in the virulence of P. aeruginosa because nonmotile mutants are less virulent than wild type (motile) strains (Monde, Drake e ta l., 1987; Drake and
Monde, 1988).
Treatment of P. aeruginosa infections is very difficult due to several reasons, including its ability to survive in hostile environments, multiple virulence factors, and intrinsic resistance to many antibiotics. P. aeruginosa strains are not susceptible to many conventional antibiotics. For example, the minimal inhibitory
20 concentration (MIC) of ampicillin against P. aeruginosa is greater than 128 pg/ml compared to just 4 pg/ml against E. co liÇ&eWido and Hancock, 1993). There are several reasons why P. aeruginosa is more resistant to many antibiotics than other
Gram-negative bacteria. First, P. aeruginosa has lower outer-membrane permeability than other Gram-negative bacteria due to some ill-defined structural differences (Nikaido and Hancock, 1986). Second, P. aeruginosa possesses several mechanisms to oppose the effect of antibiotics. For example, P. aeruginosa has at least seven penicillin-binding proteins (PBP) to inactivate the antibiotic (Bellido and
Hancock, 1993). Finally, P. aeruginosa produces both chromosomal and plasmid based 3-lactamase to hydrolyze P-Iactam based antibiotics such as ampicillin and carbenicillin (Hancock and Woodruff, 1988). Most likely, a combination of all the factors mentioned above is involved in the nature of the antibiotic resistance in P. aeruginosa. In addition to p-lactam based antibiotics, P. aeruginosa is also resistant to other types of antibiotics such as aminoglycosides and quinolone-based antibiotics. The nature of resistance to these types of antibiotics is not well understood. However, one study by Bellido and Hancock proposed that resistance against aminoglycosides is associated with a smooth-to-rough transition in lipopolysaccharide (LPS) (Bellido and Hancock, 1993). This LPS alteration decreases the affinity of LPS for aminoglycosides, resulting in lower levels of uptake across the outer membrane.
21 c. Non-proteolytic Pseudomonas aemginosa virulence factors
Pseudomonas aemginosa produces numerous virulence factors. All known
P. aemginosa virulence factors are extracellular and some are proteases while others are non-proteolytic enzymes. At least three known non-proteolytic virulence factors are secreted by P. aemginosa: exotoxin A, exoenzyme S, and phospholipase C.
However, among those three, exotoxin A is the only virulence factor which is well characterized. The functions of the other two factors have yet to be established.
I. Exotoxin A
P. aemginosa exotoxin A (ETA) was first identified 1961 and has been regarded as the most potent virulence factor produced by P. aemginosa (Liu, Abe et al., 1961; Galloway, 1993). ETA is a monomeric protein of 613 amino acids with a molecular weight of 66 kDa and is produced by most P. aemginosa strains (Bjom,
Vasil e ta l., 1977; Pollack, Taylor eta l., 1977). ETA is classified as one of the adenosine diphosphate (ADP)-ribosylating toxins; this group also includes many other potent toxins such as diphtheria toxin, cholera toxin, and pertussis toxin. All
ADP-ribosylating toxins utilize nicotinamide adenine dinucleotide (NAD) as a substrate and transfer the ADP-ribose moiety of the NAD to the target protein impairing the normal function of the protein. In the case of ETA and diphtheria toxin, the target is the diphthamide residue of the eukaryotic elongation factor 2
(EF2) and, by blocking the function of EF2, protein synthesis is inhibited in the target cell ultimately resulting in cell death (Iglewski, Liu eta l., 1977). Because by definition, both ETA and diphtheria toxin are enzymes, one molecule of the toxin is
22 sufficient to kill a host cell; this has been demonstrated in the case of diphtheria toxin (Yamaizumi, Mekada e ta l., 1978). ETA is lethal for vertebrates in very small quantities with an LDso for mice in the range of only 2.5 pg/kg (Gill, 1982).
Three-dimensional structural analysis revealed that ETA has three domains: a receptor binding domain (domain I), membrane translocation domain (domain II), and enzymatic domain (domain HI) (Allured, Collier eta l., 1986; Brandhuber,
Allured e ta l., 1988). ETA binds to the az-macroglobulin/low-density lipoprotein
(LDL)-like receptor and enters cells by receptor-mediated endocytosis (Kounnas,
Morris e ta l., 1992). Following internalization, ETA is activated within the endosome by ftirin cleavage at RXXR resulting in a 37 kDa active fragment in the cytosol (Ogata, Chaudhary e ta l., 1990; Gordon and Leppla, 1994; Gu, Gordon e t al., 1996).
Interestingly, there is a cross-reactive epitope that exists in several of the
ADP-ribosylating toxins including ETA, diphtheria toxin, cholera toxin, and pertussis toxin (Galloway, 1993). This indicates that those toxins may have evolved from a common ancestral protein that has diverged through time. Recent development further supports this idea. For example, the cholera toxin gene is believed to have originated from the filamentous bacteriophage CTXO suggesting the possibility of horizontal gene transfer among the pathogenic microbes (Waldor and Mekalanos, 1996). In addition, two “pathogenicity islands” have recently been identified in E. co//(Swenson, Bukanov eta l., 1996). A pathogenicity island is a large multigene chromosomal segment containing several virulence genes, insertion sequence elements, and phage genes. This discovery suggests that some mobile
23 elements may have facilitated the spread of virulence genes among bacteria further supporting the idea that many of the bacterial toxins may have originated from a common ancestor.
2. Exoenzyme S
In addition to ETA, P. aeruginosa secretes another ADP-ribosylating enzyme, exoenzyme S (exo-S), which exists in two forms: a 53 kDa inactive form and a 49 kDa active form (Nicas and Iglewski, 1985). Exo-S differs from ETA because the target protein for exo-S is not eukaryotic EF2. Instead the enzyme transfers ADP-ribose from NAD to a number of eukaryotic proteins including GTP- binding proteins (Cobum, Yatt e ta l., 1989; Cobum and Gill, 1992). For its enzymatic activity, exo-S requires a eukaryotic cellular factor in vitro (Cobum, Kane eta l., 1991). This factor, designated FAS (factor activating exo-S), is a member of the eukaryotic protein family collectively known as 14-3-3 proteins (Fu, Cobum e t al., 1993). Various functions have been reported for other members of the 14-3-3 family, including phospholipase A2 activity and regulation of tyrosine hydroxylase and tryptophan hydroxylase activities (Isobe, Ichimura e ta l., 1991; Zupan, Steffens eta l., 1992). However, a precise role for exo-S and its associated ADP- ribosyltransferase activity has yet to be established.
3. Phospholipase C
In addition to ETA and exo-S, P. aeruginosa also secretes a 78 kDa protein with phospholipase and hemolytic activity (Liu, 1979). This protein is designated
24 phospholipase C (PLC), and is described as a hemolytic toxin which may be associated with virulence of P. aemginosa. PLC has been shown to degrade phospholipids commonly found in eukaryotic membranes resulting in the production of diacylglycerol which can cause toxic effects in animals (Berka and
Vasil, 1982; Besterman, Duronio eta l., 1986). Initial studies also indicate that, when injected into mice, purified PLC can cause various pathologic effects including death (Berk, Brown eta l., 1987). Another study demonstrated that purified PLC induces inflammatory responses in mice (Meyers and Berk, 1990). However, the exact role of phospholipase C in the virulence of Pseudomonas aemginosa is not yet clear.
D. Pseudomonas aemginosa proteases
Several known extracellular proteases are produced by P. aemginosa. The list includes pseudolysin (elastase), alkaline protease, lysine-specific protease, LasA, and LasD. The specific functions or adaptive advantages gained by P. aemginosa by secreting these proteases are not clear. However, these proteases are believed to be involved in providing nitrogen-rich digestion products to allow growth of the organism (Steadman, Heck eta l., 1993). In this section, known characteristics of each protease will be examined.
1. Pseudolysin (elastase)
Among the many extracellular virulence factors produced by P. aemginosa, pseudolysin (elastase) has long been considered to be of major significance. It was
25 originally purified by Morihara, e ta l. and has been classified as a zinc metalloprotease (Morihara, Tsuzuki e ta l., 1965). Pseudolysin was originally termed elastase due to its ability to digest elastin tissues. However, concern has arisen regarding the appropriateness of the name ‘elastase’ for a couple of reasons. First, the proteolytic activity of elastase far exceeds its elastolytic activity in vitra, the specific elastolytic activity of P. aem ginosa elastase appears to be approximately one fourth that of Bacillus thermolysin, and one seventh that of the elastolytic activity of purified human neutrophil elastase (Peters and Galloway, 1990). In addition, P. aemginosa elastase shares significant homology vrith another zinc metalloprotease, thermolysin from Bacillus tbermoproteolyticus, in both amino acid sequence and three dimensional structure (Thayer, Flaherty eta l., 1991; Hase and Finkelstein,
1993). As a result, P. aemginosa elastase has been renamed as pseudolysin (EC
3.4.24.26) and is classified as a member of the thermolysin family (Family MH)
(Morihara, 1995).
P. aemginosa pseudolysin is a 33 kDa protein with an isoelectric point (pi) of
5.9 (Sever and Iglewski, 1988). The optimal pH range for hydrolysis is between pH
7 to 8, and the enzyme is stable at pH 6 to 10 (Morihara, Tsuzuki e ta l., 1965).
Enzymatic activity of pseudolysin is inhibited by metal chelators such as ethylenediaminetetraacetate (EDTA) and ophenanthroline, but is not affected by other inhibitors such as diisopropyl fluorophosphate (DFP) and tosyl lysyl chloromethyl ketone (TLCK). The enzyme contains one zinc atom per molecule and zinc is essential for activity (Morihara and Homma, 1985).
26 The specificity for cleavage of peptide bonds by pseudolysin has been established using substrates such as casein, hemoglobin, ovalbumin, fibrin, insulin 13- chain, and various other synthetic peptides (Morihara and Tsuzuki, 1966; Morihara,
Oka e ta l., 1971). Collectively, the results indicate that pseudolysin cleaves on the amino side of hydrophobic or aromatic amino acid residues similar to other zinc metalloproteases belonging to the thermolysin family (Morihara, 1995). For example, the substrate carbobenzoxy-Gly-X-NHz is most susceptible to digestion by pseudolysin when X is an aromatic or bulky amino acid such as phenylalanine, leucine, or tyrosine (Morihara and Tsuzuki, 1966).
In addition to the above substrates, pseudolysin also digests elastin which is the functional protein component of the elastic fiber found in connective tissues such as the artery wall and the air sacs of the lungs (Sandberg, Soskel eta l., 1981; Urry,
1983). The exact site which pseudolysin recognizes and cuts is not clear, mainly due to the highly complex structure of elastin (Galloway, 1991). However, a couple of reports suggest that elastin digestion by pseudolysin is not very specific. First, it has been reported that pseudolysin is less specific than pancreatic elastase but is capable of a more complete digestion of elastin (Saulnier, Curtil eta l., 1989); and second, elastolytic activity of pseudolysin increases several-fold when combined with LasA which lacks significant elastolytic activity by itself (Peters and Galloway, 1990). In this case, it was hypothesized that LasA first nicks the elastin structure, then pseudolysin acts on the nicked elastin to enhance elastolytic degradation. In both cases, the results suggest that pseudolysin digests elastin as a general protease, not as a specific elastase. The hypothesis that pseudolysin acts as a general protease,
27 especially in combination with other protease(s), is further supported by a recent report (Park and Galloway, 1995). When heat-killed Stapbylococcus aureus cells were used as substrate, pseudolysin enhanced the staphylolytic activity of other proteases such as LasA and LasD even though pseudolysin itself does not have any staphylolytic activity, thereby indicating that pseudolysin acts as a general protease when the cell wall structure is compromised by either LasA or LasD.
The structural gene for pseudolysin has been cloned and the amino acid sequence has also been determined (Bever and Iglewski, 1988; Fukushima,
Yamamoto e ta l., 1989). Similar to the homology noted with the three dimensional structure of thermolysin, the primary sequence of pseudolysin also exhibit great homology with thermolysin. On the basis of sequence comparison and three dimensional structure analysis, critical residues in the active site have been predicted in pseudolysin (Bever and Iglewski, 1988; Thayer, Flaherty e ta l., 1991). These residues include histidine 140, histidine 144 and glutamic acid 164 as zinc ligands, glutamic acid 141 as the active center, and tyrosine 155, histidine 223 and aspartic acid 221 for substrate binding. Therefore, pseudolysin also contains the classical
HEXXH motif found in many other metalloproteases.
Like many other bacterial proteases, pseudolysin is synthesized as a much larger precursor protein and is processed to a smaller active fragment (Kessler and
Safrin, 1988; Wandersman, 1989). As depicted in Figure 1.5, pseudolysin is initially translated as a 60 kDa prepropseudolysin and translocated to the periplasm resulting in a 56 kDa propseudolysin I. The propseudolysin I is then rapidly processed to a 36
28 Cytoplasm IM Periplasm PG OM Medium
(20 kDa)
(60 kDa) (56 kDa)
Active (36 kDa) Pseudolysin (33 kDa)
P, prepropseudolysin; Pi, propseudolysin I; Pn, propseudolysin II; P20, 20 kDa propeptide; IM, inner membrane of P. aeruginosa; OM, outer membrane; PG, pepddoglycan layer.
Figure 1.5: Proposed sequence of events involved in P. aeruginosa pseudolysin
(elastase) secretion (modified from Kessler and Safrin, 1988).
29 kDa propseudolysin H after removal of the 20 kDa propeptide. A final transport event across the outer membrane results in the 33 kDa active pseudolysin.
The cleavage from the propseudolysin I to propseudolysin H is believed to be the autoproteolytic event (Mclver, Kessler eta l., 1991; Mclver, Olsen eta l., 1993).
When the histidine 223 residue in the active site of pseudolysin is replaced with either aspartic acid or tyrosine, the elastolytic activity of the mutant strain decreases dramatically indicating that histidine 223 is required for the activity of pseudolysin.
Furthermore, the mutant pseudolysin is not processed from the propseudolysin form to the 33 kDa form. When the active 33 kDa pseudolysin is provided in trans, however, the mutant protein is processed properly to the 33 kDa form (Mclver,
Kessler e t al., 1991). These results suggest that the functional pseudolysin is required for proper processing, and the autoprocessing event is essential.
As mentioned above, pseudolysin is one of many proteases from both prokaryotes and eukaryotes which is produced as a larger inactive precursor and later activated post-translationally. Activation of these proteases often involves proteolytic removal of a propeptide region which is covalently attached to either the amino- or carboxyl terminus of the active protease. The propeptide region is thought to be involved in prevention of undesirable proteolysis (Neurath, 1984;
Wandersman, 1989) as demonstrated for subtilisin E of BadUus subtüisÇJkexmrdi,
Takagi e ta l., 1987). Recently, studies have also shown that propeptides from many unrelated proteins are essential for proper folding of the enzymes, hence the term intramolecular chaperones (IMG) (Inouye, 1991; Shinde and Inouye, 1993).
Members of the IMG family are synthesized as pre-proenzymes with an amino
30 terminal signal sequence, followed by a propeptide domain and the mature enzyme sequence (Inouye, 1991). The number of proteases belonging to the IMC family is rapidly growing and includes both eukaryotic and prokaryotic proteins such as alkaline protease from the yeast Yarrowia lypolytica (Fabre, Nicaud eta l., 1991), carboxypeptidaseY ftom SaccbaromycescerevisiaeÇ^\n\he.TandSoiemen, 1991), and a-lytic protease from the Gram-negative bacterium Lysobacter enzymogenes
(Silen, 1989).
The 20 kDa propeptide of pseudolysin is also believed to act as an intramolecular chaperone (IMC) directing pseudolysin translocation across the outer membrane as well as guiding pseudolysin into its active conformation (Mclver,
Kessler e ta l., 1995; Braun, Tommassen e ta l., 1996). When the nucleotide sequence representing the propeptide is deleted from the pseudolysin structural gene lasB, pseudolysin can not be detected in the supernatant suggesting that the propeptide region is required for secretion. Furthermore, the propeptide appears to interact with the mature pseudolysin directly since, when the independently expressed propeptide is mixed with mature pseudolysin, they co-precipitate in an immunoprécipitation assay using and-pseudolysin antibody (Mclver, Kessler eta l.,
1995). In addition, the propeptide is also considered to act as an inhibitor preventing premature pseudolysin activity in the periplasm (Kessler and Safrin, 1994).
2. LasA
LasA is another protease produced by P. aeruginosa which exhibits elastolytic and proteolytic activity. The lasA gene was first isolated from a mutant
31 strain of P. aeruginosa P AO-E64 which lacks significant elastolytic activity, yet produces pseudolysin antigenically indistinguishable from wild-type pseudolysin
(Ohman, Cryz e ta l., 1980). PAO-E64 was originally described as a temperature- sensitive mutant of the pseudolysin structural gene, but was later characterized as a mutant that contains a mutation in lasA gene, not in the pseudolysin structural gene
{lasB) (Schad, Bever e ta l., 1987). Since then, lasA has been cloned from a clinical isolate (FRD2) and complementation of mutation was achieved by addition of the lasA gene in PAO-E64 (Goldberg and Ohman, 1987a). Originally LasA was proposed to be involved in the processing of pseudolysin since a lasA negative mutant strain (FRD2128) appeared to accumulate a 47 kDa protein that reacted with anti-pseudolysin antiserum (Goldberg and Ohman, 1987b). However, the mutant strain produced and secreted pseudolysin that is biochemically identical to wild-type pseudolysin.
In 1988, Schad and Iglewski cloned and sequenced the lasA gene from PAOl and suggested that LasA may be outer membrane protein since LasA is associated with the outer membrane when expressed in E. c o li(Schad and Iglewski, 1988).
However, the published lasA sequence was later found to be incorrect and the correct sequence was published by Darzins, e ta l. (Darzins, Peters eta l., 1990). In addition, Peters and Galloway successfully purified the LasA protein from the culture supernatant fraction of strain PAOl, thereby demonstrating that LasA is not membrane-associated (Peters and Galloway, 1990). Furthermore, the authors demonstrated that LasA enhances elastolytic activity associated with several other proteases including P. aeruginosa pseudolysin, proteinase K, thermolysin, and
32 human neutrophil elastase. The authors also suggested that LasA acts directly on the elastin substrate causing modification that results in subsequent digestion of elastin by pseudolysin.
In 1992, Peters, eta l. identified P-casein as a substrate for LasA demonstrating that LasA cleaves at Asn-Lys-Lys->IIe-GIu-Lys-Phe (Peters, Park e t al., 1992). Later, LasA was found to possess staphylolytic activity since LasA caused lysis of heat-killed Staphylococcus aureus (¥Less\e.t, Safrin e ta l., 1993). A staphylolytic enzyme was first discovered in P. aeruginosa by Burke and Pattee
(Burke and Pattee, 1967) and further characterized by Lache, eta l. (Lache, Hearn e t al., 1969). This enzyme has molecular weight of 19 kDa and is active against
Staphylococcusstcaim and pentaglycine (Brito, Falcon e ta l., 1989). A second enzyme possessing staphylolytic activity was also discovered in P. aeruginosa strain
PAKS I by Camicero, eta l. (Camicero, Falcon eta l., 1990). This second staphylolytic enzyme has molecular weight of 33 kDa, an isoelectric point between
7.35 to 8.15, and an optimum pH value of 8.0 for casein hydrolysis. Kessler, etal. suggested that LasA is the first staphylolytic enzyme discovered by Burke and Pattee because of their similarities in size and activity (Kessler, Safrin e ta l., 1993).
However, evidence exists to refute this claim since the purification methods for the two proteins are quite different. The staphylolytic enzyme elutes around pH 8.0 from a CM-cellulose column when a phosphate buffer gradient from pH 6.0 to pH
9.5 is applied (Brito, Falcon eta l., 1989). However, under these conditions, LasA does not elute until the pH approaches 9.5 since the isoelectric point of LasA is 9.5.
Also, the staphylolytic enzyme can be eluted from a DEAE-Cellulose column (pH
33 8.5) with water (Lache, Hearne ta l., 1969); again, LasA can not be eluted with water under these conditions because of its high isoelectric point.
In addition to LasA, another protease with similar activity, LasD, was purified by Park and Galloway from P. aeruginosa strains PA220 and FRD2128
(Park and Galloway, 1995). The characterization of LasA and LasD is the primary focus of this dissertation, and properties of LasD as well as further characterization of LasA will be discussed in later chapters.
3. Alkaline protease
In addition to pseudolysin, P. aeruginosa also secretes alkaline protease which was originally purified and characterized by Morihara (Morihara, 1964;
Morihara and Homma, 1985). The enzyme was named alkaline protease because the optimum pH for casein hydrolysis is pH 8 to 9. Alkaline protease has a molecular weight of 49 kDa and a pi of 4.1. Alkaline protease is classified as a metalloprotease because its enzymatic activity is inhibited by metal chelators such as
EDTA and o-phenanthroline. Like pseudolysin, alkaline protease digests several denatured protein substrates such as casein, hemoglobin, and fibrin. However, pseudolysin has approximately 10 times higher specific activity than alkaline protease against casein (Morihara and Homma, 1985). When various synthetic peptides were used as substrates, benzyloxycarbonyl-Ala-Gly-Gly-Leu-Ala was the best substrate but several different synthetic peptides were also hydrolyzed by alkaline protease (Morihara, Tsuzuki eta l., 1973).
34 The secretion of alkaline protease differs from that of other P. aeruginosa proteases; pseudolysin, phospholipase C and exotoxin A are éiffected by xcp mutations, however none of the xcp mutations affect alkaline protease secretion, leading to the conclusion that alkaline protease has its own specific secretion pathway (Wretlind and Pavlovskis, 1984; Filloux, Murgier e ta l., 1987).
Interestingly, when the structural gene for alkaline protease is expressed in E. coli, the alkaline protease is synthesized and properly secreted to the medium suggesting that alkaline protease utilizes similar pathways as some E. co//extracellular proteins
(Guzzo, Murgier e ta l., 1990). Further studies revealed that alkaline protease uses a secretion mechanism similar to Erwinia cbrysantbemi^xoieases and E. co lia- hemolysin (Guzzo, Duong eta l., 1991; Guzzo, Pages e ta l., 1991).
Howe, e ta l. demonstrated a requirement for alkaline protease in establishing
P. aeruginosa PA103 comeal infections using alkaline protease-deficient mutants in a mouse eye model (Howe and Iglewski, 1984). Another study suggested that alkaline protease may play an important role in comeal keratitis since alkaline protease can degrade comeal proteins (Twining, Kirschner e ta l., 1993). By contrast, alkaline protease does not appear to be involved in lung epithelial-cell damage
(Wiener-Kronish, Sakuma eta l., 1993). Despite the implication of involvement in pathogenicity, the exact role of alkaline protease has yet to be revealed.
4. Lysine-specific protease
P. aeruginosa also produces a lysine-specific protease (Ps-1). Ps-1 has a molecular weight of 30 kDa and a pH optimum of 8 to 9. Inhibitors of cysteine
35 proteases, aspartic proteases, and metalloproteases do not inhibit the activity of Ps-1.
The activity is completely inhibited however, by the serine protease inhibitor TLCK, although DFP only partially inhibits the activity (Elliott and Cohen, 1986).
Therefore, Ps-1 can be classified as a serine protease. Ps-1 exhibits one of the most restricted specificities known for proteases: only peptide, ester, and amide bonds containing the carbonyl group of lysine are hydrolyzed by this protease (Elliott and
Cohen, 1986). Other than its restricted specificity, little is known about the lysine- specific protease or any role it may play in pathogenicity.
E. Regulation of Pseudomonas aeruginosa proteases
The expression of proteases in P. aeruginosa is not constitutive but rather is a highly regulated event. Initial studies regarding regulation of protease expression in
P. aeruginosa began with pseudolysin. Several lines of evidence indicate that the expression of pseudolysin requires an activator(s). First, the /a&g (structural gene for pseudolysin) promoter is about 1000-fold more active in P. aeruginosa than in E. coli
(Gambello and Iglewski, 1991). Although the /a&F promoter may not be effectively recognized in E. co li it is also possible that a required activator for the promoter is missing in E. coli Second, the expression of lasB is growth phase dependent, and is initiated during late logarithmic to early stationary phase. Finally, the introduction of multiple copies of the /as'^ promoter decreases the overall yield of pseudolysin suggesting a dilution effect due to multiple copies of the promoter (Passador and
Iglewski, 1995).
36 The first regulator of P. aeruginosa protease expression was cloned using strain PA103 which contains a complete lasBgene. In this strain, however, pseudolysin is not detected in either intracellular and extracellular envirorunents possibly due to the absence of a regulatory protein (Howe and Iglewski, 1984;
Gambello and Iglewski, 1991). Using a P. aeruginosa cosmid gene bank and complementation experiments, lasR was cloned and verified as the structural gene for the transcriptional activator LasR (Gambello and Iglewski, 1991). Since then,
LasR has arguably been implicated as a transcriptional activator for the expression of other proteases including LasA and alkaline protease (Toder, Gambello e ta l.,
1991; Gambello, Kaye eta l., 1993).
When the amino acid sequence of LasR was used to search the protein database for identification of homologous proteins, LasR was found to be highly similar to LuxR, a transcriptional activator of the bioluminescence genes of V/ivio
Bscùe/i (Gambello and Iglewski, 1991). Two regions of high homology exist between those two proteins: a carboxy-terminal helix-tum-helix motif (53% identity) and an amino-terminal autoinducer binding motif (36% identity). In V. Gscberi, intercellular communication is facilitated through small diffusible signal molecules known as ‘autoinducers’ composed of homoserine lactone (HSL) and an N-linked acyl side chain of varying length (Fuqua, Winans e ta l., 1994; Salmond, Bycroft e t al., 1995). For example, LuxR responds to N-(3-oxohexanoyl)-L-homoserine lactone
(OHHL) and initiates the major regulatory cascade for controlling the induction of bioluminescence in V. (Meighen, 1991).
37 In V. ffscben, autoinducers are believed to be produced by the Iw d gene product since luminescence can be restored in /uAi/'negative mutants by addition of exogenous HSL (Meighen, 1991). The current hypothesis suggests that basal transcription of lu x l leads to accumulation of a low level of HSL which subsequently binds to the amino-terminal autoinducer binding motif of LuxR. The
LuxR-HSL complex thereby stimulates transcription of the /uxregulon which contains lu x l and five structural genes {luxA-E) required for light emission in V.
Gscberi. This leads to increased levels of Luxl and HSL which further activate
LuxR. Since HSL is freely diffusible, the induction of one cell leads directly to the induction of other cells, generating a rapid response to the initial stimulus (Swift,
Bainton e ta l., 1994; Latifi, Winson e ta l., 1995). This density-dependent phenomenon is termed‘quorum sensing’. For instance, when K Æycôen cells are diluted into fresh medium, bioluminescence can not be observed until the cells reach the mid-logarithmic phase of growth (Jones, Yu e ta l., 1993). Therefore, HSL is part of an intercellular communication system that facilitates the induction of genetic régulons only when a significant population of cells has accumulated (Bainton,
Bycroft e ta l., 1992; Fuqua, Winans eta l., 1994; Swift, Bainton eta l., 1994).
P. aeruginosa also contains a gene {lasi) whose product is not only highly homologous to the lu xlg en e product (34.6% identical, 55.9% similar) but also exhibits the same function as Luxl when expressed (Passador, Cook e ta l., 1993).
The autoinducer molecules have also been identified as N-(3-oxododecanoyl)- homoserine lactone in P. aeruginosa (Pearson, Gray e ta l., 1994). Interestingly, even though this P. aeruginosa autoinducer (PAI) is very similar to V. Gscberi
38 autoinducer (VAI) in structure, they are not interchangeable. Experiments have shown that V. Gscberi lu x gene, expression can be induced by P. aeruginosa lasR, and P. aeruginosa pseudolysin expression can be stimulated by V. Æscôerf LuxR, but only in the presence of their cognate autoinducers (Gray, Passador e ta l., 1994).
In addition to V. Gscberiand P. aeruginosa, homologues of LuxR and Luxl are also found in other species including Erwinia carotovora (Jones, Yu e ta l., 1993;
Pirhonen, Flego e ta l., 1993; McGowan, Sebaihia e ta l., 1995), Agrobacterium tum efadens (Piper, Beck von Bodman e ta l., 1993; Fuqua and Winans, 1994),
Pseudomonasaureofadens(^'verson, Keppenne eta l., 1994), Enterobacter agglomerans(^w\St, Winson eta l., 1993), Rhizobium leguriiinosarum{Cvibo,
Economou e ta l., 1992), Yersinia enterocolitica (Throup, Camara e ta l., 1995), and
Serratia liquefadensÇEbeû, Winson e ta l., 1995). Therefore, the quorum sensing mechanism is used widely throughout many différent bacterial species.
The LasR-LasI system was thought to be the only global regulator, and N-(3- oxododecanoyl)-homoserine lactone was believed to be the only autoinducer molecule in P. aeruginosa. However, it soon became obvious that autoinducer-like regulation of gene expression is not that simple. First, several different autoinducerlike molecules were identified in P. aeruginosa (Pearson, Passador eta l.,
1995; Winson, Camara eta l., 1995) suggesting the complex nature of the quorum sensing system. In addition, another LuxR homolog system was identified in P. aeruginosa. The RhlR-RhH system was identified as regulatory genes affecting rhamnolipid biosurfactant synthesis (Ochsner, Koch eta l., 1994; Ochsner and
Reiser, 1995). The RMR-RhH system has recently been implicated in the expression
39 of pseudolysin using the following method (Brint and Ohman, 1995). P. aeruginosa strain PAOl was first mutated with N-methyl-N’-nitro-N-nitrosoguanidine and a pseudolysin defective mutant was chosen. When this mutant was complemented with a gene bank, rblR and r b llwere identified as responsible components for the complementation (Brint and Ohman, 1995). RhlR (27.6 kDa) exhibits 31% homology with LasR (26.6 kDa) and Rhll (22.2 kDa) shows 25% identity with LasI
(22.8 kDa). In addition, the autoinducer molecule of the RhlR-Rhll system is N- butanoyl-L-homoserine lactone which is different from the LasR-LasI system
(Winson, Camara eta l., 1995).
Because both the LasR-LasI system and the RWR-RhU system are involved in the expression of P. aeruginosa exoproducts, there is an apparent overlap between the two systems. One possible explanation is that RhlR activates LasR or alternatively that LasR activates RhlR. A recent study provides evidence that LasR is the master regulator because when activated by N-(3-oxododecanoyl)-homoserine lactone, LasR induces the expression of rbJR. In lasR negative mutants, rbIR transcription is absent, but the introduction of a functional copy of lasR complemented the lasR negative mutation and restored rbIR expression (Latifi,
Fogtino e ta l., 1996). In addition, RhlR is responsible for activating rp o S in an N- butanoyl-L-homoserine lactone-dependent manner (Figure 1.6) (Latifi, Foglino et al., 1996). RpoS (o®) is a stationary-phase sigma factor found in both E. coli and P. aeruginosa (Hengge-Aronis, 1993; Tanaka and Takahashi, 1994). It is called a stationary-phase sigma factor because the level of RpoS increases drastically at the beginning of stationary phase (Fujita, Tanaka eta l., 1994). RpoS is somehow
40 involved in the production of P. aeruginosa exoproducts including pseudolysin, alkaline protease, lipase, and lectins (Latifi, Foglino eta l., 1996). However, whether
RpoS is directly responsible for the expression of such exoproducts is not clear indicating the complex nature of this regulatory system.
41 Figure 1.6: A model for the regulation of rpo5via a hierarchical quorum sensing cascade involving LasR, RhlR, and their cognate N-acylhomoserine lactones
(modified from Latifi, e ta l., 1996).
With N-(3-oxododecanoyl)-L-homoserine lactone(OdDHL), LasR activates la s i and rblR. In turn, in the presence of N-butanoyl-L-homoserine lactone (BHL), RMR activates TpoS and rbll. LasR may also act as an inhibitor of its own transcription.
42 lasR Ia si
LasR 9 OdDHL
rblR rbU
RhlR O BHL N H 0
rpoS
Figure 1.6
43 CHAPTER 2
MATERIALS AND METHODS
A. Strains and plasmids
Bacterial strains and plasmids used in this study are listed in Tables 2.1 and
2.2, respectively. Oligonucleotides synthesized for this study were purchased from
Ransom Hill Bioscience, Inc. (Ramona, CA) and are described whenever they are used. Media and growth conditions are detailed separately whenever applicable.
B. Biochemical assays
1. Determination of elastolytic activity
Elastolytic activity was determined using elastin-Congo red (Sigma Chemical
Company, St. Louis, MO) as a substrate as previously described (Ohman, Cryz e t al., 1980; Peters and Galloway, 1990). To create a standard curve, 0 to 10 mg of elastin-Congo red was added to six 15 ml test tubes in 2 mg increments and 2 ml of
Tris-maleate buffer (0.1 M Tris-maleate, 1 mM CaCL, pH 7.0) was added to each tube. Then, lOOpg of purified pseudolysin was added to each tube and the tubes were incubated at 37°C overnight for complete digestion of elastin-Congo red. After
44 Strains Description source or Reference Pseudomonas aeraginosa strains PAOl prototroph Holloway, Krishnapillai eta l., 1979 PA103 clinical isolate; protease negative Liu, 1966 PA220 clinical isolate; high production of Pavlovskis, Pollack extracellular proteases era/,1977 PAO-LR AlasRritet; Tcf O'Dormell, 1995 FRD2128 lasA mutant; from FRD2 by gene Goldberg and replacement Ohman, 1987 AD 1825 lasA mutant; from PAOl by gene from replacement Dr. Aldis Darzins Stapbylococcus aureus strains OSU681 wildtype Department of Microbiology, Ohio State University Escberfdijacoù'stiams DH5a endA bséRlTs^m^) supE44tbi-1 Hanahan, 1983 regAlgyrAQiaV) relAl A(lacZYA-argF) ^%QlacZAM15 HBlOl S.(gpt-proA) leuB6 tbi-1 bsdS20 Boyer and Rolland- reo4 rpsL2(X^Xf) ara-14galK2 Dussoix, 1969 xly-5m tl-l supE44 mcrBs BL21(DE3) F om pTbsdSi (rs mg ) gal dan Studier and Moffatt (DE3), contains a X prophage 1986; Studier, carrying an inducible T7 RNA Rosenberg e ta l., 1990 polymerase gene BL21(DE3)pLysS F om pTbsdSs (rB'mg ) gal dan Studier and Moffatt (DE3) pLysS(CmO, BL21(DE3) 1986; Studier, containing T7 lysozyme on Rosenberg e ta l., 1990 plasmid to reduce basal activity
Table 2.1: Bacterial strains used in this study.
45 Plasmid Description Source or Reference Escben'cbia coJiYectois pUC18 Ap^; ColEl ori; Yanisch-Perron, Vieira general cloning vector e ta l., 1985 pBR322 Ap\ T(f; ColEl ori; Sutcliffe, 1979 general cloning vector pTTBlue T-vector Ap"; from pUC19; for cloning PCR purchased from products containing A overhang; Novagen blue/white selection pET-3a Ap*^; ColEl 0 / 7; contains T7 (|)10 gene Rosenberg, Lade eta l., promoter, terminator, and 1987 translation initiation signals Escbencbia - Pseudomonas chutât vectors pUCPlS Ap*^; pUCI8 derivative; contains 1.8 Schweizer, 1991 kb stabilizing fragment of pR01614 pR01614 Ap\ Ter; derived from pR01600 and Olsen, Debusscher eta l., PBR322 1982
Table 2.2: Plasmid vectors used in this study.
46 measuring absorbance at 495 nm for each tube, a standard curve could be plotted.
TheX axis was the amount of elastin-Congo red in mg and the y axis was the absorbance at 495 nm. To measure the elastolytic activity of an unknown sample, a known amount of the sample was added to a 15 ml tube containing 10 mg of elastin-
Congo red and 2 ml of the Tris-maleate buffer. After incubating for 2-15 hr at 37°C, the undigested elastin-Congo red substrate was removed by filtration through a 0.2 pm low protein binding cellulose acetate disposable filter (Schleicher & Schuell,
Keene, NH) and the absorbance of the filtrates were read at 495nm. The actual amount of elastin-Congo red digested was interpolated from the standard curve. All reactions were run in duplicate and activity was calculated as mg of elastin-Congo red digested per mg of protein per hour.
2. Determination of proteolytic activity and inhibition
General protease activity of LasA and LasD was measured using either p- casein (Sigma Chemical Company, St. Louis, MO) as a substrate or a QuantiCleave
Protease Assay Kit II (Pierce, Rockford, IL). When P-casein was used as a substrate, the protein was dissolved in 25 mM diethanolamine buffer, pH 9.5, to achieve a final concentration of 1 mg/ml. A 20 pi volume of the P-casein solution was incubated at room temperature in the presence or absence of 1 pg LasD or 0.2 pg LasA for a period of 1 hr. An aliquot was then analyzed by SDS-PAGE and the resulting gel was stained with Coomassie blue (Bio-Rad). Proteolytic activity was then visualized by the disappearance of the p-casein band in the gel.
47 When the QuantiCleave Protease Assay Kit H was used, the assay was based
on the use of succinylated casein in conjunction with trinitrobenzenesulfonic acid
(TNBSA) (Bubnis and Ofher, 1992; Hatakeyama, Kohzaki e ta l., 1992). The casein
substrate in the kit has been treated with succinic anhydride to block all the primary
amines on the surface of the protein. Therefore, when the modified casein was used
as a substrate, newly exposed primary amines could all be attributed to proteolytic
activity. The primary amine groups were detected and quantitated
spectrophotometrically with TNBSA which produces an orange-yellow color upon
reacting with a primary amine group. The assay was performed as follows; 100 pi of
succinylated casein solution (2 mg/ml in 50 mM boric acid, pH 8.5) was added to
the wells of a regular 96 well microtiter plate (Dynatech Laboratories, Chantilly,
VA). Two wells in the upper left comer of the plate were designated controls and
were filled with 50 mM borate buffer without casein. Samples (50 pi) were added to
the wells and the plate was incubated for 20 min at room temperature. Then, 50 pi
of TNBSA working solution (0.25% w/v in 50 mM borate buffer) was added to each
well and incubated for another 20 min at room temperature. Finally, the activity was determined by reading absorbance with an ELISA plate reader at 450 run.
Tosyl phenylalanyl chloromethyl ketone-trypsin (TPCK-trypsin) provided in the kit was used as a standard protease.
For infiibition studies, various inhibitors were preincubated with 1 pg LasD
or 0.2 pg LasA for 10 min at room temperature. Following preincubation, 10 pg (3- casein (1 mg/ml) was added to each sample and incubated at 37°C for 2 hr. The
48 results were analyzed by SDS-PAGE. The following inhibitor concentrations were used: ophenanthroline (10 mM), ethylenediaminetetraacetate (EDTA, 20 mM), phosphoramidon (20 pM), diisopropylfluorophosphate (DFP, 0.5 mM), Tosyl lysyl chloromethyl ketone (TLCK, 100 pg/ml), phenylmethylsulfonyl fluoride (PMSF, 2 mM), and dithiothreitol (DTT, 10 mM). Preparation of stock solutions for above inhibitors was given in Table 2.3. All inhibitors were purchased from the Sigma
Chemical company (St. Louis, MO).
3. Determination of staphylolytic activity and inhibition
The ability of LasA and LasD to lyse heat-killed Stapbylococcus aureus determined using the method previously described (Kessler, Safrin eta l., 1993) with some modifications. S. aureus sxiam OSU681 (Department of Microbiology, Ohio
State University) was cultured in LB medium for 24 hr at 37°C and the cells were collected by centrifugation. The cell pellet was resuspended in 25 mM diethanolamine buffer, pH 9.5, and were killed by heating at 100°C for 10 min. The heat-killed cells were diluted to a final optical density at 595 nm (O.D.sgs) of 1 .5-2.0.
The assay was carried out by adding 3 pg of purified LasA or LasD to a 670 pi suspension of heat-killed cells and measuring the O.D .595 at 30 min intervals for a period of 2.5 hr under different pH conditions. The subsequent decrease in the
O . D . 5 9 5 of the cell suspension provided an indirect measure of the lysis of the S. aureus ctMs.
49 Name Target Protease Concentration of the Stock Solution o-phenanthroline metallo-proteases 200 mM in methanol EDTA metallo-proteases 100 mM in HPLC grade HzO phosphoramidon some metallo-proteases 1 mM in HPLC grade HzO DFP serine proteases 100 mM in isopropanol TLCK trypsin-like serine 100 mg/ml in HPLC grade HzO proteases PMSF serine proteases 10 mM in isopropanol Aprotinin serine proteases 2 mg/ml in HPLC grade HzO Leupeptin trypsin-like serine 10 mg/ml in HPLC grade HzO proteases and some cysteine proteases Pepstatin A some aspartic proteases 5 mM in methanol
Table 2.3: Preparation of inhibitors used in this study.
50 Using heat-killed whole cells as substrates could raise some problems especially for the inhibition study because of a lot of unknown factors including inactivated proteases which may interact with inhibitors. To overcome this problem, synthetic peptides which mimicked certain cellular structures were often used. For instance, pentaglycine was used to study staphylolytic activity because the staphylolytic activity was known to be the result of a cleavage by a staphylolytic enzyme in pentaglycine which could be found in the bridge region of the cell walls of Gram positive bacteria including Stapbylococcus aureus. In this study, the pentaglycine (Sigma Chemical Company, St. Louis, MO) was modified by acétylation to quantify the reaction and also to decrease the background. The modification procedure will be described in a later section.
To measure the staphylolytic activity of LasA indirectly, 10 pi of LasA (0.1 pg/pl) was added to a 1.5 ml test tube containing 90 pi of 20 mM phosphate buffer, pH8.0, with 10 pg of N-blocked (acetylated) pentaglycine and incubated for 1 hr at
37°C. After adding 50 pi of TNBSA (0.25% in 50 mM boric acid, pH8.5) to the test tube, the tube was left for 5 min at room temperature to allow for color development. Absorbance at 450 nm was measured and the values were compared to a standard curve. The standard curve was constructed by incubating varying amounts of pentaglycine (0 to 10 pg) with 2 pg of LasA overnight at 37°C for complete digestion. Then, O.D .450 was measured and a X-Y graph was made as a standard curve. The X axis was the amount of pentaglycine and the Y axis was
O.D.450 . Staphylolytic activity of LasD was measured by the same assay except that
51 10 pi of LasD (0.1 pg/|il) was added to a 1.5 ml test tube containing 90 pi of 25 mM diethanolamine buffer (pH 9.5) instead of the phosphate buffer.
Inhibition studies were carried out using various inhibitors preincubated with
2.5 pg LasA or LasD for 10 min at room temperature prior to addition of the heat- killed cell suspension. The total volume of the reaction mixture was 500 pi. The mixture was incubated at 37°C and samples were removed at 30 min intervals for
O.D.595 determination. The final concentration of inhibitors used was phosphoramidon (10 pM), DFP (0.5 mM), o-phenanthroline (5.7 mM), EDTA (10 mM), TLCK (100 pg/ml), PMSF (2 mM), and DTT (10 mM).
4. HPLC analysis of substrate specificity using synthetic peptides and insulin (3-chain
Beckman HPLC (composed of a model 406 analog interface module, two
1 lOB pumps, and a model 166 UV detector module, controlled by a System Gold software) was used in this study. All of the peptides were purchased from Sigma
Chemical Company (St. Louis, MO). Compositions and concentrations of synthetic peptides and proteases used to analyze substrate specificity is given in Table 2.4. To determine whether the peptides were cut by various proteases (including pseudolysin, LasA, and LasD) and an uncharacterized protein (PI 5), 15 pi of each peptide was first incubated with 0.5 pg of each protease for different time intervals at different pH conditions. For example, 20 mM phosphate buffer (pH 8.0) was used for LasA and 25 mM diethanolamine buffer (pH 9.5) was used for LasD. Then 20 pi of each sample was analyzed by a reverse phase column (Bio-Sil ODS-5S with a
52 Table 2.4: Synthetic peptides and proteases utilized in HPLC analysis.
Prepared in both 20 roM phosphate buffer (pH 7.0) and 25 mM diethanolamine buffer (pH 9.5).
^ Prepared in both HPLC grade HzO (Burdick & Jackson, Muskegon, MI) and 25 mM diethanolamine buffer (pH 9.5).
^ Prepared in both 0.1 M Tris-Maleate buffer (pH 7.0) and 25 mM diethanolamine buffer (pH 9.5).
^ Prepared in 20 mM phosphate buffer (pH 8.0).
^ Prepared in 25 mM bis-Tris (pH 6.3), 20 mM phosphate buffer (pH 8.0), and 25 mM diethanolamine buffer (pH 9.5).
53 Peptide Goneentratiofn Gly 3.3 mM‘ Gly-GIy 3.3 mM‘ Gly-GIy-GIy 3.3 mM' Gly-Gly-GIy-Gly 3.3 mM' Gly-GIy-GIy-Gly-GIy 3.3 mM' Gly-Gly-GIy-Gly-GIy-GIy 3.3 mM' Gly-Pro-GIy-GIy 7mM' Lys-Lys-GIy-GIu 5.5 mM" Val-GIy-Val-Ala-Pro-Gly 4 mM" Ala-isoGIu-Lys-Ala-Ala 4.1 mM" Trp-Ala-GIy-Gly-Asp-Ala-Ser-GIy-GIu 2 mg/ml" Gly-Phe-Asp-Leu-Asn-GIy-GIy-GIy-Val-GIy 2 mg/ml" Try-Ala-Gly-Ala-Val-Val-Asn-Asp-Leu 1 mg/ml" Protease Concentration Elastase 1.2 [Lg/\if LasA 0.3 ng/pP LasD 0.2 |ig/pl" PIS 0.5 pg/pl"
Table 2.4 54 guard column, Bio-Rad, Hercules, CA) on HPLC. Eluted peaks were detected by
O.D. at 210 nm and running conditions were detailed in Table 2.5.
In addition to synthetic peptides, bovine insulin p-chain (Sigma Chemical
Company, St. Louis, MO) was also used as a substrate. In this method, 7.5 pg of insulin P-chain (0.5 mg/ml) was reacted with 0.5 pg of either pseudolysin (0.5 mg/ml), LasA (0.1 mg/ml), or LasD (0.1 mg/ml). The proteases were resuspended in either 20 mM phosphate buffer (pH 8.0, for LasA and pseudolysin) or 25 mM diethanolamine buffer (pH 9.5, for LasD). After mixing the P-chain (resuspended in either pH 8.0 or pH 9.5 buffer depending on the protease) with each protease, the samples were incubated at room temperature for 30 min before loading each sample onto an HPLC Bio-Sil 0DS-5S reverse phase column (Bio-Rad, Hercules, CA). The loading buffer A was HPLC grade water (Burdick & Jackson, Muskegon, MI) with
0.1% HPLC grade trifluoroacetic acid (TFA, Pierce, Rockford, IL) as an ion pairing reagent and the elution buffer B was 70% acetonitrile (HPLC grade, Burdick &
Jackson) in HPLC grade water with 0.1% TFA. After loading each reaction mixture at a flow rate of 0.5 ml/min for 5 min, the insulin P-chain was eluted by applying a
60 min gradient (0-100% elution buffer B) at the same flow rate.
5. N-terminal sequence determination of LasD using a PVDF membrane
N-terminal sequence analysis of purified LasD and the 30 kDa LasD precursor was performed using a method described by Matsudaira (Matsudaira,
1987). LasD (5 pg) was transferred from an SDS-PAGE gel (12%) to a PVDF
55 Table 2.5: HPLC methods used in this study for substrate specificity.
‘ Duration is the time required to reach a flow rate from the previous flow rate.
^ Buffer A is HPLC grade water (Burdick & Jackson, Muskegon, MI) with 0.1% HPLC grade trifluoroacetic acid (TFA, Pierce, Rockford, IL) as an ion pairing reagent. Buffer B is 70% Acetonitrile (HPLC grade, Burdick & Jackson) in HPLC grade water with 0.1% TFA. Duration is the time to reach current % of buffer B from the previous %.
O.D. was zeroed at 210 nm.
56 A. Method: Gly
Time Duration' Gradient^ Auto Stop End (min) Rate (min) Zero^ Data Method (ml/min) % Duration Buher A Buffer B (min)
Initial 1.00 - 100.0 0.0 - Yes --
0.00 1.00 - 100.0 0.0 - - --
7.00 1.00 - 0.0 100.0 60.00 - --
70.00 1.00 - 100.0 0.0 10.00 - Yes Yes
B. Method: Gly2
Time Flow Duration' Gradient' Auto Stop End (min) Rate Zero' Data Mediod (ml/min) (min) % Duration BufFer A BufferB (mW
Initial 1.00 - 100.0 0.0 - Yes --
0.00 1.00 - 100.0 0.0 -- --
10.00 1.00 - 0.0 100.0 0.00 - --
40.00 1.00 - 0.0 100.0 0.00 - Yes Yes
To be continued.
Table 2.5 57 Table 2.5 (cont.)
C. Method: Gly3
Time Flow Duration^ Gradient^ Auto Stop End (min) Zero^ Data Method (ml/min) % % Duration BufferB imin)
Initial 1.00 - 100.0 0.0 - Yes --
0.00 1.00 - 100.0 0.0 - ---
6.00 1.00 - 100.0 0.0 - - Yes -
7.00 0.10 - 100.0 0.0 1.00 -- Yes
D. Method: Gly5
Time Flow Duration^ Gradient^ Auto Stop End (min) Rate (min) Zero^ Data Method (ml/min) % Duration Buffer A BufferB (min)
Initial 0.50 - 100.0 0.0 - Yes - -
0.00 0.50 - 100.0 0.0 20.00 - - -
20.00 0.50 - 90.0 10.0 10.00 ---
35.00 0.50 - 0.0 100.0 5.00 -- -
36.00 0.50 - 100.0 0.0 1.00 - Yes -
51.00 0.50 - 100.0 0.0 --- Yes
To be continued. Table 2.5
58 Table 2.5 (cont.)
E. Method: GIy6
Time Flow Duràtioh- Gradient^ Auto Stop End (min) Rate (min) % % Duration Zero" Data Method (ml/min) Buffer A BüflferB (min)
Initial 1.00 - 100.0 0.0 - Yes - -
0.00 1.00 - 100.0 0.0 - - - -
5.00 1.00 - 0.0 100.0 30.00 - - -
45.00 1.00 - 100.0 0.0 5.00 - - -
50.00 0.50 - 100.0 0.0 0.00 - Yes Yes
Table 2.5 59 membrane (Millipore, Bedford, MA) using CAPS buffer (10 mM 3-
[cyciohexyIamino]-l-propanesuIphonic add, 10% methanol, pH 11). Following staining (0.1% Coomassie blue R-250 in 50% methanol) for 5 min and destaining
(50% methanol, 10% acetic add) for 10 min, the band was exdsed and washed thoroughly for 2 hr with several changes of deionized water. N-terminal sequencing was carried out at the Ohio State University Biochemical Instrumentation Center using an Applied Biosystems Gas Phase Protein Sequencer (Perkin-Elmer, Applied
Biosystems Division, Foster City, CA) with on-line 120A PTH Analyzer and model
900A Data Analysis Module. The resulting sequences were compared with the known LasA sequence and a homology search was carried out at the National
Center for Biotechnology Information (NCBI) using the Blast network service.
6. Peptide mapping by limited proteolysis
A previously published protocol (Cleveland, Fischer e ta l., 1977) was followed to partially digest a protein into several peptides. In this assay, bands from
SDS polyacrylamide gels stained with Coomassie blue were conveniently digested by placing gel slices containing these bands in the sample wells of a second SDS polyacrylamide gel and then overlaying these slices with a protease such as
Staphylococcus aureus YS protease. Digestion occurred directly in the stacking gel during the subsequent electrophoresis. The actual procedures is as follows: the stained SDS polyacrylamide gel was rinsed with cold water and individual bands were cut with razor blade. The bands were trimmed to small pieces and soaked for
30 min with occasional swirling in a solution containing 125 mM Tris-HCl, pH 6.8,
60 0.1% SDS, and 1 mM EDTA. The slices were placed at the bottom of the sample wells of a second SDS polyacrylamide gel after filling the wells with the same buffer.
Then, spaces around the gel slices were filled with buffer containing 20% glycerol.
A given amount of protease (in 10 pi of the buffer with 10% glycerol) was added to the wells and electrophoresis was performed. When the bromophenol blue dye approached the bottom of the stacking gel, the current was turned off for 30 min to allow proteolytic digestion and, then, reapplied. After finishing electrophoresis.
Western blotting or other assays were performed to analyze the limited digestion.
7. Denaturing polyacrylamide gel electrophoresis
Standard protocols were followed to run one-dimensional polyacrylamide gel electrophoreses under denaturing conditions (Laemmli, 1970; Chrambach and
Rodbard, 1971; Matsudaira and Burgess, 1978; Andrews, 1986). This type of electrophoresis separates proteins based on size upon migration through a polyacrylamide gel. Between two different single-concentration denaturing electrophoresis methods (continuous and discontinuous), only discontinuous method was used in this study for higher resolution. To run a discontinuous SDS-PAGE, either a SE 600 Vertical Slab Gel Unit for a large gel (Hoefer Scientific Instmments,
San Francisco, CA) or a SE 250 Mighty Small H for a mini gel (Hoefer Scientific
Instruments, San Francisco, CA) was assembled. Usually, 1.5 mm spacers were used for large gels and 0.75 mm spacers were used for mini gels. Separating gel solution (varying between 8-20%) was poured first, then after polymerization, stacking gel solution (4 or 6%) was poured with appropriate combs. Recipes for
61 separating and stacking gel solutions for large double gels are given in Table 2.6.
The protein sample to be analyzed was diluted 1:1 (v/v) with 2X sample buffer
(0.125 M Tris-HCl, pH6.8,4% SDS, 20% glycerol, 2% p-mercaptoethanol, and
0.01% bromphenol blue) and heated at 100°C for 3-5 min before running to insure dénaturation. Typically, when a single gel was used, the gel was run for either 2 hr at 35 mA (large gel) or 40 min at 20 mA (mini gel); when two gels were run in one apparatus, the current was doubled accordingly. The tank buffer was consisted of 25 mM Tris-HCl, pH 8.3, 192 mM glycine, and 0.1% SDS. To visualize the protein bands on the gel following electrophoresis, the gel was stained with 0.125%
Coomassie blue R-250, 50% methanol, 10% acetic acid for ca. 30 min, and destained with 20% methanol, 10% acetic acid for ca. 2 hr. To preserve a gel, destained gel was soaked overnight in 10% glycerol to prevent cracking after drying. The gel was then dried for 1 day in between 2 gel drying films (Promega, Madison, WI) and kept at room temperature.
In some cases, gradient gel electrophoresis was used for better separation of proteins with similar sizes. All the procedures and recipes were same as those of non-gradient gels except the separating gel had a gradient (usually 5-20%). The gel was poured using a mixing chamber and a P-1 peristaltic pump (Pharmacia Biotech,
Piscataway, NJ). The mixing chamber had two reservoirs - one containing 5% acrylamide solution and the other containing 20% acrylamide solution - which were cormected to each other and the gel was poured from the reservoir containing 20% solution. As a result, the bottom of the gel had highest concentration of acrylamide and the top of the gel had the lowest concentration.
62 %Gel 30%T Separating Stacldng IG%SDS^ ddHzO APS^ TEMED® 2 /7 # ^ buffer* buffer^ 7.5% 15 0.02 15 - 0.6 29.1 0.3
10% 20 15 24.1 0.3 0.02 - 0.6
12% 25 15 0.6 19.1 0.3 0.02 -
15% 30 15 0.6 14 0.3 0.02 -
4% 5 12.2 2.66 - 0.2 0.1 0.01
6% 4 5 10.6 - 0.2 0.1 0.01
‘ 30%T2.7%C represents 30% w/v acrylamide plus bis, and the bis would account for 2.7% of the total weight of the acrylamide.
^ Separating buffer is composed of 1.5 M Tris-HCl, pH8.8.
^ Stacking buffer is composed of 0.5 M Tris-HCl, pH6.8.
10% w/v sodium dodecyl sulfate in ddHzO.
^ 10% w/v ammonium sulfate in ddHzO.
* Purchased from Sigma Chemical Company (St. Louis, MO).
Table 2.6: Recipe for a large double SDS polyacrylamide gel (Unit = ml).
63 8 . Purification of P. aeruginosa pseudolysin
The procedure for pseudolysin purification is a modification of the previously published protocols (Morihara, Tsuzuki e ta l., 1965; Peters and Galloway, 1990). A frozen culture of P. aemgmosa strain PA220 was inoculated into 5 ml of PTSB medium (Ohman, Cryz eta l., 1980). PTSB medium contains 5% Bacto-Peptone
(Difco Laboratories, Detroit, MI) and 0.25% Trypticase Soy Broth (BBL
Microbiology Systems, Cockeysville, Maryland). Cultures were grown for 24 hr in a shaker (ca. 250 rpm) at 37°C. One ml of these cultures was inoculated into six 3 L
Fembach culture flasks, each containing 500 ml PTSB medium, and incubated for ca. 18 hr with shaking. After incubation, cultures were transferred to 250 ml centrifuge bottles and centrifuged at 5,000 rpm for 15 min at 4°C. The supernatants were pooled in a 4 L beaker and stored at 4°C. Ammonium sulfate was slowly added to the supernatant (ca. 5 g per min) to 60% saturation (390 g (NIL,) 2S0 4 per liter supernatant) with stirring at 4°C. After all the ammonium sulfate was added, the slurry was stirred for an additional hour and allowed to settle overnight at 4°C.
The mixture was centrifuged in 250 ml bottles at 5,000 rpm for 15 min at 4“C and the precipitate was collected and resuspended in 500 ml cold ddHzO. The sample was then centrifuged again as before and the supernatant was carefully decanted into a 2
L beaker and stored at 4°C. Mucous-like material was sometimes present in the supernatant and was removed with a Pasteur pipette. Another 500 ml of cold ddHzO was added to the pellet and mixed well by using a 25 ml pipette; the suspension was centrifuged as before and the supernatant was added to the previously collected supernatant. Again, care must be taken to prevent any mucous-
64 like material from getting into the supernatant-containing beaker. Ammonium
sulfate was added again to the pooled supernatants as before and the mixture was
allowed to stand overnight at 4°C.
The following morning, the slurry was centrifuged to collected the pellet
The pellet was then resuspended in small volume of cold ddHzO (ca. 150 ml) and
centrifuged to collect the supernatant. Cold (-20“C) acetone was added (60% v/v
concentration) to remove oily material from the sample; a white slurry was formed
immediately and this was centrifuged as before. The precipitate was collected and
resuspended in 100 ml of cold 20 mM sodium phosphate buffer (94.7ml of 20 mM
NazHP0 4 and 5.3 ml of 20 mM NaH 2? 0 4 , pH8.0). The resuspended sample was
then carefully transferred to a dialysis membrane (Spectra/Por membrane, 12-14
kDa molecular cut-off. Spectrum, Houston, Texas) which was pre-soaked in the same buffer. Dialysis continued for 2 days against 4 L of same buffer with 3 changes of buffer at 4“C.
After dialysis, the sample was centrifuged to remove any precipitate. The sample was then loaded onto a 2.5 X 35 cm DEAE-Sephacel column (Sigma
Chemical Company, St. Louis, MO) pre-equilibrated with the phosphate buffer at
4°C at a flow rate of 1 ml/min. This was followed by the application of phosphate buffer and the pass-through fractions were collected in 7 ml aliquots until O.D. at
274 nm returned to the basal level after a small but wide peak. Upon return to equilibrium, the buffer was changed to 0.1 M NaCl in 20 mM sodium phosphate buffer (pH8.0) and the buffer was passed through the column until the O.D. returns to background level again after a large peak. Fractions (7 ml) were collected and
65 analyzed by 12% SDS-PAGE. Highly pure pseudolysin was detected from the fractions corresponding to the peak. The pseudolysin containing tubes were pooled and dialyzed overnight at 4“C against 1 L of 20 mM phosphate buffer (pH 8.0).
Finally, pseudolysin was concentrated by using a stirred cell (Type 8010,10 ml capacity, Amicon, Beverly, MA) with a YM-10 membrane (10 kDa molecular cut off, Amicon, Beverly, MA). The concentration of pseudolysin was determined by
Bradford protein assay (Bradford, 1976) and the sample was stored at -70“C in 300 pi aliquots. A typical purification table was given in Table 3.1.
9. Purification of LasA
LasA was originally purified as a by-product of pseudolysin purification
(Peters and Galloway, 1990; Peters, Park eta l., 1992). The LasA protein eluted with other proteins in the pass-through fractions of a DEAE-Sephacel column. Among those fractions, LasA-containing fractions (determined by SDS-PAGE) were pooled together in a 1 L beaker and 60% ammonium sulfate precipitation was performed as before. The slurry was centrifuged and the pellet was collected. Then, the pellet was resuspended in 10 ml of 20 mM phosphate buffer (pH 8.0) and dialyzed overnight at 4“C against 3 L of the same buffer. The sample was then applied to a
CM-Cellulose column (Sigma Chemical Company, St. Louis, MO) equilibrated with the same phosphate buffer at a rate of1 ml/min and pass-through fractions were collected. In this step, one of the major contaminating proteins (15 kDa) was removed from LasA containing fractions because the 15 kDa protein was retained in the column while LasA and another 23 kDa protein passed through the column.
66 The LasA containing fractions were then pooled together and dialyzed overnight at
4°C against 3 L of 50 mM N-2-hydroxyethyl-piperazine-N’-2-ethanesulfonic acid
(HEPES) buffer (pH8.0). The resulting sample in the dialysis tubing was concentrated to less than 2 ml using the stirred cell. Finally, LasA was separated from the other protein (23 kDa LasD) using FPLC (composed of a LCC-500 controller, two P-500 pumps, a single path UV-1 monitor, a pH monitor, a MV-7 motor valve, a REC-482 two-channel recorder, and a FRAC-100 fraction collector, from Pharmacia Biotech, Piscataway, NJ) with a Mono-S cation exchange column
(Pharmacia Biotech, Piscataway, NJ) at a flow rate of 1 ml/min. Under these conditions (50 mM HEPES buffer, pH8.0), LasA was retained in the column and eluted using a shallow salt gradient (0-0.1 M NaCl in 30 minutes). During the elution of LasA, one ml fractions were collected and fractions corresponding to each peak were analyzed by SDS-PAGE and Western blotting for the presence of LasA.
The LasA sample was dialyzed overnight at 4°C against 1 L of 20 mM phosphate buffer (pH 8.0) and concentrated using the stirred cell. The purified LasA was stored at -70°C in small aliquots (50 pi) for future use. A sample purification table is shown in Table 3.2.
10. Purification of LasD
LasD was initially purified as a by-product of the pseudolysin and LasA purification (Park and Galloway, 1995). The use of strain PA220 resulted in higher yields of LasD but the protein has also been purified from other strains of P. aeruginosa such as AD 1825 and FRD 2128. LasD was purified with the same
67 procedure used for LasA purification except that the final dialysis step was omitted.
When the sample containing both LasA and LasD was applied to the FPLC Mono-
S cation exchange column, LasA was retained in the column while LasD passed through the colunm. One ml fractions were collected and analyzed by SDS-PAGE and Western blotting to detect the presence of either LasA or LasD. The sample was then concentrated using the stirred cell and stored at -70°C in 50 pi aliquots.
These two proteins (LasA and LasD) can also be separated by HPLC
(composed of a model 406 analog interface module, two HOB pumps, and a model
166 UV detector module, controlled by System Gold software, from Beckman,
Fullerton, CA) using a MA7C cation exchange column (Bio-Rad, Hercules, CA) at pH 8.0 in 20 mM sodium phosphate buffer. Using this method allows LasD to pass through the column while LasA is retained and must be eluted using a salt gradient.
Again, 1 ml fractions were collected using a Superfrac fraction collector (Pharmacia
Biotech, Piscataway, NJ) and analyzed as before by SDS-PAGE and Western blotting. HPLC methods were summarized in Table 2.7. The purified LasD was stored at -70°C in 50 pi aliquots for future use. A purification table is given in Table
3.3.
11. Radiolabeling of LasA precursor proteins
Labeling of LasA precursor was performed to study LasA processing from the precursor to the active LasA fragment. The T7 expression system, as previously described, was used with some modification (Tabor and Richardson, 1985; Studier and Moffatt, 1986; Studier, Rosenberg eta l., 1990; Groisman, Pagratis eta l., 1991)
68 Table 2.7: HPLC methods used in this study to purify LasD.
'■ Duration is the time required to reach current flow rate from the previous rate.
^ Buffer A is HPLC grade water (Burdick & Jackson, Muskegon, MI) with 0.1% HPLC grade trifluoroacetic acid (TFA)( Pierce, Rockford, IL) as an ion pairing reagent. Buffer B is 70% Acetonitrile (HPLC grade, Burdick & Jackson) in HPLC grade water with 0.1% TFA. Duration is the time to reach current % of buffer B from the previous %.
^ O.D. was zeroed at 274 nm.
69 A. Method: MA7C1
Tune :: :vFldW::;:::: Duration’ Gradient* Auto Stop End (min) ::: :: : (rain) % % Duration Zero* Data Method (ml/min) BufFer A BufFer B (rain) Initial 1.00 - 100.0 0.0 - Yes -- 0.00 5.00 1.00 100.0 0.0 - Yes - - 3.00 5.00 - 50.0 50.0 10.00 - -- 20.00 5.00 - 0.0 100.0 35.00 - -- 50.00 1.00 1.00 100.0 0.0 1.00 - Yes - 55.00 0.20 1.00 100.0 0.0 - - - Yes
B. Method: MA7C2
Time Flow Duration’ Gradient Auto Stop End (min) V: Rate (mih) Duration Zero* Data Method (ml/min) BufFer A Buffer B (min)
Initial 1 .0 0 - 1 0 0 .0 0 .0 - Yes --
0.00 2.00 1.00 1 0 0 .0 0 .0 - r Yes --
5.00 2 . 0 0 - 50.0 50.0 30.00 - --
40.00 2 . 0 0 - 0 .0 1 0 0 .0 15.00 ---
60.00 1 .0 0 1 .0 0 1 0 0 .0 0 .0 1 .0 0 - --
70.00 0 . 2 0 1 .0 0 1 0 0 .0 0 .0 - - Yes Yes
To be continued. Table 2.7 70 Table 2.7 (cont.)
C. Method: MA7C3
Time Flow Düràtibh‘ Gradient' Auto Stop End (min) Rate (min) % Duration 2 jero' Data Method (ml/min) Buffer A : Buffer B ; (min)
Initial 1 .0 0 - 1 0 0 .0 0 .0 - Yes - -
0 .0 0 2 . 0 0 0.50 1 0 0 .0 0 .0 - Yes --
2 .0 0 5.00 3.00 1 0 0 .0 0 .0 - - --
5.00 5.00 - 50.0 50.0 40.00 - --
45.00 5.00 - 0 .0 1 0 0 .0 1 0 .0 0
60.00 1.00 1.00 100.0 0 .0 1 .0 0 - --
65.00 0 . 2 0 0.50 1 0 0 .0 0 .0 - - Yes Yes
Table 2.7 71 to label the precursor. Initially, an E. o?//BL2I(DE3) strain was transformed with pET3a/(l. I kb Ase I-Bam HI lasA) purified by miniprep from a stock strain {E. coli
TBl containing pET3a/l.l kb Ase I-Bam HI lasA). The 1.1 kb Ase I-Bam HI fragment was obtained from a 2.4 kb Sal I-Hind DI fragment by PCR as described in the section for overexpression of LasA and the LasA precursor (Chapter 2.D.5).
After transformants were selected by Ampicillin resistance on LB plate (Amp 100), several positive colonies were grown in 5 ml LB medium and minipreps were performed to actually identify the vector with lasA insert by restriction digestion. A positive colony was grown overnight at 30°C in 2 ml M9 glucose media containing carbenicillin (100 p./ml). M9 glucose media was made by mixing 100 ml of lOx M9 salts (60 g Na 2HP0 4 ,30 g KH 2PO4, 5 g NaCl, 10 g NH 4CI, and ddH 2 0 to 1 L) with
0.1 ml 1 M CaCh, 2 ml 1 M MgS 0 4 , 10 ml 20% glucose (filter sterilized), and 8 8 8 ml ddH20.
After diluting to a final volume of 5 ml, the culture was incubated at 30°C for
3 hr or until O.D .600 reached between 0.35-0.7. From the 5 ml, 1.5 ml was transferred to a new tube and EPTG was added to a final concentration of 1 mM (15 pi of 100 mM EPTG stock solution) to induce the expression of T7 RNA polymerase in the strain. The culture was further incubated for 30 min at 30°C and 6 pi of rifampicin stock solution (50 mg/ml in N.N’-dimethylformamide, Sigma Chemical
Company, St. Louis, MO) was added to achieve a final concentration of 200 pg/ml.
Rifampicin was added to inhibit host E. coli'RNA. polymerase. After incubating for
20 more min at 30°C, 250 pi aliquots were placed in 1.5 ml Eppendorf tubes
72 containing 4 ni of ^H-labeled amino acid mixture (I mCi/ml, Amersham, Arlington
Heights, IL) and incubated for another 20 min at the same temperature. To check the labeled protein by SDS-PAGE, cells in one of the aliquots were lysed by adding
43.75 pi 50% trichloroacetic add (TCA). The labeled proteins from the lysed cells were predpitated on ice for 10 min and collected by centrifugation for 5 min at maximum speed (14000 RPM) in an Eppendorf microcentrifuge. The predpitated protein pellet was washed with 1 ml of cold acetone and resuspended in 30 pi of Ix
SDS sample buffer. 15% SDS-PAGE was run with 5 pi samples per well for 3 hr at
40 mA constant current at room temperature. The gel was carefully lifted from the gel cast and treated with a fluor. A fluor is an enhancer used to significantly reduce the X-ray film exposure time necessary for visualization for low energy p-emitters
(such as ^H, "C, and “S) by converting the p-particle energy into light energy. Two types of fluors were used in this study. One was a commerdally available enhancer named ENGEANCE (DuPont NEN, Boston, MA) and the other was salicylic add.
To use ENGEANCE, the gel was impregnated in the ENGEANCE solution at room temperature for 1 hr with gentle agitation. Then, the fluor was predpitated by incubating the gel in water for 30 min. After the gel was dried using a slab gel drier
(model SE 1160, Eoefer Sdentifrc Instruments, San Frandsco, CA) at 60°C for 2 hr, the gel was exposed to X-ray film at -70°C for various times (8 hr, 3 days, and a week, for instance). Finally, the film was developed and the labeled proteins were visualized. To save cost, salicylic add could be used instead of ENGEANCE. In this method, the gel was soaked in 7% acetic add for 30 min followed by washing in
73 ddHzO for 30 min. As a next step, the gel was soaked in 1 M salicylic acid for 30 min. The gel was then dried and exposed to a X-ray film as before.
C. Assays for kinetic analysis
1. Modification of pentaglycine by acétylation
Pentaglycine was one of the substrates used in this study to determine the substrate specificity of both LasA and LasD. For HPLC analysis, pentaglycine and glycine peptides of various length (glycine to hexaglycine) were purchased from
Sigma Chemical Company (St. Louis, MO) as a Glycine Kit and used directly without modification. In other cases, such as quantification of staphylolytic activity and kinetic studies, however, it was necessary to modify pentaglycine especially when trinitrobenzenesulfonic acid (TNBSA) was used to determine the activity.
TNBSA interacted strongly with any available amino group and the subsequent yellow color development was quantified by measuring absorbance at 450 nm.
Therefore, when pentaglycine was used directly, the amino group at the amino terminus of the peptide interacted with TNBSA even when the peptide was not treated with either LasA or LasD resulting in high background. The problem could be solved by blocking the amino terminus of pentaglycine with an acetyl group. As a result, by eliminating most of the background, ciny yellow color development was due to new amino groups exposed by proteolytic activity.
To acetylate pentaglycine, 420 pg (1.39 mmol) of pentaglycine was resuspended in 10 ml ddH%0 and three equivalents of triethylamine (4.17 mmol, 577
74 ^1) was added with stirring on ice. Triethylamine was used to neutralize the amine group of pentaglycine for acétylation (Figure 2.1). After the suspension was cooled on ice, two equivalents of acetyl anhydride (2.78 mmol, 260 |il) was added with stirring. The suspension was then stirred for 2 hr at room temperature and a small amount of sample (10 pi) was transferred to a filter paper every 30 min and reacted with ninhydrin. Ninhydrin interacts with free amine groups resulting in purple color development. Therefore, by reacting a small amount of sample mixture with ninhydrin, the completion of the acétylation reaction could be monitored. When the sample and ninhydrin reaction did not produce any purple color, the sample was mixed with one equivalent of triethylamine (1.39 mmol, 192 pi) and one equivalent of acetyl anhydride (1.39 mmol, 130 pi) one more time to make sure all the free amino groups were acetylated. Excess triethylamine and acetyl anhydride were then removed by extracting three times with 20 ml ethyl acetate. In this reaction, the top phase containing both triethylamine and acetyl anhydride was discarded and the bottom phase containing acetylated pentaglycine was transferred to a new flask.
The solution was then reacted with 1 M HCl until pH of the solution reached 4.0 resulting in precipitation of acetylated pentaglycine. The sample was filtered and, after removing the acidic solution, the precipitated pentaglycine was washed extensively with ddHzO to remove any remaining acid. Finally, the precipitate was dried in a vacuum dessicator at room temperature and stored at -20°C for future usage.
75 + NH3 - (Πy)s - CO2
Free up Amine for (C2Hs)3N Triethylamine Acétylation
NH 2 - (Qy)5 - CO2 îA (C 2 H 5 )3
O CH3 — C \ Acétylation O Acetyl Anhydride / CH3 — C O O + CH3 — C — NH - (ay)s — CO2 HN(C2H s)3
Acidification HCl
O
CH3 — C — NH — (Gly)s — CO2H
Figure 2.1: Modification of pentaglycine by acétylation.
76 2. Verification of acetylated pentaglycine
To verify that most of the pentaglycine was acetylated, HPLC analysis was used. First, acetylated pentaglycine was resuspended in 25 mM diethanolamine buffer (pH 9.5) to make a final concentration of 1 mg/ml. Then, 20 pi of the resuspended sample was analyzed with a Cis based Bio-Sil 0DS-5S reverse phase column (150 mm x 4 mm, Bio-Rad, Hercules, CA) at 210 nm wavelength. The modified pentaglycine was eluted at 3.14 min (Figure 4.2B). To compare the elution profile with unmodified pentaglycine, pentaglycine was resuspended in the same buffer at the final concentration of 0.66 M. The same analysis was performed, and according to Figure 4.2A, the unmodified pentaglycine was eluted at 2.19 min. This result indicated that the modified pentaglycine was eluted later than the unmodified pentaglycine because it was more hydrophobic than the unmodified one. This increase of hydrophobicity was due to the loss of the hydrophilic amino group by acétylation.
3. Kinetic analysis of enzymatic reactions
To study enzyme kinetics of LasA, acetylated pentaglycine was resuspended in 20 mM phosphate buffer (pH 8.0) at a concentration of 1 mg/ml. Then, varying amounts of the substrate (1 to 98 pg) were incubated with 0.2 pg LasA resuspended in the same buffer. Each reaction mixture was adjusted to a total reaction volume of
100 pi by adding the same phosphate buffer. After 30 min incubation at 37°C, 50 pi trinitrobenzenesulfonic acid (TNBSA, Sigma Chemical Company, St. Louis, MO)
77 working solution (10 |j1 TNBSA in 1.49 ml 50 mM boric acid, pH 8.5) was added to each reaction mixture and allowed to stand for 20 min at room temperature. Then, the O.D. was measured at 450 nm because TNBSA reacts with any newly exposed primary amine group (due to hydrolysis of acetylated pentaglycine by LasA) to produce an orange-yellow color that can be quantitated at this wavelength. The experiment was repeated twice and average values were obtained for each data
point. Kinetic graphs ( O . D . 4 5 0 vs. acetylated pentaglycine and double reciprocal
Lineweaver-Burke plot) were made using the average values, and then. Km and Vmax were calculated according to Michaelis-Menten Kinetics.
Calculation of the turnover number (kcm) was done by converting units of Vma;t
from O . D . 4 5 0 values to mM/min using the standard O . D . 4 5 0 value. To calculate the
standard O . D . 4 5 0 per unit concentration (pM) of digested acetylated pentaglycine, unmodified pentaglycine was used instead of acetylated pentaglycine. In this case , there is no difference between unmodified and modified pentaglycine substrates for
converting O . D . 4 5 0 to mM/min because O . D . 4 5 0 is dependent only on the number of available amine groups. In other words, a new amine group is produced when a molecule of acetylated pentaglycine is digested by either LasA or LasD. A molecule of unmodified pentaglycine also has an amine group when it is not treated with either enzyme. Therefore, when a molecule of acetylated pentaglycine was
hydrolyzed by LasA, the products had the same O . D . 4 5 0 as unmodified pentaglycine
which was not treated with LasA. It was easy to accurately calculate theO . D . 4 5 0 values for a unit concentration of pentaglycine solution because the number of amine groups is the same as the number of molecules. On the other hand, because
78 the amine group was blocked in acetylated pentaglycine, the O.D. 4S0 was dependent on its cleavage by either LasA or LasD, making the standard O.D .450 unreliable.
To obtain the standard O . D . 4 5 0 , 100 pi of various concentrations of
pentaglycine solutions(3.3 pM to 1.6 mM) were reacted with 50 pi TNBSA working
solution for 20 min at room temperature and the O . D . 4 5 0 value for each concentration was calculated by subtracting the background (.257 O . D . 4 5 0 ) Then, for each concentration, the O . D . 4 5 0 for 1 pM pentaglycine was calculated, and
finally, the standard O . D . 4 5 0 for 1 pM pentaglycine was obtained as an average for all values. The assay was repeated twice and average values were used for calculation. A summary of how to obtain the standard O . D . 4 5 0 for pentaglycine was given in Table 4.1.
For LasD, the same assay was performed except that the acetylated pentaglycine and LasD were resuspended in both 25 mM diethanolamine buffer (pH
9.5) and 20 mM phosphate buffer (pH 8.0) in two different sets of experiments to see the effect of different pH on its activity.
4. Kinetic analysis of inhibitors
To identify the type of inhibition for LasA, 0.5 pg LasA was preincubated with the varying amounts of ophenanthroline (50 mM in methanol) for 3 min without the substrate. Then, different amounts of the acetylated pentaglycine (5 to
75 pi of 2 mg/ml acetylated pentaglycine resuspended in 20 mM phosphate buffer, pH 8.0) were added, and the final volume was adjusted to 100 pi by adding the same
79 phosphate buffer. Three sets of reactions were performed, and the final concentration of o-phenanthroline were 0.2, 0.5, and 1 mM. After incubating for 30 min at 37“C, 50 |j1 TNBSA working solution (10 fil TNBSA in 1.49 ml 50 mM boric acid, pH 8.5) was added and allowed to stand for 20 min at room temperature before
measuring the O . D . 4 s o . The experiment was repeated twice and the average O . D . 4 5 0
was used to make a Lineweaver-Burke plot [I/O .D .450 vs. l/(acetylated pentaglycine)] for determination of the inhibition mechanism (Figure 4.4). Again, the same assay was performed except that the acetylated pentaglycine and LasD were resuspended in both 25 mM diethanolamine buffer (pH 9.5) and 20 mM phosphate buffer (pH 8.0) in two different sets of experiments to see the effect of different pH on the activity of LasD (Figures 4.5 and 4.6).
D. Assays for genetic analysis
1 . Agarose gel electrophoresis
Agarose gel electrophoresis was carried out using standard protocols
(Ausubel, Brent e t al., 1992). The powder form of agarose (Gibco BRL,
Gaithersburg, MD) was mixed in TAB buffer and heated untü the agarose was completely dissolved. The components of the 50x TAE buffer stock were 252 g Tris base, 57.1 ml glacial acetic acid, 100 ml 0.5 M EDTA (pH8.0), and ddH20 to 1 L.
To make a working Ix TAE buffer, the stock solution was diluted 50 times in ddHzO and the final concentration was 40 mM Tris-acetate, 1 mM EDTA. The concentration of agarose in the solution depended on the size of the DNA band of
80 interest. In general, higher percentage agarose gels were used for larger DNA fragments. For example, if the DNA of interest was relatively large (greater than 1-
2 kb), a 1% gel solution (w/v) was made. If the size of the DNA was smaller
(several hundred base pairs, for instance), a 1.5% gel solution was made. Once the gel solution was prepared, 8 pi of ethidium bromide stock solution ( 1 0 mg/ml) was added per 100 ml of the gel solution. The solution was then poured into a gel cast
(Bio-Rad, Hercules, CA) and allowed to solidify for 20 minutes. During this time, one volume of lOX loading buffer (20% ficoll 400, 0.1 M NazEDTA, pH 8.0,1%
SDS, 0.25% bromophenol blue, 0.25% xylene cyanol) was added to nine volumes of the DNA sample. The sample was then loaded into the solidified gel on a gel apparatus (Bio-Rad, Hercules, CA) filled with IX TAE buffer. If a mini gel apparatus was used, the gel was run for 30 min at 75 mA. If a large gel apparatus was used, the gel was run for 2 hr at 100 mA. The DNA fragments in the gel could be visualized using ultraviolet light source (Fotodyne, Hartland, WT) due to intercalation of ethidium bromide.
2. Purification of DNA by Gene cleaning
DNA restriction or PCR fragments were isolated from ethidium bromide- stained agarose gel by gene cleaning using the GeneClean n kit (Bio 101, La Jolla,
CA). Gene cleaning can be performed either directly from an aqueous solution as well as from an agarose gel. If the DNA fragment was in solution, three volumes
(v/v) of 6 M Nal was added to the test tube and allowed to stand for 5 min at room temperature or on ice. If the DNA band was in an agarose gel, the band was cut
81 directly from the gel and weighed. Three volumes (w/v) of 6 M Nal were added to the tube and the tube was placed in 55°C water bath (VWR ) for 5 min or until the agarose was completely dissolved. Silica based Glassmilk was then added to the sample depending on the amount of DNA in the tube. If there was 5 \ig or less estimated amount of DNA, 5 pi of the Glassmilk was added. An additional 1 pi of the Glassmilk was added for each 0.5 pg of DNA above 5 pg. The tube was mixed well and placed on ice to allow binding of DNA to the silica matrix. The tube was mixed every 1-2 min to ensure that the Glassmilk stayed suspended. After 5 min the tube was centrifuged and the supernatant was discarded leaving the beads with bound DNA in a pellet. The pellet was washed 3 times with 10-50 volumes of wash buffer (50% ethanol, 20 mM Tris, pH 7.5,2 mM EDTA, and 0.4 M NaCl).
Following the final wash, 20-30 pi of TE (10 mM Tris-HCl, 1 mM EDTA) was added the tube and incubated for 3 min at 55°C. Finally, the tube was centrifuged in an Eppendorf microcentrifuge for 30 sec at top speed (14000 RPM) and the supernatant containing DNA was transferred to a new tube. After adding 1 pi of
RNase A (10 mg/ml in 10 mM Tris-Cl, pH 7.5, 15 mM NaCl), the purified DNA was stored at -20°C until future use.
3. Isolation of P. aeruginosa genomic DNA
Genomic DNA from various strains of P. aeruginosa (PAOl, PA220,
PA1103, and AD2116) was isolated using a Easy-DNA Genomic DNA Isolation Kit
(Invitrogen, San Diego, CA). A protocol provided by the manufacturer was
82 followed with minor modification. Initially, 350 |ol of the lysis solution (solution A) was added to 1 ml of overnight culture and the mixture was vortexed until the content was evenly dispersed. Cells were lysed by incubating at 65“C for 10 min.
Then, 150 pi of the precipitation solution (solution B) was added and the culture was vigorously vortexed until the sample was evenly viscous and freely moved in the sample tube. Chloroform (500 pi) was added to the sample tube and vortexed vigorously as before. The sample was centrifuged for 20 min at 14,000 rpm at 4°C and the DNA-containing aqueous phase was transferred to a new tube. To the
DNA solution, I ml of cold 95% ethanol (-20°C) was added and mixed by vortexing followed by incubation on ice for 30 min. The sample was centrifuged at 14,000 rpm for 15 min at 4°C and the supernatant was removed. The pellet was washed once with 500 pi of 80% ethanol. The resulting pellet was air dried for 5 min and resuspended in 100 pi TE containing 2 pi of 2 mg/ml RNase (40 pg/ml final concentration). The tube was incubated at 37°C for 30 min to allow complete RNA digestion. Finally, the sample was stored at 4°C until future use.
4. Isolation of plasmid DNA by miniprep
Miniprep isolation of plasmid DNA was done using either a conventional alkaline lysis method or Wizard Plus Minipreps DNA Purification System
(Promega, Madison, WI). For the conventional alkaline lysis method, a standard protocol (Bimboim and Doly, 1979; Bimboim, 1983) was followed with some modifications for extraction of plasmids. Initially, 5 ml LB medium was inoculated
83 with bacteria from a frozen stock and cells were grown overnight at 37°C. The culture was transferred to 1.5 ml microfiige tubes and centrifuged for 2 min in a microcentrifuge (Brinkmann Instruments, Westbury, NY) at maximum speed
(14,000 rpm) to pellet. Cells in the each tube were then resuspended in 200 pi MP buffer (25 mM Tris-HCl, pH 8.0,10 mM EDTA) or GTE buffer (50 mM glucose, 25 mM Tris-HCl, 10 mM EDTA) and mixed well by vortexing. NaOH/SDS solution
(90 pi 0.33 M NaOH, 10 pi 10% SDS) was added (100 pi) and mixed well as before.
Tubes were then transferred to a 65°C incubator and allowed to stand for 30 min for cell lysis. The tubes were then cooled at room temperature for 10 min and 100 pi of chloroform: isoamyl alcohol (24:1) mixture was added. After mixing, 100 pi of potassium acetate solution (60 ml 6 M potassium acetate, 11.5 ml glacial acetic acid, and 28.5 ml ddHzO per 100 ml) was added and mixed again by vortexing. Cells were centrifuged for 2 min at maximum speed and the top aqueous phase was placed into a fresh tube. This layer was spun again for 2 min to pellet any residual protein which might have been withdrawn together with the top phase.
Alternatively, the top phase could be extracted one more time with 100 pi chloroform: isoamyl alcohol (24:1). Cold isopropanol (200 pi) was added to each tube and mixed gently. The tube was centrifuged for 5 min at the maximum speed in a microcentrifuge and the supernatant was removed. The pellet was washed once with 75% ethanol to remove salts, and then dried for 5 min at 65°C and resuspended in 50 pi ddHzO (or TE) with 1 pi RNase A (10 mg/ml in 10 mM Tris-Cl, pH 7.5, 15 mM NaCl). Finally, the tubes were stored at -20°C for future use.
84 When the Wizard Plus Minipreps DNA Purification System was used, a standard protocol provided by Promega was followed with minor modification.
Cells were grown in 5 ml LB medium overnight. Cells were then centrifuged at
10,000 X g for 10 min, the supernatant removed, and the tube was placed upside- down on a paper towel to remove excess media. The pellet was completely resuspended in 300 pi of Cell Resuspension Solution (50 mM Tris, pH 7.5, 10 mM
EDTA, 100 pg/ml RNase A) and transferred to a 1.5 ml microcentrifuge tube. Cell
Lysis Solution (0.2 M NaOH, 1 % SDS) was added (300 pi) to the tube and mixed by inverting the tube several times. Neutralization Solution (1.32 M potassium acetate) was added (300 pi) and mixed as before. For each miniprep, one Wizard
Minicolumn was prepared by attaching a disposable 3 ml Luer-Lock syringe barrel
(Becton Dickinson & Co., Franklin Lakes, NJ) to each column. DNA Purification
Resin was then added (1 ml) to the column and the lysate was transferred to the column after centrifugation at 10,000 x g in a microcentrifuge for 5 min. The syringe plunger was carefully inserted and gently pushed the slurry into the Minicolumn.
The syringe was detached from the Minicolumn and the plunger removed. After reattaching the syringe barrel to the Minicolumn, 2 ml of Column Wash Solution
(80 mM Potassium acetate, 8.3 mM Tris-HCl, pH 7.5,40 pM EDTA, 55% ethanol) was added to the barrel of the Minicolunm/syringe assembly. The plunger was reinserted and again pushed gently to apply the wash solution to the column. The syringe was removed and the column was transferred to a 1.5 ml microcentrifuge tube. The Minicolumn/tube assembly was centrifuged at 10,000 x g for 2 min to dry
85 the resin. The column was transferred to a new 1.5 ml microcentrifuge tube and 50
^1 of either ddHzO or TE was added to the column. After centrifugation for 20 sec at 10,000 X g, the Minicolumn was removed and the plasmid DNA in the microcentrifuge tube was stored at -20°C.
5. Overexpression of Las A and LasA precursor in E. coli
To overexpress either active LasA or the LasA precursor, the stock strain
(DH5a/pUC18 (Sma I-Hind HI) + 1.7 kb Sma I-Hind IQ lasA) was grown overnight in 5 ml LB medium containing 100 ng/ml ampicillin. The culture was then centrifuged and a miniprep was performed to isolate the vector which contains the lasA gene. The lasA gene was purified using Gene Cleaning from an agarose gel after digestion with Sma I and Hind IQ. Next, PCR was performed to incorporate primers containing additional restriction sites for convenient cloning into pET3a
(Figure 2.2). For the start primer for active LasA, the restriction sites added were
Ase I (AT/TAAT) and Sma I (CCC/GGG). Ase I creates the same cohesive end as
Nde I and also creates a unique restriction site since an Nde I site exists within the lasA gene. In addition, a start codon (ATG) was added as a part of the Ase I site to initiate translation because the portion of lasA gene for the active LasA protein does not have the start codon. In the case of the end primer for active LasA, a Bam HI site (G/GATCC) was introduced following the stop codon for convenient cloning into pET3a. For the LasA precursor construct, Ase I was also introduced in the primer for the same reason as previously described. In addition, the original TTG
86 a. Start Primer for Active LasA (33mer)‘
5 -ACT, GCC. CGG. GAT. TAA. TGG. CGC. CGC. CAT. CCA. ACC-3’ I I Start Ala Pro Pro Ser Asn Sma I Ase I b. Start Primer for LasA Precursor (32mer)‘
5’-ACT, GCC. CGG. GAT. TAA. TGC. AAC. TGC. CCA. GCG. TG-3' I I Start Gin Leu Pro Ser Val Sma I Ase I c. End Primer for both Active LasA and LasA Precursor (34mer)‘
5 -TTT, AAA, GCT, TGG, ATC, CTC. AGA. GCG,CCA,GGC,CCG,G-3’ I Stop (5’-UGA-3’) Bam HI
‘ Primers were synthesized by Ransom Hill Bioscience, Inc. (Ramona, CA).
Figure 2.2: Primers used to overexpress active LasA and the LasA precursor.
87 start codon was changed to a universal ATG codon for more efficient translation in
E. coli. The end primer was the same as that of active LasA. PCR conditions are given in Table 2.8.
After PCR, 1% agarose gel electrophoresis was performed and a band with a correct predicted size (580 bp and 1140 bp for LasA and LasA precursor, respectively) was cut directly from the gel and purified by gene cleaning. The purified PCR product was digested with Bam HI and Ase I for 2 hr at 37°C. The reaction mixture (30 pi) consisted of 15 pi PCR product (ca. 1.5 pg/plO), 3 pL lOX
REACT buffer #3 (50 mM Tris-HCl, pH 8.0, 10 mM MgCU, 100 mM NaCl)(Gibco
BRL, Grand Island, NY), 10 pi sterile ddHzO, 1 pi Bam HI (8-12 units/pl)(Gibco
BRL, Grand Island, NY), and 1 pi Ase I (2-8 units/pl)(Gibco BRL, Grand Island,
NY). One unit of all restriction enzymes used in this study is defined as the amount of enzyme required to digest 1 mg of the appropriate substrate DNA completely in
60 min under the conditions specified for that enzyme.
The vector, pET3a, was prepared from the stock strain (HMS174/pET3a) by growing the strain in 5ml LB medium overnight with 100 pg/ml ampicillin followed by miniprep isolation of the plasmid DNA. The purified pET3a was digested with
Nde I and Bam HI. Because those two restriction enzymes use different buffers, two separate digestion reactions were used. First, 10 pi of the purified pET3a (ca. 3 pg/pl) was incubated with 1 pi Nde I (4-6 units/pl)(Gibco BRL, Grand Island, NY),
1.5 pi lOX REACT buffer #2 (50 mM Tris-HCl, pH 8.0, 10 mM MgCU, 50 mM
NaCl), and 2.5 pi sterile ddHzO for 2 hr at 37°C. Agarose gel (1%) was run with the
88 PCR Reaction Mixture Name Concentration Amount Used lOX Taq buffer* 200mMTris-Ha, 10 Hi pH 8.4, 500 mM KCl dNTPMix* 500 uM 20 ul Start Primer^ 50 |xM lu l End Primer^ 50 tiM lu l MgCL* 25 mM 6 ql Template DNA (pUC18/ 1.7 1 qg/gl Ini kb Sma I-Hind DI lasA) Sterile ddHzO - 60 |il Taq Polymerase* 5 units/gl lu l Total lOOul PCR Running Conditions Steps Description Temperature Duration
1 Initial Dénaturation 95“C 2 min 2 Dénaturation 95°C 1 min 3 Annealing 55°C 1 min 4 Extension 72°C 3 min 5 Repeat Steps 2-4 - 35 cycles
6 Hold 4°C -
Purchased from Gibco BRL (Grand Island, NY).
^ Synthesized by Ransom Hill Bioscience, Inc. (Ramona, CA).
Table 2.8; PCR conditions for overexpression of active LasA and the LasA precursor.
89 sample and the digested pET3a was purified by gene cleaning. Next, 10 pi of the
Nde I digested pET3a (ca. 2 pg/pl) was incubated with 1 pi Bam HI (8-12 units/pl),
1.5 pi lOX r e a c t buffer #3 (50 mM Tris-HCl, pH 8.0, 10 mM MgCh, 100 mM
NaCl)(Gibco BRL, Grand Island, NY), and 2.5 pi ddHzO for 2 hr at 37“C. Again, the Nde I-Bam HE double digested pET3a was purified by gene cleaning after running 1% agarose gel.
Ligation was performed at 15“C overnight. The reaction mixture (15 pi) was consisted of 8 pi Bam HI-Ase I double digested lasA PCR product (ca. 1.5 pg/pl), 3 pi Bam HI-Nde I double digested pET3a (ca. 1 pg/pl), 3 pi 5X T4 DNA ligase buffer (250 mM Tris-HCl, pH 7.6, 50 mM MgCL, 5 mM ATP, 5 mM DTT, 25% polyethylene glycol-8000)(Gibco BRL, Grand Island, NY), and 1 pi T4 DNA ligase
(0.5-2 units/pl). One unit for T4 DNA ligase is defined as the amount of T4 DNA ligase which catalyzes the exchange of 1 nmol ^^P-labeled pyrophosphate into ATP in 20 min at 37°C.
After ligation, the product (pET3a//asA PCR) was verified by 1% agarose gel electrophoresis and used to transform BL21 (DE3 /pLysS) competent cells.
Preparation of competent cells and procedures for transformation will be discussed in another section. In this case, pET3a//a&4 was first transformed to TBl and plated on LB agar containing 100 pg/ml carbenicillin (CblOO) and incubated overnight at 37°C. Individual transformants were inoculated in 5 ml LB(CblOO) tubes. From the 5 ml overnight culture, 3 ml was transferred to microfiige tubes for miniprep to confirm the presence of the vector with lasA insert. After the
90 confirmation, the remaining 2 ml was used to make a stock strain. The identified vector with insert was used to transform BL21(DE3/pLysS) for overexpression.
BL21(DE3/pLysS) was not used to make a stock strain since high levels of 3- lactamase production in the strain cause rapid antibiotic degradation resulting in overgrowth of colonies lacking plasmid.
Following transformation, cells were grown in 5 ml LB(CblOO) for 3 to 4 hr until O.D .600 reached 0.6-0.7. The cells were then inoculated into 50 ml LB(CblOO) and grown for 3 to 5 hr until O.D.soo reached 0.8. Two ml of the culture was removed to serve as an uninduced control. To the rest of the culture, IPTG was added to a final concentration of 0.5 mM and the cells were incubated at 37°C for an additional 3 hr. At the end of the incubation, cells were harvested after centrifugation at 5,000 rpm for 15 min and washed once with 50 ml Tris-buffered saline (TBS). The pellet was resuspended in solution consisted of 5ml sucrose solution (30% sucrose in 10 mM Tris-HCl, pH 8.0) and 500pl lysozyme solution (1 mg/ml lysozyme in 100 mM EDTA). Following a 30 min incubation on ice, the cells were lysed by repeated freeze-thawing (3 times) in liquid nitrogen. The sample was then centrifuged at low speed (3,000 rpm) for 10 min and the supernatant was collected. The supernatant was centrifuged once again at high speed (10,000 rpm) for 15 min and the insoluble inclusion bodies were collected as a pellet. The inclusion bodies were washed repeatedly with 5ml of buffer containing 25 mM Tris-
HCl, pH8.0, 10 mM EDTA, and 2% Triton-XlOO to solubilize membrane proteins.
Finally, the pellet containing either active LasA or LasA precursor was washed repeatedly with water and stored at -70°C until use. To solubilize the protein, the
91 thawed sample was denatured in 6 M guanidine-HQ (or 8 M urea) and renatured by step-wise dilution to 0.5 M guanidine-HQ. It was impossible to remove all the urea because the protein was precipitated once the urea concentration decreases below 0.5
M. The overexpressed active LasA (or precursor LasA) was analyzed to be greater than 90% pure by SDS-PAGE.
6. Production of competent cells
A standard CaCh method was followed to make competent cells (Mandel and Higa, 1970; Dagert and Ehrlich, 1974) with minor modifications. Cells were inoculated from a frozen stock into 1-2 ml LB medium and incubated overnight at
37°C. From the overnight culture, 50 pi was inoculated again into 50 ml LB medium and incubated until O.D.eoo of 0.6. The cells were harvested at 4,000 rpm for 10 min at 4°C. After removing supernatant, cells were resuspended in 10 ml of ice-cold sterile 50 mM CaCh and stored on ice. Cells were recovered by centrifugation at 4,000 rpm for 10 min at 4°C and the tubes were dried by inversion on a paper towel for 1 min. The pellet was resuspended in 20 ml of 50 mM CaCU and kept on ice for 20 min. Following incubation, cells were spun down and the resulting pellet was resuspended in 2 ml of 50 mM CaCL, 10% glycerol. Finally, aliquots (200 pi) were dispensed into microfiige tubes and stored at -70°C.
7. Transformation of plasmid DNA
Two different methods were used for transformation of plasmid DNA into bacterial cells. One is a conventional method using competent cells generated by
92 CaClz treatment and the other was electroporation. In the CaClz method (Mandel and Higa, 1970; Dagert and Ehrlich, 1974), the DNA sample was added directly into a frozen tube of competent cells and remained on ice until the cells thawed. At this time, the tube was tapped gently to mix the cells with DNA and was put on ice.
After 1 hr, the cells were heat shocked at 42“C for 3 min and put back on ice for 5 min. LB medium (400 pi) was added and the tube was incubated for 2 hr at 37“C.
The tube was centrifuged and the supernatant was discarded. The cell pellet was resuspended in lOOp LB medium and plated on LB agar containing 100 pg/ml carbenicillin. After overnight incubation, colonies were picked up from the plate and incubated in 5 ml LB(CblOO) tubes overnight. Using half of the culture, minipreps were performed and transformation was verified by appropriate restriction digestion. After verifying transformation, a stock strain was made with the rest of the culture.
For electroporation (Dower, Miller etal., 1988; Smith and Iglewski, 1989;
Diver, Bryan etal., 1990; Kilbane and Bielaga, 1991), E. co//cells were grown in 1 L
LB medium at 37°C with shaking until O.D.goo reached 0.5 to 0.7 (early- to mid-log phase). Cells were then harvested by centrifrigation at 4000 x g for 15 min at 4°C.
After removing supernatant, cells were resuspended in 1 L cold sterile 10% glycerol and centrifuged again as before. Cells were collected and resuspended 3 times in decreasing volumes of 10% glycerol (0.5 L, 20 ml, and 2 ml). At this stage, cells could be frozen in aliquots and stored at -70°C.
To transform cells by electroporation, a Gene Puiser (with Pulse Controller) from Bio-Rad (Hercules, CA) was used. The following procedure was from the
93 operating instructions for the Gene Puiser. The first step was to prepare cold cuvettes and a chamber slide by placing them on ice. In the meantime, cells were thawed at room temperature and immediately placed on ice if they were from the frozen stock. If cells were freshly prepared, they were kept on ice. In a cold 1.5 ml polypropylene tube, 40 pi of the cold cells were mixed with 1 to 2 pi of plasmid
DNA and placed on ice until transferring contents to a chilled cuvette. The Gene
Puiser apparatus was set at 25 pF of capacitance and the Pulse Controller was set to
200 Q of resistance. For the voltage setup, the Gene Puiser apparatus was set at 2.5 kV when using 0.2 cm cuvettes. If 0.1 cm cuvettes were used, the apparatus was set at 1.8 kV. The already cold mixture of cells and DNA was transferred to a cold electroporation cuvette, and the cuvette was placed in a chilled safety chamber slide.
Following delivery of the pulse at the above settings, the cuvette was removed from the chamber and 1 ml of SOC medium (2% Bacto tryptone, 0.5% Bacto yeast extract,
10 mM NaCl, 2.5 mM KCl, 10 mM MgCU, 10 mM MgS 0 4 , 20 mM glucose) was immediately added. (The time constant was checked after the pulse to make sure that it was between 4 and 5 msec.) The cell suspension was transferred to a 1.5 ml tube and incubated at 37°C for 1 hr with shaking. The cells were plated on LB plate
(with antibiotics if necessary) and incubated overnight at 37°C. Colonies were picked and grown in 5 ml LB medium. Finally, a miniprep was performed and resulting plasmids were analyzed by restriction digestion or other appropriate methods.
94 D. Immunological and other assays
1. Production of polyclonal antisera
Anti-LasD antisera was prepared using two male New Zealand white rabbits.
For the primary injection, 200 pg of purified LasD was isolated from 12% SDS-
PAGE gels by excision and the gel slices containing LasD were lyophilized. The lyophilized LasD gel slices were finely grounded and resuspended in I ml of PBS buffer (80 g NaCl, 2 g KCl, 11.5 g NazHP 0 4 -2 H2 0 , 2 g KH 2PO4, and ddHzO to 1 L).
The resuspended LasD was mixed with the same volume of complete Freund’s adjuvant (CFA) (Freund, 1956; Harlow and Lane, 1988). To generate an emulsion between LasD and CFA, 1 ml of LasD solution was put into a glass syringe and 1 ml of the adjuvant was added to another glass syringe. After removing all air, the two syringes were connected through a luer fitting. The plunger from a syringe containing LasD was depressed driving LasD into the oil of the adjuvant. Plungers were alternately pushed for several minutes until an emulsion was formed.
Following prebleeding, each rabbit was immunized with 100 pg LasD (1 ml
LasD/adjuvant mix), injected subcutaneously into five separate locations at the shaved area. The rabbits were checked after an hour to make sure that they were awaken and not bleeding any more. Following a 4 week interval, each rabbit was immunized intramuscularly with 50 pg of purified LasD resuspended in 0.5 ml incomplete Freund’s adjuvant (IFA). A third immunization was performed at the end of the 7* week with the same amount of LasD (50 pg in 0.5 ml IF A). One week
95 following the final immunization, the rabbits were terminally bled and the sera were analyzed for specificity at a 1:5000 dilution in a Western blotting procedure.
Polyclonal antiserum was also raised against the LasA precursor to study the processing of LasA. The procedure was exactly the same as that of anti-LasD antibody. The only subtle difference was that the LasA precursor had to be obtained from SDS-PAGE gel by excision due to the presence of contaminants in the sample.
It was not necessary to get the LasD protein from SDS-PAGE because the sample was already pure. However, LasD was prepared from a gel as a precautionary measure.
2. Immunoblotting (Western blotting)
A standard protocol (Towbin, Staehelin etal., 1979; Ausubel, Brent etal.,
1992) was used for transferring proteins from a polyacrylamide gel to a supported nitrocellulose membrane (Schleicher & Schuell, Keene, NH). After running a polyacrylamide gel, the gel sandwich was disassembled and stacking gel portion was removed. A transfer cassette was assembled in a tray containing transfer buffer to cover the whole cassette. Composition of the transfer buffer was 600 ml lOx transfer buffer (60.55 g Tris, 288.4 g glycine, 20 g SDS, ddHzO to 2 L), 1.2 L methanol, and
4.2 L ddHzO. The assembly contained the gel and nitrocellulose membrane in the middle surrounded by Whatman 3MM filters (Whatman International Ltd.,
Maidstone, England) and sponges on both sides. The cassette was placed into a blotting apparatus filled with transfer buffer. Proteins were transferred from the gel to the nitrocellulose paper electrophoretically for either 1 hr (large gel) at 0.6-1.0 A
96 or 20 min (small gel) at 0.3-0.45 A. Following transfer, the nitrocellulose paper was carefully lifted from the cassette and transferred to a tray containing blocking solution composed of 2% skim milk in 50 ml TBS (100 mM Tris-HCl, pH7.5,0.9%
NaCl). After 1 hr of incubation with swirling at room temperature, the nitrocellulose paper was washed with TTBS (0.1% Tween 20 in TBS) for 5 minutes and incubated with the primary antibody (1:2000 diluted serum in TBS with 1% skim milk) for 3 hours at room temperature followed by 3 washes (5 min each) with
TTBS.
The next step is to detect the protein-antibody interaction using one of three different methods. Usually, conventional secondary antibody-Horse radish peroxidase (HRP) conjugate was used to detect proteins on the nitrocellulose membrane (Knect and Dimond, 1984). In this case, after incubating the nitrocellulose membrane with goat anti-rabbit (or donkey anti-rabbit)-HRP conjugate (Amersham Life Science Inc., Arlington Heights, IL)(25 pi conjugate in
50 ml TBS with 1% skim milk) for 2 hr at room temperature, a mixture of H 2O2 (30 pi H2O2 in 50 ml TBS) and HRP color development reagent (30 mg 4-chloro-l- naphthol in 10 ml methanol, Bio-Rad, Hercules, CA) was incubated with the nitrocellulose membrane for 30 min in the dark at room temperature. After the incubation, a purple color developed in the region where the protein of interest reacted with the primary antibody. In spite of its convenience, this method had two major problems: first, the intensity of the color was not strong enough in many cases due to its low sensitivity, and second, because the color reagent was sensitive to light, the developed color fades over time.
97 To overcome some of these problems, ‘“I-Protein A was used in place of the color reagent in some cases, especially when high sensitivity was an important issue
(Burnette 1981). Following primary antibody incubation and washing, ‘“I-Protein
A (10 pCi/|il, Amersham Life Science Inc., Arlington Heights, IL) solution (5 pi in
50 ml TBS with 1% skim milk) was incubated with the nitrocellulose membrane at room temperature with low speed swirling. After 3 hr incubation, the nitrocellulose membrane was washed 3 times with TTBS (5 min each) and once with TBS. The membrane was air dried and exposed to a X-ray film in an autoradiography cassette at -70°C for overnight before developing the film.
Even though the radioactive method solved some of the problems the color reagent had, special handling was required for radiolabeled chemicals and, therefore rendered this method inconvenient. For ease and high sensitivity, a chemiluminescent substrate was used in some cases instead of radiolabeled Protein
A (Isacsson and Wettermark, 1974). In this assay, the nitrocellulose membrane was thoroughly washed after incubating with the secondary antibody-HRP conjugate.
Then, the membrane was reacted with a chemiluminescent substrate (ECL Western blotting kit, Amersham Life Science Inc., Arlington Heights, EL) which was composed of two separate solutions (Ausubel, Brent etal., 1992). One substrate contained luminol, H 2O2, and />iodophenol. In the presence of HRP and H 2O2, luminol was oxidized and emitted blue light. The other chemical, />iodophenol, was used to increase the light output. The second part solution contained 50 mM Tris-
HCl, pH 7.5, used as a buffer. For a large membrane, 5 ml of each solution (1-2 ml for a small membrane) was mixed and the membrane was immersed in substrate for
98 1-3 min. The membrane was removed from the mixture and the reaction was visualized through exposure to X-ray film for 1-5 min depending on the intensity of the signal.
3. Fractionation of Pseudomonas aeruginosa
The fractionation procedure was performed using a previously published protocol (Cheng, Ingram etal., 1971; Hoshino, 1979) with minor modifications.
Harvested cells were washed with TMK buffer (10 mM Tris-HCl, pH7.4,1 mM KCl,
1 mM MgCU) and resuspended in 20 ml MgCU extraction buffer (10 mM Tris-HCl, pH8.4, 0.2 M MgCU) per gram of cells (wet weight). After 4 min incubation at 30°C, the cell suspension was rapidly chilled in an ice bath for 15 min. The temperature shift was repeated one more time and the supernatant was collected following centrifugation (4,000 x g) for 15 min at 4°C. The remaining pellet was rapidly dispersed in ddHzO (20 ml per 1 g wet weight) and incubated for 20 min at room temperature. The supernatant was collected following centrifugation (4,000 x g) for
15 min at 4°C and added to the previously collected supernatant. The combined supernatant was centrifuged twice at 30,000 x g for 3 min at 4°C to remove whole cells. The supernatant was then concentrated by filtration using a YM-10 filter
(Amicon, Beverly, MA) and dialyzed overnight at 4°C against TMK buffer. Finally, the resulting periplasmic fraction was stored at -70°C.
To obtain a cytoplasmic fraction, the pellet from previous stage was resuspended in 5 ml (per 40 ml original culture) 10 mM Tris-HCl, pH 8.0,100 mM
NaCl, 1 mM MgCU, and 1 mg/ml DNase. The suspension was quickly frozen in an
99 ethanol/dry ice bath and thawed at 37®C three times. After centrifugation at 3,000 x g for 15 min, the resulting cytoplasmic fraction was concentrated and dispensed in small volumes and stored at -70°C.
The pellet was washed twice in saline solution (0.9% NaCl in ddHzO) and resuspended in 5 ml (per 40 ml original culture) 10 mM Tris-HCl, pH8.0,0.1 M
NaCl, and 2% Triton X-100. The suspension was incubated for 30 min at 25°C then centrifuged at 100,000 x g for 60 min at 4°C using a ultracentrifuge (model L8-M from Beckman, Fullerton, CA). The supernatant containing the soluble membrane fraction was collected and stored at -70°C in small aliquots.
100 CHAPTERS
PURIFICATION AND CHARACTERIZATION OF LasA AND LasD
A. Introduction
Among several extracellular proteases identified in P. aeruginosa, pseudolysin (elastase) is considered to be one of the major factors for pathogenicity of P. aeruginosa. Originally purified and described by Morihara eta l (Morihara,
Tsuzuki etal., 1965), pseudolysin is classified as a zinc metalloprotease similar to many other bacterial proteases. Pseudolysin was originally named as elastase because of its ability to degrade elastin thus distinguishing it from other proteases secreted by P. aeruginosa. In 1990, another protease with elastolytic activity, LasA, was purified and it was demonstrated that full elastolytic phenotype in P. aeruginosa requires both pseudolysin and LasA (Peters and Galloway, 1990). Later, it was also demonstrated that LasA has staphylolytic activity thereby supporting a role of LasA as a virulence factor (Kessler, Safnn etal., 1993).
Originally, LasA was isolated as a by-product of pseudolysin purification, however it was an inefficient procedure because of its low yield of LasA (Peters,
1990). When the concentrated supernatant fraction was passed through a DEAE-
Sephacel column at pH 8.0, few fractions contained a relatively high concentrations
101 of LasA with minor contaminants present. Other fractions contained high concentrations of various contaminants including a 23 kDa protein. To increase the yield of purified LasA, two different approaches were undertaken in this study.
First, using a different purification procedure, it was possible to separate LasA from other contaminating proteins, including the 23 kDa protein; second, LasA was overexpressed in E. ooZ? using the T7 expression system (Chapter 5). As a result, the problem of low yield was resolved and larger quantities of LasA became available for further characterization.
After separating LasA from other contaminants, the fractions containing other proteins were further investigated and three proteins of 15 kDa, 23 kDa, and
30 kDa were purified to homogeneity. Surprisingly, it was found that the 23 kDa protein (designated LasD) also has significant protease activity. First, LasD exhibits strong proteolytic activity on several substrates even through its elastolytic activity is low; second, LasD is also a staphylolytic enzyme under alkaline conditions (pH
9.5); third, when combined with pseudolysin, LasD shows a significant increase in proteolytic activity supporting the previous hypothesis that LasA and pseudolysin work together for full proteolytic activity (Peters and Galloway, 1990).
To characterize a new enzyme such as LasD, it is necessary to investigate several aspects: the enzyme must be purified to homogeneity, physical characteristics of the enzyme should be studied, substrates for the enzyme must be identified, and finally, inhibitors of the enzyme need to be analyzed to elicit the mechanism of the enzymatic reaction. In this chapter, the purification and characterization of LasD will be discussed in detail. In addition, further
102 characterization of LasA substrate specificity using synthetic peptides and inhibitors will also be discussed.
B. Purification
1. Purification of LasA and LasD
The LasA structural gene {lasA) was first cloned from a clinically isolated mucoid P. aeruginosa strain FRDl (Goldberg and Ohman, 1987) and the sequence was reported by Schad and Iglewski (Schad and Iglewski, 1988). The nucleotide sequence has since been revised (Darzins, Peters etal., 1990), and LasA has been purified from P. aeruginosa strains (PA220 and PAOl) by Peters and Galloway as a by-product of pseudolysin purification (Peters and Galloway, 1990). Briefly, concentrated culture supernatant is dialyzed against 20 mM phosphate buffer
(pH8.0) and loaded onto a DEAE-Sephacel column. Under this condition, pseudolysin is retained in the column due to its relatively low isoelectric point (pi of
6.5). However, LasA is found in pass-through fractions since the pi of LasA is 9.5.
Pseudolysin can be eluted by applying a gradient from 0 to 0.1 M NaCl in the same phosphate buffer and the eluted pseudolysin is almost free of contaminating proteins. A typical purification table for pseudolysin is given in Table 3.1.
When LasA-containing fractions are concentrated and analyzed by SDS-
PAGE, many other proteins including a 23 kDa protein (LasD) are also present.
Originally, LasA was separated from most of the contaminating proteins using a
FPLC Mono-P colunm (Pharmacia Biotech, Piscataway, NJ) at pH9.5 since only
103 Step Volunae Total Specific Purification % (ml) Protein* (mg) Activity^ (-foldy Yield"* Total Supernatant 2810 41250 - - - First 60% ammonium 980 14480 0.2 - 35.1 sulfate precipitation Second 60% ammonium 175 4740 3.4 2.8 11.5 sulfate precipitation Acetone precipitation 120 250 7.4 6.2 0.61 and dialysis DEAE-Sephacel and 10 12 12.3 61.5 0.03 concentration
‘ Total protein as determined by the Bradford method.
^ Specific activity defined as mg of elastin-Congo red digested per mg of total protein per hr.
^ Purification (-fold) determined by dividing the specific activity of the sample by the initial specific activity of the total supernatant fraction.
Percentage yield obtained by dividing the amount of total protein by the amount of total protein in the initial supernatant fraction.
Table 3.1: A typical purification table for pseudolysin.
104 LasA, the 23 kDa protein, and a 15 kDa protein passed through the column while the other proteins were retained in the column. However, even though the Mono-P column removes many contaminating proteins, the LasA-containing pass-through fractions are still not separated from the contaminating 23 kDa and 15 kDa proteins.
Therefore, by applying Mono-P coiunm separation alone, LasA can not be separated to homogeneity. Originally to bypass this problem, LasA was separated from the 23 kDa protein by using only the early fractions following DEAE-Sephacel column chromatography because the 23 kDa protein eluted later than LasA from that column (Peters and Galloway, 1990). However, as a result of using a few early fractions instead of all LasA-containing fractions, the yield of LasA was very low
(approximately 100 pg per 3 L culture).
To overcome these problems, the FPLC Mono-P column was replaced by
CM-Cellulose (Sigma Chemical Company, St. Louis, MO) and FPLC-Mono-S columns (Bio-Rad, Hercules, CA). CM-cellulose is a strong anion exchange resin, and when used following the DEAE-Sephacel column, the 15 kDa protein is retained in the column, while both the LasA and the 23 kDa proteins pass through the column. In this way, it is possible to have pass-through fractions containing only
LasA and the 23 kDa proteins. The DEAE-Sephacel column step can be omitted if the purpose of the purification is to isolate only LasA and the 23 kDa proteins.
However, if pseudolysin is also to be purified, the DEAE-Sephacel column is necessary because elution of pure pseudolysin from the CM-Cellulose column is difficult due to the strong anionic nature of the carboxymethyl functional group.
105 The difFerence between the Mono-S and the Mono-P columns is that the
Mono-S column contains a strong cation exchange resin with -CH 2-SO3' as the charged group while Mono-P is a weak anion exchange resin. When 50 mM
HEPES buffer (pH 8.0) is used, LasA is retained in the column and can be eluted by a shallow salt gradient. However, the 23 kDa protein passes through the column and is collected as a pure protein (Figure 3.1). Purification tables of LasA and the 23 kDa protein (LasD) are provided in Tables 3.2 and 3.3, respectively, and a flow chart for the overall purification scheme is given in Figure 3.2.
As mentioned earlier, the 23 kDa protein (LasD) is a major contaminant in the initial stages of LasA purification. To increase the yield of LasA, a variety of methods were employed in addition to the FPLC Mono-S column chromatography to eliminate the 23 kDa protein from the LasA-containing fractions. For example, the 23 kDa protein can be separated from LasA by preparative isoelectrofocusing gel electrophoresis (Evans, Wilson etal., 1989; St. Clair and Sax, 1990; Petrash,
DeLucas etal., 1991) using a Rotofor Cell (Bio-Rad, Hercules, CA). Using this technique, the 23 kDa protein is separated from LasA due to their difference in pi; around 8.5 for the 23 kDa protein and 9.5 for LasA. However, there are two minor problems in this method. Due to suspected precipitation of some proteins in the separation chamber, the yield is relatively low (about 200 pg for LasA and 250 pg for the 23 kDa protein from a 3 L culture). The other problem is that both LasA and the
23 kDa protein are purified in the presence of a high ampholyte concentration even
106 Figure 3.1: Chromatographic elution profile of the LasA and LasD proteases using the FPLC Mono-S cation exchange column.
The sample containing LasA and LasD was resuspended in 50 mM HEPES buffer
(pH 8.0) and the flow rate was 1 ml/min. The LasA protease elutes under low salt conditions (0 - 0.1 M NaCl gradient) whereas LasD passes through the Mono-S column (Pharmacia Biotech, Piscataway, NJ).
107 ft 0.5-1 w
0.4 -
100
0.3 - E LasD I o d 00 d (O 0.2 - I
LasA
0.1-
40 Fraction numbers Step Volume Total Specific Purification % (ml) Protein^ (mg) Activity^ Yield?
Total Supernatant 2790 40020 1.6 - - First 60% ammonium 935 15740 5.3 3.3 39.3 sulfate precipitation Second 60% ammonium 170 6730 12 7.5 16.8 sulfate precipitation Acetone precipitation 120 430 43 26.9 1.1 and dialysis DEAE-Sephacel 220 29 90 56.3 0.07 CM-Cellulose and 2 3.5 210 131.3 0.009 concentration FPLC Mono-S and 2 1.2 780 487.5 0.003 concentration
Total protein amount as determined by the Bradford method.
^ Specific activity determined by measuring the change in O.D.sgs after incubating heat-killed S. aureus ceüs (in 20 mM phosphate buffer, pH8.0) with various fractions for 2 hr at 37°C and is expressed as AO.D.sgs per mg of total protein.
^ Purification (-fold) determined by dividing the specific activity of the sample by the specific activity of the initial total supernatant fraction.
Percentage yield calculated by dividing the amount of total protein in the sample by the amount of total protein in the initial supernatant.
Table 3.2: A typical purification table for LasA.
109 S# Volume Total Specific Purification % (rnl) Protein* (mg) Activity^ Yield* Total Supernatant 2750 36740 1.2 -- First 60% ammonium 935 15250 7.1 5.9 41.5 sulfate precipitation Second 60% ammonium 167 7760 13.4 11.2 21.1 sulfate precipitation Acetone precipitation 129 380 39 32.5 1.0 and dialysis DEAE-Sephacel 230 23 92 76.7 0.06 CM-Cellulose and 2 3.5 370 308.3 0.01 concentration FPLC Mono-S 2 2.1 819 682.5 0.006
' Total protein amount as determined by the Bradford method.
^ Specific activity determined by measuring the change in O.D .595 after incubating heat-killed S. aureus cells (in 25 mM diethanolamine buffer, pH9.5) with various fractions for 2 hr at 37°C and is expressed as AO.D. 59S per mg of total protein.
^ Purification (-fold) determined by dividing the specific activity of the sample by the specific activity of the initial total supernatant fraction.
Percentage yield calculated by dividing the amount of total protein in the sample by the amount of total protein in the initial supernatant fraction.
Table 3.3: A typical purification table for LasD.
110 Figure 3.2: A flow chart representing the purification procedures of various proteins secreted by Pseudomonas aeruginosa.
‘ All steps for ammonium sulfate salt cut, acetone precipitation, and dialysis are omitted for a simple graphical representation of the purification procedures.
Ill 5 ml Inoculum 1 3 L Culture 1 Concentration steps' I DEAE-Sephacel
0.1 M NaCl elution CM-Cellulose
[ Pass-through I Pseudolysin 0.1-0.5M Mono-S NaCl gradient
15 kDa Pass-through Protein Shallow 0-0. IM NaCl gradient elution
LasD LasA
Figure 3.2 112 though the ampholytes do not seem to interfere with the biological activities of
LasA.
The next method applied to separate LasA from the 23 kDa protein (LasD) was preparative polyacrylamide gel electrophoresis under native conditions using the Model 491 Prep Cell from Bio-Rad (Hercules, CA) (Fountoulakis, Takacs Di
Lorenzo etal., 1993; Kyriakopoulos, Kalckloesch etal., 1993). In this method,
LasA and the 23 kDa protein can be separated even though they are very similar in size (22 kDa for LasA) because non-denaturing conditions allow for separation based not only on sizes but conformation and net charge of the proteins as well.
Denaturing conditions are avoided because it is very difficult to remove SDS following protein elution.
The 23 kDa protein can also be purified by HPLC using the MA7C cation exchange column (Bio-Rad, Hercules, CA) following DEAE-Sephacel column chromatography. MA7C is a nonporous column containing negatively charged carboxyl groups as the functional group. When this column is used, LasA and other contaminating proteins are retained in the column even at a rapid flow rate (2-5 ml/min) while the 23 kDa LasD passes through the column. LasA is then eluted under low salt conditions (less than 0.1 M NaCl). Various running conditions are detailed in the Materials and Methods section.
2. Production of anti-LasD and anti-LasA precursor polyclonal antisera
In order to verify that each of these proteins is unique, specific rabbit antisera against each protein was prepared using two male New Zealand rabbits for each
113 protein. The antisera were subsequently used in immunoblotting analysis to confirm the lack of any cross-reactivity (Figure 3.3). Significantly, use of the anti-LasD antiserum revealed the existence of a 30 kDa protein in one of the fractions eluted following CM-Cellulose chromatography (Figure 3.3, panel B, lane 3). However, the 30 kDa protein is undetectable in the culture supernatant fraction probably due to its low concentration (Figure 3.3, panel B, lane 4). This 30 kDa protein is apparently cleaved or fragmented during purification, resulting in the formation of the active 23 kDa LasD fragment, and therefore, may represent a precursor form of the LasD protein. The peak containing the 30 kDa protein does not reveal the presence of any contaminating protein (Figure 3.3, panel A, lane 3). Further studies are presented in the discussion section of this chapter. The anti-Las A failed to recognize LasD, confirming that LasA and LasD are two separate, unique proteins
(Figure 3.3, panel C, lane 1). Furthermore, LasD is detected in the concentrated supernatant protein fraction from the LasA mutant strain FRD 2128. This is conclusive proof that LasA and LasD are two different proteins (Figure 3.3, panel B, lane 5). However, LasA is not detected in the same sample, confirming that the
LasA' mutant strain indeed does not produce LasA (Figure 3.3, panel C, lane 5).
C. Physical characteristics of LasA and LasD
1. pH and temperature optima of LasA and LasD
As demonstrated in the previous section, LasA and LasD display many similarities in substrate specificity. However, the staphylolydc activities of these two
114 Figure 3.3: SDS-PAGE and immunoblot analyses of the purified LasA and LasD proteases following final chromatographic separation.
A. SDS-PAGE (12%) separation and staining of the purified LasA and LasD
proteases, including a 30 kDa protein. Lane 1,2 pg LasD; lane 2, 1 pg LasA;
lane 3, 1 pg 30 kDa protein; lane 4,10 pg concentrated supernatant from P.
aeruginosa strain PA220; lane 5, 10 pg concentrated supernatant from P.
aeruginosa LasA strain FBD2128; and lane 6, prestained protein molecular
weight standard.
B. Autoradiograph from immunoblot of the purified proteases as above using
specific anti-LasD antisera (1:2000) and [‘“I]-labeled Staphylococcus aureus
Protein A.
C. Autoradiograph from Immunoblot of the purified proteases as above using
specific anti-LasA antisera (1:2000) and [*“I]-labeled Staphylococcus aureus
Protein A.
115 -43 kD a
-29 kD a
-21 kD a •14 kD a •••V- : • ' . • / S - f e S i t S f e
1 2 3 4 5 6
B
LasD- LasA-
1 2 3 4 5 1 2 3 4 5
Figure 3.3
116 proteases can be distinguished on the basis of their respective pH optima, which indicate that the LasD protease is incapable of staphylolytic activity at neutral pH and is only functional in this regard under higher pH conditions, e.g. pH9.5 (Figure
3.4). This factor is crucial in the identification of the separate existence of these two staphylolytic proteases and further demonstrates that these two proteins are distinct proteins. If there is no difference between the specificities of LasA and LasD, one could argue that the staphylolytic activity is due to contamination of the LasD sample by LasA, or vice versa. However, because of their difference in pH optima, the existence of two separate staphylolytic enzymes can be confirmed.
In addition to their difference in pH optima, LasA and LasD also exhibit differences in their optimum temperatures for staphylolytic activity. As shown in
Figure 3.5, when the staphylolytic activity is measured at pH 7.0, LasA shows the strongest activity at 37°C. Surprisingly, LasD shows the strongest activity at 55°C when measured at pH 9.5 further suggesting that LasA and LasD are two totally different proteases.
2. Determination of N-terminal amino acid sequence of LasD
A further indication of the unique nature of LasD was revealed when the N- terminal sequence of the protein was determined. Briefly, LasD was first transferred from an SDS-PAGE gel to PVDF (polyvinylidene difluoride) membrane and stained. The stained band was then excised and the N-terminal sequence was determined using an Applied Biosystems (Foster City, CA) Gas Phase Protein
Sequencer. Fifteen N-terminal residues were identified after the initial sequencing.
117 1.6
•LasA ■LasD
0.8 -
4.75 7 8.5 9.5 10.5 pH
Figure 3.4: Effect of pH* on staphylolytic activity^ of LasA and LasD.
* * Buffers used in the assay are: 0.1 M Tris-HCl, pH4.75 0.1 M Tris-maleate, pH7.0 0.1 M Tris-HCl, pH8.5 0.025 M diethanolamine-HCl, pH9.5 0.025 M diethanolamine-HCl, pHlO.5
^ Staphylolytic activity is represented as a function of 1/O.D. 595 . Since the higher O . D . 5 9 5 values represent the lower staphylolytic activity (cells are not lysed), the
higher 1/ O . D . 5 9 5 values represent higher staphylolytic activity.
118 4.5
■LasD •LasA
1.5
0 10 20 30 40 50 60 70 80 Temperature (°C)
Figure 3.5: Temperature optima for staphylolytic activity'"^ of LasA and LasD.
‘ Staphylolytic activity was measured in the presence of 3 pg purified LasA or LasD using a 670 pi suspension of heat-killed S. aureus in either 20 mM phosphate buffer (pH 7.0, for LasA) or 25 mM diethanolamine buffer (pH 9.5, for LasD) at various temperatures.
^ Staphylolytic activity was represented as a function of 1 / O . D . 5 9 5 . Since the higher
O . D . 5 9 5 viues represent the lower staphylolytic activity (cells are not lysed), the
higher 1/ O . D . 5 9 5 values represent the higher staphylolytic activity.
119 To verify this sequence, the procedure was repeated and 26 N-terminal amino acid residues were identified. The sequence is His - Gly - Ser - Met - GIu - Thr - Pro - Pro
- Ser - Arg - Val - Tyr - Gly - Cys (or other modified amino acid) - Phe - Leu - Glu -
Gly - Pro -Glu - Asn - Pro - Lys - X - Ala - Ala. Using this sequence, peptide data base searches were carried out routinely at the National Center for Biotechnology
Information (NCBI) using the BLAST network service. Although there were several proteins with an indicated minor homology in amino acid composition, these analyses failed to reveal any protein with significant homology to LasD.
As mentioned previously, immunoblotting using anti-LasD identified a 30 kDa protein which cross-reacted strongly with the LasD specific antiserum. The N- terminal sequence of the 30 kDa protein was also determined using the same method. In this case, only 9 residues were identified with the sequence Ala - X - Ser
- Met - Glu - Thr - Gly - Pro - Ala. Again, database analysis revealed no significant homology with other known proteins.
D. Substrate analysis of LasA and LasD
1. Staphylolytic activity of LasA and LasD
The purified LasD protease clearly demonstrates the ability to lyse heat-killed
Staphylococcus aureus axià, therefore, may be designated as a staphylolytic protease
(Burke and Pattee, 1967; Lache, Hearn etal., 1969; Brito, Falcon etal., 1989;
Camicero, Falcon etal., 1990), analogous to the LasA protease. In Figure 3.6, both
LasA and LasD demonstrate strong staphylolytic activity at pH 9.5 after a 30 min
120 Figure 3.6: Staphylolytic activity*’^ of LasA and LasD.
^ Staphylolytic activity was measured in the presence of 3 pg purified LasA, elastase, or LasD using a 670 pi suspension of heat-killed S. aureus in either 20 mM phosphate buffer (pH 8.0, for LasA and pseudolysin) or 25 mM diethanolamine buffer (pH9.5, for LasD) for at 37°C.
^ Staphylolytic activity was measured spectrophotometrically. The O.D .595 values drop when cells are lysed. As a result. The higher O.D.595 values represent the lower staphylolytic activity and the lower O.D.595 values represent the higher staphylolytic activity.
121 1 . 5 -
E c aiu> u> « d d
0 . 5 -
0 0 . 51 1 . 52 2 . 5 3 3 . 5 Time (Hr)
LasD LasA + Pseudolysin
LasD +Pseudolysin Pseudolysin
LasA Untreated Control
Figure 3.6 122 incubation with heat-killed S. aureus. Initially, LasA exhibits more than two-fold higher activity than LasD. Following a 3 hr incubation, however, both proteases shows similar activities. Signiticantly, the ability of both LasA and LasD to lyse heat-killed staphylococci is enhanced in the presence of pseudolysin, which itself exhibits no staphylolytic activity. The enhancement of staphylolytic activity by pseudolysin supports the previous hypothesis regarding the function of LasA (Peters and Galloway, 1990). We have suggested that the elastolytic activity of pseudolysin is enhanced by LasA because LasA nicks the elastin substrate thereby exposing elastin to further degradation in the presence of other general proteases, such as pseudolysin. By the same reasoning, when heat-killed S. aureus cdl wall is being digested by LasA, pseudolysin enhances the staphylolytic activity by acting as a general protease. In the absence of LasA, fewer substrate sites for pseudolysin are exposed to the enzyme, and pseudolysin is unable to destroy the cell wall.
2. Peptide substrate specificity of LasA and LasD
The use of the staphylolytic assay, while convenient, does not provide a well- defined system for the analysis of substrate specificity or reaction kinetics. It has previously been demonstrated that staphylolytic proteases are capable of cleaving pentaglycine as a substrate (Lache, Heam etal., 1969). Also, Kessler etal. have recently demonstrated that the LasA protease cleaves synthetic pentaglycine
(Kessler, Safrin etal., 1993). In order to extend our studies using a more defined assay system, we have analyzed both LasA and LasD for the ability to hydrolyze synthetic peptide substrates including pentaglycine. Hydrolysis of several peptide
123 substrates was monitored using high pressure liquid chromatography (HPLC) analysis as described in the Materials and Methods section. Figures 3.7 through 3.10 show HPLC profiles representing hydrolysis of pentaglycine by LasA and LasD. In
Figure 3.7,20 p.1 of 1 mg/ml glycine standards (di-, tri-, and pentaglycine) were analyzed using an HPLC reverse phase column (Bio-Sil 0DS-5S with guard column, Bio-Rad, Hercules, CA) at a flow rate of 1 ml/min. To determine whether pentaglycine was a substrate for LasA, 15 pi of Img/ml pentaglycine (in 20 mM phosphate buffer, pH 8.0) was incubated with 0.5 pg of LasA for 1 hr at room temperature, the sample was run on the reverse phase column at a flow rate of 0.5 ml/min and O.D .210 was measured to detect eluted peptides (Figure 3.8). Two peaks corresponding to diglycine and triglycine were detected at similar times (4.37 min and 4.66 min, respectively) as the diglycine and triglycine (4.44 min and 4.80 min, respectively) in the glycine standard profile (Figure 3.7). To verify that each peak was indeed representing either diglycine or triglycine, diglycine (Figure 3.9A) or triglycine (Figure 3.9B) was added following digestion of pentaglycine with LasA for
1 hr at room temperature. In Figure 3.9A, the O.D .210 value of the first peak corresponding to diglycine increased several-fold compared to that of Figure 3.8 confirming that the first peak is diglycine. Likewise, in Figure 3.9B, when triglycine was added, the O.D .210 value of the corresponding peak increased in the same pattern suggesting that the second peak is indeed triglycine. Therefore, LasA cuts pentaglycine producing di- and tri-glycine. In Figure 3.10, the same assays were repeated for LasD under similar conditions. However, pentaglycine was not digested by LasD when both of them were resuspended in 20 mM phosphate buffer
124 Figure 3.7: HPLC analysis of glycine standard.
The glycine standard was composed of di-, tri-, and pentaglycine (1 mg/ml each in
20 mM phosphate buffer, pH 8.0). An HPLC reverse phase column (Bio-Sil ODS-
58 with a guard column, Bio-Rad, Hercules, CA) was used, to analyze 20 pi of the glycine standard at a flow rate of 0.5 ml/min. Under this condition, di-, tri-, pentaglycine are eluted at 4.44,4.80, and 5.60 min, respectively.
125 O.D. at 210 nm oooro oooro 0000*0
oo Tf
-S I
LL'l
oooro oooro OOOO'O O.D. at 210 nm
Figure 3.7
126 Figure 3.8: HPLC analysis of pentaglycine digestion by LasA.
Both pentaglycine (1 mg/ml) and LasA (0.1 mg/ml) were resuspended in 20 mM phosphate buffer (pH 8.0). For digestion assay, 15 pl of pentaglycine was incubated with 5 pi of LasA at room temperature for 1 hr. The sample was analyzed using an HPLC reverse phase column (Bio-Sil ODS-5S with a guard column, Bio-Rad, Hercules, CA) at a flow rate of 0.5 ml/min. Because the column was designed to analyze small peptides only, large proteins such as LasA were retained in the guard column to prevent clogging of the ODS-5S column. As a result, the elution profile contained only peptides.
127 O.D. at 210 nm oooro oooro oooro oooro OOOO'O
•S S I
oooro oooro oooro oooro OOOO'O O.D. at 210 nm
Figure 3.8
128 Figure 3.9: Verification of diglycine and triglycine by spiking experiments.
A. For the diglycine spiking experiment, 15 pi of pentaglycine (1 mg/ml) was
incubated with 5 pi of LasA (0.1 mg/ml) at room temperature for I hr. Then, 10
pi of diglycine (1 mg/ml) was added to the sample immediately before analysis.
An HPLC reverse phase column (Bio-Sil 0DS-5S with a guard column, Bio-
Rad, Hercules, CA) was used and 20 pi of the sample was loaded at a flow rate
of 0.5 ml/min.
B. For the triglycine spiking assay, 15 pi of pentaglycine (1 mg/ml) was incubated
with 5 pi of LasA (0.1 mg/ml) at room temperature for 1 hr. Then, 10 pi of
triglycine (1 mg/ml) was added to the sample immediately before analysis. The
rurming condition was same as (A).
129 4.34
Gly2
4.67
Gly3 Q d
c> -
0.00 2.00 4.00 6.00 7.06 T im e (m in ) B
4.69
I
Q d Gly3
4.38
Gly2
0.00 2.00 4.00 6.00 7.12 T im e (m in)
Figure 3.9 130 Figure 3.10: HPLC analysis of pentaglycine digestion by LasD.
A. Both pentaglycine (1 mg/ml) and LasD (0.1 mg/ml) were resuspended in 20
mM phosphate buffer (pH 8.0). For the digestion assay, 15 |il pentaglycine was
incubated with 5 pi LasD at room temperature for 1 hr. The sample was
analyzed using a HPLC reverse phase colunm (Bio-Sil 0DS-5S with a guard
column, Bio-Rad, Hercules, CA) at a flow rate of 0.5 ml/min.
B. Both pentaglycine (1 mg/ml) and LasD (0.1 mg/ml) were resuspended in 25 mM
diethanolamine buffer (pH 9.5). For the digestion assay, 15 pi pentaglycine was
incubated with 5 pi LasD at room temperature for 1 hr. The sample was
analyzed by HPLC reverse phase column (Bio-Sil 0DS-5S with a guard column,
Bio-Rad, Hercules, CA) under the same condition as (A).
131 5.41
_ § E I
A § o
Tim e (min) B 4.75
4.39 5.43 I S
A § §
•3 .0 0 2.00 Time (min)
Figure 3.10 132 at pH 8.0 and incubated at room temperature for 1 hr before HPLC analysis (Figure
3. IDA). To determine whether different pH conditions affect activity, both
pentaglycine and LasD were resuspended in 25 mM diethanolamine buffer at pH 9.5
for the same reaction. In this case, pentaglycine was digested by LasD although the
activity was somewhat lower than that of LasA (Figure 3.10B). Pseudolysin was
also analyzed with pentaglycine, but no activity was observed under any pH
condition (data not shown).
Similar experiments were performed with other peptides and analyzed by the
same HPLC method. For example, when 7.5 pg of bovine insulin P-chain (0.5
mg/ml) was reacted with 0.5 pg of either pseudolysin (0.5 mg/ml), LasA (0.1
mg/ml), or LasD (0.1 mg/ml), all three proteases cut the P-chain but at different places (Figure 3.11). As summarized in Table 3.4, both LasA and LasD exhibit similar peptide specificities. Based upon tfiis analysis, both LasA and LasD proteases appear to cleave peptides at internal diglycine sites and require a synthetic peptide substrate of at least four residues in length, as the triglycyl peptide is not hydrolyzed.
3. Digestion studies with P-casein
Even though peptide analysis is one of the methods used to identify substrates for enzymes, synthetic peptides often fail to form the secondary structures required for some enzymes. To overcome this limitation, various other proteins were used to find possible substrates for both LasA and LasD. Among those examined, P-casein
133 Figure 3.11: HPLC analysis of insulin p-chain digestion by various P. aerugmosa proteases.
A. Insulin P-chain standard.
B. Insulin P-chain digested by pseudolysin.
C. Insulin P-chain digested by LasA.
D. Insulin p-chain digested by LasD.
134 I
a Q d d
0.00 T im e (min)
B
§ §
2 5 .0 0 T im e (min)
To be continued. Figure 3.11 135 Figure 3.11 (cont.)
C
m m c
i §
s-
O.CO 2 0 .0 0 4 0 .0 0 T im e (m in) D
i
i
ta § g
S
0.00 2 0 .0 0 4 0 .0 0 T im e (m in)
Figure 3.11 136 Substrate LasD* LasA* (% Activity) (%: Activity) Insulin P-chain 56 57 Triglycine 0 0 Tetraglycine 40 49 Pentaglycine 62 97 Hexaglycine 70 93 Trp-Ala-Gly-Gly-Asp-Ala-Ser-Gly-Glu 55 68 Gly-Phe-Asp-Leu-Asn-Gly-Gly-Gly-Val-Gly 64 70 Tyr-Ala-Gly-Ala-Val-Val-Asn-Asp-Leu 0 0 Gly-Pro-Gly-Gly 0 0 Lys-Lys-Gly-Glu 0 0 P-casein 100" 100" Heat-killed Staphylococcus aureus 53" 69"
‘ Activity is reported, as the percentage ratio of the product peak area in relation to the total peak area (products and substrates) as determined by HPLC analysis.
^ The percentage activity was determined from the band density on SDS-PAGE gels using laser densitometer scanning analysis.
^ The percentage activity was determined from the change in O.D.sgs, [(O.D.control - O.D.sample)/O.D.control x 100].
Table 3.4: Substrate specificity of LasD and LasA.
137 was identified as a substrate for both proteases. Specifically, both LasA and LasD cut p-casein twice producing two internal fragments of 25 kDa and 18 kDa (Figures
3.12 and 3.13, respectively). Surprisingly, N-terminal sequencing of the fragments revealed that both proteases cut P-casein at same places. The N-terminal sequence of the 25 kDa fragment was De-Gly-Lys-Phe-Gln-Ser suggesting that the P-casein cleavage site is Asn-Lys-Lys->Ile-Gly-Lys-Phe-Gln-Ser. The N-terminal sequence of the 18 kDa fragment identified that the second cleavage sited is Ala-Leu-
Ala^Arg-Glu-Leu-Glu-Glu.
E. Inhibitors of LasA and LasD
1. Inhibitor study using heat-killed S. aureus ceUs as a substrate
To identify inhibitors of LasA and LasD, various protease inhibitors were used. Characteristics of various inhibitors used in this study are summarized in
Table 2.5. According to Figures 3.14 and 3.15, the staphylolytic activity of LasD as well as LasA is clearly inhibited in the presence of o-phenanthroline and dithiothreitol (DTT). In both figures, staphylolytic activity is represented by a decrease in O.D. at 595 nm due to the lysis of S. aureus ceüs by either LasA or
LasD. Therefore, inhibition of the activity was demonstrated by constant O.D .595 values throughout the incubation period in the presence of an inhibitor such as o- phenanthroline. o-Phenanthroline is a small (M.W. 198.2) metaUoprotease inhibitor which inhibits many metalloproteases and metal-activated proteases. Both LasA and LasD showed no activity when 6.7 mM ophenanthroline was introduced before
138 Figure 3.12: Digestion of P-casein by LasA.
Both LasA (0.1 mg/ml) and P-casein (2 mg/ml) were resuspended in 20 mM phosphate buffer (pH 8.0). Following addition of 10 pi P-casein and 5 pl of LasA, the reaction mixture was incubated at 37°C for various times before adding SDS- sample buffer and boiling for 3 min. Each sample was then analyzed by 12% SDS-
PAGE. Lane 1, protein molecular weight marker; lane 2, p-casein control without incubation; lane 3, LasA control without incubation; lane 4, P-casein + LasA without incubation; lane 5, P-casein + LasA, 15 min incubation; lane 6, P-casein +
LasA, 30 min incubation; lane 7, p-casein + LasA, 1 hr incubation; lane 8, p-casein
+ LasA, 2 hr incubation; and lane 9, P-casein control, 2 hr incubation.
139 43 kDa-
•P-casein 29 kDa.
■LasA 18 kDa.
1 2 3 4 5 6 7 8 9
Figure 3.12
140 Figure 3.13: Digestion of P-casein by LasD.
The P-casein substrate in 25 tuM diethanolamine buffer (pH 9.5) was digested with I
\i% LasD at room temperature for varying times. The reaction was stopped by addition of SDS-sampIe buffer and boiling for 3 min. Each sample was then analyzed by 12 % SDS-PAGE. Lane 1, 5 pg P-casein; lane 2,1 pg LasD; lane 3, P- casein + LasD, no incubation; lane 4, p-casein + LasD, 15 min incubation; lane 5, P- casein + LasD, 30 min incubation; lane 6, p-casein -t- LasD, 1 hr incubation; lane 7,
P-casein -i- LasD, 2 hr incubation; and lane 8, protein molecular weight marker.
141 ■43 kDa
P-casein- ■29 kDa
LasD" ■18 kDa ■14 kDa
8
Figure 3.13
142 Figure 3.14: Effect of inhibitors on staphylolytic activity of LasA.
The staphylolytic assay described in the Materials and Methods section was carried out at pH 8.0 in 20 mM phosphate buffer with 2.5 pg LasA in the presence of various inhibitors at selected concentrations. Heat-killed S. aureus ceMs with neither
LasA nor any inhibitor were used as a control. “No Inhibitor” indicates that heat- killed S. aureus ceWs were incubated with LasA in the absence of any inhibitor.
143 1.4
1.2 ♦ Control ♦ - - No Inhibitor 1 % -A— 10 mM EDTA (Q ■a— 2mMPMSF Q 0.8 d lOOug/mlTLCK 0.6 -e— 10 mM DTT # 0.4 01 2 3 Time (Hr)
B
1.4
1.2 ■Control
A - - No Inhibitor 1 % 0 - —10 uM Phosphoramidon % Q 0.8 -a— 1 mM DPP d ■A— 6.7 mM o-Phenanthroline 0.6
0.4 0 1 2 3 Time (Hr)
Figure 3.14 144 Figure 3.15: EfiFect of inhibitors on staphylolytic activity of LasD.
The staphylolytic assay described in the Materials and Methods stcûon was carried out at pH 9.5 in 25 mM diethanolamine buffer with 2.5 pg LasD in the presence of various inhibitors at selected concentrations. Heat-killed S. aureus céls with neither
LasD nor any inhibitor were used as a control. “No Inhibitor” indicates that heat- killed S. aureus céü-S were incubated with LasD in the absence of any inhibitor.
145 1.4 1.2 ♦ Control ♦ * *No Inhibitor -A— 20 mM EDTA S 0.8 - -H— 2 mM PMSF ^ 0.6 - lOOug/mlTLCK 0.4 - -e— lOmMDTT
0.2 --
0 I 2 3 4 Time (H t)
B
1.4 •Control 1.2 # - -No Inhibitor g (3 - O - - 10 uM Phosphoramidon A 0.8 -H— O.SmMDFP o 0.6 -A — 6.7 mM o-PhenanthroIine
0.4 0 1 2 3 Time (H t)
Figure 3.15
146 adding the substrate (heat-killed S. aureus cells). Interestingly, the presence of 10 mM ethylenediaminetetraacetate (EDTA) did not inhibit staphylolytic activity as previously reported for LasA by Kessler et al. (Kessler, Safrin e t ed., 1993). This result was contradictory to that of ophenanthroline since both chemicals are metaUoprotease inhibitors. However, this problem may be explained by the difference between these two metal chelators: o-phenanthroline has a much higher affinity for zinc than for calcium, but EDTA has higher affinity for calcium than for zinc (Salvesen and Nagase, 1990). Therefore, even though both o-phenanthroline and EDTA chelate zinc ions, o-phenanthroline may be a more efficient inhibitor because of its high zinc affinity. Dithiothreitol (DTT) was used as a control because it is a strong denaturing agent. In the presence of 10 mM DTT, both enzymes were expected to be denatured and therefore, neither proteases demonstrated any staphylolytic activity. Of aU the serine protease inhibitors tested, none appeared to affect the staphylolytic activities of either LasA or LasD.
2. Inhibitor analysis using 3-casein as a substrate
In addition to heat-kiUed S. aureus, P-casein was also used to identify inhibitors for both LasA and LasD. Surprisingly, hydrolysis of P-casein by LasA or
LasD is inhibited in the presence of serine protease inhibitors such as phenylmethanesulphonyl fluoride (PMSF), diisopropyUluorophosphate (DPP), and tosyl lysyl chloromethyl ketone (TLCK), whereas metaUoprotease inhibitors such as ophenanthroline and EDTA do not appear to inhibit this reaction (Figure 3.16).
This result is obviously contradictory to the inhibition assay using heat-kiUed S.
147 Figure 3.16; Digestion of (3-casein with either LasD or LasA in the presence of inhibitors.
A. Beta-casein (5 pg) in 25 mM diethanolamine buffer, pH 9.5, was digested at 37°C
with 1 pg LasD for 2 hr in the presence of the following inhibitors. Lane 1,
control, no incubation; lane 2,2 hr incubation, no inhibitor; lane 3, 10 mM o-
phenanthroline; lane 4,10 mM EDTA; lane 5,20 pM phosphoramidon; lane 6,
2 mM PMSF; lane 7,0.5 mM DFP; lane 8,100 pg/ml TLCK; lane 9, 10 mM
DTT; and lane 10, protein molecular weight marker. The reaction mixture was
inactivated by boiling and analyzed using 12% SDS-PAGE.
B. Beta-casein (5 pg) in 25 mM diethanolamine buffer, pH 9.5, was digested at 37°C
with 0.5 pg LasA for 2 hr in the presence of inhibitors. The lanes contained the
same components as in (A). The reaction mixture was inactivated by boiling and
analyzed using 12% SDS-PAGE.
148 ■43 kDa P-casein- ■29 kDa LasD' ■18 kDa ■14 kDa
1 2 3 8 10
B
-43 kDa p-casein- -29 kDa
LasA' 18 kDa 14 kDa
1 2 5 6 8 9 10
Figure 3.16
149 aureus céüs as the substrate. Three possible explanations may be considered to account for the discrepancies. First, both LasA and LasD have two distinct active sites: one for metaUoprotease and one for serine protease. However, it is very unlikely that either protease has two active sites because the sizes of LasA and LasD
(22 kDa and 23 kDa, respectively) are not large enough to contain two active sites.
The second possibiUty is that both LasA and LasD preparations are contaminated by a common protease. In that case, one activity is due to either LasA or LasD and the other activity is due to the contaminant. However, as mentioned earlier, it is very unlikely that either of the proteases is contaminated because of evidence based on protein sequencing, immunoblotting, densitometer assay, and chromatographic profiles. The third and most likely possibility is that both LasA and LasD are serine proteases which require metal ions (primarily zinc ions) for their function.
Therefore, it is possible that neither LasA nor LasD can be classified as a metaUoprotease or a serine protease similar to Lysobacterenzymogenes^-\yûc protease (Whitaker, 1965; Whitaker and Roy, 1967; Peters, 1990).
In addition to serine protease inhibitors and metaUoprotease inhibitors, the role of cysteine protease inhibitors were also examined. However, use of the cysteine protease inhibitor cystatin faUed to inhibit both P-casein digestion and staphylolytic activity, indicating that neither LasA nor LasD appear to belong to the cysteine protease class of proteases (data not shown).
150 F. Discussion
The purpose of the purification strategies applied in this study was to optimize yield of the purified enzyme in the shortest time. It took 9 to II days to purify either pseudolysin, LasA, or LasD (Table 3.5). It was desirable to reduce the number steps especially dialysis and ammonium sulfate concentration steps because a portion of sample was lost each step. However, all of the dialysis steps were necessary because of the changes in buffer conditions. Likewise, ammonium sulfate salt cuts were also necessary. For the first two salt cuts, different methods of concentration were not practical because of mucous material in the sample. If there was no mucous material in the sample, a large scale concentrator could be used.
However, the mucous material was likely to block the filter in the concentrator. The third salt cut in LasA and LasD purification could be avoided by using a 50 ml stirred cell (Amicon, Beverly, MA) with a YM-IG membrane (10 kDa molecular cut off, Amicon, Beverly, MA).
In the production of the antiserum against LasD, the serum was used directly without purifying IgG fractions because the serum itself exhibited no cross-reactivity.
Using this serum for immunoblotting, the existence of a 30 kDa protein was revealed. It is not likely that the 30 kDa protein cross-reacted with the serum due to contamination of the LasD sample used in the production of the serum; a band corresponding to LasD was cut directly from an SDS-PAGE gel making the sample free from contamination. Therefore, the most probable explanation is that the 30 kDa protein is a precursor of LasD which is subsequently processed to active LasD.
The N-terminal sequence of both LasA and the 30 kDa protein was determined and
151 Time Steps Pseudolysin LasA LasD Day 1 5 ml culture Day 2 3 L culture Day 3 1"* salt cut Day 4 2“*^ salt cut Day 5 day 1 of 1® dialysis Day 6 day 2 of 1“ dialysis Day 7 DEAE-Sephacel DEAE-Sephacel column column and 3”* salt cut Day 8 elution and 2"“ dialysis 2"** dialysis Day 9 concentration and CM-Cellulose column analysis and 3"* dialysis Day 10 concentration, Mono-S Mono-S column and column, elution, and 4* concentration dialysis Day 11 concentration and analysis analysis
Table 3.5: Purification time schedule for pseudolysin, LasA, and LasD.
152 both sequences showed no homology with any known protein suggesting that both are undescribed proteins.
Substrate specificity of both LasA and LasD was examined using various substrates including heat-killed S. aureus ceUs, P-casein, bovine insulin P-chain, and synthetic peptides. Both proteases showed similar activity in many cases. However, heat-killed S. aureus cells were sensitive to LasD only at alkaline condition (pH 9.5).
In contrast, the same substrate was digested by LasA under a wide range of pH conditions. For bovine insulin P-chain, even though both proteases cut the substrate, the HPLC profile produced different peaks suggesting that LasA and
LasD cut the P-chain at different places.
Inhibitor studies were also conducted using various substrates and inhibitors.
The results were difficult to interpret since both proteases were inhibited by different inhibitors when different substrates were used. When heat-killed S. aureus cells were used as the substrate, both proteases were apparently inhibited by the metaUoprotease inhibitor ophenanthroline. However, when P-casein was used as a substrate, serine protease inhibitors such as PMSF and TLCK were able to inhibit the enzymatic reactions of both proteases. These results suggest three possible conclusions. First, both LasA and LasD act as both metaUoproteases and serine proteases. Second, both are serine proteases which require structural zinc molecules, and finaUy, both are contaminated by a common protease which is either a serine protease or a metaUoprotease. Among these possibiUties, it is most likely that the second assumption (serine protease with structural zinc molecules) is correct based
153 on several lines of evidence which will be discussed later. However, there is not enough evidence to conclude whether both LasA and LasD can be classified as serine proteases.
154 CHAPTER 4
KINETIC ANALYSIS OF LasA AND LasD
A. Introduction
The kinetic analysis of LasA and LasD is based on the assumption that the enzyme kinetics of both LasA and LasD follow Michaelis-Menten kinetics (Fersht,
1985). In other words, when an enzyme is combined with a large amount of substrate, there is an initial pre-steady state followed by an eventual steady state
(Figure 4.1). During the steady state, the rates of enzymatic reactions ÇVmix, Km, and kcat) can be calculated. However, it is also important to look at the initial pre-steady state of the enzymatic reaction. In this case, an enzymatic reaction can be divided into three stages. In Figure 4.1, Region 1 represents the first stage of the reaction where the concentration of substrate [S] is very small relative to the concentration of the enzyme. At this stage, as [S] decreases, the velocity decreases proportionally and is referred to as “first-order kinetics”. On the other hand. Region 3 represents the third stage where very high concentration of substrate is present relative to that of the enzyme. At this stage, [S] decreases with time but the velocity remains constant; this is called a region of “zero-order kinetics”. The second region. Region 2, represents the region of mixed first- and zero-order kinetics. In this study, as an
155 max
Region 3: zero-order kinetics (steady state)
Region 2: mixed kinetics max
TRegion 1: first-order kinetics (pre-steady state)
Substrate concentration [S]
Figure 4.1: Plot of initial velocity versus substrate concentration (modified from
Segel, 1968).
156 indication of the pre-steady state, the first order constant k was calculated for both enzymes. This constant represents the initial rate constant of the enzymatic reaction
(Cleland, 1990).
There are several possible graphical representations of enzyme kinetics.
Among those, the Lineweaver-Burk plot and the Eadie-Hofttee plot are the most common (Fersht, 1985). The Lineweaver-Burk plot is a double reciprocal plot (1/v vs. 1/[S]) and the Eadie-Hofstee plot is a graphical representation of v vs. v/[S].
Both plots have their own advantages and disadvantages. For example, the
Lineweaver-Burk plot has a disadvantage compared to the Eadie-Hofstee plot by compressing the data points at higher concentration of substrates and emphasizing the low concentration region since it is a reciprocal plot. However, at a given [S], it is easier to read the velocity (v) in the Lineweaver-Burk plot than in the Eadie-
Hofstee plot. In this study, the Lineweaver-Burk plot was used for all graphical representations including those of inhibitor studies.
In addition to the rates of enzymatic reactions, inhibition kinetics were also observed in this study. As well as being inactivated by heat or extreme pH, enzymes may also be inactivated by inhibitors. There are four main types of inhibition: competitive, uncompetitive, noncompetitive, and mixed inhibition (Fersht, 1985). If an inhibitor binds to the active site of the enzyme and prevents the substrate from binding, the inhibitor is a competitive inhibitor. In this case, the inhibition affects
Km, but not Vmax. On the other hand, uncompetitive inhibition is observed when an inhibitor binds only to the enzyme-substrate complex, thereby altering both Km and
Vmax. In other cases, an inhibitor that binds to both free enzyme and enzyme-
157 substrate complex affects only Vmax resulting in noncompetitive inhibition. More common, the Km of the substrate is also affected by a noncompetitive inhibitor showing a mixed inhibition. For this study, a known inhibitor, o-phenanthroline, was used in both cases and the Lineweaver-Burk plot was used to determine the mode of inhibition.
B. Kinetic analysis of LasA
1. Acétylation of pentaglycine
One of the ways to characterize an enzyme is to study its kinetics, and a key component of kinetic studies is a good substrate. For LasA, the substrate of choice was pentaglycine due to its rapid hydrolysis by LasA. However, native pentaglycine could not be used because the assay is based on the use of a substrate in conjunction with trinitrobenzenesulfonic acid (TNBSA). TNBSA reacts with an exposed primary amine to produce an orange-yellow color that can be quantitated by measuring O.D. at 450 nm (Habeeb, 1966; Bubnis and Ofher, 1992). Therefore, when native pentaglycine is used, TNBSA will react with it even when there is no hydrolysis by an enzyme because of the exposed primary amine group at its N- terminus. To prevent this high background, native pentaglycine was modified by acétylation. To acetylate the amino terminus of pentaglycine, the substrate was first treated with triethylamine to free up its amino group for acétylation. Then, acetyl anhydride was used to acetylate the amino group. Finally, the modified
158 pentaglycine was dried in a dessicator at room temperature and its integrity was checked by HPLC.
In Figure 4.2, the result of HPLC analysis of the modified pentaglycine using a 150 mm x 4 mm Bio-Sil ODS-5S reverse phase column (Bio-Rad, Hercules, CA) is shown. According to the figure, modified pentaglycine was eluted later (Figure
4.2B, 3.14 min) than the unmodified pentaglycine control (Figure 4.2A, 2.19 min) representing the decrease in hydrophobicity due to the loss of the hydrophilic amino groups.
2. Mathematical analysis of the enzymatic reaction
Following acétylation, varying amounts (10 to 98 pg) of the modified pentaglycine substrate was incubated with 0.2 pg LasA for varying times to determine the appropriate linear range for increasing enzymatic activity. Data from the 30 minute incubation time was chosen for subsequent calculation of kinetic parameters because the reaction increases linearly between 0 to 1 hr incubation time.
Therefore, it is assumed that, at 30 min incubation, the enzymatic activity represents the linear portion (0-1 hr) of the reaction. Figure 4.3 A illustrates that LasA- catalyzed cleavage of acetylated pentaglycine followed normal Michaelis-Menten kinetics, with the measured reaction velocity (represented by O . D . 4 5 0 ) being dependent on substrate concentration. Vnax is defined as the maximal number of moles of product formed per minute and Kn, is the substrate concentration at which the reaction rate is half-maximal (Fersht, 1985; Creighton, 1993). In this study, was initially measured as a maximal O . D . value at 450 nm ( O . D . 4 5 0 ) because the
159 Figure 4.2: HPLC elution profiles of unmodified (A) and acetylated (B) pentaglycines.
Each peptide was resuspended in 20 mM phosphate buffer (pH 8.0) at a concentration of 1 mg/ml. Then, 20 pJ of each sample was loaded onto a Bio-Sil
0DS-5S reverse phase column (Bio-Rad, Hercules, CA) in separate experiments at a flow rate of 1 ml/min. O.D .210 was used to detect the elution of the peptides.
160 2 .1 9
Unmodified Pentaglycine
1 2 § §
0.00 2 .0 0 Tim e (min)
B
3.14
Acetylated Pentaglycine - i
* a A § o
0.00 .5 .0 0 Tim e (min)
Figure 4.2 161 Figure 4.3: Kinetic analysis of acetylated pentaglycine digestion by LasA.
To determine enzyme kinetics of LasA, acetylated pentaglycine was resuspended in
20 mM phosphate buffer (pH 8.0) at a concentration of 1 mg/ml. Varying amounts of the substrate (1 to 98 pg) were incubated with 0.2 pg LasA resuspended in the same buffer. Each reaction mixture was adjusted to a total reaction volume of 100 pi by addition of phosphate buffer. After 30 min incubation at 37°C, 50 pi trinitrobenzenesulfonic acid (TNBSA) working solution (10 pi TNBSA in 1.49 ml 50 mM boric acid, pH 8.5) was added to each reaction mixture; after 20 min at room temperature, the O.D. was measured at 450 nm. The assay was performed in duplicate and average values were obtained for each data point (A). A Lineweaver-
Burk plot (B) was derived from data in (A).
162 2.5 ■
o « .,5 - ci d
0.5 -
0 20 40 60 80 100 Acetylated Pentaglycine (ng)
B
1.2
î o 0-8
% 0.6 q é 0.4
0.2
- 0.2 - 0.1 0 0.1 0.2 0.3 0.4 0.5
1/(Acetylated Pentaglycine) (l/|ig )
Figure 4.3
163 amount of product formed is directly proportional to O.D. 4S0. Then the O.D.450 was converted to jiM using standard O.D.450 values derived from various concentrations of pentaglycine solutions reacted with TNBSA for 20 min. As a result, O.D .450 value of 1 represents 606.1 pM of acetylated pentaglycine which is completely digested by
either LasA or LasD. How to calculate the standard O . D . 4 5 0 for pentaglycine is summarized in Table 4.1.
To calculate Km and Vmax, Figure 4.3A was modified to a double-reciprocal
Lineweaver-Burk plot (Figure 4.3B). In this plot, the x axis represents 1/(acetylated pentaglycine) and the y axis represents 1/O.D.450. A straight line derived from data points was extended until y = 0. The equation for the strait line was y=0.30228 +
1.6999Xand, when y = 0, x was -0.1778. According to Michaelis-Menten kinetics, thisX value (-0.1778) is equal to -1/Km. Therefore:
1/Km = 0.1778 = 1/ [acetylated pentaglycine]
Km = [acetylated pentaglycine] at ^ Vmax = 1/0.1778 = 5.624 pg
This Km was then converted to molarity by considering the volume (100 pi) and molecular weight (345.3) of the acetylated pentaglycine:
1 pg/100 pi = 0.01 g/L = 0.01 g/L X lM/(345.3 g/L) = 2.896 x 10 " M
Therefore, Km = 5.624 pg/lOOpl = 5.624 X 2.896 x 10 " M = 1.629 x 10"* M
To calculate Vmax, the y value when x = 0 was chosen because, in the Michaelis-
Menten equation, this value represents 1/Vmax:
when X = 0, y= 0.30228 = 1/Vmax, Vmax = 3.308 O . D . 4 5 0
1 O . D . 4 5 0 = 606.1 pM acetylated pentaglycine 3.308 O.D.450 = 3.308 x 606.1 = 2005 pM 164 Pentaglycine (jiM) Average O.D .4 5 0 O.D.4 5 0 - OX>.4 so per pM Background penta^ydne (x IQ"*)
0 0.257 0 0
33 0.315 0.058 1.748
6 6 0.354 0.097 1.464
99 0.407 0.150 1.513
132 0.475 0.218 1.652
165 0.544 0.287 1.739
330 0.858 0.601 1.821
1650 2 . 6 6 8 2.411 1.461
Standard O.D .450 per pM pentaglycine* (1.623 X10*) ±(0.116 X10*)
‘ Standard O.D. 4S0 is the average of all O . D . 4 5 0 per |iM pentaglycine values except the highest and the lowest values.
Table 4.1: Calculation of standard O.D .450 for pentaglycine.
165 However, because the reaction time was 30 min, the actual Vma% was:
Vmax = 2005 |iM/30 min = 66.83 liM/min
The turnover number (kc«) of an enzyme is defined as the maximum moles of substrate utilized per mole of enzyme per minute under optimum assay conditions
(Fersht, 1985; Creghton, 1993) . When [ S ] » K m , all LasA molecules ( 0.2 pg in 100 pi reaction mix) are involved in the reaction and the reaction velocity v is approximately same as Vmax- Therefore:
0.2 pg LasA/100 pi = 2 mg/L = (2 x 10'" g)/[l M/(21000 g/M)] = 9.524 X 10 * M
Vmax = 66.83 pM/min = 66.83 x 10^ M/min
kcat = Vmax/[LasA] = 66.83 x 10^ M acetylated pentaglycine/min 9.524 X 10 * M LasA = 701.7 M acetylated pentaglycine per IM LasA per min
When the substrate concentration [ S ] is very small ( [ S ] « K m ) , the enzymatic reaction follows the first-order kinetics. In this case, the first order rate constant k can be approximated as k = Vmax/Km. Therefore:
k = (66.83 X 10^ M/min)/(1.629 x W M) = 4.102 x 10 ' min '
A summary of LasA kinetic data is provided in Table 4.2.
3. Inhibition kinetics of LasA
To characterize what type of inhibition is occurring for LasA, different concentrations (0.2, 0.5, and 1 mM) of the previously described inhibitor o phenanthroline was used with increasing amounts of acetylated pentaglycine. In
166 this assay, the inhibitor was preincubated with LasA for 3 min before adding acetylated pentaglycine. Following addition of acetylated pentaglycine, the reaction was continued for 30 min. TNBSA was added and O.D .450 was measured after 20 min. Finally, data was presented as a double reciprocal Lineweaver-Burk plot. As shown in Figure 4.4, when the concentration of ophenanthroline changes, the y intercept (1/Vmax) remains constant indicating that Vmax is constant. However, the value for the x intercept (- 1/ K m ) increases as the concentration of ophenanthroline increases indicating that Km increases with inhibitor concentration. This pattern of inhibition is typical of competitive inhibition; constant Vmax but higher Km.
Therefore, it is concluded that inhibition of LasA by ophenanthroline is competitive inhibition.
C. Kinetic analysis of LasD
1 . Mathematical analysis of the enzymatic reaction
For LasD, the same assay was performed to determine kinetic information for the enzyme using acetylated pentaglycine. The only difference, in this case, was that both LasD and acetylated pentaglycine were resuspended in either 25 mM diethanolamine buffer (pH 9.5) or 20 mM phosphate buffer (pH 8.0) because it has been previously demonstrated that LasD is pH sensitive when certain substrates, such as heat-killedS. aureus cells are used. The results show that, as expected, an alkaline pH condition (pH 9.5) is optimal for LasD activity. However, under a lower pH condition (pH 8.0), LasD still exhibited a considerable degree of activity
167 Figure 4.4: Lineweaver-Burk plot showing competitive inhibition of LasA by o phenanthroline.
To identify the type of inhibition for LasA, 0.5 pg LasA was preincubated with varying amounts of ophenanthroline (50 mM in methanol) for 3 min without the substrate. Different amounts of acetylated pentaglycine (5 to 75 pi of 2 mg/ml acetylated pentaglycine resuspended in 20 mM phosphate buffer, pH 8.0) were added, and the final volume was adjusted to 100 pi by addition of phosphate buffer.
After incubating for 30 min at 37“C, 50 pi TNBSA working solution (10 pi TNBSA in 1.49 ml 50 mM boric acid, pH 8.5) was added; after 20 min at room temperature, the O.D.450 was measured. The assay was performed in duplicate and the average
O.D.450 was used to make a Lineweaver-Burk plot.
168 6
■ 0.2 mM 5 # 0.5 mM ▲ 1 mM I 4 § % 3 Q d 2
1
0 I — \ ^ -I l ' I ' I -0.08 -0.06 -0.04 -0.02 0 0.02 0.04 0.06 0.08 0.1 1/(Acetylated Pentaglycine) (1/pig)
Figure 4.4
169 (Figure 4.5). Kinetic values (Km, Vmax, and kcat) for both pH conditions are provided in Table 4.2.
a) alkaline pH condition (25 mM diethanolamine buffer, pH 9.5)
To calculate Km and Vmax, in Figure 4.5B, a straight line derived from value points was extended until y = 0. The equation for the line was y = 2.5928 + 32.582x and when y = 0, x value was - 0.0796. Since this x value represents - 1/ K m :
1/ K m = 0.0796 = 1/(amount of acetylated pentaglycine)
Km = amount of acetylated pentaglycine at 14 Vmax = 1/0.0796 = 12.56 pg.
This Km was then converted to molarity by considering the volume (100 pi) and MW of acetylated pentaglycine (345.3):
1 pg/100 pi = 2.896 X 10 " M
Km = 12.56 pg/100 pi = 12.56 x 2.896 x 10" M = 3.637 x 10^ M
For Vmax, the y value when x = 0 is chosen because the value represents 1/Vmax according to the Michaelis-Menten equation:
when X = 0, y= 2.5928 = 1/Vmax, Vmax = 0.3857 O.D. 4so
1 O.D .450 = 606.1 pM acetylated pentaglycine 0.3857 O.D .450 = 0.3857 x 606.1 = 233.8 pM
However, because the reaction time was 30 min, the actual Vmax was:
Vmax = 233.8 pM/30 min = 7.793 pM/min
When [ S ] » K m , all LasD molecules are involved in the reaction and, therefore, the reaction velocity is approximately same as Vmax. As a result, kcat could be calculated in the following equations:
170 Figure 4.5: îCinetic analysis of acetylated pentaglycine digestion by LasD.
Acetylated pentaglycine was resuspended in either 20 mM phosphate buffer (pH 8.0) or 25 mM diethanolamine buffer (pH 9.5) at a concentration of 1 mg/ml. Then, varying amounts of the substrate (1 to 98 pg) were incubated with 0.2 pg LasD resuspended in the same buffer. Each reaction mixture was adjusted to a total reaction volume of 100 pi by adding the same phosphate buffer. After 30 min incubation at 37°C, 50 pi trinitrobenzenesulfonic (TNBSA) working solution (10 pi
TNBSA in 1.49 ml 50 mM boric acid, pH 8.5) was added to each reaction mixture; after 20 min at room temperature, the O.D. was measured at 450 nm. The assay was performed in duplicate and average values were obtained for each data point
(A). A Lineweaver-Burk plot (B) was derived from data in (A).
171 0.45 0.4 0.35 0.3 pH9.5 o pHS.O 0.25 (8 0.2 D d 0.15 0.1 0.05 0 0 20 40 60 80 100 Acetylated Pentaglycine (^g)
B
80
pH9.5 pH8.0 60
§ eo 40 Q d
20
Ol------,------^ -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 1/(Acetylated Pentaglycine) (1/pg)
Figure 4.5
172 0.2 pg LasD/100 pi = (2x 10" g/L) x {1 M/(23000 g/L)} = 8.696 x 10 * M
kat = Vmax/[LasD] = 7.793 x 10^ M acetylated pentaglycine/min 8.696 X 10* M LasD = 89.62 M acetylated pentaglycine per 1 M LasD per min (when [ S ] » K m )
The first order constant k could be estimated when [S] is much smaller than Km:
k = V m a x / K m = (7.793 X 10^ M/min)/(3.637 x lO^* M) = 2.143 x 10 * min '
b) lower pH condition (20 mM phosphate buffer, pH 8.0)
Calculation was by the same method as described as above. The equation for the straight line for pH 8.0 in Figure 4.5B was y = 4.5594 + 64.4I4x. When y = 0, x
= 0.0708 and y = 4.5594 when x = 0. Therefore:
Km = 1/0.0708 = 14.13 pg/100 pi = 14.13 x 2.896 x 10 " M = 4.092 x 10"* M
1/Vmax = 4.5594
V m a x = 0.2193 O.D.450 = 0.2193 X 606.1 = 132.9 pM/30 min = 4.431 pM/min
kcat = Vmax/[LasD] = 4.432 x 10~" M acetylated pentaglycine/min 8.696 X 10* M LasA
= 50.97 M acetylated pentaglycine per 1 M LasD per min ( [ S ] » K m )
k = V m a x / K m = (4.431 X 10^/(4.092 x 10"») = 1.083 x 10 * min ( [ S ] « K m )
2. Inhibition kinetics of LasD
To determine the mechanism of inhibition, the same assay as that for LasA was repeated with varying concentrations of ophenanthroline (0.2, 0.5, and 1 mM) except that both LasD and acetylated pentaglycine were resuspended in 25 mM diethanolamine buffer (pH 9.5) instead of 20 mM phosphate buffer (pH 8.0) because the activity of LasD at pH 9.5 was much higher than at pH 8.0 (Figure 4.5A). After
173 measuring O . D . 4 5 0 , a double-reciprocal Lineweaver-Burk plot was drawn and the
result showed that, just like LasA, the inhibition was competitive. According to
Figure 4.6, the Vmax did not change when the concentration of o-phenanthroline was
increased. However, the Km values increased as the concentrations of the inhibitor
were increased from 0.2 mM to 1 mM as is typical for a competitive inhibitor.
Therefore, it appears that LasD as well as LasA are competitively inhibited by o
phenanthroline.
D. Discussion
Initially, the biggest problem in studying the kinetics of LasA and LasD was
the lack of a substrate which could be easily quantitated. Even though several
substrates were known for both LasA and LasD, all of them were not suitable for the
kinetic study. For example, heat-killed S. aureus cells were not a good choice because the substrate was not clearly defined. Another substrate, P-casein could be
used if the rate of the enzymatic reaction was represented by the degradation of P- casein at a given concentration or time. The level of degradation could be calculated by measuring its density on SDS-PAGE gel using a densitometer. However, it is not practical to use SDS-PAGE gels especially at very high levels of substrate concentration. If the substrate concentration was lowered, the whole calculation would be dependent on the accuracy of the reaction mixture in each well, and the accuracy varied from one gel to another. Synthetic peptides, including pentaglycine, could also be used by employing HPLC as a way to measure the enzymatic reaction.
174 Figure 4.6: Lineweaver-Burk plot showing competitive inhibition of LasD by o- phenanthroline.
To identify the type of inhibition for LasD, 0.5 pg LasD was preincubated with varying amounts of o-phenanthroline (50 mM in methanol) for 3 min without the substrate. Different amounts of acetylated pentaglycine (5 to 75 pi of 2 mg/ml acetylated pentaglycine resuspended in either 20 mM phosphate buffer, pH 8.0, or
25 mM diethanolamine buffer, pH 9.5) were added, and the final volume was adjusted to 100 pi by addition of appropriate buffer. After incubating for 30 min at
37“C, 50 pi TNBSA working solution (10 pi TNBSA in 1.49 ml 50 mM boric acid, pH 8.5) was added; after 20 min at room temperature, the O . D . was measured at 450 nm. The assay was performed in duplicate and the average O . D . 4 5 0 was used to make a Lineweaver-Burke plot.
175 4 0
■ 0.2 m M 35 • 0.5 m M
* ■ 1 m M 30 I I 25 13 20 A o. 15
10
5
0 “ T I I “r~ ” I ' I ^ " I -0.08 -0.06 -0.04 -0.02 0 0.02 0.04 0.06 0.08 0.1 1/(Acetylated Pentaglycine) (l/|ig)
Figure 4.6
176 However, calculating the area of each peak on an HPLC profile is inaccurate mainly because the basal line of each peak was ambiguous.
To overcome these problems, acetylated pentaglycine was used as a substrate.
There are three major advantages of using this substrate. First, when the acetylated pentaglycine is hydrolyzed by either LasA or LasD, it is easy to measure the reaction using TNBSA because TNBSA only reacts with newly exposed amino groups resulting in an orange-yellow color. The color change could be quantitated very easily by measuring O . D . at 450 nm. Second, by blocking the existing amino group of the pentaglycine molecule, the reaction has virtually no background because the acetylated amino groups are not available for TNBSA. Third, it is easy to calculate the concentration of both substrates and products because the O.D. value is directly proportional to the concentration of products. In other words, because both LasA and LasD cut the acetylated pentaglycine only once, one molecule of acetylated pentaglycine will give same O . D . 4 5 0 as one molecule of unmodified pentaglycine when it was hydrolyzed by either enzyme because both of the molecules have only one amino group for TNBSA. Therefore, by measuring the O . D . 4 5 0 of the known concentration of unmodified pentaglycine, it was possible to precisely calculate the amount of acetylated pentaglycine which was digested during a reaction.
Determining the Km value of an enzymatic reaction is important for two reasons (Stryer, 1988). First, because Km represents the substrate concentration when half of the active sites are filled, the fraction of active sites occupied by the enzyme at a given substrate concentration, f^, can be calculated from the equation fEs = v/Vmax = [S]/([S] + Km). Second, when ka (dissociation constant of an ES
177 complex to products) is much lower than kz (dissociation constant of an ES complex back to the initial state), Km is equal to kz. In other words, Km can be used to measure the strength of an ES complex; the higher the Km, the lower the strength of the complex. The Km values of most enzymes are between 0.1 pM to 100 mM
(Stryer, 1988) and the Km values of LasA and LasD, 162.9 pM and 363.7 pM (pH
9.5), respectively, are relatively low indicating that the strength of the LasA (or
LasD)-acetylated pentaglycine complex is relatively high (Table 4.2).
The kcat values of a reaction when substrate is saturating ([ S ]» K m ) indicate the maximum turnover rate of the enzyme. Typically, kcat values under these conditions lie between 1 to KT* per second (Stryer, 1988). The higher kat value of
LasA (701.7 M per M LasA per min) relative to that of LasD (89.62 M per M LasD per min at pH 9.5) indicates that the acetylated pentaglycine is better substrate for
LasA than LasD (Table 4.2). Also the kat value of LasD at pH 9.5 was higher than that of pH 8.0 conforming our previous result that LasD has optimal activity under high pH conditions.
For inhibition studies, initially 1 to 10 mM of ophenanthroline was used because that was the effective concentration of the inhibitor. However, at concentrations over 1 mM, most of the enzyme became inactive making the data unusable. Therefore, 0.1 to 1 mM concentrations were used successfully and the data showed that Km increased when higher concentration of inhibitor was used for both LasA and LasD . On the other hand, the Vmax remained constant regardless of the inhibitor concentration. These results indicate that the inhibition was competitive in both cases.
178 Enzyme LasA LasD (pH 8:0) pH 9.5 pHS-0 3.308 0.3857 0.2193 (O.D.4S o)^ Vmax 66.83 7.793 4.431 (uM/min)^ mKm 1.629 X 10-* 3.637 X 10“ 4.092 X 10“ iQat when [S]»K a 707.1 89.62 50.97 (rSI/rEj/miny* kwhen[S]«Km 4.102 X 10-' 2.143 X 10'" 1.083 X 10'" (min')^
‘ Maximum velocity expressed as O.D. 450 .
^ Maximum velocity expressed as molar concentration of acetylated pentaglycine digested per minute.
^ Michaelis constant expressed as molar concentration of acetylated pentaglycine at (6 Vmax.
Turnover number expressed as molar concentration of acetylated pentaglycine digested per mole of LasA (or LasD) per minute.
^ First order rate constant per minute.
Table 4.2: Kinetic analysis of LasA and LasD using acetylated pentaglycine as a substrate.
179 CHAPTERS
PROCESSING OF LasA
A. Introduction
In addition to enzymatic activity, it is also important to study how proteases are activated. Bacterial proteases are commonly produced as a larger, inactive precursor and later processed to a smaller, active fragment. For example, P. aeruginosa pseudolysin is produced as a 60 kDa prepropseudolysin and after three processing events, results in a 33 kDa active fragment (Figure 1.5) (Kessler and
Saftin, 1988). LasA is produced as a 41 kDa precursor and later processed to a 22 kDa protein as predicted by sequence analysis (Darzins, Peters eta l., 1990).
However, it is not known what kind of factor(s) are required to process LasA from its precursor form to an active form. The biggest challenge in the study of LasA processing is that the LasA precursor has not been purified, probably because of its rapid processing to the active fragment. To overcome this problem, the LasA precursor was overexpressed in E. coli and the overexpressed proteins was used to study processing events and activity of the LasA precursor. In addition, the active fragment of LasA was also overexpressed to obtain a high yield of purified protein.
180 To express the LasA precursor in E. coli, the T7 overexpression system was employed (Rosenberg, Lade eta !., 1987). The RNA polymerase of T7 is very selective for specific promoters that are rarely encountered in sequences unrelated to
T7 DNA. E. co li strains BL21(DE3) and BL21(DE3)/pLysS were used as a host strains for transformation. The strain BL21(DE3) is deficient in the Ion protease gene, and also lacks the om pTgene for outer membrane protease that can degrade proteins during purification. In the strain, bacteriophage DE3 is a lamda phage derivative that has the immunity region of phage 21 and carries a DNA fragment containing the la d gene, the lacU VS promoter, the beginning of the lacZ gene, and the gene for T7 RNA polymerase. This fragment is inserted into the Zof gene, and because the in tg en e is inactivated, DE3 needs a helper for either integration into or excision from the chromosome. Once a DE3 lysogen is formed, the only promoter known to direct transcription of the T7 RNA polymerase gene is the lacUVS promoter, which is induced by isopropyl P-o-thiogalactopyranoside (IPTG). In the strain BL21(DE3)/pLysS, a T7 lysozyme gene was introduced on a plasmid (pLysS) to decrease the basal level of T7 RNA polymerase activity. When produced from a cloned gene inside of a cell, T7 lysozyme binds to T7 RNA polymerase and inhibits transcription and as a result, decreases the background.
The vector used for the overexpression was pET3a (Novagen, Madison, WT)
(Studier and Moffatt, 1986; Rosenberg, Lade eta l., 1987; Studier, Rosenberg eta l.,
1990). The plasmid vector was developed for cloning and expressing target genes under the control of T7 promoter. The vector contains a T7 promoter inserted into the Bam HI site of pBR322 and the promoter is derived from the (|) 10 promoter
181 which is one of six strong promoters in T7 DNA that have an identical nucleotide sequence from the positions -17 to +6. In addition to the T7 promoter, pET3a also contains a T7 transcription terminator (T(j)) for effective termination of transcription at the 3’ end of the target gene.
The overall overexpression scheme is illustrated in Figure 5.1. In this figure, the T7 RNA polymerase is produced by E. co//when the /ac promoter is induced by
IPTG. In turn, the T7 RNA polymerase recognizes the T7 promoter ((j)10 promoter) on the pET vector and transcribes the target gene inserted downstream of the promoter.
B. Overexpression of the LasA precursor and active LasA
Overexpression of both active LasA and the LasA precursor was conducted as described in the Materials and Methods. To determine the level of induction, the transformed cells were grown in the presence or absence of IPTG. Figure 5.2A shows the result of the comparison on 10% SDS-PAGE. As indicated in lanes 1 (2 hr with induction) and 2 (2 hr without induction), the level of induction was not very high. After 3 hours, however, the induced cells (lane 3) produced much more LasA precursor than the uninduced cells (lane 4). Figure 5.2B shows the same assay in immunoblotting using the anti-LasA precursor antibody in 1:2000 dilution.
Although the level of expression was very high in BL21(DE3), the basal level of the T7 RNA polymerase activity was also very high in the strain. To reduce the basal activity, another strain BL21(DE3)/pLysS was used. However, the level of overexpression was also relatively low compared to BL21(DE3) due to the low T7
182 IPTG Induction
RNA Polymerase Gene for T7 RNA Pofymerase
T 7R N A Polymerase lac Repressor Promoter
T7 Promoto*
Genome mTTTTl
Host Cell
Figure 5.1: Graphical representation of the T7 overexpression system (modified from the 1996-1997 Novagen catalog, page 27).
183 Figure 5.2: Overexpression of the LasA precursor in BL21(DE3).
A. Overexpression of the LasA precursor analyzed by 10 % SDS-PAGE. Lane 1,2
hr incubation, induced; lane 2,2 hr incubation, uninduced; lane 3, 3 hr
incubation, induced; and lane 4, 3 hr incubation, uninduced.
B. Immunoblotting with anü-LasA precursor polyclonal antibody. The lane
representation is the same as in (A).
184 ^ 4 3 kDa LasA Precursor -29 kDa
•18 kDa
1 2 3
B
LasA Precursor
Figure 5.2
185 RNA polymerase activity. Figure 5.3 A shows the overexpression profile of both
LasA and the LasA precursor. The assay in the Materials and Methods'was followed in this experiment with minor modification. Cells were transformed with either pET3a/(l. 1 kb Ase I-Bam HIlasA) for the LasA precursor or pET3a/(660 bp
Ase I-Bam HI lasA) for active LasA. After transformation, cells were grown for 2 hours in 5 ml LB without induction. Then, 0.5 ml of each culture was inoculated to new tubes as a uninduced control. For the rest of the culture, IPTG (0.5 mM) was added for induction and both induced and uninduced cells were grown 3 hr at 37“C.
The inclusion bodies were obtained from the induced cells by previously described method and supernatants were obtained from the uninduced cells by centrifugation.
After measuring the protein concentration of the total cell lysate by the Bradford method (Bradford, 1976), the same amount of protein (5 pg) was added to each lane for comparison. As indicated in lanes 2 and 4 of Figure 5.3A, the induced cells produced much more LasA precursor (lane 2) and LasA (lane 4) than uninduced cells (lane 1 and 3, respectively). The level of activity appears very low compared to that of Figure 5.2A. However, unlike to Figure 2A in which same volumes of the samples were used in each lane, same amounts of proteins were used in this figure.
Inclusion bodies from the induced cells are diluted several-fold with 0.5 M guanidine-HCl. As a result, the level of overexpression was not represented appropriately. After concentrating and thoroughly washing the inclusion bodies, highly concentrated proteins were obtained. Figure 5.3B shows immunoblot analysis of the proteins in Figure 5.3A.
186 Figure 5.3: Overexpression of the LasA precursor and active LasA in
BL21(DE3)/pLysS.
A. SDS-PAGE analysis (10 %) of the overexpressed proteins. Lane 1, supernatant
from uninduced BL21(DE3)/pLysS transformed with pET3a/(l.l kb Ase I-Bam
HI lasA) for LasA precursor overexpression; lane 2, same construct as lane 1 but
induced with 0.5 mM IPTG; lane 3, supernatant from uninduced
BL21(DE3)/pLysS transformed with pET3a/(660 bp Ase I-Bam HI lasA)\ lane
4, same construct as lane 3 but induced with 0.5 mM IPTG; lane 5, Ipg native
LasA control; and lane 6 , low molecular weight protein standards.
B. Immunoblotting analysis of the overexpressed proteins using and-LasA
polyclonal antibody at 1:2000 dilution. All lanes are same as (A) except that the
lane for protein standards (lane 6 ) is not present.
187 LasA Precursor ■43 kDa •29 kDa ■LasA ■18 kDa
B
LasA _ Precursor
■LasA
Figure 5.3 188 C. Processing of LasA by LasD.
To study LasA processing, P. aeruginosa was fractionated using a previously published protocol with some modification (Cheng, Ingram e ta l., 1971; Hoshino,
1979). After obtaining cytoplasmic fraction, periplasmic fraction, and culture supernatant, the overexpressed LasA was analyzed with each fraction to examine the possible processing event. To identify the LasA precursor in each reaction mixture, the overexpressed LasA precursor was radiolabeled using ^H-labeled amino acid mix (Amersham Life Science Inc., Arlington Heights, DL) during induction with
IPTG. A previously published protocol (Groisman, Pagratis eta l., 1991) was used with some modifications for this procedure.
After labeling the LasA precursor with %, the labeled protein was incubated with different P. aeruginosa fractions to examine the role of each fraction in LasA processing. Each sample was analyzed by 10 % SDS-PAGE followed by treatment with EN^HANCE (DuPont NEN, Boston, MA) before exposing to a X-ray film for 3 days at -70°C as described in the Materials and Methods. As shown in Figure 5.4, most of the labeled LasA precursor was still present after incubating with the cytoplasmic fraction (lane 1). However, the amount of the labeled protein decreases significantly when the protein is incubated with the periplasmic fraction (lane 3) and the protein completely disappeared after incubating with the supernatant fraction
(lane 5). The precursor control (lane 7) stayed intact indicating that it is stable at
37°C. This result may indicate that the processing event starts in the periplasmic space but finishes in the supernatant. However, the result suggests possible problem with the assay. If the precursor protein is indeed processed in the supernatant, a
189 Figure 5.4: Involvement of various P. aeruginosa fractions in processing the LasA precursor.
Each sample was incubated for 1 hr at 37“C before analysis. Lane 1 , cytoplasmic fraction + labeled LasA precursor; lane 2, cytoplasmic fraction control; lane 3, periplasmic fraction + labeled LasA precursor; lane 4, periplasmic fraction control; lane 5, supernatant fraction + labeled LasA precursor; lane 6 , supernatant fraction control; and lane 7, labeled LasA precursor control.
190 LasA Precursor
m
3 4
Figure 5.4
191 band corresponding to the active LasA protein should appear after incubation.
Instead, in lane 5, there was no band corresponding to the active LasA protein and all of the labeled precursor protein disappeared after incubation indicating that the precursor may be digested by protease(s) present in the supernatant fraction rather than processed to an active fragment by a specific factor(s). In this case, the conformation of the overexpressed LasA precursor maybe incorrect since the recombinant protein is refolded after dénaturation by a reducing agent (guanidine-
HCl). As a result of the wrong conformation of the overexpressed protein, different amino acid residues may be exposed to the surface of the protein and hydrolyzed by protease(s) present in the supernatant fraction. Alternatively, sensitivity may be a problem since a low P-emitter (^H) was used as a label for the assay. This issue will be discussed in detail later. However, the investigation was continued assuming that the processing event observed in the supernatant fraction is real. Therefore, the detection assay was changed from a radioactivity-dependent method to an immunoblotting analysis to see whether the ^H labeled protein was the reason why there was no observable active LasA fragment after the processing event.
To find out which factor in the supernatant is responsible for the disappearance of the LasA precursor, the LasA precursor was incubated with various supernatant proteins separately. In this case, 1 pg of either pseudolysin,
LasA, LasD, or the 15 kDa protein resuspended in 20 mM phosphate buffer (pH 8.0) was incubated with 5 pg of the LasA precursor for 1 hr at 37°C. The samples were analyzed by immunoblotting using the anti-LasA polyclonal antiserum at a 1:2000 dilution. Anti-LasA was used because it recognizes both LasA and the LasA
192 precursor. On the other hand, anti-LasA precursor antibody recognizes only the precursor protein. Therefore, by using the anti-LasA antibody, the appearance of the active LasA protein as well as the disappearance of the LasA precursor protein could be monitored more closely. Also, enhanced chemiluminescence (ECL)
(Amersham Life Science Inc., Arlington Heights, EL) was used instead of conventional color reagents for detection to increase sensitivity. Figure 5.5 shows the result of the immunoblot. In lane 1, when pseudolysin was incubated with the precursor, no LasA band was detected. In lane 2, when LasD was incubated with the precursor, a band corresponding to active LasA appeared after 1 hr incubation at
37°C indicating that LasD is involved in the processing of the LasA precursor. Lane
3 shows the result of the reaction between the LasA precursor and LasA. Some of the LasA precursor seems to have been digested by LasA because new bands appeared in the lane compared to lane 1. In lane 4, the LasA precursor was incubated with the 15 kDa protein. Again, there was no indication of processing in this lane. Therefore, the result of this assay suggests that among various proteases purified, LasD is the only protein which causes the processing of the LasA precursor to the active LasA protein. The resulting active fragment of LasA from lane 2 was isolated and the N-terminal sequence was found to be the same as that of native
LasA (Ala - Pro - Pro - Ser - Asn - Leu - Met).
To further examine the role of LasD in LasA processing, the LasA precursor was incubated with LasD in a series of reactions with different reaction times. In this assay, both LasD and the LasA precursor were resuspended in 25 mM diethanolamine buffer (pH 9.5). Then 3 pg of the LasA precursor was mixed with 1
193 Figure 5.5: Immunoblot analysis of various proteases involved in LasA processing.
All samples were incubated for 1 hr at 37“C before analysis. Anti-LasA polyclonal antibody (1:2000 dilution) and the ECL kit were used for detection. After incubating the nitrocellulose membrane with donkey anti-rabbit Ig-horseradish peroxidase conjugate, the membrane was reacted with the ECL reagents for 20 sec before exposing to a X-ray film for 30 sec. Lane 1, pseudolysin + LasA precursor; lane 2,
LasD + LasA precursor; lane 3, LasA 4- LasA precursor; and lane 4,15 kDa protein
+ LasA precursor.
194 LasA Precursor
LasA
Figure 5.5
195 Hg LasD and the total volume was adjusted to 15 ^1 with the same bulfer. The samples were incubated at 37“C for different times (0 to 2 hr) before analysis by immunoblotting. Again, anti-LasA polyclonal antiserum and the ECL kit were used
(45 sec incubation, 15 sec exposure) for the blotting. In Figure 5.6, lane 1 represents the LasA precursor control after 2 hr incubation at room temperature. Lanes 2 and
3 contain the LasA precursor with either 0.5 |ig LasA (lane 2) or 1 pg LasD (lane 3), respectively. Both lanes were used as either positive (lane 2) or negative (lane 3) controls and were not incubated. Lanes 4 through 9 represent the reaction between
LasD and the LasA precursor after different incubation times (0 to 2 hr) at room temperature. The incubation temperature in this experiment was room temperature instead of 37°C to slow down the reaction since the goal of this experiment was to examine the processing event gradually over time. In lane 4, after only a 10 min incubation, active LasA begins to appear indicating that the processing event is rapid. Note the ratio between the LasA precursor and LasD in this assay was 5 pg: 1 pg. In situ, the ratio of LasA to LasD is much smaller than 5:1; growth curve analysis suggests the ratio is actually closer to 1:1. As a result, the real processing event may be faster than what is observed in the figure, explaining why the LasA precursor was never purified in the native form. Besides the LasA band, another smaller band began to appear after 30 min incubation (Figure 5.6, lane 6 ). This band is obviously related to LasA because the band is recognized by the LasA specific antiserum. The identity of the band is not clear, but it is probably the result of processing at different sites, which gave rise to a smaller product. Lane 9
196 Figure 5.6: Immunoblotting showing the involvement of LasD in LasA processing.
Anti-LasA polyclonal antibody (1:2000) and the ECL kit was used in this assay.
After incubating the nitrocellulose membrane with the donkey anti-rabbit Ig- horseradish peroxidase conjugate, the membrane was reacted with the ECL reagents for 45 sec before exposing to a X-ray film for 15 sec. Lane 1, LasA precursor control, 2 hr incubation at room temperature; lane 2, LasA precursor + LasA, no incubation; lane 3, LasA precursor + LasD, no incubation; lane 4, LasA precursor +
LasD, 10 min incubation at room temperature; lane 5, LasA precursor + LasD, 20 min incubation at room temperature; lane 6 , LasA precursor + LasD, 30 min incubation at room temperature; lane 7, LasA precursor + LasD, 45 min incubation at room temperature; lane 8 , LasA precursor + LasD, 1 hr incubation at room temperature; and lane 9, LasA precursor + LasD, 2 hr incubation at room temperature.
197 LasA Precursor
•LasA
8
Figure 5.6
198 indicates the end of the processing after 2 hr incubation at room temperature.
Interestingly, the smaller band observed in lanes 6 through 8 is not detected in this lane indicating that the smaller product is either degraded or digested further by
LasD. This experiment serves as further proof that LasD is actually involved in the processing of LasA precursor and the processing event is proportional to the incubation time.
Finally, to find out whether the processed LasA has any activity, samples from Figure 5.5 were tested for the staphylolytic activity. In this experiment, the pH was adjusted to 7.0 with 0.1 M Tris-maleate buffer to distinguish the activity of processed LasA from that of LasD because LasD itself does not have any staphylolytic activity at pH 7.0 (Figure 3.4). To adjust the pH, 15 pi of each sample
from Figure 5.5 was mixed with 85 pi of heat-killed S. aureusceUs ( O . D . 5 9 5 value of
0.5) resuspended in 0.1 M Tris-maleate buffer (pH 7.0). Then the reaction mixtures were incubated for 2 hr at 37°C before measuring the staphylolytic activity by observing O . D . 5 9 5 . In Figure 5.7, the staphylolytic activity is represented as a function of 1 / O . D . 5 9 5 . Therefore, the higher value represents the higher staphylolytic activity. According to the figure, the sample containing LasD clearly exhibits staphylolytic activity. In other words, the active LasA generated by processing of the LasA precursor by LasD is enzymatically active. The activity is not due to the unprocessed LasA precursor since the LasA precursor only shows similar 1/ O . D . 5 9 5 value with the control (heat-killed S. aureus ceUs resuspended in
0.1 M Tris-maleate buffer, pH 8.0, and incubated for 1 hr at 37°C). As expected, the highest 1/ O . D . 5 9 5 value is from the sample containing LasA since LasA itself has
199 high staphylolytic activity. Therefore, the sample shows high activity without the processed LasA precursor.
D. Discussion
In this chapter, the role of LasD in LasA processing was examined. To study the processing event, purified LasA precursor is essential. However, the LasA precursor can not be isolated from P. aeruginosa cultures. Olsen and Ohman previously reported the existence of the LasA precursor in the supernatant when 100
|iM ZnCL was added to the supernatant (Olson and Ohman, 1992). The authors used TSBD medium (Trypticase soy broth treated with Chelex-100 and dialyzed against deionized water containing 1 % glycerol and 0.05 M monosodium glutamate) in which metal ions are removed by treating with an ion exchange resin
Chelex-100 (Bio-Rad, Hercules, CA). Although the authors did not purify the observed LasA precursor, it is the first and the only report to date indicating the presence of the LasA precursor in any P. aeruginosa fraction. In an attempt to purify the precursor, the experiment was repeated using the PAOl strain. However, we were unable to duplicate the experiment. There was no observable presence of the LasA precursor in the supernatant.
To overcome this problem, the LasA precursor was expressed in E. co//using the T7 overexpression system. Using the recombinant protein, it was possible to study processing events. However, we should be cautious about using this overexpressed protein since it may not represent true folding pattern of the native
LasA precursor. Since the overexpressed protein is denatured by 6 M guanidine-
2 0 0 HCl and renatured by step-wise dialysis, the majority of the refolded proteins may represent an incorrect conformation. The only way to investigate this possibility is to examine the enzymatic activity of the recombinant protein. However, the fact that the overexpressed LasA precursor lacks any enzymatic activity does not mean that the protein is in wrong conformation because precursor forms of most of the enzymes are inactive. For example, when various precursor forms of pseudolysin were tested for activity, none of them were active until they are fully processed to the active 33 kDa form (Kessler and Safrin, 1988; Kessler and Safrin, 1994). Therefore, the integrity of the overexpressed LasA precursor can only be assessed indirectly by examining the activity of the products after the precursor is processed (Figure 5.7).
However, it is possible to examine the integrity of the overexpressed active LasA fragment by measuring staphylolytic activity. In Figure 5.7, the recombinant active
LasA fragment clearly exhibits relatively strong staphylolytic activity suggesting that at least a portion of the recombinant protein has refolded to its native form.
When the overexpressed LasA precursor was incubated with various fractions of P. aeruginosa, addition of the supernatant fraction causes the disappearance of the precursor protein suggesting the possible processing event occurs in the supernatant (Figure 5.4). Further analysis of LasA processing event revealed that LasD is involved in the processing event (Figures 5.5 and 5.6).
However, in Figure 5.4, the LasA precursor started to disappear in the presence of the periplasmic space fraction. This may indicate that LasD is also processed and activated in the periplasmic space since LasD is the only factor involved in the LasA processing event even though the result may be due to another artifact unrelated to
201 Figure 5.7: Staphylolytic activity of the processed LasA precursor by various proteases.
Staphylolytic activity is expressed as a function of I/O .D .595. The higher
O.D.595 values represent a decrease in staphylolytic activity (cells are not lysed);
alternatively the higher /1 O . D . 5 9 5 values represent the higher staphylolytic activity.
Control is the heat-killed S. aureus cells resuspended in 0.1 M Tris-maleate buffer
(pH 7.0). The control is also incubated for 1 hr at 37°C.
2 0 2 the processing event such as the instability of the overexpressed LasA precursor in the sample.
Another unclear result in Figure 5.4 is that active LasA was not observed in the sample containing the LasA precursor and the supernatant fraction Gane 5).
This result may suggest two possibilities. First, the overexpressed precursor is completely digested by proteases present in the supernatant. If this is the case, the result of the assay is unreliable because what we are observing is simply the degradation of the unnaturally refolded LasA precursor by proteases when labeled amino acids are provided along with glucose as nutrients. The second possibility is that the labeled protein emits a low energy signal when it is cleaved and, as a result, the active LasA band is not detected after the processing event.
Because both hypotheses are feasible, immunoblotting was used instead as a detection method to eliminate the possible effect of the labeled amino acid mix.
Therefore, it was possible to identify LasD as the factor responsible for the processing event (Figures 5.5 and 5.6). Furthermore, the active LasA protein resulted from the processing of the LasA precursor has been demonstrated to be enzymatically active (Figure 5.7). Therefore, we conclude that LasD is responsible for processing LasA from an inactive precursor form (41 kDa) to an active form (22 kDa) in the supernatant.
203 CHAPTER 6
CONCLUSIONS
Pseudomonas aeruginosa produces several extracellular proteases including pseudolysin (elastase), LasA, and alkaline protease. Among them, pseudolysin
(elastase) is the most well characterized protease primarily because of its supported role in pathogenesis due to its ability to digest elastin fibers commonly found in connective tissues such as the air sacs of the lungs. LasA was first purified and characterized by Peters and Galloway as a second protease possessing elastolytic activity (Peters and Galloway, 1990) and various substrates for LasA have been identified (Peters, Park e ta l, 1992; Kessler, Safrin eta l., 1993; Park and Galloway,
1995).
During the course of this study, a new protease, LasD, has been purified and characterized from Pseudomonas aeruginosa strains. Detailed properties of LasD such as purification, physical properties, substrates, kinetic analysis, and involvement of LasD in LasA processing are addressed. Surprisingly, LasD shares many properties with LasA. LasA and LasD have very similar molecular weights,
22 kDa and 23 kDa, respectively. In addition, both enzymes have common substrates such as (3-casein, heat-killed Staphylococcus aureus cdh, and various
204 synthetic peptides. However, overwhelming evidence suggests LasD is distinct from
LasA. First, the N-terminal sequence of LasD shares no homology with either active LasA or the LasA precursor sequence. Second, LasA and LasD are purified under different conditions (Figure 3.1). Third, P. aerugmosa LasA knock-out mutant strains AD 1825 and FRD2128 do not produce LasA, yet produce LasD
(Figure 3.3). Fourth, specific antisera to each protease do not show any cross reactivity (Figure 3.3, panels B and C). Fifth, LasA and LasD have different pH and temperature optima (Figures 3.4 and 3.5, respectively). Sixth, LasD does not have any elastolytic activity whereas LasA has elastolytic activity. Seventh, LasA and
LasD cut the insulin p-chain at different sites (Figure 3.11, panels C and D). Eighth, pentaglycine is a better substrate for LasA than LasD (Figures 3.8 and 3.10, respectively). Finally, when P-casein is used as a substrate, TLCK is the most effective inhibitor against LasA whereas PMSF is the most effective inhibitor of
LasD (Figure 3.16).
During this study, P-casein was identified as a substrate for LasA as well as for LasD. This finding is in apparent contrast to results reported by Kessler e t al. who suggested LasA does not cleave P-casein (Kessler, Safhn e ta l., 1993). One possible explanation of this discrepancy is that both the purified LasA and LasD samples are contaminated by another protease, a common contaminant in both
LasA and LasD samples. However, in this study, three lines of evidence suggest that both proteases are pure. First, HPLC analysis of both the LasA and LasD samples using various columns such as a Bio-Gel MA7C cation exchange column
205 (Bio-Rad, Hercules, CA) and Bio-Gel DEAE-5-PW anion exchange column (Bio-
Rad, Hercules, CA) failed to identify any minor peak or contaminating material.
Second, N-terminal sequence analysis of both the LasA and LasD samples did not reveal any other residues in each sequencing cycle. Finally, laser densitometer scanning analysis of both samples on an SDS-PAGE gel did not detect any other bands. These results suggest that the LasA and LasD preparations used in these studies are free from any protein contaminants and that the digestion of P-casein can be attributed specifically to LasA or LasD. However, Kessler, e t al. only employed a
DEAE-cellulose column to purify LasA (Kessler, Safrin e t al., 1993). Under these conditions, it is impossible to purify LasA to homogeneity due to the weak anionic nature of the DEAE-cellulose resin. Therefore, the LasA sample eluted from the
DEAE-cellulose column most likely contains other minor contaminants. To prevent possible interference from the contaminants, Kessler, e t al. used serine protease inhibitors to inhibit the activity of any contaminating serine protease (Kessler, Safrin eta l., 1993). As a result, P-casein was not implicated as a substrate for LasA since the activity of LasA is inhibited in the presence of high concentration of the serine protease inhibitor when P-casein is used as substrate.
Another question arises regarding the classification of the LasA and LasD proteases. When heat-killed S. aureus cells are used as a substrate, activities of both proteases are inhibited by o-phenanthroline which acts as a metalloprotease inhibitor by chelating zinc atoms (Figures 3.14 and 3.15). When p-casein is the substrate however, both proteases are apparently inhibited by serine protease inhibitors such
206 as PMSF and DPP (Figure 3.16). Kessler, eta l. classified LasA as a metalloprotease based on the absence of pentaglycine hydrolysis by LasA in the presence of o phenanthroline (Kessler, 1995). However, LasA does not have any common metalloprotease motif such as HEXXH, although it has a possible HXH motif
(Rawlings and Barrett, 1995). A few proposed metalloproteases such as
Acbromobacter lyticus p-lytic endopeptidase contain this HXH motif, but zinc ligands have not been identified in those cases (Rawlings and Barrett, 1995). In addition, the occurrence of random HXH sequences is very high because it is a simple combination. Furthermore, a site-directed mutagenesis study on the HXH motif by Ohman, e ta l. proved inconclusive (unpublished data). Thus it has not been firmly established that LasA is a metalloprotease.
In addition to the possible metalloprotease motif, LasA also contains a possible serine protease catalytic triad. According to computer modeling analysis using HyperChem software (Autodesk, Inc., Sausalito, CA), Ser 138, His 120(122), and Asp 96 are located close together and form a possible catalytic triad. In addition, Gly 136 and Gly 140 are located near the catalytic triad, possibly aiding in the formation of a tetrahedral intermediate. These two residues are important since glycine residues tend to be conserved in the vicinity of the catalytic serine residue to form the GXSXG motif found in many serine proteases (Brenner, 1988). In addition, most known serine proteases contain two glycines proximal to the active site serine residue (Table 6.1). As a result, it is difficult to classify LasA (or LasD) as a metalloprotease or serine protease. To answer this question, site-directed
207 Family Active Site Serine Motif SA SI .. G . SG.... S2 .. G . SG .... S3 .. G . SG.... S5 .. G . SG .... S6 .. G . SG .... S30 .. G . SG.... S7 .. G . SG .... S29 .. G . SGG ... S31 .. G . SG .... SB S8 ..G.S SC S9 .. G . S .GG. SIO .. G . S .. G . SIS .. G . S .. G . SF S24 .. G . S .... G S26 • • • • S • • • • • SG S14 Cjr ...... S16 .. G . S . G ... S21 .... S....
Table 6.1: Conserved glycine residues near the active site serine residue in various families of serine proteases (modified from Rawlings and Barrett, 1995).
208 mutagenesis must be employed; by mutating Ser 138 to other amino acids, it may be possible to identify whether the serine residue is essential for activity.
Logical progression of the LasD project requires cloning of the LasD structural gene. For more than a year however, we have been unsuccessful in attempts to clone this gene possibly due to excessive redundancy in the genetic code representing the N-terminal amino acid sequence. However, using a variety of techniques such as colony hybridization and Southern blotting, several possible positive colonies containing cosmid vectors with EcoR I fragments of P. aeruginosa
PAOl genomic DNA were isolated. Successful isolation of the LasD structural gene would facilitate studies to determine the role of LasD in pathogenicity of P. aeruginosa using LasD negative mutants and observing the resulting phenotype. It would also be interesting to compare sequences for likely identification of an active site.
Like most bacterial extracellular proteases, both LasA and LasD may be involved in providing nitrogen-rich digestion products to facilitate growth of the organism (Steadman, Heck e ta l., 1993). In addition, LasA, LasD, and pseudolysin seem to work together efficiently. As depicted in Figure 6.1, pseudolysin, LasA, and possibly LasD are positively regulated by LasR. Following secretion, LasD is involved in the processing LasA from its inactive precursor form to an active form.
The active LasA subsequently nicks the elastin tissue providing a better substrate for pseudolysin. In addition, both LasA and LasD can lyse S. an/eus thereby providing cell debris available as a potential nutrient source for P. aeruginosa and allowing
Pseudomonas to survive in a competitive environment. Other substrates such as
209 p. aeruginosa LasR lasB Prepropseudolysin lasA ProLasA lasD ProLasD (?)
Other Substrates Other Substrates ProLasA (41 kDa) Active LasD (23 kDa) Active Pseudolysin (33 kDa)
Active LasA (22 kDa) liges ted Substrai
S. aureus Elastin Elastin | (Intact) i^(Nicked)|
^^sed 5. aureus cell d ^ ^ ^
Figure 6.1; Model illustrating possible roles of P. aeruginosa proteases.
2 1 0 various proteins and peptides are also digested by the proteases to be used as nutrients. Therefore, these proteases all work together to access nutrients and allow survival in environments where nutrients are scarce.
211 LIST OF REFERENCES
Allured, V. S., R. J. Collier, eta l. (1986). “Structure of exotoxin A of Pseudomonas aeruginosa at 3.0 A resolution.” Proc. Natl. Acad. Sci. USA 83: 1320-1324.
Andrews, A. T. (1986). Electrophoresis: theory, techniques, and biochemical and clinical applications. New York, Oxford Univ. Press.
Ausubel, F. M., R. Brent, e ta l., Eds. (1992). Current Protocols in Molecular Biology. New York, Green Publishing Associates and Wiley-Interscience: 2.5.1- 2.5.8.
Bachovchin, W. W. and J. D. Roberts (1978). “Nitrogen-15 nuclear magnetic resonance spectroscopy. The state of histidine in the catalytic triad of alpha-lytic protease. Implications for the charge-relay mechanism of peptide-bond cleavage by serine proteases.” J. Am. Chem. Soc. 100: 8041-8047.
Bainton, N. J., B. W. Bycroft, e ta l. (1992). “A general role for the /uxautoinducer in bacterial cell signaling: control of antibiotic synthesis in Erwinia." Gene 116: 87- 91.
Baker, N., V. Mino, e ta l. (1991). Pseudomonas aeruginosa exoenzyme S is an adhesin.” Infect. Immun. 59: 2859-2863.
Barrett, A. (1994). “Classification of peptidases.” Meth. Enzvmol. 244: 1-15.
Barrett, A. J., A. A. Kembhavi, eta l. (1982). “L-trans-Epoxysuccinyl-leucylamido(4- guanidino)butane (E-64) and its analogues as inhibitors of cysteine proteinases including cathepsins B, H, and L.” Biochem. J. 201: 189-198.
Bellido, F. and R. E. W. Hancock (1993). Susceptibility and resistance of Pseudomonas aeruginosa to antimicrobial agents. Pseudomonas aeruginosa as an opportunistic pathogen. M. Campa, M. Bendinelli and H. Friedman. New York, Plenum Press; 321-348.
Berk, R. S., D. Brown, etal. (1987). “To wVo studies with two phospholipase C fractions from Pseudomonas aeruginosa." Infect. Immun. 55: 1728-1730. 212 Berka, R. M. and M. Vasil (1982). “Phospholipase C (heat-Iabile hemolysin) of Pseudomonas aeruginosa: Purification and preliminary characterization.” L Bacteriol. 152: 239-245.
Besterman, J. M., V. Duronio, e ta l. (1986). “Rapid formation of diacylglycerol from phophatidylcholine: a pathway for generation of a second messenger.” Proc. Natl. Acad. Sci. USA 83: 6785-6789.
Bever, R. A. and B. H. Iglewski (1988). “Molecular characterization and nucleotide sequence of the Pseudomonas aeruginosa elastase structural gene.” J. Bacteriol. 170: 4309-4314.
Bimboim, H. C. (1983). “A rapid alkaline extraction method for the isolation of plasmid DNA.” Meth. Enzymol. 100:243-255.
Bimboim, H. C. and J. Doly (1979). “A rapid alkaline extraction procedure for screening recombinant plasmid DNA.” Nucl. Acids Res. 7: 1513-1523.
Bjom, M. J., M. L. Vasil, eta l. (1977). “Incidence of exotoxin production by Pseudomonassçedesi" Infect, Tmmun, 16: 362-366.
Botzenhart, K. and G. Doting (1993). Ecology and epidermiology of Pseudomonas aeruginosa. Pseudomonas aeruginosa as an opportunistic pathogen. M. Campa, M. Bendinelli and H. Friedman. New York, Plenum Press: 1-18.
Boyer, H. W. and D. Rolland-DussoLx (1969). “A complementation analysis of the restriction and modification of DNA in Escbericbia 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.
Brandhuber, B. J., V. S. Allured, eta l. (1988). “Mapping the enzymatic active site of Pseudomonas aeruginosa exotoxin A.” Proteins 3: 146-154.
Braun, P., J. Tommassen, e ta l. (1996). “Role of the propeptide in folding and secretion of elastase of Pseudomonas aeruginosa." Mol. Microbiol. 19: 297-306.
Brenner, S. (1988). “The molecular evolution of genes and proteins: a tale of two serines.” Nature 334: 528-30.
Brint, J. M. and D. E. Ohman (1995). “Synthesis of multiple exoproducts in Pseudomonas aeruginosa is under the control of RhlR-RMI, another set of regulators in strain PAOl with homology to the autoinducer-responsive LuxR-LuxI family.” J. Bacteriol. 177: 7155-7163.
213 Brito, N., M. A. Falcon, eta l. (1989). “Purification and peptidase activity of a bacteriolytic extracellular enzyme from Pseudomonas aeruginosa." Res. Microbiol. 140: 125-137.
Bubnis, W. A. and I. Ofher, C.M. (1992). “The determination of epsilon-amino groups in soluble and poorly soluble proteinaceous materials by a spectrophotometric method using trinitrobenzenesulfonic acid.” Anal. Biochem. 207(1): 129-133.
Burke, M. E. and P. A. Pattee (1967). “Purification and characterization of a staphylolytic enzyme from Pseudomonas aeruginosa." J. Bacteriol. 93: 860-865.
Burnette, W. N. (1981). “Western blotting: Electrophoretic transfer of proteins from sodium dodecyl sulfate-polyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and radioiodinated protein A.” Anal. Biochem. 112: 195-203.
Bye, M. R., J. M. Ewig, eta l. (1994). “Cystic fibrosis.” Lung 172: 251-270.
Camicero, A., M. Falcon, e ta l. (1990). “A protease with staphylolytic activity from Pseudomonas aerugmosa PAKSl.” FEMS Microbiol. Lett. 70: 171-176.
Carter, P. and J. A. Wells (1988). “Dissecting the catalytic triad of a serine protease.” Nature 332: 564-568.
Chasan, R. and K. V. Anderson (1989). “The role of Ester, an apparent serine protease, in organizing the dorsa-ventral pattern of the Drosophila embryo.” Cell 56: 391-400.
Cheng, K.-J., J. M. Ingram, e ta l. (1971). “Interactions of alkaline phosphatase and the cell wall of Pseudomonas aeruginosa." J. Bacteriol. 107: 325-336.
Chrambach, A. and D. Rodbard (1971). “Polyacrylamide gel electrophoresis.” Science 172:440-451.
Cleland, W. W. (1990). Steady-state kinetics. The enzymes. D. S. Sigman and P. D. Boyer. San Diego, Academic Press, Inc.: 99-158.
Cleveland, D. W., S. G. Fischer, eta l. (1977). “Peptide mapping by limited proteolysis in sodium dodecyl sulfate and analysis by gel electrophoresis.” J. Biol. Chem. 252: 1102-1106.
Cobum, J. and D. M. Gill (1992). Exoenzyme S ADP-ribosylates GTP-binding proteins and requires a eukaryotic protein for activity. Bacterial protein toxins. B. Witholt, J. E. Alouf, G. J. Boulnois e ta l., Stuttgart, Gustav Fischer Verlag: 234-236.
214 Cobum, J., A. V. Kane, e ta l. (1991). ''Pseudomonasaerugmosaex.oex\x^vo& S requires an eukaryotic protein for ADP-ribosyltransferase activity.” J. Biol. Chem. 266: 6438-6446.
Cobum, J., R. T. Yatt, e ta l. (1989). “Several GTP-binding proteins, including p21^" h-ras, are preferred substrates for Pseudomonas aeruginosa exoenzyme S.” J. Biol. Chem. 264: 9004-9008.
Coleman, J. E. and P. Gettins (1983). Molecular properties and mechanism of alkaline phosphatase. Zinc enzymes. T. G. Spiro. New York, John Wiley & Sons. 5: 153-217.
Colman, P. M., J. N. Jansonius, etal. (1972). “The structure of thermolysin: an electron density map at 2-3 Â resolution.” J. Mol. Biol. 70: 701-724.
Creighton, T. E. (1993). Proteins - stmctures and molecular properties. New York, W. H. Freeman and Company.
Cubo, M. T., A. Economou, eta l. (1992). “Molecular characterization and regulation of the rhizosphere-expressed genes rbiA B C R that can influence nodulation by Rhizobium leguminosarum." J. Bacteriol. 174: 4026-4035.
Dagert, M. and S. D. Ehrlich (1974). “Prolonged incubation in calcium chloride improves competence of Escherichia co//cells.” Gene 6:23-28.
Darzins, A., J. E. Peters, e ta l. (1990). “Revised nucleotide sequence of the lasA gene from Pseudomonas aeruginosa PAOl.” Nucl. Acids Res. 18(21): 6444.
Dayhoff, M. O., W. C. Barker, etal. (1983). “Establishing homologies in protein sequences.” Meth. Enzymol. 91: 524-545.
Diver, J. M., L. E. Bryan, e ta l. (1990). “Transformation of Pseudomonas aeruginosa by electroporation.” Anal. Biochem. 189: 75-79.
Dower, W. J., J. F. Miller, eta l. (1988). “High efficiency transformation of E. colihy high voltage electroporation.” Nucl. Acids Res. 16: 6127-6145.
Drake, D. and T. C. Montie (1988). “Flagella, motility and invasive virulence of Pseudomonas aeruginosa." J. Gen. Microbiology. 134: 43-52.
Dunn, B. M. (1990). Determination of protease mechanism. Proteolvtic enzvmes: a practical approach. R. J. Beynon and J. S. Bond. Oxford, IRL Press: 57-81.
215 Eberl, L., M. K. Winson, eta l. (1995). “Involvement of N-butanoly-L-homoserine lactone autoinducers in controlling the multicellular behavior of Serrada liqueSadens." Mol. Microbiol. 20:127-136.
Elliott, B. W. and C. Cohen (1986). “Isolation and characterization of a lysine- specific protease from Pseudomonas aerugmosa." J. Biol. Chem. 261: 11259-11265.
Evans, C. H., A. C. Wilson, eta l. (1989). “Preparative isoelectric focusing in ampholine electrofocusing columns versus immobile polyacrylamide gel for the purification of biologically active leukoregulin.” Anal. Biochem. 177: 358-363.
Fabre, E., J. Nicaud, e ta l. (1991). “Role of the proregion in the production and secretion of the Yarrowia lypolytica alkaline extracellular protease” J. Biol. Chem. 266: 3782-3790.
Fersht, A. (1985). Enzyme structure and mechanism. New York, W. H. Freeman and Company; 98-120.
Filloux, A., M. Murgier, eta l. (1987). “Characterization of two Pseudomonas aeruginosa mutants with defective secretion of extracellular proteins and comparison with other mutants.” FEMS Microbiol. Lett. 40: 159-163.
Fountoulakis, M., E. Takacs Di Lorenzo, eta l. (1993). “Purification of interferon gamma-interferon gamma receptor complexes by preparative electrophoresis on native gels.” Anal. Biochem. 208:270-276.
Freund, J. (1956). “The mode of action of immunologic adjuvants.” Adv. Tuberc. Res. 7: 130-148.
Fu, H., J. Cobum, e ta l. (1993). “The eukaryotic host factor that activates exoenzyme S of Pseudomonas aeruginosa is a member of the 14-3-3 protein family.” Proc. Natl. Acad. Sci. USA 90: 2320-2324.
Fujita, M., K. Tanaka, eta l. (1994). “Transcriptional regulation of the principal sigma factor genes, rpoD a n d rp o Sin Pseudomonas aeruginosa is controlled according to the growth phase.” Mol. Microbiol. 13: 1071-1077.
Fukushima, J., S. Yamamoto, eta l. (1989). “Structural gene and complete amino acid sequence of Pseudomonas aeruginosa EFO 3455 elastase.” J. Bacteriol. 171: 1698-1704.
Fuqua, W. C. and S. C. Winans (1994). “A LuxR-LuxI type regulatory system activates Agrobacterium Ti plasmid conjugal transfer in fiie presence of a plant tumor metabolite.” J. Bacteriol. 176: 2796-2806.
216 Fuqua, W. C., S. C. Winans, eta l. (1994). “Quorum sensing in bacteria: the LuxR- LuxI family of cell density responsive transcriptional regulators.” J. Bacteriol. 176: 269-275.
Galloway, D. R. (1993). Role of exotoxins in the pathogenesis of P. aeruginosa infections. Pseudomonas aeruginosa as an opportunistic pathogen. M. Campa, M. Bendinelli and H. Friedman. New York, Plenum Press: 107-127.
Galloway, D. R. (1991). Pseudomonas aeruginosa elastase and elastolysis revisited: recent developments.” Mol. Microbiol. 5: 2315-2321.
Gambello, M. J. and B. H. Iglewski (1991). “Cloning and characterization of the Pseudomonas aeruginosa lasR gene, a transcriptional activator of elastase expression.” J. Bacteriol. 173: 3000-3009.
Gambello, M. J., S. Kaye, eta l. (1993). “LasR of Pseudomonas aeruginosa is a transcriptional activator of the alkaline protease gene (api) and an enhancer of exotoxin A expression.” Infect. Immun. 61: 1180-1184.
Gill, D. M. (1982). “Bacterial toxins: A table of lethal amounts.” Microbiol. Rev. 46: 86-94.
Goldberg, J. B. and D. E. Ohman (1987a). “Cloning and transcriptional regulation of the elastase las A Gene in mucoid and nonmucoid Pseudomonas aeruginosa.” L Bacteriol. 169: 1349-1351.
Goldberg, J. B. and D. E. Ohman (1987b). “Activation of an elastase precursor by thelasA gene product of Pseudomonas aeruginosa” J. Bacteriol. 169:4532-4539.
Gordon, V. M. and S. H. Leppla (1994). “Proteolytic activation of bacterial toxins: role of bacterial and host cell proteases.” Infect Immun. 62: 333-340.
Govan, J. W. R. and V. Deretic (1996). “Microbial pathogenesis in cystic fibrosis; mucoid Psudomonas aeruginosa and Burkbolderia cepacia.” Microbiol. Rev. 60: 539-574.
Gray, K. M., L. Passador, eta l. (1994). “Interchangeability and specificity of components from the quorum-sensing regulatory systems of Vibrio Rscberi and Pseudomonas aeruginosa.” J. Bacteriol. 176: 3076-3080.
Groisman, E. A., N. Pagratis, eta l. (1991). “Genomic mapping and protein coding region identification using bacteriophage Mu.” Gene 99: 1-7.
Gu, M., V. M. Gordon, etal. (1996). “Furin regulates both the activation of Pseudomonas exoioyin. A and the quantity of the toxin receptor expressed on target cells.” Infect. Immun 64: 524-527. 217 Guzzo, J., F. Duong, eta l. (1991). “The secretion genes of Pseudomonas aeruginosa alkaline protease are functionally related to those of Erwinia cbrysantbemi proteases and Escbericbia coZ/a-hemolysin.” Mol. Microbiol. 5: 447-453.
Guzzo, J., M. Murgier, eta l. (1990). “Cloning of the Pseudomonas aeruginosa alkaline protease gene and secretion of the protease into the medium by Escbericbia coli." J. Bacteriol. 172: 942-948.
Guzzo, J., J. M. Pages, eta l. (1991). “Pseudomonas aeruginosa alkaline protease: evidence for secretion genes and study of secretion mechanism.” J. Bacteriol. 173: 5290-5297.
Habeeb, A. F. S. A. (1966). “Determination of free amino groups in proteins by trinitrobenzenesulfonic acid.” Anal. Biochem. 14: 328-336.
Hanahan, D. (1983). “Studies on transformation of Escbericbia co//with plasmids.” J. Mol. Biol. 166: 557-580.
Hancock, R. E. W. and W. A. Woodruff (1988). “Roles of porin and p-lactamase in P-Iactam resistance of Pseudomonas aeruginosa." Rev. Infect. Dis. 10: 770-775.
Harlow, E. and D. Lane (1988). Antibodies: a laboratory manual. Cold Spring Harbor Laboratory: 96-100.
Hartsuck, J. A. and J. Tang (1972). “The carboxylate ion in the active center of pepsin.” J. Biol. Chem. 247: 2575-2580.
Hase, C. C. and R. A. Finkelstein (1993). “Bacterial extracellular zinc-containing metalloproteases.” Microbiol. Rev. 57: 823-837.
Hatakeyama, T., H. Kohzaki, e ta l. (1992). “A microassay for proteases using succinylcasein as a substrate.” Anal. Biochem. 204: 181-184.
Hazlett, L. D., M. M. Moon, e ta l. (1991). “Analysis of adhesion, piliation, protease production and ocular infectivity of several P. aeruginosa strains.” Curr. Eve Res. 10: 351-362.
Hengge-Aronis, R. (1993). “Survival of hunger and stress: the role of rpoS'm early stationary phase regulation in Escbericbia coli." Cell 72: 165-168.
Hofmann, T., B. M. Dunn, e ta l. (1984). “Cryoenzymology of penicillopepsin; with an appendix: mechanism of action of aspartyl proteinases.” Biochemistry 23: 5247- 5256.
Holloway, B. W.,V. ïCrishnapillai, eta l. (1979). “Chromosomal genetics of Pseudomonas." Microbiol. Rev. 43: 73-102. 218 Holmes, M. A. and B. W. Matthews (1982). “Structure of thermolysin refined at 1.6 A resolution.” J. Mol. Biol. 160: 623-639.
Hoshino, T. (1979). “Transport systems for branched-chain amino adds in Pseudomonas aerugmosa." J. Bacteriol. 139: 705-712.
Howe, T. R. and B. H. Iglewski (1 9 8 4 ). “Isolation and characterization of alkaline protease-defident mutants of Pseudomonas aerugmosa in vitro and in a mouse eye model ” Infect. Tmmun 43: 1 0 5 8 -1 0 6 3 .
Howe, T. R. and B. H. Iglewski (1984). “Isolation and characterization of alkaline protease defident mutants of Pseudomonas aeruginosa in vitro and in a mouse eye model ” Tnfeef Tmmun 43: 1058-1063.
Huber, R. and W. Bode (1978). “Structural basis of the activation and action of trypsin.” Acc. Chem. Res. 11: 114-122.
Iglewski, B. H., P. V. Liu, eta l. (1977). “Mechanism of action o f Pseudomonas aerugmosa exotoxin A is adenosine diphosphate-ribosylation of mammalian elongation factor 2 in vitro and in vivo." Infect, Tmmun, 15: 138-144.
Dcemura, H., H. Takagi, eta l. (1987). “Requirement of pro-sequence for the production of active subtilisin E in Escbericbia coli." J. Biol. Chem. 262: 7859-7864.
Inouye, M. (1991). “Intramolecular chaperone: the role of the pro-peptide in protein folding.” Enzyme 45: 314-321.
Irvin, R. T. (1993). Attachment and colonization of Pseudomonas aeruginosa: role of the surface structure. Pseudomonas aeruginosa as an opportunistic pathogen. M. Campa, M. Bendinelli and H. Friedman. New York, Plenum Press: 19-42.
Isacsson, V. and G. Wettermark (1974). “Chemiluminescence in analytical chemistry.” Anal. Chim. acta. 68: 339-362.
Isobe, T., T. Ichimura, eta l. (1991). “Distinct forms of the protein kinase-dependent activator of tyrosine and tryptophan hydroxylases.” J. Mol. Biol. 217: 125-132.
Jiang, W. and J. S. Bond (1992). “Families of metalloendopeptidases and their relationship.” FEBS Letters 312: 110-114.
Jones, S., B. Yu, e ta l. (1993). “The autoinducer regulates the production of exoenzyme virulence determinants in Erwinia carotovora and Pseudomonas aeruginosa." EMBO J. 12: 2477-2482.
Kalwant, S. and A. G. Porter (1991). “Purification and characterization of human prolyl endopeptidase.” Biochem. J. 276:237-244. 219 Kessler, E., E. Safrin, eta l. (1993). “Secreted LasA of Pseudomonas aerugmosa is a staphylolytic protease.” J. Biol. Chem. 268: 7503-7508.
Kessler, E. and M. Safrin (1988). “Partial purification and characterization of an inactive precursor of Pseudomonas aerugmosa elastase.” J. Bacteriol. 170:1215- 1219.
Kessler, E. and M. Safnn (1988). “Synthesis, processing, and transport of Pseudomonas aerugmosa elastase.” J. Bacteriol. 170: 5241-5247.
Kessler, E. and M. Safnn (1994). “The propeptide of Pseudomonas aerugmosa elastase acts as an elastase inhibitor.” J. Biol. Chem. 269:22726-22731.
Kessler, E., M. Safnn, eta l. (1993). “Inhibitors and specificity of aerugmosa LasA.” Am. Soc. Microbiol. Proc. Abst 93: B-328.
Kessler, S. P. andD. R. Galloway (1992). Pseudomonas aerugmosa exoto^àn. A. interaction with eukaryotic elongation factor 2.” J. Biol. Chem. 267: 19107-19 111.
Kilbane, J. I. and B. A. Bielaga (1991). “Instantaneous gene transfer from donor to recipient microorganisms via electroporation.” BioTechniques 10: 354-365.
Knect, D. A. and R. L. Dimond (1984). “Visualization of antigenic proteins on Western blots.” Anal. Biochem. 136:180-184.
Kounnas, M. S., R. E. Morris, eta l. (1992). “The alpha 2-macroglobulin receptor/low density hpoprotein receptor-related protein binds and internalizes PseudomonasexoXoxm A.” J. Biol. Oiem. 267: 12420-12423.
Kyriakopoulos, A., M. Kalckloesch, eta l. (1993). “Studies on 16 kDa selenium- containing proteins enriched by means of preparative electrophoresis.” Electrophoresis 14: 108-111.
Lache, M., W. R. Hearn, eta l. (1969). “Specificity of a bacteriolytic enzyme from Pseudomonas aerugmosa." J. Bacteriol. 100: 254-259.
Lache, M., W. R. Hearn, etal. (1969). “Specificity of a bacteriolytic enzyme from Pseudomonas aerugmosa." J. Bacteriol. 100: 254-259.
Laemmli, U. K. (1970). “Cleavage of structural proteins during the assembly of the head of bacteriophage T4.” Nature 227: 680-685.
Latifi, A., M. Foglino, eta l. (1996). “A hierarchical quorum-sensing cascade in Pseudomonas aerugmosa links the transcriptional activators LasR and RhIR (VsmR) to expression of the stationary-phase sigma factor RpoS.” Mol Microbiol. 21: 1137- 1146. 2 2 0 Latifi, A., M. K. Winson, e ta l. (1995). “Multiple homologues of LuxR and Luxl control expression of virulence determinants and secondary metabolites through quorum sensing in Pseudomonas aeruginosa PAOl.” Mol. Microbiol. 17: 333-343.
Lipman, D. J. and W. R. Pearson (1985). “Rapid and sensitive protein similarity searches.” Science 227: 1435-1441.
Lipscomb, W. N., J. A. Hartsuck, eta l. (1969). “The structure of carboxypeptidase A. DC. The x-ray diffraction results in the light of the chemical sequence.” Proc. Natl. Acad. Sci. USA 64: 28-35.
Liu, P. V. (1979). Toxins oï Psudomonas aerugmosa. Pseudomonas aeruginosa - Clinical manifestations of infection and current therapv. R. G. Doggett. New York, Academic Press; 63-88.
Liu, P. V. (1966). “The roles of various fractions of Pseudomonas aeruginosa in its pathogenesis, n . Effects of lecithinasein protease.” J. Infect. Pis. 116: 112-116.
Liu, P. V., Y. Abe, eta l. (1961). “The role of various fractions of Pseudomonas aeruginosa in its pathogenesis.” J. Infect. Pis. 108:218-228.
Mandel, M. and A. Higa (1970). “Calcium dependent bacteriophage DNA infection.” J. Mol. Biol. 53: 159-162.
Matsudaira, P. (1987). “Sequence from picomole quantities of proteins electroblotted onto polyvinylidene difluoride membranes.” J. Biol. Chem. 262: 10035-10038.
Matsudaira, P. T. and D. R. Burgess (1978). “SDS microslab linear gradient polyacrylamide gel electrophoresis.” Anal. Biochem. 87: 386-396.
McGowan, S., M. Sebaihia, eta l. (1995). “Carbapenem antibiotic production in Erwinia carotovora is regulated by CarR, a homologue of the LuxR transcriptional activator.” Microbiologv 141: 541-550.
Mclver, K. S., E. Kessler, eta l. (1991). “Substitution of active site His-223 in Pseudomonas aeruginosa elastase and expression of the mutated lasB alleles in Escbericbia co//shows evidence for autoproteolytic processing of proelastase.” L Bacteriol. 173: 7781-7789.
Mclver, K. S., E. Kessler, eta l. (1995). “The elastase propeptide functions as an intramolecular chaperone required for elastase activity and secretion in Pseudomonas aeruginosa." Mol. Microbiol. 18:877-889.
221 Mclver, K. S., J. C. OIsen, eta l. (1993). Pseudomonas aeruginosa lasBl mutants produce an elastase, substituted at active-site His-223, that is defective in activity, processing, and secretion." J. Bacteriol. 175:4008-4015.
Meighen, E. A. (1991). “Molecular biology of bacterial bioluminescence." Microbiol. Rev. 55: 123-142.
Meyers, D. J. and R. S. Berk (1990). “Characterization of phospholipase C from Pseudomonas aeruginosa as a potent inflammatory agent.” Infect. Immun. 58: 659- 666.
Montie, T. C., D. Drake, eta l. (1987). “Mobility, virulence and protection with a flagella vaccine against Pseudomonas aeruginosa infection.” Antibiot. Chemother. 39: 233-248.
Morihara, K. (1964). “Production of elastase and proteinase by Pseudomonas aerugmosa." J. Bacteriol. 88: 745-757.
Morihara, K. (1995). “Pseudolysin and other pathogen endopeptidases of thermolysin familv.” Meth. Enzvmol. 248: 2^-253.
Morihara, K. and J. Y. Homma (1985). Pseudomonas^toied&çs. Bacterial enzvmes and virulence. I. A. Holder. Boca Baton, CRC Press: 42-77.
Morihara, K., T. Oka, eta l. (1971). “Comparative study of various serine proteinases from microorganisms: specificity with oligopeptides." Arch. Biochem. Biophvs. 146: 297-305.
Morihara, K. and H. Tsuzuki (1966). “Substrate specificity of elastolytic and nonelastolytic proteinases from Pseudomonas aeruginosa.” Arch. Biochem. Biophvs. 114: 158-165.
Morihara, K., H. Tsuzuki, eta l. (1973). “On the specificity of Pseudomonas aeruginosa alkaline proteinase with synthetic peptides." Biochim. Biophvs. Acta. 309: 414-429.
Morihara, K., H. Tsuzuki, eta l. (1965). “Pseudomonas aeruginosa elastase, isolation, crystallization, and preliminary characterization.” J. Biol. Chem. 240: 3295-3304.
Neurath, H. (1984). “Evolution of proteolytic enzymes." Science 224: 350-357.
Neurath, H. (1985). “proteolytic enzymes, past and present." Fed. Proc. 44: 2907- 2913.
2 2 2 Nicas, T. L and B. H. Iglewski (1985). “Contribution of exoenzyme S to the virulence of Pseudomonas aerugmosa." Antibiot. Chemother. 36: 40-48.
Nikaido, H. and R. E. W. Hancock (1986). Outer membrane permeability of Pseudomonas aeruginosa. The bacteria. J. R. Sokatch. Orlando, Academic Press. 10: 145-193.
Nomenclature Committee of the International Union of Biochemistry and Molecular Biology (1992). Enzyme nomenclature 1992. Orlando, Academic Press.
Ochsner, U. A., A. K. Koch, e ta l. (1994). “Isolation and characterization of a regulatory gene affecting rhamnolipid biosurfactant synthesis in Pseudomonas aerugmosa." J. Bacteriol. 176: 2044-2054.
Ochsner, U. A. and J. Reiser (1995). “Autoinducer-mediated regulation of rhamnolipid biosurfactant synthesis in Pseudomonas aerugmosa." Proc. natl. Acad. Sci. USA 92: 6424-6428.
O’Donnell, L. C. (1995). Analysis of a s- and trans- acting factors involved in regulation of LasA production, a second elastase produced by Pseudomonas aerugmosa. Department of Microbiology. Columbus, Ohio State University.
Ogata, M., V. K. Chaudhary, e ta l. (1990). “Processing of Pseudomonas exotoxin by a cellular protease results in the generation of a 37,000-Da toxin fragment that is translocated to the cytosol.” J. Biol. Chem. 265:20678-20685.
Ohman, D. E., S. J. Cryz, e ta l. (1980). “Isolation and characterization of a Pseudomonas aerugmosa PAO that produces altered elastase.” J. Bacteriol. 142: 836-842.
Olson, J. C. and D. E. Ohman (1992). “Efficient production and processing of elastase and LasA by Pseudomonas aerugmosa require zinc and calcium ions.” L Bacteriol. 174: 4140-4147.
Olsen, R. H., G. Debusscher, e ta l. (1982). “Development of broad-host-range vectors and gene banks: self-cloning of the Pseudomonas aeruginosa PAO chromosomes.” J. Bacteriol. 150: 60-69.
Park, S. J. and D. R. Galloway (1995). “Purification and characterization of LasD: a second staphylolytic proteinase produced by Pseudomonas aeruginosa." Mol. Microbiol. 16: 263-270.
Parkes, C., A. A. Kembhavi, e ta l. (1985). “Calpain inhibition by peptide epoxides.” Biochem. J. 230: 509-516.
223 Passador, L., J. M. Cook, etal. (1993). “Expression of Pseudomonas aerugmosa virulence genes requires cell-to-cell communication.” Science 260: 1127-1130.
Passador, L. and B. H. Iglewski (1995). Quorum sensing and virulence gene regulation in Pseudomonas aeruginosa. Virulence mechanism of bacterial pathogens. Roth. J. A. Washington, D C., American Society for Microbiology: 65- 78.
Pavlovskis, O. R., M. Pollack, etal. (1977). “Passive protection by antitoxin in experimental Pseudomonas aeruginosa bum infections.” Infect. TmmunoL 18: 596- 602.
Pearson, J., K. M. Gray, etal. (1994). “Stmcture of the autoinducer required for expression of Pseudomonas aeruginosa vimlence genes.” Proc. Natl. Acad. Sci. USA 91: 197-201.
Pearson, J. P., L. Passador, etal. (1995). “A second N-acylhomoserine lactone signal produced by Pseudomonas aeruginosa." Proc. Natl. Acad. Sci. USA 92: 1490- 1494.
Pearson, W. R. andD. J. Lipman (1988). “Improved tools for biological sequence comparison.” Proc. Natl. Acad. Sci. USA 85: 2444-2448.
Perona, J. and C. S. Craik (1995). “Stmctural basis of substrate specificity in the serine proteases.” Protein Science 4: 337-360.
Peters, J. E. (1990). Pseudomonas aeruginosa LasA protein: purification, characterization, and its role in enhancement of elastase activity. Department of Microbiology. Columbus, OH, The Ohio State University: 81-84.
Peters, J. E. and D. R. Galloway (1990). “Purification and characterization of an active fragment of the LasA protein from Pseudomonas aeruginosa', enhancement of elastase activity.” J. Bacteriol. 172: 2236-2240.
Peters, J. E., S. J. Park, etal. (1992). “Further studies of Pseudomonas aeruginosa LasA: analysis of specificity.” Mol. Microbiol. 6: 1155-1162.
Petrash, J. M., L. J. DeLucas, etal. (1991). “Resolving isoforms of aldose reductase by preparative isoelectric focusing in the Rotofor cell.” Electrophoresis 12(1): 84-90.
Pierson, S. L. I., V. D. Keppenne, etal. (1994). “Phenazine antibiotic biosynthesis in Pseudomonas aureofaciens regulated is by PhzR in response to cell density.” L Bacteriol. 176: 3966-3974.
224 Piper, K. R., S. Beck von Bodman, etal. (1993). “Conjugation factor of Agrobacterium ft//ne/âae»5 regulates Ti plasmid transfer by autoinduction.” Nature 362: 448-450.
Pirhonen, M., D. Flego, etal. (1993). “A small diffusible signal molecule is responsible for the global control of virulence and exoenzyme production in the plant pathogen Erwinia carotovora." EMBO J. 12:2467-2476.
Polgar, L. (1989). Mechanisms of protease action. Boca Raton, CRC Press: 157-182.
Pollack, M. (1990). Pseudomonas aeruginosa. Principles and practice of infectious disease. G. L. Mandell, R. G. Douglas and J. E. Beimett. New York, Churchill Lvingstone, Inc.: 1673-1691.
Pollack, M., N. S. Taylor, etal. (1977). “Exotoxin production by clinical isolates of Pseudomonas aeruginosa.” Infect. Immun. 15: 776-780.
Rawlings, N. D. and A. J. Barrett (1993). “Evolutionary families of peptidases.” Biochem. J. 290: 205-218.
Rawlings, N. D. and A. J. Barrett (1994a). “Families of serine peptidases.” Meth. Enzymol. 244: 19-61.
Rawlings, N. D. and A. J. Barrett (1994b). “Families of cysteine peptidases.” Meth. Enzymol. 244: 461-486.
Rawlings, N. D. and A. J. Barrett (1995). “Evolutionary families of metalloproteases.” Meth. Enzymol. 248: 183-228.
Reeck, N. D., C. de Haen, etal. (1987). “"Homology" in proteins and nucleic acids: a terminology muddle and a way out of it.” Cell 50: 667.
Rhame, F. S. (1980). The ecology and epidermiology of Pseudomonas aeruginosa. Pseudomonas aeruginosa, the organism, disease it causes, and their treatment. L. D. Sabath. Bern, Hans Huber Publishers: 31-51.
Rosenberg, A. H., B. N. Lade, etal. (1987). “Vectors for selective expression of cloned DNAs by T7 RNA polymerase.” Gene 56: 125-135.
Salmond, G. P. C., B. W. Bycroft, etal. (1995). “The bacterial 'enigma': cracking the code of cell-cell communication.” Mol. Microbiol. 16: 615-624.
Salvesen, G. and H. Nagase (1990). Inhibition of proteolytic enzymes. Proteolvtic enzymes: a practical approach. R. J. Beynon and J. S. Bond. Oxford, ERL Press: 83- 104.
225 Sandberg, L. B., N. T. Soskel, eta l. (1981). “Elastin structure, biosynthesis, and relation to disease states.” New England Journal of Medicine 304: 566-579.
Saulnier, J. M., F. M. Curtil, eta l. (1989). “Elastolytic activity of Pseudomonas aeruginosa elastase.” Biochim. Biophvs. Acta. 995:285-290.
Schad, P. A., R. A. Bever, eta l. (1987). “Cloning and characterization of elastase genes from Pseudomonas aeruginosa.” J. Bacteriol. 169: 2691-2696.
Schad, P. A. and B. H. Iglewski (1988). “Nucleotide sequence and expression in Escbericbia coli of thePseudomonas aeruginosa lasA gene.” J. Bacteriol. 170: 2784- 2789.
Schweizer, H. P. (1991).'"Escbericbia-PseudomonasshuVile. vectors derived from pUC18/19.” Gene 97: 109-112.
Segel, I. H. (1968). Biochemical Calculations. New York, John Wiley & Sons, Inc.: 368.
Shinde, U. and M. Inouye (1993). “Intramolecular chaperones and protein folding.” Trends. Biochem. Sci. 18:4^-446.
Silen, J. L. (1989). “Analysis of prepro-a-lytic protease expression in Escbericbia reveals that the pro region is required for activity.” J. Bacteriol. 171: 1320-1325.
Smith, A. W. and B. H. Iglewski (1989). “Transformation of Pseudomonas aeruginosa by electroporation.” Nucl. Acids Res. 17: 10509.
Smith, C. L. and R. DeLotto (1994). “Ventralizing signal determined by protease activation in Drosophila embryogenesis.” Nature 368: 548-551.
St. Clair, N. L. and M. Sax (1990). “Free-solution isoelectric focusing for the purification of Staphylococcus aureusevAexotoyàn. Cl.” Protein Expression & Purification 1: 97-103.
Stark, W., R. A. Pauptit, eta l. (1992). “The structure of neutral protease from Bacillus cereusdX 0.2-nm resolution.” Eur. J. Biochem. 207: 781-791.
Steadman, R., L. W. Heck, eta l. (1993). The role of proteases in the pathogenesis of Pseudomonas aeruginosa infections. Pseudomonas aeruginosa as an opportunistic pathogen. M. Campa, M. Bendinelli and H. Friedman. New York, Plenum Press: 129-143.
Stryer, L. (1988). Biochemistry. New York, W. H. Freeman and Company: 177-198.
226 Studier, F. W. and B. A. MofFatt (1986). “Use of bacteriophage T7 RNA polymerase to direct selective high-level expression of cloned genes.” J. Mol. Biol. 189: 113-130.
Studier, F. W., A. H. Rosenberg, eta l. (1990). “Use of T7 RNA polymerase to direct expression of cloned genes.” Meth. Enzvmol. 185: 60-89.
Sutcliffe, J. G. (1979). “Complete nucleotide sequence of the Escbericbia coli plasmid pBR322.” Cold Spring Harbor Svmp. Quant. Biol. 43: 77-90.
Swenson, D. L., N. O. Bukanov, e ta l. (1996). “Two pathogenicity islands in urophthogenic Escbericbia coli J96: cosmid cloning and sample sequencing.” Infect. Tmmun. 64: 3736-3743.
Swift, S., N. J. Bainton, eta l. (1994). “Gram negative bacterial communication by N-acyl homoserine lactons: a universal language?” Trends. Microbiol. 2: 193-198.
Swift, S., M. K. Winson, eta l. (1993). “A novel strategy for the isolation of lu x l homologues: evidence for the widespread distribution of a LuxR:LuxI superfamily in enteric bacteria.” Mol. Microbiol. 10: 511-520.
Tabor, S. and C. C. Richardson (1985). “A bacteriophage T7 RNA polymerase/promoter system for controlled exclusive expression of specific genes.” Proc. Natl. Acad. Sci. USA 82: 1074-1078.
Tanaka, K. and H. Takahashi (1994). “Cloning, analysis and expression of an q>oS homologue gene from Pseudomonas aeruginosa PAOl." Gene 150: 81-85.
Tang, J. (1979). “Evolution in the structure and function of carboxyl proteases.” Mol. Cell. Biochem. 26: 93-109.
Thayer, M. M., K. M. Flaherty,eta l. (1991). “three-dimensional structure of the elastase of Pseudomonas aeruginosa at 1.5-Â resolution.” J. Biol. Chem. 266: 2864- 2871.
Throup, J., M. Camara, etal. (1995). “Characterization of the y e n l/y e n R locus from Yersinia enterocolitica mediating the synthesis of two N-acylhomoserine lactone signal molecules.” Mol. Microbiol. 17: 345-356.
Toder, D. S., M. J. GambeUo, eta l. (1991). Pseudomonas aeruginosa LasA; a second elastase under the transcriptional control of lasR." Mol. Microbiol. 5:2003- 2010.
Towbin, H., T. Staehelin, etal. (1979). “Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets; procedure and some applications.” Proc. Natl. Acad. Sci. USA 76: 4350-4354. 227 Twining, S. S., S. E. Kirschner, etal. (1993). “Effect of Pseudomonas aerugmosa elastase, alkaline protease and exotoxin A on comeal proteinases and proteins.” Invest. Ophthalmol. Vis. Sci. 34: 2699-2712.
Umezawa, H. (1976). “Structures and activities of protease inhibitors of microbial origin.” Meth. Enzymol. 45: 678-695.
Urry, D. W. (1983). “What is elastin; what is not.” Ultrastmctural Pathologv 4: 227- 251.
Vallee, B. L. (1983). Zinc in biology and biochemistry. Zinc enzvmes. T. G. Spiro. New York, John Wiley & Sons. 5: 1-24.
Vallee, B. L. and D. S. Auld (1990). “Zinc coordination, function, and stmcture of zinc enzymes and other proteins.” Biochemistrv 29: 5647-5659.
Waldor, M. K. and Mekalanos, J. J. (1996). “Lysogenic conversion by a filamentous phage encoding cholera toxin.” Science 272: 1910-1914.
Wandersman, C. (1989). “Secretion, processing and activation of bacterial extracellular proteases.” Mol. Microbiol. 3: 1825-1831.
Whitaker, D. R. (1965). “Lytic enzymes of Sorangium sp.. Isolation and enzymatic properties of the alpha and beta-lytic proteases.” Can. J. Biochem. 43: 1935-1954.
Whitaker, D. R. and C. Roy (1967). “Concerning the nature of the alpha- and beta- lytic proteases of Sorangium sp..” Can. J. Biochem. 45: 911-916.
Wiener-Kronish, J. P., T. Sakuma, etal. (1993). “Aveolar epithelial injury and plueral empyema in acute P. aeruginosa pneumonia in anesthetized rabbits.” L Appl. Physiol. 75: 1661-1669.
Winson, M. K., M. Camara, etal. (1995). “Multiple N-acyl-L-homoserine lactone signal molecules regulate production of vimlence determinants and secondary metabolites in Pseudomonas aeruginosa." Pro. Natl. Acad. Sci. USA 92: 9427-9431.
Winther, J. R. and P. Sorensen (1991). “Propeptide of carboxypeptidase Y provides a chaperone-like function as well as inhibition of the enzyme activity.” Proc. Natl. Acad. Sci. USA 88: 9330-9334.
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.
Yamaizumi, M., E. Mekada, etal. (1978). “One molecule of diphtheria toxin fragment A introduced into a cell can kill the cell.” Cell 15: 245-250. 228 Yanisch-Perron, S., J. Vieira, eta l. (1985). “Improved M13 cloning vectors and host strains: nucleotide sequence of the M13mpl8 and pUC19 vectors.” Gene 33: 103- 119.
Zupan, L. A., D. L. Steffens, eta l. (1992). “Cloning and expression of a human 14- 3-3 protein mediating phospholipolysis. Indication of an arachidonoyl-enzyme intermediate during catalysis.” J. Biol. Chem. 267: 8707-8710.
229